1986_PMI_Linear_and_Conversion_Products_Data_Book 1986 PMI Linear And Conversion Products Data Book

User Manual: 1986_PMI_Linear_and_Conversion_Products_Data_Book

Open the PDF directly: View PDF PDF.
Page Count: 899

Download1986_PMI_Linear_and_Conversion_Products_Data_Book 1986 PMI Linear And Conversion Products Data Book
Open PDF In BrowserView PDF
Alpha- .

numerIC

Index
Part
Number

Page

ADC-910 ............. 12-7
ADC8208 ........... 1219
AMP-Ol ......... , ... , .. 6-5
AMP-OS ............. 6-27
BUF-03 ................ 7-4
CMP-OJ ................ 8-6
CMP-02., ............ 8-14
CMP-04 .............. 8-22
CMP-05 .............. 8-30
CMP-07 , ............. 8-37
CMP-404 ............ 8-40
DAC-OJ .............. 11-10
DAC-02 .............. 11-14
DAC-03 ............. 11-14
DAC-05 ........ , .... 11-14
DAC-06 .............. 11-19
DAC-08 ............. 11-23
DAC-IO ............. 11-34
DAC-20 ............. 11-42
DAC-86 ............. II-50
DAC-88 ............. 11-58
DAC-89 .............. 11-67
DAC-100 ............. 11-76
DAC-208 ........... 11-84
DAC-210 ............ 11-88
DAC-312 ............ 11-92
DAC-888 ........... 1l-104
DAC-1408 .......... 11-116
DAC-1508 .......... 11-116
DAC-80l2 .......... 11-122
DAC-8212 .......... 11-131
DAC-8408 .......... 11-136
DMX-88 ........... , .13-74
GAP-OJ .............. 15-~
JM3851O/10l04 ..... 5-273
JM3851O/10106 ..... 5-276
JM3851O/ll004 ..... 5-279
JM3851O/ll30l .... 11-209
JM3851O/ll302 .... 11-209
JM3851O/ll40l ..... 5-282
JM38510/11402 ..... 5-282

JM3851O/11403 ..... 5-282
JM3851O/11404 ..... 5-282
JM3851O/11405 ..... 5-282
JM3851O/11406 ..... 5-282
JM38510/13501 ..... 5-292
JM3851O/13502 ..... 5-292
JM38510/13503 ..... 5-295
MAT-OJ ................ 9-5
MAT-02 .............. 9-11
MAT-04 ............. 9-23
MUX-08 ............ 13-48
MUX-16 ............. 13-59
MUX-24 ............ 13-48
MUX-28 ............ 13-59
MUX-88 ............ 13-69
OP-OJ ................. 5-12
OP-02 ................ .5-18
OP-04 ................ 5-26
01'-05 ................ 5-34
01'-06 ................ 5-43
OP-07 ................ 5-51
OP-08 ................ 5-61
OP-09 ................ 5-68
01'-10 ................. 5-75
01'-11 ................ 5-68
01'-12 ................ 5-87
OP-14 ................ 5-26
OP-15 ................ 5-91
01'-16 ................ 5-91
OP-17 ................ 5-91
OP-20 ............... 5-105
01'-21 ................ 5-1ll
01'-22 ............... ,5-117
01'-27 ............... 5-128
OP-32 ............... 5-140
OP-37 ............... 5-152
01'-41 ............... 5-164
OP-43 ................ 5-175
OP-50 ............... 5-180
OP-77 ................ 5-191
01'-90 ............... 5-197
OP-207 ............. 5-199
01'-215 .............. 5-205
OP-220 ............. 5-212
01'-221 .............. 5-220
OP-227 .............. 5-228
OP-400 ............. 5-239
OP-420 ............. 5-241
OP-421 .............. 5-246
PKD-OJ .............. 15-19
PM-108 .............. 5-252

PM-Ill ............... 8-48
PM-119 ............... 8-54
PM-139 ............... 8-56
PM-146 ............ ,.5-256
PM-155 .............. 5-259
PM-156 .............. 5-259
PM-157 .............. 5-259
PM-208 ............. 5-252
PM-211 ............... 8-48
PM-219 .............. 8-54
PM-246 ............. 5-256
PM-308 ... , ......... 5-252
PM-319 .............. 8-54
PM-339 .............. 8-56
PM-355 ............. 5-259
PM-356 ............. 5-259
PM-357 ............. 5-259
PM-562 ............. 11-140
PM-565 ............ 11-148
PM-725 ............. 5-265
PM-741 .............. 5-268
PM-747 ............. 5-270
PM-2108 ............ 5-252
PM-7226 ............ 11-149
PM-7524 ............ 11-150
PM-7528 ............ 11-159
PM-7533 ............ 11-175
PM-7541 ............ 11-185
PM-7542 ............ 11-195
PM-7543 ............ 11-196
PM-7545 ............ 11-197
PM-7548 ........... 11-208
PM-7645 ............ 11-197
REF-OJ ............... 10-4
REF-02 ............... 10-11
REF-03 ............ , .. 10-19
REF-OS .............. 10-21
REF-IO .............. 10-22
RPT-82 .............. 16-4
RPT-83 .............. 16-4
SMP-IO ............... 14-6
SMP-11 ............... 14-6
SMP-81 .............. 14-15
SW-OJ ................. 13-7
SW-02 ................. 13-7
SW-05 ..... , .......... 13-14
SW-06 ............... 13-22
SW-20l ............. 13-33
SW-202 ............. 13-33
SW-751O ........ , .... 13-40
SW-7511 ............. 13-40

I,

Linear and
Conversion
Products
1986/1987
Data Book

The PMI
Commitment
PMI is committed to
building long-term
customer relationships
resulting in mutual
growth.
At PMI we dedicate
ourselves to leadership
in customer service,
quality; and technology.
Our goal is flawless
performance and
professional excellence.

Copyright@ 1986
Precision Monolithics Inc.
PMI reserves the right to make changes to the products
contained in this data book to improve performance, relIabIlity,
or manufacturability. Consequently, contact PMI for the latest
available specifications and performance data.
Although every effort has been made to ensure accuracy of the
information contained in this data book, PMI assumes no
responsibility for inadvertent errors.
PMI assumes no responsibility for the use of any circuits
described herein and makes no representation that they are free
of patent infringement.
The products in this catalog are manufactured under one or
more of the following patents: 4,055,773; 4,056,740; 4,068,254;
4,088,905; 4,092,639; 4,109,215; 4,118,699; 4,131,884; 4,138,671;
4,142,117; 4,168,528; 4,210,830; 4,228,367; 4,260,911; 4,272,656;
4,285,051; 4,333,047; 4,340,851; 4,374,335; 4,449,067; 4,471,321;
4,503,381; 4,538,115; 4,542,349.

Precision Monolithics Inc.
Life Support and Nuclear Facility Applications Policy
As a general policy, Precision Monolithics Inc. (PMI) does not
recommend the use of any of its products In (a) life support
applications where failure or malfunction of the PMI product
can be reasonably expected to cause failure of the life support
device or to significantly affect Its safety or effectiveness, or (b)
any nuclear facility applications. PMI Will not knowingly sell
its products for use in such applicatIOns unless it receives In
writing assurances satisfactory to PMI that (a) the risks of
injUry or damage have been minimized (b) the customer
assumes all such risks, and (c) the liability of PMI is adequately
protected under the circumstances.
Examples of devices considered to be life support deVices are
neonatal oxygen analyzers, nerve stimulators (whether used for
anesthesia, pain relief, or other purposes), autotransfusion
devices, blood pumps, defibrillators, arrhythmia detectors and
alarms, pacemakers, hemodialysis systems, peritoneal dialysis
systems, neonatal ventilator incubators, ventilators for both
adults and infants, anesthesia ventilators, and infuSIOn pumps,
as well as other devices designated as "cntical" by the FDA. The
above are examples only and are not intended to be conclusive
or exclusive of any other life support device.
Examples of nuclear facility applicatIOns are applications in
[a) a nuclear reactor, or [b) any device designed or used in
connection With the handling, processing, packaging,
preparation, utilizatIOn, fabricating, allOYing, storing, or
disposal of fissionable matenal or waste products thereof.

Table of Contents

1

Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers t
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18
~~--

--~-

---

~-

~

-~---~

---~---

-

---

TABLE OF
CONTENTS
PI I. t.: I .... HIII ]\101101 It

Section

h

1~'"

I Ill:

2

Ordering Information
2-2

Package Products Part
Numbering System

2-3

Dice Part
Numbering System

2-4

Dice Information

2-5

Reliability Information

2-6

Discontinued Device Types
Ordering Guide

Section

3

Section

5 continued

5-26

OP.04/OP.14
Dual Matched High-Performance
Operational Amplifiers

5-34

Instrumentation
Operational Amplifier
5-43

Introduction

3-3

Processing

Section

5-51

5-61

4
5-68

5

5-4

Definitions

5-6

Selection Guide

5-12

OP-Ol

5-87

OP-12
Precision Low-Input-Current
Operational Amplifier

5-91

OP.15/0P.16/0P.17
Precision JFET-Input
Operational Amplifiers

Inverting High-Speed
Operational Amplifier
5-18

OP·I0
Dual Matched Instrumentation
Operational Amplifier

Operational Amplifiers
Introduction

OP·09/0p·n
Quad Matched 741-1)rpe
Operational Amplifiers

Alphanumeric Direct
Replacement Guide

5-4

OP·08
Precision Low-Input-Current
Operational Amplifier

5-75
Section

OP·07
Ultra-Low Offset Voltage
Operational Amplifier

Industry Cross Reference
4-2

OP·06
High-Gain Instrumentation
Operational Amplifier

Product Assurance Program
3-3

OP·05

5-105

OP·02

OP-20
Micropower Operational Amplifier

General-Purpose
Operational Amplifier

5-111

OP·21
Low-Power Operational Amplifier

1-2

TABLE OF
CONTENTS
Pll(I<"IOlll\'lollolnhlt:"IIH:

Section

5 continued

Section

5 continued

5-117

OP·21

5-212

OP·220

Programmable Micropower
Operational Amplifier

Dual Micropower Operational Amplifier
5-220

5-128

OP·27
5-228

5-239

5-175

5-241

OP-41
5-246

OP-43

PM-I0SA/PM-20SA/PM-30SAI
PM-I0S/PM.20S/pM·30S1
5-252 PM·210SA/PM·21OS

Quad Low-Power Operational Amplifier

Low-Input-Current
Operational Amplifiers

OP-50
5-256

PM·155A/PM-355A/pM.1551
PM.156A/PM·356A/PM.1561
5-259 PM-157A/PM·357A/PM.157

OP·90
Low-Voltage Micropower
Operational Amplifier

5-199

Monolithic JFET-Input
Operational Amplifiers

OP-207

5-265

Dual Ultra-Low Vas Matched
Operational Amplifier

PM-725
Instrumentation Operational Amplifier

5-268
5-205

PM-146/PM.246
Programmable Quad
Operational Amplifiers

OP·77
Next Generation OP-07 (Ultra-Low
Offset Voltage Operational Amplifier)

5-197

OP·421

Low-Bias-Current JFET
Operational Amplifier

High-Output-Current
Operational Amplifier
5-191

OP-420
Quad Micropower Operational Amplifier

Low-Bias-Current, Fast JFET
Operational Amplifier
5-180

OP-400
Quad Low-Offset Low-Power
Operational Amplifier

OP-37
Low-Noise Precision High-Speed
Operational Amplifier

5-164

OP·227
Dual Low-Noise Low-Offset
Instrumentation Operational Amplifier

OP·32
High-Speed Programmable Micropower
Operational Amplifier

5-152

OP·221
Dual Low-Power Operational Amplifier

Low-Noise Precision
Operational Amplifier
5-140

~

0

OP·215

PM-741
Compensated Operational Amplifier

Dual Precision JFET-Input
Operational Amplifier
1·3

~--

------

~----

----

-

-

------~--

~--~--

~----~-

U

~

0
~

~

TABLE OF
CONTENTS
Plll:I..,IOIl 1\101l011thH: ...

Section

5-270

IIH:

S continued

Section

PM·747
Dual Compensated
Operational Amplifier

5-273

5-279

7-3

Introduction

7-4

BUF·03

J~851O/10104

High-Speed Voltage Follower/Buffer

JAN Single Low-Input-Current
Operational Amplifier

5-276

7

Voltage Followers/Buffers

Section

8

J~8510/10106

Voltage Comparators

JAN Dual Low-Input-Current
Operational Amplifier

8-3

Introduction

8-3

Definitions

JM3851O/11004
JAN Quad 741-1}rpe
Operational Amplifier

8-6

J~851O/11401/11402/11403/11404/

8-14

CMP·Ol
Fast Precision Comparator

5-282 11405/11406
JAN JFET-Input Operational Amplifiers

CMP·02
Low-Input-Current
Precision Comparator

5-292 JM38510/13501/13502
Ultra-Low Offset Voltage
Operational Amplifiers

8-22

CMP·04
Quad Low-Power Precision Comparator

8-30

5-295 JM38510/13503
Low-Noise Precision
Operational Amplifier

CMP·05
High-Speed Precision Comparator

8-37

CMP·07
Very High-Speed Comparator

Section'

8-40

Instrumentation Amplifiers
6-3

Introduction

6-3

Definitions

6-5

8-48

PM.111/PM.211
Precision Voltage Comparators

AMP· 01

8- 54

Low-Noise Precision
Instrumentation Amplifier

PM.119/PM.2.19/PM.319
Dual Comparators

8-56
6-27

CMP.404
Quad Low-Power Precision Comparator

AMP·05

PM.139/PM.139A/PM.339A
Quad Low-Power Voltage Comparators

Fast-Settling JFET
Instrumentation Amplifier

1-4

TABLE OF
CONTENTS
PrecIsion MOllo]ithics Inc.
C/)

Section

9

Section

11

~

Matched Transistor Arrays

Digital-to-Analog Converters

9-3

Introduction

11-4

Introduction

U

9-3

Definitions

11-4

Definitions

0

9-4

Parameter Comparison Tables

11-8

Selection Guide

9-5

MAT-Ol
Matched Monolithic Dual Transistor

11-10

DAC-Ol
6-Bit Voltage-Output D/A Converter

~

MAT-02

11-14

DAC-02/DAC-03/DAC-OS
lO-Bit-Plus-Sign Voltage-Output
D/A Converters

Z

0

9-11

Low-Noise, Matched Dual
Monolithic Transistor
9-23

MAT-04

11-19

Section

10

11-23

DAC-08
8-Bit High-Speed Multiplying
D/A Converter

11-34

DAC-lO
lO-Bit High-Speed Multiplying
D/A Converter

11-42

DAC-20
2-Digit BCD High-Speed Multiplying
D/A Converter

Voltage References
10-3

Introduction

10-3

Definitions

10-4

REF-Ol
+10V Precision Voltage Reference

10-11

REF-02

+5V Precision Voltage
Reference/Temperature Transducer
10-19

REF-03
+2.5V Precision Voltage Reference

10-21

REF-OS
+5V Precision Voltage Reference

10-22

DAC-06
Two's-Complement lO-Bit
Voltage-Output D/A Converter

Matched Monolithic Quad Transistor

11-50

DAC-86
COMDAC® Companding
D/A Converter

11-58

DAC-88
COMDAC® Companding
D/A Converter

REF-lO

11-67

+ 10V Precision Voltage Reference

DAC-89
COMDAC® Companding
D/A Converter

1-5

f.l-<

tj

TABLE OF
CONTENTS
Precision Mo])olithics Inc.

Section

11

11-76

DAC·I00

Section

contmued

11-149

10-Bit Current-Output D/A Converter
11-84

DAC·208
DAC·210

11-159 PM·7528
Dual 8-Bit Buffered Multiplying CMOS
D/A Converter

DAC·312
12-Bit High-Speed Multiplying
D/A Converter

11-175

11-104 DAC·888
BYTEDAC® 8-Bit High-Speed
"Microprocessor Compatible"
Multiplying D/A Converter
11-116

11-185

PM·7541
CMOS 12-Bit Monolithic Multiplying
D/A Converter

DAC·1508A/1408A

11-195 PM·7542
12-Bit Multiplying CMOS
D/A Converter

11-122 DAC·8012
CMOS 12-Bit Multiplying D/A
Converter "With Memory"

11-196 PM·7543
12-Bit Serial-Input Multiplying CMOS
D/A Converter

DAC·8212
Dual 12-Bit Buffered Multiplying
CMOS D/A Converter

11-197

11-136 DAC·8408
Quad 8-Bit Multiplying CMOS D/A
Converter With Memory

PM.7545/PM.7645
12-Bit Buffered Multiplying CMOS
D/A Converters

11-208 PM·7548
12-Bit Multiplying CMOS
D/A Converter

11-140 PM·562
12-Bit Multiplying Current-Output
D/A Converter
11-148

PM·7533
CMOS Low Cost lO-Bit Multiplying
D/A Converter

8-Bit Multiplying D/A Converters

11-131

PM·7226

11-150 PM·7524
CMOS 8-Bit Buffered Multiplying
D/A Converter

11-Bit Voltage-Output D/A Converter
11-92

continued

Quad 8-Bit CMOS D/A Converter
With Voltage Output

9-Bit Voltage-Output D/A Converter
11-88

11

11-209 JM38510/11301/11302
JAN 8-Bit Multiplying D/A Converters

PM·565A
Complete High-Speed 12-Bit
Monolithic D/A Converter

1-6

TABLE OF
CONTENTS
Precision Monolithics Inc.

Section

12

Analog-to-Digital Converters
12-3

Introduction

12-3

Definitions

12-7

ADC-910
High-Speed "Microprocessor
Compatible" A/D Converter

12-19

ADC-8208
Microprocessor-Compatible 8-Bit
CMOS A/D Converter

Section

Analog Switches/Multiplexers
Introduction

13-3

Definitions

13-7

SW-Ol/SW-02
Quad SPST JFET Analog Switches

13-14

SW-05
Dual SPST JFET Analog Switch

13

13-69

MUX-88
8-Channel Analog Multiplexer For
PCM CODECs

13-74

DMX-88
8-Channel Analog De-Multiplexer

Section

14

contmued

Sample-and-Hold Amplifiers

13

13-3

Section

14-3

Introduction

14-3

Definitions

14-6

SMP-IO/SMP-ll
Low-Droop-Rate/Accurate Sampleand-Hold Amplifiers

14-15

SMP-81
Telecommunications Sample-andHold Amplifier

Section

15

Special Functions

SW-06
Quad SPST JFET Analog Switch

15-3

Introduction

15-3

Definitions

13-33

SW-201/SW -202
Quad SPST JFET Analog Switches

15-4

13-40

SW-7510/SW-7511
Quad SPST JFET Analog Switches

GAP-Ol
General-Purpose Analog Signal
Processing Subsystem

15-19

PKD-Ol
Monolithic Peak Detector

13-22

13-48

MUX-08/MUX-24
8-Channel/Dual 4-Channel JFET
Analog Multiplexers

13-59

MUX-16lMUX-28
16-Channel/Dual 8-Channel JFET
Analog Multiplexers

1-7

TABLE OF
CONTENTS
Precision Monolithics Inc.

Section

16

Section

Communications Products
16-3
16-4

Introduction

17-17
17-17
17-18
17-18

RPT-82/RPT-83

Package Information

17-19
17-19
17-19
17-19
17-20
17-20
17-20
17-20

Metal Cans
6-Lead TO-78 Metal Can
8-Lead TO-99 Metal Can
lO-Lead TO-IOO Metal Can

17-4
17-4
17-5
17-5
17-6
17-7
17-8

8 -Lead Ceramic DIP
14-Lead Ceramic DIP
16-Lead Ceramic DIP
18 -Lead Ceramic DIP
20-Lead Ceramic DIP
24- Lead Ceramic DIP
28 -Lead Ceramic DIP

Ceramic DIPs

17-13
17-13
17-14
17-14
17-15
17-16

lO-Lead Flatpack
lO-Lead Flatpack, Bottom-Brazed
14-Lead Flatpack
14-Lead Flatpack, Bottom-Brazed
16-Lead Flatpack
16-Lead Flatpack, Bottom-Brazed
24- Lead Flatpack
24-Lead Flatpack, Bottom-Brazed

Leadless Chip Carriers
17-21
17-22

20-Position Chip Carrier
28-Position Chip Carrier

Section

18

Sales Offices, Representatives,
and Distributors

Side-Brazed DIPs
17-9
17-9
17-10
17-10
17-11
17-12

10-Lead Cerpack
14-Lead Cerpack
16-Lead Cerpack
24-Lead Cerpack

Flatpacks

17

17-3
17-3
17-3

continued

Cerpacks

PCM Repeaters
Section

17

14-Lead Side-Brazed DIP
16-Lead Side-Brazed DIP
18-Lead Side-Brazed DIP
20-Lead Side-Brazed DIP
24-Lead Side-Brazed DIP
28-Lead Side-Brazed DIP

18-2

Sales Offices, Representatives
North America

18-4

Authorized Distributors
North America

18-6

Sales Offices, Representatives
International

Epoxy DIPs

18-7

Authorized Distributors
International

8-Lead Epoxy DIP
14-Lead Epoxy DIP
16-Lead Epoxy DIP
18 -Lead Epoxy DIP
20-Lead Epoxy DIP
24- Lead Epoxy DIP

1-8

Table of Contents 1
Ordering Infornlation 2

Product Assurance Program 3
Industry Cross Reference 41
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

ORDERING
INFORMATION
Precision MOl1olithics Inc.

PACKAGED PRODUCTS PART NUMBERING SYSTEM
, . . . . - - - - - - - - - Device Type Be Model Number

r--------

IiI

~-----

CAC-08 BI

_ _ __
Q /883

!

I

L . . ._ _ _ _

Burn-In Option
Electrical Grade
Package Suffix
MIL-STD-883, Class B, Revision C Option
PMI -55 0 C to +125 0 C devices are available with MIL-STD-883,
Class B, Revision C screening as standard products. To order an 883
part, simply add the designation 1883 to the part number. For example,
the DAC-08AQ, screened to the 883 requirements would be ordered as
a DAC-08AQ/883. Contact factory for 883 device specifications.
Package Type
S = Not used
T = 28 lead Ceramic DIP
TC = 28 position LCC'
U = Not used
o = Not used
V = 24 lead Ceramic DIP
P = Epoxy DIP
X = 18 lead Ceramic DIP
Q = 16 lead Ceramic DIP
V = 14 lead Ceramic DIP
R = 20 lead Ceramic DIP
Z = 8 lead Ceramic DIP
RC = 20 position LCC'
"Available on certain 883 products
Note: See more complete listing on page 17-2.

H

= 6 lead TO-78 Can

= 8 lead TO-99 Can
K = 10 lead TO-100 Can

J

L.....-_ _ _ _

1 -_ _ _ _ _ _

Select electrical grade from data sheet.
PMI offers all 00/70°C and -25°/+85°C devices with burn-in per method
1015.5 of MIL-STD-883, 160 hours at +125°C or 80 hours at +150°C at
PMI's option. Parts with this option are specified with the letters BI added between the model number and the electrical grade. For example, to
order DAC-08EQ with burn-in, the part number is DAC-08BIEQ.

' - - - - - - - - - - Device types are listed on next page. Select model number from
product listings.
MIL-M-38S10
PMI's factory is certified to produce JAN parts per MIL-M-38510. Consult factory for availability of specific slash sheet part numbers. At this
writing, devices with Part 1 or Part 2 approval include PM-155A,
PM-156A, PM-157A, PM-108A, PM-210BA, PM-4136, DAC-08, OP-07,
and OP-27. Other types are being qualified.
See Table of Contents for JAN data sheet listings.

2·2

ORDERING
INFORMATION
I'n:cisiou MOlloli;thics Inc.

DICE PART NUMBERING SYSTEM

r----------- Device Type Be Model Number
r-------- Electrical Grade
N = Top, G = Middle, GR = Lowest

r------- Temperature-Tested Ole Option (125
r---- Visual Criteria

0

C)

A = MIL-STD-883, Method 2010, Test Condition A.
B = MIL-STD-883, Method 2010, Test Condition B.

r--- Backing Suffix

__ 1

-I

C = Plain, CG = Gold Backed

CAe-OS N T B C

I .

All PMI dice are available with ellher plain backing or, for a 50% adder
plus a $250 lot charge, a 1-micron-thick eutectic-bonded gold backing.
Several PMI devices are available with min/max wafer probe limitsfordice
tested at +125 0 C. See specific data sheet.

' - - - - - - - - Data sheets give specific available grades for each PMI device.

1----------- Device 1\tpe
ADC
AMP
aUF
CMP
DAC
DMX
GAP
JAN
MAT
MUX
OP
PKD
PM
REF
RPT
SMP
SW

-

Analog-to-Digital Converter
Instrumentation Amplifier
Buffer (Voltage Follower)
Comparator
Digital-to-Analog Converter
De-Multiplexer
General-Purpose Analog Processor
MIL-M-38510 Slash Sheet
Matched Transistor
Multiplexer
Proprietary Operational Amplifier
Peak Detector
Second-Source, Industry Specifications
Voltage Reference
PCM Line Repeater
Sample-and-Hold Amplifier
Analog Switch

' - - - - - - - - - - - Select model number from product listings.
2-3

ORDERING
INFORMATION
Precision Monolithics Inc.

DICE INFORMATION

is separated into individual dice. Due to variations in assembly methods and normal yield loss,
PMI does not guarantee specifications after
packaging for standard dice. Sample assembly
and testing in standard PMI packages to specified LTPD's and min/max specifications are
available at extra cost. Consult factory for dice
lot qualification negotiation.

Triple Passivation

Triple Passivation is a three-step process which
provides superior reliability and protection for
all PMI integrated circuits. First, a specially
treated thermal silicon dioxide layer is grown.
This protects the junctions and also attracts any
residual ionic impurities to the top surface of the
oxide, where they are held fixed. Next, a layer of
silicon nitride is applied to prevent the entry of
any potential contamination or impurities. The
third step is the thick glass overcoat layer which
leaves only the bonding pads exposed. This
"glassivation" protects the die from damage during assembly and is especially important in minimizing yield loss during shipment and assembly
of dice for hybrid circuits.

Shipping

Protection during shipment is provided by a wafflepack carrier with anti-static shield and
cushioning strip. In addition, the wafflepack is
vacuum sealed in a polyethylene bag.

Military/Aerospace Applications
PMI devices are widely used in military and
aerospace programs. A partial listing includes:

Quality Assurance

PMI believes that quality and reliability must be
built into the product; no amount of testing can
replace these inherent properties. For this reason, devices are fabricated and processed with
many exclusive processes and controls added to
improve quality and reliability. The integrity of
aluminum metallization is confirmed by
sampling wafer lots using Scanning Electron
Microscope (SEM) examinations per Method
2018 specifications.

Military Aircraft

F-4
8-18
Sikorsky UH-60A
A-1O
Sikorsky SH-3
YC-15
P-3
S-3A
KC-10

Mechanical Information

Missile/Spacecraft

Aluminum metallization with a nominal thickness of 10,000 angstroms is standard for all
devices. Die thickness is 19 mils minimum to 21
mils maximum. Minimum bonding pad size is 4.0
mils X 4.0 mils for all devices.

Viking
(Mars Orbitor)
Aerosat
Harm Missile
DSCS-3
Sparrow Missile
Trident
TDRSS
TV SAT
Intelsat 5
Minuteman

Visual Inspection

All dice are 100% visually inspected to the applicable visual criteria per MIL-STD-883 Method
2010, Condition 8.
Electrical Testing

All dice are 100% tested to the 25° C DC wafer
test limits shown in this catalog before the wafer

2-4

F-111
F-15
8-52
F-18
Alpha Jet
E-3A
F-16
F-5
Tornado
Milstar
Voyager
(Jupiter/Saturn)
Stinger Missile
Standard Missile"
Tiros-N
Cruise Missile
Ariane
Roland
Eurosat
Space Shuttle

ORDERING
INFORMATION
Precision Monolithics Inc.

Electronic Systems
Omega
Tram
Aims (MK86)
F-16
(Ground Support)
Pave Spike
AWACS
Miscellaneous
RFP Model 35 #13
Project 4620
Heads-Up Display
DST 1860
Walleye
PMS
Aerial Surveillance
Camera

ICs to the unique in-house speCifications of
their customers. PMI recognizes your special
needs and welcomes the opportunities provided
by the military/aerospace industry. Hi-rei is a
cornerstone of PMl's business and we will continue to offer the extra processing that your
applications require.

A4KU
MK-48
8-52 Radar Mod
Pathfinder Radar
MK-46
Seaguard
Gepard

Radiation Resistance
As a leading supplier of precision linear ICs to
the military/aerospace industry, PMI is supportive of the system designer's needs for readily
available, standard components that are radiation resistant. A number of standard PMI linear
integrated circuits have characteristically demonstrated good resistance to radiation. These
devices have been subjected to radiation levels
necessary to perform effectively in military/aerospace radiation environments, and they are
now being used in a number of demanding military and space programs.

Cutty Sark
Compass Tie
System 27
ACM
VCS
Naval Submarine
Periscope

RELIABILITY INFORMATION
MIL-STD-883C
PM I standard "883" parts are man ufactu red to
be in full compliance with all MIL-STD-883
requirements. See Section 3 for more details.

Experiments to isolate the processing mechanisms that led to PMl's increased radiation
hardness characteristics have pointed heavily
toward the use of a silicon nitride passivation
layer. While we do not believe that this process
is the only radiation hardening advantage of PMI
devices, it does add a great deal to survivability.

Specials
At PMI, we have a proven track record for handling "customer specials". Many IC manufacturers shy away from processing precision linear

For more information request PMl's "Radiation
Resistance" brochure.

2-5

z

9

~

2
z
......

ORDERING
INFORMATION

JPMI)
Precision Monolithics Inc.

DISCONTINUED DEVICE
TYPES ORDERING GUIDE

Note 2. DAC-76 -55 0 CI+ 1250 C types may be ordered as
DAC-86 specials.
Note 3. DAC-89EX has idling currents on the outputs
requiring matched load resistors. DAC-89EX has
higher speed and accuracy and is an improved,
direct, pin-lor-pin replacement lor DAC-87 in most
designs.

Between 1978 and 1985 some device types,
individual grades, and package options were
discontinued. This guide is provided to help the
designer to select an appropriate alternative
device.
Alternative Device "TYpe
OP-07 connected as a
voltage follower.
BUF-02
OP-16 connected as a
voltage follower.
DAC-04
DAC-06 nearest grade.
DAC-76
DAC-86 nearest grade.
DAC-78
DAC-88EX.
DAC-87
DAC-89EX.
DAC-101 DAC-100 "03" nearest
grade.
DAC-206 DAC-01 nearest grade.
DAC-808 DAC-888 is the most
similar device.
OP-03
OP-04 with externallyconnected V+ pins.
OP-18
LM101 is the most
similar device.
OP-19
MC17418 is the most
similar device.
OP-24
OP-27GP, improved
replacement.
OP-34
OP-37GP, improved
replacement.
PM-1458 OP-14 nearest grade.
PM-1558 OP-14 nearest grade.
PM-4136 OP-09 nearest grade.
888-725 OP-06 or PM-725
nearest grade.
888-741
OP-02 nearest grade.
888-747 OP-04 or PM-747
nearest grade.
888-1458 OP-14 nearest grade.
888-1558 OP-14 nearest grade.
8W-03
8W-06 is a functional
replacement.
8W-04
8W-06 is a functional
replacement.

"TYpe
BUF-01

Note(s)

1
1,2
1
3

Package "TYpe

Affected
Device Type

"K" (TO-100)

PM-747.

"N" (Flatpack)

DAC-100N9.

"Y" (14-pin DIP) 8ee list below.

Remarks
Available
on 8pecials.
Available
on 8pecials.
Available
on 8pecials.

"Y" Package Option: Affected Device Types
CMP-01
OP-02
PM-725
CMP-02
OP-06
PM-741
OP-01
OP-220
The "Y" package option was discontinued for
some products. They are all available in other
package types including "J" (TO-99), "P" (Epoxy
DIP), or "Z" (Ceramic 8-pin Mini-DIP). 8ee
individual data sheets for available package
options for each device.
Discontinued-Electrical-Grade Ordering Guide
The following electrical grades were discontinued but are available with different
specifications in the same packages. 8ee
individual data sheets.
CMP-01 B
MAT-01 F
PM-255
PM-256
CMP-02B
OP-01 E
PM-257
DAC-05B
OP-01 F
PM-339
DAC-05F
OP-04GR
PM-355
DAC-20E
OP-08B
PM-356
DAC-88C
OP-08F
PM-357
DAC-89C
OP-09C
PM-2208
DAC-10001
OP-09G
PM-2308
DAC-10002
PM-239
REF-Om
DAC-100DD05 PM-239A
DAC-100DD06
Five volt output versions of DAC-02 and DAC-05,
"X2" suffix, were discontinued. They may be

ordered as DAC-03X2 specials.
8ixty volt breakdown versions of MAT -01 were
discontinued (MAT-01H and MAT-01FH).

Note 1. Direct, pin-lor-pin replacement. No design
changes required.
2-6

Table of Contents 1
Ordering Information 2

Industry Cross Reference 4
Operational Amplifiers 5

I

i

6:

I

Instrumentation Amplifiers

Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

PRODUCT
ASSURANCE
Precision Monolithics Inc.

3-3

Introduction

3-3

Processing

3-2

PRODUCT
ASSURANCE
Precision Monolithics Inc.

INTRODUCTION

nearest PMI Sales Office or the Literature
Department for the latest issue of the PMI Reliability Bulletin.

PMI has long been recognized as a High Quality/Reliability supplier of Commercial, Industrial
and Military/Aerospace Products. The PMI Product Assurance Department plays a vital role in
controlling processes to ensure the manufacture
of highly reliable, cost-effective product, and
to make certain that all pertinent customer
specifications and requirements are met.

The Reliability Department also performs failure
analyses as required.
Program Management - The primary responsibility of the Program Management Department
is to ensure that the MIL-M-38510 JAN Program
and other special customer program requirements
are met. This is accomplished by monitoring the
in-house procedures used to define each process step of a particular program. If necessary,
baselining documentation is written detailing
specific procedures and processing flows. A
Configuration Control System consisting of
maintenance of PMI standard baselining for
each device type, as well as notification to customers of major process and product changes,
are also responsibilities of this group.

ORGANIZATION
Product Assurance Department of PMI is composed of four functional departments: Process
Quality Control, Quality Assurance, Reliability,
and Program Management.

RESPONSIBILITIES
Process Control- The primary responsibility of
the Process Control Department is to establish
and maintain effective controls over process
integrity by monitoring manufacturing processes
and equipment operation; to provide real-time feedback of information concerning the status of
these controls; and to initiate statistically
valid techniques to further improve quality and
reliability levels. These concepts are used
extensively throughout all manufacturing
processes.

Contact the nearest PMI Sales Office or the Literature Department for a copy of the comprehensive PMI Product Assurance Manual.

QUALITY LEVELS
PMI processes to stringent quality standards.
Quality guarantees range from parts-per-million on Standard Product to imposed Quality
Levels dictated by customer specification on
custom orders.
Current information on Quality Levels is
available upon request; contact the nearest PMI
Sales Office or the Literature Department.

Quality Assurance (Standard and HI-Rei) - The
primary responsibility of the Quality Assurance
Department is to assure that the delivered product meets PMI or Customer Product Standards
of reliability and quality. Process monitors and
gate inspections are designed so that all devices
are properly tested and required sample tests are
performed prior to shipment. Inspection records
and reports concerning monitor and inspection
data keep all cognizant personnel fully informed
about the status of the quality level of products
going through final test operations.

PROCESSING
The cornerstone of the manufacturing of PMI
hermetic products is the strict adherence to all
requirements of MIL-STD-883, Level B, for our
"883" product line. All PMI hermetic products,
be they "full 883" or not, receive the benefit of
MIL-STD-883 processing through assembly.
The manufacture of plastic devices is inherently
different from hermetic in the area of assembly.
Automation of the assembly line has produced a
tightly process-controlled product that requires
few interim inspections from wafer fabrication to
pre-mold visual. Plastic product may also be
obtained with a burn-in (BI) option (see Section
2, Ordering Information for further details).

Reliability - The Reliability Department assures
a high and consistent reliability of PMI products.
The Reliability Department establishes, defines,
and maintains evaluation programs to determine process/product reliability. The Reliability
Department will issue periodic reports on the
results of all evaluation testing. Contact the

3-3

:§
~
C)

~
~
~

u~

~
~
......
Q

~
~

PRODUCT
ASSURANCE
DATA SHEET SPECIFICATIONS
PMI standard product is guaranteed to meet the
published limits under the test conditions shown
in the data sheet. Where practical, PM I performs
100% testing of the indicated parameters;
however, following accepted industry practice,
certain parameters may be guaranteed by
sample testing or by using design and/or
characterization data.
PMI provides separate data sheets for all "883"
products in strict conformance with MIL-STD883, Method 5005.8 and MIL-M-38510F,
Appendix B. Interim Electrical Test Parameters
(pre-burn-in), Final Electrical Test 'Parameters,
Group A Test Parameters, and guidelines used
for PDA calculations, are all detailed in tabular
form on the "883" data sheets.
It Is highly recommended that the "883" data sheet
be used as a baseline for new military or
aerospace Source Control Drawings. Consult
your sales representative to obtain these "883"
data sheets.

Upon completion of this internal re-qualification, PMI informs all customers who have
requested process change notification with a
complete description of the change, along with
applicable reliability or characterization data.
Upon request, PMI will assist customers in their
internal re-qualification effort.
STANDARD PROCESS FLOWS
(JAN 38510. Cia •• B. 883 Cia., B; PMllndu,trlaIiCommerclal)
010 Flbrlcallon

Incoming Inspection 01 all orltlcal raw materials used In fabrication, (I.e. wafers,
masks, dopant. chemicals).

Scanning electron microscope/x-ray dlsburslve system controls oxide profiles,
diffUSion profiles and metallization Integrity,
Ole sort - automated eleotrlcal teat at room or elevated temperature as required,

Ole preparation - wafer saw

Ole sort visual
ToA••• mbly,
H.rmetfa

Semiautomatic die attach

TESTING
Testing of electrical parameters is generally
performed using pulse testing techniques on
automated test equipment. Unless otherwise
specified, chip temperature remains close to the
ambient temperature.
PROCESS CHANGE NOTIFICATION
PMI reviews all process, product, and package
changes for possible impact on form, fit, or
function. All major changes are submitted for a
re-qualification, which may include electrical,
mechanical, and/or thermal characterization.
Where applicable, reliability re-qualification is
performed.

(eutectic)

PI.,II.
Automatic die attach (epoxy)

Semiautomatic bonding
(aluminum wire)

Automatic bonding (gold wire)

FlowS 1, 2, 3

F're-mold visual

Continued on facing page.

F'lsstlc molding
F'ost-mold cure
Temperature cycle
Electrical test (Flow 4)
Continued on faCing page,

PRODUCT
ASSURANCE
......ccision MOllolithics Inc.
Flow 1
MIL·SID·883
MEIHOD 8004 CLASS B/
JAN 3B810

Flow 2
MIL·SID·8B3
MEIHOD 8004 CLASS B
(Soo Nolo 1)

Flow 3
PMI SIANDARD
HERMEIIC DEVICES

Flow 4
PMI SIANDARD
PLASIIC DEVICES

Ellclrlcal 100t
+28'C (Optional)

Burn-In Test
Mulll·Optlon

Final Eleclrlcal lest
al+25'C
(See

ote3)

I

NOlES:
1. M IL·883 product Is a..omblod/lesled porlho roqulre·
mont. ot MIL·M·38510 and MIL·SID·883, Rov, C, Class
B, as well as te.ted !=Ief PMI electrical test programs.
PDA = 5'110 @ 25'C (Subgroup 1).
Note: In·hous.t.st .paclflcatlons are aVClllable upon

reque.t.

2. Standard prOduct Is available with burn-In options.
Multiple burn-In conditions are available.
3. Standard product II tested at 25°C with programs

guard banded tor all temperatures.
4. As required by dotallod 'po.llication.

QAPlanl
Clearance
country of Origin AssemblyCodes: United States = 5, Korea"" K. Philippines = D, Puerto Rico = P

3·5

QAPlanl
Clearance

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

ALPHANUMERIC DIRECT
REPLACEMENT GUIDE
Precision Monolithics Inc.

Device

PMI Replacement

Device

PMI Replacement

Device

PMI Replacement

AD10SAH
AD10SH
AD140S-7D
AD140S-8D
AD150S-8D

PM10SAJ
PM10SJ
DAC140SA7Q
DAC140SASQ
DAC150SASQ

AD7524KN
AD7524LN
AD7524SD
AD7524TD
AD7524UD

PM7524HP
PM7524GP
PM7524BQ
PM7524BQ
PM7524AQ

ADDACOSCD
ADDACOSD
ADDACOSED
ADDACOSHD
ADDAC100JD

DACOSCQ
DACOSQ
DACOSEQ
DACOSHQ
DAC100BCQ7

AD20SAH
AD20SH
AD30SAH
AD30SAN
AD30SH

PM20SAJ
PM20SJ
PM30SAJ
PM30SAP
PM30SJ

AD752SAQ
AD752SBQ
AD752SCQ
AD752SJN
AD752SKN

PM752SFR
PM752SFR
PM752SER
PM752SHP
PM752SHP

ADDAC100KD
ADDAC100LD
ADDAC100SD
ADG200AA
ADG200AP

DAC100ABQ7
DAC100AAQ7
DAC100BCQ5
SW05BK
SW05BY

AD562AD
AD562KD
AD562SD
AD7226KN
AD7226TD

PM562FV
PM562HV
PM562AV
PM7226GP
PM7226AR

AD752SLN
AD752SSD
AD752STD
AD752SUD
AD7533AD

PM752SGP
PM752SBR
PM752SBR
PM752SAR
PM7533FQ

ADG200BA
ADG200BP
ADG200CJ
ADG201AP
ADG201BP

SW05FK
SW05FY
SW05GP
SW201BQ
SW201FQ

;,D7506JD
AD7506JN
AD7506KD
AD7506KN
AD7506SD

MUX16ET
MUX16ET
MUX16ET
MUX16ET
MUX16BT

AD7533BD
AD7533CD
AD7533JN
AD7533KN
AD7533LN

PM7533FQ
PM7533EQ
PM7533HP
PM7533HP
PM7533GP

ADG201CJ
ADOP07AH
ADOP07CH
ADOP07CN
ADOP07DH

SW201GP
OP07AJ
OP07CJ
OP07CP
OP07DJ

AD7506TD
AD7507JD
AD7507JN
AD7507KD
AD7507KN

MUX16BT
MUX2SET
MUX2SET
MUX2SET
MUX2SET

AD7533SD
AD7533TD
AD7533UD
AD7541AD
AD7541BD

PM7533BQ
PM7533BQ
PM7533AQ
PM7541FX
PM7541EX

ADOP07DN
ADOP07EH
ADOP07EN
ADOP07H
Am140SL6

OP07DP
OP07EJ
OP07EP
OP07J
DAC140SA6Q

AD7507SD
AD7507TD
AD7510DIJD
AD7510DIKD
AD7510DISD

MUX2SBT
MUX2SBT
SW7510FQ
SW7510FQ
SW7510BQ

AD7541JN
AD7541KN
AD7541SD
AD7541TD
AD7542BD

PM7541HP
PM7541GP
PM7541BX
PM7541AX
PM7542FR

Am140SL7
Am140SLS
Am140SN6
Am140SN7
Am140SNS

DAC140SA7Q
DAC140SASQ
DAC140SA6P
DAC140SA7P
DAC140SASP

AD7511DIJD
AD7511DIKD
AD7511DISD
AD7511DITD
AD7520JD

SW7511FQ
SW7511FQ
SW7511BQ
SW7511BQ
PM7533FQ

AD7542GBD
AD7542GKN
AD7542GTD
AD7542KN
AD7542TD

PM7542ER
PM7542GP
PM7542AR
PM7542HP
PM7542BR

Am150SLS
Am6012DC
Am6012DM
Am6012PC
Am6070ADC

DAC150SASQ
DAC312FR
DAC312BR
DAC312FR
DACS6CX

AD7520JN
AD7520KD
AD7520KN
AD7520LD
AD7520LN

PM7533HP
PM7533FQ
PM7533HP
PM7533EQ
PM7533GP

AD7543BD
AD7543GBD
AD7543GKN
AD7543GTD
AD7543KN

PM7543FR
PM7543ER
PM7543GP
PM7543AR
PM7543HP

Am6070DC
Am6072DM
Am6S5DL
Am6S5DM
Am6S5HL

DACS6EX
DACSSEX
CMP07FQ
CMP07BQ
CMP07FK

AD7520SD
AD7520TD
AD7520UD
AD7521JD
AD7521JN

PM7533BQ
PM7533BQ
PM7533AQ
PM7541FX
PM7541HP

AD7543TD
AD7545CQ
AD7545GCQ
AD7545GLN
AD7545GUD

PM7543BR
PM7545FR
PM7545ER
PM7545GP
PM7545AR

Am6S5HM
CA10SAT
CA10ST
CA145SE
CA145SG

CMP07BK
PM10SAJ
PM10SJ
OP14CP
OP14EP

AD7521KD
AD7521LD
AD7521LN
AD7521SD
AD7521TD

PM7541FX
PM7541FX
PM7541HP
PM7541BX
PM7541BX

AD7545LN
AD7545TD
AD7545UD
AD754SAQ
AD754SBQ

PM7545HP
PM7545CR
PM7545BR
PM754SFR
PM754SER

CA145SS
CA145ST
CA155SS
CA155ST
CA20SAT

OP14EP
OP14C
OP14J
OP14J
PM20SAJ

AD7521UD
AD7524AD
AD7524BD
AD7524CD
AD7524JN

PM7541BX
PM7524FQ
PM7524FQ
PM7524EQ
PM7524HP

AD754SJN
AD754SKN
AD754SSD
AD754STD
ADDACOSAD

PM754SHP
PM754SGP
PM754SBR
PM754SAR
DACOSAQ

CA20ST
CA30SAT
CA30SE
CA30ST
CA339AD

PM20SJ
PM30SAJ
PM30SP
PM30SJ
PM 339AY

4·2

ALPHANUMERIC DIRECT
REPLACEMENT GUIDE
Precision Monolithics Inc.

Device

PMI Replacement

Device

PMI Replacement

Device

PMI Replacement

CA339AG
CA747CE
CA747CG
CA747CT
CA747T

CMP04FP
OP04CY
OP04CY
OP04CK
OP04K

DG202CJ
DG202CK
DG506AR
DG506BR
DG506CJ

SW202GP
SW202FQ
MUX16AT
MUX16ET
MUX16FT

LF13201N
LF13202D
LF13202N
LF13333D
LF13333N

SW201GP
SW202FQ
SW202GP
SW06FQ
SW06GP

DAC0800LCJ
DAC0800LCN
DAC0801LCJ
DAC0801LCN
DAC0802LCJ

DAC08EQ
DAC08EP
DAC08CQ
DAC08CP
DAC08HQ

DG507AR
DG507BR
DG507CJ
DG508AP
DG508BP

MUX28AT
MUX28ET
MUX28FT
MUX08AQ
MUX08EQ

LF13508D
LF13508N
LF13509D
LF13509N
LF155AH

MUX08EQ
MUX08EP
MUX24EQ
MUX24EQ
PM155AJ

DAC0802LCN
DAC0806LCJ
DACOB06LCN
DAC0807LCJ
DAC0807LCN

DAC08HP
DAC1408A6Q
DAC1408A6P
DAC140BA7Q
DAC1408A7P

DG508CJ
DG509AP
DG509BP
DG509CJ
HI-200-2

MUX08FP
MUX24AQ
MUX24EQ
MUX24FP
SW05BY

LF155AJ-8
LF155H
LF156AH
LF156AJ-8
LF156H

PM155AZ
PM155J
PM156AJ
PM156AZ
PM156J

DAC0808LCJ
DAC0808LCN
DAC0808LD
DAC08ADM
DAC08CDC

DAC140BA8Q
DAC1408A8P
DAC1508A8Q
DAC08AQ
DAC08CQ

HI-200-4
HI-200-5
HI-201-2
HI-201-4
HI-201-5

SW05FY
SW05GP
SW201BQ
SW201FQ
SW201GP

LF156J-8
LF157AH
LF157AJ-8
LF157H
LF351H

PM156Z
PM157AJ
PM157AZ
PM157J
OP15FJ

DAC08CN
DAC08DM
DACOBEDC
DAC08EF
DACOBEN

DAC08CQ
DAC08Q
DAC08EQ
DACOBEQ
DAC08EP

HI1-506-2
HI1-506-5
HI1-506A-2
HI1-506A-5
HI1-507-2

MUX16BT
MUX16FT
MUX16BT
MUX16ET
MUX28BT

LF353H
LF355AH
LF355AJ-8
LF355BJ
LF356AH

OP215FJ
PM355AJ
PM355AZ
PM355Z
PM356AJ

DAC08F
DAC08HC
DAC08HD
DAC08HF
DAC08HN

DAC08Q
DAC08HP
DAC08HQ
DAC08HQ
DAC08HP

HI1-507-5
HI1-507A-2
HI1-507A-5
HI1-562A-2
HI1-562A-5

MUX28ET
MUX28AT
MUX28ET
PM562AV
PM562AV

LF356AJ-8
LF357AH
LF357AJ-8
LF411ACH
LF411ACN

PM356AZ
PM357AJ
PM357AZ
OP15EJ
OP15EZ

DAC10BDM
DAC10CDM
DAC10FCC
DAC10GDC
DAC6012ADC

DAC10BX
DAC10CX
DAC10FX
DAC10GX
DAC312FR

HI1-562A-8
HI1-7541AD-4
H11-7541 BD-4
HI1-7541JD-5
HI1-7541KD-5

PM562HVl883
PM7541 FX
PM7541 EX
PM7541HP
PM7541GP

LF411AMH
LM108AD
LM108AH
LM108AJ-8
LM108D

OP15AJ
PM108AZ
PM108AJ
PM108AZ
PM108Z

DAC6012ADM
DAC6012DC
DAC6012DM
DG200AA
DG200AAA

DAC312BR
DAC312FR
DAC312BR
SW05BK
SW05BK

HI1-7541SD-2
HI1-7541TD-2
HI2-200-2
HI2-200-4
HI3-506-5

PM7541BX
PM7541AX
SW05BK
SW05FK
MUX16FT

LM108H
LM108J-8
LM139AD
LM139AF
LM139AJ

PM108J
PM108Z
PM139AY
PM139AY
PM139AY

DG200AAK
DG200ABA
DG200ABK
DG200ACJ
DG200AP

SW05BY
SW05FK
SW05FY
SW05GP
SW05BY

HI3-506A-5
HI3-507-5
HI3-507A-5
HI3-508A-5
HI3-509A-5

MUX16ET
MUX28FT
MUX28ET
MUXOBEP
MUX24EP

LM139D
LM139F
LM139J
LM1458AH
LM1458AJ-8

PM139Y
PM139Y
PM139Y
OP14EJ
OP14EZ

DG200BA
DG200BP
DG200CJ
DG201AAK
DG201ABK

SW05FK
SW05BY
SW05GP
SW201BQ
SW201FQ

HI4-508A-2
HI4-508A-5
HI4-509A-2
HI4-509A-5
LF11201D

MUX08BQ
MUX08EQ
MUX24BQ
MUX24FQ
SW201BQ

LM1458AN
LM1458H
LM1458J
LM1458N
LM1558AH

OP14EP
OP14CJ
OP14CZ
OP14DP
OP14AJ

DG201AP
DG201BP
DG201CJ
DG202AK
DG202BK

SW201BQ
SW201FQ
SW201GP
SW202BQ
SW202FQ

LF11202D
LF11333D
LF11508D
LF11509D
LF13201 D

SW202BQ
SW06BQ
MUX08AQ
MUX24AQ
SW201FQ

LM1558AJ
LM194H
LM208AD
LM208AH
LM208AN

OP14AZ
MAT02AH
PM20BAZ
PM208AJ
PM208AZ

~

U

~
~
~

~
(/)
(/)

~
U
~

~

::J
Cl

Z

>-+

4-3

---~~-~----

-

--------

-

--~------~---~-----

--

-~"-

--

------

-----

._----.

-

-- -

----

--.

------

ALPHANUMERIC DIRECT
REPLACEMENT GUIDE
Precision Monolithics Inc.

Device

PMI Replacement

Device

PMI Replacement

Device

PMI Replacement

LM208D
LM208H
LM208N
LM239AD
LM239D

PM208Z
PM208J
PM208Z
CMP04FY
CMP04FY

MC1504U10
MC1504U5
MC15088F
MC1508L8
MC1558G

REF01AZ
REF02AZ
DAC15088Q
DAC1508A8Q
OP14J

MP7506KD
MP7506TD
MP7507JD
MP7507JN
MP7507KD

MUX16ET
MUX16BT
MUX28ET
MUX28FT
MUX28ET

LM239F
LM239N
LM258P
LM2901F
LM2901J

CMP04FY
CMP04FP
OP221GZ
CMP04FY
CMP04BY

MC1558NG
MC1558NU
MC1558U
MC1741CG
MC1741CP1

OP14AJ
OP14AZ
OP14Z
OP02CJ
OP02CP

MP7507KN
MP7507SD
MP7507TD
MP7508DIJD
MP7508DIJN

MUX28FT
MUX28BT
MUX28BT
MUX08EQ
MUX08EP

LM2901N
LM308AD
LM308AH
LM308AJ-8
LM308AN

CMP04FP
PM308AZ
PM308AJ
PM308AZ
PM308AP

MC1741CU
MC1741G
MC1741NCG
MC1741NCP1
MC1741NG

OP02HJ
OP02J
OP02Z
OP02HZ
OP02J

MP7508DIKD
MP7508DISD
MP7524AD
MP7524BD
MP7524CD

MUX08EQ
MUX08AQ
PM7524FQ
PM7524FQ
PM7524EQ

LM308D
LM308H
LM308J-8
LM3302N
LM339AD

PM308Z
PM308J
PM308Z
CMP04FP
PM339AY

MC1741NU
MC1741SCG
MC1741SCP1
MC1741SG
MC1741U

OP02Z
OP01HJ
OP01CP
OP01J
OP02Z

MP7524JN
MP7524KN
MP7524LN
MP7524SD
MP7524TD

PM7524HP
PM7524HP
PM7524GP
PM7524BQ
PM7524BQ

LM339AF
LM339AJ
LM339AN
LM339D
LM339N

PM339AY
PM339AY
CMP04FP
PM339AY
CMP04FP

MC1747CL
MC1747CP2
MC1747G
MC1747L
MC3302N

OP04EY
OP04EY
OP04BK
OP04BY
CMP04FP

MP7524UD
MP7528BD
MP7528CD
MP7528KN
MP7528LN

PM7524AQ
PM7528FR
PM7528ER
PM7528HP
PM7528GP

LM358JG
LM394H
LM725AH
LM725CH
LM725CN

OP221HZ
MAT02EH
OP06AJ
OP06EJ
OP06GZ

MC35001AG
MC35001AU
MC35001BG
MC35001BU
MC4741CL

OP16AJ
OP16AZ
OP16GJ
OP16BZ
OP11GY

MP7528TD
MP7528UD
MP7533BD
MP7533CD
MP7533KN

PM7528BR
PM7528AR
PM7533FQ
PM7533EQ
PM7533HP

LM725H
LM747AH
LM747AJ
LM747CH
LM747CJ

OP06BJ
OP04K
OP04AY
OP04CK
OP04CY

MC4741CP
MC4741L
MC7506KN
MC7506SD
MP200DIAA

OP11GP
OP11CY
MUX16FT
MUX16BT
SW05BK

MP7533LN
MP7533TD
MP7533UD
MP7541TD
MP7621AD

PM7533GP
PM7533BQ
PM7533AQ
PM7541AX
PM7541FX

LM747CN
LM747EJ
LM747EN
LM747H
LM747J

OP04DY
OP04EY
OP04CY
OP04BK
OP04Y

MP200DIAP
MP200DIBA
MP200DIBP
MP201DIAP
MP201DIBP

SW05BY
SW05FK
SW05BY
SW201BQ
SW201FQ

MP7621BD
MP7621JN
MP7621KN
MP7621SD
MP7621TD

PM7541EX
PM7541HP
PM7541GP
PM7541BX
PM7541AX

MC14087F
MC14088F
MC1408F
MC1408L6
MC1408L7

DAC14087Q
DAC14088Q
DAC1408A7Q
DAC1408A6Q
DAC1408A7Q

MP201DICJ
MP4136CY
MP4136Y
MP5520AD
MP5520AZ

SW201GP
OP09FY
OP09BY
DAC01Y
DAC01AY

MP7623AD
MP7623BD
MP7623JN
MP7623KN
MP7623SD

PM7541FX
PM7541EX
PM7541HP
PM7541GP
PM7541BX

MC1408L8
MC1458CG
MC1458CP1
MC1458CU
MC1458G

DAC1408A8Q
OP14CJ
OP14CP
OP14CZ
OP14CJ

MP5520BD
MP5520CD
MP5520DD
MP5520FD
MP5520HD

DAC01BY
DAC01CY
DAC01DY
DAC01FY
DAC01HY

MP7623TD
MP7645CD
MP7645LN
MP7645UD
MPC4D

PM7541AX
PM7645ER
PM7645GP
PM7645AR
MUX24FQ

MC1458N
MC1458NP1
MC1458NU
MC1458P
MC1458P1

OP14CP
OP14EP
OP14EZ
OP14CP
OP14DP

MP562AD/BIN
MP562KD/BIN
MP562SD/BIN
MP7506JD
MP7506JN

PM562FV
PM562HV
PM562AV
MUX16ET
MUX16FT

MPC8S
MPOP01AJ
MPOP01AZ
MPOP01CJ
MPOP01CP

MUX08FQ
OP01AJ
OP01AZ
OP01CJ
OP01CP

4-4

ALPHANUMERIC DIRECT
REPLACEMENT GUIDE
Precision M0]101ithics Inc.

Device

PMI Replacement

Device

PMI Replacement

Device

PMI Replacement

MPOP01CZ
MPOP01GJ
MPOP01GZ
MPOP01HJ
MPOP01HP

OP01CZ
OP01GJ
OP01GZ
OP01HJ
OP01HP

MPOP12FJ
MPOP12FZ
MPOP12GJ
MPOP12GZ
MPOP27AJ

OP12FJ
OP12FZ
OP12GJ
OP12GZ
OP27AJ

MPREF02HP
MPREF02HZ
MPREF02J
MPREF02Z
NE5007E

REF02HP
REF02HZ
REF02J
REF02Z
DAC08CQ

MPOP01HZ
MPOP02AJ
MPOP02AZ
MPOP02BJ
MPOP02BZ

OP01HZ
OP02AJ
OP02AZ
OP02BJ
OP02BZ

MPOP27AZ
MPOP27BJ
MPOP27BZ
MPOP27CJ
MPOP27CZ

OP27AZ
OP27BJ
OP27BZ
OP27CJ
OP27CZ

NE5008F
NE5009F
NE532H
NE532N
OP07AH

DAC08EQ
DAC08HQ
OP221CJ
OP221GZ
OP07AJ

MPOP02CJ
MPOP02CP
MPOP02CZ
MPOP02DJ
MPOP02DP

OP02CJ
OP02CP
OP02CZ
OP02DJ
OP02DP

MPOP27EJ
MPOP27EP
MPOP27EZ
MPOP27FP
MPOP27FZ

OP27EJ
OP27EP
OP27EZ
OP27FP
OP27FZ

OP07AH/883
OP07AJ8
OP07 AJ8/883
OP07AT
OP07CH

OP07AJ/883
OP07AZ
OP07AZ/883
OP07AJ
OP07CJ

MPOP02DZ
MPOP02EJ
MPOP02EP
MPOP02EZ
MPOP02J

OP02DZ
OP02EJ
OP02EP
OP02EZ
OP02J

MPOP27GJ
MPOP27GP
MPOP27GZ
MPOP37AJ
MPOP37AZ

OP27GJ
OP27GP
OP27GZ
OP37AJ
OP37AZ

OP07CJ8
OP07CJG
OP07CN8
OP07CNB
OP07CT

OP07CZ
OP07CZ
OP07CP
OP07CP
OP07CJ

MPOP05AJ
MPOP05AZ
MPOP05BJ
MPOP05CJ
MPOP05CZ

OP05AJ
OP05AZ
OP05BJ
OP05CJ
OP05CZ

MPOP37BJ
MPOP37BZ
MPOP37CJ
MPOP37CZ
MPOP37EJ

OP37BJ
OP37BZ
OP37CJ
OP37CZ
OP37EJ

OP07DE
OP07DH
OP07DN8
OP07DNB
OP07DT

OP07Z
OP07DJ
OP07DP
OP07DP
OP07DJ

MPOP05DJ
MPOP05EJ
MPOP05EZ
MPOP05J
MPOP05Z

OP05DJ
OP05EJ
OP05EZ
OP05J
OP05Z

MPOP37EP
MPOP37EZ
MPOP37FJ
MPOP37FP
MPOP37GJ

OP37EP
OP37EZ
OP37FJ
OP37FP
OP37GJ

OP07EDE
OP07EH
OP07EJ8
OP07EJG
OP07EN8

OP07EZ
OP07EJ
OP07EZ
OP07EZ
OP07EP

MPOP07AJ
MPOP07AZ
MPOP07CJ
MPOP07CP
MPOP07CZ

OP07AJ
OP07AZ
OP07CJ
OP07CP
OP07CZ

MPOP37GP
MPOP37GZ
MPREF01AJ
MPREF01AZ
MPREF01CJ

OP37GP
OP37GZ
REF01AJ
REF01AZ
REF01CJ

OP07ET
OP07H
OP07H/883
OP07J8
OP07J8/883

OP07EJ
OP07J
OP07J/883
OP07Z
OP07Z/883

MPOP07DJ
MPOP07DP
MPOP07DZ
MPOP07EJ
MPOP07EP

OP07DJ
OP07DP
OP07DZ
OP07EJ
OP07EP

MPREF01CP
MPREF01CZ
MPREF01EJ
MPREF01EZ
MPREF01HJ

REF01CP
REF01CZ
REF01EJ
REF01EZ
REF01HJ

OP07T
OP12AJG
OP12BJG
OP12CJG
OP12EJG

OP07J
OP12AZ
OP12BZ
OP12CZ
OP12EZ

MPOP07EZ
MPOP07J
MPOP10CY
MPOP10EY
MPOP11AY

OP07EZ
OP07J
OP10CY
OP10EY
OP11AY

MPREF01HP
MPREF01HZ
MPREF01J
MPREF01Z
MPREF02AJ

REF01HP
REF01HZ
REF01J
REF01Z
REF02AJ

OP12FJG
OP12GJG
OP15AH
OP15BH
OP15CH

OP12FZ
OP12GZ
OP15AJ
OP15BJ
OP15CJ

MPOP11BY
MPOP11EY
MPOP12AJ
MPOP12AZ
MPOP12BJ

OP11BY
OP11EY
OP12AJ
OP12AZ
OP12BJ

MPREF02AZ
MPREF02CJ
MPREF02CP
MPREF02CZ
MPREF02DJ

REF02AZ
REF02CJ
REF02CP
REF02CZ
REF02DJ

OP15EH
OP15FH
OP15GH
OP16AH
OP16BH

OP15EJ
OP15FJ
OP15GJ
OP16AJ
OP16BJ

MPOP12BZ
MPOP12CJ
MPOP12CZ
MPOP12EJ
MPOP12EZ

OP12BZ
OP12CJ
OP12CZ
OP12EJ
OP12EZ

MPREF02DP
MPREF02DZ
MPREF02EJ
MPREF02EZ
MPREF02HJ

REF02DP
REF02DZ
REF02EJ
REF02EZ
REF02HJ

OP16CH
OP16EH
OP16FH
OP16GH
OP227AJ

OP16CJ
OP16EJ
OP16FJ
OP16GJ
OP227AY

4-5

--

----~-

------

--

-----~--~------

----

--

----

~---

.--

~

U

~
~

~
~

~
t/)
t/)

~
U
~
!;;
:::J
Cl

Z

ALPHANUMERIC DIRECT
REPLACEMENT GUIDE
Precisioll Monolithics Inc.

Device

PMI Replacement

Device

PMI Replacement

Device

PMI Replacement

OP227CJ
OP227EJ
OP227FJ
OP227GJ
OP27AT

OP227CY
OP227EY
OP227FY
OP227GY
OP27AJ

REF-01CDE
REF-01CT
REF-01DDE
REF-01DE
REF-01EDE

REF01CZ
REF01 CJ
REF01HZ
REF01Z
REF01EZ

TSC9496CJ
TSC9496EE
XRC277
.uA0801CDC
.uA0801CPC

REF01HP
REF01 EJ
RPT82FQ
DAC08EQ
DAC08CP

OP27BDE
OP27BT
OP27BT/883
OP27CDE
OP27CH

OP27BZ
OP27BJ
OP27BJ/883
OP27CZ
OP27CJ

REF-01ET
REF-01HT
REF-01T
REF-02ADE
REF-02AT

REF01EJ
REF01HJ
REF01J
REF02AZ
REF02AJ

.uA0801DM
.uA0801EDC
.uA0802ADC
.uA0802APC
.uA0802BDC

DAC08Q
DAC08EQ
DAC1408A8Q
DAC1408A8P
DAC1408A7Q

OP27CH/883
OP27CJ8
OP27CJ8/883
OP27CT
OP27EDE

OP27CJ/883
OP27CZ
OP27CZ/883
OP27CJ
OP27EZ

REF-02CDE
REF-02CT
REF-02DDE
REF-02DE
REF-02EDE

REF02CZ
REF02CJ
REF02HZ
REF02Z
REF02EZ

.uA0802BPC
.uA0802CDC
.uA0802CPC
.uA0802DM
.uA0802DMQB

DAC1408A7P
DAC1408A6Q
DAC1408A6P
DAC1508A8Q
DAC1508A8Q/883

OP27EH
OP27EJ8
OP27EN8
OP27ET
OP27FT

OP27EJ
OP27EZ
OP27EP
OP27EJ
OP27FJ

REF-02ET
REF-02HT
REF-02T
REF01AH
REF01CH

REF02EJ
REF02HJ
REF02J
REF01AJ
REF01CJ

.uA0802EPC
.uA108AH
.uA1 08H
.uA1 39ADM
.uA139DM

DAC08EP
PM108AJ
PM108J
PM139AY
PM139Y

OP27GDE
OP27GH
OP27GJ8
OP27GN8
OP27GT

OP27GZ
OP27GJ
OP27GZ
OP27GP
OP27GJ

REF01CN8
REF01EH
REF01H
REF01HH
REF01HN8

REF01CP
REF01 EJ
REF01J
REF01HJ
REF01HP

.uA208AH
.uA208H
.uA308AH
.uA308H
.uA3303PC

PM208AJ
PM208J
PM308AJ
PM308J
OP11FP

OP37ADE
OP37AH
OP37AH/883
OP37AJ8
OP37AJ8/883

OP37AZ
OP37AJ
OP37AJ/883
OP37AZ
OP37AZ/883

REF02AH
REF02CH
REF02CN8
REF02DH
REF02DN8

REF02AJ
REF02CJ
REF02CP
REF02DJ
REF02DP

.uA339ADC
.uA714EHC
.uA714HC
.uA714HM
.uA714LHC

PM339AY
OP07EJ
OP07CJ
OP07AJ
OP07DJ

OP37AT
OP37BDE
OP37BT
OP37BT/883
OP37CDE

OP37AJ
OP37BZ
OP37BJ
OP37BJ/883
OP37CZ

REF02EH
REF02H
REF02HH
REF02HN8
RM725T

REF02EJ
REF02J
REF02HJ
REF02HP
PM725J

.uA725HC
.uA725HM
.uA725TC
.uA741CJG
.uPC1251 C

PM725CJ
PM725J
PM725CP
PM741CZ
OP220GZ

OP37CH
OP37CH/883
OP37CJ8
OP37CJ8/883
OP37CT

OP37CJ
OP37CJ/883
OP37CZ
OP37CZ/883
OP37CJ

RM741T
RM747DC
SA1458N
SA741
SE532H

PM741J
OP04Y
OP14BZ
OP02BZ
OP221 BJ

.uPC1251D
.uPC151C
.uPC151 D
.uPC154D
.uPC156D

OP220FZ
OP02BZ
OP02CZ
OP05CZ
PM208Z

OP37EDE
OP37EH
OP37EJ8
OP37EN8
OP37ET

OP37EZ
OP37EJ
OP37EZ
OP37EJ
OP37EJ

SE5534AFE
SE5534FE
SSS1408A-6Q
SSS1408A-7Q
SSS1408A-8Q

OP27BZ
OP27CZ
DAC1408A6Q
DAC1408A7Q
DAC1408A7Q

.uPC251C
.uPC251 D
.uPC339C
.uPC358C
.uPC4581 C

OP14EP
OP14EZ
CMP04FP
OP221HZ
OP15GZ

OP37GDE
OP37GH
OP37GJ8
OP37GN8
OP37GT

OP37GZ
OP37GJ
OP37GZ
OP37GP
OP37GJ

SSS1508A-8Q
TSC7541BD
TSC7541JN
TSC7541KN
TSC7541SD

DAC1508A8Q
PM7541EX
PM7541HP
PM7541GP
PM7541BX

.uPC603D
.uPC610D
.uPC624C
.uPC624D
.uPC648D

DAC01CY
DAC02CCX1
DAC08EP
DAC08EQ
DAC312FR

RC1458NB
RC3302DB
RC725T
REF-01ADE
REF-01AT

OP14CP
CMP04FP
PM725CJ
REF01AZ
REF01AJ

TSC7541TD
TSC9495CE
TSC9495CJ
TSC9495EE
TSC9496CE

PM7541AX
REF02HJ
REF02HP
REF02EJ
REF01HJ

.uPC801C

OP15FZ

4-6

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

OPERATIONAL
AMPLIFIERS
Precision Monolithics Inc.

5-4

Introduction

5-4

Definitions

5-6

Selection Guide

5-12

OP-Ol
Inverting High-Speed
Operational Amplifier

5-18

5-91

5-105
5-111

5-43

5-51

OP-02
5-117

OP-04/0P-14
5-128

op-os
Instrumentation
Operational Amplifier

5-140

5-75

5-87

OP-27

OP-06
High-Gain Instrumentation
Operational Amplifier

5-152

OP-32
High-Speed Programmable Micropower
Operational Amplifier

OP-37
Low-Noise Precision High-Speed
Operational Amplifier

OP-07
5-164

OP-41
Low-Bias-Current JFET
Operational Amplifier

OP-08
Precision Low-Input-Current
Operational Amplifier

5-68

OP-22
Programmable Micropower
Operational Amplifier
Low-Noise Precision
Operational Amplifier

Ultra-Low Offset Voltage
Operational Amplifier
5-61

OP-21
Low-Power Operational Amplifier

Dual Matched High-Performance
Operational Amplifiers
5-34

OP-20
Micropower Operational Amplifier

General-Purpose
Operational Amplifier
5-26

OP-lS/0P-16/0P-17
Precision JFET-Input
Operational Amplifiers

5-175

OP-43
Low-Bias-Current, Fast JFET
Operational Amplifier

Quad Matched 741-Type
Operational Amplifiers

5-180

OP-I0
Dual Matched Instrumentation
Operational Amplifier

OP-SO
High-Output-Current
Operational Amplifier

5-191

OP-12
Precision Low-Input-Current
Operational Amplifier

OP-77
Next Generation OP-07 (Ultra-Low
Offset Voltage Operational Amplifier)

5-197

OP-09/0P-ll

OP-90
Low-Voltage Micropower
Operational Amplifier

5-2

OPERATIONAL
AMPLIFIERS
Pn'CI;"IOIl MOIlolnh,,:s Inc

5-199

OP-207

5-265
5-268

5-205

OP-215
5-270

OP-220
5-273

OP-221
5-276

OP-227

5-279

OP-400

JM38510/11401/11402/11403/11404/

OP-420

5-282 11405/11406
JAN JFET-Input Operational Amplifiers

Quad Micropower Operational Amplifier
5-246

OP-421

5-292 JM38510/13501/13502
Ultra-Low Offset Voltage
Operational Amplifiers

Quad Low-Power Operational Amplifier

PM-I08A/PM-208A/PM-308A/
PM-I08/PM-208/PM-308/
5-252 PM-2108A/PM-2108

5-295

Low-Input-Current
Operational Amplifiers
5-256

IM38510/11004
JAN Quad 741-Type
Operational Amplifier

Quad Low-Offset Low-Power
Operational Amplifier
5-241

JM38510/10106
JAN Dual Low-Input-Current
Operational Amplifier

Dual Low-Noise Low-Offset
Instrumentation Operational Amplifier
5-239

JM38510/10104
JAN Single Low-Input-Current
Operational Amplifier

Dual Low-Power Operational Amplifier
5-228

PM-747
Dual Compensated
Operational Amplifier

Dual Micropower Operational Amplifier
5-220

PM-741
Compensated Operational Amplifier

Dual Precision JFET-Input
Operational Amplifier
5-212

PM-725
Instrumentation Operational Amplifier

Dual Ultra-Low Vos Matched
Operational Amplifier

IM3851O/13503
Low-Noise Precision
Operational Amplifier

PM-1461PM-246
Programmable Quad
Operational Amplifiers

PM-155A/PM-355A/PM-1551
PM-156A/PM-356A/PM-1561
5-259 PM-157A/PM-357A/PM-157
Monolithic JFET-Input
Operational Amplifiers

5-3

~
!-I-l
......
~

2
~

~

0......

~C3

OPERATIONAL
AMPLIFIERS
Precision MOllolith ics Inc.

INTRODUCTION

High Accuracy
OP-05, OP-06,
OP-07, OP-77
OP-27, OP-37, OP-50
OP-227, OP-207, OP-10
OP-08, OP-12
OP-50

Precision Monolithics pioneered in the development of low-offset, high-gain operational
amplifiers for use in precision applications. A
proprietary linear bipolar process with nitride
passivation was developed to achieve low noise,
enhanced long-term reliability, and improved resistance to radiation effects. PMI operational
amplifier processing capability includes JFET
and super-beta devices as well as standard NPN
and PNP devices. A zener-zap trimming technique was designed to reduce input offset voltage at the wafer testing stage. Offset trimming
is performed in discrete steps by applying highcurrent pulses through automatically-selected
zener diodes. High-current pulsing shorts the
zener which is parallel-connected to a trim resistor. Zener-zap trimming provides a very reliable
and stable reduction of input offset voltage. PMI
has developed many innovative operationalamplifier circuit designs based on their lownoise, low-drift processing in combination with
zener-zap offset trimming.
The table below summarizes the PMI families
of operational amplifiers. All feature low input
offset voltage, low drift, and high open-loop
gain.

Low Vos, High Gain
Low Vos, Low Noise
Dual
Low IB, Low Power
High Output Current

Low Power, Low Input-Bias-Current
OP-21
Single
OP-221
Dual
OP-41
JFET Input
PM-108/208/308,

PM-2108 (Dual)
Micropower
OP-20
OP-22/32

OP-220
OP-420,OP-421

Low IB, LM-108 Type
Singles, Low ISY
Programmable
Dual
Quad

DEFINITIONS
Average Bias Current Drift (TCla) - The ratio
of change in input bias current to a change in
temperature.
Average Offset Current Drift (TClos) - The ratio
of change in input offset current to a change in
temperature.
Average Offset Voltage Drift (TCVos) - The
ratio of change in input offset voltage to a
change in temperature.

Operational Amplifiers
General Purpose, Bipolar Input
OP-02
Single
OP-04, OP-14
Dual
Quad
OP-09, OP-11
OP-01
High Speed, Inverting

Average Offset Voltage Drift With External
Trimming (TCVOSN) - The ratio of the change
in input offset voltage to a change in temperature with the input offset voltage trimmed
to zero at room temperature.

General Purpose, JFET Input
PM-155, PM-156,
PM-157
Standard JFET Input
38510 Versions
MIL-Grade, 38510

Common-Mode Input Resistance (RlnCM)
- The ratio of input voltage range to the change
in input bias current over this range.

Precision, High-Speed Improved JFET Input
OP-15, OP-16, OP-17,
OP-43, OP-215 (Dual)

Common-Mode Rejection Ratio (CMRR)
- The ratio of the common-mode voltage range
(CMVR) to the peak-to-peak change in equivalent input offset voltage (CME) over this
range. CMRR is specified for a specific CMVR.
CMRR = 20 10glO (CMVR/CME)

5·4

OPERATIONAL
AMPLIFIERS
Precision Monolithics Inc.

Gain-Bandwidth Product (GBW) - The
frequency at which the open-loop gain equals
unity.
Input Bias Current (Ia) - The average of the
currents into the two input terminals when the
output is at zero volts with no load. IB is
measured at VCM = O.

Power Dissipation (Pd) - The total power
dissipated in the amplifier with the output at
zero volts with no load.

Input Noise Current (I np _p) - The peak-to-peak
noise current within a specified frequency band.

Slew Rate (SR) - The ratio of a change in
output voltage to the minimum time required to
effect this change under large-signal drive conditions. Slew rate may be specified separately
for positive and negative-going changes.

Power Supply Rejection Ratio (PSRR) - The
inverse ratio of change in input offset voltage to
a change in power supply voltage. PSRR can be
specified in dB or jJ.VIV.

Input Noise Current Density (In) - The rms
noise current in a 1Hz band centered on a
specified frequency.

Supply Current (ISY) - The current required
from the power supply to operate the amplifier
with no load and the output at zero volts.

Input Noise Voltage (enp-p) - The peak-to-peak
noise voltage within a specified frequency band.
Input Noise Voltage Density (en) - The rms
noise voltage in a 1Hz band centered on a
specified frequency.

Unity-Gain Closed-Loop Bandwidth (BW)
-The frequency at which the magnitude of the
small-signal voltage gain of the amplifier, operated closed-loop as a unity-gain follower, is
3dB below unity.

Input Offset Current (los) - The difference
between the currents into the two input
terminals when the output is at zero volts with
no load.

MATCHING PARAMETER DEFINITIONS

Input Resistance-Differential Mode (RIN) - The
ratio of small-signal change in input voltage to a
change in input current at either input terminal
with the other grounded.

Input Offset Voltage Match (~Vos) - The difference between the offset voltages of side A and
side B (VasA - VOSB)· If VaSA = VOSB, the net
differential offset voltage at the output of the
amplifier pair equals zero.
Input Offset Voltage Tracking (TC~Vos)
- The ratio of change in ~VOS to a change in
temperature.

Input Voltage Range (IVR) - The range of input
voltage for which the device will operate as a
linear amplifier.

Average Nonlnvertlng Bias Current (Ia+) - The
average of the side A and side B noninverting
input bias currents:

Large-Signal Voltage Gain (Avo) - The ratio
of change in output voltage (over a specified
range) to a change in input voltage.

IBA+ + IBB+
2
Noninverting Input Offset Current (los +) - The
difference between the noninverting input bias
currents of side A and side B; (IBA+ -IBB+).
Inverting Input Offset Current (los -) - The
difference between the inverting input bias
currents of side A and side B; (I BA- - I BB- ).
Average Drift Of Noninverting Bias Current
(TCla+) - The ratio of change in noninverting
bias current to a change in temperature.

Input Offset Voltage (Vos) - The voltage which
must be applied between the input terminals to
obtain zero output voltage with no load.

Open-Loop Output Resistance (Ro) - The
small-signal driving-point resistance of the
output terminal with respect to ground at a
specified quiescent DC output voltage and
current.
Output Voltage Swing (Vo) - The peak output
voltage that can be obtained without clipping
into a specified load resistance.

5-5

II

OPERATIONAL
AMPLIFIERS
Precision Monolithics Inc.

Average Drift of Noninverting Offset Current
(TClos+) - The ratio of change in noninverting
offset current to a change in temperature.

rate needs. Distortion is often critical for audio
use and fast settling may be essential in a dataconversion application.

Common-Mode Rejection-Ratio Match (aCMRR)
- The difference between the common-mode
rejection ratios (expressed in volt/volt of side A
and side B. aCMRR in dB = 20 10glO (aCMRR in
volt/volt).

Output Load - Op amps are sometimes called
upon to drive long cables, storage capacitors,
transformers, or other semiconductors. Highspeed circuits generally require low-impedance
feedback elements and the load is usually low
impedance; therefore, relatively high output
current drive is needed for high-speed circuits.

Power Supply Rejection-Ratio Match (aPSRR)
- The difference between the power supply
rejection ratios (expressed in volt/volt) of side A
and side B. aPSRR in dB = 20 10glO (aPSRR in
volt/volt).
Channel Separation - The ratio of change in
offset voltage of one channel to a change in
output voltage in the second channel.

Environmental Conditions - Temperature
range and power supply characteristics are very
important factors. Power supply drain is often
critical in battery-powered equipment, process
control systems, and satellites. In addition,
op-amp package type is generally dictated by
environmental and cost factors. Another factor
to consider is the electrical environment. Minimize accuracy degradation from unavoidable
ground noise and power supply fluctuations by
choosing an op amp with high CMRR and
PSRR.

SELECTION PRINCIPLES
Selecting an operational amplifier can be a
frustrating experience. The choice of circuit
configuration, and of associated component
values, interrelates to the choice of op amp for
a given application. Op amps are specified as
open-loop devices, but in a circuit application
they generally have feedback applied. The
designer must predict the closed-loop circuit
performance as determined by his choice of op
amp and choice of circuit configuration (and
component tolerances). Detailed literature is
available on circuit configurations to accomplish
particular analog circuit functions using op
amps. This Selection Guide gives recommended
guidelines and a design strategy for selecting op
amps to best meet your needs.

2. Use the published op-amp specifications
and characterization graphs.
PMI provides comprehensive specification
tables with well-defined test conditions for
operation at 25° C and over specific temperature ranges. The "Typical Performance Curves"
show the characteristic response of an op amp to
variations in frequency, temperature, supply
voltage, or load impedance. Since op amps often
perform much better than indicated by their mini
max specification limits, the designer may be
tempted to ask for special selection to tighter
limits. Although sometimes necessary, special
selection tends to be costly. A better strategy is
to select a standard op amp that meets the
application need on a worst-case basis. Careful
initial selection of a high-performance standard
op amp will provide predictable circuit performance on a continuing basis.

The first design steps are to:
1. Completely define the design objectives.
Input Signal- Determine the signal level,
frequency content, and impedance of the input.
Accuracy Required - For lillear amplification,
this consists of limits for offset, gain error, and
nonlinearity. Establish your bandwidth and slew

5-6

OPERATIONAL
AMPLIFIERS
Precision Monolithics Inc.

Two factors will generally dictate the op amp
choice:
1. The loop gain (excess of open-lOOp gain
over closed-loop gain) must be sufficient at
the highest frequency of interest. For example,
if 1.0% accuracy at 10kHz is required when
operating at closed-loop gain of 10, then the op
amp must have an open-loop gain of at least
1000 at 10kHz (10/1000 = 1%). When operating
at high closed-loop gains, decompensated op
amps generally offer the advantage of better
gain bandwidth product without a price/
performance penalty.

Selection Process
Operational amplifiers can be divided into four
basic functional categories:
Category

Primary Characteristics

General Purpose

"741" types.

High Accuracy

Low input offsets
(Vas < 1mV), high DC
gain, high CMRR, and
low noise. Leading part
types are OP-05, OP-07,
OP-27, OP-50, and
OP-77.

High Speed

Optimized for high slew
rate, high gain-bandwidth, and fast settling
time.

Low Power,
Wide Supply Range

Low supply drain
(Isy < 1 mAl, wide input
and output voltage range.
Includes micropower
(Isy < 100J.LA) units for
battery operation.

2. Slew rate must be high enough to follow the
fastest signal input without causing distortion or
other anomalies. Slew-rate symmetry, linearity,
and overload recovery should be considered.
The detrimental effects of slew-rate limiting can
be subtle; it is best to avoid trouble by choosing
an op amp with at least a 20% safety margin in
minimum slew-rate.
High speed implies a need for high output
current. Applications such as audio amplifiers,
active filters, DAC-output amplifiers, and fast
integrators often require high output currents for
driving feedback capacitors or low-impedance networks. Driving such capacitive loads as long
cables or storage capacitors at high frequency
requires high output currents.

There can be overlap between some categories, while others are mutually exclusive.
For example, the PMI OP-22 covers both "High
Accuracy" and "Low Power". However, "High
Speed" and "Low Power" tend to be mutually
exclusive; it is difficult to simultaneously
optimize both speed and power.

DC Considerations
If the frequency requirements are relatively
modest (SR < 10VlJ.Ls) and the circuit requires
closed-loop gain above unity, choose a "High
Accuracy" op amp. These op amps feature:

Economics is another important dimension of
the selection process. The "General Purpose"
category is generally lowest in cost, but a "High
Accuracy" op amp with low input-offset-voltage
may be more cost effective if it eliminates the
need for external trimming components. The
PMI high-accuracy OP-07 is often used in place
of general-purpose 741-types because of its
low input-offset-voltage and high gain.

Low Input Offsets
Low Input Offset
Voltage
Low Input Bias
Current
- Bipolar Input
Stage
- JFET Input
Stage

AC Considerations
Consideration of AC requirements for an
application is a good starting pOint in the op
amp selection process. If high frequency
(GBW> 10MHz, SR > 10VlJ.Ls) is the primary
concern, then the choice quickly narrows down
to the "High Speed" category.
5-7

Vas':501mV,
TCVas':50 2J.LV/o C

18':50 100nA (18':50 10nA
is desirable)
18':50 200pA

II

OPERATIONAL
AMPLIFIERS
Precision MOllolithics Inc.

High Open-Loop
DC Gain
High Common-Mode
Rejection
Low Input Noise
at 100Hz

AVOL;:::: 1,000,000

The OP-41/43 JFET-input op amps have less
than ±5pA of input bias current, and the OP15/16/17 JFET-input op amps have bias currents
specified to within ±50pA. At high temperature,
however, the input bias current for JFET-input
op amps rises to a level which is typically 3
orders of magnitude higher than the roomtemperature value.
The OP-05, OP-07, OP-77, and OP-50 are bipolar op amps which have input bias cancellation
circuitry that significantly reduces input-biascurrent, and maintain low input-bias-current
levels over temperature. For example, the OP05/07/77 op amps have bias-current limits set at
±4nA over the full Military temperature range.
The OP-08, which is an improved LM108, has a
superbeta input stage which also provides very
good high-temperature high-impedance operation (Is S; 3nA at 125°C).
The OP-22/32 programmable micropower op
amps have PNP input stages which also have
very low input bias current over temperature.
Their input-bias-current remains below 5nA over
the full Military temperature range.

CMR;:::: 100dB
en

S;

15nVly'HZ

A leading op amp in the High-Accuracy
category is the PMI OP-77 with these key
specifications:
Vas S; 25,..tV
Is S; ±2nA, los S; 2nA
AVOL ;:::: 5,000,000
CMRR;:::: 114dB
en S; 11 nVly'HZ at 1000Hz
The OP-77 performs very well even at high
closed-loop gains. For example, consider a
non inverting configuration with a closed-loop
gain of 100. Assume a signal source with a range
of ±0.1V and source impedance of 10kO. If the
feedback resistances are chosen to be relatively
low, then the maximum offset caused by input
bias current will be 4.4nA (Is + 10512 = 4.4nA)
multiplied by the 10kO source resistance, or
44~V. Total input offset, even without external
offset nulling, will be less than 119~V (approximately 0.12% of full-scale). The effect of CMRR
is negligible in this example; the ±0.1V input
divided by 110dB of common-mode rejection is
only 0.3~V referred-to-input. Gain error factor is
1/(1 + AVCL/AvoL), which is a gain error of
approximately AVCL/AVOL. The DC gain error at
AVCL of 100 will be less than 10015,000,000, a
0.002%-of-full-scale gain error. The worst-case
sum of offset and gain errors for this example is
only 0.12% of full-scale, and is achieved without
any external trimming of offset or gain.

Low noise, always desirable, is sometimes the
primary consideration. In many high-gain
active-filter or audio-amplifier applications, low
noise can be more important than DC offset.
These are the three basic rules for obtaining low
noise:
Using low
impedances minimizes the effect of current
noise flowing through the source impedance,
reduces resistor thermal noise, and reduces stray
pick-up of RF noise.
2. Restrict the system bandwidth - Noise
outside the frequency range of interest can usually
be attentuated by filtering. Block high-frequency power-supply noise from the signal path
by use of decoupling capacitors at the op-amp
supply inputs.
1. Design with low impedances -

Selection of a specific op-amp type within the
High-Accuracy category is generally determined
by impedance levels of the input signal and
feedback elements. High impedances
(Rs> 10kO) imply a need for an op amp with low
input bias currents. This need for low bias current can be met through use of FET-input op
amps, or by using bipolar-input op amps specifically deSigned for low input-bias-current.

3. Select a low-noise op amp - Some op amps,

such as the bipolar-input OP-27, are designed
for minimum noise. The input stage current is
set to a relati~IY high value which reduces input
noise (5.5nVl Hz max at 10Hz). Output swing

5·8

OPERATIONAL
AMPLIFIERS
Pn."ClS10ll

M0l101ithics Inc.

is increased to ± 10V into 600n to allow the use
of low-impedance, low-noise feedback
elements.

The PMlline of low-power, wide-supply-range
op amps all feature high open-loop gain, low
input offsets, and high CMRR. They can provide
high accuracy even at high closed-loop gain.

Power Supply Considerations
The op-amp power-supply requirements are the
next factors to consider. If the circuit is to be
operated from a battery, such as in portable
instruments, missiles, or spacecraft, then narrow
the selection to the "Low-Power, Wide Supply
Range" category. Low-power op amps are
designed for minimum quiescent supply current.
Speed is traded off for lower power consumption and output drive is generally reduced. The
input and output stages are designed for linear
operation over a wide voltage range which is
very helpful for single-power-supply operation.

The low-power family includes programmable
micropower op amps that offer the designer
another dimension in circuit design. The quiescent supply current is set by an external resistor
which allows the circuit designer to trade off
quiescent supply current against speed. Since
the quiescent current directly controls slew rate
and gain-bandwidth product, these programmable op amps are easily frequencycompensated in such circuits as active filters,
oscillators, or multi-stage instrumentation
amplifiers.

DC ERROR CALCULATIONS FOR STANDARD CONFIGURATIONS
NONINVERTING CONFIGURATION

r--- ,
I INPUT I

I

I

I

Rs I
I
I
IVI'U
I
I _ _ .JI
~

(1+ ~~) [vos + (R~~R~2 - RS) Is + (R~~R~2 + Rs) I~S ]

Output Offset =
Special Cases:

Max Output Offset = (1

+

~~)

(vos + Rs lOS) if Rs =

R~~ ~2

- (1+ ~~) [vos + IR~\R~2 -Rslls] if 10soC Is
(lOS)].
R1 R'2
R2) [
- (1 + "Fi1 Vos + Rs Is + "'2 If Rs > R1 + R2
Note: Is is the average of the Input bias currents and los is the difference.

Gain Error -

(1

+ ~~)

A:O +

C~RR

• where Avo = Open-Loop Gain and Avo> (1
5-9

+ ~~)

OPERATIONAL
AMPLIFIERS
PreClsiOIl MOllolithlcS Inc.

INVERTING CONFIGURATION
INPUT
r----.,
I

I

I
I
I VI 'V
IL ____ .JI

Output Offset = (1 +

:~) [V os + (R~~ R~2 - RC)

Ie +

(R~~R~2 + Rc) I~S]

if Rs - 0

Special Cases:
MaxOutputOffset= (1+

:~)

(VOS+RCIOS) ifRc= :::2R2

- (1+ :~) [VOS+
- (1 +

IR~1+R~2-Rclle]

if 10s4!i Ie

:~) [VOS + R~1+R~2 (Ie + I~S)] if Rc = 0

Note: Ie is the average of the input bias currents and los is the difference.
1 + RRs
Gain Error - (1 + RR2) -A
1
VO
1

~
R1

,where Avo = Open-Loop Gain and Rs 4!i R1

5-10

OPERATIONAL
AMPLIFIERS
PrecisloJl Monolithics Inc.

input voltage noise. The OP-08 and OP-12 provide an excellent combination of low input-biascurrent, low offset voltage, and moderate power
drain. The OP-07 and its improved version, the
OP-77 offer a selection of the lowest offset
voltages (60,.LV max to 250lo'V max) combined
with low input-bias-current (±2nA max to
±12nA max) and have become an industry
standard for high-precision applications.

OPERATIONAL AMPLIFIER
SELECTION GUIDE
The Operational Amplifier Selection Guide
chart highlights PMl's line of operational amplifiers. The matrix indicates the most essential
parametric differences for each product group.
"General Purpose" op amps are usually the least
expensive and are recommended for applications where impedance levels are relatively
low, closed-loop gain is low, and speed
requirements are moderate. JFET inputs provide
lower input-bias-currents and better bandwidth
than standard bipolar inputs, but input voltage
offsets and noise are generally better for the
bipolar input amplifiers.
The "High Accuracy" category presents the best
amplifiers for high-gain applications. A combination of low input-offset-voltage, high openloop gain, and high CMRR provide excellent DC
accuracy even at high closed-loop gain. The
OP-27, OP-37 and OP-50 are best for minimum
General Purpose
Bipolar

Monolithic Technology

Precision
HighSpeed
JFET
Input

JFET
Packages
Single
Dual
Quad
Ultra Low Ollset, Vas < 200"V
Low Offset, Vas < lmV
Low Bias Current
Is" 100pA
High Gain, AYOL;o, 1 Million V/V
High Slew Rate, SR > 10V/"s
Low Power
Isy < 1mA/Ampllfler
Isy < 1OOIlA/Amplifler
1kHz)

en "SnV/JRZ
In" 0 25pA/JRZ

High
Accuracy

Low Power,
Low InputBiasCurrent

Micropower

•••••• •••••••••••
••••••
•
•
• • • • • •• •••• • • • • •
•
•
••
•
•
•
•
••
•
• ••
•••••
••••
• •• • •
•
••••••
•
•• ••• ••
•
•
•
••• •
••••• •
• •
•
• • • ••• ••
• •
• •• • •
•••
••••••• • •
••••

Bipolar Input

Low Noise (10

JFET

PMI is a leader in op amps featuring low power
consumption. The OP-20/21/22132 are
micropower op amps that operate with only a
few microamps of supply drain. PSRR and
CMRR are high, and the input-voltage-range is
wide. Such features work together to make
these amplifiers ideal for battery-powered
applications or for operation from a single
supply voltage. The OP-22/32 can be
programmed to operate over any supply current
from 11o'A to 400lo'A and is excellent for batterypowered designs.

•

•

•

•

•

5-11

OP-Ol
INVERTING HIGH -SPEED
OPERATIONAL AMPLIFIER
Precision MOIlolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

and excellent DC input characteristics. An internal feedforward frequency compensation network provides simplicity of application - no external capacitors are required for
stable, high-speed performance. The fast output response Is
achieved without sacrifice of input bias current or power
consumption. A 250kHz typical power bandwidth is attained
with a small-signal bandwidth of only 2.5MHz, thus board
layout is non-critical. The OP-01 is completely protected at
both input and output, fits standard 741 sockets, and is offset
nulled with a 10k!} potentiometer.

Fast Settling Time ................... 1J's to 0.1% Max
High Slew Rate .......................... 12V1J's Min
Po_r Bandwidth ........................ 150kHz Min
Low Power Consumption .........•........ 90mW Max
Excellent DC Specifications
Internally Compensated
Ideal DAC Output Amplifier
MIL-STD-SS3 Processing Available
Fits Standard 741 Sockets
Low Cost

The fast output response combined with excellent settling
time makes the OP-011deal for use as a OfA converter output
amplifier.

ORDERING INFORMATIONt
PACKAGE
TA= 25°C
YosMAX
(mY)

HEFIMETIC
TO-"
I-PIN

HEFIMETIC
DIP
I-PIN

0.7
0.7
5.0
5.0

OP01J'
OP01HJ
OP01GJ'
OP01CJ

OP01Z'
OP01HZ
OP01GZ'
OP01CZ

PLASTIC
DIP
I-PIN
OP01HP
OP01CP

PIN CONNECTIONS

OPEFIATING
TEMPEFlATUFlE
flANGE

BAL98

MIL
COM
MIL
COM

N.C. 7

v+

sour

-IN 2

'For devices processed In total compliance to MIL-STD-883. add 1883 after
part number. Consult factory tor 883 data sheet.
tAli commercial and Industrial temperature range parts are available with
burn-In. For ordering information see 1986 Data Book, Section 2.

+IN 3

EPOXY MINI-DIP
(P-Sutflx)
&
S-PIN HERMETIC DIP
(Z-Sufflx)

GENERAL DESCRIPTION
The OP-01 series of monolithic inverting high-speed operational amplifiers combines high slew rate, fast settling time

5 BAL

4

v- (CASEI

TO-SS
(J-Sufflx)

SIMPLIFIED SCHEMATIC
V+

*Q1. QZ, Q3 AND Q4 FORM A
THERMALL V CROSS-COUPLED
QUAD. as. 0&1, 06 AND 06 1
COMPRISE A SIMILAR
THERMALLY CROSS-COUPLED
QUAD.

R'O
~---+--<>OUTPUT

RD

R,

RS

R2

BALANCE

5-12

1/86, Rev. A

---------Im

OP-01INVERTING HIGH-SPEED OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

(Note 2)

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300·C

Total Supply Voltage, OP-01, OP-01 H, OP-01 N, OP-01 NT,
OP-01G, OP-01GT .............................. ±22V
OP-01G, OP-01C, OP-01GR ..................... ±20V
Power Dissipation (Note 1) ...................... 500mW
Differential Input Voltage .......................... ±30V
Input Voltage (Note 3) •••••••••••••••••••••••••••• ±15V
Short-Circuit Duration •••••••••••••••••••••••• Indefinite
Operating Temperature Range
OP-01, OP-01G ..................... -55·C to +125·C
OP-01H,OP-01C ....................... O·C to +70·C
DICE Junction Temperature (Tj) ••••••• -6S·C to +150·C
Storage Temperature Range
J and Z Packages.. .. .. .. • .. • .. .. ... -65· C to + 150· C
P Package ......................... -65·C to +125·C

ELECTRICAL CHARACTERISTICS at Vs = ±

15V, TA

NOTES:
1. See table lor maximum ambient temperature rating and derating lactor.

PACKAGE TYPE

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE
7.1mW'oC

TO-99 (J)

80°C

8-Pin Hermetic DIP (Z)

75°C

6.7mW'oC

8-Pln Plastic DIP (P)

35°C

5.6mW'oC

2. Absolute maximum ratlngaapplyto both packaged parlsand DICE, unless
otherwisa noted.
3. For supply voltages less Ihan ±15V, Ihe maximum Input vollage Is the
supply voltage.

= 25·C, unless otherwise noted.
OP-01
OP-01H

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

RsS20kO

Input Ollsal Current

los

Input Bias Currant

Is

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VCM =±10V
RsS 20kO

PSRR

Vs = ±5V to ±20V
RsS 20kO

Output Voltage Swing

Vo

RL ;;,5kO
RL;;'2kO

Large-Signal
Voltage Gal n

Avo

RL;;'2kO
Vo =±10V

Power Consumption

Power Supply
Rejection Ratio

MIN

TYP

MAX

0.3

MIN

~

TYP

MAX

UNITS

0.7

2.0

5.0

mV

0.5

2.0

2.0

20

nA

18

30

25

100

nA

±12

±13

±12

±13

V

85

110

80

100

dB

10

100

60

150

I'VN

±12.5
±12.0

±13.5
±13.0

±12.5
±12.0

±13.5
±13.0

V

50

100

25

75

VlmV

Pd

VOUT=O

50

90

50

90

mW

Settling Time
to 0.1% (Summing
Node Error)

ts

Av =-1
(Notes 1, 2)
VIN =5V

0.7

1.0

0.7

1.0

I'S

Slew Rate
(Notes 2, 3)

SR

Av =-I,
Rs =3k to 5kO

12

18

12

18

VII's

Large-Signal
Bandwidth
(Notes 3, 4)

150

250

150

250

kHz

Smail-Signal
Bandwidth
(Notes 3, 4)

1.5

2.5

1.5

2.5

MHz

150

ns

Risatime

Ir

Overshoot

OS

Av =-l
VIN = 50mV

150

%

NOTES:
1. RL= 25kO; CL = 50pF. See Settling Time Test Circuil.
2. Sample tested.
3. See applications Information.
4. Guaranteed by design.

5-13

ffi

OP-01G
OP-01C

1/86, Rev. A

::l
~

~

~

-~

0

;;.I.l

ts

~ OP-01 INVERTING HIGH-SPEED OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ±15V, -55·C:5 TA:5 +125·C for OP-01, OP-01G and O·C:5 TA:5 +70·C for
OP-01 H, OP-01 C, unless otherwise noted.
OP-01
OP-01H
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs" 20kn

Input Offset Current

los

Input Bias Current

Is

Input Voltage Range

IVR

MIN

TYP

MAX

TYP

MAX

UNITS

0.4

1.0

3.0

6.0

mV

4

4

40

nA

50

50

200

nA

30

Common-Mode
Rejection Ratio

CMRR

VcM =±10V
Rs" 20kn

Power Supply
Rejection Ratio

PSRR

Vs = ±5V to ±20V
Rs" 20kn

Large-Signal
Voltage Gain

Avo

Output Voltage Swing

Vo

Offset Voltage
Drift (Note 1)

TCVos

RL~2kn

RL~5kn

RL~2kn

±13

±10

±13

V

85

110

80

100

dB

100

60

150

",VN

30

80

15

50

VlmV

±125
±12.0

±13.5
±13.0

±12.5
±120

±13.5
±13.0

V

2

Rs ,,5kn

MIN

±10

10

Vo =±10V

OP-01G
OP-01C

8

5

20

,..VloC

NOTE:

1. Sample tested.

5-14

1/86, Rev. A

----------l~ OP-01 INVERTING HIGH-SPEED OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS

(125°C TESTED DICE AVAILABLE)

1.
2.
3.
4.
5.
8.

NULL
INVERTING INPUT
NON INVERTING INPUT
VNULL
OUTPUT

7. V+

For additional DICE information refer to
1986 Data Book, Section 2.
DIE SIZE 0.046 X 0.042 Inch, 1932 sq. mils
(1.17X 1.07 mm,1.25 sq. mm)

WAFER TEST LIMITS

at Vs= ±15V, TA=25°C for Op·01N, OP·01G and OP·01GR devices; T A =125°C for OP·01NT and

OP·01GT devices, unless otherwise noted.

PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

Input Offset Current

los

Input Bias Current

Ie

Input Voltage Range

IVR

OP-01NT
LIMIT

OP-01N
LIMIT

OP-01GT
LIMIT

OP-01G
LIMIT

OP-01GR
LIMIT

UNITS

1.0

0.7

3.0

2.0

5.0

mVMAX

Rs" 20kO

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio

PSRR

Output Voltage
Swing

VOM

RL " 5kO
RL" 2kO

Large-Signal
Voltage Gain

Avo

RL" 2kO
Vo =±10V

Power Consumption

Pd

VOUT=O

VcM =±10V
Rs" 20kO
Vs = ±5V to ±20V
Rs" 20kO

4

2

10

20

nAMAX

50

30

100

50

100

nAMAX

±10

±12

±10

±12

±12

VMIN

85

85

80

80

80

dBMIN

60

80

100

100

150

"VN MAX

±12.5
±12.0

±12.5
±12.0

±12.5
±12.0

±12.5
±12.0

±12.5
±120

VMIN

30

50

25

50

25

VlmV MIN

90

90

mWMAX

90

NOTES:
For 2S Q C characteristics of NT & GT devices, see N & G characteristics respectIVely
Electrical tests are performed at wafer probe to the limIts shown Due to vanatlons in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotIate specIfications based on dice lot qualification through sample lot assembly and testIng.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V, TA =

25° C, unless otherwise noted.
ALL GRADES
TYPICAL

UNITS

AVCL =-1. Rs =3kOt05kO

18

VI,..

V ,N = 5V
Av=-1
RL = 2kO (See Settling Time Test CirCUIt)
C L =50pF

1.0

"s

Large-Signal
Bandwidth

250

kHz

Smail-SIgnal
BandWIdth

2.5

MHz

Risetlme

150

ns

PARAMETER

SYMBOL CONDITIONS

Slew Rate

SR

Settling Time to
ts
0.1%
(Summing Node Error)

5-15

1/86, Rev. A

----------I!EMD

OP-01 INVERTING HIGH-SPEED OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
LARGE-SIGNAL
PULSE RESPONSE

SMALL-SIGNAL
PULSE RESPONSE

TIME (500ms!DIV)
Vs

e

±15V, AV

TIME (100ms/DIV)

=-1, AL .. 2kn. CL ,. 50pF

Vs "':t15V, AV '" -1, AL

UNITY-GAIN BANDWIDTH
V8 SOURCE RESISTANCE
I _I
1,
2.

3.
4.
6.

-2

r-

0'1

II

I
I
I

~

'm
3

R

~
w

• >;

";:!

~ 4

0

1\1

0,10

0

\

1\

6

1.00

10.0

FREQUENCY (MHz)

-,

III

2.
3.

4.

RS'" 10kO

INVERTING

•

AV;;;' 60
3. NON·

3

INVERTING
AV" 10

I

I

4. NON·
INVERTING

~
I

1\

1\
...l

0.10

\\
[

~O~;AGE

FOLLOWER)

2~)~

I

Y

o 111111111 111111111 II!I
o 111111111 111111111 1111
0.001 0,01

1, INVERTING

A V "-'
2 2. NON.

I

•
RS" 100.0
RS,. lkn
RS" 3.3kO I

1\.1 M

jI.,[

2

,.

'2

,

1r,

60

LARGE-SIGNAL OUTPUT
SWING V8 FREQUENCY
II

I

~

iii 8 0
l!
z

_'U

\

I
I

'00

I

......

-4

0.01

'20

II

1111

AS. 2,6kO
RS·6.0krl
RS" 10kn
RS" 33kG
RS. 100kn

Or--- 1"'\0{

OPEN-LOOP GAIN
V8 FREQUENCY

=2kn, CL .. 50pF

Ulill

1.00

10,0

100

1000 10,000

0,001

0.10

1.00

10,0

FREQUENCY (MHz)

PREQUENCY (kHz)

APPLICATIONS INFORMATION

0,01

FAST INVERTING AMPLIFIER

The OP-01 incorporates an internal feed-forward compensation network to provide fast slewing and settling times in all
inverting and mOderate-to-high-gain noninverting applications. Unity-gain bandwidth is a function of the total equivalent source resistance seen by the inverting terminal. Proper
choice of this resistance will allow the user to maxmlze
bandwidth while assuring proper stability. The equivalentinverting-terminal~resistance is defined as R'N II RF, and it
must be greater than 3.3kO to assure stability in all closedloop gain configurations including unity gain. Should
R,N" RF:5 3.3kO, a resistor (Rs) may be placed between the
inverting input and the sum node to provide the required
resistance. (See Fast Inverting Amplifier Diagram.) Lower
values of total equivalent resistance may be used to improve
bandwidth in higher closed-loop gain configurations, as
indicated by the Open-Loop Gain vs. Frequency plot.

Rp

Reo= RS+ RIN II RF

FOR AV = -1, Rea ;';'3.3kn
Rp = Rea

5-16

1/86, Rev. A

_ _ _ _ _ _ _ _~~ OP-01 INVERTING HIGH-SPEED OPERATIONAL AMPLIFIER

TYPICAL APPLICATIONS

SETTLING-TIME TEST CIRCUIT
Settling time may be measured using the circuit shown
below. This circuit Incorporates the "false sum node" technique to produce accurate, repeatable results. For a 5V Input
step, 0.1 % settling will be achieved when the false sum node
settles to within ±2.5mV of Its final value. The oscilloscope
used for observation of the false sum node should have wide
bandwidth, fast overload recovery time, and be used with a
low capacity probe (S 10pF, including strays). A Tektronix
7504 scope with a 7A 11 probe or equivalent Issuggested. The
pulse generator should have a 50n output impedance and be
capable of a 5V rise time in S 20ns with ringing less than
2.5mV after 0.5I's. Measurements to 0.1 % require RINto equal
RFwithin 0.01%; Rsand Reare used as trimming resistors to
achieve this matching.

VA TO SCOPE CIN - l0pF

rAS

INPUT

rAI

A7

&kll

..7kG

~

SOn:l:1%

aBOn±li%

E2

~

"',BV

1¥

24011

~

OUTPUT

Bkll
to.6%
A3

~

PRECISION POWER-BOOSTER CIRCUIT

AI
1&00:t:1%

A2

:1:0.8%
AS

LSB

TVPICAL SETTLING
TIME O.Bj.l1

~

AF - . ,

10kn A 10kO
to.I5%
:1:0.8%

Al

Msa

0-0- - - - - - - - - - ,

AIN - . ,

1aOOt'"

FAST VOLTAGE-OUTPUT D/A CONVERTER

O. '~F:I:'O%
""BV
7 ~

2W

o•.en

RL>
2kll

10kO

9

CL C;;
50pF
10ka

l&kll

VOUT

I

AL
4 O,1J.1F:t:1""

~

6

100n

20kll

0:-

0:-

-lav
2400

TYPICAL PERFORMANCE

OFFSET NULLING CIRCUIT

SLEW RATE--------------"" 18V/j.ls
01% SETTLING-------4J.1s (RL ·500n)
aUIESCENT SUPPLY CURRENT---15mA

v+

V-

5·17

1/86, Rev. A

~
f.I.l

~

OP-02
GENERAL-PURPOSE
OPERATIONAL AMPLIFIER
Precision Monolithics Inc.

c

FEATURES

guaranteed over the full operating temperature range.
Precision Monolithics' exclusive Silicon-Nitride "Triple
Passivation" process reduces "popcorn noise." A thermallysymmetrical input-stage design provides low input offset
voltage drift and insensitivity to output load conditions.

• Excellent DC Specifications
• Low Noise ............................. O.65/Np _p Typ
• Low Drift (TeVos) ........................ 81lVl" C Max
• Silicon-Nitride Passivation
• 125 0 C Tested Dice Available
• "Premium" 741 Replacement

The OP-02 is a direct replacement for the 741. It is ideal for
upgrading existing designs where accuracy improvements
are required and for eliminating special low-drift or low-noise
selected types.

ORDERING INFORMATIONj'
PACKAGE
TA = 25°C
VosMAX
(mV)

TO-99
a-PIN

HERMETIC
DIP
a-PIN

0.5
0.5
2.0
2.0
5.0
5.0

OP02AJ'
OP02EJ
OP02J'
OP02CJ
OP02BJ'
OP02DJ

OP02AZ'
OP02EZ
OP02Z'
OP02CZ
OP02BZ'
OP02DZ

* For devices processed

PIN CONNECTIONS
PLASTIC
DIP
a-PIN

OPERATING
TEMPERATURE
RANGE

OAfd,'NC'v.

MIL
COM
MIL
COM
MIL
COM

OP02EP
OP02CP
OP02DP

-IN 2

'.

+IN 3

bOUT

5 BAL

4 V- (CASel

TO-99
(J-Suffix)

total compliance to MIL-STO-883, Add 1883 after
part number. Consult factory for 883 data sheet
In

tAil commercial and Industnal temperature range parts /:Ire available with
burn-In. For orderrng Information see 1986 Data Book, Section 2

GENERAL DESCRIPTION
This high-performance general-purpose operational amplifier provides significant Improvements over industry-standard
and "premium" 741 types while maintaining pln-for-pin
compatibility, ease of application, and low cost Key specifications, such as Vos. los, lB. CMRR, PSRR, and Avo are

SIMPLIFIED SCHEMATIC

f --

THERMALLY CROSS CDUPLfD
QUAD U!:>, 05, Qfj AND U6'
COMPRISE. A SIMILAR THER
MALL YCROSS COlJPLEDOUAO
o---.,~-

EPOXY MINI-DIP
(P-Suffix)

---------------------'

-- .-.------ - - - - - - - _ . _ - - - - - - - - - - - - - - - - - - - - - - ,

"Q1, Q2, 03 AND Q4 FORM II

··IN

8-PIN HERMETIC DIP
(Z-Suffix)

~----

U8

_._-

--

-

~

r

09

- -__

----

- -----.-.-..:::-

OUT

Rll

5-18

1/86, Rev. A

-----------l~ OP-02 GENERAL-PURPOSE OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

(Note 2)

Supply Voltage ••••••••••••••••••••••••••••••••••• ±22V

NOTES:
1 See table for maximum ambient temperature rating and deratmg factor.

Power Dissipation (Note 1) ...................... 500mW
Differential I n put Voltage • • • • • • • • • • . • • • • • • . • • • • • • •• ± 30V
Input Voltage ......................... '" Supply Voltage
Output Short-Circuit Duration •••••••••••..•.•• Indefinite

PACKAGE TYPE

Operating Temperature Range
OP-02A, OP-02, OP-02B

••••••••••••.

-55°C to +125°C

OP-02E, OP-02C, OP-02D •••••••••••••••• O°C to +70°C
Storage Temperature Range •••••••••••

-65°C to +150°C

at Vs

7.1mW/'C

TO-99 (JI

ao'c
36'C

56mW/'C

8-Pln Hermetic DIP (ZI

75'C

6.7mW/'C

Absolute maximum ratings apply to both packaged parts and DICE. unless
otherwise noted.

= ±15V, TA = 25°C, unless otherwise noted.

OP-02A
OP-02E
PARAMETER

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

a-Pm PlastiC DIP (PI

Lead Temperature (Soldering, 60 sec) ••.•••••••••• 300°C
DICE Junction Temperature (Tj ) •.••••• -65°C to +150°C

ELECTRICAL CHARACTERISTICS

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs" 20kO

Input Offset Current

los

0.5

Input Bias Current

Is

18

Input ResistanceDifferential-Mode

R'N

(Note21

MIN

OP-02
OP-02C

TYP

MAX

0.3

0.5

MIN

TYP

20

30

OP-02B
OP-02D

MAX

MIN

TYP

MAX

UNITS

3

5

mV

5

5

25

nA

50

30

100

nA

34

5.7

2.0

5.2

±10

±13

±10

±13

±10

±13

V

85

100

80

95

70

85

dB

3.5

MO

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VCM = ±10V
Rs" 20kO

Power Supply
Rejection Ratio

PSRR

Vs = ±5 to ±20V
Rs" 20kO

Output Voltage Swing

Vo

RLe: 2kO

±12

±13

±12

±13

±12

±13

V

Large-Signal
Voltage Gain

Avo

RLe: 2kO
Vo =±10V

100

250

50

200

25

150

VlmV

Power Consumption

Pd

Vo=OV

Input Noise Voltage

8 np _p

O.IHz to 10Hz

Input Noise
Voltage Density

en

fo= 10Hz
fo= 100Hz
fo= 1000Hz

Input Noise Current

i np _p

0.1 Hz to 10Hz

Input NOise
Current Density

'n

fo= 10Hz
fo= 100Hz
fo= 1000Hz

Slew Rate

SR

Large-Signal
Bandwidth

(Note 11
Vo= 20Vp _p
(Note 11

Closed-Loop
Bandwidth

BW

Rlsetlme

tr

Overshoot

as

AVCL = +1

(Note 31
AVCL =+1
Y,N = 50mV (Note 11

10

40

30

60

70

50

100

100

50

90

150

90

IfoVIV

mW

0.65

0.65

0.65

IfoVp_p

25
22
21

25
22
21

25
22
21

nVl$.

12.8

12.8

12.8

pA p

1.4
0.7
0.4

14
07
0.4

1.4
0.7
0.4

pAl$.

0.25

0.5

4

0.5

0.25

0.5

VII's

8

8

4

8

kHz

1.3

1.3

13

MHz

200

0.25

e-

350
10

(Note 11

200

350
10

200

350

ns

10

%

NOTES:
1. Sample tested.
2. Guaranteed by input bias current.
3. Guaranteed by risetlme.

5-19

1/86, Rev. A

~
~
.....
>-LI

::sp..
~

~

0
.....

~

~

0

IfMD

OP-02 GENERAL-PURPOSE OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS

at Vs = ± 15V, -55' C :5 TA:5 + 125' C, unless otherwise noted.

OP-02A
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

RsS 20kn

TCVos

Rs =50n

Average Input Offset
Voltage Drift (Note 1)
Input Offset Current

los

Average Input Offset
Current Drift (Note 1)

TClos

Input Bias Current

Ie

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VCM=±10V
RsS 20kn

Power Supply
Rejection Ratio

PSRR

Vs = ±5 to ±20V
RsS 20kn

Large-Signal
Voltage Gain

Avo

Output Voltage Swing

Vo

RL~2kn

Vo=±10V
RL~2kn

MIN

TYP

OP-02

MAX

MIN

MAX

TYP

MAX

UNITS

1.4

3

3

6

mV

8

4

10

8

20

p'vrc

5

2

10

5

50

nA

7.5

75

15

150

30

300

pArC

30

60

40

100

50

200

nA

0.5
2

±13

±10

±13

±10

±13

V

80

95

80

95

70

85

dB

10

30

60

Input Offset Voltage

Vos

Rs S20kn

Average Input Offset
Voltage Drift (Note 1)

TCVos

Rs= 50n

Input Offset Current
Average Input Offset
Current Drift (Note 1)

100

100

150

p.VIV

50

100

25

60

25

60

VlmV

±12

±13

±12

±13

±10

±13

V

unless otherwise noted.

OP-02E
SYMBOL CONDITIONS

MIN

±10

ELECTRICAL CHARACTERISTICS at VS= ±15V, O'C:5 TA:5 +70'C,
PARAMETER

OP-02B

TYP

MIN

TYP

OP-02C

MAX

MIN

0.4

OP-02D

TYP

MAX

TYP

MAX

UNITS

1.2

3

3

6

mV

MIN

2

8

4

10

8

20

p'vrc

los

0.7

4

1.4

10

5

60

nA

TClos

7.5

120

15

250

70

500

pA/'C

Input Bias Current

Ie

22

50

25

100

50

200

nA

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VCM = ±10V
RsS20kn

Power Supply
Rejection Ratio

PSRR

Vs = ±5 to ±20V
RsS20kn

large-Signal
Voltage Gain

Avo

Output Voltage Swing

Vo

RL~2kn

Vo =±10V
RL~2kn

±10

±13

±10

±13

±10

±13

V

60

100

80

90

70

85

dB

10

30

60

100

100

150

p.VIV

50

100

25

60

15

25

VImV

±12

±13

±12

±13

±10

±13

V

NOTE:
1. Sample tested.

5-20

1/86, Rev. A

-----------l~ OP-02 GENERAL-PURPOSE OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS (125°C TESTED DICE AVAILABLE)
1.
2.
3.
4.
5.
6.

NULL
INYERTING INPUT
NONINYERTING INPUT
YNULL
OUTPUT

7. Y+
DIE SIZE 0.046 X 0.042 Inch, 1932 sq. mils
(1.17 X 1.07 mm, 1.25 sq. mm)

WAFER TEST LIMITS at Vs = ± 15V, TA = 25° C for
OP-02GT devices, unless otherwise noted.
PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

Input Offset Current

los

Input Bias Current

I.

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio
Output Voltage Swing

For additional DICE information refer to
1986 Data Book, Section 2.

OP-02N, OP-02G

and

OP-02GR

devices;

TA

= 125° C for OP-02NT and

OP-02NT

OP-02N

OP-02GT

OP-02G

OP-02GR

LIMIT

LIMIT

LIMIT

LIMIT

LIMIT

0.5

3

RsS 20kO

UNITS
mVMAX

3

25

nAMAX

50

30

60

50

200

nAMAX

±13

±13

±13

±13

±13

VMIN

VCM = ±10V
RsS 20kO

80

85

80

80

70

dBMIN

PSRR

Vs = ±5V to ±20V
RsS 20kO

60

60

100

100

150

I'V/v MAX

±12

±12

±12

±12

±12

VMIN

50

100

25

50

25

VimV MIN

90

90

mWMAX

Vo

RL" 2kO

Large-Signal
Voltage Gain

Avo

RL" 2kO
Vo= ±10V

Power Consumption

Pd

Vo=OV

90

NOTE:
For 25°C characteristics of NT and GT devices, see Nand G characteristics, respectively.
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging Is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at

PARAMETER

SYMBOL

Input Resistance
Differential-Mode

RIN

Input Noise Voltage

enp •p

0.1 Hz to 10Hz

Input Noise
Voltage Density

en

fo= 10Hz
fo= 100Hz
fo= 1000Hz

Input Noise Current

I np _p

0.1 Hz to 10Hz

Input Noise
Cu rrent Density

in

fo= 10Hz
fO = 100Hz
fo= 1000Hz

Slew Rate

SR

Large-Signal
Bandwidth

Vs

= ±15V, TA = 25° C, unless otherwise noted.

CONDITIONS

OP-02NT
OP-02N

OP-02GT
OP-02G

OP-02GR

TYPICAL

TYPICAL

TYPICAL

UNITS

5.7

5.2

3.5

MO

0.65

0.65

0.65

I'Vp.p

25
22
21

25
22
21

25
22
21

nViVHz

12.8

12.8

12.8

pA p_p

1.4
0.7
0.4

1.4
0.7
0.4

1.4
0.7
0.4

pAlVHz

0.5

0.5

0.5

ViI's

Vo= 20Vp_p

8

kHz

Closed-Loop
Bandwidth

BW

AVCL = +1

1.3

1.3

1.3

MHz

Risetime

tr

Av=+1
VIN = 50mV

200

200

200

ns

Overshoot

OS

15

15

15

%

8

I'V/oC

30

pAlOC

Average Input Offset
Voltage Drift
Average Input Offset
Cu rrent Drift

TCVos

Rs= 5000
(Note 1)
7.5

TClos

5-21

15

1/86, Rev_ A

f&......

....
::sp..

~

~

0
......

~

~

0

------------IIEM!) OP-02 GENERAL-PURPOSE OPERATIONAL AMPLIFIER
BURN-IN CIRCUIT

OFFSET NULLING CIRCUIT
~

+20V

-----2N7

•

OP·02

'V·
-20V

TYPICAL PERFORMANCE CHARACTERISTICS
INPUT SPOT NOISE
VOLTAGE VI FREQUENCY

INPUT WIDEBAND NOISE VI
BANDWIDTH (0.1 Hz TO
FREQUENCY INDICATED)

INPUT SPOT NOISE
CURRENT VI FREQUENCY

1000

10

100
Vs = 115V

TA "2SOC
RS son

Imlll

Vs -

~

.tl~~~

r-

TA=2SoC

Vs .t15V
TA"'+25"C

TYPICAL

V
TYPICAL
1.0

1

1.0

10

100

1000

0.01

0.10

FREQUENCY 1Hz)

10

~

1000
TA

25"e ~

f=

.-

....... ,.......

10

i

2'00

S!

~

Ii:

i

'/

Ii

1.0

a:

~

0.1

o
0.1

100

10

POWER CONSUMPTION
VI POWER SUPPLY

100

. / NEGATIVE SWING

10

1.0

BANDWIDTH (kHz)

VI TEMPER~TURE

= :t:15V

12

1000

100

DIFFERENTIAL
INPUT RESISTANCE

~OS'~'~E ~,~~

~ 25"le
V;

T

14

1.0

FREQUENCY (Hz)

OUTPUT VOLTAGE VI
LOAD RESISTANCE

,.

O. 1
0.1

0.1

0.10

0.01

1.0
LOAD RESISTANCE TO GROUND (kO)

10

-60

-20

+20

+60

TEMPERATURE (eC)

5-22

+100

+140

10

/

1.0
20
40
60
TOTAL SUPPLY VOLTAGE, V+ TO V- (VOL TSI

1/86, Rev. A

-------------l!rHD

OP-02 GENERAL-PURPOSE OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
POWER CONSUMPTION
vs TEMPERATURE
35

50
Vs;c ±15V

I
\

OPEN-LOOP GAIN
vs TEMPERATURE

OUTPUT SHORT·CIRCUIT
CURRENT vs TIME

1

I

OP 02A

I-- f--(D VIN(PIN 3) =-

30

t- t--.

TA.+d~Gl
is'
"Iv

300

r--

-10mV, Vo" +15V VO" -15V

® VIN(PIN 3}" +10mV,

45

- -- -- ---r--._._-t--

\
-

._--'-----

.-

1\
1\\

25

~p OZE,C

I'.....
I-- -

.-

20

--- I--I--

+ 20

-20

+60

+140

+100

I-- -

175

--

2

....0

-20

INPUT OFFSET CURRENT
vs TEMPERATURE

+140

10
60

OP 02

l - I--

100

20

'\

1,5

075

-

10

OP 02A

050

e:..-'-

OP 02E

'"
..........

50

'\
"-...

05

I--

o

-20

+20

+60

+100

+140

40

'" U::t-N~~r
I'<

025

30

"

QP.02

" ".........

20

r-.....

+20

-20

-60

OP 02C

+60

+100

OPEN-LOOP GAIN vs
POWER SUPPLY VOLTAGE

TA-

1
25'C

=

10kH

110

t---.

~p\~~~\~ ~p\~;~\\

J

-60

-20

~P~\2\~\~P~

100
90

1

80

:

200

I--- ~

=

!

110

I

60

50
±5

±10

t:15

POWER SUPPLY (VOLTS)

,20

Ii

!

"

I

I
100

lk

5-23

+140

1IIIIInttt

olW~~I

10k

100k

'I

\

I
I

\

I
100

I,
I

\

I
10

lk

!Ii

"!ili'

i \

I

"

' ~I
I

I

\

70

50

lilil:!

T';'~ 25"C!~1

80

60

II
FREQUENCY (Hz)

+100

lill

~pl~~~I~ dp'~~;11

I

I
10

100
90

70

I

Opl:'AL

-

+60

PSRR vs FREQUENCY
120

±lSV

I

100

+20

TEMPERATURE (OC)

t~~I~~!cl
Vs

r-

OP-02E

+140

~P~2-

I--

o

CMRR vs FREQUENCY
120

OP 02e

.........
..........

10

TEMPERATURE (OC)

RL

+100

INPUT BIAS CURRENT
VB TEMPERATURE

25

OP 02e -

-

+60

+20

TEMPERATURE ('C)

50H

125

300

'\.

~15V

150 - R S

-60

\

50

,

o

UNTRIMMED OFFSET VOLTAGE
vs TEMPERATURE

:'\.

--

- I - I--

TIME FROM OUTPUT BEING SHORTED (MINUTES)

TEMPERATURE (OC)

i= Vs

r-

r--- -

1"( ..........

1"-

/'

/

. -I - -

"'-...
15

35
-60

-

\ [\.

40

V~' ';5V
"",

10k

I

"!I i
lOOk

FREQUENCY (Hz)

1/86, Rev. A

------------1~ OP·02 GENERAL·PURPOSE OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
MAXIMUM UNDISTORTED
OUTPUT VI FREQUENCY

~llIIJ!lc
VS=:l:15V

1\

,, iii 'ITTJ
I

j

1\

III

- r-.

80

'00

'\

TA·I

'\

i

II

'\
'\

: i;

10
100
FREQUENCY (kHz)

"'"

'\

III!

-40
'000

TA.2!OC
VS-:t16V -

80

vs·.,sv -

'\

11\1

25'6

II

III

,

11::1

'I

\

o

'20
I

CLOSED·LOOP RESPONSE
FOR VARIOUS GAIN
CONFIGURATIONS

OPEN·LOOP
FREQUENCY RESPONSE

""\

"'

-20
0.1

1.0

10

100
1k
10k 'OOk
FREQUENCY 1Hz!

1M

10M

10

100

1k

10k

lOOk

1M

10M

FREaUENCV (Hz)

TYPICAL APPLICATIONS
HIGH-8TABILITY VOLTAGE REFERENCE
'mA

A3

A2

6,4M2
+15V

>'::"""--+-"-j- - - 0 VOUT· +'ov
2mA

paR R2· 3.6kH AND

A3·6.4kB,
Vour· 1.5826V Z

A'

IF V

1.SkO
L-_ _ _ _ _ _ _

~

z

OR

6.4V, THEN

Vour" +10V

Vz

B.4V

ABSOLUTE VALUE CIRCUIT

A3

'Okll

A.

10kn

A5

DESIGN EQUATIONS

10ka

POSITIVE INPUT

D'
FD3333

1.

VA" 0, 02 OFF, D1 ON

2.

EO.(-E~N,R3).(1f)"EIN~

3. With R1 .. R3
EIN

R1

= R4"

R6: EO" EIN

4. Vas error Included: EO" EIN + 2Vos

10kn
D2

F03333

RZ
10kn

NEGATIVE INPUT
1. 01 OFF, 02 ON

2.

-~I,N.~ + R3 ~~4

3,

EO·VA

4,

With R3 .. R4" AS: EO" 1.5VA

_(R2~~R(~; ~~3(!·6~4~IN

5,

EO"

6.

With R1 .. R2 .. R3 .. R4: EO" -EIN

7.

VOS error Included:
EO" -EIN + 1.6VOS2 - O.5VOS1
For both Inputs: EO = + IEINI

8.

5·24

(1+R32~4)

1/86, Rev. A

-----------I~ OP·02 GENERAL·PURPOSE OPERATIONAL AMPLIFIER

TYPICAL APPLICATIONS
DAC·08 OUTPUT AMPLIFIER

INPUT/OUTPUT TABLE
B1 B2 B3 B4 B5 BI B7 B8 10mA

6,DDOkO

CAe·os
6,OkO

1.992

FULL-SCALE
-2 LSB

1.964 -9.920

HALF-SCALE
HSB

0

HALF-SCALE

0

0
0

0
0

0

HALF-SCALE
-LSB

FOR COMPl.EMENTARY OUTPUT (OPERATION ASA NEGATIVE
l.OGIC DAe) CONNECT NON.INVERTING INPUT OF OP·AMP TO

TO (PIN 2), CONNECT 10 (PIN 4) TO GROUND.

EO

FULL·SCALE
-1 LSB

ZERO-SCALE
HSB
ZERO-SCALE

0

0

0

0

0

0

0

0

-9.960

1.008

-5.040

1.000

-5.000

0.992

-4.960

0.0008 -0.040
0

0

0.000

0.000

~
~

tt:
e2

~

g~

~

~

()

5·25

1/86, Rev. A

OP-04/0P-14
DUAL MATCHED HIGH-PERFORMANCE
OPERATIONAL AMPLIFIERS
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

pin-for-pin compatibility, ease of application, and low cost.
Key speCifications, such as Vos, los, Ie, CMRR, PSRR and
Avo, are guaranteed over the full operating temperature
range. Precision Monolithics' exclusive Silicon-Nitride "Triple
Passivation" process reduces "popcorn noise". A thermallysymmetrical input stage design provides low TCVos, TClos,
and insensitivity to output load conditions. This series is ideal
for upgrading existing designs where accuracy improvements are desired. For more stringent requirements, refer to
the OP-207, OP-220, or OP-221 dual-matched operational
amplifier data sheets.

Excellent DC Input Specifications
Matched Vos and CMRR
OP-14 Fits Standard 1458/1558 Sockets
Internally Compensated
Low Noise
Low Drift
Low Cost
00 CI +70 0 C and -55 0 CI +125 0 C Models
SIlicon-Nitride Passivation
Models with MIL-STD-883 Class B Processing
Available From Stock

PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE
TA ~ 25' C -:;::;::-;:::-=~H;;:E:::R,;,;,M;.:E,;,;,T;,,:IC~=_ _..;P:..;L::.:A:;;S:;.T:.:IC:.., OPERATING
vos
TO-99 TO-l00
DIP
DIP
TEMPERATURE
(mV)
0.75
0.75
20
2.0
5.0
50

a-PIN

l0-PIN

a-PIN

l4-PIN

8-PIN

OP14AJ'OP04AK' OP14AZ' OP04AY'
OP14EJ
OP14EZ OP04EY OP14EP
OP14J' OP04K' OP14Z' OP04Y'
OP14CJ OP04CK OP14CZ OP04CY OP14CP
OP14BJ'OP04BK' OP14BZ' OP04BY'
OP140J
OP14DZ OP040Y OP140P

RANGE
MIL
COM
MIL
COM
MIL
COM

-IN (A)

1

+IN (A)

2

SAL (A)

3

SAL (8)

5

-IN (8)

7

B

BAL (B)

8-PIN HERMETIC DIP
(Z-Sufflx) OP-14

"

14-PIN HERMETIC DIP
(Y-Sufflx)
OP-04

EPOXY MINI-DIP
(P-Sufflx) OP-14

OUT

v+

I~A1
9 OUT IBI

(A) 2

8 V+ (8)

A
-IN (A) 3

B

-+

-

v+

8

10 N.C.

'For devices processed In total compliance to MIL-STO-883, add /883 after
part number. Consult factory for 883 data sheet.
t All commercial and industnal temperature range parts are available With
burn~ln. For ordering information see 1986 Data Book, Section 2.

OUT

IAI~+

7 OUT IBI

-IN (A) 2

6 -IN (8)

7 -IN (8)
+IN (A) 3

+IN (A) 4

GENERAL DESCRIPTION
The OP-04/0P-14 series of dual general-purpose operational
amplifiers provides significant improvements over industrystandard 747 and 1458/1558 (OP-14) types while maintaining

S

v-

5 +IN (B)

6 +IN (8)

4

TO-100 (K-Sufflx)
OP-04

v-

TO-99 (J-Sufflx)
OP-14

SIMPLIFIED SCHEMATIC (Each Amplifier)
"01,02,03 AND 04 FORM A
THERMALLY CROSS·COUPLEO
OUAD 05,05 1,06 AND 06 1
COMPRISE A SIMILAR
THERMALLY CROSS·COUPLED
OUAD
-IN

o--------t:====---,

+IN

.9~

R1

RB

R3

BALANCE
BALANCE ACCESSIBLE WITH OP·04 IN 14·PIN HERMETIC DIP ONLY

5-26

1/86, Rev, A

--------1~

OP-04/0P-14 DUAL MATCHED HIGH-PERFORMANCE OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS

(Note 2)

Supply Voltage ................................... ±22V
Internal Power Dissipation (Note 1) ••....•...••.. 500mW
Differential Input Voltage .......................... ±30V
Input Voltage .•••••••...••..........•.••• Supply Voltage
Output Short-Circuit Duration ••••••.•.••••.••• Indefinite
Storage Temperature Range
J, K, Y, and Z Packages •.•••.•.••••. -65°C to +150°C
P Package ••••.•...•••••••••••••••. -65°C to +125°C
Lead Temperature Range (Soldering, 60 sec) •••••• 300°C
Operating Temperature Range
A, Plain, B-Suffix ................... -55°C to +125°C
E, C, D-Suffix .......................... O°C to +70°C
DICE Junction Temperature (Tj) •.•.••. -65°C to +150°C

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

14-Pin Hermetic DIP (Y)
OP-04

NOTES:

100'C

10.0mW/·C

TO-100 (K)
OP-04

80'C

71mWrC

TO-99 (J) OP-14

80'C

7.1mW/·C

8-Pin Hermetic DIP (Z)
OP-14

75'C

6.7mWrC

8-Pin Plastic DIP (P)
OP-14

36'C

5.6mW/·C

2.

1. See table for maximum ambient temperature rating and derating factor

MATCHING CHARACTERISTICS

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

Absolute maximum ratings apply to both packaged parts and DICE, unless
otherwise noted.

~

i-J-l

at Vs = ± 15V, TA = 25° C, unless otherwise noted.

I-<

J;..I.;

OP-04A OP-04E
OP-14A OP-14E
TYP

MIN

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage
Match

t;,Vos

Rs:S 20kn

Common-Mode Rejection
Aatio Match

t;,CMRR

VcM =±10V, As:S 100n

1--4

OP-040P-04C
OP-140P-14C

MAX

MIN

TYP

~
~

MAX

UNITS

0.3
94

mV

106

94

106

dB

OP-04A OP-04E
OP-14A OP-14E
SYMBOL

CONDITIONS

Input Offset Voltage
Match

t;,Vos

Rs :S20kn

Common-Mode Aejection
Ratio Match

t;,CMAA

VCM = ±10V, As:S 100n

MIN

90

ELECTRICAL CHARACTERISTICS (Each Amplifier)

~

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

As :S20kn

Input Offset Current

los

0.5

Input Bias Current

I.

18

Input Resistance -

Differential-Mode

R'N

Input Voltage Range

IVA

(Note 3)

CMRA

VCM =±10V
As:S 20kn

Power Supply
Aejection Aatio

PSAA

Vs = ±5V to ±20V
As:S 20kn

Output Voltage Swing

Vo

ALe: 2kn

Common-Mode
Aejection Aatio

MIN

TYP

MAX

05

1.5

i-J-l

()

OP-04 OP-04C
OP-140P-14C
MIN

100

90

TYP

MAX

UNITS

15

mV

100

dB

at Vs = ± 15V, TA = 25° C, unless otherwise noted.

OP-04A10P-14A
PARAMETER

TYP

MAX

0.3

075

OP-04/0P-14
MIN

50

TYP

20

OP-04B/OP-14B

MAX

MIN

75

TYP

MAX

UNITS

3

5

mV

25

nA

100

nA

30

135

2.0

7.5

±10

±13

±10

±13

±10

±13

V

85

100

80

95

70

85

dB

60

10
±12

~

0
1--4

MATCHING CHARACTERISTICS at Vs= ±15V, -55°C::5 T A ::5 +125°C for OP-04A, OP-14A, OP-04 and OP-14,
O°C::5 TA::5 70°C for OP-04E, OP-14E, OP-04C and OP-14C, unless otherwise noted.

PARAMETER

~

30
±12

±13

5-27

Mn

±13

100

100
±12

±13

150

~VIV

V

1/86, Rev. A

~ OP-04/0P-14 DUAL MATCHED HIGH-PERFORMANCE OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS (Each Amplifier)

at Vs

= ±15V. TA = 25°C. unless otherwise noted. (Continued)

OP-04A10P-14A
PARAMETER

SYMBOL

CONDITIONS

Large-Signal
Voltage Gam

Ava

RL2: 2kO
Va = ±10V

Power Consumption
INote2)

Pd

Vo=OV

Input Noise Voltage

ene-e

0.1 Hz to 10Hz

Input Noise Voltage
Density

en

10= 10Hz
10 = 100Hz
10 = 1000Hz

Input Noise Current

i ne-e

0.1 Hz to 10Hz

Input Noise Current
Density

in

10= 10Hz
10 = 100Hz
10= 1000Hz

Channel Separation

CS

Slew Rate I Note 1 )

MIN

TYP

100

250

MAX

50

50

SR

Av=+1, V,N=50mV p_p
RL = 2kO, C L = 50pF

12.8

12.8

pAe-~

1.4
0.7
0.4

1.4
0.7
0.4

1.4
0.7
0.4

pAiy'HZ

100

260

0.25

1.0
350

CONDITIONS

Input Offset Voltage

Vas

RsS 20kO

Average Input Offset
Voltage Drift I Note 1)

TCVos

Rs= 500

80
0.5

0.25

8

4

1.3

1.0

260

at Vs

MIN

los
TClos

Input Bias Current

Ie

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VcM =±10V
Rs S20kO

Power Supply
Rejection Ratio

PSRR

Vs = ±5V to ±20V
RsS 20kO

Large-Signal
Voltage Gain

Ava

RL2: 2kO
Vo =±10V

Output Voltage Swing

Vo

RL2: 2kO

0.5

VII's
kHz

1.3

MHz

260

350

ns

10

5

10

%

= ±15V. -55° C:5 TA:5 +125° C. unless otherwise noted.

TYP

MAX

0.4

1.5

OP-04/0P-14
MIN

7.5

120

30

60

OP-04B/OP-14B

TYP

MAX

1.2

3

4

10

8

20

I'V/oC

10

10

50

nA

15

250

70

500

pA/oC

40

100

50

200

nA

10

Input Offset Current
Average Input Olfset
Current Drift INote 1)

dB

350

10

OP-04A10P-14A
SYMBOL

mW

12.8

ELECTRICAL CHARACTERISTICS'(Each Amplifier)
PARAMETER

90

nVly'HZ

1.3

OS

50

I'V~.~

1.0

Overshoot I Note 1)

90

V/mV

25
22
21

8

Av = +1. V'N = 50mV Pop
RL = 2kO. C L = 50pF

200

UNITS

0.65

4

tr

25

MAX

25
22
21

Va = 20V p_p

Rlsetime (Note 1)

TYP

0.65

0.5

AVCL = +1.0

200

MIN

25
22
21

0.25

BW

OP-04B/OP-14B

MAX

0.65

RL = 2kO.
C L = 100pF

Closed-Loop
Bandwidth INote 4)

TYP

50

90

100

Large-Signal
Bandwidth INote 1)

OP-04/0P-14
MIN

MIN

TYP

MAX

UNITS
mV

±10

±13

±10

±13

±10

±13

V

80

100

80

95

70

85

dB

10

30

60

100

100

150

J.lVIV

50

100

25

60

25

60

V/mV

±12

±13

±12

±13

±10

±13

V

NOTES:
1. Sample tested.
2. Power dissipation per amplifier.
3 Guaranteed by input bias current
4. Guaranteed by maximum rJsetlme.

5-28

1/86, Rev. A

~ OP-04/0P-14 DUAL MATCHED HIGH-PERFORMANCE OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS (Each Amplifier) at Vs = ±15V, TA = 25° C, unless otherwise noted.
OP-04E/OP-14E
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs" 20kO

Input Offset Current

los

Input Bias Current

Ie

Input ResistanceDifferential-Mode

R'N

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio

PSRR

Output Voltage Swing

MIN

TYP

MAX

0.3

0.75

OP-04C/OP-14C
MIN

TYP

MAX

OP-04D/OP-14D
MIN

TYP

(Note 3)

VCM =±10V
Rs" 20kO

UNITS

mV

0.5
18

MAX

50

20

75

30

25

nA

100

nA

2.0

7.5

±10

±13

±10

±13

±10

±13

V

85

100

80

95

70

85

dB

Vs = ±5V to ±20V

1.35

10

Rs" 20kO

MO

30

60

100

100

150

!'V/v

Vo

RL ", 2kO

±12

±13

±12

±13

±12

±13

V

Large-Signal
Voltage Gal n

Avo

R L ", 2kO
Vo =±10V

100

250

50

200

25

150

V/mV

Power Consumption
(Note 2)

Pd

Vo=OV

Input Noise Voltage

en~e

0.1 Hz to 10Hz

Input Noise Voltage
Density

en

fo= 10Hz
fo = 100Hz
fo= 1000Hz

Input Noise Current

Ine' e

0.1 Hz to 10Hz

Input Noise Current
Density

In

fo= 10Hz
fo= 100Hz
fo= 1000Hz

Channel Separation

CS

Slew Rate (Note 1)

SR

Large-Signal
Bandwidth (Note 1 )

,......

50

90

Vo= 20Vp_p

0.25

50

50

90

mW

0.65

!,Ve-e

25
22
21

25
22
21

25
22
21

nV/v"HZ

12.8

12.8

12.8

pA e' e

1.4
0.7
0.4

1.4
0.7
0.4

1.4
0.7
0.4

pA/v"HZ

100
0.5

0.25

Closed-Loop
Bandwidth (Note 4)

BW

AVCL = +1

Risetime (Note 1)

t,

Av = +1, V,N = 50mV
RL = 2kO, C L = 50pF

260

350

Overshoot (Note 1)

OS

Av= +1, V,N = 50mV
RL = 2kO, C L = 50pF

5

10

1.3

08

80
0.5

0.25

8

4

13

0.8

260

350
10

dB
0.5

V/!'s
kHz

1.3
260

MHz
350

ns

10

%

NOTES:
1. Sample tested.
2. Power dissipation per amplifier.
3. Guaranteed by input bias current.
4. Guaranteed by maximum risetime.

5-29

~

::s
~
~

0.65

4
0.8

90

0.65

100
RL = 2kO, CL= 100pF

~
~

1/86, Rev. A

~0

,......

~

~

--------l~ OP-04/0P-14 DUAL MATCHED HIGH-PERFORMANCE OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS (Each Amplifier) at Vs = ±15V, 0° C :5 TA:5 +70° C, unless otherwise noted.
OP-04E/OP-14E
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

RsS20kO

Average Input Offset
Voltage Drift (Note 1)

TCVos

Rs= 500

Input Offset Current

los

Average Input Offset
Current Drift (Note 1)

MIN

TYP

MAX

0.4

OP-04C/OP-14C
MIN

OP-04D/OP-14D

TYP

MAX

TYP

MAX

UNITS

1.5

1.2

3

3

6

mV

8

4

10

8

20

p.V/oC

10

2

10

10

50

nA

MIN

TClos

7.5

120

15

250

70

500

pA/oC

Input Bias Current

19

30

60

40

100

50

200

nA

Input Voltage Range

IVR

Common-Mode

CMRR

VCM =±10V
RsS 20kO

Power Supply
Rejection Ratto

PSRR

Vs = ±5V to ±20V
RsS 20kO

Large-Signal
Voltage Gain

Ava

Rejection Ratio

Output Voltage Swing

RL~

2kO
Vo= ±10V
RL~

Va

2kO

±10

±13

±10

±13

±10

±13

V

80

100

80

95

70

85

dB

10

50
±12

30

60

100

100

150

p.VN

100

25

60

15

25

VlmV

±13

±12

±13

±10

±13

V

NOTES:
1. Sample tested.
2. Power dissipation per amplifier.
3. Guaranteed by design.

BURN-IN CIRCUIT (1/2 of OP-04, OP-14)

OFFSET ADJUST CIRCUIT
V+

V+

+20V

-

~
~
-20V

vPIN NUMBERS ARE FOR "Y" PACKAGE ONLY

5-30

1/86, Rev. A

_ _ _ _ _-;~ OP-04/0P-14 DUAL MATCHED HIGH-PERFORMANCE OPERATIONAL AMPLIFIERS

DICE CHARACTERISTICS
OP-14

OP-04

DIE SIZE 0.080 x 0.050 Inch, 4000 sq. mils
(2.03 x 1.27 mm, 2.58 sq. mm)
For additional DICE information refer to
1986 Data Book, Section 2.
1. INVERTING INPUT (A)
2. NONINVERTING INPUT (A)

8.
9.
10.
11.
12.
13.
14.

3. BALANCE (A)
4. V5. BALANCE (B)
6. NONINVERTING INPUT (B)
7. INVERTING INPUT (B)

BALANCE (B)
V+
OUTPUT (B)
V+
OUTPUT (A)
V+
BALANCE (A)

1. INVERTING INPUT (A)
2. NON INVERTING INPUT (A)
3. BALANCE (A)
4. V5. BALANCE (B)
6. NON INVERTING INPUT (B)
7. INVERTING INPUT (B)

8. BALANCE (B)
9. V+(B)
10. OUTPUT (B)
11. NO CONNECTIONS
12. OUTPUT (A)
13. V+(A)
14. BALANCE (A)

NOTE: 9. 11 and 13 are Internally connected.

WAFER TEST LIMITS at Vs

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs$ 20kO

Input Offset Voltage
Match

tJ.Vos

Rs$ 20kO

Input Offset Current

los

Input Bias Current

I.

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

~
~

= ± 15V, TA =25°C, unless otherwise noted.

CMRR

OP-04G
OP-14G

OP-14GR

LIMIT

LIMIT

LIMIT

UNITS

6

mVMAX
mVMAX

Rs$ 20kO
VCM~±10V

tJ.CMRR

Power Supply
Rejection Ratio

PSRR

Vs ~ ±5V to ±20V
Rs$ 20kO

Output Voltage SWing

Vo

Large-Signal
Voltage Gain

1:.1...

0.75

VCM~±10V

Common-Mode Rejection
Ratio Match

......

OP-04N
OP-14N

200

nAMAX

50

75

500

nA MAX

±10

±10

±10

VMIN

85

80

70

dBMIN

94

94

60

100

150

!J.VIV MAX

RL " 10kO
RL " 2kO

±12
±12

±12
±12

±12
±10

VMIN

Avo

RL" 2kO
Vo~ ±10V

100

50

25

V/mV MIN

Power Consumption
(Both Amplifiers)

Pd

VOUT~ 0

170

170

180

mWMAX

Channel Separation

CS

100

100

Rs$1000

dBMIN

dB MIN

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal Yield loss, Yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ±15V,

PARAMETER

SYMBOL

CONDITIONS

TA

= 25° C, unless otherwise noted.

OP-04N
OP-14N

OP-04G
OP-14G

OP-14GR

TYPICAL

TYPICAL

TYPICAL

UNITS

200

200

200

ns

Av~+1

Rlsetime

tr

Overshoot

OS

Slew Rate

SR

V'N~

50mV
RL ~ 2kll
C L ~ 50pF
Av~+1

V'N~

50mV
RL ~ 2kll
C L ~ 50pF

%

RL ~ 2kO
C L ~ 100pF

0.25

5·31

0.25

VII'S MIN

1/86, Rev. A

::s
~
~

~

0
......

t:
~

0

--------1!EMD

OP-04/0P-14 DUAL MATCHED HIGH-PERFORMANCE OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (Each Amplifier)
UNTRIMMED OFFSET VOLTAGE
VI TEMPERATURE

INPUT OFFSET CURRENT
VI TEMPERATURE

INPUT BIAS CURRENT
VI TEMPERATURE

2.6

70
60

OP.(J4
OP·14

'\

'\

OP.(J4C
OP·14C

'\
OP.(J4A

.......

~~~= OP·I'A

.......

I,

i'-

I'...

~~E

-20

+20
+80
+100
TEMPERATURE rC}

+140

-80

OPEN-LOOP GAIN VB
POWER SUPPLY VOLTAGE
300

-

-20

+20
+60
+100
TEMPERATURE rC}

I"'-....

"

'00

so

IL

OP.CJ4A & OP-04E

OP·14A 81 OP,'4E

OP·'4

r--

OP~~P')4A
+140

W~!U

TA·2lfC

111111111

'00

\

OP·0481
OP.(J4C

0P·,4&
OP·14C

op.,. "OP·14C \

60

70

60

o

OP-04

/

+20
+80
+100
TEMPERATURE (OCI

110

70

o

-20

-

r-.

CMRR VB FREQUENCY

;:~':2&"C

1IIIIIIrrnI

J.lll[l~

OP-04C
OP·14C

CP.(J4E
OP·,4E

-60

+'40

11111111111

110

"- .......
.......

o

'20

OP·04A 81 QP.()4E
OP 14A 81 OP·14E

TA "2SOC

r-...

'0

17"

PSRR VB FREQUENCY
'20

RL" 10k

t-t-

OP.04A
OP-14A

o
-60

CP·l'

~I-

OP·04E

I'-..

I....

OP.Q4C - OP.o4

r--.~

:t6

:1:10

:1:1&

'20

POWER SUPPLY (VOLTS)

1000.

INPUT SPOT NOISE
VOLTAGE VI FREQUENCY

60

,

60

\

I\.

60

'0

100

1k

'Ok

FREQUENCY (Hz)

'00k

'0

100
1k
FREQUENCY (Hz)

'Ok

'OOk

INPUT WIDEBAND NOISE VI
BANDWIDTH (0.1 Hz TO
FREQUENCY INDICATED)

INPUT SPOT NOISE
CURRENTvsFREQUENCY

'0

=
-

~

~IOO

. '0_,,_

Vs '" t15V

TA·2&"C

/

I
~

~
~

0.0'

0.'0

'.0

'0

FREQUENCY (Hz)

'00

'000

O"OL.,O""J.WO"','-OI.l..l.OWJ''',O.uJ.ww...I.J..wJIIL..I.~
FREQUENCY (Hz)

5·32

0,

,
0.,

1,0
10
BANOWIDTH (kHz)

'00

1/86, Rev. A

------I~ OP-04/0P-14 DUAL MATCHED HIGH-PERFORMANCE OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (Each Amplifier)

OPEN-LOOP GAIN
vs TEMPERATURE

,

300

,

'20

I--

VS .. :t15V

250

~200
z

§"

./

'"

/

"

-

'00

" "1"'-

so

t\.
'\

" 150

CLOSED-LOOP RESPONSE
FOR VARIOUS
GAIN CONFIGURATIONS

OPEN-LOOP
FREQUENCY RESPONSE

40

'\

Vs ."'6V '
TA-2S·C

~

60

+20
+80
+'00
TEMPERATURE rCI

+140

1.0

10

100

lk

10k lOOk

1M

'0

10M

~

NEGATIVE SWING

'OM

FREQUENCY (Hz)

a

k- ~

10

~

[I

ffl

0:

\

'l
r--..

~

0.'

'.0

0.'

1000

'0

-60

-20

+20

POWER CONSUMPTION
VI TEMPERATURE
35
VS=±15V

.1

I

\

I

+100

-

OP·04A
OP 14A

~A· 25~C
Vs ='±15V

1--(1) VIN)PIN

3)1.

-,Lv. vL=+,L -

@V1N(PIN3),,+10mV, Vo

\

1

~\

OP·04E

,OP.14E
/

I'- "'"-

+140

OUTPUT SHORT-CIRCUIT
CURRENT vs TIME

50

TA" 2(fC

+60

TEMPERATURE (-C)

LOAD RESISTANCE TO GROUND (kn!

POWER CONSUMPTION
vs POWER SUPPLY
'000

~ 1.0

o

100

FREQUENCY (kHz)

'0

"

1M

"'

II!

~

"

lOOk

INPUT RESISTANCE
VI TEMPERATURE

!

\

1Otc.

POSITIVE SWING

Vs= :l:15V

TA""26"C

a:

1k

'00

, -'

TA'" 2SOC

'4

VS" ±16V

r

100

OUTPUT VOLTAGE vs
LOAD RESISTANCE

,.

11111111

'" '"

-20
0.1

FREQUENCY (Hz)

MAXIMUM UNDISTORTED
OUTPUTvsFREQUENCY

10

"-

-

-40

-20

VS'":t15V
TA '" 25°C -

"""

~ 100

....0

I

80

\
"'"- I--

-15V

,

"

r---....
..........

'.0

o

20

40

TOTAL SUPPLY VOLTAGE. V+ TOV- (VOLTS)

60

36
-60

-20

+20

+60

TEMPERATURE

5-33

+100

rei

.140

'6 o

TIME FROM OUTPUT BEING SHORTED (MINUTES)

1/86, Rev. A

op-os

INSTRUMENTATION
OPERATIONAL AMPLIFIER
Precision MOllolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•
•

voltage and bias current combined with very high levels of gain,
input impedance, CMRR, and PSRR.

Low Noise • • .. • • • • • • • • • • • •• 0.6/lVp-p Max, 0.1 to 10Hz
Low Drift VI. Temperature •••••••••••••• 0.5/lVloC Max
Low Drift VI. Time •••••••••••••••••• 0.2/lVlMonth Typ
Low Blal Current ....................... 2.0nA Max
High CMRR ............................ 114dB Min
High PSRR ............................ 100dB Min
High Gain ............................ 300,000 Min
High RIN Differential ...................... 30MCl Min
High RIN CM ........................... 200GCl Typ
Internally Compenseted •••••••••• Steble to 500pF Load
Fits 725, 108A and 741 Sockets
125° C Temperature Tested Dice

The OP-05 is a direct replacement in 725, 108A, and unnulled 741
sockets allowing instant system performance improvement
without redesign. The OP-05 is an excellent choice for a wide
variety of applications including strain gauge and thermocouple
bridges, high-gain active filters, buffers, integrators, and sampleand-hold amplifiers. For dual-matched versions, refer to the
OP-207 and OP-10 data sheets.

PIN CONNECTIONS

VOSTS~7V+

ORDERING INFORMATIONt
PACKAGE
HERMETIC
TA - 25°C
PLASTIC OPERATING
DIP
V08MAX TO...
DIP
TEMPERATURE
(mV)
8-PlN
8-PlN
14-PlN
a-PIN
RANGE
Of'OSAJ- OPOSAZ' Of'OSAY'
0.15
MIL
0.5
OPOSJ'
OPOSZ'
OP05Y'
MIL
0.5
OP05EJ
OPOSEZ
OP05EY OP-DSEP
COM
1.3
OP05CJ OPOSCZ OPOSCY OP05CP
COM

+

-IN 2

60UT

+IN 3

5 N,C.

•

v-leASE)

TO-99
(J-Sufflx)

EPOXY MINI-DIP
(P-Sufflx)

"

S-PIN HERMETIC DIP
(Z-Sufflx)

• For devices processed In total compliance to MIL-STO-883, add 1883 after
part number. Consult factory for 883 data sheet
tAli commercial and Industrial temperature range parts are available with
burn-In. For ordenng information see 1986 Data Book, Section 2.

14-PIN HERMETIC DIP
(Y -Suffix)"
GENERAL DESCRIPTION
'This package not
recommended for
new designs.

The OP-05 series of monolithic instrumentation operational
amplifiers combine excellent performance in low-signal-Ievel
applications with the simplicity of use of a fully-protected,
internally-compensated opamp. The OP-05 has low input offset

SIMPLIFIED SCHEMATIC

W~--~--------~------~----4I--------~--~------~------~------r-~---r--~
R.B

R1B

a1.
R9

OB

OUTPUT

~--4--t::.la.

NON

INVERTING
INPUT

027

INVERTINGo-...;.;..."...J'----~----I---........------_K
INPUT

025

R10

028

RB

V-~~----------------------~--~--~--~

5-34

______~__~--~

1/86, Rev, A

-----------l~ OP-OS INSTRUMENTATION OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS (Note 3)

NOTES:
1. See table for maximum ambient temperature rating and derating factor

Supply Voltage ................................... ±22V
Internal Power Dissipation (Note 1) .............. SOOmW
Differential Input Voltage .......................... ±30V
Input Voltage (Note 2) ............................ ±22V
Output Short-Circuit Duration ................. Indefinite
Storage Temperature Range
J, Y, and Z Packages ................ -6SoC to +1S0°C
P Package ..............•.......... -6SoC to +12SoC
Operating Temperature Range
OP-OSA,OP-OS ..................... -SsoC to +12SoC
OP-OSE, OP-OSC ....................... 0° C to + 70° C
Lead Temperature Range (Soldering, 60 sec) ...... 300°C
DICE Junction Temperature. . . . • . . . . .. -6So C to + 1S00 C

ELECTRICAL CHARACTERISTICS at Vs = ±1SV,

TA =

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

PACKAGE TYPE

SO'C

7.1mW/'C

14-Pin HermetiC DIP (VI

100'C

100mW/'C

S-Pin Hermetic DIP (ll

75'C

6.7mW/'C

S-Pln Plastic DIP (PI

36'C

5.6mW/'C

TO-99 (JI

2.
3.

For supply voltages less than ±22V, the absolute maximum Input voltage IS
equal to the supply voltage.
Absolute maximum ratings apply to both packaged parts and DICE, unless
otherWise noted

2So C, unless otherwise noted.
OP-05A

PARAMETER

SYMBOL

Input Offset Voltage

Vas

Long-Term Input Offset
Voltage Stability

AVos/Time

Input Offset Current

los

Input Bias Current

18

Input Noise Voltage (Note 2)

e np _p

en

Input Noise Voltage Density
(Note 2)

CONDITIONS

0.2

TYP

MAX

UNITS

015

0.2

0.5

mV

1.0

02

1.0

~V/Mo

2.0

1.0

2.S

nA

±2.0

±10

±3.0

nA

0.1 Hz to 10Hz

0.35

0.6

0.35

0.6

~Vp_p

fo= 10Hz
fo= 100Hz
fo= 1000Hz

10.3
10.0
9.6

IS.0
13.0
11.0

14

30

14

30

pA p_p

0.32
0.14
0.12

O.SO
0.23
0.17

0.32
0.14
0.12

O.SO
0.23
0.17

pAlJHZ

Input NOise Current DenSity
(Note 21

In

fo= 10Hz
fo= 100Hz
fo= 1000Hz

R 1NCM

007

MIN

0.7

O.IHz to 10Hz

Input ReslstanceCommon-Mode

MAX

(Note 3)

Input Voltage Range

IVR

Common-Mode Rejection Ratro

CMRR

VCM = ±13 5V

Power Supply Rejection Ratio

PSRR

Vs= ±3V to ±ISV

Large-Signal Voltage Gain

Ava

RL2: 2kO, Vo= ±10V
RL 2: 5000, Va = ±0.5V
Vs = ±3V (Note 31

Output Voltage SWing

Va

RL2: 10kO
RL2: 2kO
RL2: lkO

30

10.3
10.0
9.6

IS.0
13.0
110

nVlJHZ

60

MO

200

200

GO

±13.5 ±14.0

±135 ±140

126

300
150

20

114

126

500

200

500

500

150

500

4

10

±12.5 ±130
±12.0 ±12.S
±105 ±12.0

4

V
dB
10

~V/v

VlmV

±12.5 ±13.0
±120 ±12S
±10.5 ±12.0

V

SleWing Rate (Note 2)

SR

RL 2: 2kO

0.1

0.3

0.1

0.3

V/~s

Closed-Loop Bandwidth
(Note 2)

BW

AVCL = +10

0.4

0.6

04

0.6

MHz

Open-Loop Output ReSistance

Ro

Vo=O,lo=O

60

Pd

No load
Vs = ±3V, No load

90

Power Consumption
Offset Adjustment Range

Rp= 20kO

60
120
6

90
4

0
120
6

4

NOTES:
1. Long-term Input offset voltage stability refers to the averaged trend line
of Vosvs. Time over extended periods after the first 30 days of operation
Excluding the initial hour of operation, changes In Vas during the first 30

operating days are tYPically
Sample tested.
Guaranteed by design.

..

---'-------------------~---~

----_._-------

mW
mV

2.5~V.

Refer to typical performance curve.

1/86, Rev. A

5-35

-

>J,:I
,......
~

::sp..
~

<:

.....:I

~

0,......

~

>J,:I

SO

114

II
~

OP-05

TYP

±0.7

i np _p

R'N

MIN

(Note 1)

Input Noise Current
(Note 2)

Input Resistance Differential-Mode

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

~--~----

p..

0

-----------l~ OP-OS INSTRUMENTATION OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ±15V, -55 0 C S; TA S; +125 0 C, unless otherwise noted.
OP-OSA
PARAMETER

SYMBOL

Input Offset Voltage

Vos

Average Input Offset Voltage
Drift Without External Trim
With External Trim

TCVOS
TCVOSn

Input Offset Current

los

Average Input Offset Current
Drift

TCI OS

(Note 2)

TCl a

(Note 2)

MIN

CONDITIONS

(Note 2)
Rp= 20kO (Note 3)

MAX

TYP

MAX

UNITS

0.10

0.24

0.3

0.7

mV

0.3
0.2

0.9
0.5

0.7
0.3

2.0
1.0

/IoViOC

1.0

4.0

1.8

5.6

nA

50

pA/OC

±1

±4

±2

±6

nA

25

13

50

pAlOC

±13.0 ±13.5

Input Voltage Range

IVR

Common-Mode Rejection Ratio

CMRR

VCf" = ±13.0V

Power Supply Rejection Ratio

PSRR

VS=±3Vto±18V

Large-Signal Voltage Gain

AVO

RL ", 2kO, Vo= ±10V

MIN

25

Input Bias Current
Average Input Bias Current
Drift

OP-OS

TYP

110

123

200

400

5

±13.0 ±13.5
123

150

400

V/mV

±12.0 ±12.6

V

20

±12.0 ±12.6

Output Voltage Swing

V

110

dB
20

ELECTRICAL CHARACTERISTICS at Vs = ±15V, TA = 25 0 C, unless otherwise noted.

Op-osc

OP-OSE
PARAMETER

SYMBOL

Input Offset Voltage

Vos

Long-Term Input Offset
Voltage Stability

LlVOs/Time

Input Offset Current

CONDITIONS

MIN

(Notes 1. 2)

lOS

I nput Bias Current

TYP

MAX

TYP

MAX

UNITS

0.2

0.5

0.3

1.3

mV

0.3

1.5

0.4

2.0

1.2

3.8

1.8

6.0

nA

±1.2

±4.0

±1.8

±7.0

nA

MIN

Input Noise Voltage (Note 2)

0.1 Hz to 10Hz

0.35

0.6

0.38

0.65

Input Noise Voltage Density
(Note2)

fO= 10Hz
fo= 100Hz
fo= 1000Hz

10.3
10.0
9.6

18.0
13.0
11.0

10.5
10.2
9.8

20.0
13.5
11.5

nVly"RZ

14

30

15

35

pA p_p

0.32
0.14
0.12

0.80
0.23
0.17

0.35
0.15
0.13

0.90
0.27
0.18

pA/y"RZ

Input Noise Current (Note 2)

inp_p

0.1 Hz to 10Hz

Input Noise Current Density
(Note 2)

fO= 10Hz
fo= 100Hz
fo= 1000Hz

Input Resistance Differential-Made

(Note 3)

15

IVR

Common-Mode Rejection Ratio

CMRR

Power Supply Rejection Ratio

PSRR

VS=±3Vto±18V

AvO

RL ", 2kO. Vo= ±10V
RL ", 5000. Vo= ±0.5V
Vs = ±3V (Note 3)

Large-Signal Voltage Gain

MO

160

120

GO

±13.5 ±14.0

±13.0 ±14.0

110

123

100

120

200

500

120

400

150

500

100

400

5

RL ",10kO
RL ",2kO
RL ",lkO

Output Voltage Swing

8

33

Input ResistanceCommon-Mode
Input Voltage Range

50

20

±12.5 ±13.0
±12.0 ±12.8
±10.5 ±12.0

32

Slewing Rate (Note 2)

SR

RL ="' 2kO

0.1

0.3

0.1

0.3

BW

AveL = +1.0

0.4

0.6

0.4

0.6

Open-Loop Output Resistance

VO=O.IO=O

60

Power Consumption

No load
Vs = ±3V. No load

90

120

4

6

Offset Adjustment Range

Rp= 20kO

NOTE: See notes on previous page.

5-36

V/mV

±12.0 ±13.0
±".5 ±12.8
±12.0

Closed-Loop BandWidth
(Note 2)

4

V
dB

V

MHz

o

60
95

150

8
4

mW
mV

1/86, Rev. A

------------1~ OP-05 INSTRUMENTATION OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs= ±15V, Q·G:S TA:S +7Q·G, unless otherwise noted.
OP-05C

OP-05E
PARAMETER

SYMBOL

Input Offset Voltage

Vas

Average Input Offset Voltage
Drift Without External Trim
W,th External Trim

TCVos
TCVoSn

Input Offset Current

los

Average Input Offset Current

Drift

TClos

(Note 2)
Rp= 20kfl (Note 3)

(Note 2)

Input BIas Current

Ie

Average Input Bias Current
Drift

TCl a

Input Voltage Range

IVR

Common-Mode RejectIon RetIa

CMRR

VCM = ±13.0V

Power Supply Rejection Ratio

PSRR

Vs= ±3Vto ±18V

Large-SIgnal Voltage GaIn

Ava

RL 2:2kfl, Vo =±10V

Va

RL2: 2kfl

Output Voltage Swing

MIN

CONDITIONS

(Note 2)

TVP

MAX

TVP

MAX

UNITS

0.25

0.6

0.35

1.6

mV

0.7
0.2

2.0
0.6

1.3
0.4

4.5
1.5

p'vrc

1.4

53

2.0

8.0

nA

8

35

12

50

pAl"C

±15

±5.5

±2.2

±9.0

nA

13

35

18

50

pAloC

±13.0 ±13.5
107

123

180

450

MIN

±13.0 ±13.5

ffi

120

p.VN

......

100

400

VlmV

±11.0 ±126

V

::l
p..

32

±120 ±12.6

V

97

10

dB
51

~

~

NOTES:
1. Long-Term I nput Offset Voltage Stability refers to the averaged trend line
of Vas vs. Time over extended periods after the first 30 days of operation.
Excluding the initial hour of operation, changes In Vas during the first 30
operating days are typically 2.5p.V. Refer to typical performance curve.
2. Sample tested.
3. Guaranteed by design.

~

0......

~
~

C5

5-37

1/86, Rev. A

-----------l~ OP-05 INSTRUMENTATION OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS (125° C TESTED DICE AVAILABLE)
1. BALANCE
2. INVERTING INPUT
3. NONINVERTING INPUT

4. V5. NO CONNECTION
6. OUTPUT
7. V+
8. BALANCE
For additional DICE information refer to

1986 Data Book, Section 2.

DIE SIZE 0.100 X 0.051 Inch, 5100 sq. mils
(2.54 X 1.30 mm, 3.29 sq. mm)

WAFER TEST LIMITS

at Vs = ± 15V, TA = 25°C for OP·05N, OP·05G and OP·05GR devices; TA = 125°C for OP·05NT and

OP·05GT devices, unless otherwise noted.

OP·OSNT

OP·OSN

OP·OSGT

Op·OSG

OP·OSGR

LIMIT

LIMIT

LIMIT

LIMIT

LIMIT

UNITS

Vas

0.25

0.15

0.7

0.5

1.3

mVMAX

PARAMETER

SYMBOL CONDITIONS

Input Ollset Voltage
Input Offset Current

los

4.0

2.0

5.7

3.8

6.0

nAMAX

Input Bias Current

I.

±4

±2

±6

±4

±7

nAMAX

Input Resistance
Dillerential Mode

R'N

(Note 2)

±13.S

±13.0

±13.S

±13.0

VMIN

110

114

110

110

100

dBMIN

20

10

20

20

30

~VNMAX

RL = 10kO
RL = 2kO
RL =lkO

±12.0

±12.S
±12.0
±10.S

±12.0

±12.S
±12.0
±10.S

±12.0
±11.S

VMIN

RL = 2kO
Vo =±10V

200

200

150

200

120

VlmVMIN

±30

±30

±30

±30

±30

V MAX

120

150

mWMAX

IVR
CMRR

VCM = ± 13.5V at +25° C
VCM = ± 13.0 at + 125' C

Power Supply
Rejection Ratio

PSRR

Vs = ±3V to ±18V

Output Voltage Swing

Va

Large-Signal
Voltage Gai n

Avo

Dillerential Input
Voltage
Pd

MOMIN

±13.0

Input Voltage Range
Common-Mode
Rejection Ratio

Power Consumption

15

20

120

Vour = OV

NOTES:
1. For 25° C characteristics of NT & GT devices see N & G characteristics
respectively.

2. Guaranteed by design.

Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS

at Vs= ±15V. TA= +25°C, unless otherwise noted.

OP·OSNT

OP·OSN

OP·OSGT

Op·OSG

OP·OSGR

TYPICAL

TYPICAL

TYPICAL

TYPICAL

TYPICAL

UNITS

Rs" 500

0.3

0.3

0.7

0.7

1.2

~VloC

Rs" 500, Rp = 20kO

0.2

0.2

0.3

0.3

0.4

~VloC

8

8

12

pAloC

0.3

0.3

0.3

0.3

VI~s

0.6

06

0.6

0.6

MHz

PARAMETER

SYMBOL CONDITIONS

Average Input
Offset Voltage Drift

TCVos

Nulled Input
Offset Voltage Drift

TCVOSn

Average Input
Ollset Current Drift

TClos

Slew Rate

SR

RL" 2kO

0.3

Closed-Loop
Bandwidth

BW

AVCL = +1

0.6

5-38

1/86, Rev. A

-----------I~ OP-05 INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
TRIMMED OFFSET
VOLTAGE VB TEMPERATURE

16

1.0

30
Vas TRIMMED TO <5f.1V AT 25"C
NULLING POT" 20k11

-

I--

2 OP-05E
3 OP-05

I

4. OP-05C

I

(il
Q

I

1\ a>

12
OP-05C

/

1 OP-05A

~

-

JI5

I ~/
~\ I~'/

""-

I" IF
o

-60

..
reJ

100

~

1 DAY 1WK 1 MO 6MO

....

2

0

~
~

LONG-TERM

ERROR

DRIFT
ERROR BAND

-12

o

1

1

1

1

1

1

1

1

1

1

5 ~rr~rH~rH~-H~-Hffi*-H~~.!IIII

1

1

1.0

10

100

1

2

1k

1

~ ~20

V~.j'5)-

TIME (SEC)

~

:;

VB

....

i(

0:
0
....
1.0

~

ie

II I 11111

Ii!a:

Ii!a:

UNTRIMMED -550 C TO 125°C

a:
0
a: 0.1

~

~

a:
a:
0
a:

ffi

-

10

0.01

100

UIllJU1k I IlL

10k

":E
lOOk

MATCHED OR UNMATCHED SOURCE RESISTANCE (n)

11

12

0.5

I-+-+-+--+-:-i-/--"..
:E

TRIMMED _55°C TO 125°C

9

ill

~

,.~

8

:c

:: l1'eTO 711'e

*

UNTRIMMED 25°C

ffi

~

10

:;
!

-sSOc TO 125°C
Vs '" ±15V

7

(CURVES ARE SYMMETRICAL
ABOUT ZERO FOR .6.VOS < 0)

MAXIMUM ERROR
SOURCE RESISTANCE
QP-05A

6

1.0

10k

10

.§.

5

t

TIME (HOURS)

VB

4

TRIMMED
OFFSET VOLTAGE DRIFT

c:;

w

1

40

20

3

TIME (MONTHS)

~M~

~~~~~~~~~~~-W~.

0.1

100

TA "'70°C

8

~~~~~ffiOO-HffiOO-H~-H~

0,01

50

r-- ~
r-- ~~~
: 10 r-- "Ai 1 1
r-~
/1' 1 1 1
5 5
IMMERSEO
IN 7rfJC OIL BATH
~ r-- - OEVleE

-2

0.001

I--

20

~ 15

WARM·UP
DRIFT
BAND

....

RS = 5051
Vs "" ±15V

TA '" 25°C

~
o

~~~~{)~INE

-16

25

g

w

~

~ . .c..0,3p.V/mo TREND LINE

30

~

~

1

-

OFFSET VOLTAGE CHANGE
DUE TO THERMAL SHOCK

4f+ttttI1H-11tt1iiH'iffttm-'Tt_...,..,._...,..,._+!J!IIII

....

05E

TEMPERATURE ("C)

w

g

I--r

....o.r".-. •.•'j"O·2;v/mo TREND LlNEJ
~~~ 3JNlrno TREND LINEr¥~~~LINE
,
V'.
,
i

f-- c--

OP-05A

-50

OFFSET
VOLTAGE DRIFT WITH TIME

~

--;

0,01

TEMPERATURE

1 SEC 1 MIN 10MIN

~~~~~LINE~~~""~""""'':':~

OP-06

I-"

V

IS /

.--r-

-

/

\\

~

TYPICAL OFFSET
VOLTAGE STABILITY VB TIME

UNTRIMMED OFFSET
VOLTAGE VB TEMPERATURE

I

I~~' RI' ~R121

lk
10k
SOURCE RESISTANCE (.11)

5-39

lOOk

,.::>
X
""

UNTRIMMED -55"C ITJ

112~n

UNTRIMMED 25"C

01

,/

II 11111
TRIMMED -55"C TO 126°C
RS -R 1 -R 2

II 11111
0.0\00

lk

111111
10k

lOOk

MATCHED SOURCE RESISTANCE (n)

1/86, Rev. A

I

--

-----------I~ OP-G5 INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
MAXIMUM ERROR

INPUT BIAS CURRENT
VI TEMPERATURE

VI SOURCE RESISTANCE
10

1

OP..QSC
O'c TO 70'C

i

UNTRIMMED'c;.C to 70'C

12 '.0

UNTRIMMED 2&'C

iI
i

0.'

TRIMMED

~ ±16~

Vs

VS~±,15IV

C TO 70'

100

-

VS

~±16J

•

OP·08

,

::::::: ~OBIE /

........ ......

RS- R, - R2

0.0'

~I -

", "-

i-'

INPUT OFFSET CURRENT
VI TEMPERATURE

1111111

~loP..Q81

OP.QIA

./

r--.

IOP-,

1k
10k
MATCHED SOURCE RESISTANCE 1111

10Dk

o

-&0

so

100

INPUT WIDEBAND NOISE
VI BANDWIDTH (0.1 Hz
TO FREQUENCY INDICATED)
,0

=

;t; r-- ?
".

~
10

".,

,00

TEMPERATURa (Oe)

TEMPERATURE I'C)

OP-05 LOW FREQUENCY NOISE

OP.Q&C

::::---

0

-110

,21

......

1000

VOLTAGE NOISE DENSITY
VI FREQUENCY
EEEm!';C=;:=:;:====g

V. . :l:lIV
_ TA- +28"C

~

ISEE NOISE TEST CIRCUIT)

I---"
0.1
.00

1k

100k

10k

'0
FREQUENCY 1Hz)

BANDWIDTH 1Hz)

120
110

,00

.j

J~~I ~I~B"~ 'JJ.lll
Vs ':3~ ±'BV

j ~~~~

TO

11lilllL
-OP..QSC

aD

OPEN-LOOP GAIN VI
POWER SUPPLY VOLTAGE

CMRR VI FREQUENCY

PSRR VI FREQUENCY
,30
120

JJJlt
OP-O&C

110

~

J~~lllm
~

t~I~~\.J

1000

IIII

_I

100

/

~so

1\

70

so
&0

0.'

'.0

_I_

I- TA-+2B"C

TA - +2S'c
VCM -±13,&V

..........

~

10

lk

~

70

~
10
'00
FREQUENCY 1Hz)

....... ......

10k

80

'.0

,0

100

'k

FREQUENCY 1Hz)

5-40

,Ok

lOOk

o

o

:1:10
±IS
±S
POWER SUPPLY VOLTAGE (VOLTS)

±20

1/86, Rev. A

------------I~ OP-05 INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

OPEN-LOOP GAIN
VB FREQUENCY
120

r-

"

Z

~ 40

j

"

AV.1L
AV·

[\

0,'

,0

'00

'k

,Ok 'Oak

-20

'M 'OM

FREQUENCY IHI)

,J

'0 ,00

3.

::::: TA-+26"C

,

"-1'-

1k
10k
100k
FREQUENCY IH.)

r\

1\

o

1M

tL
t?

V

-20

-'0

V

"~ -10
~

! -20

,.

./

V

~

I
o

40
60
TOTAL SUPPLY VOLTAGE, V+ TO V- (VOL. TS)

20

20

VS·±16V TA",26°C._

-30
-30

i'

-20

-10

10

I

I

20

'M

10k
lOOk
FREaUENCV {H;cl

MAXIMUM OUTPUT VOLTAOE
VI LOAD RESISTANCE

-3.

AT IVDIFFI S 'ov, Ilel ~~~~ :g::~:f) ~

,,-

',0

'k

10M

S'.A IOP·OBCI ...,

/

11111
VS=±15V
TA" +250 C :

INPUT BIAS CURRENT VI
DIFFERENTIAL INPUT VOLTAGE

POWER CONSUMPTION
VI POWER SUPPLY
1000

~

AV· ,

-40

~

VS=±15V_
TA '" 26°C

80

I'\.

2B

I I

t--

AV.10~

~

~

'00
-

TA" +26"C

I"

80

~

VSI=±15~

i'--

MAXIMUM OUTPUT SWING
VB FREQUENCY

CLOSED-LOOP GAIN
VB FREQUENCY

2.

i

I-

Ii;

i2

II:

~

,B

a'"
.~

! ,.

!

"

~

"
z
>=

ffi

>
!

30
30

DIFFERENTIAL INPUT VOLTAGE {VOLTS}

~

::>

TAI.+2~d I

II I
I I I! III

VS.±'BV
VIN "±10mV

POSITIVE SWING

NEGATIVE SWING

1/1/

rl

i

•

100

1k

10k

1.0AC RESISTANCE TO GROUND (m

TYPICAL LOW·FREQUENCY NOISE TEST CIRCUIT·

TYPICAL OFFSET VOLTAGE TEST CIRCUIT

+15V

200kO

100.11

600

,oon

~----~---oVo

INPUT REFERRED NOISE '"

~ .. :;,~~~m

.

2oonV/em

~10Hz

FILTER

* OBSERVATION TIME LIMITED TO 10 SECONDS MAXIMUM.

5·41

1/86, Rev. A

----------l1£MD

OP-GS INSTRUMENTATION OPERATIONAL AMPLIFIER

FSET NULLING CIRCUIT

Offset stability can be degraded by stray thermoelectric
voltages arising from dissimilar metals at the contacts to the
input terminals. Best operation will be obtained when both
input contacts are maintained at the same temperature,
preferably close to the temperature of the device's package.

v+

TYPICAL APPLICATIONS

OUTPUT
INPUT

STABLE, HIGH-IMPEDANCE BUFFER
SR = O.25V/.usec

NOTE: PIN OUTS SHOWN FOR J, P, AND Z PACKAGES.

BURN-IN CIRCUIT
ZIN

::gg~~o9n

OUTPUT NOISE .. 0 35/JV P-P TYP

OUTPUT OFFSET .. O.2mV TVP

IN = ±1.0nA

OUTPUT

BANDWIDTH .. 600kHz

HIGH IMPEDANCE, HIGH COMMON-MODE REJECTION
INSTRUMENTATION AMPLIFIER
A4

2kn

RS
20k.l1

NOTE: PIN OUTS SHOWN FOR J, P, AND Z PACKAGES.

APPLICATIONS INFORMATION
OP-05 series devices may be fitted directly to 725 and
10S/10SA Series sockets with or without removal of external
compensation components. Additionally, the OP-05 may be
fitted to unnulled 741 series sockets. However, if conventional
741 nulling circuitry is in use, it should be modified or
removed to enable proper OP-05 operation. The OP-05
provides stable operation with load capacitance of up to
500pF and ±10V swings; larger capacitances should be
decoupled with a 500 resistor.

ADJUST R7 FOR MAXIMUM CMRR

5-42

1/86, Rev. A

OP-06
HIGH -GAIN INSTRUMENTATION
OPERATIONAL AMPLIFIER
Precision Monohthics Inc.

FEATURES
• Very High Voltage Gain ................ 1,OOOV/mV Min
• Low Offaet Voltage and Offaet Current
• Low Drift va. Temperature
(TCVoa) •......•.........•.......... O.8!,V/oC Max
• Low Input Voltage and Current Nolae
• Low Offset Voltage Drift with Time
• High Common-Mode Rejection ............ 120dB Typ
• High Power Supply Rejection •....•..••.... 2!'v/V Max
• Wide Supply Range ................... ±3.0V to ±22V
• MIL-STD-883 Proceaalng Available
• Slew Rate to .........•....................... 100V/!'a

Superior DC input characteristics include very low offset
voltage and current, extremely high open-loop gain, low 1/f
and wideband noise, and low "popcorn" noise. Low offset
voltage drift is improved by a nulling technique that optimizes
TeVos performance when Vos is nulled to zero. Very high
common-mode and power supply rejection enable accurate
performance in noisy environments.
Flexible external compensation provides wide-bandwidth
and high slew rate operation in high closed-loop gain
applications. Excellent long-term stability, and compatibility with MIL-STD-883 processing, make the OP-06 an
excellent choice for high-reliability applications. For example,
process control and aerospace applications; including strain
gauge and thermocouple amplifiers, low-noise audio amplifiers, and instrumentation amplifiers. The OP-06 is a direct
replacement for all 725 types providing superior DC and
noise performance plus the unique feature of complete Input
differential Yoltage and output short-circuit protection.

ORDERING INFORMATIONt
PACKAGE

TA =2SoC
Voa MAX
(mV)
0.2
0.2
0.5
0.5

1.3
1.3

TO-99
a-PIN

HERMETIC
DIP
a-PIN

OPERATING
TEMPERATURE
RANGE

OP06EJ
OP06AJ*
OP06FJ
OP06BJ*
OP06GJ
OP06CJ*

OP06EZ
OP06AZ*
OP06FZ
OP06BZ*
OP06GZ
OP06CZ*

COM
MIL
COM
MIL
COM
MIL

PIN CONNECTIONS

VOSeTn7V+

·For devices processed In total compliance to MIL-STD-BB3, add IBB3 after
part number. Consult factory for BB3 data sheet.
t All commercial and Industrial temperature range parts are available with
burn-In. For ordering Information see 19B6 Data Book, Section 2.

-IN 2

+

6 OUT

+IN 3

5 COMP

•

V-leASE)

GENERAL DESCRIPTION
The OP-06 monolithic instrumentation operational amplifier
is designed for accurate high-gain amplification of low level
signals. High common-mode rejection reduces Signal
degradation when large common-mode voltages are present.

I-PIN HERMETIC DIP
(Z-SuHlx)

TO-99 (J-Sufflx)

SIMPLIFIED SCHEMATIC

R,9""
_ _-

.....--0 OUTPUT

R20**

R3

'-I--l-......r

026

*027,028, R21, R22, COM·
PRISE THE INPUT PROTEC·
TION CIRCUIT
R5

R6

Rl0

R11

R'5

R'6

R17

"'023,029, R,9, R20 COM·
PRISE THE OUTPUT PRO·
TECTION CIRCUIT.

V-

S·43

- - - _..•. _--, ' - _ . , - - - - - -

1/86, Rev. A

--------f~ OP-G6 HIGH-GAIN INSTRUMENTATION OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

NOTES:
1. See table for maximum ambient temperature rating and derating factor.

(Note 3)

Supply Voltage ••••••••••••••••••••••••••••••••••• ±22V
Internal Power Dissipation (Note 1) •••••••••••••• 500mW

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99(J)

SOOC

7.1 mW'o C

S-Pin Hermetic DIP (Z)

75°C

6.7mVrC

Differential Input Voltage •••••••••••••••••••••••••• ±30V
PACKAGE TYPE

Input Voltage (Note 2) ..•..•••..•••.•••...•••••••. ±22V
Output Short-Circuit Duration ••••••••••••••••• Indefinite
Storage Temperature Range. • • • • • • • • ••

-65' C to + 150' C

Operating Temperature Range
OP-06A, OP-06B, OP-06C ••••••••••• -55'C to +125'C

2. Forsupply voltages less than ±22V, the absolute maximum input voltage Is
equal to the supply voltage.
3. Absolute ratings apply to both DICE and packaged parts, unless otherwise
noted.

OP-06E, OP-06F, OP-06G ••••••••••••••• O'C to +70'C
Lead Temperature Range (Soldering, 60 sec) •••••• 300' C
DICE Junction Temperature •••••••••••

-65' C to + 150' C

ELECTRICAL CHARACTERISTICS at Vs = ±

15V, TA

= 25' C, unless otherwise noted.

OP-06A/E
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Rs S 2OkO (Note 2)

Input Offset Current
Input Bias Current

Input Noise Voltage
Density

Input Noise Current

Density
Input Resistance

MIN

OP 08B/F

OP-08C/G

o

TYP

MAX

0.06

los
Ie

MIN

TYP

MAX

MIN

TYP

MAX

UNITS

0.2

0.2

0.5

0.4

1.3

mV

0.3

2.0

0.75

5.0

2

13

nA

30

70

30

so

40

110

nA

en

fO= 10Hz (Note 1)
10= 100Hz (Note 1)
10= 1000Hz (Note 1)

90
8.0
7.0

15.0
9.0
7.5

9.0
8.0
7.0

15.0
9.0
7.5

9.0
8.0
7.0

15.0
9.0
7.5

nVl.JHz

In

10= 10Hz (Note 1)
10= 100Hz (Note 1)
10= 1000Hz (Note 1)

0.5
0.25
0.15

1.2
0.8
0.25

0.5
0.25
0.15

1.2
0.6
0.25

0.8
0.3
0.2

1.4
0.7
0.3

pAl.JHz

R'N

(Note 3)

0·8

1.8

0.7

1.8

05

1.5

MO

Large-Signal
Voltage Gain

Ava

Rl" 2kO,
Vo=±10V

1,000

3,000

1,000

3,000

500

3,000

VlmV

Output Voltage
Swing

Va

Rl " 10kO
Rl " 2kO
Rl " lkO

±12.5
±120
±11.0

±13.0
±128
±12.5

±12.5
±12.0
±11.0

±13.0
:1:12.8
±12.5

±12.0
±11.5

±13.0
±12.8
±12.0

V

Input Voltage Range

IVR

±13.5

:1:14.0

±13.5

±14.0

:1:13.5

±14.0

V

114

120

114

120

110

115

dB

CMRR

VCM=±13.5V
Rss2OkO

Power Supply
Rejection Ratio

PSRR

VS =±3Vto ±18V
Rs S2OkO

Power Consumption

Pd

Common-Mode

Rejection Ratio

Large-Signal
Voltage Gain

Ava

Rl ,,5000, (Note 3)
Vo=±0.5V
Vs=±3V

Power Consumption

Pd

Vs=±3V

100

0.5

2.0

10

5.0

2.0

10

pVN

90

120

90

120

110

150

mW

100

600

4

6

so

600

4

6

600

VlmV

4

mW

NOTES:
1. Sample tested.
2. Thermoelectric voltages generated by dissimilar metals at the contacts to
the input terminals can degrade drift performance. Both Sides of the
contacts should be kept at approximately the same temperature. All
temperature gradients should be minimized.
3. Guaranteed by design.

5-44

1/86, Rev. A

---------1~ OP-06 HIGH-GAIN INSTRUMENTATION OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs= ±15V, -55·C ~ TA~ +125·C, unless otherwise noted.
OP-06B

OP-06A
PARAMETER

MIN

TYP

MAX

0.08

RS = 500 (Note. 1. 2)

RS = 500 (Note. 2. 3)
Rp= 20kO

SYMBOL

CONDITIONS

Vos

Rs" 20kO (Note 2)

TCVos

TeVosn

MIN

OP-06e

TYP

MAX

0.28

0.3

0.3

0.8

0.2

0.25
0.8

4.0

22
40

60
120

MIN

TYP

MAX

UNITS

07

0.5

16

mV

07

2.0

14

45

~VI"C

0.6

028

1.0

0.5

15

10

06
2.0

40
180

2.0
3.0

25

90

14

150

pA/OC

25
45

70
180

35
45

110
180

nA

Input Offset Voltage
(Without external

trim)
Average Input Offset
Voltage Drift (Without external tnm)
Average Input Offset

Voltage Drift (With
external tnm)
Input Offset Current

Average Input Offset
Current Dnft
Input Bias Current
Common-Mode

Rejection Ratio
Power Supply

Rejection Ratio

lOS

TCIOS

20

18

CMRR

PSRR

VCM = ±13 5V
Rs" 20kO

109

112

109

95

110

Vs = ±3V to ±18V

dB

15

Rs" 20kO

AvO

TAMAX
TAMIN

Output Voltage

Swmg

1.000
700

3.500
2.000

1.000

3.500

400

3.200

700

1.800

300

1.700

±120

±12.6

±12.0

±126

±11.0

±12.6

ELECTRICAL CHARACTERISTICS at Vs= ±15V, Q·C ~

~VIV

SYMBOL

CONDITIONS

Vos

RS" 20kn (Note 2)

TCVos

TCVosn

MIN

V/mV

V

TA~ 7Q·C, unless otherwise noted.

OP-06E
PARAMETER

112

nA

VO =±10V;R L ?:2kO

Large-Signal
Voltage Gain

(Note 1)

15

OP-06F

TYP

MAX

008

RS = 500 (Note. 1. 2)

RS = 500 (Note. 2. 3)
Rp= 20kO

MIN

OP-06G

TYP

MAX

028

0.25

03

08

0.2

025
08

40

MIN

TYP

MAX

UNITS

06

0.5

1.6

mV

07

20

14

4.5

~V/OC

0.6

028

10

0.5

1.5

~V/OC

10

065
20

5.0
180

20
30

15
25

nA

90

14

150

pAlOC

80
180

35
45

110

Input Off.et Voltage
(Without external

trim)
Average Input Offset
Voltage Drift (Without external tnm)
Average Input Offset
Voltage Dnft (With
external trim)
Input Offset Current

lOS

Average Input Offset
Current Dnft
Input BIBS Current

Common-Mode
Rejection RatiO
Power Supply
Rejection Ratio

TCIOS

Output Voltage
Swing

20

18

CMRR

PSRR

VCM =±135V

Rs ,,20kO

109

22

60

40

120

112

Vs = ±3V to ±18V

109

10

Rs" 20kO

30
45

50

112

15

95

7.0

180

110

30

nA

dB

15

~VIV

VO=±10V. RL 2!2kO

Large-Signal
Voltage Gain

(Note 1)

AVO

TAMAX
TAMIN

1.000
800

3.500
2.000

1.000
800

3.500
1.800

400

3.200

300

1.700

±120

±126

±120

±126

±110

±126

V/mV

V

NOTES:
1.
2.

Sample tested.
Thermoelectric voltages generated by dissimilar metals at the contacts to
the Input terminals can degrade drift performance. Both sides of the

3

5-45

contacts should be kept at approximately the same temperature. All
temperature gradients should be minimized.
Guaranteed by Input bias current

1/86, Rev. A

---------llEHD

OP-06 HIGH-GAIN INSTRUMENTATION OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS
1.
2.
3.
4.
5.
6.

NULL
INVERTING INPUT
NONINVERTING INPUT
VCOMPENSATION
OUTPUT

7. V+
8. NUll

For additional DICE information refer to
1986 Data Book, Section 2.

DIE SIZE 0.094 X 0.050 Inch, 4700sq. mils
(2.39 X 1.27 mm, 3.03 sq. mm)

WAFER TEST LIMITS

at

Vs = ± 15V,

TA

= 25° C for OP-06N, OP-06G and OP-06GR devices; TA = 125° C for OP-06NT and

OP-06GT devices, unless otherwise noted. (Note 2)

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

RsS 20kO

Input Offset Current

los

Input Bias Current

18

Input Resistance
Differential Mode

R'N

Input Voltage Range

IVR

OP-06NT
LIMIT

OP-06N
LIMIT

OP-06GT
LIMIT

OP-06G
LIMIT

OP-06GR
LIMIT

UNITS

0.3

0.2

0.7

0.5

13

mVMAX

5

13

nA MAX

60

70

70

80

110

nAMAX

0.7

0.5

MOMIN

0.8

(Note II

Common-Mode
Rejection Ratio

CMRR

VCM =±13.5
RsS 20kO

Power Supply
Rejection Ratio

PSRR

Vs= ±3V to ±18V
RsS 20kO

±13.0

±13.5

±13.0

±13.5

±13.5

VMIN

108

114

108

114

110

dB MIN

10

p.VIV MAX

±12.0

±12.5
±12.0
±11.0

±12.0
±115

VMIN

6

Output Voltage Swing

Vo

RL"= 10kO
RL"= 2kO
RL"= lkO

±12.0

±12.5
±12.0
±11.0

Large-Signal
Voltage Gam

Avo

RL"= 2kO
Vo= ±10V

1000

1000

800

1000

500

VlmV MIN

±30

±30

±30

±30

±30

V MAX

120

150

mWMAX

Differential Input
Voltage
Power Consumption
(VOUT = OVI

Pd

120

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS

at

Vs = ±15V,

TA

= +25°C, unless otherwise noted.

OP-06NT
TYPICAL

OP-06N
TYPICAL

OP-06GT
TYPICAL

OP-06G
TYPICAL

OP-06GR
TYPICAL

UNITS

R s S500

0.3

0.3

0.7

0.7

1.4

p.VI'C

RsS 50kO
Rp= 20kO

0.2

0.2

0.28

0.28

0.5

p.V/'C

3

3

8

8

14

pA/'C

PARAMETER

SYMBOL

CONDITIONS

Average Input Offset
Voltage Drift

TCVos

Nulled Input Offset
Voltage Dnft

TCVoSn

Average Input Offset
Current Drift

TClos

NOTES:
I. Guaranteed by input bias current.
2. For +25'C specifications of OP-OSNT and OP-06GT, see OP-OSN and
OP-06G respectively.

5-46

1/86, Rev. A

---------I~ OP-08 HIGH-GAIN INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
SLEW RATE
USING RECOMMENDED
COMPENSATION NETWORKS

OPEN-LOOP RESPONSE FOR
VALUES OF COMPENSATION
ISO ,--....,..__-....,..--,----,---,
120

100

t--""~-"...2_----t---t--j

CLOSED-LOOP FREQUENCY
RESPONSE FOR VALUES OF
COMPENSATION

8Or----c:::::::r-r--r-T"""---,

Vs· :t1SV
TA-26'IC

8Oi--....- -.....;;::-t-+-',....;;:---j

10

1..0

!~ @~--~---..~~t-~~--;

~

0.1

c

0,01

"

~

'"

§ ~I---+--~~~+-~~--j
~ o~---~----.l-.~=+-+-"""'..-l

NO COMPENSATION NEEDED

-20~--~--+-~-+-+-~~

IIIIIIIIIIIINI
-30 1.':0--::::-----:I::-k--:I~Ok:---:1:-:00::-k--:-::.L...-:-!,OM

III

0.001

N

1.0

FREQUENCY (Hz)

FREQUENCY COMPENSATION
COMPENSATION VALUES

Avel

R1

C1

(0)

(j.lF)

10000

10k

50pF

1000

470

0.001

100

47

0.01

10

27

0.05

270

1

10

0.05

39

0.0015
0.02

COMPENSATION CIRCUIT (J or Z PACKAGE)

OFFSET VOLTAGE

TRIM
USE R3 (= 51.11) WITH
CAPACITIVE LOAD
R3

*THE COMPENSATION NETWORK (R1, ell SHOULD BE RE·
TURNED TO THE V-TERMINAL. IF THE NETWORK IS RE·
TURNED TO GROUND, SERIOUS DEGRADATION OF POWER
SUPPLY REJECTION PERFORMANCE WITH FREQUENCY
WILL OCCUR. SEe TYPICAL CURVES (PSRR vs FREQUENCY).

**THE TRIMMING POTENTIOMETER SHOULD BE 20krl. FOR
OPTIMUM NULLED OFFSET VOLTAGE DRIFT see TYPI·
CAL CURVES (TRIMMED OFFSET VOLTAGE DRIFT AS A
FUNCTION OF TRIMMING POTENTIOMETER)

5-47

I

C,

FREQUENCY
R, COMPENSATION

V- OR GND·

1/86, Rev. A

---------IIEMD

OP·06 HIGH·GAIN INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
TRIMMED OFFSET VOLTAGE
V8 TEMPERATURE
30

~

VOSTRIMMEDTO

!
>

~
o

~6"VAT25'~

NULLINO POT'" 20kn

1
1. OP.O~A

\

"

~

o

w
310

~ !)

I..'

~

",'I

~
~

w

"~
5
>

I

I

75

100

0.Q1

100

125

-75 -50

....
k-'

~
!!

f=:::::--

~

--

-

f-

-0.6

op·OeA -

OP·06B

-II

1+-1

°r 1- 0J_061

-25

25

50

o

125

130

I

Op 06B

I I

I

Rp .. 50kn
r- ~~
r-- ~n

1'0..

Vs

-75

-50

-25

Vs

75

100

125

II ~ll

~

110

-75 -50

1\

20

~~~ERNAL

1

III

75

100

125

4

lOOk

1M

NEGATIVE

i-'

1\

I

/1

\

/
i

I

vS,7,~~,·V
FREQUENCY (Hz)

50

II I
pU,J,U

~~'= ~5~6'

70

TA" 25°C

10k

25

MAXIMUM OUTPUT VOLTAGE
V8 LOAD RESISTANCE

5

1k

0

TEMPERATURE (OC)

3

I: .
80

100

-25

100
\

90

10

t--

'(

1. OP-06G
2. OP-OSA, OP-06B, OP-06E
OP-06F (OOC TO 70°C)
3. ALL GRADES

NO
COMPENSATION

~ ~

1

VS" ±16V

l"'-

I"'- ,..,..-}

I
I

PSRR VI FREQUENCY
(OP-06B, OP-06E)

90

70

~

~

,I ±3V

rc)

120

I

80

1000

lOa

.........

-..,.,

o

25
50
TEMPERATURE

~±l.V
1-""'"

~1P-06El

r- --

FREQUENCY

I
I
I

10

I

OP·06G

0.1

II

0.1

INPUT BIAS CURRENT
vs TEMPERATURE

~OS~

~

IOP.06G •

OP·06G

11111111

0.01

TIME (HOURS)

"- .LJ-

OP-OSA, OP-OSB, OP-06E, OP-OSF

120

III~[

-S
0,001

1

0,4
0.6
0.8
1.0
1.2
UNTRIMMED OFFSET VOLTAGE Vas (mV) (CURVES
ARE SYMMETRICAL ABOUT ZERO FOR VOS < 0)

V8

~
WARM·UP DRIFT ERROR BAND

Vs' ±16V=

02

CMRR

r-I
LONG TERM DRIFT
ERROR BAND
(NON·CUMMULATIVE)

60

V

.Y

06

-1.0

i

l"i-"

OFFSET CURRENT
VB TEMPERATURE

....
.R'i lOt'
Ap·2Okn

.............. 1--., I-

...

Vs "'±15V

TEMPERATURE (OC)

0

..... 1--" ~

11111111

g

~

son

10

to.•

~
~

OP-06E--/

Vs" ±16V

Rp" SkU

a:o

VOs~20~VATT"O " HOURS

3

2

-

RS"

1.0

~

"~

~
":ll: -3

- OP·OSA

TRIMMED OFFSET VOLTAGE
DRIFT AS A FUNCTION OF
TRIMMING POTENTIOMETER
(Rp) SIZE AND Vos

isfa

-

I-

~~~~R ~JJ~~I,I~s ~u ~~~ID~~~I~~.IOI

~w

w

a 25 50 75
TEMPERATURE (DC)

-25

0. 1

~

-I"- OP.OS,:--I
i I

o

'ill

-50

-Or

i

~\ /AV
"-\ //7
-75

-

- r- OP.06G

i

/ J
1/ il 1/
CD ~~ /
L I).!.!
/~

\

1.0

1

V

/

2. OP-06E
3. OP·06G
4. OP·O C

20

tt

VS .. :t15V

OFFSET VOLTAGE DRIFT
WITH TIME

OFFSET VOLTAGE
V8 TEMPERATURE

60

10
1
2
3.
4.
5

100
C1
C1
C1
C1
R2
C1
R2

1k
10k
FREQUENCY (Hz)

100k

1M

100

TA = 2SoC
Vs '" ±15V
VIN" ±10mV

I 11111

10k

1k
LOAD RESISTANCE TO GROUND (n)

0.OO1pF, R1 .. 47051 FROM PIN 5 TO V0.1pF, R1 '" 5R TO V0.OO1pF: R1 = 470R FROM PIN 5 TO GND
0.05pF, R1 = 100, C2 = OD2pF,
39.11 TO V0.05pF, R1 = 10.11, C2 = O.02pF,
39.11 TO GND

5·48

1/86, Rev. A

---------I~ OP·06 HIGH·GAIN INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
VOLTAGE NOISE DENSITY
VI FREQUENCY

INPUT WIDEBAND NOISE
VI BANDWIDTH

1000

NOISE FIGURE VI
SOURCE RESISTANCE
12

'00
;:: (O,Q1Hz TO FREQUENCY INDICATED)

R, • 470n, C, • o.oO'.F TO

_

~
~
oz

-

RS" 10kn

r-N.
RS" 0

_

III
'Hz

!

~
~

V

1.0

10Hz

10

0.'

'k

100

FREQUENCY (Hz)

OPEN·LOOP GAIN
TEMPERATURE

E

~

lk
10k
SOURCE RESISTANCE (n)

OPEN·LOOP GAIN VB
SUPPLY VOLTAGE

MAXIMUM OUTPUT SWING
VB FREQUENCY
'0

Ll

20

!: 16

~

,,12

II
J

II

1\
J

11

;

*

~ 4

106

,oS

L-_"-_"-_.l-_..!..._..!...--J
-80
-40
0
40
80
120
160
TEMPERATURE (OC)

0

IAT lVD,IFF
I.-

o

I\

0

,1, ~V. I

"30nA, OP-OSB/A/F
"40nA, OP-06G ONL~

1\

V

o

±S

±10

:1:15

±20

POWER CONSUMPTION
SUPPLY VOLTAGE
-30

'00

'0

'25

~

V

J 1
J 1
J J

v

-'0

20

01 FFERENTIAL INPUT VOLTAGE (VOLTS)

'M

~:: ;~;~-20

"- f"."
b-"

/

=---

t:::::::

30
30

TA " 25°C

I

i I I

'0

20
30
40
60
60
TOTAL SUPPLY VOLTAGE, V+ TO V- (VOLTS)

5·49

~~N"(~~5~1"

.1 .1

-10mV,_

1 1

VIN (PIN 3) '" +10mV, _
Vo '" -16V

20

'0

',000

lOOk

46

,

-10

~

JJ"'-.

OUTPUT SHORT·CIRCUIT
CURRENT

VB

1000

GAIN ..

'00

~

1k
10k
FREQUENCY (Hz)

t-

1/

-20

GAIN ..

SUPPLY VOLTAGE (VOLTS)

INPUT BIAS CURRENT VB
DIFFERENTIAL
INPUT VOLTAGE

V

GAIN - 10,000

GAIN'"

::>

V

~~

-, 0

lOOk

o

~
o

./"'"

r06~~§§

-30
-30

GAIN .. 1

E24
i.--'""'

~

-20

'00

lOOk

VA - 2SOC

~

z

10k

2.

'07

100Hz

1kHz

BANDWIDTH (Hz)

lk

VB

'07~m

Vs" ±15V

I--

Jd'1

~

TA " 26°C

10

VS" ±1SV

1.

~

TA" 26°C

I

1.0
1.0

ill

'0

---

Rs·60kn

ISii

v-

~

,0

'6

o

,

1 1 1 1-

TIME FROM OUTPUT BEING SHORTED (MINUTES)

1/86, Rev. A

---------I~ OP-06 HIGH-GAIN INSTRUMENTATION OPERATIONAL AMPLIFIER

GUARANTEED PERFORMANCE CHARACTERISTICS

l
~
g

.fa
~..
~

10

~

OP.()8B
, -sSOC TO 125°C

~
1.0

~

S

y UNTRIMMED -5S0c TO 125°C

0.1

ill

II IIIII U-

~

ill

W TRIMMED -66°C TO 12SOC

::E 0.01
100

~
c

."'
Y UNTRIMMED

0.1

II IIIII

II

lk
10k
SOURCE RESISTANCE, Rt {OJ

~

II lIill

0

II III

:e

tOOk

0.01
100

Wn ij'lutTtl1
lk

10k

tOOk

SOURCE RESISTANCE, Rt (n)

op·06G

OOc TO 7efC
Y UNTRIMMED OOc TO 7fPC

1.0

These graphs depict maximum error
referred to the Input as a function of
source resistance (R, I. Curves Ware
shown with Vos trimmed at +25'C and
include errors due to Vos and los over
the indicated temperature range. Curves
Y and Z plot maximum errors with Vos
not trimmed.

Z UNTRIMMED 2&oC

0.1

ill

W TRIMMED DoC TO 70°C

"" =t
X
"
"

~

"~

lk
10k
SOURCE RESISTANCE, Rt (0)

0:
0:

*.

/

!

0.1

10

~

.fil

i

1111
100

>-

~

0:
0:

W TRIMMED -55"C TO 126°C

::I 0.01

tOOk

: Z UNT~I~~~.~ 2.'C .

$

126°C

I IIIII

UNT~'MMED ----------.I-l
......
.....

18

24

15

mW

~

0......

~

>.I-l
p.,

COMPENSATION CIRCUITS

0
ALTERNATE
R2

STANDARD
R2

R1

R1
OUTPUT

R3

OUTPUT

R3

Cf;;' Rl ~lR2 Co

CS'" 100pF

Co '" 30pF

IMPROVES REJECTION OF POWER
SUPPLY BY A FACTOR OF TEN

OFFSET VOLTAGE TEST CIRCUIT

BURN-IN CIRCUIT

200kn

>--""'--0 Va

5-63

1/86, Rev. A

!fHD

OP-08 PRECISION LOW-INPUT-CURRENT OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ±15V for G Grade and Vs = ±20V for
otherwise noted.

Grade,

E

Ooe:5 TA:5

OP-OSE
PARAMETER

SYMBOL

Input Offset Voltage

Vos

Average Input
Offset Vollage
Drift

TCVos

Input Offset Current

los

Average Input
Offset Current
Drift

TClos

CONDITIONS

MIN

(Note 1)

(Note 1)

Input Blal CUrrent

Is

Input Voltage Range

IVR

Vs· ±15V

Common-Mode
Rejection Ratio

CMRR

VOM - ±13.5V

Power Supply
Rejection Ratio

PSRR

Vs-UVto±15V

+70 o e, unless

OP-OSG

TYP

MAX

TYP

MAX

UNITS

0.10

0.26

0.32

1.4

mV

0.50

2.5

1.5

10

p.V/'C

0.08

0.30

0.12

6.5

nA

0.50

2.5

2.0

50

pAl'C

1.0

2.6

1.4

8.5

nA

MIN

±13.5

±14.0

±13.5

±14.0

V

100

118

80

112

dB

2

10

3

100

/AVIV

RL~2kO,

Large-Signal
Voltage ' ,
,I
y, ,CS = 'OOO~""""
I

80

,. r-....

20

'20

240

b,j.,

INPUT BIAS CURRENT
AND INPUT OFFSET CURRENT
VB TEMPERATURE

OPEN-LOOP GAIN
VB SUPPLY VOLTAGE

'=2.·C

015

0'0

40

20

,.

005

90
10M

l- t- I--

-55°C

I-"'i:'"
100

--

"0

±tS

SUPPLY VOLTAGE (VOLTS)

5-86

'20

o

J""-.. ........

-55 -35 -15

5

25

46

65

85

105

125

TEMPERATURE (OC)

1/86, Rev. A

---------I~ OP-08 PRECISION LOW-INPUT-CURRENT OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

SUPPLY CURRENT
vs SUPPLY VOLTAGE

POWER SUPPLY
REJECTION RATIO (PSRR)
va FREQUENCY

SUPPLY CURRENT
vs TEMPERATURE

600

600 . - . . , - - , - , - - . . . , - ; - , - - - , - - , - - ,
NO LOAD

NO LOAD

500

<
..:;. 400
>-

~

a

300

;

-55':;'"

t

ffi

400

II-t;:;f-::t:+=+~~JII

200

I--+-+-t---+--+-t---+--+-I

~

::>

"~

125°C

~

iii

<

25"C

l---- r-

~

200

100

'5

±10

±15

±20

SUPPLY VOLTAGE (VOLTS)

o~~~-~~-=~~~~~
___
5
~

~

~

~

1~

-20 ' - - - ' - - - ' - - - ' - - - - ' - - - - '
100

1~

11<

10k

APPLICATIONS INFORMATION

100k

1M

10M

FREQUENCY (Hz)

TEMPERATURE (OC)

TYPICAL APPLICATION

The op-oa series has very low input offset and bias currents;
the user is cautioned that printed circuit board leakage
currents can produce significant errors, especially at high
board temperatures. Careful attention to board layout and
cleaning procedure is needed to take full advantage of the
oP-Qa performance. Board leakage is minimized byencircling
the input pins with a guard ring maintained at the same
potential as the inputs. This guard ring should be driven by a
low impedance source, such as an amplifier's output
or ground.

BILATERAL CURRENT SOURCE
R2
2Me!

Rl
4M51
R3

':"
4Me!

RS«A3

RS
lke!

RS

~'L

3Op'
R4
2Me!
'VE

RL
ER,

IF Rl=R3 AND R2=R4+R5 THEN

RaRs

IL IS INDEPENDENT OF VARIATIONS
IN RL

'LS!'':"

5-67

1/86, Rev. A

OP-09/0P-ll
QUAD MATCHED
741-TYPE OPERATIONAL AMPLIFIERS
Prcci sion Monolithics Inc.

fier is designed to have equal positive-going and negativegoing slew rates. This is an important consideration for good
audio system performance.

FEATURES
•
•
•
•
•
•
•
•
•
•
•

Guaranteed Vos ......•.............•..... SOOI'V Max
Guaranteed Matched CMRR ................ 94dB Min
Guaranteed Matched Vos .................. 7S0l'V Max
RC/RM4136 Direct Replacement (OP-09)
LM148/LM348 Direct Replacement (OP-11)
Low Noise
Silicon-Nitride PasslYatlon
Internal Frequency Compensation
Low Crossoyer Distortion
Continuous Short-Circuit Protection
Low Input Bias Current

Each of the four amplifiers has the proven OP-02 advantages
of low noise, low drift, and excellent long-term stability. Precision Monolithics' exclusive Silicon-Nitride "Triple Passivation" process reduces "popcorn noise", provides high reliability, and assures long-term stability of parameters.
The OP-09 and OP-11 are ideal for use in designs requiring
minimum space and cost while maintaining OP-02-type
performance.
OP-09's and OP-11's with processing per the requirements of
MIL-STD-883 are available. For dual-741-type versions, see
the OP-04/14 data sheet.

ORDERING INFORMATIONt
TA = 25°C
VosMAX
(mY)

HERMETIC
DIP
14-PIN

05

OP-09AY'
OP-llAY'

0.5

OP-09EY
OP-llEY

2.5

OP-09BY'
OP-llBY'

25

OP-09FY
OP-lIFY

5.0

OP-l1CY'

50

OP-lIGY

EPOXY
DIP
14·PIN

LCC

OPERATING
TEMPERATURE
RANGE

OPI1AAC/883
OP-lIEP

PIN CONNECTIONS
OP-09

MIL

OP-11

COM
MIL

OP-09FP
OP·l1FP

COM

OP-llGP

COM

MIL

14-PIN HERMETIC DIP (V-Suffix)
EPOXV DIP (P-Suffix)

'For devices processed In total compliance to MIL-STD-883, add 1883 after
part number. Consult factory for 883 data sheet
tAli commerCial and Industrial temperature range parts are available with
burn-In. For ordering information see 1986 Data Book, Section 2.

GENERAL DESCRIPTION
The OP-09 and OP-11 provide four matched 741-type operational amplifiers in asingle 14-pin DIP package. The OP-11 is
pin compatible with the LM148, LM348, RM4156, and HA4741
amplifiers. The OP-09 is pin compatible with the RM4136 and
RC4136. The amplifiers are matched for common-mode
rejection ratio and offset voltage which is very important in
designing instrumentation amplifiers. In addition, the ampli-

OP-11ARC/883
LCC
(RC-Sufflx)

"z

SIMPLIFIED SCHEMATIC (One 01 Four Amplifiers Is Shown)

(-lIN

(+)INo--t-------+---'

'----J--+--+-o OUTPUT

5-68

1/86, Rev. A

----------I~ OP-09/0P-11 QUAD MATCHED 741-TYPE OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS (Note 2)
Supply Voltage .•.••......•......••••..•.......... ±22V
OP-09GR and OP-11GR (Only) ...........•...... ±18V
Internal Power Dissipation (Note 1)
RC, V-Package ............................... 800mW
P-Package ................................... 500mW
Differential Input Voltage .............•.........•.. ±30V
Input Voltage .........•.....••.......•... Supply Voltage
Output Short-Circuit Duration ....•...•...... Continuous
(One Amplifier Only)
Storage Temperature Range
RC, V-Package ..................... -65·C to +150·C
P-Package ......................... -65·C to +125·C
Lead Temperature Range (Soldering, 60 sec) ...... 300· C
DICE Junction Temperature (Tj ) ••••• -65· C to + 150· C

OP-11A,OP-11B,
OP-11C,OP-11ARC ................. -55·C to +125·C
OP-11E, OP-11F, OP-11G ................ 0·Cto+70·C
NOTES:
1.

See table for maximum ambient temperature and derating factor.
MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

14-Pm Hermetic DIP (YI

70'C

10.0mW/'C

14-Pm Plastic DIP (PI

42'C

SmW/'C

LCC (RCI

70'C

7.8mW/'C

PACKAGE TYPE

2

Operating Temperature Range
OP-09A,OP-09B .........•......•.• -55·C to +125·C
OP-09E,OP-09F ....•.........•.•....... o·e to +70·e

Absolute maximum ratingsapply to both DICE and packaged parts. unlesa
otherwise noted.

~

~
j:.l..j

.....,

::l
p.,

MATCHING CHARACTERISTICS at Vs = ±15V, TA = +25·e, Rs S 100n, unless otherwise noted.
OP-09A, OP-09E
OP-11A,OP-11E
PARAMETER

SYMBOL

Input Offaet Voltage Match

~Vos

Common-Mode Rejection
Ratio Match

~CMRR

CONDITIONS

MIN

VcM =±12V
VcM =±12V

Q4

TYP

MAX

0.5

~

OP-09a, OP-09F
OP-11a, OP-11F
MIN

TVP

MAX

0.75

0.8

2.0

mV

20

1
120

20

I'VIV
dB

120

Q4

UNITS

O·C S TAS +70·e for OP-09E, OP-09F, OP-11 E and OP-11 F, RsS 100n. unless otherwise noted.

SYMBOL

Input Offset Voltage Match

~Vos

Common-Mode Rejection
Rallo Match

~CMRR

CONDITIONS

V cM =±12V
V cM =±12V

-

OP-09a, OP-09F
OP-11a,OP-11F

MIN

MIN

Q4

5-69

---~."-~

OP-09A, OP-09E
OP-11A,OP-11E
TVP

MAX

TYP

MAX

O.S

1.0

1.0

2.5

mV

3.2
110

20

3.2
110

20

I'ViV
dB

94

~

~

MATCHING CHARACTERISTICS at Vs= ±15V. -55·e S TAS +125·e for OP-09A, OP-09B, OP-11A and OP-11B,

PARAMETER

~

9

UNITS

1/86, Rev. A

~ OP·09/0P·11 QUAD MATCHED 741·TYPE OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS (Each Amplifier) at Vs = ±15V TA = 25°C, unless otherwise noted.
OP·09A1E
OP·11A1E
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

RsS tOkO

Input Offset Current
Input BIas Current

TYP

MAX

0.3

los

MAX

TYP

MAX

UNITS

0.5

0.6

2.5

t.2

5.0

mV

55

20

25

50

75

200

nA

Is

180

300

300

500

300

500

nA

Input ReSIstance
DifferentIal Mode

R'N

017

0.29

0.1

0.17

0.1

017

MO

Input Voltage Range

IVR

±12

±13

±12

±13

±12

±13

V

Common-Mode
Rejection Ratio

CMRR

VcM =±12V, RsS 10kO

100

120

100

120

70

100

dB

Power Supply
RejectIon Ratio

PSRR

Vs = ±5 to ±15V,
RsS 10kO

4

MIN

OP·11C/G

TYP

(Note3)

MIN

OP·09B/F
OP·11B/F

32

4

MIN

10

32

100

"VIV

Output Voltage Swing Vo

RL ,,2kO

±11

±13

±11

±13

±11

±13

V

Large-Signal Voltage
GaIn

Avo

RL S2kO, Vo =±10V

100

650

100

650

50

500

V/mV

Power Consumption
(Note 1)

Pd

Vo=OV

105
0.7

0.7

07

"VI!:~

18
14
12

18
14
12

18
14
12

nVl..;HZ

Input Noise Voltage

180

123

180

210

340

mW

enf:!-~

0.1 Hz to 10Hz

Input Noise Voltage
Density

en

fo= 10Hz
fo= 100Hz
fo= 1000Hz

Input Noise Current

'np-p

o 1Hz to 10Hz

Input NOIse Current
Density

In

fo= 10Hz
fo= 100Hz
fo= 1000Hz

Channel SeparatIon

CS

100

130

100

130

130

dB

Slew Rate (Note 2)

SR

07

10

0.7

1.0

0.7

10

VI"s

Va = 20V p_p

11

16

11

16

11

16

kHz

BW

AVCL =+1.0

2.4

3.0

2.4

30

2.4

3.0

MHz

Rlsetime (Note 2)

tr

Av = +1, V ,N = 50mV

Overshoot (Note 2)

OS

Large-Signal
Bandwidth (Note 2)
Closed-Loop
Bandwidth (Note 4)

17

17

17

pA~_~

1.8
1.5
1.2

1.8
1.5
1.2

1.8
15
12

pAl..;HZ

110

145

110

145

110

145

ns

15

25

15

25

15

25

%

NOTES:
1. Total dissipation for all four amplifiers in package
2. Sample tested.
3. Guaranteed by Input bIas current.
4. Guaranteed by ,"setlme.

5·70

1/86, Rev. A

---------l~ OP-09/0P-11 QUAD MATCHED 741-TYPE OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS (Each Amplifier) at Vs = ±15V, -55° C:'S TA:'S +125° C, unless otherwise noted.
OP-09B
OP-11B

OP-09A
OP-11A
PARAMETER

SYMBOL

Input Offset Voltage

Vas

Average Input Offset
Voltage Drift (Note 3) TCVos
Input Offset Current

MIN

CONDITIONS

Rs $10kn

los

Average Input Offset TCI
Current Drift (Note 3)
as
Input Bias Current

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VCM =±12V,

PSRR

Vs = ±5 to ±15V,
Rs $10kn

Ava

R L ;"2kn, Vo =±10V

Power Supply
Rejection Ratio

Large-Signal
Voltage Gain

Rs

$10kn

Output Voltage SWing Va
Power Consumption

TYP

MAX

UNITS

1.0

10

35

15

60

mV

10

40

15

40

04
20

MIN

jJ.V/'C

20

40

40

60

250

300

0.1

03

0.3

0.6

03

06

200

375

400

650

400

800

nA

nA

±12

±13

±12

±13

±12

±13

V

100

120

100

120

70

100

dB

32

32

10

100

50

250

50

250

25

100

V/mV

±11

±13

±11

±13

±11

±13

v

115

Vo=OV

(Note 1)

MAX

MAX

MIN

OP-11C

TYP

TYP

115

200

200

250

400

mW

ELECTRICAL CHARACTERISTICS (Each Amplifier) at Vs = ±15V, DOC:'S TA:'S +7DoC, unless otherwise noted.
OP-09E
OP-11E
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Average Input Offset
Voltage Drift

TCVos

Input Offset Current

los

MIN

TYP

MAX

Rs $10kn

0.4

Rs $10kn

Average Input Offset
Current Drift (Note 3) TClos
Input BIBS Current

Input Voltage Range
Common-Mode
Rejection Ratio

Power Supply
Rejection Ratio

Large-Signal
Voltage Gain

IVR
CMRR

VCM =±12V,

PSRR

Vs = ±5 to ±15V,
Rs $10kn

Rs

$10kn

Ava

Output Voltage Swing Va
Power Consumption

(Note 1)

OP-09F
OP-11F

Vo=OV

MIN

OP-11G

TYP

MAX

MIN

0.8

08

30

TYP

MAX

UNITS

1.5

60

mV

2.0

10

4.0

15

4.0

14

30

40

60

250

300

nA

01

03

03

06

0.3

06

nA/'C

200

350

400

550

400

600

nA

±12

±13

±12

±13

±12

±13

V

100

120

100

120

70

100

dB

4

32

32

10

100

50

250

50

250

25

100

V/mV

±11

±13

±11

±13

±11

±13

V

115

200

115

200

250

400

mW

NOTES:
Total diSSipation for all four amplifiers in package.

2

Sample tested.

3.

Guaranteed but not tested.

5-71

1/86, Rev_ A

---------I1fMD
DICE CHARACTERISTICS

OP-08l0P-11 QUAD MATCHED 741-TYPE OPERATIONAL AMPLIFIERS

(125° C

TESTED DICE AVAILABLE)

OP-09

OP-11
1.
2.
3.
4.
5.
6.

1.
2.
3.
4.
5.
6.
7.

OUTPUT (A)
INVERTING INPUT (A)
NONINVERTING INPUT (A)
V+
NON INVERTING INPUT (B)
INVERTING INPUT (B)
OUTPUT (B)
B. OUTPUT (C)
9. INVERTING INPUT (C)
10. NONINVERTING INPUT (C)
11. V12. NON INVERTING INPUT (0)
13. INVERTING INPUT (D)
14. OUTPUT (D)
15. V+

INVERTING INPUT (A)
NONINVERTING INPUT (A)
OUTPUT (A)
OUTPUT (B)
NONINVERTING INPUT (B)
INVERTING INPUT (B)

7. VB. INVERTING INPUT (C)
9. NON INVERTING INPUT (C)
10. OUTPUT (C)

11.
12.
13.
14.
15.

V+
OUTPUT (0)
NONINVERTING INPUT (0)
INVERTING INPUT (D)
V+

DIE SIZE 0.085 X 0.070 Inch, 5950 sq. mill
(2.16 X 1.78 mm, 3.84 sq. mm)

DIE SIZE 0.085 X 0.070 Inch, 5950 sq. mill
(2.16 X 1.78 mm, 3.84 sq. mm)

NOTE:
Ellher or both V+ pads may be used without any change In performance.

For addilional DICE Information refer to
1986 Data Book, Section 2.

WAFER TEST LIMITS at Vs= ±15V, T A =25°C for OP·09/11N, OP·09/11G and OP·09/11GR devices; TA=125°C for
OP·09/11 NT and OP·09/11 GT devices, unless otherwise noted.

PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

Input Offset Current

los

Input Bias Current

Ie

Input Voltage Range

IVR

OP-09NT
OP-11NT

OP-09N
OP-11N

OP-09GT
OP-11GT

OP-11G

OP-09GR
OP-11GR

LIMIT

LIMIT

LIMIT

LIMIT

LIMIT

UNITS

1.0

0.5

3.5

2.5

5.0

mVMAX

RsS 10kO

20

20

50

50

200

nAMAX

300

300

500

500

500

nAMAX

±12

±12

±12

±12

±12

VMIN

100

100

100

100

70

dBMIN

32

32

32

32

100

I'VNMAX

Common-Mode
Rejection Ratio

CMRR

VcM =±12V
RsS 10kO

Power Supply
Rejection Ratio

PSRR

Vs =±5Vto±15V
RsS 10kO

Output Voltage Swing

Vo

RL 210kO
RL =2kO

±11
±11

±12
±11

±11
±11

±12
±11

±11
±11

VMIN

Avo

RL22kO
Vo =±10V

50

100

50

100

50

V/mVMIN

Pd

VOUT=O
No Load

200

ISO

200

ISO

340

mWMAX

Large-Signal
Voltage Gain
Power Consumption
(Four Amplifiers)

NOTES:
For 25°C characteristics of NT & GT devices, see N & G characteristics, respectively.
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging Is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at

PARAMETER

SYMBOL CONDITIONS

Slew Rate

SR

Unity Gain Bandwidth

GBW

Channel Separation

CS

Vs

=±15V, TA =+25°C, unless otherwise noted.

OP-08NT
OP·11NT

OP-08N
OP-11N

OP·09GT
OP-11GT

OP-11G

OP-09GR
OP-11GR

TYPICAL

TYPICAL

TYPICAL

TYPICAL

TYPICAL

UNITS

2

2

2

2

2

MHz

130

130

130

130

130

dB

Av=1
RL 22kO

Av= 100
f= 10kHz
Rs= lkO

VI"s

S-72

1/86, Rev. A

----------1~ OP·09/0P·11 QUAD MATCHED 741·TVPE OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS
INPUT OFFSET VOLTAGE
VI TEMPERATURE

BIAS CURRENT VI
TEMPERATURE

OFFSET CURRENT VI
TEMPERATURE
300

-0.10

~w

i
~

o

... ~

-0.20

-030

,

...... ......
~

r-.... ~

-0.40

-0.60
-80 -40 -20

0

20

40

60

.;'~

.....

......

0
-60 -40 -20

80 100 120 140

.....
......

0

20

40

80

80 100 120 140

1.4

\

~

PH~i

eo

I

\

GAIN

I

I-- -

\

.....

\

VS" :t16V

TA-+25"C

20

-<10 -20

0

20

40

eo eo

o

100 120 140

0.1

TEMPERATURErC)

OPEN·LOOP GAIN V8
SUPPLY VOLTAGE
BOO

TA'+25l C
700 -RL-2kn

/'
600

~ 500

V

2

:c
400

~300

~

1.0

10

100

1k

10k

100k

13

e5l

1.2

0:

f5

1. 1

90

w

10

\,
\

\

~
45

136

\

RL- 2k

CL"00pF

o

15

-80 -40 -20

~

'"~

eo

VS'" ±15V
RL -2kO -

I

!-"-

100

100

~

o

80 100 120 140

NORMALIZED SLEW RATE
AND BANDWIDTH
VI TEMPERATURE

200

"

"""

100

OPEN-LOOP GAIN AND PHASE
VI FREQUENCY

40

!

t

60

",

iii

I-

40

...... ~

""~

",

OPEN·LOOP GAIN VI
TEMPERATURE

I

-eo

1I

-~~

200

TEMPERATURE (OC)

/

300

20

~

0:
0:

TEMPERATURE (OC)

120

400

0

1

TEMPERATURE (OC)

700

&00

~

", ~

800

600

V~""5V

IV
)±15J

vl . t}sv

BO

~

~>
~

I

CMRR

-

0.7

80

0:
0:

i!i

--80 -40 -20

10M

FREQUENCY

11111111111

120

VS'"±15V

r-...

TA '" 25°C

100

,
I

I

80

i

60

20

20

o

o

10

100

----

._-------

1k

FREQUENCY (Hz)

5·73

._.

-

111111111
Vs" :t15V

TA '" 25°C

i

60

1.0

0 20 40 80 aD 1ao 120 140
TEMPERATURE I·C)

PSRR VI FREQUENCY

100

20

.....

140

120

j

",

BANDWIDTH " -

O.B
1M

140

100

/

V8

SLEW RATE

"-

FREQUENCV (Hz)

/

6
10
1&
POWER SUPPLY VOLTAGE "VI

~~

08

40

o

Va· :t16V

0.9

40

200

....

1- ~

10k

lOOk

o

1.0

10

100
1k
FREQUENCY IH.I

10k

101M<

1/86, Rev. A

-----------1m

OP-09/0P-11 QUAD MATCHED 741-TYPE OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS
CHANNEL SEPARATION va
FREQUENCY

NOISE CURRENT DENSITY
va FREQUENCY

NOISE VOLTAGE DENSITY
va FREQUENCY
1000

V S .. ;t16V

TA=+2!1'C

INPUT~V1
10kf.!

CS -20109 1DOV,

~_

r--

?

Vz
VZ

1

100

1k

10k

lOOk

01
10

FREQUENCY (Hz)

L--L~Llllli-~-L~WL__LL~Uill

10

"'"

100
1k
FREQUENCY (Hz)

1k

10k

FREQUENCY (Hz)

MAXIMUM OUTPUT SWING
va FREQUENCY

VOLTAGE FOLLOWER
PULSE RESPONSE

TRANSIENT RESPONSE

100

28

~

+20

~
";:

~(::

3

V S =±15V

TA =+2SOC

\

~ 20

1

IIIII

1\

.

2

RL .. 2k.!l

o

0

~ 16

~-1

~w

....

~-2
~-3

-20

I

,.

~

g8
5

200

400

2

800

4

8

8

10
TIME

TIME (ns)

12

14 18

,8

5o 4

~

(~sl

~f....

o

OUTPUT VOLTAGE va
LOAD RESISTANCE

lOOk
FREQUENCY (Hz)

---

POSITIVE SWING

I(

Ij

(

NEGATIVE SWING

I I I
Vs ~ ±16V

-

1/

-

I-:--

r-

-

Vs - ±16V

.......
.......

ffi

~

o

o
10
LOAD RESISTANCE TO GROUND (kSlI

III

TA '" +26°C

I

TA '" 2!i"C

~

0.1

POWER CONSUMPTION va
TEMPERATURE

QUIESCENT CURRENT VI
SUPPLY VOLTAGE

16

10

o

10

20

30

TOTAL SUPPLY VOLTAGE (VOLTS)

5-74

1M

10k

lk

.......

r-.

110

100
--60 ....0 -20

0

20

40

80

80

100 120 140

TEMPERTURE I-C)

1/86, Rev. A

OP-IO
DUAL MATCHED INSTRUMENTATION
OPERATIONAL AMPLIFIER
Prec ISlon 1\1011olIthH'::s Inc

FEATURES
•
•
•
•
•
•
•
•
•
•
•

is provided between channels of the dual operational
amplifier.

Extremely Tight Matching
Excellent Individual Amplifier Parameters
Offlet Voltage Match ..................... O.18mV Max
Offset Voltage Match VI Temp •...•.•.•.. O.8p.V/ o C Max
Common-Mode Rejection Match .......•..• 114dB Min
Power Supply Rejection Match ....•..•.... 100dB Min
Blal Current Match ....................... 3.0nA Max
Low Nolle .............................. O.6p.Vp•p Max
Low Blal Current ......................... 3.0nA Max
High Common-Mode Input Impedance •... 200GO Typ
Excellent Channel Separation ...•.•....•.. 126dB Min

The excellent specifications of the individual amplifiers
and tight matching over temperature enable construction of
high-performance instrumentation amplifiers. The deSigner
can achieve the guaranteed specifications because the
common package eliminates temperature differentials which
occur in designs using separately housed amplifiers.
Matching between channels is provided on all critical parameters including offset voltage, tracking of offset voltage vs.
temperature, non inverting bias currents, and common-mode
and power-supply rejection ratios. The individual amplifiers
feature extremely low offset voltage, offset voltage drift, low
noise voltage, low bias current, internal compensation and
input/output protection.

ORDERING INFORMATIONt
TA=25°C
VosMAX
(mV)

HERMETIC
DIP
14-PIN

OPERATING
TEMPERATURE
RANGE

0.5
0.5
0.5
0.5

OP10AY'
OP10EY
OP10Y'
OP10CY

MIL
COM
MIL
COM

PIN CONNECTIONS

14-PIN CERAMIC DIP
(V-Suffix)

• For devices processed In total compliance to MIL-STD-883, add 1883 after
part number. Consult factory for 883 data sheet
tAli commercial and Industrial temperature range parts are available with
burn-In For ordering information see 1986 Data Book, Section 2.

GENERAL DESCRIPTION
NOTE:
Device may be operated even if insertion is reversed; this is due to
inherent symmetry of pin locations olampliliers A and B.

The OP-10 series of dual-matched Instrumentation operational amplifiers consists of two independent monolithic
high-performance operational amplifiers in a single 14-pin
dual-in-line package. Tight matching of critical parameters

SIMPLIFIED SCHEMATIC (1/2 OP-10)
V+o-----~--------~----~----_.------~--._----~------~----~~~--~--,

RS

OUTPUT
Rl0

+IN

-IN

-------+----<--------f,.,

O--'N'..-<........

v-o-------------~~--------------------~--+_--~~~~------4_--~--~

5·75

1/86, Rev. A

---------l~ OP·10 DUAL MATCHED INSTRUMENTAnON OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

DICE Junction Temperature (Tj) ••••••• -65'C to +150'C
Lead Temperature Range (Soldering, 60 sec) •••••• 300'C

Supply Voltage ••••••••••••••••••••••••••••••••••• ±22V
Internal Power Dissipation (Note 1) •••••••••••••• 500mW
Differential Input Voltage •••••••••••••••••••••••••• ±30V
Input Voltage (Note 2) •••••••••••••••••••••••••••• ±22V
Output Short·Circuit Duration ••••••••••••••••• Indefinite
Storage Temperature Range ••••••••••• -65'C to +150'C
Operating Temperature Range
Op·10A,OP-10
-55'C to +125'C

.....................

at Vs

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

Dual-in-Llne (V)

106'C

11.3mW/·C

NOTES:
1. See table for maximum ambient temperature rating and derating factor.
2. For supply voltages less than +22V, the absolute maximum Input voltage Is
equal to the supply voltage.

OP-10E,OP·10C •••••••••••••••••••••••• O'C to +70'C

INDIVIDUAL AMPLIFIER CHARACTERISTICS

PACKAGE TYPE

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

= ± 15V, TA = 25' C, unless otherwise noted.
Op·10A

PARAMETER

SVMBOL

Input Offset Voltage

Vas

Long-Term Input Offset
Voltage Stability

<1VosfTlme

MIN

CONDITIONS

(Notes 1. 2)

OP·10

TVP

MAX

MIN

TVP

MAX

0.2

0.5

0.2

0.5

mV

0.25

1.0

0.25

1.0

",V/Mo
nA

UNITS

Input Offset Current

los

1.0

2.8

1.0

2.8

Input Bias Current

Ie

±1

±3

±1

±3

nA

Input Noise Voltage

enp-e

(Note 2) O.IHz to 10Hz

0.35

0.6

0.35

0.6

",Ve:e

Input Noise Voltage
Density

en

fo= 10Hz
(Note 2) fa = 100Hz
fo= 1000Hz

10.3
10.0
9.6

18.0
13.0
11.0

10.3
10.0
9.6

18.0
13.0
11.0

nVl../HZ'

Input Noise Current

Ine-e

(Note 2) O.IHz to 10Hz

Input Noise Current
Density

In

fo= 10Hz
(Note 2) fo = 100Hz
fo= 1000Hz

Input Resistance Differential-Mode

RIN

(Note 3)

Input Resistance Common-Mode

RINCM

20

14

30

14

30

pAe:e

0.32
0.14
0.12

0.80
0.23
0.17

0.32
0.14
0.12

0.80
0.23
0.17

pA/../HZ'

60

20

200

Input Voltage Range

IVR

Common·Mode Rejection
Ratio

CMRR

VCM =±13V

Power Supply Rejection
Ratio

PSRR

Vs = ±3V to ±18V

Large-Signal Voltage
Gain

Ava

RL" 2kll, Vo = ±10V
RL" soon, Vo = ±0.5V,
Vs = ±3V [Note 3)

Output Voltage Swing

Va

RL" 10kn
RL ,,2kO
RL"lkO

60

MO

200

GO

±13

±14

±13

±14

V

110

126

110

126

dB

4

10

4

10

p.VIV

200

500

200

500

150

500

150

500

±12.5
±12.0
±10.5

±13.0
±12.8
±12.0

±12.5
±12.0
±10.5

±13.0
±12.8
±12.0

V

VlmV

Slew Rate

SR

RL" 2kO

0.17

0.17

Vlp.s

Closed-Loop Bandwidth

BW

AVCL =+1.0

0.6

0.6

MHz

Open-Loop Output
Resistance

Ro

Vo=O.lo=O

80

Pd

Each Amplifier
Vs =±3V

90
4

Rp= 20kO

±4

±4

mV

8

8

pF

Power Consumption
Offset Adjustment Range
Input CapaCitance

C1N

NOTES:
1. Long·Term Input Offset Voltage Stability refers to the averaged trend line
of Vos vs. Time over extended periods after the first 30 days of operation.
Excluding the Initial hour of operation. changes in Vos during the first 30
operating days are typically 2.5",V - refer to typical parformance curves.

80
120

6

90
4

0
120
6

mW

2. Sample tested.
3. Guaranteed by design.

5·76

1/86, Rev. A

~

OP·10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

INDIVIDUAL AMPLIFIER CHARACTERISTICS at Vs= ±15V, -55°C:5 TA:5 +125°C, unless otherwise noted.
Op·10A
PARAMETER

SYMBOL

Input Ollset Voltage

Vos

Average Input Ollset
Voltage Drift
Without External Trim

TCVos
TCVosn

With External Trim
Input OIIset Current
Average Input OIIaet
Current Drift

MIN

MAX

TVP

MAX

UNITS

0.3

0.7

0.3

0.7

mV

(Note 2)

0.7

2.0

0.7

2.0

,.V/oC

Rp= 20kO (Note 3)

0.3

1.0

0.3

1.0

,.VloC

1.8

5.6

1.8

5.6

nA

los
TClos

Input Bias Current

la

Average Input Blaa
Current Drift

TCl a

OP·10

TVP

CONDITIONS

(Note 2)

(Note 2)

I nput Voltage Range

IVR

Common·Mode Rejection
Ratio

CMRR

VcM =±13V

Power Supply Rejection
Ratio

PSRR

Vs=±3Vto±18V

Large-Signal Voltage
Gain

Avo

RL~

Output Voltage Swing

Vo

RL~2kO

MIN

8

50

8

50

pArC

±2

±8

±2

±8

nA

13

50

13

50

pA/oC

±13.0

±13.5

±130

±13.5

V

106

123

106

123

dB

loLl

5

2kO, Vo= ±10V

MATCHING CHARACTERISTICS at Vs =

±15V, TA = 25°C,

20

5

,.VN

20

150

400

150

400

VlmV

±12.0

±12.8

±12.0

±12.6

V

unless otherwise noted.
OP·10

PARAMETER

SYMBOL

TVP

MAX

TVP

MAX

UNITS

Input Offset Voltage

AVos

007

0.18

0.12

0.5

mV

Average Nonlnvertlng
Bias Current

la+

±1.0

±3.0

±1.3

±4.5

nA

Nonlnvertlng Offset
Current

Ios+

0.8

2.8

1.1

4.5

nA

Inverting Offset Current

loS-

0.8

28

1.1

4.5

nA

Common-Mode Rejection
Ratio Match

ACMRR

VcM =±13V

Power Supply Rejection
Ratio Match

APSRR

Vs =±3Vto ±18V

Channel Separation

CS

(Note 2)

CONDITIONS

MATCHING CHARACTERISTICS at Vs =

MIN

114

123

126

140

3

±15V, -55°C:5 TA:5 +125°C,

MIN

106

120

126

140

10

4

SYMBOL

Input Offset Voltage
Match

AVos

With External Trim

CONDITIONS

TCAVos

(Note 2)

TCAVosn

Rp = 20kO (Note 3)
Channel A only

MIN

NOTES:
1. Long-Term Input Ollset Voltage Stability refers to the averaged trend line
of Vos vs. Time over extended periods after the IIrst30 days of operation.
excluding the Initial hour of operation, changes In Vos during the IIrst30
operating days are typically 2.5"V - refer to typical performance curves.

-~
~

dB

,.VN

20

dB

OP·10

TVP

MAX

01

0.3

0.45
0.3

MIN

TVP

MAX

UNITS

0.2

0.9

mV

1.3

0.9

2.5

"VI"C

0.8

04

1.2

"VI"C

2. Sample tested.
3. Guaranteed by deSIgn

1/86, Rev. A

5·n
~-~---

~

unless otherwise noted.

Op·10A
PARAMETER

~

0

OP·10A

Input Offset Voltage
Tracking
Without External Trim

-2
ffi

~--~-

----------

--~-

--

-

-- - - -

~ OP·10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

MATCHING CHARACTERISTICS at Vs =± 15V. -55°C ~ TA ~ +125° C. unless otherwise noted. (Continued)
OP·10A
PARAMETER

SYMBOL

Average Nonlnverting
Bias Current

IB+

Average Drift of
Nonlnvertlng
B,as Current

TCI B+

Noninvertlng Offset
Current

loot

Average Drift of
Nonlnvertlng
Offset Current

TClos+

MIN

CONDITIONS

(Note 2)

(Note 2)

Inverting Offset Current

los-

Common-Mode ReJeclion
Ratio Match

ACMRR

VcM =±13V

Power Supply Rejection
Rallo Match

APSRR

Vs = ±3V to ±18V

108

OP·10

TVP

MAX

±2.0

TVP

MAX

UNITS

±6.0

±2.4

±8.0

nA

10

40

15

2.0

6.5

2.4

12

50

18

2.0

6.5

2.4

120

6

MIN

103

pAl"C

9.0

pArC
9.0

117

20

nA

nA
dB

32

/LVIV

INDIVIDUAL AMPLIFIER CHARACTERISTICS at Vs =± 15V. TA = 25° C. unless otherwise noted.
Op·10E
PARAMETER

SYMBOL

Input Offset Voltage

Vos

Long-Term Input Offset
Voltage Stability

AVos/Time

Input Offset Current

los

Input Bias Current

IB

Input NOise Voltage

enp~p

Input Noise Voltage
Density
Input Noise Current

CONDITIONS

MIN

(Notes 1, 2)

OP·10C

TVP

MAX

TVP

MAX

UNITS

0.2

0.5

0.2

0.5

mV

0.3

1.5

0.5

MIN

/Lv/Mo

12

38

1.8

6.0

±1.2

±4.0

±1.8

±7.0

nA

(Note 2) O.IHz to 10Hz

0.35

0.6

0.38

0.65

/LV~_~

en

fo= 10Hz
(Note 2) fo = 100Hz
fo= 1000Hz

10.3
10.0
9.6

18.0
13.0
11.0

10.5
10.2
9.8

20.0
13.5
11.5

nV/..jHZ

In~-~

(Note 2) 0.1 Hz to 10Hz

Input Noise Current
Density

In

fo= 10Hz
(Note 2) fo= 100Hz
fo= 1000Hz

Input Resistance Differential-Mode

R'N

(Note 3)

Input Resistance Common-Mode

R'NCM

Input Voltage Range

IVR

Common-Mode Rejection
Rallo

CMRR

VCM =±13V

Power Supply Rejecllon
Ratio

PSRR

Vs =±3Vto±18V

Large-Signal Voltage
Gain

Avo

RL ;" 2kll. Vo = ±10V
RL ;" 5000, Vo = ±O 5V,
Vs = ±3V (Note 31

Output Voltage SWing

Vo

RL ;" 10kO
RL ;"2kO
RL ;"lkO

nA

14

30

15

35

pA~::E

0.32
0.14
0.12

0.80
0.23
017

0.35
0.15
0.13

0.90
0.27
0.18

pAl..jHZ

15

50

±13

±14

±13

±14

V

106

123

100

120

dB

8

160

4

20

33

MO

120

GO

10

200

500

120

400

150

500

100

400

±12.5
±12.0
±10.5

±13.0
±128
±12.0

±12.0
±11.5

±13.0
±12.8
±12.0

32

/LV/V

V/mV

NOTES:
1. Long-Term Input Offset Voltage Stability refers to the avereged trend line
of Vos vs. Time over extended periods after the IIrst30 days of operation.
Excluding the Inillal hour of operation, changes in Vosdurlng the IIrst30
operating days are typically 2.5/LV - refer to typical performance curves.

V

2. Sample tested
3. Guarantaed by design.

5·78

1/86, Rev. A

----------l~ OP-10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

INDIVIDUAL AMPLIFIER CHARACTERISTICS at Vs= ±15V, QOC ~ TA~ +7QoC, unless otherwise noted. (Continued)
OP-10E
PARAMETER

SYMBOL

CONDITIONS

Slewing Rate

SR

RL<" 2kO

Closed-Loop Bandwidth

BW

Open-Loop Output
Resistance
Power Consumption

TYP

OP-10C
MAX

MIN

TYP

MAX

UNITS

0.17

0.17

V/~s

AveL =+10

0.6

0.6

MHz

Ro

Vo= 0.10= 0

60

60

0

Pd

Each Amplifier
Vs= ±3V

90
4

Rp= 20kO

±4

Offset AdJustment Range
Input Capacitance

MIN

120
6

95
4

150
8

±4

mW
mV
pF

C,N

INDIVIDUAL AMPLIFIER CHARACTERISTICS at Vs= ±15V, QOC ~ TA~ +7QoC, unless otherwise noted.
OP-10E
PARAMETER

SYMBOL

Input Offset Voltage

Vas

CONDITIONS

MIN

OP-10C

TYP

MAX

0.25

06

MIN

TYP

MAX

UNITS

0.35

1.6

mV

Without External Trim

TCVos

With External Trim

TCVOSn

Input Offset Current

los

Average Input Offset
CUrrent Drift

TClos

Input Bias Current

la

Current Dnft
Input Voltage Range
Common~Mode

Rejection

Ratio
Power Supply Rejection
Ratio
Large-Signal Voltage
Gain

Output Voltage Swing

i:2
~
......
~

::l
~

Average Input Offset
Voltage Drift

Average Input Bias

II

Rp= 20kO

(Note 2)

0.7

2.0

1.2

4.5

~V/oC

(Note 3)

0.3

1.0

0.4

1.5

~V/oC

1.4

53

2.0

8.0

nA

8

50

12

50

pA/oC

±15

±55

±2.2

±9.0

nA

13

50

18

50

pA/oC

(Note 2)

TCl a

(Note 2)

IVR
CMRR

VeM = ±13V

PSRR

Vs = ±3V to ±18V

Ava

RL<" 2kO, Vo= ±10V

Va

RL <" 2kO

±130

±135

±13.0

±13.5

V

103

123

97

120

dB

32

10

51

~VIV

100

400

100

400

V/mV

±12.0

±12.6

±11.0

±12.6

V

NOTES:
1. Long-Term Input Offset Voltage Stability refers to the averaged trend line
of Vas vs Time over extended penods after the first 30 days of operation.
Excludmg the initial hour of operation, changes in Vas during the first 30
operating days are typically 2.Sp.V - refer to tYPical performance curves.
2. Sample tested
3. Guaranteed by design

5-79

1/86, Rev_ A

~

~

0
......

~
~
~

0

IfHD

OP-10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

MATCHING CHARACTERISTICS at Vs = ± 15V, TA = 25° C, unless otherwise noted.
OP·10E
PARAMETER

SYMBOL

CONDITIONS

MIN

OP·10C

TYP

MAX

MIN

0.5

TYP
0.3

mV

MAX

UNITS

Input Offsst Voltage
Match

AVos

0.12

Average Nonlnvertlng
Bias Current

Ie+

±1.3

±4.5

±2.0

nA

Nonlnvertlng Offset
Current

los+

1.1

4.5

1.a

nA

1.1

4.5

1.a

nA

117

dB

Inverting Offset Current

loS""

Common-Mode Aejectlon
Aatlo Match

ACMAA

VCM =±13V

Power Supply Aejectlon
Aatlo Match

APSAA

Vs=±3Vto±1av

Channel Separation

CS

(Note II

MATCHING CHARACTERISTICS at Vs= ±15V, DOC ~

106

120
20

126

TA$

p.VlV

140

120

SYMBOL

Input Offset Voltage
Match

t!.Vos

Input Offset Voltage
Tracking
Without External Trim
With External Trim

dB

+7DoC, unless otherwise noted.
OP·10E

PARAMETER

137

MIN

CONDITIONS

OP·10C

TYP

MAX

0.1S

0.7

0.4

MIN

TYP

MAX

UNITS
mV

TCAVos

(Note II

0.9

2.3

1.3

p.VI'C

TCt!.VOSn

AL = 20kO
Channel A Only (Note 21

0.3

0.9

0.6

p.VI'C

±2.0

±6.0

±2.S

nA

12

40

18

pAl'C

2.0

6.0

2.8

nA
pAl'C

Average Nonlnverting
Bias Current

18+

Average Drift of
Nonlnverting Bias Current

TCI 8+

Nonlnvertlng Offset
Current

18+

Average Drift of
Nonlnvertlng Offset Current

TClos+

(Note 11

(Note II

Input Offset Current

los-

Common-Mode Aejection
Aatlo Match

t!.CMAA

VcM =±13V

Power Supply Aejection
Aatlo Match

APSAA

Vs= ±3V to ±1aV

103

15

50

20

2.0

6.0

2.a

nA

114

dB

117
6

32

p.VIV

NOTES:
I. Sample tested.
2. Guaranteed by design.

5-80

1/86, Rev. A

--------100

OP·10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

BURN·IN CIRCUIT

OFFSET NULLING CIRCUIT

V+
NC

14 I

20kn

,

14

'::- .....

3

;r

4

+18V 0 - -

13

>-+''''----<> OUT A

INPUT

N.C.

' -_ _-\,'c::.'_ _-oV_

12

-lav

QP·10

QP·10

5

~

11

l1 . .

10

";"

7 8

r---~---ov11

•

N,C.

>--~---oOUT B

INPUT

9
20kS1

N.C

V+

TYPICAL PERFORMANCE CHARACTERISTICS

MATCHING CHARACTERISTICS
TRIMMED OFFSET VOLTAGE
MATCH vs TEMPERATURE
40

=

r- Vas TRIMMED TO 

+15V

-55 C TO +12S"C
VS'" ±15V

~

i<
~

I

100.11

OP-l0E

Vs'" ±15V

III

~
0

lOon

aOc TO 70"C

OP-l0A

Q

.§

UNTRIMMED

55"C TO

+12~C

-

III

~

~

UN~RI,JJJ~ O"c

I 11111111

10

UNTRIMMED 25°C
01
TRIMMED

~

TO 70"C

UNTRIMMED 2s<'C

I I LII V

1

sSOC TO +12!i"'C

~

~

INPUT REFERENCE NOISE '"

25~~O '" ~~.~b~ '" lOOnV/em

TRIMMED

x
~

~

I 11111111

001

II

II

lk

100

10k

00 1
100

TOOk

INPUT WIDEBAND NOISE
VB BANDWIDTH
(0.1 Hz TO FREQUENCY
INDICATED)

1000

10
AS1 - RS2 - 200kn

Vs - +15V
TA -+2S"C

THERMAL NOISE OF SOURCE
RESISTORS INCLUDED

~

EXCLUDED

t--.

100

'>

."!
w

w
~

0z

/

~

w

'"~

lOOk

10k

SOURCE RESISTANCE (n)

VOLTAGE NOISE DENSITY
VB FREQUENCY

~

I

lk

SOURCE RESISTANCE (.\1)

OP-10 LOW FREQUENCY NOISE

oCc TO 70°C

J I liW

~

10

0

>

--

5
z
~

RS'" 0

1.0

~

--

-V s = 15V

-Yf5°f
1

10
1.0

10

1
100

1k

100

BANDWIDTH (Hz)

OFFSET VOLTAGE DRIFT
WITH TIME

TYPICAL OFFSET VOLTAGE
STABILITY vs TIME

TOOk

10k

1k

FREQUENCY (Hz)

TRIMMED OFFSET VOLTAGE
VB TEMPERATURE
30

'>

'> 4I+HmFH11tIII~-mtltm-mttnC11'1't1III!C11'm1IIt-ij4111
."!
w

'"~
~
~

~
0

>

~

0

WARM-UP
0

DRIFT

LONG·TERM

ERROR
BAND

DRIFT

~0

15

l.I.

~

-3

~

-4

G -4rt~rH~rH~rH~rHffl$-Hffl$~,ffift

~

_6~~~~~~~~~~-W~-U~

10

10

5-82

100

lk

M

~

10k

:;

~

3)

2)

(3 /

\ \

w

-5

TIME (HOURS)

\

10

>

Vf~~~rH~rH~rH*I-H*I'

01

/

/

1
I

CD

w

:;

-2

0001001

I

-

0

ERROR BAND

-1

o

20

~

~j~ g~=~gc

3

>
~

-2

TIME (MONTHS)

'"~

Vas TRIMMED TO < 5J,1V AT 25"C
NULLING POT'" 20kn
1
1/20P-l0E
2
1/20P-l0A

f-25

0

2

f-

tl

."!
w

5

~

1 &l
10

IF

a
-75

-50

-25

25

50

75

100 125

TEMPERATURE (OC)

1/86, Rev. A

---------I~ OP-10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

CMRR
130

IlllIi

'20

FREQUENCY

VB

PSRR
'20

Jllll!l~

1/20P-l0C

110

1/20P-l0C

90

t\

BO

80

70

~

~

1k

'00

_I

~

AV=10

20

~
0'

lOOk

'0

OPEN-LOOP GAIN
VB FREQUENCY

t--

I"

BO
60
40

1k

'00

'Ok

"- ~

-40

10

100

lk

lOOk

1M

10M

-

r-....

600

.........

r-....

200

o

±s

±10

:l:1S

±20

MAXIMUM OUTPUT VOLTAGE
vs LOAD RESISTANCE
20

-

r--

o
-75

-50

-25

25

75

100

POWER CONSUMPTION
POWER SUPPLY

125

VB

1000
TA =+2SoC

LoLI l1E

'5

50

TEMPERATURE (OC)

II II

TA=+2SJC
Vs = ±15V
VIN = ±10mV

=

Vs ±1SV
TA '" +25"C

20

'OM

200

MAXIMUM OUTPUT SWING
vs FREQUENCY
24

I.,...-

400

POWER SUPPLY VOLTAGE (VOLTS)

I-

\

1M

BOO

FREQUENCY (Hz)

2B

lOOk

Vs = ±15V

400

o

10k

1000
TA = +25~C

/f-""

10k

1k

'00

OPEN-LOOP GAIN vs
TEMPERATURE

600

""
"

10

OPEN-LOOP GAIN vs POWER
SUPPLY VOLTAGE

800

-20

"'-

-20

FREOUENCY (Hz)

Vs = ±15V
TA '" +25°C_

20

0'

'0

lOaD

~

~

FREQUENCY (Hz)

fREaUENCY (Hz)

'00

~

J

AV'" 100

60

'Ok

J±15V

AV=~

60

50
10

Vs

TA '" 2SoC

BO

AV=l

70
60
10

'00

40

90

,20

I'~~ = 25Jb'l

llllIJ

'00

100

CLOSED-LOOP GAIN
VB FREQUENCY

FREQUENCY

1;;JlliUI

~

110

VB

E

V

100
NEGATIVE SWING

'6

'0
'2

1\
o
1k

10k

lOOk

FREQUENCY (Hz)

'M

o

'00

1/,/
II

IL

10

1k
LOAD RESISTANCE TO GROUND (n)

5-83

10k

'0 o

I
20

40

60

TOTAL SUPPLY VOLTAGE, V+ TO V- (VOLTS)

1/86, Rev. A

--------I~ OP-10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
INPUT BIAS CURRENT VI
DIFFERENTIAL INPUT
VOLTAGE

OFFSET VOLTAGE CHANGE
DUE TO THERMAL SHOCK
30

;;
-"
w

~g

IlL L I
IlL L L

25

.1

~ 15 - f -

!:

-~

10

~

w

~

< 6J,1V AT 25°C

~

-20

S1

a

~

V

V iL

10

~
~

~ -10

;:

~ -20

60

~
80

100

TIME (SEC)

,

/

ffi

DEVICE IMMERSED
IN 700C OIL BATH

40

Vs! ±15v'

- -c-

~
--_-0 OUTPUT

R3

2kn

R2

Skn

Rl + R2) R6
VOUT=V 1N ( 1+ R 3 R4
= lOa
GAIN LIN = 0 002%
SLEW RATE = 25V/j.ls
PSAR = 112dB
GAIN

R5

2kn
lo-----+--l+

R7

20kn

'F~=~
THEN CMRR = 120dB
ADJUST R7 FOR MAXIMUM CMRR

5-85

1/86, Rev. A

~

I-I-l

......
~

~

~

~

0
......

~
I-I-l

C5

---------l1fHD

OP-10 DUAL MATCHED INSTRUMENTATION OPERATIONAL AMPLIFIER

CMRR VB FREQUENCY
INSTRUMENTATION AMPLIFIER (3 OP-AMP DESIGN)

PRECISION DUAL TRACKING VOLTAGE REFERENCES
USING OP-10

125
120

R'

V+
115

i'-

m

110

~

""u

'05

""

2

\j

,

'00

2

95

3

AS - 100kn,
BALANCED
AS = loon, lkn,
UNBALANCED

RS = 20kn,
BAI~fNCED

90

'0

10

R2

R4

R5

,
\
R6

\

R6=

~!: ~;
V2

\
100

1k

FREQUENCY (Hz)

PRECISION DUAL TRACKING VOLTAGE REFERENCES
USING OP-10

INSTRUMENTATION AMPLIFIER (2 OP-AMP DESIGN)

Precision dual tracking voltage references using a single
reference source are easily constructed using OP-10. These
references exhibit low noise, excellent stability vs. temperature and time, and have excellent power supply rejection.

V+

Rl

In the circuit shown, R3 should be adjusted to set IREF to
operate VREF at its minimum temperature coefficient current.
Proper circuit start-up is assured by Rz, Z1, and 01.
V1 = VREF (1

R4

99kn

VOUT

(1+~)

+ R2)
R1

Vl

V2

~=~

±O.005%

V2=V1 (-R5)
R4

Output Impedance (dIL:1.0mA-5.0mA) ...... 0.25 x 10-30

5-86

1/86, Rev. A

OP-12
PRECISION LOW- INPUT- CURRENT
OPERATIONAL AMPLIFIER (INTERNALLY COMPENSATED)
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

•
•
•
•
•
•
•
•
•
•
•

The PMI OP-12 is an improved version of the popular LM108A
low-power op amp. The OP-12 is internally compensated and
its chip dimensions are only 42 X 58 mils. Offset voltage is
lower; the total worst-case input offset voltage over-55° C to
+ 125° C for the OP-12A is only 350"V. In addition, the OP-12
drives a 2kO load which is five times the output current
capability of the 108A. Thisexcellent performance is achieved
by applying PMI's ion-implanted super-beta process and
on-chip zener-zap trimming capabilities. The internal compensation makes this op amp ideal for hybrid assembly
applications.

Low Offset Voltage ........................ 150"V Max
Low Offset Voltage Drift ................ 2.5"Vlo C Max
Load Current Capability ..................... 5mA Min
Internal Frequency Compensation
125°C Temperature Tested Die
Low Offset Current ...................... 200pA Max
Low Bias Current ......................... 2.0nA Max
Low Power Consumption .......... 18mW Max@±15V
High Common-Mode Input Range .......... ±13V Min
MIL-STD-883 Class B Processing Available
Silicon-Nitride Passivation

ORDERING INFORMATIONt

PIN CONNECTIONS

NC
NCS'V'

PACKAGE
TA = 25°C
VosMAX
(mV)

HERMETIC
TO-99
8-PIN

HERMETIC
DIP
8-PIN

OPERATING
TEMPERATURE
RANGE

0.15
0.15
0.30
0.30
1.0
1.0

OP12AJ'
OP12EJ
OP12BJ'
OP12FJ
OP12CJ'
OP12GJ

OP12AZ'
OP12EZ
OP12BZ'
OP12FZ
OP12CZ'
OP12GZ

MIL
COM
MIL
COM
MIL
COM

8

"+

-IN 2

+IN3

5NC

v-

• For deVices processed In total compliance to M I L-STD-883. add /883 after
part number Consult factory for 883 data sheet
t All commercial and industnal temperature range parts are available with
burn-In. For ordenng Information see 1986 Data Book, Section 2.

60UT

4
(CASE)

TO-99 (J-Suffix)

8-PIN HERMETIC MINI-DIP
(Z-Sufflx)

SIMPLIFIED SCHEMATIC

R5

OUTPUT
L---+--~-o6

R6

5-87

1/86, Rev. A

--------IIEHD

OP·12 PRECISION LOW·INPUT·CURRENT OPERATIONAL AMPLIFIER

NOTES:
1. See table for maximum ambient temperature rating and derating factor.

ABSOLUTE MAXIMUM RATINGS (Note 4)
Supply Voltage
OP·12A, Op·12B,
OP·12E, Op·12F, All DICE except GR •••••••••••• ±20V
OP·12C, OP·12G, GR DICE Only ................ ± 18V
Operating Temperature Range
Op·12A, Op·12B, OP·12C ••••••••••• -55°C to +125°C
OP·12E, Op·12F, Op·12G •••••••••••••••• O°C to +70°C
Storage Temperature Range ••••••••••• -65°C to +150°C
Lead Temperature Range (Soldering, 60 sec) •••••• 300°C
Internal Power Dissiption (Note 1) ............... 500mW
Differentiallnpul Current (Note 2) •••••••••.••••• ±10mA
Input Voltage (Note 3) ............................ ±15V
Output Short·Clrcuil Duration ••••••••••••••••• Indefinite
DICE Junction Temperature (Tj) ••••••• -65°C to +150°C

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (J)

80'C

7,1mW/'C

8-Pin Hermetic DIP (Z)

75'C

8,7mW/'C

PACKAGE TYPE

2, The inputs are shunted with back-to-back diodes for overvoltage protection, Therefore. excessive current will flow If a differential Input voltage In
excess of 1V Is applied between the Inputs unless some limiting resistance
Is provided,
3, For supply voltages less than ± 15V, the absolute maximum Input voltage Is
equal to the supply voltage,
4, Absolute maximum ratings apply to both DICE and packaged parts, unless
otherwise noted,

ELECTRICAL CHARACTERISTICSatVs=±20Vand TA=25°CforA, B, Eand Fgrades, Vs=±15V, andTA =25°CforCand
G grades, unless otherwise noled.
OP·12A/E
TYP
MAX
MIN

OP·12B/F
TYP
MIN
MAX

OP·12C/G
MIN
TYP
MAX

Vos

0,07

0,15

0,18

0,30

0,25

1,0

mV

Input Offset Current

los

0,05

0,20

0,05

0,20

0,08

0,50

nA

Input Bias Current

18

0,8

2,0

0,8

2,0

1,0

5,0

nA

Input ResistanceDifferential-Mode

R'N

(Note 1)

PARAMETER

SYMBOL

Input Offset Voltage

CONDITIONS

UNITS

26

70

26

70

10

50

Mil

Input Voltage Range

IVR

Vs= ±15V

±13

±14

±13

±14

±13

±14

V

Common-Mode
Rejection Ratio

CMRR

VCM = ±13V

104

120

104

120

84

116

dB

Power Supply
Rejection Ratio

±13
±10

±14
±12

±13
±10

±14
±12

±13
±10

±14
±12

BO

300

BO

300

40

250

50

150

50

150

PSRR

Vs= ±5V to ±15V

Output Voltage
Swing

Vo

RL2: 10kll, Vs =±15V
RL 2:2kll. Vs =±15V

63

Large-Signal
Voltage Gain

Avo

RL 2: 10kll
Vo =±10V
RL 2: 2kll
Vo =±10V

Power Consumption

Pd

Vs = ± 15V. No Load
Vs =±5V. No Load

Input Noise Voltage

en~pE!

0,1 Hz to 10Hz

0,9

0,9

O,g

",V

Input Noise
Voltage Density

en

fo= 10Hz
fo= 100Hz
fo= 1000Hz

22
21
20

22
21
20

22
21
20

nVlVH%

Input Noise Current

",VlV
V

VlmV
100
1B
6

18

12
4

24

mW

e-e

Ine-e

0,1 Hz to 10Hz

3

pA e-e

Input Noise
Current Density

In

fo= 10Hz
fo = 100Hz
fo= 1000Hz

0,15
0,14
0,13

0,15
0,14
0,13

0.15
0,14
0,13

pA/VH%

Slew Rate

SR

RL 2: 2kll

0,12

0,12

0.12

VI",s

Closed-Loop
Bandwidth

BW

AVCL = +1

0,80

O,BO

O,BO

MHz

Open-Loop Output
Resistance

Ro

Vo=O,lo=O

200

200

200

Il

NOTE:
1, Guaranteed by input bias current.

5·88

1/86, Rev. A

IfMD

OP-12 PRECISION LOW-INPUT -CURRENT OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs =± 15V, for C grade, Vs =±20V for A and B grades, -55· C ~ TA~ +125· C, unless
otherwise noted.
OP-12A
PARAMETER

SYMBOL

Input Offset Voltage

CONDITIONS

MIN

OP-12B

TYP

MAX

Vos

0.12

Average Input Offset
Voltage Drift

TCVos

MIN

OP-12C

TYP

MAX

MIN

TYP

MAX

UNITS

0.35

0.28

0.60

0.40

2.0

mV

0.50

2.5

1.0

3.5

1.5

10

~V/'C

Input Offset Current

los

0.12

0.40

0.12

0.40

0.18

1.0

nA

Average Input Offset
Current Drift

TClos

0.50

2.5

0.50

2.5

1.0

5.0

pA/'C

Input Bias Current

Ie

1.2

3.0

1.2

3.0

1.8

10

nA

Input Voltage Range

IVR

Vs= ±15V

±13

±14

±13

±14

±13

±14

V

Common-Mode
Rejection Ratio

CMRR

VOM = ±13V

100

116

100

116

80

112

dB

Power Supply
Rejection Ratio

PSRR

Vs = ±5 to ±15V

Large-Signal
Voltage Gain

Avo

R L ", 5kO
Vo =±10V

Output Voltage
Swing

Vo

RL ", 10kO, Vs =±15V
RL "'5kO. Vs =±15V

Power Consumption

Pd

Vs = ±15V, No Load

ELECTRICAL CHARACTERISTICS at Vs
otherwise noted.

4

10

10

100

p.VIV

120

40

120

15

80

V/mV

±13
±10

±14
±13

±13
±10

±14
±13

±13
±10

±14
±12

V

18

9

9

18

15

24

mW

~

OP-12F

OP-12G

SYMBOL

TYP

MAX

TYP

MAX

TYP

MAX

UNITS

Vos

0.10

0.26

0.23

0.45

0.32

1.4

mV

Average Input Offset
Voltage Drift

TCVos

0.50

2.5

1.0

3.5

1.5

10

p.V/'C

Input Offset Current

los

0.08

0.30

0.11

0.60

0.12

0.70

nA

Average Input Offset
Current Drift

TClos

0.50

2.5

1.0

5.0

1.0

5.0

pA/'C

1.0

2.6

1.2

5.2

1.4

6.5

nA

MIN

MIN

MIN

Input Bias Current

Ie

Input Voltage Range

IVR

Vs =±15V

±13

±14

±13

±14

±13

±14

V

Common-Mode
Rejection Ratio

CMRR

VOM =±13V

100

116

100

116

80

112

dB

Power Supply
Rejection Ratio

PSRR

Vs= ±5 to ±15V

Large-Signal
Voltage Gain

Avo

RL ", 10kO
Vo= ±10V
R L ", 2kO
Vo =±10V

Vo

Power Consumption

Pd

~

-~

= ±15V for G grade, Vs = ±20V for E and F grades, O·C ~ TA ~ 70·C, unless 0

PARAMETER

R L ", 10kO
Vs =±15V
RL ", 5kO
Vs= ±15V

....

::l
p...

Input Offset Voltage

Output Voltage
Swing

~

~

40

OP-12E
CONDITIONS

I-

4

10

10

100

60

200

60

200

25

100

25

100

±13

±14

±13

±14

±13

±14

±12

±10

±12

±10

±12

25

p.VIV

150
V/mV
80

V
±10

Vs = ± 15V. No Load

9

18

18

15

24

mW

For typical performance characteristics, see OP-08 data sheet. Assume
Co=30pF.

5-89

1/86, Rev. A

~

15

I
I

-----------l1fMD

OP-12 PRECISION LOW-INPUT-CURRENT OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS (125°C TESTED DICE AVAILABLE)
1.
2.
3.
4.
6.
7.
8.
DIE SIZE 0.058 X 0.042 Inch, 2436 sq. mils
(1.47 X 1.07 mm, 1.57 sq. mm)

WAFER TEST LIMITS

at Vs=

NO CONNECTION
INVERTING INPUT
NON INVERTING INPUT
VOUTPUT
V+
NO CONNECTION

For additional DICE informalion refer 10
1986 Dala Book, Seclion 2.

± 15V, TA = 25°C for Op·12N, Op·12G and Op·12GR devices; TA = 125°C for OP·12NT and

OP-12GT devices, unless otherwise noted. (Note 2)

OP-12NT

OP-12N

OP-12GT

OP-12G

OP-12GR

LIMIT

LIMIT

LIMIT

LIMIT

LIMIT

Vos

035

015

0.6

03

Input Offset Current

los

02

02

0.2

0.2

0.5

Input Bias Current

Ie

Input Voltage Range

IVR

±13

±13

±13

±13

±13

VMIN

Common-Mode
Rejection Ratio

CMRR

VCM = ±13V

100

104

100

104

84

dBMIN

Power Supply
Rejection Ratio

PSRR

Vs

Output Voltage Swing

Vo

R L <: 10kfl
R L ;" 2kfl
R L ;"5kfl

PARAMETER

SYMBOL

I nput Offset Voltage

2

=

nAMAX
nAMAX

10

±5V to ±15V

UNITS
mVMAX

10

±13

±13
±10

±10

±13

63
±13
±10

±13
±10

80
50

40

~V/v

MAX

VMIN

±10

R L ;" 10kfl. Vo= ±10V
R L <:2kfl, Vo =±10V
R L ;"5kfl, Vo =±10V

80

R'N

(Note II

26

26

26

26

10

MflMIN

ISY

0
VOUT=O

0.6

06

0.6

0.6

0.8

mAMAX

Large-Signal
Voltage Gain

Avo

Input ResIstance
Supply Current

CONDITIONS

80
50

40

10UT=

NOTES:
1. Guaranteed by design.

80

VlmV MIN

40

2.

For 25°C specifications of OP-12NT and OP-12GT, see OP-12N and
OP-12G , respectively.

Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS

at Vs

= ± 15V, unless otherwise noted.

OP-12NT

OP-12N

OP-12GT

OP-12G

OP-12GR

TYPICAL

TYPICAL

TYPICAL

TYPICAL

TYPICAL

UNITS

TCVos

05

0.5

10

1.0

15

~V;oC

TClos

05

0.5

10

10

10

pA/oC

PARAMETER

SYMBOL

Average In put Offset
Voltage Drift
Average Input Offset
Current Dnft

CONDITIONS

5-90

1/86, Rev. A

OP-15/ OP-16/ OP-17
PRECISION JFET-INPUT
OPERATIONAL AMPLIFIERS
Precision Monolithics Inc.

FEATURES (All Devices)

op amps. All devices offer offset voltages as low as 0.5mV with
TCVosguaranteed to 5!,VloC. A unique input bias cancellation
circuit reduces the IB by a factor of 10 over conventional
designs. In addition, PMI specifies I Band loswith the devices
warmed up and operating at 25°C ambient.

• Significant Performance Advantages over LF155. 156 and
157 Devices.
• Low Input Offset Voltage ................•.. 500!'V Max
• Low Input Offset Voltage Drift ................ 2.0!,Vlo C
• Minimum Slew Rate Guaranteed on All Models
• Temperature-Compensated Input Bias Currents
• Guaranteed Input Bias Current @ 125°C
• Bias Current Specified WARMED UP Over Temperature
• Internal Compensation
• Low Input Noise Current ....•............ 0.01pAlVHz
• High Common-Mode Rejection Ratio ............ 100dB
• Models With MIL-STD-883 Processing Available
• 125°C Temperature Tested DICE

These devices were designed to provide real precision
performance along with high speed. Although they can be
nulled, the design objective was to provide low offset-voltage
without nulling. Systems generally become more cost effective as the number of trim circuits is decreased. PMI achieves
this performance by use of an improved Bipolar compatible
J FET process coupled with on-chip, zener-zap offset trimmi ng.
The OP-15 provides an excellent combination of high speed
and low input offset voltage. In addition, the OP-15 offers the
speed of the 156A op amp with the power dissipation of a
155A. The combination of a low input offset voltage of 500!'V,
slew rate of 13V1!,s, and settling timeof 1200ns toO.1% makes the
OP-15 an op amp of both preCision and speed. The additional
features of low supply current coupled with an input bias current
of 9nA at 125° C ambient (not junction) temperature makes the
OP-15 ideal for a wide range of applications.

OP-1S
•
•
•
•

156 Speed With 155 Dissipation ........•... (80mW Typ)
Wide Bandwidth ................................ 6MHz
High Slew Rate ..............•.•....•.......... 13V1!,s
Fast Settling to ±0.1% .....•..•...............• 1200ns

OP-16
• Higher Slew Rate .............••...•........... 25V1!,s
• Faster Settling to ±0.1% ....•.................... 900ns
• Wider Bandwidth ....•.•..•....•..............•. 8M Hz

The OP-16 features a slew rate of 25V1!,s and a settling time of
900ns to 0.1% which represents a significant improvement in
speed over the 156. Also, the OP-16 has all the DC features of
the OP-15.

OP-17
• Highest Slew Rate .......•........•...•.. , •.... 60Vl!,s
• Fastest Settling to ±0.1% ........................ 600ns
• Highest Gain Bandwidth Product •.•............ 30M Hz

The OP-17 has a slew rate of 60Vl!,s and is the best choice
for applications requiring high closed-loop gain with high
speed. Applications include high-speed amplifiers for current
output DACs, active filters, sample-and-hold buffers, and
photocell amplifiers.

GENERAL DESCRIPTION
The PMI JFET-input series of devices offer clear advantages
over industry-genenc devices and are superior in both cost
and performance to many dielectrically-isolated and hybrid

See the OP-215 data sheet for a dual configuration of the
OP-15.

SIMPLIFIED SCHEMATIC
'"NOTE
R7, RS ARE ELECTRONICALLY
ADJUSTED ON CHIP FOR
MINIMUM OFFSET VOLTAGE

022

,---+----+--o (61 OUTPUT

V-14Io--....................- - - - -.............-

......-+-.......- - - - + - - -.......___..........- -........-

5-91

........---'

1/86, Rev. A

i.

I

I

I

II

--------l~ OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

ORDERING INFORMATIONt

ABSOLUTE MAXIMUM RATINGS (Note 2)
Supply Voltage
All Devices Except C, G (Packaged) & GR Grades .. ±22V
C, G (Packaged) & GR Grades .................. ±18V
Internal Power Dissipation (Note 1) .............. 500mW
Operating Temperature
A, B, & C Grades ................... -55·C to +125·C
E, F & G Grades ........................ O·C to +70·C
Maximum Junction Temperature ................ +150·C
DICE Junction Temperature (Tj) ....... -65·C to +150·C
Differential Input Voltage
All Devices Except C, G (Packaged) & GR Grades .. ±40V
C, G (Packaged) & GR Grades .................. ±30V
Input Voltage (Note 3)
All Devices Except C, G (Packaged) & GR Grades .. ±20V
C, G (Packaged) & GR Grades .................. ±16V
Input Voltage
OP-15A, OP-15B, OP-15E, OP-15F ............... ±20V
OP-15C,OP-15G ............................... ±16V
OP-16A, OP-16B, OP-16E, OP-16F ............... ±20V
OP-16C,OP-16G ............................... ±16V
OP-17A, OP-17B, OP-17E, OP-17F ............... ±20V
OP-17C,OP-17G ............................... ±16V
Output Short-Circuit Duration ................. Indefinite
Storage Temperature Range ........... -65·C to +150·C
Lead Temperature Range (Soldering, 60 sec) .... +300·C

PACKAGE
TO-89
8-PIN

8-PIN
HERMETIC
DIP

OPERATING
TEMPERATURE
RANGE

0.5

OP15AJ'
OP16AJ'
OP17AJ'

OP15AZ'
OP16AZ'
OP17AZ'

MIL

0.5

OP15EJ
OP16EJ
OP17EJ

OP15EZ
OP16EZ
OP17EZ

COM

1.0

OP15BJ'
OP16BJ'
OP17BJ'

OP15BZ'
OP16BZ'
OP17BZ'

MIL

1.0

OP15FJ
OP16FJ
OP17FJ

OP15FZ
OP16FZ
OP17FZ

COM

3.0

OP15CJ'
OP16CJ'
OP17CJ'

OP15CZ'
OP16CZ'
OP17CZ'

MIL

3.0

OP15GJ
OP16GJ
OP17GJ

OP15GZ
OP16GZ
OP17GZ

COM

TA = 2S·C
VOSMAX
(mV)

'For devices processed in total compliance to MIL-STD-883, add 1883 after
part number. Consult factory for 883 data sheet.
t All commercial and industrial temperature range parts are available with
burn-in. For ordering information see 1986 Data Book, Section 2.

PIN CONNECTIONS

NOTES:

1. See table for maximum ambient temperature rating and derating factor.

N.C.

•

BALe7V+
~

-IN 2

+IN 3

PACKAGE TYPE

6 OUT

5 BAL

v-

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (J)

SO·C

7.1mWfOC

Hermetic S-Pin Dip (Z)

75·C

6.7mW/·C

4
(CASE)

2.

TO-99
(J-Suffix)

8-PIN HERMETIC DIP
(Z-Sufflx)

3

5-92

Absolute maximum ratings apply to both packaged parts and DICE. unless
otherwise noted.
Unless otherwise speCified the absolute maximum negative input voltage is
equal to the negative power-supply voltage.

1/86, Rev, A

----------1~

OP·15/0P·16/0P·17 PRECISION JFET·INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, TA = 25°C, unless otherwise noted.
Op·15A1E
Op·16A1E
OP·17A1E
PARAMETeR

SYMBOL CONDITIONS

Input Offset Voltage

Vas

T1= 25'C (Note I)
Input Offset Current

los

Device Operating

Tj= 25'C (Note I)
Device Operating

Input Bias Current

la

MIN

Rs= 500

T1=25'C (Note I)
Device Operating
TI = 25'C (Note I)
Device Operating

Op·15B/F
Op·16B/F
OP·17B/F

TYP

MAX

0.2

0.5

MIN

10
22
10
25

Op·15
OP-D16/0P-17

OP-16/0P-17

Ava

Output Voltage
Swing
Supply Current

Slew Rate

RL " 2kO

ISY

OP-15
OP-16/0P-17

SR

OP-15
OP-16
OP-17

10
18
45

OP-15
OP-16
OP-17

4.0
6.0
20

AVCL = +1 (Note 3)

Gain Bandwidth
Product

GBW

Closed-Loop
Bandwidth

CLBW

AVCL = +1
AVCL = +5

Settling Time

Input Voltage Range

(Note 3)

2.7
4.6

MAX

UNITS

0.5

30

mV

20
40
20
50

12
20
12
20

50
100
50
125

pA

±30 ±100
±40 ±200
±30 ±100
±40 ±250

±60
±80
±60
±80

±200
±400
±200
±500

pA

10'2

10'2

o

200

VlmV

±12
±13
±11 ±12.7

V

220

4.0
7.0

2.7
4.6

MIN

50

±12
±13
±11 ±12.7
4.0
7.0

2.8
4.8

60

7.5
12
35

11
21
50

25

17
40

6.0
8.0
30

3.5
5.5
15

5.7
7.6
28

3.0
5.0
11

5.4
7.2
26

13
25

14

OP-15
OP-16
OP-17

TYP

1.0

6

75

±12
±13
±11 ±12.7

RL = 10kO
RL = 2kO

AVCL = +5 (Note 3)

240

100

Vo= ±10V

MAX

0.4

10

Input Resistance
Large·Signal
Voltage Gain

TYP

10

±15
±50
±18 ±110
±15 ±50
±20 ±130

OP-15

OP·15C/G
Op·16C/G
OP·17C/G

9

19
11

13
18
10

12
17

MHz

MHz

9

OP-15

4.5
1.5
1.2

4.5
1.5
1.2

4.7
1.6
1.3

to 0.01%
to 0 05% (Note 2)
to 0.10%

OP-16

3.8
1.2
0.9

3.8
1.2
0.9

4.0
1.3
1.0

to 0.01%
to 0.05% (Note 4)
to 0.10%

OP-17

1.5
0.7
0.6

1.5
0.7
0.6

1.6
0.8
0.7

±10.5

mA

9

to 0.01%
to 0.05% (Note 2)
to 0.10%

IVR

5.0
8.0

±10.5

±10.3

V

Common-Mode
Rejection Ratio

CMRR

VCM =±10.5V
VcM =±10.3V

Power Supply
Rejection Ratio

PSRR

Vs= ±10V to ±18V
Vs =±10Vto±15V

10

Input Noise
Voltage Density

fO= 100Hz
fo= 1000Hz

20
15

20
15

20
15

nV/y"Hz

Input Noise
Current Density

fO= 100Hz
fo= 1000Hz

0,01

0,01

0,01

0.01

0.01
0.01

pAly"Hz

3

3

3

pF

100

86

86

100
82

51

10

51
10

Input Capacitance
NOTES:
1. Input bias current is specified for two different conditions. The TJ= 25'C
specification is with the junction at ambient temperature; the Device
Operating specification is with the device operating in a warmed-up

dB

96
80

inverting input pin on the amplifier) to settle to within a specified percent of
its final value from the time a 10V step input is applied to the inverter. See
settling time test circuit.
3. Sample tested.
4. Settling time is defined hereforaA v =-5connection with R F= 2kO.lt isthe
time required forthe error voltage (the voltage at the inverting input pin on
the amplifier) to settle to within 0.01% of its final value from the time a 2V
step input is applied to the inverter. See settling time test circuit.

condition at 25° C ambient. The warmed-up bias current value is correlated
to the junction temperature value via the curves of lavs TJ and IBvs TA. PMI
hass bias current compensation circuit which gives Improved bias current
over the standard JFET input opamps. IBand losare measured at VCM=O.
2. Settling time is defined herefora unity gain inverter connection using 2kO
resistors. It is the time required for the error voltage (the voltage at the

5·93

1/86, Rev. A

----------IIEMD

OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at Vs =

± 15V, -55°C :S TA:S 125°C, unless otherwise noted.
OP-15A
OP-16A
OP-17A

PARAMETER

SYMBOL CONDITIONS

I nput Offset Voltage

Vos

Average In put
Offset Voltsge Drift
Without External
Trim
With External
Trim

TCVos

Input Offset
Current (Note 1)

Input Bias
Current (Note 1 )

Input Voltage Range
Common-Mode
Rejection Ratio
Power Supply
Rejection Ratio

MIN

OP-15B
OP-16B
OP-17B

TYP

MAX

TYP

MAX

TYP

MAX

UNITS

0.4

0.9

0.7

2.0

0.9

4.5

mV

2

5

3

10

4

15

Rs=50n

MIN

OP-15C
OP-16C
OP-17C
MIN

(Note 2)

I'VI'C
TCVOSn

Rp= l00kn
Tj= 125'C
TA = 125'C
Device Operating

los

Ie

OP-15

3

4

0.6

4.0

0.8

6.0

1.0

0.8

7.0

1.2

11

1.5

17

0.6

4.0

0.8

6.0

1.0

9.0

1.0

8.5

1.3

14.5

1.7

22

9.0

nA
TJ =125'C
TA = 125'C
Device Operating

OP-16/0P-17

TJ =125'C
TA = 125'C
DevIce Operating

OP-15

±1.2

±5.0

±1.5

±7.5

±1.8

±10

±1.7

±90

±2.2

±14

±2.7

±19

±1.2

±5.0

±1.5

±7.5

±1.8

±10

±2.0

±11

±2.5

±18

±3.0

±25

nA
TJ = 125'C
TA = 125'C
DevIce Operating

IVR
CMRR

2

OP-16/0P-17
±104

±10.4

85

VCI~=±104V

97

85

±10.25

VCM = ± 10.25V

80
15

57

V

97
15

dB

93

57

PSRR

Vs= ±10V to ±18V
Vs =±10Vto±15V

Large-Signal
Voltage Gain

Avo

RL ,,2kn
Vo =±10V

35

120

30

110

25

100

V/mV

Output Voltage
Swing

Vo

RL " 10kn

±12

±13

±12

±13

±12

±13

V

23

100

I'VN

NOTES:
1. Input bias current Is specified for two different condItIons. The TJ = 25' C
speCification is with the junction at ambient temperature; the Device
Operating speCification Is with the device operating In a warmed-up
condition at 25'C ambient. The warmed-up bias current value Is correlated to the junction temperature value via the curves of Ie vs TJ and Ie vs
TA• PMI hasa bias current compensation circuit which gives Improved bias
current over the stsndard JFET input op amps. Ie and los are measured
at VCM=O.
2. Sample tested.

5-94

1/86, Rev. A

~

OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at Vs = ±15V,

PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

Average Input
Offset Voltage Drift
Without External
Trim

With External

Tnm

OP-15F
OP-16F
OP-17F
MIN TYP MAX

OP-15G
OP-16G
OP-17G
MIN TYP MAX

03

0.55

1.5

07

3

10

3.S

UNITS
mV

(Note 2)
15
~VI'C

TCVOSn

Rp= 100kn

3

OP-15

Device Operating

los

OP-16/0P-17

OP-15

Device Operating

I.

4

0.04

0.30

006

0.45

O.OS

0.65

0.06

055

OOS

O.SO

0.10

1.2

0.04

030

0.06

0.45

OOS

065

007

070

010

11

015

1.7

nA
TJ = 70'C
TA = 70'C

TJ = 70'C
TA = 70'C

±010 ±0.40

±012 ±060

±0.14 ±O.SO

±013 ±o 75

±016

±0.19

±11

~
~
......
~
......

......:l

±15
nA

T J = 70'C
TA = 70'C

OP-16/0P-17

Device Operating

Input Voltage Range

075

TCVos

Device Operating

Input Bias
Current (Note 1)

unless otherwise noted.

OP-15E
OP-16E
OP-17E
MIN TYP MAX

Rs= son

T J = 70'C
TA = 70'C
Input Offset
Current (Note 1 )

ooe:s TA:S 70 oe,

±010 ±040

±012 ±060

±015 ±0.90

±020

±104

IVR

85

±1.4

±10.4
98

85

±0.14 ±OSO
±025

-<
......:l

±20

±10.25

V

98

Common-Mode
Rejection Ratio

CMRR

VCM = ±10 4V
VCM = ± 10.25V

Power Supply
Rejection Ratio

PSRR

Vs = ±10V to ±18V
Vs = ±10V to ±15V

Large-Signal
Voltage Gain

Avo

Rl2: 2kn
Vo= ±10V

65

200

50

180

35

160

VlmV

Output Voltage
Swing

Vo

Rl2: 10kn

±12

±13

±12

±13

±12

±13

V

80
13

57

13

dB

94

57
20

100

~VIV

~

0......

~

~

NOTES:
Input bias current IS specified for two different conditions The T J= 25° C
specification IS with the Junction at ambient temperature, the Device
Operating specification IS With the device operating In a warmed-up
condition at 25°C ambient. The warmed-up biBS current value IS correlated to the Junction temperature value via the curves of IBvS T J and 18 vs
T A- PMI hasa bl8S current compensation circuit which gives improved bias
current over the standard JFET input op amps 1B and los are measured

atVCM=O.
Sample tested.

5-95

~

::E

1/86, Rev. A

C5

--------I~ OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

DICE CHARACTERISTICS (125'C TESTED DICE AVAILABLE)

OP-16

OP-15

DIE SIZE 0.068 X 0.056 Inch, 3808 sq. mils
(1.73 X 1.42mm, 2.46 sq. mm)
1.
2.
3.
4.
5.
6.

BALANCE
INVERTING INPUT
NONINVERTING INPUT
VBALANCE
OUTPUT

7. V+

OP-17

DIE SIZE 0.068 X 0.056 Inch, 3808 sq. mils
(1.73 X 1.42mm, 2.46 sq. mm)

DIE SIZE 0.068 X 0.056 Inch, 3808 sq. mils
(1.73 X 1.42mm, 2.46 sq. mm)

5. BALANCE
6. OUTPUT

1.
2.
3.
4.
5.
6.

7. V+

7. V+

1. BALANCE
2. INVERTING INPUT
3. NON INVERTING INPUT

4. V-

BALANCE
INVERTING INPUT
NONINVERTING INPUT
VBALANCE
OUTPUT

For addillonal DICE Informallon refer 10 1986 Data Book, Secllon 2.

WAFER TEST LIMITS at Vs= ± 15V, TA = 25'C for OP-15/16/17N, OP-15/16/17G and OP-15/16/17GR devices; TA = 125' C for
OP-15/16/17NT and OP-15/16/17GT devices, unless otherwise noted.
OP-15NT
OP-16NT
OP-17NT

OP-15N
OP-16N
OP-17N

OP-15GT
OP-16GT
OP-17GT

OP-15G
OP-16G
OP-17G

OP-15GR
OP-16GR
OP-17GR

PARAMETER

SYMBOL CONDITIONS

LIMIT

LIMIT

LIMIT

LIMIT

LIMIT

UNITS

Input Offset Voltage

Vos

Rs= 500

0.9

0.5

2.0

1.0

3.0

mVMAX

Large-Signal
Voltage Gain

Avo

Vo =±10V
RL = 2kO

35

100

30

75

50

VlmV MIN

Input Voltage Range

IVR

±10.4

±10.5

±10.4

±10.5

±10.3

VMIN

Common-Mode
Rejection Ratio

CMRR

VCM = ±IVR

85

86

85

86

82

dB MIN

Power Supply
Rejection Ratio

PSRR

Vs = ± 10V to ±20V
Vs = ±10V to ±15V

57

51

57

51

Output Voltage
Swing

Vo

RL = 10kO
RL = 2kO

Supply Current

ISY

OP-15
OP-16,OP-17

Input Bias Current

19

OP-15
OP-16, OP-17

±9
±11

±14
±18

nAMAX

Input Offset Current

los

OP-15
OP-16, OP·17

7.0
8.5

11,0
14,5

nAMAX

80
±12

±12
±11

±12

±12
±11
4

±12
±11

p.VIV MAX
VMIN

mAMAX

NOTES:
For2S'C characteristics of OP-15/16/17NT and OP-15/16/17GT, see OP-1S/16/17N
and OP-15/16/17G characteristics, respectively.
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal Yield loss, yield after packaging Is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

5-96

1/86, Rev. A

~ OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

TYPICAL ELECTRICAL CHARACTERISTICS at Vs =

PARAMETER

SYMBOL CONDITIONS

Average Input Offset
Drift Unnulled

TCVos

Average Input Offset
Drift Nulled

TCVOSn

Input Offset Current

los

Input Bias Current

IB

Slew Rate

SR

ts

GBW

Closed-Loop
Bandwidth

CLBW

C, unless otherwise noted.

OP-1SNT
OP-16NT
OP-17NT

OP-1SN
OP-16N
OP-17N

OP-1SGT
OP-16GT
OP-17GT

OP-1SG
OP-16G
OP-17G

OP-1SGR
OP-16GR
OP-17GR

TYPICAL

TYPICAL

TYPICAL

TYPICAL

TYPICAL

Rp= 100kn

UNITS

~V/'C

3
3

pA

±15

±15

±15

±15

±15

pA

OP-15
OP-16
OP-17

13
25
60

13
25
60

11
21
50

11
21
50

9
17
40

V/~s

to 0.01%
to 0.05%
to 0.10%

OP-15

4.5
1.5
1.2

4.5
1.5
1.2

4.5
1.5
1.2

4.5
1.5
1.2

4.7
1.6
1.3

to O.Q1%
to 0.05%
to 0.10%

OP-16

3.8
1.2
0.9

3.8
1.2
0.9

3.8
1.2
0.9

3.8
1.2
0.9

4.0
1.3
1.0

OP-17

1.5
0.7
0.6

1.5
0.7
0.6

1.5
0.7
0.6

1.5
0.7
0.6

1.6
OS
0.7

OP-15
OP-16
OP-17

6.0
8.0
30

6.0
8.0
30

5.7
7.6
28

5.7
7.6
28

5.4
7.2
26

MHz

OP-15
OP-16
OP-17

14
19
11

14
19
11

13
18
10

13
1S
10

12
17
9

MHz

20
15

20
15

20
15

20
15

20
15

nVl..jHZ

0.Q1
0.01

0.01
0.01

0.Q1
0.01

0.Q1
0.Q1

0.Q1
0.Q1

pAl..jHZ

3

3

3

AVCL = +1

to 0.01%
to 0.05%
to 0.10%
Gain Bandwidth
Product

0

~VI'C

AVCL = +5

Settling Time
(see settling time
test circuits)

±15V, TA = +25

AVCL = +1
AVCL = +5

Input Noise Voltage
Density

en

f= 100Hz
f = 1000Hz

Input Noise Current
Density

in

f = 100Hz
f = 1000Hz

Input Capacitance

C 'N

~.......
~s

~
~

~....:I

~

0
.......

~

...:I

~

pF

NOTES:
For 25' C characteristics of OP-15/16/1 7NT and OP-15/16/1 7GT. see OP-15/16/1 7N
and OP-15/16/17G characteristics, respectively.

5-97

.....

1/86, Rev. A

0

--------I~ OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (OP-15/0P-16/0P-17)

MAXIMUM OUTPUT SWING
vs LOAD RESISTANCE
0

I IIII

7

~ ~1

4

,
1

f---

1111

VOLTAGE NOISE vs
SOURCE RESISTANCE

±20;------r-----,------;-----~

w

'-'
Z

±15 ~--+--___t--____;,r_-___T_--I

~

"~ "0
0

.,

+l~~~C

2

~--~--_+--+--~-____J

100

;i

+25°C

5

1000 ; - - - - - , - - - - - , - - - - - . . , - - - - - , - - - - ,

~

IU~ ",15J

V

_550C

g

I
I

COMMON-MODE INPUT
VOLTAGE RANGE
vs SUPPLY VOLTAGE

>

10r-----r----+----+--~~~~
~--+--__,!'---------,,L-+_----I

>-

;;;
w

.j;,
~

"0~

o 11-____""'--~-*Ci)AMPLIFIER

,.,.
0

100

lk

10k

lOOk

NEGATIVE

°0~----±~5~----±~,0~----±1~5-----±~2-0

0

<.>

SUPPLY VOLTAGE (VOLTS)

OUTPUT LOAD RESISTANCE (n)

±1:WARMEO-UP
vs"" IN FREE AI1±±
±15V
TA"'25°C

80

I

I

I

I

70 CDUNDERCANCELLED 18

!>-

60

~

40
30

iii0:

50

<.>

~
iii
>-

I

I

= +16pA @VCM '" ~

CD PERFECTLY CANCLLED IB = a @VCM = 0
CD OVERCANCELLED IS = -16pA \@VCM,=

--r-----1

'1 , .A'JaT/

I
----T-t-

20

.,

'

(i)

10

e--+--

f-

;;;

V

V
v'/

~

0~~~~~~~~~~s:S~~~~URED
__
____
____
___ J
~

1M

~

10M

lG

lOG

40
RL eo 2k
TA = 25°C

>"

SOak r---------SSoC
400k
;;- 300k
25°C
200k
w

>-

g
""
i

~

100M

OUTPUT VOLTAGE SWING
vs SUPPLY VOLTAGE

----

-

1----125'C

100k

V

/

RL'2kn-

~

1;

./

-10

lOOk

SOURCE RESISTANCE (n)

1M

§

,

G)/

001L-____

OPEN-LOOP
VOLTAGE GAIN
vs SUPPLY VOLTAGE

INPUT BIAS CURRENT
vs COMMON-MODE VOLTAGE
100
90

NOISE

0N~~~;ON RESISTOR

Z

V

/

L

-20
10k

-30

-12-10-8 -6 -4 -2

0

2

4

6

81012

±5

8

"

7

!

"

±s
±10
±15
SUPPLY VOLTAGE (VOLTS)

".['.----l'-II

'20

'r r I I I /~)V

,",,[\\----jl----j

t;;

5
z

o

laOnA

IIIIII
IIIIII

g ,;".
o~

o

INPUT BIAS CURRENT vs
AMBIENT TEMPERATURE (UNITS
ARE WARMED-UP IN FREE AIR)

OFFSET VOLTAGE DRIFT
vs TEMPERATURE OF
REPRESENTATIVE UNITS

NULLED OFFSET
VOLTAGE DRIFT
vs POTENTIOMETER SIZE

P

±20

±l0
±lS
SUPPLY VOLTAGE (VOLTS)

INPUT COMMON-MODE VOLTAGE (VOLTS)

V

MAX
~~
V' /i:H

1P-rrJ

,/,/

V

100pA

/

-4F--+--~--+-~---+--~--1
-6 L--L__-L__L-~__-L__-L~
-50

-25

o

25

50

TEMPERATURE ec)

5-98

75

100

125

I I

vV'

Vs = ±15V

.- ,/

VS'" ±1SV

III

10pA
10

30

50

70

90

110

130

150

AMBIENT TEMPERATURE (OC)

1/86, Rev. A

---------l~ OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (OP-1S/0P-16/0P-17)
BIAS CURRENT vs TIME
IN FREE AIR
(OP-1S)
'oA

'oA

ISY = ; ~~~ ~g~ ¥:pxc'tuR~i~s
15SA MAX

lDOpA

OP-1SAMAX
lDOpA

lDnA

J

155ATVV

OP-15AMA~V

156A/1S7A TYP

'oA

~~ ~ V VV

= 25"C

ISY '" 6.7mA FOR MAX CURVES
5.0mA FOR TVP CURVES

100pA

OP.,J TYP

V40

60

80

100

120

140

20

TIME AFTER POWER APPLIED (SEC)

60

80

100

120

'~6~,.LLtV / v
V~/
VI~ v-V vtrrr

V:':'5ATYP

'"

10pA
140

'0

30

50

70

90

110

130

150

AMBIENT TEMPERATURE (OC)

TIME AFTER POWER APPLIED (SEC)

INPUT BIAS CURRENT vs
AMBIENT TEMPERATURE (UNITS
ARE WARMED-UP IN FREE AIR)
(OP-16/0P-17)
lDOnA

40

~,/

V

,

I I

10pA

20

/~

"''Y v"...

OP-16A/OP-17A TVP

t-"

V

/~

Vs - +15V

TA

./

o

,JJJ

156A/1S7 A MAX

7 ,L;;p'6A/Or-17~ L

15SA TYP

10pA

lDOnA

-

Vs'" ±15V
TA = 2S"C

V r-

INPUT BIAS CURRENT vs
AMBIENT TEMPERATURE (UNITS
ARE WARMED-UP IN FREE AIR)
(OP-1S)

BIAS CURRENT vs TIME
IN FREE AIR
(OP-16/0P-17)

SUPPLY CURRENT

SUPPLY CURRENT

vs SUPPLY VOLTAGE

vs SUPPLY VOLTAGE
(OP-16/0P-17)

(OP-1S)
35r---r---,---,---,

55~----r-----.-----'-----'

_30~--~--r---r--~

50~----~----+-----4---~

Y

lanA

156A/157A TY

'oA

"

OP-'6A/

V V
,~ :?~ V v~-:lLLlp

lDOpA

1
~

::;
~
-'

"
~

2.5

1-------::-5=5'-::-Cp.-=-b--=f------/
2SoC

_55°C

~

25°C

::>

~ 20~--~--r---r--~

"V v"
30

50

40
125°C

10pA
10

4.5

125°C

70

90

110

130

JUNCTION TEMPERATURE ("C)

150

'5~--~--~--~-~
±s
±10
±15
o
±20
SUPPLY VOLTAGE (VOLTS)

3.5
0

±5

±10

±15

±20

SUPPLY VOLTAGE (VOLTS)

TYPICAL PERFORMANCE CHARACTERISTICS (OP-1S)
LARGE-SIGNAL
TRANSIENT RESPONSE

SMALL-SIGNAL
TRANSIENT RESPONSE

SETTLING TIME

0.5

'.0

'.5

2_0

25

SETTLING TIME (/-Isec)

5-99

1/86, Rev. A

---------f1EMD

OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (OP-1S)

BANDWIDTH VI
TEMPERATURE

CLOSED-LOOP BANDWIDTH
AND PHASE SHIFT VI FREQUENCY
18
16

"12

\..

\

m 10
:s 8
z

;;

-

PHASE

90

~AJGI~ ~ ~l

100

Vs = ±15V

120

TA = 25°C

w

\

-10

\ \
1M

!:i

12

~

~

70

OJ 80
a: 50

~a:

~

~

"

I"-

100

~

'\.
1

~TIVE

~

20

:s

J.,

AV
Vs" ±15V

-

-- r--

POllTlVE

-60

-26

60

76

80

~

80

~a:

100

i

40

~"

20

"'"10k

v

"

1\

,

1Q

100

1k

10k

tOOk

1M

10M 100M

VOLTAGE NOISE DENSITY
VI FREQUENCY
,.0

11111111

~ 120

7'

~

80

!!l0

80

I

It~lt~

~ 100
>
>-

T .. 26°C

~

z

1/1 CORNER FREQUENC

w

~g

~
10M

"

10M 100M

\.

VS" :t1SV
TA" 2ir'C

OUTPUT IMPEDANCE
VI FREQUENCY

"-

1M

1M

FREQUENCY(Hz)

t\.

lOOk

tOOk

\

o

125

SUPPLY

I'\.

10k

~
a:

fe)

T

"-

........

0

z

25

~6"C-

l"-

1k

100

iii

AMBIENT TEMPERATURE

TAo

FREQUENCY (Hz)

100

COMMON-MODE REJECTION
RATIO VI FREQUENCY

11111Il\00

20

lk

10

FR EQUENCY 1Hz)

100 r-1'"T1"TT1!1""-I'"T1"T11Tor-o-n"T11Tor-o-nrmm

10

100

I\.

20

~

0L-~--~--~--~~--~~

10M

I"-

30

'\.

126

"- .OS'~'VE""- "-

NEGATIVE
SUPPLY

r--- r-----

10

16

60

10

1M

-- " "

I\.

40

0

40

o

c5
>

SLEW RATE

30

POWER-SUPPLY REJECTION
RATIO VI FREQUENCY

90
80

60

[\.

w

VI TEMPERATURE

~ 40

FREQUENCY (Hz)

o
~

26

-26

1 r-i

tOOk

~100

80

"
"
;!

s

TA -26"C-

-20

-60

60

o

120

- --

PRODUCT

80

...... .....

110

CLlED-Lla. I ._

......... ~IDTHAV=+l

;;

70

~
~ 4

iii

Z

GAIN BANDWIDTH

t-.....

100

~

< 5%

TEMPERATURE reI

1\

.

±20V IS

o

100M

!i

~

< VS "

I--

VS. ±16V
TA-26°C
AV" +1

1\

~ 20

a

~

12

:!

II

10M
FREQUENCY (Hz)

r-,

;24
18

~
z

16

MAXIMUM OUTPUT SWING
VI FREQUENCY

2.

g

........

AV = +1

\.

-6

;

160

20

:;

200

!\ \

-6

150

190

"

-4

~

180

-2

BANDWIDTH VARIATION FROM
±5V

140

170

\

g

2.

V~ ±1lv

r- ........

VS'" ±15V

130

r-- r- \AV> 10 \

6

"
~

120

28

110

!\

OPEN-LOOP GAIN
VI FREQUENCY

10k

lOOk
FREQUENCV (Hz)

5-100

1M

10M

40
20

o

1

10

100

1k

'Ok

FREQUENCY IHzl

1/86, Rev. A

--------I~ OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (OP-16)

LARGE-SIGNAL
TRANSIENT RESPONSE

SMALL-SIGNAL
TRANSIENT RESPONSE

SETTLING TIME
10

~

~
~

"IE
0

'"z
~

w

'~"

0

>
>-

-.

~:J
0

-10

D.

0

,.

10

20

2.

SETTLING TIME (,lIsee)

CLOSED-LOOP BANDWIDTH
AND PHASE SHIFT
VI FREQUENCY
m
W

\.\.

N
12

~

6
4

g

2
0

00

28

120

+++ ~:(

1 00
1W

2.

100

Vs = ±15V

1~
140

20

80

1 50

16

60

12

.0

\..
\.
\

8 r-AV> 10

tlI
~

PHAJE

\.

aw

i

-

iA

1

26"C

170
180

'\. \
\.
\. \
\

-.

-6
-8

100

1M

10M

100M

28

0
-50

_I

i

=18
~

8

~

~ 4

AV=+l
VS =±15V -

o
10M

-60

tOOk 1M

........

-25

25

---------------------

50

75

100M

1\
1\

\

Vs = ±15V
TA = 25"C

AMBIENT TEMPERATURE (OC)

10M

I\.

20

5-101

--

10k

40

10

"

lk

60

~"'30 ':::-",..
~~E
~
., 20

1M

100

COMMON-MODE REJECTION
RATIO VB FREQUENCY
100

~50

FREQUENCY (Hz)

10

FREQUENCY (Hz)

SLEW RATE
TEMPERATURE

-"
Z40

1"-

1

125

80

1\

o

tOOk

"\.
100

N~GATIJE

so

a~ 12
6

VB

1 1

¥!: ~~~~
AV=+l

~
'"
z 20

I"'\

-20
25
7S
50
TEMPERATURE (OC)

-25

70

I-- t-

'\

BANDWIDTH VARIATION FROM
±5V ... Vs ... ±20V IS < 5%

MAXIMUM OUTPUT SWING
VI FREQUENCY
~24

I~

20

FREQUENCY (Hz)

;;;

[\

200

\. 1\

-10

vJ . ±lsl

TA= 26"C- f--

160

{HASE

~

- r--..

1~

11

AV" +1

OPEN-LOOP GAIN
VB FREQUENCY

BANDWIDTH VI
TEMPERATURE

100

125

o

1

10

100

lk

10k

tOOk

1M

10M 100M

FREOUENCY(Hz)

1/86, Rev. A

-------- - - - - - - - - - - - - - - -

-

----------

---------ilEMD

OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (OP-16)
POWER-SUPPLY REJECTION
RATIO VI FREQUENCY

~

60

- '" '"

50

I--- "---SUPPLY

i!:

40

a:

30

120

iii

110

.E 100
0
;:: 90

"a:

z

80

~

70

0

~

::>
"!

!i!

"'"

"'-

NEGATIVE

"-

~ 100
~

~

80

"-

III

60

SUPPLY

f'-

10

100

1k

10k

tOOk

IWllU
T

= 25°C

c

oz

"'-

1M

1/f CORNER FREQUENC

w

~

40

g

20

~

o

1111111

~ 120

/'

"- I'\.

10

140

T
POSI~IVE-

"'- t"\.

20

"

IJJI11L

TA=~5°C-

"'-

VOLTAGE NOISE DENSITY
VI FREQUENCY

100 r-rnmn,---rnl11Trr-rnl11Trr-rn"1T111l

"""1
I

"""

f'-

OUTPUT IMPEDANCE
VI FREQUENCY

o

10M

tOOk

10k

FREQUENCY (Hz)

1M

10

1

10M

100

1k

10k

FREQUENCY (Hz)

FREQUENCY (Hz)

TYPICAL PERFORMANCE CHARACTERISTICS (OP-17)
SMALL-SIGNAL
TRANSIENT RESPONSE

LARGE-SIGNAL
TRANSIENT RESPONSE

SETTLING TIME
+10 . . - - . . . . - - . . . . , . - - - - ; . - - - , - - - - - ,

~

~
~

Vs = ±15V

TA" 26°C
AV= -5

+5~41~~~--+---r--1

~

il:

i"

a ~-+----+--+---+----;

w

~g

~~~~~~--+---+----;

>-

::>

§
CLOSED-LOOP BANDWIDTH
AND PHASE SHIFT
VI FREQUENCY
28

"\. I.....

Vs = ±15V
TA = 25"C

'\.
l\.

2.

18

AV

RS =

12

90

50

110

Vs

"'-

130

PHASE MARGIN" 63"
150

=1 +5

\

soon

Rf = 200kO

170

\

.\ \
\\
\\

20

~

120

±15V

"

~AIN ~ANDWI~TH

-

f-CLOSEDL:

190

FREQUENCY (Hz)

80

r- r--

-... ~+5J.

BANDWIDTH,

-

'\

"

"\.

t"\.

20

""

Vs = ±15V

FROM

-so

"\.

40

r- TA = 25°C

BANDWIDTH VARIATION

100M

,

80

10

±5V.;;;: Vs.;;;: ±lO V IS

10M

I-- 1---",,"

I

100

" " PRODUCT

30

\\
1M

OPEN-LOOP
FREQUENCY RESPONSE

BANDWIDTH VI
TEMPERATURE

40

AV"~O

20

-10 L-_.w..._-L_-"-.L....._-L_.....J
2.5
05
10
15
20
15
SETTLING TIME (psecl

-25

< 5%

25

50

-20
75

TEMPERATURE (OC)

5-102

100

125

1

1

10

100

1k

10k

tOOk

1M

10M 100M

FREQUENCY (Hz)

1/86, Rev. A

-------.. .1fHD

OP·15/0P·16/0P·17 PRECISION JFET·INPUT OPERATIONAL AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS (OP-17)
SLEW RATE VI
TEMPERATURE

MAXIMUM OUTPUT SWING
VI FREQUENCY
28

110
Vs = ±15V' I

~

~

"z
~
>-

24

I
!

20

I

I
16

::>

:=
6

I

12

""

I

~

IiI

!\
\

i

4

'I

I
lOOk

II

90

i

80

II

r\

!

100

II

I

~
"
~

~~: !~DC 7

...........

............

-50

-25

25

50

75

--

20

Vs = t15V
TA=25"C

I I

10

125

100

100

"

!

i

1'['
;':

i

I

~
40

1.0

%
!< 20

.,

1;

/

1k

10k

t
I

-

'\

-- - -

lk
10k lOOk
FREQUENCY (Hz)

1M

10M 100M

IIIIII I

~ 120

~

I
I

80

w
5"

60
40

g

20

t

o

1

10M

1M

TA=Z5"i

1/1 CORNER FREQUENC

w
"

i

I

100k

~

~

,

I

I

100

Z

I

I

IW11IU

>
>-

I

AV = 10

!

,/

01
10M

il

/

I

/

I'

,

~

.r

./1

I1II1111

A~'~ 100

§ 60

1M

!

140

10

0

i\

.J-

+-- \

f--

Vs = ±15V

>=
a: 80
z

1

\

VOLTAGE NOISE
vs FREQUENCY

100

.!:. 100
0

10k
lOOk
FREQUENCY (Hz)

w
c 40

o
~
z
o
~

OUTPUT IMPEDANCE
vs FREQUENCY
TA=25"C

1k

-t

a:

8

TA=2S"C

100

- - -- - - C--

;J

AMBIENT TEMPERATURE (OC)

.--,---r--.,----,--......,.--,

10

j:: 60

ill

40
10M

\

a:

50

iii

::>

-- ---

80

z
o

........... r-.

POWER·SUPPLY REJECTION
RATIO VI FREQUENCY

ll:

~

r--... r--.....

-

""""-

~

VS= ±15V-

r--...

60

"

100

_I

AV = +5

~ '-!OSITIVE

FREQUENCY {Hz}

120

I

~EGATIVE

70

,

1M

r-.....

COMMON-MODE REJECTION
RATIO VI FREQUENCY

10

100

1k

10k

FREQUENCY (Hz)

FREQUENCY (Hz)

BASIC CONNECTIONS
INPUT OFFSET VOLTAGE NULLING

SETTLING·TIME TEST CIRCUIT - OP-15/0P-16
2kH 0 1%
+15V

V'

:oy o---.-~2k'VnV'0\r'%-+---"-I
5kU

01%
VOUT

-15V

5kU 0 1%

2N4416

+15V

NOTE Vas CAN BE TRIMMED WITH POTENTIOMETERS
RANGING FROM 10kSl TO lMn FOR MOST UNITS
TevOS WILL BE MINIMUM WHEN Vas IS ADJUSTED
WITH A 100kn POTENTIOMETER

5·103

1/86, Rev. A

II

_ _ _ _ _ _ _----j~ OP-15/0P-16/0P-17 PRECISION JFET-INPUT OPERATIONAL AMPLIFIERS

SETTLING-TIME TEST CIRCUIT - OP-17
2kO 0.1%

TYPICAL APPLICATIONS
CURRENT-TO-VOLTAGE AMPLIFIER OUTPUT
A2 5kn

DIGITAL
INPUTS

+10V

MSB

LSB

RAEF
5kn

-

'4

'6

+15V

-1SV

~15V

APPLICATIONS INFORMATION
DYNAMIC OPERATING CONSIDERATIONS
As with most amplifiers, care should be taken with lead dress,
component placement, and supply decoupling in order to
ensure stability. For example, resistors from the output to an
Input should be placed with the body close to the input to
minimize "pick-up" and maximize the frequency of the
feedback pole by minimizing the capacitance from the Input
to ground.
A feedback pole is created when the feedback around any
amplifier is resistive. The parallel resistance and capacitance

from the Input of the device (usually the Inverting Input) to
AC ground set the frequency of the pole. In many Instances
the frequency of this pole Is much greater than the expected
3dB frequency of the closed-loop gain and consequently
there is negligible effect on stability margin. However, If the
feedback pole Is less than approximately six times the
expected 3dB frequency, a lead capacitor should be placed
from the output to the negative Input of the op amp. The value
of the added capacitor should be such that the RC timeconstant of this capacitor and the resistance It parallels Is
greater than, or equal to, the original feedback pole time
constant.

5-104

1/86, Rev. A

OP-20
MICRO POWER OPERATIONAL AMPLIFIER
(SINGLE OR DUAL SUPPLY)
Precision MOllolithlcs Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

+30V. or from dual supplies of ±2.5V to ± 15V. The input
voltage range extends to the negative rail. therefore input
signals down to zero volts can be accomodated when
operating from a single supply.

Low Supply Current ........................ 55",A Max
Single-Supply Operation ................ +5V to +30V
Dual-Supply Operation. . . . . . .. . . . . . . .. ±2.5V to ± 15V
Low Input Offset Voltage .................. 250",V Max
Low Input Offset Voltage Drift .......... 1.5",V/oC Max
High Common-Mode Input Range ... V- to V+ (-1.5V)
High CMRR and PSRR .................... 100dB Min
High Open-Loop Gain . . . . . . . . . . . . . . . . . . . .. 120dB Min
No External Components Required
741 Pinout and Nulling

Precision performance in high-gain applications is readily
obtained when using the OP-20. The B/F grade features a
maximum input offset voltage of 250",V. minimum CMRR of
95dB. and open-loop gain of over 500.000. Quiescent supply
current is a maximum of only 55",A at ±2.5V or BO",A at ± 15V.
The low input offset. high gain. and low power consumption
brings precision performance to portable instruments. satellites. missile control systems. and many other batterypowered applications.

ORDERING INFORMATIONt
PACKAGE
T A =25'C
VosMAX
(MV)
250
250
250
500
500
500
1000

TO-99
a-PIN

HERMETIC
DIP
a-PIN

OP20BJ'
OP20FJ

OP20BZ'
OP20FZ

OP20CJ'
OP20GJ

OP20CZ'
OP20GZ

OP20HJ

OP20HZ

PLASTIC
DIP
a-PIN

PIN CONNECTIONS

OPERATING
TEMPERATURE
RANGE

OP20FP

OP20GP
OP20HP

N.C
8

MIL
IND
COM
MIL
IND
COM
COM

_I::QV: OUT
HN~BAL

TO-99
(J-Sufflx)

4

v- (CAse)

'For devices processed in total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet.
tAil commercial and industrial temperature range parts are available with
burn-in. For ordering Information see 1986 Data Book, Section 2.

EPOXY MINI-DIP
(P-Sulflx)
And
a-PIN HERMETIC DIP
(Z-Sulflx)

GENERAL DESCRIPTION
The OP-20 is a monolithic mlcropower operational amplifier
that can be operated from a Single power supply of +5V to

SIMPLIFIED SCHEMATIC
r-----~------.-----------~--~--~--~--------~----~--~~--~-

1\\

II
\

l/

3100
w

V

"~

80

1'..

r--....

>

~

0

~

;r

120

1111111

f-

Vs = ±15V
RL = 25kn

100

TA = 2SoC

\

0

II

CLOSED-LOOP GAIN
VB FREQUENCY

INPUT OFFSET VOLTAGE
POWER SUPPLY VOLTAGE

TRIMMED OFFSET VOLTAGE
VB TEMPERATURE

60
40

80

--

AVCL = 1000

60

1111 I 1111
AVCL = 100

r-

40

20

~

~~I
1111 I I
AVCL = 1

20

NULLED TO ZERO OFFSET AT 25°C

-25
-75

WITH 10k POT

-25

25

o
175

125

75

o

TEMPERATURE (Oel

-

700

~s = ~15v'

12

. / I--"""

...

70

V~ = ±i5V

50

/

300

-50

50
TEMPERATURE (Oe)

100

150

100k

10k

- --

:=t===
...... ~~
TA = 12SoC

500

100

10
100
1k
FREQUENCY (Hz)

1.0

SUPPLY CURRENT
SUPPLY VOLTAGE

60

200

I

111111

VB

600

400

-100

-20
001

±15

INPUT OFFSET CURRENT
VB TEMPERATURE

INPUT BIAS CURRENT
VB TEMPERATURE
14

±5
±10
SUPPLY VOLTAGE (VOLTS)

I'-.

I

- -"

./

40

,/

30

,/

20

I---- J.--"
~

...,.;

--

TA = 2soC ' -

I I I

~55~

10

o
-100

-50

50
TEMPERATURE (Oe)

SMALL-SIGNAL TRANSIENT RESPONSE

100

150

±5

±10

±1S

SUPPLY VOLTAGE (VOLTS)

LARGE-SIGNAL TRANSIENT RESPONSE

OUTPUT

OUTPUT

Rl

CL

25"~100PF

5-109

1/86, Rev. A

------------t~ OP-20 MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
VOLTAGE NOISE DENSITY
vs FREQUENCY

MAXIMUM OUTPUT VOLTAGE
VB LOAD RESISTANCE
1000

CURRENT NOISE DENSITY
vs FREQUENCY

,---r--,---,---,---'

10

'"~

1

OL-_~

1k

10k

01

lOOk

__

~_~

10

__

100

0.0 1
01

~_~

lk

10k

10

100

1k

10k

FREQUENCY (Hz)

FREQUENCY (Hz)

R LOAD (nJ

.....

TYPICAL APPLICATIONS
TEMPERATURE SENSOR
r------~--------~-----------._-----o+9VOLTS

''''

1kn

100kn

">-""'--0 OUTPUT

DESIGN EQUATIONS
A,VBE

=..!5..!
q

A.VBE = 85,8
~

In

In

(~)
Ie'

('e2)
iC1

!.uVrK]

VOUT'" 10H.6.VBE)
IF

-=

C

1soc

~

60

TA'"
VS=±15V-

28
24

~

40

Vs = ±15V
RL = 10kn

~

20

~

20

16
12

~
"

10

100

1k

10k

lOOk

lk

FREQUENCY (Hz)

10k

lOOk

1M

FREQUENCY (Hz)

5-115

1/86, Rev. A

-------------I~ OP-21 LOW-POWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
VOLTAGE NOISE DENSITY
VB FREQUENCY
1000 r---r----,r--"T""--.,-----,

~

i1:

~
w

~

10

~
!1:

~

1001-"'C-,,-t---I1---+----t----i

""

1.0

~

~

Q

w

'-t---~---+

a
z

CURRENT NOISE DENSITY
vs FREQUENCY

'"

~
~
a:

10 t---t---II---+----t----i

o. 1

'"

r--....

a:

B
0,0 1
01

10~1--~-~1~0-~10-0-~1k---J,0k
FREQUENCY (Hz)

1.

100

1k

10k

FREQUENCY (Hz)

NONINVERTING
SMALL-SIGNAL RESPONSE

NONINVERTING
LARGE-SIGNAL RESPONSE

OUTPUT

40kD.

10kn

100pF

OUTPUT

100pF

5-116

1/86, Rev. A

OP-22
PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER
(SINGLE OR DUAL SUPPLY)
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•

Programmable Supply Current .......... lilA to 400ilA
Single Supply Operation ................ + 3V to + 30V
Dual Supply Operation ................ ± 1.5V to ± l5V
Low Input Offset Voltage ...................... lOOIlV
Low Input Offset Voltage Drift ............. O.75IlV/oC
High Common-Mode Input Range ... V- to V+ (-1.5V)
High CMRR and PSRR ........................ l15dB
High Open-Loop Gain ..................... l800V/mV
±30V Input Overvoltage Protection
Unity-Gain Stable
LM4250 Pinout and Nulling

used in circuits with high closed-loop gain. The low offsets
and high gain accuracy of the OP-22 bring precision performance to the micropower field.

GENERAL DESCRIPTION

The OP-22 is a versatile op amp designed for operation from
battery or solar-cell power sources. Supply current is programmable over a range of lilA to 400ilA with a single external resistor. Input voltage range is very wide and extends
down to the negative rail, thus the common-mode input voltage range includes ground when operating from a single
supply voltage. This ability to provide high DC performance
over a wide input range is particularly useful in single-battery
applications. In addition, the OP-22 is characterized over a
wide supply range of ±1.5V-to ±15V, or +3V to +30V for
single supply.

The OP-22 is a monolithic micropower operational amplifier
designed to provide excellent accuracy in high-gain applications. Offsets are very low which generally eliminates any
need for external nulling of Vos. The OP-22 is internally
compensated and unity-gain stable. It also features high
open-loop gain, CMRR, and PSRR. This assures good gain
accuracy and rejection of powersupplyvariationseven when

The OP-22 pin-out and offset nulling are identical to the
LM4250 and many other micropower operational amplifiers.
This functional commonality allows easy upgrading of system performance. By selection of set resistor value, the
circuit designer can readily use the OP-22 in place of such
amplifiers as the LM108, LMl12, LM4250,IlA776, and ICL8021
in high-gain, low-frequency applications.

SIMPLIFIED SCHEMATIC

r-------~------~--------~----------~--~--------~--~------~--~--~--Ov.

OUTPUT

-IN

.,N 0---+---+-------+------1~...J

+--

c,

~~+_------_H~-+--------~--+---~--~------~--------+-------~--~---*--Ov-

5-117

1/86, Rev. A

---------I1fMD

OP-22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

(Note 2)
NOTES:

Supply Voltage ..••••.•••.•••.•.•••••••••••••••••• ± 18V
Power Dissipation (Note 1) •.••••••••••••••••.••• SOOmW

See table for maximum ambient temperature rating.

Differential Input Voltage .•••••••••••••••••.••••••• ±30V

Absolute raMgsapplyto both DICE and packaged parts. unless otherWise
noted.

Input Voltage ••••.•••••.•... " ••••••••••• Supply Voltage
Storage Temperature Range
J and Z Packages. • • • . • . . . • • • . • • •• ••

-6So C to + lS0° C

Operating Temperature Range
OP-22A, OP-22B (J or Z package)

-SsoC to +12SoC

OP-22E, OP-22F (J or Z package)

•.•
••••

MAXIMUM AMBIENT
TEMPERATURE
Vs = ±15V
and ISET= 10~A

-2SoC to +8So C

OP-22HJ, OP-22HZ •••••••••••••••.•.•.• O°C to + 70°C
Lead Temperature Range (Soldering, 60 sec) •••••• 300°C
DICE Junction Temperature ••••••••••• -6SoC to +lS0°C

ELECTRICAL CHARACTERISTICS
PARAMETER

SYMBOL

CONDITIONS

OP-22A/E
TYP
MAX
MIN

OP-22B/F
MIN
TYP
MAX

100

200

Vas

Input Offset CUrrent

los

VeM=O

02

Input Bias Current

18

ISET= I~A, VeM =0
ISET = 10"A, VeM = 0

26
19

Input Voltage Range

IVR

V+ =+5V,
V-=OV
Vs = ±15V

Rejection Ratio
Power Supply
Rejection Ratio

(Note 11

CMRR

Vs =±15V

(Note2)

-15V

PSRR
(Note21

S;

124°C
124°C

at Vs = ± 1.SV to ± lSV, 1 ILA S ISET S 10ILA. TA = +2So C, unless otherwise noted.

Input Offset Voltage

Common-Mode

TO-99IJ)
8-Pin Hermetic DIP IZ)

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

VCM :5 +13 5V

300

MIN

500

400

03
3.0
24

30

75
35

40
30

0/35

0/35

0/35

-15/+135

-15/+135

100

95

115

105

85

18

18

1000

05

-15/+135

Vs =±15Vto±15V,
and v-= av,

OP-22H
TYP
MAX

UNITS
~V

nA
10
50

nA

V

95

10

dB

32

"ViV

V+ = 3V to 30V
Vs = ±15V,

'SET = l,uA.

Large-Signal
Voltage Gain

Ava

1000

1800

500

900

250

500

V/mV

1000

1800

500

900

300

500

V/mV

±O 8

±082

±O 8

±082

±075

±O 8

V

±14

±142

±14

±14.2

±135

±14

V

RL = lOOk!!
Vs = ±15V,
ISET= 10,uA,

RL = 10k!!
Vs =±15V,
Output Voltage

Va

Swmg

ISET = I"A, RL = lOOk!! &
ISET = 10"A, RL = 10k!!
Vs =±15V,
ISET= l"A, RL = lOOk!! &
'SET = lO,uA, RL = 10kC}

AveL = +10,

Closed-Loop
BandWidth

BW

Slew Rate

SR

Vs = ±15V,

250

250

250

kHz

008

008

008

V/"s

ISET = 10"A, RL = 10k!!
Vs = ±15V,

Supply Current
No Load

ISY

ISET = 10"A,
RL = 10kn
Vs = ±15V, ISET = l"A
Vs = ± 15V, ISET = 10"A

15
150

17
170

16
160

19
190

18
180

21
210

"A

Vs = ±1 5V, ISET = l,uA

105
105

125
125

14
140

16
160

17
170

20
200

"A

Vs =±1 5V, ISET = 10"A
NOTES:
Sample tested for Single-supply operation, 100% tested for dual-supply
operation

2, Measured With Vas unnutled and ISET constant

5-118

1/86, Rev. A

~

OP-22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs= ±1.5V to ±15V, 1p.A S ISETS 10p.A, -55°C S TAS+125°C forOP-22AJ/AZ and
OP-22BJ/BZ, -25°C S TA S +85°C for OP-22EJ/EZ and OP-22FJ/FZ, and O°C S TA S +70°C for OP-22HJ and OP-22HZ,
unless otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

TCVos

Unnulled

OP-22A/E
MIN
TYP MAX

OP-22B/F
MIN
TYP MAX

075

15

10

20

15

30

"V/'C

175

400

350

600

500

1200

"V
nA

25

pA/'C

10
50

nA

MIN

OP-22H
TYP MAX

UNITS

Average Input
Offset Voltage

Dnft (Note 1)
Input Offset Voltage

Vas

Input Offset Current

los

VCM=O

TClos

(NOle 1)

Average Input Offset
Current Drift
Input Bias Current
Input Voltage
Range

VS~±15V

Power Supply
Rejection Ratio

-15V:::; VCM :5 +13 2V

(Note 3)

ISET ~ 1"A
ISET ~ 10"A

PSRR

Voltage Gam

Output Voltage

V

99
105

80
80

90
90

dB

32

10

10

32

32

56

p,V/v

400

200

400

100

250

V/mV

1000

300

750

150

300

V/mV

±O 75

±O 65

±O 75

±O 6

±07

200

V

Vs

No Load

ISY

±136

±136

±138

±130

±138

±135

V

Vs == ±15V. ISET = l,uA
Vs = ±15V, ISET -= 10,uA

16
160

18
180

17
170

20
200

20
200

25
250

"A

Vs =±15V, I SET 1.uA
Vs = ± 1 5V, ISET = 10,uA

12
120

14
140

15
150

18
180

19
190

25
250

"A

'=;

NOTES:
Sample tested
2 VCM~ 1 5V

3

ORDERING INFORMATIONt

Measured With Vas unnulled and I SET constant

PIN CONNECTIONS

PACKAGE

(/LV)

TO-99
8-PIN

HERMETIC
DIP
a-PIN

OPERATING
TEMPERATURE
RANGE

300
300
500
500
1000

OP22AJ'
OP22EJ
OP22BJ'
OP22FJ
OP22HJ

OP22AZ'
OP22EZ
OP22BZ'
OP22FZ
OP22HZ

MIL
IND
MIL
IND
COM

TA = 25°C
VOSMAX

P;

~
~

0......

~
~

P;

0

ISET = 10,uA, RL = lOki!

Supply Current

~......

P-<
......
......:I

......:I

V+= 3V to 30V INote 21

= ±15V,
ISET ~ 1"A. RL ~ 100kn &

SWing

0/3.2
-15/+132

80
86

105
115

V- ~ OV.

Vs::O ± ISV,
bOO
I SET ;:; 10jJ.A, RL 10k!!
------,----_.
Vs = ±1 5V,
±o 65
'SET = 1/J-A, RL "'- 10m,!} &
' SET ::;; 10p.A, Rl = 10k!}

Va

4.5
34

Vs::::: ± 1 5V to ± 15V &

'SET =- 1J1A, RL _. 100kn

Ava

7.5
35

0/32
-15/+132

80
90

Vs = ±15V,

Large-Signal

33
27

= ±15V

CMRR

(Note 31

15

5
30

0/32
-15/+132

V+~+5V. V-~OV

Vs

Common-Mode
Rejection Ratio

28
21

ISET ~ 10"A. VCM ~ 0

IVR

05

10

ISET ::::: lJ.tA, VCM :::: 0

18

03

02

.. NOTE DO NOT SHORT PIN 8 TO v- OR GROUND
SEE APPLICATIONS INFORMATION

ISET*
8

~e·

IN 2

- ! -

+IN 3

BAL

6 OUT

5 BAL
4

v- (CASE)

'For devices processed in total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet

TO-99
(J-Sufflx)

tAli commercial and Industrial temperature range parts are available With
burn-in. For ordering information see 1986 Data Book, Section 2

5-119

8-PIN HERMETIC DIP
(Z-Sufflx)

1/86, Rev, A

---------I~ OP-22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS

1. BALANCE
2. INVERTING INPUT
3. NON INVERTING INPUT

4. V5.
6.
7.
8.

For additional DICE Information refer to
1986 Data Book, Section 2.

DIE SIZE 0.069 X 0.049 Inch, 3381 sq. mils
(1.75 X 1.24 mm, 2.18 sq. mm)

WAFER TEST LIMITS at Vs = ±1.5V to

±15V, 11'A:5 ISET:5 10I'A, TA

PARAMETER

SYMBOL

Input Offset Voltage

VOS

Input Offset Current

los

(Note 11

Input Bias Current

18

ISET= 11'A
ISET= 1Ol'A

Input Voltage Range

IVR

V+ = +5V, V- = OV
Vs=±15V

Common-Mode
Rejection Ratio

CMRR

Vs = ±15V, -15V S VCM S +13,5V
(Note 2)

Power Supply
Rejection Ratio

PSRR

Vs =±1,5Vto±15V
V- = OV, V+ = 3V to 30V (Note 21

Large-Signal
Voltage Gain

Output Voltage
Swing

Supply Current
No Load

Avo

Vo

ISY

BALANCE
OUTPUT
V+
ISET

CONDITIONS

(Note 11

= 25° C, unless otherwise noted.

OP-22N

OP-22G

OP-22GR

LIMIT

LIMIT

LIMIT

UNITS

300

500

1000

I'VMAX

3

nAMAX

30

7,5
35

10
50

nAMAX

0/3,5
-15/+13,5

0/3,5
-151+13,5

0/3,5
-15/+13,5

VMIN

100

95

85

dB MIN

18

32

I'VIV MIN

Vs= ±15V,
ISET= 11'A, RL = 100kO,

1000

500

250

V/mV MIN

Vs = ±15V,
ISET= 10I'A, RL = 10kO,

1000

500

300

V/mV MIN

Vs = ±1,5V,
ISET= 11'A. RL = 100kO &
ISET= 10I'A, RL = 10kO,

±0,8

±0,8

±0,75

VMIN

Vs =±15V,
ISET = 11'A, RL = 100kO &
ISET = lO/lA, RL = 10kO,

±14

±14

±13,5

VMIN

Vs =±15V,l sET =1I'A,
Vs = ±15V, ISET = lO"A

17
170

19
190

21
210

I'AMAX

Vs =±1,5V, ISET= 11'A.
Vs = ±1,5V, ISET= 1OI'A,

12,5
125

16
160

20
200

~MAX

NOTES:
1, VCM=O
2, Measured with Vas un nulled and ISET held constant.
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice, Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing,

TYPICAL ELECTRICAL CHARACTERISTICS at
noted,
PARAMETER

SYMBOL

CONDITIONS

Average Input
Offset Voltage Drift

TCVos

Unnulled

Large-Signal
Voltage Gain

Avo

Vs= ±15V
ISET = 1/lA, RL = 100kO &
ISET = 10I'A, RL = 10kll

Vs

= ±1.5V to ±15V, 11'A :5ISET:510I'A, TA = +25°C, unless otherwise

5-120

OP-22N

OP-22G

OP-22GR

TYPICAL

TYPICAL

TYPICAL

UNITS

1,0

1,5

2,5

I'V/'C

1800

900

500

V/mV

1/86, Rev. A

----------1~ OP·22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

,.

1000

'00
TA .. 25 c

IJs 1'!5V I

e
Vs • ±1.5V

VS·±15~

ASET TO GND

;;

j

100

I-

Vs· ±16V
RSETTOV-=

10

¥

VS ·±15V

v:.VS ·t1.5V

.3

~

VB

TA -25 Ge

TA ·26 C

'00 F=

BIAS CURRENT
SET CURRENT

SUPPLY CURRENT
VB SET CURRENT

SET CURRENT
SET RESISTOR

VB

ill

'0

'"

a'"

a'"'"

..

~

t;; ,

~

Vs" ±1.SV
RSET TO GND

0,'

~

10

Vs· ±1.6V
RSET TO V-

0.01

100k

'Ok

1M

10M

seT RESISTOR

,~

0,'

0,'

100M

,
'0
SET CURRENT (,uAI

~m

OFFSET CURRENT
VB SET CURRENT
1000

¥:: i~o6CV

0.1

'00

1

'000

'30

TO t16V

I I IIIII

12.

~

'00

~

EAT Vs· ±1.5V TO ±16V
122
1260

./

~5'

~

o

,,

-

/

'0

11.

_55 0

-

1'1

I 1.56' I

,

OP'22~

116

I

"'
110

'D.
'06

'D'
'02

ts .1±1,~vlid ;,I~v
I I IIIII
I

I 11111

"'

HAT Vs· ±1.5V TO t16V

110

, I I IIIII '0

",

'00

PHASE MARGIN

vs SET CURRENT
Vs = ±15V

90

T A"" 2s Q e

VB

,
,

OP·22F

I
I

OP·22H

SLEW RATE
SET CURRENT

¥!: ~~~~

-+-++I+H--t--l-+-+++t-H

.0 ~+:R:+:I:tttt-It111tt11

I

'00

SET CURRENT tuA)

SET CURRENT (pA)

COMMON·MODE REJECTION
VB SET CURRENT
11.

I

'0

'00

seT CURRENT (IlA)

'20

FAT

11.

25'

~

I I IIIII

120

it , 0

'0

F ATVS· 3V TO 30V

12.

100

TA _2e G e

EATVS-3VT030V

12.

>

'00

POWER SUPPLY REJECTION
VB SET CURRENT

OFFSET VOLTAGE
VB SET CURRENT

I

10

seT CURRENT !,uA)

0,0

/

,

/

1
Vs .1±15V
T A " 2sC e

'0
SET CURRENT (IlA)

'00

10
SET CURRENT ("A)

5·121

'00

0.00

,
0,'

,

'0

'00

SET CURRENT (MA)

1/86, Rev. A

_ _ _ _ _ _ _ _---j~ OP·22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
GAIN·BANDWIDTH PRODUCT
vs SET CURRENT
1M

Vs
TA

--

±15V

1:':-

2SOC

15 a

70

145

60

140

1/

13a
12 5

/

!I

40

.....
i---

"'" f-/\.. r-...
~

30

~~ ~"'\.
~ !\:.OiJA
10,AJ
~'\ ~'\.

'SET = l,.A
20

12a

I

Vs = ±15V
TA '" 2SOC
AVCL = 100

50

135

I

lOOk

FREQUENCY RESPONSE
vs SET CURRENT

OPEN·LOOP GAIN
vs SET CURRENT

5,A

10k

10
11 5

"

11 a

1k

V

01

1
10
SET CURRENT CuA)

-10

105

II

I

10a

CLOSED·LOOP
FREQUENCY RESPONSE
120

140

60

-.........

40
20

,

120
100

-

100

10k

1k

100k

12a

......

01

....... .......

"'......

10

40

40

! 111111

vllJU~1

30

I

I111I11

TA = 25°C

30

"
>'"

25

"z

vJ=U,IW

20

~

>-

~

15

1111111

1.5

~

::>
0

~

10

~

25

ISET = l/JA
I

10

'1/

a

"

a

10k

a

01

10

"' "

100

1k

10k

FREQUENCY (Hz)

VOLTAGE NOISE
vs FREQUENCY

+15=\

.II +5-,

il;

20

=10'j,\

a

100k

35

1-

'SET

1k

100

"

a

MAXIMUM OUTPUT CURRENT
vs SET CURRENT AT
Vs = ±15V, +5 AND ±1.5

PEAK-TO·PEAK
OUTPUT SWING
vs LOAD RESISTANCE
35

10a

""-

FREQUENCY (Hz)

FREQUENCY (Hz)

~

"'-rOSIT1VE

NEGATIV~

a

1M

ISET'" 10,.A

14a

20

-60
10

"
>'"

Vs = ±15V

16a

40

-40

1M

18a
Vs'" ±15V
ISET = lO.uA

60

.........

""

lOOk

COMMON-MODE REJECTION
vs FREQUENCY

80

"'

-20

10k

POWER SUPPLY REJECTION
vs FREQUENCY
160

80

lk

FREQUENCY (Hz)

180

~!: ~~~~

100

SET CURRENT (pA)

ISET = 10iJ.A

100

~

100

10

100

-20
10

1-

"z
>-

~

::>

~

+15

<1

10k

.3

>~

~

::>

"::>>-

V': \01'

v-:: ~

1k

rJ....\. -15
-15

v-::

j:
::>

0

100

"-~

0

05

a
100

a
1k

10k

lOOk

LOAD RESISTANCE (nJ

1M

10
01

10
SET CURRENT (pAl

5·122

100

10 L-...I....Ll..LL1.llL.....L...I...Wllll._Ll.illllll
0.1
10
100
FREQUENCY (Hz)

1/86, Rev. A

---------IIEHD

OP·22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
CURRENT NOISE
vs FREQUENCY

SUPPLY CURRENT
vs TEMPERATURE

10

17
Vs - ±15V
TA - 2SOC

Vs" ±15V

~
~

16

V
15

:::::-

~ 14

~

---

BIAS CURRENT
vs TEMPERATURE

'SET

~ l/lA

'SET" lOIlA

II

10

11

iil

'SET'" lilA

01
01

>
~

100

~

150

"
>w
~

140

'SET - 10ilA

ISET

--

'SET = lOMA Vs

=

±1 s v - I - - 20

[--.

3

~~'SET -

r-....

15

:0

2

I--,:;;,;,A Js,,,15VTb,;~ f',

10

lOjlA Vs

+15V

I--

[\

~

1,A

............

is

9
-75

-50

-25

25

50

75

100

o

90
125

-75

-50

-25

~

1
>-

~

::>

is

o
25
50
75
TEMPERATURE (Oel

lOa

!

">-w

"~

100

FREQUENCY (Hz)

"~
~

1
>1l

~, 10

10

4

~

120
Vs = 1 5V

25

5

~

130

~

12

160

'SET = lOJjA

>~ 13

i3

30

170

125

-~
~
.......
>.Lo

~

(:l;

OFFSET CURRENT
vs TEMPERATURE

iSET - 1 TO 10pA

Vs = ±1 5V TO ±15V
'SET = 10pA

80
60

>~

40

1M

20

-- r-- - ~

0
.......

"""" r--

~

VS-+3V_

~

~
(:l;

r-

::>

tu" -20

if

.......... t-..,

~ -40

0

lOOk

........

-60

AV" +1, C L = 30pF
'SET = 50l-lA

-80

Vs = ±15V

-100
-75

-50

-25

25

~

~

10M

100

!

SMALL·SIGNAL
TRANSIENT RESPONSE

OPEN·LOOP GAIN
vs TEMPERATURE

50

75

100

125

10k
-75

-45

TEMPERATURE (Oel

SMALL·SIGNAL
TRANSIENT RESPONSE

-15

15

45

75

105

135

TEMPERATURE (Oel

SMALL-SIGNAL
TRANSIENT RESPONSE

SMALL·SIGNAL
TRANSIENT RESPONSE

AV '" +1, CL '" 30pF

AV = +1, CL = 30pF

AV = +1, C L = 30pF

'SET = lilA
Vs'" ±15V

ISET

Vs" ±15V

'SET = 100tlA
Vs = ±lSV

=

lO/1A

5·123

1/86, Rev. A

----------l1fHD

OP-22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
SMALL-SIGNAL
TRANSIENT RESPONSE

SMALL-SIGNAL
TRANSIENT RESPONSE

SMALL-SIGNAL
TRANSIENT RESPONSE

AV '" +10, CL '" 30pF

'SET = lpA
Vs = ±15V

APPLICATIONS INFORMATION
OP-22 series units may be inserted directly into LM4250,
!LA776 and ICL8021 sockets with or without removal of
external nulling components. The value of set resistor for a
given supply current varies between types and the manufacturer's data sheets should be consulted for this information.
Table 1 compares set resistor values for the OP-22 and the
LM4250. (RSETconnected to V-).

Biasing the OP-22 with a fixed resistor produces a supply
current approximately proportional to supply voltage. In
applications where a constant drain is required with varying
supply, RSET can be replaced by current generators. Two
suggested arrangements are shown below:

la)

TABLE 1
Supply Current VS, Set Resistor for OP-22 and LM4250
ISY= 1O!LA
VSUPPLY
±1.5V
±3.0V
±5.0V
±12V
±15V
ISET

OP-22
2.2Mfl
6.8Mfl
13Mfl
33Mfl
43Mfl
O.67!'A

LM4250
1.3Mfl
2.7Mfl
4.7Mfl
12Mfl
15Mfl
1.B!,A

ISY= 30"A
OP-22
6BOkfl
2.2Mfl
4.3Mfl
11Mfl
15Mfl
20!,A

LM4250
430kfl
910kfl
1.5Mfl
3.9Mfl
5.1Mfl
6.0!'A

ISY= 100"A
OP-22
220kfl
680kfl
1.3Mfl
3.3Mfl
4.3Mfl
6.7!'A

LM4250
120kfl
270kfl
470kfl
1.2Mfl
1.5Mfl
20!,A

Rl

v-

Ib)

SET-RESISTOR SELECTION
10Mn

The value of set resistor for selected supply current may be
calculated using the "Supply current vs. Set current" curve
and the formula;
RSET= (VSUPPLY - 2V SE) .....•...•.............•..... (1)
ISET

2,
2N930

Alternatively, the "Supply Current vs. Set Current" graph may
be used in conjunction with the "Set Current vs. Set Resistor"
graph. VSUPPLY in formula (1) refers to the total supply
voltage with RSET connected between pin 8 and negative
supply. RSET may be connected to ground in which case
VSUPPLY in (1) is the positive supply.

CAUTION: Shorting of pin 8 to negative supply or ground
will cause excessive ISET which in turn will cause excessive
supply current to flow. ISET should always be limited.

5-124

1/86, Rev. A

----------I1fHD

OP-22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

OFFSET VOLTAGE ADJUSTMENT

APPLICATIONS CIRCUITS

The offset vOltage can be trimmed to zero using a 100kO
potentiometer (see offset nulling circuit). Adjustment range
is approximately ±5mV. Resolution of the nulling can be
increased by using a smaller pot in conjunction with fixed
resistors as shown below.

A micropower bandgap voltage reference operating at a
quiescent current of 15/LA may be constructed using an OP22 and a MAT-01 dual transistor (see Figure 1). The circuit
provides a 1.23V reference with better performance than
micropower I.C. shunt regulators and has the advantages of
being a series regulator.
MICROPOWER 1.23 VOLT BANDGAP REFERENCE

39kn

39kn

20kn

PIN 1 o-----v'w----vW~--___v'VV'---__o PIN 5

SUPPLY

+3V TO +30V

1-

lN914

This arrangement has a ±500/LV adjustment range. Offset
nulling of the OP-22 has negligible effect on the value of
TCVos·

Vour'" 1 23V
a-SmA

OFFSET NULLING CIRCUIT
v+
22Mn

OUTPUT
INPUT

ov
QUIESCENT CURRENT, 15~A AT 5V INCREASING TO 20J.,lA AT lOV
OUTPUT VOLTAGE TEMPERATURE COEFFICIENT (OVER OC TO 70"C), 20ppml"C TYP
LINE REGULATION, 0 01%/VOL T
LOAD REGULATION, 000l%/mA

100kn

RSH

Figure 1

v-

GATED MICROPOWER AMPLIFIER
+5V

GAIN'" 50

BURN-IN CIRCUIT*
500kn

+18V
470nF

o--------j I----w>l---+--<-""""
OUTPUT

2!"-- 7
6

OP-22

3+/'
v.
~

INPUT

OUTPUT

OV
36Mn

...............
-lav
Figure 2
·Other CIrcuits may apply at PMl's discretion

5-125

1/86, Rev. A

---------I~ OP-22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

MICROPOWER INSTRUMENTATION AMPLIFIER - POWER DRAIN ~ 3mW WITH ±5 VOLT SUPPLIES
r----.------------------~------._----------------------~~--------------O~v
OFFSET ADJUST

1M!!

1k!!
300kO
01%

l00nF

3OOk!!
01%

MAXIMUM
OUTPUT :t3V
RL ;> 3OkO

22Mn

50pF
OV

50pF

SOp'

-5V

-5V

VOLTAGE GAIN

111m-

Rt = R4
R2 = R3

20kfl:

DUAL TRANSISTOR

MAT.Q1AH

-IN

FD333

R2

R3

R4

1kn01%

1kno 1%

500kn 0 1%

499kn 0 1%

INPUT BIAS CURRENT

2kn CMRR ADJUST

OV
+INo-----------------______________________________________

< lDnA

CMRR> 100dB "80Hz
PSRR > 70de
TC Vas < 02p.V/"C
NONLINEAR lTV < 0 001%
(Av = 500)

~

Figure 3

In Figure 2, the OP-22 is used as a gated amplifier where
power consumption and bandwidth are controllable. Rs can
be selected for a specific lower-power operation or omitted
so the amplifier can be completely shut down.

TWO TERMINAL 4-20mA TRANSMITTER
+5 VOLT

REFERENCE

2mAMAX

A micropower instrumentation amplifier that consumes less
than 3mW with ±5V supplies is shown in Figure 3. Offset
voltage drift is less than 0.2".v/oC and common-mode
input range is ±3V with CMRR of over 100dB at 60Hz.

5kU 1%

Process control systems use two-wire 4-20mA current transmitters when sending analog signals through noisy environments. The "zero" or "offset" current of 4mA may be used to
power the transmitter signal conditioning amplifiers and/or
excite a d.c. transducer. This allows remote signal conditioning without having a remote power source. Power is provided
at the receiving end where the signal current is monitored by
a precision son resistor. The 4-20mA transmitter shown in
Figure 4 has high stability, excellent linearity, and generates
the 4-20mA current output. A 5V reference is available for
powering transducers and micropower amplifiers at a maximum current of 2mA.

SIGNAL
SOURCE
0-10OmV

lOon 1%
80kn 1%

~----~~-------------------110
COMPLIANCE
LINEARITY

Figure 4

5-126

10 VOLT TO 40 VOLT
.

.

I

0002% OF SPAN

LINE REJECTION, OFFSET
0 002%/VOL T
SPAN.
. DOO6%NOLT
TEMPERATURE COEFFICIENTS. OFFSET
. 0002%fC
(_25°C TO +8SoCI
SPAN
a D01%jOe

1/86, Rev. A

-----------l~ OP-22 PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

MICROPOWER WIEN-BRIDGE OSCILLATOR (Pd < 500ILW)

-=-9V

~----------------~--~VOUT
3Vp-p AT 1kHz

Al - OP·22
01, D2 - lN914

THD <05%ATVour=3Vpp

Figure 5

MICROPOWER 5 VOLT REGULATOR

QUIESCENT CURRENT"" 50,uA AT 9V
LOAD REGULATION, 0 001% I rnA
LINE REGULATION, 0 01% / VOLT
TEMPERATURE COEFFICIENT, 40ppm;oC (_25°C TO +85 C)
Q

+6 5V';;; VIN .;;;; +30V

r-----t----o

301kD
1%

100kn
1%
1Mn

l1kH
1%

D1

301kn
1%

+5V
OUTPUT
(10mAMAX)

10M.I1

D2

SET

OUTPUT

20kD

4---~----~--~-----------4----~--~-------4------o0V

Al, A2 - DP·22, Q1 - 2N2222
D1, 02 - MAT·Ol OR MATCHED DIODES

Figure 6

Ie regulators typically draw 2mA to SmA quiescent current
compared to only SOILA with this discrete implementation.
Maximum load current is 10mA as shown, and can be
increased by changing Ql to a power transistor and proportionately increasing the set current of A2.

Figure S shows a micropower Wien-bridge oscillator
designed for battery-powered instrumentation. Output level
is controlled by nonlinear elements Dl and D2. When
adjusted for 3V p-p output, the distortion level is below O.S%
at 1 kHz.
The S volt regulator in Figure 6 is intended for instrumentation requiring good power efficiency. Low-power 3-terminal

5-127

1/86, Rev. A

OP-27
LOW-NOISE PRECISION
OPERATIONAL AMPLIFIER
PrCClSl0U

Monolithics Inc

FEATURES
L

•
•

N I

ow

{ ... 80nV

(0.1 Hz to 10Hz)

p_p
0................. ...............
3nV/.jiiZ'

•

Low Drift ......•..•...•...........•........ 0.2p.VfO C
d
{ . . . . . . .. 2.8V/ p.8 Slew Rite
HI h S
g
pee.. .. .. .. .. .. .. .. .. 8MHz Glln Blndwldth

•
•
•
•

Low Vas ....................................... 10p.V
Excellent CMRR ............... 126dB It VCM of ±11V
High Open-Loop Glln .... .. .. .. .. .. .. .. ... 1.8 Million
Fits 725, OP-07, OP-05, AD510, AD517, 5534A locketl

ORDERING INFORMATIONt

A low input bias current of ±10nA Is achieved by use of a
bias-current-cancellation circuit. Over the military temperature range, this circuit typically holds IB and los to ±20nA
and 15nA respectively.
The output stage has good load driving capability. Aguaranteed swing of ± 10V Into 600n and low output distortion make
the OP-27 an excellent choice for professional audio applications.
PSRR and CMRR exceed 120dB. These characteristics,
coupled with long-term drift of 0.2p.Vlmonth, allow the circuit
designer to achieve performance levels previously attained
only by discrete designs.

PACKAGE
TA = 25'C HIRMETIC HERMETIC
TO-It
DIP
Voa MAX
(pV)
a·PIN
a·PIN
25
26

60
60
100
100

OP27AJ'
OP27EJ
OP27BJ'
OP27FJ
OP27CJ'
OP27GJ

PLASTIC
DIP
a·PIN

OP27AZ'
OP27EZ
OP27SZ'
OP27FZ
OP27CZ'
OP27GZ

LCC

OP27EP
OP27BRC/883
OP27FP
OP27GP

OPERATING
TEMPERATURE
RANGE

PIN CONNECTIONS

MIL
INO/COM
MIL
INO/COM
MIL
INO/COM

BALttiBAL.7V+
-IN2

lOUT

+IN 3

'For device. processed In total compliance to MIL-STD-883. add 1883 after
part number. Consult factory for 883 data sheet.
tAli commercial and Industrial temperature range parts are available with
burn-In. For ordering Information see 1986 Data Book. Section 2.

6 N.C.
4 V-ICASE)

TO-SS
(J-Sufflx)

8-PIN HERMETIC DIP
(Z-Sufflx)
EPOXY MINI-DIP
(P-Sufflx)

GENERAL DESCRIPTION
The OP-27 precision operational amplifier combines the low
offset and drift of the OP-07 with both high-speed and lownoise. Offsets down to 25p.V and drift of 0.6p.V/oC maximum
make the OP-27 Ideal for precision Instrumentation applications. Exceptionally low noise, en = 3.5nV/y'Hz", at 10Hz, a
low 1/f noise corner frequency of 2.7Hz, and high gain (1.8
million), allow accurate high-gain amplification of low-level
signals. A gain-bandwidth product 018MHz and a 2.8V1p.sec
slew rate provides excellent dynamic accuracy In high-speed
data-acquisition systems.

OP-27BRC/883
LCC PACKAGE
(RC-Sulllx)

SIMPLIFIED SCHEMATIC
r-----~--------~------~--~--~----~------~~~_4~v.

OUTPUT
NON·

INVERTING

INPUT!o:'I-t--t"""1-t---lE:-i~
INVERTING
INPUT<>-H..........--<~_ _ _ _ _

-+____--'

• R1 &. A2 ARE PERMANENTL V ADJUSTED
AT WAFER TEST FOR MINIMUM

OFFSET VOL TAoe

~---------+--~----+-------~+-"""~~v-

5-12S

1/86, Rev. A

-----------l1rHD OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER
Low cost, high-volume production of OP-27 is achieved by
using an on-chip zener-zap trimming network. This reliable
and stable offset trimming scheme has proved its effectiveness over many years of production history.

Operating Temperature Range
OP-27A, OP-27B, OP-27C (J, Z, RC) .. ,. -55° C to +125° C
OP-27E, OP-27F, OP-27G (J, Z) ....... -25°C to +85°C
OP-27E, OP-27F, OP-27G (P) ............ 0° C to +70° C
Lead Temperature Range (Soldering, 60 sec) ...... 300° C
DICE Junction Temperature ......... " -65° C to + 150° C

The OP-27 provides excellent performance in low-noise
high-accuracy amplification of low-level Signals. Applications include stable integrators, precision summing amplifiers, precision voltage-threshold detectors, comparators,
and professional audio circuits such as tape-head and
microphone preamplifiers.

NOTES:

1.

The OP-27 is a direct replacement for725, OP-06, OP-07 and
OP-05 amplifiers; 741 types may be directly replaced by
removing the 741 's nulling potentiometer.

See table for maximum ambient temperature rating and derating factor.
MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

PACKAGE TYPE

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (JI

BO'C

7.1mW/'C

B-Pin Hermetic DI P (ZI

7S'C

6.7mW/'C

ABSOLUTE MAXIMUM RATINGS (Note 4)

B-Pin PlastiC DIP (PI

62'C

S.6mW/'C

Supply Voltage ................................... ±22V
Internal Power Dissipation (Note 1) .............. 500mW
Input Voltage (Note 3) ............................ ±22V
Output Short-Circuit Duration ................. Indefinite
Differential Input Voltage (Note 2) ................ ±0.7V
Differential Input Current (Note 2) ............... ±25mA
Storage Temperature Range. . . . . . . . . .. -65° C to + 150° C

LCC

BO'C

7BmW/'C

2.

The OP-27's inputs are protected by back-to-back diodes. Current limiting

resistors are not used in order to achieve low noise. If differential input
3.
4.

voltage exceeds ±0.7V, the input current should be limited to 2SmA.
For supply voltages less than ±22V, the absolute maximum input voltage is
equal to the supply voltage.
Absolute maximum ratings apply to both DICE and packaged parts, unless

otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs = ±15V, TA = 25° C, unless otherwise noted.
OP-27A/E
PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

Long~Term

Vos

Stability

loS

Input Bias Current

Ie

Input Nol.e Voltage

e np _p

Input Noise

Input Nol.e
Current Density

en

In

Input ReslstanceDifferential-Mode

Input ReslstanceCommon-Mode
Input Voltage Range
Common-Mode
Rejection Ratio
Power Supply
Rejection Ratio

Large-Signal
Voltage Gain

RIN

Slew Rate

OP-27B/F
MAX

TYP

MAX

UNITS

25

20

60

30

100

"V

1.0

0.3

1.5

0.4

2.0

"VIM a

50

12

75

nA

MAX

10
0.2

MIN

35
±40

±12

±5S

±15

±80

nA

O.OB

0.18

0.08

0.18

0.09

0.25

"VP-P

fo

3.5

5.5

3.5

5.5

3.8

8.0

fo

3.1

4.5

3.1

4.5

3.3

5.8

3.0

3.8

3.0

3.8

3.2

4.5

0.1 Hz to 10Hz
(Note. 3, 61

= 10Hz (Note 31
=30Hz (Note 31
fo = 1000Hz (Note 31
fo = 10Hz (Notes 3,81
fo = 30Hz (Notes 3.61
fo = 1000Hz INotes 3,81

1.7

4.0

1.7

4.0

1.7

1.0

2.3

1.0

2.3

0.4

0.6

0.4

0.6

1.0
0.4

1.3

INote 71

0.94

= ±11V

VOM

PSRR

Vs

Avo

RL " 2kll, Vo = ±10V
RL c eoon, Vo = ±10V

pA/yHz

0.7

Mil

Gn

±11.0

±12,3

±110

±12.3

±11.0

±12.3

V

114

128

106

123

100

120

dB

=±4V to ±18V

10

10

20

1000

1800

1000

1800

700

800

1500

800

1500

600

1500

RL " 2kll

±12.0

±13.8

±12.0

±13.8

±11.5

±13.5

RL " 80011

±10.0

±11.5

±10.0

±115

±10.0

±ns

17

2.8

17

2.8

1.7

28

RL " 2kn INote 41

5-129

-------~-~-------~-

~---~-.~--

~

nV/yHz

0.8

2.5

CMRR

SR

MIN

±10

IVR

Vo

OP-27C/G

TYP

TYP

R'NOM

Output Voltage

Swing

(Note tl

Vos/Tlme (Note. 2, 31

Input Offset Current

Voltage Density

MIN

1500

"VIV

VlmV

V
VI"s

1/86, Rev. A

-----------I~ OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ±15V, TA = 25°C, unless otherwise noted. (Continued)
OP-27A/E
PARAMETER

SYMBOL CONDITIONS

Gam Bandwidth Prod GBW
Open-Loop Output

(Note41

MIN

TYP

50

80

_P_o_w_e_rC_o_n_s_u_m~p_tlo_n___P~d______V~e~_______________________ 90

Offset Adjustment

Rp~

Range

MIN

TYP

50

80

70

Ve~O,le~O

Resistance

OP-27B/F
MAX

10kn

OP-27C/G
MAX

MIN

TYP

50

80

MHz

70

n

70
140

90

140

100

±40

±40

NOTES:
1 Input offset voltage measurements are performed - 05 seconds after
application of power. AlE grades guaranteed fully warmed-up
2. Long-term input offset veltage stability refers to the average trend line of
Ves vs. Time over extended penods after the first 30 days of operallon
Excluding the initial hour of operation, changes In Ves dUring the first 30

3

MAX

170

±4.0

UNITS

mW
mV

days are typically 2 5/lV - refer to typical performance curve.
Sample tested

4. Guaranteed by design
5

See test CIrcuit and frequency response curve for 0.1 Hz to 10Hz tester.

6
7.

See test CircUit for current nOise measurement
Guaranteed by Input bias current.

ELECTRICAL CHARACTERISTICS for Vs = ±15V, -55°C:s TA:S +125°C, unless otherwise noted.
OP-27B

OP-27A

TCVes
TCVeSn
los
Ie

Input Voltage Range

IVR

Common-Mode
Rejection RatiO

Power Supply
ReJectton RatiO

CMRR

PSRR

Vs

AVO

R L ,,2kll,

Large-Signal

Voltage Gam

70

300

06

03

13

0.4

18

!lVl'C

02

Average Input

Input BIBS Current

200

(Note21
(Note 31

Ves

Input Offset Current

MAX

50

30

Input Offset Voltage

Offset Dnft

TYP

60

(Note II

CONDITIONS

~

Output Voltage
SWing

MIN

UNITS

15

50

22

85

30

135

nA

±20

±60

±28

±95

±35

±150

nA

±103

±115

±103

±115

±10 2

±11.5

V

108

122

100

119

94

116

dB

16

±4 5V to ±18V

VO~±10V

MIN

OP-27C
MAX

MAX

SYMBOL

MIN

TYP

TYP

PARAMETER

4

20

51

!lVN

600

1200

500

1000

300

800

VlmV

±115

±135

±110

±132

±105

±130

v

ELECTRICAL CHARACTERISTICS for Vs = ±15V. -25°C:S TA:S +85°C for OP-27J and OP-27Z. O°C:s TA:S +70°C for
OP-27P. unless otherwise noted.
OP-27F

OP-27E
PARAMETER

SYMBOL

Input Offset Voltage

Ves

Average Input

TCVes
TCVOS n

Offset Drift
Input Offset Current

los

Input BIBS Current

18

Input Voltage Range

IVR

Common-Mode
Rejection RatiO

CMRR

Power Supply
Rejection RatiO

PSRR

Large-Signal
Voltage Gain

AvO

Output Voltage
SWing

Vo

CONDITIONS

MIN

(Note21
(Note 31

Vs

~

OP-27G

TYP

MAX

TYP

MAX

50

40

140

55

220

06

03

13

04

18

!lVI'C

TYP

MAX

20
02

MIN

MIN

UNITS

10

50

14

85

20

135

nA

±14

±60

±1a

±95

±25

±150

nA

±105

±118

±105

±11a

±10 5

±11.8

V

110

124

102

121

96

118

dB

±4 5Vto ±18V

16

15

750

1500

±117

±136

NOTES:

2

32

!lVN

700

1300

450

1000

VlmV

±114

±135

±110

±13.3

V

The TeVos performance IS within the specifications unnulled or when
nulled with Rp ~ 8kll to 20kll TCVes IS 100% tested for AlE grades,
sample tested for B/C/F/G grades
Guaranteed by deSign

Input offset voltage measurements are performed by automated test
equipment approximately 05 seconds after appllcallon of power AlE
grades guaranteed fully warmed-up

5-130

1/86, Rev. A

------------1~ OP-27 LOW-NOISE PRECISION OPERATIONAL AM_P_L_IF_I_ER
_ _ _ _ _ _ _ _ _ __

DICE CHARACTERISTICS
1. NULL
2. (-) INPUT
3. (+) INPUT

4. V6. OUTPUT

7. V+
8. NULL
For additional DICE inlormation reler to
1986 Data Book, Section 2.

DIE SIZE 0.054 X 0.108 Inch, 5832 sq. mila
(1.37 X 2.74mm, 3.76 sq. mm)

WAFER TEST LIMITS at Vs = ± 15V, T A = 25 0 C for
OP-27GT devices, unless otherwise noted.
SYMBOL

CONDITIONS

Input Offset Voltage

Vos

(Note 1)

Input Offset Current

los

PARAMETER

Input Bias Current

I.

Input Voltage Range

IVR

Common-Made
Rejection RatiO

CMRR

VCM = IVR

Power Supply
Rejection RatiO

PSRR

Vs =±4Vto±18V

Large-Signal
Voltage Gain

Avo

R L ,,2kO, Vo =±10V
RL"aOOO, Vo=±10V

Output Voltage Swing

Vo

RL" 2kO
RL" 6000

Power Consumption

Pd

Vo=O

OP-27N, OP-27G,

and

OP-27GR

devices;

TA

= 125

0

C for

OP-27NT

and

OP-27NT
LIMIT

OP-27N
LIMIT

OP-27GT
LIMIT

60

35

200

60

100

p.VMAX

50

35

85

50

75

nAMAX
nAMAX

OP-27G OP-27GR
LIMIT
LIMIT

UNITS

±60

±40

±95

±55

±80

±103

±11

±10.3

±11

±11

VMIN

108

114

100

106

100

dBMIN

10

20

p.VIV MAX

1000
800

800

±12.0
±10.0

±11.5
±10.0

VMIN

140

170

mWMAX

10
1000

600

500

600
±11.5

±12.0
±10.0

±110

140

700

VlmV MIN

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V, TA = +25 0 C, unless otherwise noted.
SYMBOL

CONDITIONS

Average Input Offset
Voltage Dnft

TCVosor
TCVoSn

Nulled or Unnulled
Rp = 8kO to 20kO

Average Input Offset
Current Dnft

TClos

Average Input Bias
Current Dnft

TCI.

Input NOise
Voltage DenSity

en

fo= 10Hz
fo= 30Hz
fo= 1000Hz

Input NOise
Current DenSity

In

fo= 10Hz
fo= 30Hz
fo= 1000Hz

Input NOise Voltage

e np _p

o 1Hz to 10Hz

Slew Rate

SR

R L,,2kO

Gain BandWidth Prod uct

GBW

OP-27N
TYPICAL

OP-27G
TYPICAL

OP-27GR
TYPICAL

UNITS

02

03

04

p.VI'C

80

130

180

pAl'C

100

160

200

pAl'C

35
31
30

35
31
30

3.8
33
32

nVl,jHZ

17
10
04

17
10
04

17
10
0.4

pAl,jHZ

008

008

0.09

p.Vp-p

28

28

28

VII'S

8

8

8

MHz

NOTE:
1 Input offset voltage measurements are performed by automated test
equipment approximately 0.5 seconds after appllcallon of power.

5-131

~

::l
I=l-<

~

~

0>-<

~
f:J.:t

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to variations In assembly methods and normal yield loss, yield after packaging IS not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

PARAMETER

ffi

>-<

1/86, Rev. A

C5

------------l~ OP·27 LOW·NOISE PRECISION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
0.1 Hz TO 10Hzp _p NOISE TESTER
FREQUENCY RESPONSE

A COMPARISON OF
OP AMP VOLTAGE
NOISE SPECTRA

VOLTAGE NOISE DENSITY
vs FREQUENCY

100 1""""1-n-mrrr-TTnnnr"T"l-mmr---rTTlm11

l~~~!OO
~

100

-I";

741

TA-2S0C

I¥

6

I.,--+-+H+++l+-H++f++t+

Vs '" ±15V

'55 P-d-f-t++ttIt--H-+ttttft--H-+ttttH
>
~

4

f-->.cl,4

30

i,tZ..:LJ

>-

i<

1

:;:

-25

25

50

75

100

125

0

-

60

Vs

'\

r-....

50

'\

I"'\.

±15V-

-1 0

10

25

~
9>u

"
""
0

--

it

100

lk
10k lOOk
FREQUENCY (Hz)

1M

t-SLEW

"1'-.

10M 100M

20
15

1\

-25

25

50

TEMPERATURE fC)

5-133

75

50

75

z
<1

100

125

TnTT

100

125

iil

PHASE \ "
MARGIN

t"
~

"" 700

160 w

\

lBO

~

200

r-..
1M

100

140

~

-5

80

120.-

1\
II

-10

-50

O;"27A

25

TA = 25"c
Vs ~ ±15V

~Ll

10

b
~
z
:iI

"
-75

--

~

8 J:

7

I'\.

-25

OP-278

GAIN, PHASE SHIFT vs
FREQUENCY

10

l-- GBW

'\

0

~ .......... -

TEMPERATURE (OC)

SLEW RATE, GAIN-BANDWIDTH
PRODUCT, PHASE MARGIN vs
TEMPERATURE

1'\

o

10

,","",

OP·27C

-75 -50

150

I--"M

I"

20

0

OP-278

01--$1

'"w
5

~

30

TEMPERATURE (OC)

I'\.

70

~

a:
a h

OP-27A

13
0,-

1'\

I I

Vs =±15V-

40
1
>-

III

OPEN-LOOP GAIN vs
FREQUENCY

0

50

~

TIME (SEC)

11 0

INPUT OFFSET CURRENT
vs TEMPERATURE

OP-27C

ufe OIL BATH

20

TIME AFTER POWER ON (MINUTES)

\ \ '\

10

0

-

-...

INPUT BIAS CURRENT
vs TEMPERATURE

~

tu

-

TIME (MONTHS)

V~.~'5)-

I I I I I I
I I I I I

w

OP·27 C/G

-4

OP·27C

OFFSET VOLTAGE CHANGE
DUE TO THERMAL SHOCK

;;

-

A.
-2

I

100

Vs - .,6V

/

./'" h.

I

......

75

T1-+2J

f---

-6

....... OP-278

!"

T~VOSI

-60

~

OP 27A

!'.....

-40

- v . . . .....

-4

I

r::o-

...."

......:;:

-2
OP 27A

>~ -2 0

WARM-UP OFFSET
VOLTAGE DRIFT

I-..

.Y

Opi 7A

0

g

LONG-TERM OFFSET
VOLTAGE DRIFT OF SIX
REPRESENTATIVE UNITS

10M
FREQUENCY (Hz)

220
100M

1/86, Rev. A

------------t1fMD

OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
OPEN-LOOP VOLTAGE GAIN
vs SUPPLY VOLTAGE
28

I

5

o -TA =

.A"

24

Vs

V
/";..-

RL=600n

/

±15V

14

- poslmJE

12

-

16

V
NEGATIVE
SWING

J/

12

IV

~

SWING

10

f/
\

05

00
10

=

16

20

~V

0

18

II ~~Io 2l,ll

RL '" 2kS1
2SOC

MAXIMUM OUTPUT VOLTAGE
vs LOAD RESISTANCE

MAXIMUM OUTPUT SWING
vs FREQUENCY

20

30

40

o

50

10k

1k

TOTAL SUPPLY VOLTAGE (VOLTS)

SMALL-SIGNAL OVERSHOOT
vs CAPACITIVE LOAD

2 mV

It II

-2
10M

lOOk
1M
FREQUENCY (Hz)

100

1k

10k

LOAD RESISTANCE (n)

SMALL-SIGNAL TRANSIENT
RESPONSE

100

TA=2SoC
Vs '" ±15V

'----

t---

LARGE-SIGNAL TRANSIENT
RESPONSE
2

1(; 00 S

I

--

I,gS

+5V

50mV

80

I
-'"

60

/

40

/

20

o

Vs eo ±15V
VIN=100mV
AV"'+l

V

o

OV

OV

500

1000

I

I

1500

2000

-50mV

11111

,;;::

~

=

-5V

AVCl = +1, C L = lSpF

I

I

;;==

I

I

VS'" ±15V

AVCL = +1
Vs = ±15V

TA '" 25°C

TA "'25°C

I

=

-

2500

CAPACITIVE LOAD (pF)

SHORT-CIRCUIT CURRENT
vs TIME
60

COMMON-MODE INPUT RANGE
vs SUPPLY VOLTAGE

CMRR vs FREQUENCY

16r-----~-----r----_r~~~

140

IIIIIIIIH

50

40

30

"-

~

TA = +2SOC

TA'" 25 C
Vs = ±1SV

120

Vs '" ±15V

VCM '" ±10V

~sc~1

-

12 I-----+_--

:\
100

Iscf+)

10

o

I\.

80

0

1

TIME FROM OUTPUT SHORTED TO GROUND (MINUTES)

60
100

1k

I

-12~----t-----1-~~~~---1
-16~

10k

lOOk

1M

o

____

~

±5

____

~

±lO

____

~

____- J

±1S

±20

SUPPLY VOLTAGE (VOLTS)

FREQUENCY (Hz)

5-134

1/86, Rev. A

-----------I~ OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
VOLTAGE NOISE TEST CIRCUIT (0.1 Hz-TO-10Hz)

LOW-FREQUENCY NOISE

O.lIJF
120
80

40

"'r'

""1
-::-

NOTE

-'0

SCOPE
Xl
RIN = lMn

4.3kO

-80

-120
-::- 110kO

o 1Hz TO 10Hz PEAK-TO-PEAK NOISE

ALL CAPACITOR VALUES ARE FOR
NON POLARIZED CAPACITORS ONLY.

NOTE:
Observation time limited to 10 seconds.

OPEN-LOOP VOLTAGE GAIN VB
LOAD RESISTANCE
24

'2
~ 20

f-f--

~Alel,~!bl
Vs

=

~ 140

±15V

1!

z 18

<1
~

PSRR VB FREQUENCY
160

if

16

o
~

Y

'"~ 100

~ '4

~

>12

'"~

~ 8

o

is
g

Z
~
o

~

10
08

06

0'

100

:>

~

ffi

II

~

lk

10k

100k

1=

TA

12 0 ........

25O~-

~

O-paSITI\~

NEGATIVE

~UPPLY

SUPPl Y

60

~

'0
20

10

LOAD RESISTANCE (il)

100

lk

10k

~

lOOk

~

1M

10M 100M

FREQUENCY (Hz)

(see Offset Nulling Circuit). Other potentiometer values from
1kO to 1 MO can be used with a slight degradation (0.1 to
0.2p.V/oC) of TCVos. Trimming to a value other than zero
creates a drift of approximately (Vos/300) p.V/ oC. For example, the change in TCVos will be 0.33p.V;o C if Vos is adjusted
to 100p.V. The offset-voltage adjustment range with a 10kO
potentiometer is ±4mV. If smaller adjustment range is required, the nulling sensitivity can be reduced by using a
smaller pot in conjuction with fixed resistors. For example,
the network below will have a ±280p.V adjustment range.

APPLICATIONS INFORMATION
OP-27 Series units may be inserted directly into 725, OP-06,
OP-07 and OP-05 sockets with or without removal of external
compensation or nulling components. Additionally, the OP27 may be fitted to unnulled 741-type sockets; however, if
conventional 741 nulling circuitry is in use, it should be modified or removed to ensure correct OP-27 operation. OP-27
offset voltage may be nulled to zero (or other desired setting)
using a potentiometer (see Offset Nulling Circuit).
The OP-27 provides stable operation with load capacitances
of up to 2000pF and ± 10V swings; larger capacitances should
be decoupled with a 500 resistor inside the feedback loop.
The OP-27 is unity-gain stable.

47k!1

Thermoelectric voltages generated by dissimilar metals at
the input terminal contacts can degrade the drift performance. Best operation will be obtained when both input
contacts are maintained at the same temperature.

lknpOT

47kn

v+

OFFSET VOLTAGE ADJUSTMENT
The input offset voltage of the OP-27 is trimmed at wafer
level. However, if further adjustment of Vos is necessary, a
10kO trim potentiometer may be used. TCVos is not degraded

5-135

NOISE MEASUREMENTS
To measu re the 80nV peak-to-peak noise specification of the
OP-27 in the 0.1 Hz to 10Hz range, the following precautions
must be observed:
(1) The device has to be warmed-up for at least five minutes.
As shown in the warm-up drift curve, the offset voltage

1/86, Rev. A

----------IIEM!) OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER
typically changes 4/LV dueto increasing chip temperature
after power-up. In the 10-second measurement interval,
these temperature-induced effects can exceed tens-ofnanovolts.

bias-current cancellation circuit. The OP-27A/E has IB and
los of only ±40nA and 3SnA respectively at 25° C. This is
particularly important when the input has a high sourceresistance. In addition, many audio amplifier designers
prefer to use direct coupling. The high IB' Vos, TeVos of
previous designs have made direct coupling difficult, if not
impossible, to use.

(2) For similar reasons, the device has to be well-shielded
from air currents. Shielding minimizes thermocouple
effects.

Voltage noise is inversely proportional to the square-root of
bias current, but current noise is proportional to the squareroot of bias current. The OP-27's noise advantage disappears
when high source-resistors are used. Figures 1, 2, and 3
compare OP-27 observed total noise with the noise performance of other devices in different circuit applications.

(3) Sudden motion in the vicinity of the device can also "feedthrough" to increase the observed noise.
(4) The test time to measure 0.1 Hz-to-10Hz noise should not
exceed 10 seconds. As shown in the noise-testerfrequencyresponse curve, the 0.1 Hz corner is defined by only one
zero. The test time of 10 seconds acts as an additional
zero to eliminate noise contributions from the frequency
band below 0.1 Hz.

Total noise = [(Voltage nOise)2
(resistor noise)2]112

+ (current

noise X RS)2

+

(5) A noise-voltage-density test is recommended when
measuring noise on a large number of units. A 10Hz
noise-voltage-density measurement will correlate well
with a 0.1 Hz-to-10Hz peak-to-peak noise reading, since
both results are determined by the white noise and the
location of the 1/f corner frequency.

Figure 1 shows noise-versus-source-resistance at 1000Hz.
The same plot applies to wideband noise. To use th is plot, just
multiply the vertical scale by the square-root of the
bandwidth.

UNITY-GAIN BUFFER APPLICATIONS
When Rt:51000and the input is driven with a fast, large signal
pulse (> 1V), the output waveform will look as shown in the
pulsed operation diagram below.

NOISE vs SOURCE RESISTANCE
(INCLUDING RESISTOR NOISE)
AT 1000Hz.
100

During the fast feedthrough-like portion of the output, the
input protection diodes effectively short the output to the
input and a current, limited only by the output short-circuit
protection, will be drawn by the signal generator. With
Rt ;:: 5000, the output is capable of handling the current
requirements (IL:5 20mA at 10V); the amplifier will stay in its
active mode and a smooth transition will occur.

50

~;jtI

OP-08/10B

~

mp-oJ

2

1 Rs UNMATCHED

5534

e 9 RS=RS1=10k,RS2=O
2 RS MATCHED
eg RS=10k.RS1~RS2"5k

When Rt > 2kO, a pole will be created with Rt and the
amplifier's input capacitance (8pF) that creates additional
phase shift and reduces phase margin. A small capacitor
(20 to SOpF) in parallel with Rt will eliminate this problem.

OP-27/37

W-~SISTOR
NOISE ONLY
1

50 100

500

1k

itt>
RS2

5k

10k

50k

RS - SOURCE RESISTANCE (m

PULSED OPERATION
Figure 1

At Rs < 1kO, the OP-27's low voltage noise is maintained.
With Rs> 1kO, total noise increases, but is dominated by the
resistor noise rather than current or voltage noise. It is only
beyond Rsof 20kO that current noise starts to dominate. The
argument can be made that current noise is not importanttor
applications with low-to-moderate source resistances. The
crossover between the OP-27 and OP-07 and OP-08 noise
occurs in the 1S-to-40kO region.
Figure 2 shows the 0.1 Hz-to-10Hz peak-to-peak noise. Here
the picture is less favorable; resistor noise is negligible, current noise becomes important because it is inversely proportional to the square-root of frequency. The crossover with the
OP-07 occurs in the 3-to-SkO range depending on whether
balanced or unbalanced source resistors are used (at3kO the
IB' los error also can be three times the Vos spec.).

COMMENTS ON NOISE
The OP-27 is a very low-noise monolithic op amp. The outstanding input voltage noise characteristics of the OP-27 are
achieved mainly by operating the input stage at a high quiescent current. The input bias and offset currents, which would
normally increase, are held to reasonable values by the input-

5-136

1/86, Rev. A

----------I~ OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

10Hz NOISE vs
SOURCE RESISTANCE
(INCLUDES RESISTOR NOISE).

PEAK-TO-PEAK NOISE (0.1 to
10Hz) VB SOURCE RESISTANCE
(INCLUDES RESISTOR NOISE).
1k

100
OP-08/108

553..

500

50

h2

III
OP-07

II:

~

IIII

100

V

~l-07

OP-27/37

, RS UNMATCHED

1 RSUNMATCHED
e!l RS"RS1=lOk,RS2-o
2 RSMATCHED

e9

~~::Ig~~~
100

~.

r~r-2TJ,

5k

10k

1
50

50k

RS - SOURCE RESISTANCE (n)

Figure 3

Therefore, for low-frequency applications, the OP-07 is better than the OP-27/37 when Rs> 3ko.. The only exception is
when gain error is important. Figure 3 illustrates the 10Hz
noise. As expected, the results are between the previous two
figures.

<5000

Typically used in low-frequency
applications.

Magnetic
tapehead

<15000

Low Ie very Important to reduce
self·magnetization problems when
direct coupling is used. OP-27 Ie
can be neglected.

Magnetic
phonograph
cartridges

<15000

Similar need for low Ie in direct
coupled applications. OP-27 will not
introduce any self·magnetization
problem.

Linear variable
differential
transformer

<15000

Used in rugged servo-feedback
applications. Bandwidth of interest is
400Hz to 5kHz.

500

100

1k

Sk

10k

50k

AS - SOURCE RESISTANCE (n)

Figure 4 is an example of a phono pre-amplifier circuit using
the OP-27 for A 1; R1-R2-C1-C2 form a very accurate RIAA
network with standard component values. The popular method
to accomplish RIAA phono equalization is to employ
frequency-dependent feedback around a high-quality gain
block. Properly chosen, an RC network can provide the three
necessary time constants of 3180, 318, and 75/,s.1

COMMENTS

Strain gauge

As2

AUDIO APPLICATIONS

Table 1
SOURCE
IMPEDANCE

~.

rOnpl~~~

The following applications information has been abstracted
from a PMI article in the 12/20/80 issue of Electronic DeSign
magazine and updated.

For reference, typical source resistances of some signal
sources are listed in Table 1.

DEVICE

e 9 RS~10k.RS1~AS2~5k

HRESISTOR

""

500 1k

RsmRS1~lOk,RS2"O

2 RSMATCHED

a 9 RS"',Ok,RSl"'RS2 a Sk

10
50

t/

V

553'

50

Figure 2

~

op-oa/1os

V

For initial equalization accuracy ana stability, precision
metal-film resistors and film capacitors of polystyrene or
polypropylene are recommended since they have low voltage
coefficients, dissipation factors, and dielectric absorption. 4
(High-K ceramic capacitors should be avoided here, though
low-K ceramics-such as NPO types, which have excellent
dissi pation factors, and somewhat lower dielectric absorptioncan be considered for small values.)

C4 (2)

(+1220~(~1

R5

P---9~
-=-

IF ROllOFF
OUT

IN

C3

OPEN-LOOP GAIN
OUTPUT

FREQUENCY
AT:

OP-07

OP-27

OP-37

3Hz

100dB

124dB

125dB

10Hz

100dB

120dB

125dB

30Hz

90dB

110dB

124dB

R1

97 Sk.n

R3

100.11

G = 1kHz GAIN

For further information regarding noise calculations, see
"Minimization of Noise in Op-Amp Applications", Application
Note AN-15.

=0101 (1+*)

Figure 4

5-137

= 98 677 (39 9 dB) AS SHOWN

1/86, Rev. A

----------l1fMD

OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

The OP-27 brings a 3.2nV/y'HZ voltage noise and 0.45
pAly'HZ current noise to this circuit. To minimize noise
from other sources, Rs is set to a value of 1000, which
generates a voltage noise of 1.3nVly'HZ. The noise increases the 3.2nVly'HZ of the amplifier by only 0.7dB. With
a 1kO source, the circuit noise measures 63dB below a 1mV
reference level, unweighted, in a 20kHz noise bandwidth.

The network values of the configuration yield a 50dB gain at
1kHz, and the dc gain is greater than 70dB. Thus, the worstcase output offset is just over 500mV. A single 0.471'F output
capacitor can block this level without affecting the dynamic
range.
The tape head can be coupled directly to the amplifier input,
since the worst-case bias current of BOnA with a 400mH, 100
I'in. head (such as the PRB2H7K) will not be troublesome.

Gain (G) of the circuit at 1kHz can be calculated by the
expression:

One potential tape-head problem is presented by amplifier
bias-current transients which can magnetize a head. The
OP-27 and OP-37 are free of bias-current transients upon
power up or power down. However, it is always advantageous
to control the speed of power supply rise and fall, to eliminate transients.

G = 0.101 (1+ : : )
For the values shown, the gain is just under 100 (or 40dB).
Lower gains can be accommodated by increasing Rs, but
gains higher than 40dB will show more equalization errors
because of the BMHz gain-bandwidth of the OP-27.

In addition, the dc resistance of the head should be carefully
controlled, and preferably below 1kO. For this configuration, the bias-current-induced offset voltage can be greater
than the 100l'V maximum offset if the head resistance is not
sufficiently controlled.

This circuit is capable of very low distortion over its entire
range, generally below 0.01% at levels up to 7V rms. At 3V
output levels, it will produce less than 0.03% total harmonic
distortion at frequencies up to 20kHz.
Capacitor Cs and resistor R4 form a simple -6dB-per-octave
rumble filter, with a corner at 22Hz. As an option, the switchselected shunt capacitor C4, a non polarized electrolytic,
bypasses the low-frequency roll off. Placing the rumble filter's high-pass action after the preamp has the desirable
result of discriminating against the RIAA-amplified lowfrequency noise components and pickup-produced lowfrequency disturbances.

A simple, but effective, fixed-gain transformerless microphone preamp (Fig. 6) amplifies differential signals from lowimpedance microphones by 50dB, and has an input impedance of 2kO. Because of the high working gain of the circuit,
an OP-37 helps to preserve bandwidth, which will be 110kHz.
As the OP-37 is a decompensated device (minimum stable
gain of 5), a dummy resistor, Rp, may be necessary, if the
microphone is to be unplugged. Otherwise the 100% feedback from the open input may cause the amplifier to oscillate.

A preamplifier for NAB tape playback is similar to an RIAA
phono preamp, though more gain is typically demanded,
along with equalization requiring a heavy low-frequency
boost. The circuit in Fig. 4 can be readily modified for tape
use, as shown by Fig. 5.

Common-mode input-noise rejection will depend upon the
match of the bridge-resistor ratios. Either close-tolerance
(0.1%) types should be used, or R4Shouid be trimmed for best
CMRR. All resistors should be metal-film types for best stability and low noise.
Noise performance of this circuit is limited more by the input
resistors R1 and R2 than by the op amp, as R1 and R2 each
generate a 4nVly'HZ noise, while the op amp generates a
3.2nVly'HZ noise. The rms sum of these predominant noise
sources will be about 6nVly'HZ, equivalent to 0.91'V in a
20kHz noise bandwidth, or nearly 61 dB below a 1mV input
signal. Measurements confirm this predicted performance.

a 47J.1F
TAPE

HEAD

R,

c,
R1

313kn

R2

O.OlMF

f

5kn

loon

R1

T1 '" 31BOMS

lkSl

R3

316kn

T2 '" 50ps

Cl

R6

5.u F

loon

Figure 5
LOW IMPEDANCE
MICROPHONE INPUT ~

While the tape-equalization requirement has a flat highfrequency gain above 3kHz (T2 = 50l's), the amplifier need
not be stabilized for unity gain. The decompensated OP-37
provides a greater bandwidth and slew rate. For many applications, the idealized time constants shown may require
trimming of R1 and R2 to optimize frequency response for
non ideal tape-head performance and other factors. s

5-138

(Z= 50 TO 200n)

~,,~
Rl R2

~

R7

1

OUTPUT

10kn

R2

R4

1k"

316M2

Figure 6

1/86, Rev_ A

------------t1£HD

OP-27 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

For applications demanding appreciably lower noise, a highquality microphone-transformer-coupled preamp (Fig. 7)
incorporates the internally-compensated OP-27. Tl is a
JE-11SK-E 1S00/1SkO transformer which provides an optimum source resistance for the OP-27 device. The circuit has
an overall gain of 40dB, the product of the transformer's
voltage setup and the op amp's voltage gain.

eliminated in such cases, but is desirable for higher gains to
eliminate switching transients.
Capacitor C2 and resistor R2 form a 21's time constant in this
circuit, as recommended for optimum transient response by
the transformer manufacturer. With C2 in use, AI must have
unity-gain stability. For situations where the 21's time constant is not necessary, C2 can be deleted, allowing the faster
OP-37 to be employed.
Some comment on noise is appropriate to understand the
capability of this circuit. A 1500 resistor and Rl and R2 gain
resistors connected to a noiseless amplifier will generate 220
nV of noise in a 20kHz bandwidth, or 73dB below a 1mV
reference level. Any practical amplifier can only approach
this noise level; it can never exceed it. With the OP-27 and Tl
specified, the additional noise degradation will be close to
3.6dB (or-69.S referenced to 1mV).

C2
1800pF
R1

R2

121n

11oof!

OUTPUT

References

150[2
SOURCE

*T1 '" JENSEN JE-l 1SK-E

1.

Llpshltz. S.P, "On RIAA Equalization Networks," JAES, Vol 27, June 1979,
p.458-481.

2

Jung, W G , IC Op Amp Cookbook, 2nd Ed., H.W. Sams and Company,
1980

JENSEN TRANSFORMERS

10735 Burbank Blvd
N Hollywood, CA 91601

Jung, W G , Audio IC Op Amp ApplicatIOns, 2nd Ed, H W. Sams and
Company, 1978

Figure 7
4

Gain may be trimmed to other levels, if desired, by adjusting
R2 or R1 . Because of the low offset voltage of the OP-27, the
output offset of this circuit will be very low, 1.7mVor less, for a
40dB gain. The typical output blocking capacitor can be
BURN-IN CIRCUIT

Jung, W G , and Marsh, R.M , "Picking Capacitors," Audio, February &
March, 1980.

5. Otala, M , "Feedback-Generated Phase Nonlinearity in Audio Amplifiers,"
London AES Convention, March 1980, preprint 1976.
6. Stout, D.F, and Kaufman, Moo Handbook of Operational Amplifier Circuit
Design, New York, McGraw Hill, 1976

OFFSET NULLING CIRCUIT

7

6

> - - ' - - - 0 OUTPUT

5-139

1/86, Rev. A

II

II

OP-32
HIGH-SPEED (AveL 2: 10) PROGRAMMABLE MICROPOWER
OPERATIONAL AMPLIFIER (SINGLE OR DUAL SUPPLY)
['recision Monolithics Inc

FEATURES
•
•
•
•
•
•
•
•
•
•
•
•

Programmable Supply Current ............ 1j.1A to 2mA
Single Supply Operation ................ +3V to +30V
Dual Supply Operation ................ ±1.5V to ±15V
Low Input Offset Voltage ...................... 100j.lV
Low Input Offset Voltage Drift .............. 0.5j.1V1°C
High Com/TIon-Mode Input Range ... V- to V+ (-1.5V)
High CMRA and PSRR ........................ 115dB
High Open-Loop Gain . . . . . . . . . . . . . . . . . . . .. 2000VlmV
±30V Input Overvoltage Protection
Fast ............................ 1V1j.1s@ ISY= 300j.lA
LM4250 Pinout
Compensated for Minimum Gain of 10

ORDERING INFORMATIONt
PACKAGE
EPOXY
HERMETIC
DIP
DIP
8-PIN
8-PIN

TA=25'C
YosMAX
(IlY)
300
300

OP32EP

500
500
1000

OP32FP
OP32GP

OPERATING
TEMPERATURE
RANGE

OP32AZ'
OP32EZ
OP32BZ'
OP32FZ
OP32GZ

MIL
INO
MIL
INO
INO

low, and both are stable with changes in temperature, supply
voltage, and set current. High CMRR and PSRR ensure
precision performance when the OP-32 is used with an
unregulated battery or vehicular electrical system.
The wide input voltage range, including the negative supply
or ground, allows use in single-battery applications. The
OP-32 is characterized over a wide supply range of ± 1.5V to
± 15V. This guarantees predictable performance with any
commonly available supply.
The ability to operate at relatively high speed with low power
consumption makes this amplifier ideal for remote applications where power is limited. The programmability allows
each amplifier in a system to be set for the minimum power
consumption necessary for each specific application.
Programmability also makes it possible to adjust the bandwidth and phase shift.
The OP-32 pinout is identical to the LM4250 and many other
micropower operational amplifiers. This allows easy upgrading of system performance.

PIN CONNECTIONS
8-PIN EPOXY DIP
(P-Sufflx)

* For devices processed in total compliance to M1L-STD-883, add /883 after
part number. Consult factory for 883 data sheet.

8-PIN HERMETIC DIP
(Z-Suffix)

t All

commercial and industnal temperature range parts are available with
burn-In. For ordering information see 1986 Data Book, Section 2

GENERAL DESCRIPTION

NOTE: DO ~ SHORT PIN 8 TO v- OR GROUND
seE APPLICATIONS INFORMATION

The OP-32 is a high-speed, high-gain programmable operational amplifier. Both offset voltage and offset current are

SIMPLIFIED SCHEMATIC
r-----~------~------~--------~--~----~~~------~--~~--ov+

+---t-~_f::c_---l'::____,t

Vs = ±1.5V
10

100

i

0

/.

fI"

~

10

~

's= ±1

0

r-f-rn
I- liTII\)
NEGATIVE-

1

10

~-55"
I

If

100

1000

"/

1/

01
100

10

SUPPLY CURRENT ("AI

OPEN-LOOP GAIN
vs SUPPLY CURRENT
150

100
Vs:; ±15V

ISY

TA = 2SOC

450J.lA

fo""'"

~

-

~

10

I--

+15

~

I~

,..-

+10

I,...-

'SY - 150j./A

+5

-5

TA"'25°C

145

--

-10

140
135
Vs· ±15V

130

v! J5l

120

=

115
110

100l1li1&11

105
'0 ,L ....J.....u.ll!.u.,0~J.....l..J..!.Ull,0':..0....J....J....JUJ.J.lJI,000

-15

100
10

1000

100
SUPPLY CURRENT (j.lA)

SUPPLY CURRENT (pA)

1/86, Rev. A

5-145

..

~---

I

I

125

COMMON-MODE VOLTAGE (VOLTS)

-----~----------

100

SUPPLY CURRENT (pAl

MAXIMUM OUTPUT CURRENT
VB SUPPLY CURRENT AT
Vs= ±15V, +5V AND ±1.5V

BIAS CURRENT vs
COMMON-MODE VOLTAGE

1

100

1000

SUPPLY CURRENT (j.lA)

--

---~-~

----=--~---

-

--- ----------------

.~-

-- --

-----~--~----

--::--

---_.-

--------1~ OP·32 HIGH·SPEED PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL DC OPEN-LOOP INPUT-OUTPUT CHARACTERISTICS
Isy= 1mA, RL = 100kO

Isy = 1mA, RL = 10kO
20
15
10

'>

-"
~

;:
~

-5
-10
-15

-25 -20 -15 -10

-5

0

5

10

15

20

25

-20 -15 -10

OUTPUT (VOL lS)

-5

0

5

10

15

20

25

-25 -20 -15 -10

OUTPUT (VOLTS)

-5

0

5

10

15

20

25

OUTPUT (VOL IS)

ISY = 100ILA, RL = 2kO
20

20

20

15

15

15

10

10

'>

~

-"
~

;:

~

10

'>
-"
~

~

;:

-5

~

-10

;:

-5

~

-10

-15

-15

-20

-20
-25 -20 -15 -10 -5

5

10

15

-15
-20
-25 -20 -15 -10 -5

20 25

-5
-10

5

10

15

20

25

-25 -20 -15 -10 -5

OUTPUT (VOLTS)

OUTPUT (VOLTS)

5

10

15

20 25

OUTPUT (VOLTS)

TEST CIRCUIT
20
Vy

15
10

~
~

;:

~

-5
-10
-15

":"

-20

-25 -20 -15 -10

-5

5

10

OUTPUT (VOL TS)

15

20

25

-25 -20 -15 -10 -5

o

5

10

15

20

25

R2»Rl, R3

R2

Vy '" -

Ra

-=

1
AVOL Your

OUTPUT (VOL TS)

5·146

1/86, Rev. A

-------I~ OP-32 HIGH-SPEED PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL AC PERFORMANCE CHARACTERISTICS
GAIN-BANDWIDTH PRODUCT
vs SUPPLY CURRENT

POWER SUPPLY REJECTION
vs SUPPLY CURRENT

SLEW RATE vs
SUPPLY CURRENT

10M

130
Vs

T A =2SoC

OP32E AT Vs = 3V TO 30V

+15V

128

TA - 2S C C

I

126

V

1M

I 1111111

OP32F AT Vs '" 3V TO 30V

124
QP32E AT Vs =- ±1 5V TO ±15V

122

I

120

I 1111111

QP32F AT Vs = ±1 5V TO ±15V

~

lOOk

118

I

100

140

150

COMMON-MODE REJECTION
vs SUPPLY CURRENT
120

v~ ±15~

140

116

'SY = 150l1A

114
130

"

80
60

....

100

20

90

lOOk

10k

1M

-75

-50

-25

FREQUENCY (Hz)

10

II

TA == 25°C

f-

'SY = lmA

oulp1l

01

.......

001

35

10

30

03

25

01

20

003

1

001

1

0003

DISTORTION..",

0003

i-"

0001

o
1k

10k

FREQUENCY (Hz)

75

"

100

104
T A =2S0C
Vs = ±15V

102
100
10

125

100k

VOLTAGE NOISE
vs FREQUENCY

25

-

00003
100

1000

1000

30

'SY = lOOfJA

.......

100

SUPPLY CURRENT lilA)

35

T~O;5~~'11

DISTORTION

20

1\i\'"

l

1k

0
z

10

"~

101b
10k

100

~,

o
100k

0

;

5~A I.

','!v ;5"~ I
0

w

'SY = 150.uA

0

>

SWING

FREQUENCY (Hz)

5-147

15

OUTPUT

I,~y

~
~

0001

1,..001-'

00003
100

50

r- Jp32~

106

TOTAL HARMONIC
DISTORTION vs FREQUENCY

SWING
003

25

108

TEMPERATURE (OC)

TOTAL HARMONIC
DISTORTION vs FREQUENCY

03

110

........
~ r......

40

1k

I

r-lp32~

112
'SY - 150flA

120

110

100

QP32E

118

=

TA =25°C

....

SUPPLY CURRENT I/lA)

COMMON-MODE REJECTION
vs TEMPERATURE

:~ll~l±l~~1I1

120

100

10

SUPPLY CURRENT (MAl

COMMON-MODE REJECTION
vs FREQUENCY

10

I I 1111111

1000

SUPPLY CURRENT (IlA)

o

L

I 1111111

QP32G AT Vs = ±1 5V TO ±15V

110
10

1

100

II IIJIIJI

114
112

IL

10k

116

ISY = 1 SmA

10
01

W In
10

100

FREQUENCY (Hz)

1/86, Rev. A

-------I!EMD

OP-32 HIGH-SPEED PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL AC PERFORMANCE CHARACTERISTICS
SMALL-SIGNAL TRANSIENT RESPONSE vs SUPPLY CURRENT
Isy = 1.51'A

ov

ov

ov

ov

ov

ov

ov

ov

ov

ov

ISy=1.5mA

TEST CIRCUIT
+1SV

10kH

ov
>----+---~OUTPUT

RSET

100kn

ov
-15V

5-148

1/86, Rev. A

-------t~ OP-32 HIGH-SPEED PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL AC PERFORMANCE CHARACTERISTICS
LARGE-SIGNAL TRANSIENT RESPONSE vs SUPPLY CURRENT
ISY= 1.5",A

OV

ov

OV

ov

ov

ov
RFt

RFt = 11kn, R F2 " lQOkn, Al = lMn

RFl '" 110kn, RF2 '" lMn, RL "OPEN

=

11kf':!, RF2 = 100kn, RL

=

lMn

~
>J..l
........
>J-.

~

~

ISY= 150",A

ISY= 750",A

ISY= 450",A

~

~

OV

OV

0
........

~

OV

>J..l
~

OV

OV

RFt = , lkn, RF2 = 10kn, RL = 100kn

0

OV

RFt = 1 lkn, RF2 = 10kn, RL

Isy=1.5mA

= 100kn

RFt = , lkn, RF2 = 10kn, RL = lOOk!,,:!

TEST CIRCUIT
+15V

OV

> - -.......~---o OUTPUT
ov

5-149

1/86, Rev. A

-------l~ OP-32 HIGH-SPEED PROGRAMMABLE MICROPOWER OPERATIONAL AMPLIFIER

APPLICATIONS INFORMATION

CURRENT SETTING CIRCUITS

SETTING SUPPLY CURRENT
The op amp power supply current is determined by the
current flowing out of pin 8. Pin 8 isat the V+ voltage less two
diode drops. which is approximately V+ minus 1.1 V. Do not
connect pin 8 to ground or V- without a set resistor in series
or excessive supply current will be drawn which may damage
the OP-32.

V.--------~~---------------

The set resistor value is selected to make the power supply
current optimum for the specific application. Adjusting the
OP-32 power supply current determines the slew-rate.
bandwidth. and the output current limits (see Performance
Characteristics). The supply current is nominally 15 times the
set current and the set resistor value is calculated from:
Rs =

(VSUPPLY - 1.1 V)
ISET

2N411BA

Rl

V-------~--~~------~----_

(a)

• where Isy"" 151sET

(See graph below)
v+--------~----------~------

Note that the set resistor can go to either negative supply or
to ground. If the set resistor goes to negative supply. then
VSUPPLY= (V+) - (V-). For a single-supply circuit. VSUPPLY is
simply (V+). If the supply voltage varies widely. set current
can be stabilized with circuits (a). (b). or (c).

10Mn

The relationship between supply voltage. supply current and
set current can be approximated by:
2,

.!g ""10+ (V+)-(V-) (TA=250C)

2N930A

6

ISET

R1

The ratio -IISY increases with temperature by approximately
SET
0.05%/oc.

V- - -

------4>___1----____......._ _ _ _ - -

(b)
SUPPLY CURRENT vs TEMPERATURE
1

_I

7
61-- Vs = ±15V -

5

170

.I

r- l,uA

160

l.---r-r
t:.-I-~A

150

4

140

3

130

2

120

1!--Vr '5V 0
9
~

~

~

k

r- 10,uA

v+-------~----------~-----R2

:2''"
"

-;

1
>-

1)

110

*
~

R1

il:

T
~

100

V--------4--~~

~

m

LED

iil

______~-----

90

~

rn

(e)

TEMPERATURE (OC)

INPUT BIAS CURRENT
Input bias current varies directly with set current. The set
current required for a given supply current ranges from
Isy!10.5 at ± 1.5V supply voltage to I sy!15 at ± 15V. Therefore.
I swill be highest at the minimum supply voltage condition of
±1.5V (or 3V) for any given supply current.

5-150

1/86, Rev. A

--------I~ OP-32 HIGH-SPEED PROGRAMMABLE MICRO POWER OPERATIONAL AMPLIFIER

BATTERY-POWERED, GAIN-OF-100 AMPLIFIER

OFFSET NULLING CIRCUIT

R,

v.

5.06kn

R2
500k.l1
.9V

>----0 OUTPUT

INPUT

Vo

Vo = 100VIN

v-

Figure 1

Using an OP-32B/F, we can expect an IB + los/2 of less than
S.SnA when operating at ISY of 1S",A, so the input offset
caused by IBR1 will be negligible (S.SnA x S.OSkO - 43",V).

OFFSET VOLTAGE ADJUSTMENT
The offset voltage can be trimmed to zero using a 100kO
potentiometer (see offset nulling circuit). Adjusting the pot
wiper towards pin S causes the output to go positive.
Adjustment range is approximately ±SmVat Vs= ±1SV. The
Vos adjust range is proportional to supply voltage. Resolution of the nulling can be increased by using a smaller pot in
conjunction with fixed resistors.
If power supply voltages vary widely and the set current is
established by a resistor, the op amp supply currents will vary
in proportion to the supply voltage changes. Vas will remain
almost constant with supply current changes if the null pins
(1 and S) are not used. If a Vos adjust pot is used, current
variations may flow through the offset pot causing an
apparent Vos change. If a Vas adjust pot is used in combination with widely-varying supply voltages, a set-current
stabilizer circuit as shown in (a), (b), or (c) is recommended.

The set resistor Rs needed for a supply current of 1S",A is
calculated from:
Rs =

AVCL= 1 + R,
100 = 1 + SOOkO
R1
:. R1 = S.OSkO

1SV-1.1V
1",A

Offset voltage adjustment is optional. An OP-32B/F has
maximum input offset voltage of SOO",V which would cause
an output offset voltage of SOmV. Drift over temperature is
very low, typically less than 1.0",Vlo C, and is guaranteed to be
less than 2.0",Vlo C. PSRR is also low, only S",VIV, so battery
voltage change_has negligible effect on offset.
Most micropower programmable op amps lose open-loop
gain and CMRR at low supply currents. The OP-32 design
overcomes these limitations so accuracy is maintained at
supply currents of only a few microamps. The OP-32B/F
used in this example has a minimum open-loop gain of over
117dB. Gain error due to finite open-loop gain will be less
than 10017S0,000, which is only 133 PPM. CMRR will typically
be 110dB, an error of 3PPM. Gain accuracy of the circuit is
almost entirely dependent on the accuracy of the R1/R2 ratio;
the op amp contributes less than 0.01S% gain error.

BATTERY-POWERED, GAIN-OF-100 AMPLIFIER
The simple non inverting amplifier circuit shown in Figure 1
provides an accurate gain-of-100 while operating from a pair
of 9V batteries. The circuit requires only 1S",A of supply
current. Slew-rate is approximately O.OSVI",sec and output
swing is ±SV.

R2

VSUPPLy-1.1V
Isy/1S

:. Rs = 1S.9MO

APPLICATIONS EXAMPLE

A value of SOOkO was chosen for R2' For a gain of 100, R1 is
calculated as:

-9V

Considering all error sources, this simple x 100 batterypowered circuit using an OP-32B/F is capable of achieving
excellent accuracy. Without external adjustments of any
kind, output offset will be less than S4mV and gain accuracy
will be better than ±0.01S% (exclusive of R2/R1 error). Gain
linearity, slew-rate symmetry, and stability over temperature
are all excellent with the OP-32, making circuit performance
very predictable.

5-151

1/86, Rev. A

OP-37
LOW-NOISE PRECISION HIGH-SPEED
OPERATIONAL AMPLIFIER (AVCL 2: 5)
Precision Monolithics Inc.

FEATURES
• Low Noise .. . . . . . . . . . . . . . . .. 80nV pop (0.1 Hz to 10Hz)
.. ... ....... ............ 3nV/y'HZ at 1kHz
• Low Drift .................................. 0.2/JoV;o C
• High Speed ......................... 17V1/Jos Slew Rate
................... 63MHz Gain Bandwidth
• Low Input Offset Voltage ....................... 10/JoV
• Excellent CMRR ... 126dB (Common-Voltage of ± 11V)
• High Open-Loop Gain .. . . . . . . . . . . . . . . . . . .. 1,8 Million
• Replaces 725, OP-05, OP-06, OP-07, AD510, AD517,
SE5534In Gains >5

PACKAGE

25
25

60
60
100
100

OP37AJ'
OP37EJ
OP37BJ'
OP37FJ
OP37CJ'
OP37GJ

OP37AZ'
OP37EZ
OP37BZ'
OP37FZ
OP37CZ'
OP37GZ

The low input bias current of± 10nA and offset current of7nAare
achieved by using a bias-current-cancellation circuit. Over
the military temperature range this typically holds I B and los
to ±20nA and 15nA respectively.
The output stage has good load driving capability. A guaranteed swing of ± 10V into 600n and low output distortion make
the OP-37 an excellent choice for professional audio
applications.
PSRR and CMRR exceed 120dB. These characteristics,
coupled with long-term drift of 0.2/JoVimonth, allow the circuit

ORDERING INFORMATIONt
TA = 25'C HERMETIC HERMETIC
DIP
vos MAX TO-99
8-PIN
8-PIN
(.V)

(e n =3.5nVlVHZ at 10Hz),a low 1/f noise corner frequency of
2.7Hz, and the high gain of 1.8 million, allow accurate high-gain
amplification of low-level signals .

PLASTIC
DIP
8·PIN

LCC

OPERATING
TEMPERATURE
RANGE

OP37EP
OP37BRC/883
OP37FP
OP37GP

PIN CONNECTIONS

'-::':~.::"'

MIL
INO/COM
MIL
IND/COM
MIL
IND/COM

"NVNC
4

• For dev,ces processed ,n total compliance to M IL·STD·883. add /883 after
part number. Consult factory for 883 data sheet.
t All commerCial and industrial temperature range parts are avaIlable with
burn-In. For ordering Information see 1986 Data Book, Section 2.

v- (CASE)

8-PIN HERMETIC DIP

(Z-Sufflx)

TO·99

EPOXY MINI-DIP

(J-Sufflx)

(P-Sufflx)

GENERAL DESCRIPTION
The OP-37 provides the same high performance as the OP-27,
but the design is optimized for circuits with gains greater
than five. This design change increases slew rate to 17V1/Josec
and gain-bandwidth product to 63MHz.

OP-37BRC/883
LCC PACKAGE
(RC-Suffix)

The OP-37 provides the low offset and drift of the OP-07 plus
higher speed and lower noise. Offsets down to 25/JoV and drift
of 0.6/JoVioC maximum make the OP-37 ideal for precision
instrumentation applications. Exceptionally low noise

SIMPLIFIED SCHEMATIC
r-------~--------~--------~--~--~r-----~------~--~~--~-ov,

OUTPUT
NON~

INVERTING

INPUT (0.':..1+-+~r-+----j~-k.
INVERTING

INPUT H

~~4--+-+

______

~

______

~

5-152

1/86, Rev. A

----------1~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

designer to achieve performance levels previously attained
only by discrete designs.

Operating Temperature Range
OP-37A, OP-37B, OP-37C (J, Z, RC) .... -55°Cto+125°C
OP-37E, OP-37F, OP-37G (J, Z) ...... -25°C to +B5°C
OP-37E, OP-37F, OP-37G (P) ............ O°C to +70°C
Lead Temperature Range (Soldering, 60 sec) ...... 300°C
DICE Junction Temperature ........... -65°C to +150°C

Low-cost, high-volume production of the OP-37 is achieved by
using on-chip zener-zap trimming. This reliable and stable
offset trimming scheme has proved its effectiveness over
many years of production history.

NOTES:

The OP-37 brings low-noise instrumentation-type performance to such diverse applications as microphone, tapehead, and RIAA phono preamplifiers, high-speed signal conditioning for data acquisition systems, and wide-bandwidth
instrumentation.

1.

See table for maximum ambient temperature rating and derating factor.
MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

PACKAGE TYPE

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99IJI

80'C

7.1mW/'C

8-Pin Hermetic DIP IZI

75'C

S.7mW/'C

ABSOLUTE MAXIMUM RATINGS (Note 4)

8-Pin Plastic DIP (P)

62'C

5.SmW/'C

Supply Voltage ................................... ±22V
Internal Power Dissipation (Note 1) .............. 500mW
Input Voltage (Note 3) ............................ ±22V
Output Short-Circuit Duration ................. Indefinite
Differential Input Voltage (Note 2) ................ ±0.7V
Differential Input Current (Note 2) ............... ±25mA
Storage Temperature Range ........... -65° C to +150° C

LCC

80'C

7.8mW/'C

2.

The OP-37's Inputs are protected by back-to-back diodes. Current limiting

resistors are not used
3.
4.

In

order to achieve low noise. If differential input

voltage exceeds ±O.7V, the input current should be limited to 25,mA.
For supply voltages less than ±22V, the absolute maximum input voltage IS
equal to the supply voltage.
Absolute maximum ratings apply to both DICE and packaged parts, unless
otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs = ±15V, TA = 25° C, unless otherwise noted.
OP-37B/F

OP-37A/E
PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vas

(Note 1)

Long-Term Vas
Stability

Vas/Time

(Notes 2,3)

los

Input BIBS Current

18

Input NOise Voltage

8 np _p

o 1Hz to 10Hz
(Notes 3,5)

to =

Voltage DenSity

fa
fo

'n

Input ReslstanceCommon-Mode
Input Voltage Range
Common-Mode
Rejection Ratio

Power Supply
Rejection Ratio

~
~

10Hz INotes 3, 6)
30Hz I Notes 3, 6)

fo=1000Hz (Notes3,6)

Input ReslstanceDifferential-Mode

10Hz (Note 3)

fo =30Hz (Note 3)
to ~ 1000Hz INote 3)

en

Input Noise
Current DenSity

Voltage Gain
Output Voltage
Swing

MAX

10

MIN

MAX

TYP

MAX

25

20

60

30

100

"V

0.2

10

0.3

15

04

2.0

"V/Mo

50

12

75

nA

±10

±40

±12

±55

±15

±80

nA

008

018

008

018

009

025

"Vp-p

35
31

55
45

35
31

55
45

3.8
33

nV/y"HZ

30

3.8

30

38

32

8.0
56
45

17

40

17

40

1.7

10
0.4

23
06

10

23
06

10
04

0.4
094

IVR
CMRR

VCM = ±11V

PSSR

Vs

~

SR
GBW

0.7

'--"~'--~--

--~-~---

M!!

±123

±11.0

±12.3

V

114

126

106

123

100

120

dB

10
1000

~

fo~

1800
1500

800
250

RL" 2k!!INote 4)
fo

10kHz (Note 4)

10
1000

700

800
250

±13.8
±11.5

±12.0
±10.0

---~~--

~

20

1800
1500

700

1500

400

1500

700

200

500

±13.8
±11.5

±11.5
±10.0

±13.5
±11.5

"V/V

VimV

V

11

17

11

17

11

17

V/"s

45

63
40

45

63
40

45

63
40

MHz

lMHz

---:---~--'~-----

G!!

±11.0

1/86, Rev. A

5-153

-

pA/y"HZ
06

±123

±12.0
±10.0

RL " 2k!!
RL " 600!!

Slew Rate

UNITS

±110

±4V to ±18V

RL " lk!!, Vo ~ ±10V
Avo
RL ~ 600!!, Va ~ ±IV, Vs ~ ±4V, INote 4)

Gain Bandwidth Prod.

MIN

2.5

R 1NCM

Va

OP-37C/G

TYP

13

(Note 7)

R'N

RL" 2k!!, Vo ~ ±10V

large-Signal

TYP

35

Input Offset Current

Input NOise

MIN

--

,-----

----

----,--- '--

--- -

----

.--

--

---------I~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ±15V. TA = 25°C. unless otherwise noted. (Continued)
OP-37A1E
PARAMETER

SYMBOL CONDITIONS

Open-Loop Output

MIN

TYP

Ro

Resistance

Power Consumption

TYP

OP-37C/G

MAX

MIN

70

140

90

Rp = 10k!!

Range

MIN

70

Vo=O

Offset Adjustment

OP-37B/F

MAX

MAX

UNITS

170

mW

!!

70

140

90

±40

TYP

100
±40

±4Q

mV

NOTES:
Input offset voltage measurements are performed by automated test
equrpment approximately 0.5 seconds after application of power AlE
grades guaranteed fully warmed up.
Long-term input offset voltage stability refers to the average trend line of
Vas VB. Time over extended penods after the first 30 days 01 operation.
Excluding the rnitlal hour 01 operation, changes in Vas dunng the first 30

days are typically 2.51-lV - refer to tYPical performance curve
Sample tested
Guaranteed by design.
5. See test circUIt and frequency response curve for 0 1Hz to 10Hz tester.
6 See test CircUit for current noise measurement.
Guaranteed by Input bias current.

3.

ELECTRICAL CHARACTERISTICS for Vs = ± 15V. -55°C:s TA:S +125°C. unless otherwise noted.
OP-37A
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

(Note 11

Average Input

TCVos

(Note 2)

Offset Dnft

TCVOs n

(Note3)

Input Offset Current

los

Input Bias Current

Is

Input Voltage Range

IVR

Common-Mode
Rejection Aatlo
Power Supply
Rejection RatiO

CMRR

VCM == ±10V

PSRR

vs = ±4 5V to ± l8V

MIN

SWing

30
02

OP-37C

TYP

MAX

TYP

MAX

60

50

200

70

300

06

03

13

04

18

MIN

MIN

UNITS

15

50

22

85

30

135

nA

±20

±60

±28

±95

±35

±150

nA

±1t 5

±10.3

±11.5

±10.2

±115

v

108

122

100

119

94

116

dB

16

20

51

600

1200

500

1000

300

800

VlmV

±115

±13.5

±110

±132

±10 5

±13.0

v

AvO

Output Voltage

MAX

±103

Large-Signal
Voltage Gain

OP-37B

TYP

ELECTRICAL CHARACTERISTICS for Vs

= ±15V. -25°C:S TA:S +85°C for OP-37J and OP-37Z. O°C:s TA:S +70° C for

OP-37P. unless otherwise noted.
OP-37E
PARAMETER

SYMBOL

Input Offset Voltage

Vas

Average Input
Offset Drift

TCVOS
TCVoSn

Input Offset Current

los

Input Bias Current

Ie

Input Voltage Range

IVA

Common-Mode
Rejection Aatio

Power Supply
Rejection Ratio

Output Voltage
SWing

MIN

(Note 21
(Note3)

CMRR

PSRR

Vs = ±4.5V to ± 18V

AVO

RL 2: 2k!!. Vo = ±10V

Large-Signal
Voltage Gam

CONDITIONS

OP-37F
MAX

TYP

MAX

50

40

140

55

220

06

03

13

04

18

MAX

20
0.2

MIN

MIN

UNITS

10

50

14

85

20

135

nA

±14

±60

±18

±95

±25

±150

nA

±10 5

±11.8

±10.5

±118

±10.5

±118

v

110

124

102

121

96

118

dB

15

16

32

750

1500

700

1300

450

1000

VlmV

±117

±136

±114

±13.5

±11.0

±133

v

NOTES:
1.

OP-37G

TYP

TYP

2.

The Tevos performance is within the specifications unnulled or when
nulled with Rp =8kil to 20k!!. TCVos is 100% tested for AlE grades. sample
tested for BICIFIG grades.
3. Guaranteed by design.

Input offset voltage measurements are performed by automated test
equipment approximately 0.5 seconds after application of power. AlE
grades guaranteed fully warmed up.

5-154

1/86, Rev. A

---------I~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS
DIE SIZE 0.054 X 0.096 Inch, 5184 sq. mils
(1.37 X 2.44 mm, 3.35 sq. mm)

1.
2.
3.
4.
6.
7.
8.

For additional DICE information refer to
1986 Data Book, Section 2.

NULL
(-) INPUT
(+) INPUT
VOUTPUT
V+
NULL

WAFER TEST LIMITS at Vs = ± 15V, TA = 25° C for OP-37N, OP-37G and OP-37GR devices; TA = 125° C for OP-37NT and
OP-37GT devices, unless otherwise noted.
OP-37NT
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

(Note 1)

Input Offset Current

los

Input 818S Current

Ie

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VcM =±11V

Power Supply
Rejection Ratio

PSRR

TA =25'C, Vs =±4Vto±18V
TA = 125'C, Vs = ±4 5V to ±18V

Large-Signal
Voltage Gain

Ava

Output Voltage Swing

Va

Power Consumption

Pd

RL~2kO,
RL~

RL~
RL~

Vo =±10V
1kO, Vo =±10V

OP-37G OP-37GR

LIMIT

LIMIT

LIMIT

LIMIT

UNITS

60

35

200

60

100

~VMAX

50

35

85

50

75

nAMAX

±60

±40

±95

±55

±80

nAMAX

±103

±11

±10.3

±11

±11

V MIN

108

114

100

106

100

dB MIN

10
16

10

10
20

10

20

600

1000
800

500

1000
800

700

±12.0
±100

±11.0

±120
±100

±11.5
±10.0

VMIN

140

170

mWMAX

±11.5

2kO
6000

OP-37N OP-37GT

LIMIT

140

Vo=O

~VIV

MAX

VlmV MIN

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ±15V,

TA =

OP-37NT
CONDITIONS

Average Input Offset
Voltage Drift

TCVos or

Nulled or Unnulled
Rp = 8kO to 20kO

TeVoSn

+25°C, unless otherwise noted.
OP-37N OP-37GT

OP-37G OP-37GR

TYPICAL

TYPICAL

TYPICAL

TYPICAL

TYPICAL

UNITS

0.2

02

03

03

04

~V/'C

80

80

130

130

180

pAl'C

100

100

160

160

200

pAl'C

nV,jHZ

Average Input Offset
Current Dnft

TClos

Average Input Bias
Cu rrent Drift

TCl e

Input NOise
Voltage DenSity

en

fo= 10Hz
fa = 30Hz
fa = 1000Hz

3.5
31
30

35
31
30

35
31
3.0

35
3.1
30

38
33
32

Input Noise
Current Density

in

fa = 10Hz
fa = 30Hz
fa = 1000Hz

1.7
1.0
04

1.7
10
04

17
10
0.4

17
10
04

17
10

0.08

008

008

008

17

17

17

17

17

VI"s

63

63

63

63

63

MHz

Input NOise Voltage

e np _p

a 1Hz to 10Hz

Slew Rate

SR

RL~

Gain Bandwidth Product

GBW

fo= 10kHz

2kO

pAlVHZ

04
009
~

J.N p _p

-------.~---

NOTE:
Input offset voltage measurements are performed by automated test
equipment approximately a 5 seconds after application of power

R~\I_

5-155

------

"'~-

---

---~-----=-

>.L..

::s
P;

~.....:I

~

0......

~

r::r:

~

0

Electncal tests are performed at wafer probe to the limits shown Due to vanatlons In assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice Consult factory to negotiate specificatIOns based on dice lot qualification through sample lot asembly and testing

SYMBOL

~
......

P;

NOTES:
For25'C charactenstlcsof OP-37NT and OP-37GT deVices, see OP-37N and
OP~37G charactenstlcs, respectively

PARAMETER

~

..- - - - - - -

~~-----

--..

-

~

~

-.

-~-

---

fit

----------l~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

NOISE-TESTER FREQUENCY
RESPONSE (O.1Hz TO 10Hz)
100
90

-

80

\\

'0
9

l~

o2

60

6

,

4

is

>

,

0'

, 0

,

,

,

I

.......

lOW NOISE

AUDIO

III CORNEA
27Hz /

I/f CORNER

'" 2 7Hz

~~ , OP AMP
I IIIII

IJ UllL

INSTRUMENTATION

1
100

1000

10

FREQUENCY (Hz)

INPUT WIDEBAND VOLTAGE
NOISE vs BANDWIDTH (O.1Hz
TO FREQUENCY INDICATED)

III

AVOIO RANGE
TO 20kHz

RANGE, TO DC

'0

100

FREQUENCY (Hz)

1/1 CORNER

-- -

~ ~OP-37

2

I

10

-

l/f CORNER

n'i'IIII~IW FIIIIIIIIICY ITl'l il l GAiNI

30
00'

741

.......

3

"iO

I

TEST TIME OF 10sec MUST BE USED TO

40

r-

TA '" 2S C
Vs = ±15V
D

,>5 I'
~
~

l-

'00

8
7

70

50

A COMPARISON OF
OP AMP VOLTAGE
NOISE SPECTRA

VOLTAGE NOISE DENSITY
VB FREQUENCY

1000

'00

FREQUENCY 1Hz)

TOTAL NOISE vs SOURCE
RESISTANCE

VOLTAGE NOISE DENSITY
VB TEMPERATURE

100 F~=Ff=RR'fFF====="!1

~TA=250C

~ ~

15,V--t-;rtttt'-_ ~ .. rl-_V_'r-=_±t-

1---I--H-rI+++1--

RS '" 2Rl

0',,_
lk

10k

~

~

"~
o

> 21---i--t---r-+--t---t--1

,
100

lOOk

BANDWIDTH (Hz)

VOLTAGE NOISE DENSITY
VB SUPPLY VOLTAGE

lk
SOURCE RESISTANCE

10k

-

-50

-25

25

0

50

75

'00

125

TEMPERATURE (DC)

(m

SUPPLY CURRENT vs
SUPPLY VOLTAGE

CURRENT NOISE DENSITY
VB FREQUENCY
50

100
T!

3 P-i--t-=.,...=-AT 1kHz

w

= 2SoC
40

AT 10Hz

I--

AT

~

lkH~

I

30

20
Ilf CORNER

r- - " , 140Hz
01

'0

20

30

TOTAL SUPPLY VOL TAGE (V+ - V-I (VOLTS)

40

11111111
10

TA "" _5SD C

I

100
lk
FREQUENCY (Hz)

5-156

10
10k

15

25

35

45

TOTAL SUPPLY VOLTAGE (VOLTS)

1/86, Rev. A

-------4~ OP-37

LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

LONG-TERM OFFSET
VOLTAGE DRIFT OF SIX
REPRESENTATIVE UNITS

OFFSET VOLTAGE DRIFT OF
EIGHT REPRESENTATIVE
UNITS V8 TEMPERATURE
60

40

'>
.3

20

~

w

~

...~~

.;::",

-

/

aP-37C
OP-376

V /'

V

- -/'

V .....,<..

~ 2

/

/'

-40

TRIMMING WITH"'
10k POT DOES
NOT CHANGE

T~VOS,

-60
-75

50

25

,

,

0

25

~ -6
:::

o

I

OP-37C/G

6

./'

"~

I
I

5

.L"..

0

-

'"

i'-'

-2
-4

........ OP-37C

100

-6

125 150 175

o

TEMPERATURE (OC)

~

30

I
I

~

~ 25

g
t;;

2SOC

~

~ 15

f-If-f--

}j
z
<

r+-

10

f-r-

Ii

w

40

Vs

=

I I

Vs = ±15V-

±15V

~

i \ OP-37C

OP-37C

[\,1\,

,,~

RESPONSE

i' "- ....
"r--.

BA' I I
IIf I I I

OP-37B

~kI...u

I

40

60

80

100

-25

80

TAI= +2StC

75
70

RL ;;:,. 2kS1

"'"

BO
60
40

65

'"

60

'\.

55

r\.

~

C

\

102

103

104

10 5

FREQUENCY (Hz)

10 6

20

~

15

oj
107

108

Vs ""5V

75

100

125

....-

~

150

-75

-50

-25

-

10
-50

-25

90

60

85

I'..
50

75
70

i

OP 37A

25

50

40

Vs

75

100

60

r- 55

I. ~,l~ II

125

-80

-100

TA -"+2SoC

1-10-

-120
PHASE

30

\

65

-...

0

GAIN, PHASE SHIFT vs
FREQUENCY

80

r- ~W

OP-37B

TEMPERATURE (oC)



I

75

v

~

o ·4

OP-37B

........

50

"~

,-'"

0

OP-37A

r-..

-20

i!;

w

A
"., OPT

WARM-UP OFFSET
VOLTAGE DRIFT

-140

MARGIN

'" 71°
-160

20
AV '" 5

-180

10
"

50

SLEW

-200
45

40

a

+25

+50

+75

TEMPERATURE (OC)

5-157

+100

+125

-10

lOOk

-220
1M

10M

100M

FREQUENCY (Hz)

1/86, Rev. A

---------I~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
OPEN-LOOP VOLTAGE GAIN
VB SUPPLY VOLTAGE

MAXIMUM OUTPUT SWING
V8 FREQUENCY
28

2S

18

v}= 1"lsU

Rl ·2k.l1

.Y

-TALsoc

20

MAXIMUM OUTPUT VOLTAGE
V8 LOAD RESISTANCE

/'

/

~ 24

o
2:.

Rl=lkn

/'-

16

TA = 2S'e

f-

1\

20

poJITIJe

I-

~

" 16

.;'

~

~ /'"

NEGATIVE
SWING

:11
"

IV

~

8

~

4

"..

os

//

12

~

~
~

00

t-

o
10
20
30
40
TOTAL SUPPLY VOLTAGE (VOLTS)

105

TA'" 2SoC
Vs = ±15V

I III

~2

104

50

100

106

10k

lk
LOAD RESISTANCE (11)

FREQUENCY (Hz)

SMALL-SIGNAL OVERSHOOT
VB CAPACITIVE LOAD

:;;;..-

SWING

LARGE-SIGNAL TRANSIENT
RESPONSE

SMALL-SIGNAL TRANSIENT
RESPONSE

80
Vs = ±15V
VIN =20mV

+50mV

AV = +5 (lkn, 260m

BO

I-

0
0

~

i;

40

I

I-

~

a:
~

20

o

+10V

. / i'-"

~

V

-TOV

-somv • • • •

I
o

ov

ov

Vs"' ±15V
Av = +5 (1kil, 2500)
TA = +25°C

Vs = ±15V
AV = +5 (1kil. 250n)
TA =+2s<'C

1000
1600
CAPACITIVE LOAD (pFJ

500

2000

SHORT-CIRCUIT CURRENT
VB TIME
140

60

<'50

120

S
I-

"-

a')
~ 40
::>

"'-=

u

I-

:;

~ 30

U
a:

""-

lBr-----r-----;------r~~-,

l!ll!I"Jvi I

r-

12r-------1---

TA =+2SoC
VCM = tlDV

I"

T A = +250
Vs = ±15V

r--.'sc~1

COMMON-MODE INPUT RANGE
VB SUPPLY VOLTAGE

CMRR VI FREQUENCY

ac~
ll:

iii 100
:!1.
a:
a:

Ej

~

~,

z

~ 20

-41-----~~~+----_+_----__1

! ~81------I----,3~

ISC (+)

""o

41-----:i;d~~+----+----_l

.. 01----7'-+-----+----_+_----__1

\

80

8

60

___

-121------\----+-~~..---_l

10

o

TIME FROM OUTPUT SHORTED TO GROUND (MINUTES)

-16 L....____- ' -____- ' -____--'-____--.J

40
lk

10k

lOOk
FREQUENCY (Hz)

5-158

1M

10M

±S

±15

±20

SUPPLY VOLTAGE (VOLTS)

1/86, Rev. A

---------l~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
OPEN-LOOP VOLTAGE GAIN
vs LOAD RESISTANCE

LOW-FREQUENCY NOISE

NOISE TEST CIRCUIT (O.1Hz TO 10Hz)
120

~

80

~

40

o
z
w
t:l

~

o
>

VOLTAGE GAIN

- 50,000

-40

-80
-120

a 1Hz TO

10Hz PEAK-TO-PEAK NOISE

1--H1'+t+ttt+-+-H-iffl-t-+-H-I+ttti

04L-~~~L-~~llllL-~-U~

NOTE:
NOTE

0.6

lOa

lk

Observation time limited to 10 seconds,

ALL CAPACITOR VALUES ARE FOR
NON POLARIZED CAPACITORS ONLY

10k

LOAD RESISTANCE

100k

1m

SLEW RATE
PSRR vs FREQUENCY
160

140
120

100
80

TA

-~

1=

25O~_

SLEW RATE vs LOAD
19

t--

TA = J25 0 C

AV '" +5

15

NEGATIVE

~UPPLY

17

~

16

60

0
20

lk

10k

~ ~

AVCL = +5

VO" 20V p-p

SUPPLY

100

20

tAl. ~Wl"
Vs = ±15V

18

-POSITI~

10

vs SUPPLY VOLTAGE

p~

---::

FALL

,.

V
~

100k

~

1M

10M 100M

FREQUENCY (Hz)

15
100

lk

10k

100k

o

±S

'3

LOAD RESISTANCE en)

APPLICATIONS INFORMATION

Thermoelectric voltages generated by dissimilar metals at
the input terminal contacts can degrade the drift performance. Best operation will be obtained when both input
contacts are maintained at the same temperature.

±l2

±15

±18

±2l

OFFSET NULLING CIRCUIT

OP-37 Series units may be inserted directly into 725, OP-06,
OP-07, and OP-05 sockets with or without removal of external
compensation or nulling components. Additionally, the OP37 may be fitted to unnulled 741-type sockets; however, if
conventional 741 nulling circuitry is in use, it should be modified or removed to ensure correct OP-37 operation. OP-37
offset voltage may be nulled to zero (or other desired setting)
using a potentiometer (see offset nulling circuit).
The OP-37 provides stable operation with load capacitances
of up to 1000pF and ± 10V swings; larger capacitances should
be decoupled with a 500 resistor inside the feedback loop.
Closed-loop gain must be at least five. For closed-loop gain
between five to ten, the designer should consider both the
OP-27 and the OP-37. For gains above ten, the OP-37 has a
clear advantage over the unity-gain-stable OP-27.

±9

SUPPLY VOLTAGE (VOLTS)

>...

~-'------{)v+

>-7...'6____.0 OUTPUT

OFFSET VOLTAGE ADJUSTMENT
The input offset voltage of the OP-37 is trimmed at wafer
level. However, if further adjustment of Vos is necessary, a
10kO trim potentiometer may be used. TCVosis not degraded
(see offset nulling circuit). Other potentiometer values from
1kO to 1MO can be used with a slight degradation (0.1 to
0.2p.V/o C) of TCVos. Trimming to a value other than zero
creates a drift of approximately (Vosl300) p.V/ o C. For exam-

5·159

1/86, Rev. A

II

_ _ _ _ _ _ _;~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

pie, the change In TeVos will be 0.33~Vloe If Vos is adjusted
to 100~V. The offset-voltage adjustment range with a 10kO
potentiometer is ±4mV. If smaller adjustment range Is required, the nulling sensitivity can be reduced by using a
smaller pot in conjunction with fixed resistors. For example,
the network below will have a ±280~V adjustment range.

4.7kG

1kGPOT

4.7kG

INSTRUMENTATION AMPLIFIER
A three-op-amp instrumentation amplifier provides high gain
and wide bandwidth. The Input noise of the circuit below Is
4.9nVlJ"HZ. The gain of the input stage is set at 25 and the
gain of the second stage is 40; overall gain Is 1000. The
amplifier bandwidth of 800kHz Is extraordinarily good for a
preciSion Instrumentation amplifier. Set to a gain of 1000, this
yields a gain-bandwidth product of 800M Hz. The full-power
bandwidth for a 20Vp _p output is 250kHz. Potentiometer R7
provides quadrature trimming to optimize the instrumentation amplifier's Ae common-mode rejection.

BURN-IN CIRCUIT

"6 0 1%
soon

"8

20kn 0.1%

R6

R9
198kSl

soon 0 1%

NOISE MEASUREMENTS
To measure the 80nV peak-to-peak noise specification of the
OP-37 in the 0.1 Hz to 10Hz range, the following precautions
must be observed:

• TRIM R2 FOR AVCL = 1000
• TRIM R10 FOR de CMRR
• TRIM R7 FOR MINIMUM Your
AT V CM = 20V p_p, 10KHz

(1) The device has to be warmed-up for at leastfive minutes.
As shown In the warm-up drift curve, the offset voltage
typically changes 4~V due to increasing chip temperature after power-up. In the 10 second measurement interval, these temperature-induced effects can exceed tensof- nanovolts.

,40

l~ -l

,
"
"

,20

(2) For similar reasons, the device has to be well-shielded
from air currents. Shielding minimizes thermocouple
effects.

r-.

iii 100

RS = lkG,
BALANCED

:s

""iIj

"s = ,00 3kO. The only exception is
when gain error is Important. Figure 3 illustrates the 10Hz
noise. As expected, the results are between the previous two
figures.

2 RSMATCHED
egRs=lOk,RS1=RS2-5k

];

~~:~I~~~
10
50 100

Figure 2

500 1k

100

,=iY>,

Figure 2 shows the 0,1 Hz-to-10Hz peak-to-peak noise. Here
the picture Is less favorable; resistor noise is negligible, current noise becomes important because it is inversely proportional to the square-root of frequency. The crossover with the
OP-07 occurs in the 3-to 5kO range depending on whether
balanced or unbalanced source resistors are used (at3kO the
IB' los error also can be three times the Vos spec.).

553.
QP-07

~ol,sr I~~~IT

beyond Rs of 20kO that current noise starts to dominate. The
argument can be made that current noise is not important for
applications with low-to-moderate source resistances. The
crossover between the OP-37 and OP-07 and OP-08 noise
occurs In the 15-to-40kO region.

PEAK-TO-PEAK NOISE (0.1 to
10Hz) VB SOURCE RESISTANCE
(INCLUDES RESISTOR NOISE)

500

8gRS"lOk,Rs,"Rs2"6k

~RESISTOR

RS2

500 1k

2RSMATCHED

fT'J.

itt>

~~SISTOR
NOISE ONLY
1

1 As UNMATCHED

egRs-Rs''''Ok,RS2aQ

2 RSMATCHED
.gRS"'Ok,RS,-RS2006k

=iY>,
R52

500 1k
5k 10k
RS - SOURCE RESISTANCE (m

50k

For reference, typical source resistances of some signal
sources are listed in Table 1.

prefer to use direct coupling. The high IB' Tevos of previous
designs have made direct coupling difficult, If not Impossible,
to use.

Table 1
DEVICE

Voltage noise Is Inversely proportional to the square-root of
bias current, but current noise Is proportional to the squareroot of bias current. The OP-37's noise advantage disappears
when high source-resistors are used. Figures 1, 2, and 3
compare OP-37 observed total noise with the noise performance of other devices In different circuit applications.
Total noise = f(Voltage nolse)2
(reSistor nolse 2 ]1/2

SOURCE
IMPEDANCE
<5000

Typically used In low-frequency
applicatIons,

MagnetIc
tapehead

<15000

Low Ie very Important to reduce
self-magnetization problems when
direct coupling IS used, OP-37 Ie
can be neglected

Magnetic
phonograph
cartridges

<15000

SimIlar need for low Ie In dIrect
coupled applications, OP-37 WIll not
Introduce any self-magnetIZatIon
problem,

Linear vanable
dIfferentIal
transformer

<15000

Used in rugged servo-feedback
applications Bandwidth of onterest
is 400Hz to 5kHz,

+ (current noise x RS)2 +

Figure 1 shows nolse-versus-source-reslstance at 1000Hz.
The same plot applies to wide band noise. To use this plot,
just multiply the vertical scale by the square-root of the
bandwidth.
At Rs<1 kO, the OP-37's low voltage noise is maintained, With
Rs< 1kO, total noise Increases, but Is dominated by the resistor noise rather than current or voltage noise. It Is only

5-161

COMMENTS

Straon gauge

1/86, Rev. A

II

--------I~ OP-37 LOW-NOISE PRECISION HIGH-SPEED OPERATIONAL AMPLIFIER

AUDIO APPLICATIONS

The following applications information has been abstracted
from a PMI article in the 12120/80 issue of Electronic Design
magazine and updated.

0412)
220~F

~~
,___ 6 ,~okn-:;;.
LF ROLLOFF

":'

OUT
C3

This circuit is capable of very low distortion over Its entire
range, generally below 0.01% at levels up to 7V rms. At 3V
output levels, it will produce less than 0.03% total harmonic
distortion at frequencies up to 20kHz.
Capacitor C3 and resistor R4 form a simple -6dB-per-octave
rumble filter, with a corner at 22Hz. As an option, the switchselected shunt capacitor C4, a non polarized electrolytic,
bypasses the low-frequency roll off. Placing the rumble filter's high-pass action after the preamp has the desirable
result of discriminating against the RIAA-amplifled lowfrequency noise components and pickup-produced lowfrequency disturbances.
A preamplifier for NAB tape playback is similar to an RIAA
phono preamp, though more gain Is typically demanded,
along with equalization requiring a heavy low-frequency
boost. The circuit in Fig. 4 can be readily modified for tape
use, as shown by Fig. 5.

O.47~F

C2
O.Ot~F

R3

,001000

pF

150

0

250

>1000

250

>1000

ps

-----"

Open-Loop Output
Resistance

150

Ro

150

NOTES:
1 Warmed up VCM = 0
2 Guaranteed by CMR test
3 Guaranteed but not tested

'OP-41G SPECIFICATIONS ARE SUBJECT TO CHANGE AT TIME OF INTRODUCTION

ELECTRICAL CHARACTERISTICS at Vs = ±15V, TA = -55° C/+125° C, unless otherwise noted.
OP-41A
PARAMETER

SYMBOL

Offset Voltage

CONDITIONS

MIN

OP-41B

TYP

MAX

Vas

400

1000

TCVos

25

MIN

TYP

MAX

UNITS

600

2000

/J.V

35

10

/J.vrc

Temperature
Coefficient of
Input Offset

Voltage
Offset Current

los

(Note 1)

40

1000

50

2000

pA

Bias Current

18

(Note 1)

4000

7500

4500

15000

pA

Ava

RL = 2kO
Va = ±lOV

Output Voltage
SWing

Va

RL = 2kO

Supply Current

ISY

Open-Loop

1000

5000

500

3000

VlmV

±120

±125

±115

±12.5

V

Va = OV

75

IVR

(Note 2)

±11

+15
-11 5

±11

+15
-115

V

CMR

VCM = ±11V

95

105

85

100

dB

PSRR

Vs = ±10V to ±20V

Short CirCUit
Output Current

Isc

Short Circuit to Ground

Slew Rate

SR

Voltage Gain

Input Voltage
Range
Common-Mode
Rejection
Power Supply

Rejection Ratio

Gain Bandwidth
Power Bandwidth
Capacitive Load

Stability

12

75

40

±6

GBW
BWp
Av = +1
(Note 3)

100

+12
-17

±36

±6

12

mA

10

100

pV/v

+12
-17

±36

mA

13

13

Vips

500

500

kHz

20

20

kHz

>1000

pF

>1000

100

NOTES:
Warmed up. VCM = 0
Guaranteed by CMR test

Guaranteed but not tested

5-166

1/86, Rev. A

---

-------

~ OP-41

LOW-BIAS-CURRENT JFET OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs

= ±15V,

TA

= -25°C/+85°C

for ElF grades and O°C/70°C for G grade, unless

otherwise noted.
OP-41E
PARAMETER

SYMBOL

Offset Voltage

CONDITIONS

OP-41G*

OP-41F

TYP

MAX

Vos

250

1000

TCVos

3.5

MIN

MIN

TYP

MAX

500

2000

MIN

TYP

MAX

UNITS

500

2000

IJ V

Temperature
Coefficient of
Input Offset
Voltage
Offset Current

los

(Note 1)

Bias Current

Is

(Note 1)

Open-Loop
Voltage Gain

Avo

RL
Vo

=2kO
=±10V

Output Voltage
Swing

Vo

RL

=2kO

Supply Current

ISY

Vo

Input Voltage
Range
Common~Mode

Rejection
Power Supply
Rejection Ratio

240

IJVloC

75

100

10

200

20

500

300

1000

100

pA
500

pA

1000

5000

500

4000

300

3000

VlmV

±12.0

±126

±115

±125

±11

±126

V

=OV

75

IVR

(Note 2)

±110

+15
-115

±11

+15
-115

±11

+15
-115

V

CMR

VCM

95

110

85

100

85

100

dB

PSRR

Vs

Short CIrcuIt

=±1W

=±10V to ±20V

Short CIrcuit

Output Current

7.5

Isc

±6

to Ground

75

12

5-

40

+16
-18

±36

±6

1.2

10

100

+16
-18

±36

75

±6

1.2

mA

10

100

IJVIV

+20
-18

±36

mA

Slew Rate

SR

13

13

13

Gain Bandwidth

GBW

500

500

500

kHz

Power Bandwidth

BWp

20

20

20

kHz

>1000

pF

Av = +1
(Note 3)

Capacitive Load
Stability

100

>1000

100

>1000

100

VII'S

NOTES:
VCM = 0
2. Guaranteed by CMR test.
3. Guaranteed but not tested.
'OP-41G SPECIFICATIONS ARE SUBJECT TO CHANGE AT TIME OF INTRODUCTION

1. Warmed up.

BURN-IN CIRCUIT

+18V

10k!}

+3Vo----''Mr--,
10kfl.

-lav

5-167

1/86, Rev. A

II
ffi.....
i:l-<

::sp..,
~

~

0.....

~

~

p..,

0

_ _ _ _ _ _ _ _ _--;~ OP-41 LOW-BIAS-CURRENT JFET OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS

1.
2.
3.
4.

OFFSET VOLTAGE NULL
INVERTING INPUT
NONINVERTING INPUT
NEGATIVE SUPPLY
5. OFFSET VOLTAGE NULL
6. AMPLIFIER OUTPUT
7. POSITIVE SUPPLY
For additional DICE information refer to
1986 Data Book, Section 2.

DIE SIZE 0.103 X 0.074 inch, 7622 sq. mils
(2.62 X 1.88mm, 4.92 sq. mm)

WAFER TEST LIMITS at Vs = ±15V, TA = 25° C, unless otherwise noted.
PARAMETER

SYMBOL

Offset Voltage

Vas

Bias Current

CONDITIONS

OP-41N
LIMIT

UNITS

1000

/-IV MAX

18

(Note 1)

20

pA MAX

Open-Loop Voltage Gain

Ava

RL = 2kO

500

VlmV MIN

Output Voltage SWing

Va

RL = 2kO

±12

VMIN

Supply Current

ISY

Va = OV

12

mAMAX

Input Voltage Range

IVR

(Note 2)

±11

VMIN

Common-Mode
RejectIon

CMR

VCM = ±11V

90

dB MIN

Power Supply
Rejection RatiO

PSRR

Vs = ±10V to ±20V

80

/-IVIV MAX

Isc

Short Circuit to Ground

±6
±36

mAMIN
mAMAX

Short CirCUit

Output Current

Slew Rate
Capacitive Load

Stability

SR
Av = +1

VI/-Is MIN
(Note 3)

250

pF MIN

NOTES:
1 VCM =0
2 Guaranteed by CMR test
3 Guaranteed but not tested
Electrical tests are performed at wafer probe to the limits shown Due to vanatlons In assembly methods and normal Yield loss, Yield after packaging IS not guaranteed
for standard product dice Consult factory to negotiate specifications based on dice lot qualificatIon through sample lot assembly and testing

5-168

1/86, Rev_ A

-----------I1fMD

OP-41 LOW-BIAS-CURRENT JFET OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
BIAS CURRENT vs
TEMPERATURE

OFFSET CURRENT vs
TEMPERATURE

BIAS CURRENT vs
COMMON-MODE VOLTAGE

'a'

'00

/

'a'

!...
~a:

::>

'02

!...

/'/

'0'

u

~
a;

/

/
V

/

'a'
10-1

V

10-2

-50

-75

~

-25

25

50

75

100

-'5

-10

w

"
:I:

u

1\

-600
-500

w

:;"
~

-400

0
> -300

~

>
3

\

0

-100

8

~

\

6

"~
o
>

~

,

o

±10

'5

±15

'20

'00
840

820

80

::> 800

60

u

-- -

~

720
700
'4

IV

-

1...
i:la:

'8

'25

~~

±16

820

780

r- ........

760
740

I

720

'0

'2

700
-75

-50

'""'-.........
\.

"

80

"-

40

±20

0'

PHASE

'00

100

lk

FREQUENCY (Hz)

5-169

10k

,'0
,
,
,
20

"

lOOk

50

75

1M

-r--...

'20

'00
80

~

10

'40

90

' - MARGIN
'" 77"

25

100

125

POWER SUPPLY REJECTION
vs FREQUENCY
70

~

-25

TEMPERATURE (c GI

60

20

SUPPlVVOLTAGE (VOLTS)

840

u

~ij;

-20
±12

'00

a:
::> 800

'40

860

740

75

--

OPEN-LOOP GAIN AND
PHASE vs FREQUENCY

900

~

50

SUPPLY CURRENT
vs TEMPERATURE

TIME AFTER SUPPLY TURN-ON (MIN)

880

~

25

860

o

±20

SUPPLY CURRENT

ij; 760

/

a

880

o

vs SUPPLY VOLTAGE

780

i

2

SUPPLY VOLTAGE (VOLTS)

~

v

a:

a

...

'5

900

Ii:

1\
\

-200

~

0'

'0

-5

'0

Z

~

~

WARM-UP DRIFT vs TIME

-800
-700

/

COMMON-MODE VOLTAGE (VOLTS)

OFFSET VOLTAGE vs
SUPPLY VOLTAGE

>
3

::>

u

o

a

125

TEMPERATURE (c c)

10

~

~

io"""

.V

V

/

60

'" " , " '\
-PSR

40

30

20

40

o

,

+PSR

~

i\.

"'- ~
10

100

lk

10k

lOOk

1M

FREQUENCY (Hz)

1/86, Rev. A

_ _ _ _ _ _ _ _ _----;~ OP-41 LOW-BIAS-CURRENT JFET OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
MAXIMUM OUTPUT SWING
VI FREQUENCY

COMMON-MODE REJECTION
VI FREQUENCY

14

,

120 ~!±!!I:IH-~-++I+ItHf--t-tlfJttttt-ttttttlt
100

I.

30

140 r-r"l"T11mr-T"T1"1'TT111"""T"nmrm-TTTmr-r1rmm

-pJs'u~
OUTPUT

13

;;;

\

H+H+llII-+H+I1I!I..-l

MAXIMUM OUTPUT VOLTAGE
VI LOAD RESISTANCE

~

~ 12

o

C

11

"'~

10

~

9

...

8

g

6

o
>

1\!,

it

~
~~¥~J~VE_

7

IfJ
o
10k

lk

100k

FREQUENCY (Hz)

FREQUENCY (Hz)

TOTAL HARMONIC
DISTORTION VI FREQUENCY

SLEW RATE VI
TEMPERATURE

Vo = 10Vp-p

¥/

~
I.
~

01

001

~ -I-

i-"

::;:... ~VCL

~-

~

1-0...
i"""-o .......:

!<"'a: '4
~ 13

=10

~,

-

POSITIVE

-

NEGATIVE

12

0.001

00001
100

11

III
1k

/
-2
-4

-6
-8

10
-75

10k

AV ·-1
RL'" 2.Skn

T001%/A
A ~OO.J,%

16

AveL ·'00

lOOk

SETTLING TIME
10

17

10

10k

lk

RL = 2kn

RL = 10kU

10

1.
LOAD RESISTANCE 1m

18

100

3
100

-10
-50

-25

25

50

75

100

125

o

I~ ,TO 0.01%

TOO1~

""

t'....

10

12

SETTLING TIME (,..IS)

TEMPERATURE (coC)

FREQUENCY (Hz)

"" '"

SMALL-SIGNAL OVERSHOOT
VI CAPACITIVE LOAD
40
3.

~

30

!!

2.

8ili

20

~

16

Z

a:

,

~

"

II

10

o

/X

~

il

o

500

1000

1600

2000

2500

3000

3500

CAPACITIVE LOAD (pF)

5-170

1/86, Rev. A

-----------I~ OP-41 LOW-BIAS-CURRENT JFET OPERATIONAL AMPLIFIER

OFFSET VOLTAGE ADJUSTMENT

FIGURE 3: SMALL-SIGNAL TRANSIENT RESPONSE

Offset voltage is adjusted by a potentiometer of 10kO to 100kO
resistance. This potentiometer should be connected between
pins 1 and 5 with the wiper connected to the V- supply. (See
Figure 1.) Nulling Vos will change TCVos by no more than
5fJV/o C per millivolt of Vos change
FIGURE 1: INPUT OFFSET VOLTAGE NULLING
v'

FIGURE 4: SMALL-SIGNAL TRANSIENT RESPONSE WITH
1000pF LOAD
vNOTE
Vas CAN BE TRIMMED WITH POTENTIOMETER
RANGING FROM 10kn TO 100kn

APPLICATIONS INFORMATION
TYPICAL AC PERFORMANCE CHARACTERISTICS
Figure 2 shows the overload recovery time after the output
saturates at each supply. A high degree of slew-rate symmetry
is maintained even during severe input overload The photo
also shows the well controlled linear characteristics of the
amplifier and freedom from oscillations The OP-41 's symmetry
greatly reduces the generation of large DC components in the
output when the amplifier is overdriven. This significantly
reduces system recovery time after an overload.

FIGURE 5: WIDE-DYNAMiC-RANGE LIGHT DETECTOR

""4pF

109 n

Figure 3 shows the unity-gain small-signal transient response
of the OP-41. Note the clean symmetrical waveform.
Figure 4 illustrates the high degree of stability even when
loaded with 1000pF at unity-gain. Heavy capacitive loading will
cause stability problems with many amplifiers.

>---<>--. Va'" 0 5V!nW

Figure 5 illustrates the use of the OP-41 in a high sensitivity,
wide-dynamlc-range light detector. ThiS circuit Will produce an
output voltage proportional to the light input over a60dB range.
FIGURE 2: OVERLOAD RECOVERY TIME AT Av = 10

CMR MEASUREMENT METHODS
Two separate methods are used to measure the CMR. The first
method is used over the range of 10Hz to 20kHz. This method
grounds the input circuitry and applies the common-mode
signal to the remainder of the op amp, Figure 6.
The AMP-01 eliminates loading on the output stage. This
assures that the OP-41 output is not required to deliver current
into the feedback circuit. The effects of the DUT open-loop gain
changing with frequency are therefore significantly reduced.
The Circuit does not require tight resistor matching. DC data
sheet limits may be verified using this method. Circuit accuracy
is dependent on the high CMR of the AMP-01.

5-171

1/86, Rev. A

------------l!EMD

OP-41 LOW-BIAS-CURRENT JFET OPERATIONAL AMPLIFIER

FIGURE 6: CIRCUIT USED TO MEASURE CMR FROM 10Hz
TO 20kHz

transmitters. Loss of accuracy can also occur from surface and
bulk leakages in printed circuit boards. Both of these conditions
can be avoided by the following methods.
Hum and RF pickup are eliminated or reduced by keeping all
high Impedance leads, including feedback resistor leads, inside
shielded enclosures. In addition to shielding, power supply
lines should be bypassed where they pass through the shielding. This will prevent noise from being retransmitted from the
power supply lines inside the shielded enclosure.

9kn

1kH

1kH

AMP-01
AV= 1

Noise can also be created by the flexing of coax cable. These
signals can be caused by mechanical vibrations inside or
outside the shielding. Prevention consists of securely supporting all high-impedance shielded lines to prevent motion.
Printed circuit board leakage currents can easily exceed the
OP-41 bias currents or the incoming signal. Leakage currents
can be minimized by using Teflon insulators to support wires
instead of using PC traces. An alternate method is guarding the
high impedance traces. When the OP-41 is in the inverting
mode, the signal traces should have grounded guard traces on
both sides, Figure 8. The opposite side of the board should be
used as a ground plane and shield, if not otherwise used. A
ground plane is implemented by leaving copper on all areas
that are not being used for signal or power conduction. Ground
connection should be made to all areas of isolated copper. In
the noninverting configuration, the OP-41 's output signal or a
portion of it should be used to drive the guard traces, Figure 9
When the guard drive voltage is equal to the input signal,
leakage currents will be effectively eliminated.

An alternate circuit may be used to make high-frequency
measurements from 2kHz to 500kHz, Figure 7. The 2kHz to
20kHz data overlap can be used to verify the accuracy of the
respective test methods.
This method drives the input stage with the test signal and
requires an accurate ratio of resistors, R4/R3 = R1/R2. To
measure CMR to 100dB requires ratio matching to better than
10ppm. For this reason, it is not practical to use the second
method at low frequencies where CMR is greater than 80-100dB.
The OUT output is normally connected directly to R4 which
may cause problems. If the OUT is not buffered with a broadband low-output-impedance amplifier, the frequency-dependent output impedance of the OUT, in series with R4, rapidly
unbalances the resistor ratios. This causes frequency dependent errors. The OP-27 provides good performance over the
range of frequencies used.

FIGURE 8: CURRENT-TO-VOLTAGE CONVERTER

r--------..,
I
I
GUARD"./

I
I

FIGURE 7: CIRCUIT USED TO MEASURE CMR FROM 2kHz
TO 500kHz

r-----J
I
I

2
>,~~-oOUT

rOGAIN/PHASE METER
R4

R3

>-+~·~~IN/PHASE

1kn

FIGURE 9: VERY HIGH IMPEDANCE NON INVERTING
AMPLIFIER

METER

R2

1kn
"v

R1

9kn

GUARDING AND SHIELDING

In applications where the input is at high impedance, careful
shielding is required to prevent hum pickup from power line
sources or detection of RF from radio stations and nearby radar

5-172

1/86, Rev. A

-----------I~ OP-41 LOW-BIAS-CURRENT JFET OPERATIONAL AMPLIFIER

The High Q Notch Filter benefits from the low bias current and
high input impedance of the OP-41 , Figure 10. These features
enable small value capacitors and large resistors to be used in
this 60Hz notch filter. The 5pA bias current only develops 100/lV
across R1 and R2.
FIGURE 10: HIGH

FIGURE 12: AMPLIFIER FOR PIEZOELECTRICTRANSDUCERS
~--o---O-------------jI~

250n

+IN~--W~-------+-_+----'

I

IL

I
I
I
I
I
I

____________ _

----------------~

V-

*NOTE SEPARATE SUPPLIES FOR OUTPUT STAGE

Manufactured under the follOWing patents 4.471,321 and 4.503.381

5-180

1/86, Rev. A

-------------I~ OP-50 HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

The OP-50 is stable for closed-loop gains above 50, and can
be externally compensated for closed-loop gains in the range
of 5 to 50. The amplifier is designed for use in high-gain
and/or high-output-current applications. For example, an
OP-07 coupled with an output buffer can be replaced by a
single OP-50 amplifier.

A thermal-shutdown circuit protects the OP-50 from overdissipation. When the die temperature reaches approximately
165°C, the output stage automatically shuts down. The
amplifier input stage remains fully operational, thereby protecting the signal source from any loading changes caused
by a complete shutdown.

lon-implanted superbeta transistors, combined with a patented input bias current cancellation circuit, provide an input
bias current of only 5nA and input offset current of 1nA. Over
the full military temperature range, input bias current and
input offset current for an A-grade device does not exceed
8nA and 3nA, respectively. Input offset voltages are trimmed
to a maximum of 25/N (A/E grades) and 1OOl'V (B/F grades)
using PMI's zener-zapping technique. This low offset eliminates the need for an offset trimpot in most applications.

COMPENSATION FOR GAINS BETWEEN 5 AND 50
The OP-50 can be compensated for inverting gains between
5 and 50 using a series resistor and capacitor. These values
can be adjusted to minimize overshoot for a given application. The recommended compensation is'

Lowvoltage-noise, typically4.5nV/VHZ at 1kHz, isachieved
in the OP-50 with minimum sacrifice of input protection.
Overload protection is provided by input resistors of 2500
and emitter-base diodes. The input resistors provide current
limit protection against differential inputs of up to ±10V; and
the diodes prevent avalanche breakdown which could
degrade the IB' los, and matching of the input stage transistors. External resistors can be added to the input to guard
against higher input voltages; however, the added resistors
will degrade noise voltage performance. When minimum
noise voltage is required, source resistance should be kept
below a few hundred ohms.

GAIN RANGE

RC

5 <; AveL <; 20
20 <; AveL <; 50
AveL 2: 50

56011

3.3kll

47nF
1nF

No compensation reqUired

COMPENSATION

Separate output-stage power supply pins are provided on the
OP-50 to allow control of device power dissipation and output voltage swing. The maximum voltage which may be
applied across the power supply pins is ±18V. The guaranteed specifications are based on operating both stages at
±15V; however, there is minimal effect on DC performance
when the main amplifier is operated at ±15V and the output
stage is operated at a reduced voltage. When operating both
the main amplifier and the output stage at the same voltages,
the corresponding power supply pins may be tied together.
Decoupling capacitors are recommended between the power
supply pins and analog ground. It is necessary to use decoupiing capacitors on each power supply pin when operating
the output stage at supply voltages less than the amplifier
supply voltage. Do not operate the output-stage negative
power supply pin at a more negative voltage than the negative supply pin (V-).

ABSOLUTE MAXIMUM RATINGS (Note 1)
Supply Voltage (Note 2) ........................... ±18V
Internal Power Dissipation (Note 3) .............. 500mW
Input Voltage ............................ Supply Voltage
Differential Input Voltage (Note 4) ................. ±10V
Differential Input Current (Note 4) ............... ±20mA
Output Short-Circuit Duration ................. Indefinite
Storage Temperature Range ............ -65°C to +150°C
Operating Temperature Range
OP-50A, B ........................... -55°C to +125°C
OP-50E, F ..................... , ., ... ,. -25°C to +85°C
Lead Temperature (Soldering, 60 sec) ............. 300°C
DICE Junction Temperature (TJ) •••••••• -65°C to +150°C

A thermally-symmetric die layout, which differs from other
op amp designs by the positioning of more devices along the
center line, provides the OP-50 with a thermal drift of less
than 0.3I'V;CC. This layout feature is critical to the maintenance of high open-loop gain when driving large-current
loads and dissipating hundreds of milliwatts in the device.
The use of a heatsink is recommended to reduce internal
temperature rise when operating at high output power levels.
The use of standard dual-in-line package heatsinks will help
to dissipate heat to the environment. Other techniques, such
as the use of external voltage-dropping resistors, allow heat
to be dissipated outside of the package. See Figure 5, "Driving
500 Loads", in the applications section.

PACKAGE TYPE

14-Pin

Hermetic DIP (Y)

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

106°C

11.3mW/oC

NOTES:
,
Absolute ratings apply to both DICE and packaged parts, unless
2

otherWise noted
Supply voltage rating applies to all power supply pinS No device pins
should be connected to a voltage more negative than the supply

to V-, pin 5
3
4

5-181

See table for maximum ambient temperature ratmg and derating factor
The OP-50's Inputs are protected by 250!! senes resistors and protection
diodes If the differential Input voltage exceeds ±10V, the Input current
must be limited to ±20mA

1/86, Rev. A

II

~

OP-SO HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at V+ = +VOP = +15V, V- = -Vop =-15V, TA = 25°C, no compensation, unless otherwise
noted.
OP-50A/E
MAX

TYP

MAX

Vas

10

25

50

100

~V

18

±1

±5

±1

±10

nA

SYMBOL

Input Offset Voltage
Input Bias Current

MIN

OP-508/F

TYP

PARAMETER

CONDITIONS

MIN

01

0.1

UNITS

Input Offset Current

los

Input Voltage Aange

IVA

CMAA:> 100dB

Va

AL:> 50 on
A L :>501l (Note 1)

±13
±25

±134
±40

±13
±25

±13,4
±40

Output Voltage SWing

Va

V+ = +Vop = +5V,
V- = -Vop = -5V
AL = 5001l
AL = 501l

±3.5
±25

±38
±28

±3.5
±25

±38
±28

Slew Rate

SA

AL:> 2kll
Ac= 56011
C c = 47nF

25

30

25

30

V/~s

CMAA

VCM = ±10V

126

140

110

120

dB

PSAA

Vs = ±5V to ±15V

05

~V!V

Ava

Vo =±lOV, RL = 1kll

10

20

75

15

V/~V

AVCL = 50 (Note 2)

15

25

15

25

MHz

±10

±25

±10

±25

mV

Output Voltage SWing

Common-Mode
Rejection Ratio

Power Supply
Rejection Ratio
Large-Signal

Voltage Gain
Galn~Bandwidth

Product

GBW

Offset Voltage
Aange Adjust
Input NOise Voltage

±12

0.1

Ap=100kll
€np-p

f = 0,1 Hz to 10Hz

012

NOise Voltage Density

eo

f= 10Hz
f = 1kHz (Note 3)

55
4,5

NOise Current

Inp-p

05

No Load

26

3.3

POSitive Current Limit

+lse

Output shorted to Ground

60

95

120

Negative Current limit

-Ise

Output shorted to Ground

60

85

120

55
4.5

Common-Mode
Input Resistance
Capacitive Load

Capability

Settling-Time

jN p _p
85
60

0.3
023

nVljHZ

pAljHZ

26

3,3

mA

60

95

120

mA

60

85

120

mA

Differential-Mode
Input ReSistance

V

pA p. p

ISY

Current

10

V

f = 0 1Hz to 10Hz

i = 100Hz
f = 1kHz

NOise Current Density

V

012
85
60

0.3
0,23

QUiescent Supply

nA

±12

Mfl

RIND

20

R 1NCM

CL

AVCL:> 5
Ac= 56011 (Note 2)
Cc=47nF

t,

Settling to 0,01%, Va = 20Vp_p
A VCL = 500
AVCL = 1000

10

20

10

30
60

Gil

nF

30
60

~s

NOTES:
Guaranteed by current limit tests
2 Guaranteed by deSign
3, Sample tested,

5-182

1/86, Rev. A

-----------l~ OP-50 HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at V+ =

+VOP = +15V, V- = -Vop = -15V, -25°C:'O: TA:'O: +85°C, no compensation, unless

otherwise noted.

OP-50E
CONDITIONS

MIN

OP-50F

TYP

MAX

MIN

TYP

MAX

UNITS

20

45

50

150

~V

015

0.3

03

PARAMETER

SYMBOL

Input Offset Voltage

Vas

Input Offset
Voltage Drift

TCVos

Input Bias Current

la

±2

±7

±2

±25

nA

Input Offset Current

los

0.2

2.5

0.2

20

nA

Input Offset
Current Drift

Input Bias
Current Dnft

(Note 1)

~V/'C

pArC

TClos
TCl a

20

50

Input Voltage Range

IVR

CMRR;e 100dB

Output Voltage SWing

Va

RL;e 500!!

±12

±134

±12

±134

V

CMRR

VCM~

120

130

105

120

dB

PSRR

Vs ~ ±5V to ±15V

05

ISY

No Load

28

28

rnA

15

15

V/MV

Common-Mode
Rejection Ratio

Power Supply
Rejection Ratio

QUiescent Supply
Current
Open-Loop Gam

±115

pArC

±10V

±11.5

V

~

VOUT~

Ava

RL ~

±10V,
(Note 2)
1k!!

1 25

05

1 25

MVIV

ELECTRICAL CHARACTERISTICS atV+=+Vop=+15V,

V-=-Vop =-15V, -55°C:'O:TA :'O:+125°C, no compensation, unless

otherwise noted.

OP-50A
MAX

TYP

MAX

20

55

50

200

015

03

03

Is

±2

±8

±2

±20

nA

los

05

3

05

12

nA

Input Offset Voltage

Vos

Input Offset
Voltage Drift

TCVos

Input Bias Current

Input Offset Current
Current Dnft
Input Bias

Current Dnft

OP-50B

TYP

SYMBOL

MIN

CONDITIONS

MIN

UNITS

MV
MV/'C

pA/'C

Telos
20

TCl a

50

±115

pArC

±115

Input Voltage Range

IVR

CMRR;e fOOdB

Output Voltage Swing

Va

RL;e soon

±12

±132

±12

±132

V

CMRR

VCM~

120

130

105

120

dB

Power Supply
Reiection Ratio

PSRR

Vs ~ ±5V to ±15V

0.5

Quiescent Supply
Current

ISY

No Load

2.8

2.8

rnA

Ava

Vo= ±10V,
(Note 1)
RL = 1kll

10

10

V/MV

Common-Mode

Rejection Ratio

Open-Loop Gain

±10V

1.25

V

0.5

1.25

MVIV

NOTE:

1. Tested at +125'C, guaranteed by design at -55'C.

1/86, Rev. A

5-183

- - - -. .

~~~~-

~

~
~

2. Guaranteed by design

PARAMETER

~

9

NOTES:
TeVos tested on E grade, guaranteed by design on F grade specification

Input Offset

~

......

-~----

-~-

O~

------------l~ OP-50 HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS

1.
2.
5.
6.
7.
9.
10.
11.
12.
13.
14.
15.

NONINVERTING INPUT
INVERTING INPUT
VOUTPUT
-VOP
V+
+Vop
COMPENSATION
COMPENSATION
NULL
NULL
V- (OPTIONAL BONDING PAD)'

For addilional DICE informalion refer 10
1986 Dala Book, Section 2.
DIE SIZE 0.110 X 0.148 inch, 16,280 sq. mils
(2.79 X 3.76 mm, 10.50 sq. mm)

WAFER TEST LIMITS at V+ = +VOP = +15V, V- = -Vop = -15V, TA = 25°C, no compensation, unless otherwise noted.
OP-SOG
CONDITIONS

PARAMETER

SYMBOL

Input Offset Voltage

Vas

LIMIT

100

~V

Input Bias Current

18

±10

nAMAX

UNITS
MAX

Input Offset Current

los

Output Voltage SWing

Va

RL " soon

Output Voltage SWing

Va

V+ ~ +VoP ~ +5V. VRL ~ soon
RL ~ son

Common-Mode
Rejection Ratio

CMRR

VCM == ±10V

Power Supply
Rejection Ratio

PSRR

Vs ~ ±5V to ±15V

Large-Signal
Voltage Gain

Ava

Va

±10V. RL ~ 1kn

7.5

Positive Current Limit

+Isc

Output shorted to Ground

60

mAMIN

Negative CUrrent LimIt

-I se

Output shorted to Ground

60

mAMIN

QUiescent Supply
Current

ISY

No Load

3.3

mAMAX

nAMAX

~

~

-Vop

~

±13

VMIN

±35
±25

V MIN

110

dBMIN

-5V

~V/V

V/~V

MAX

MIN

NOTE:
Electrical tests are performed at wafer probe to the limIts shown Due to vanatlons In assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

5-184

1/86, Rev. A

----------11EH!>

OP·50 HIGH·OUTPUT·CURRENT OPERATIONAL AMPLIFIER

TYPICAL ELECTRICAL CHARACTERISTICS at V+ = +Vop= +15V, V-= -Vop= -15V, TA = 25°C, no compensation, unless
otherwise noted.
OP·50G
PARAMETER

SYMBOL

CONDITIONS

Slew Rate

SR

RL ", 2kn
Rc = S60n
C c =47nF

NOise Voltage Density

e,

f= 10Hz
f= 1kHz

Input Noise Voltage

enp-p

i,

NOIse Current Density

TYPICAL

UNITS
V/~s

5.5
4.5

nV/VHZ

f = 0.1Hz to 10Hz

012

jJ.V p_p

f= 10Hz
f = 1kHz

02
0.15

pA/VHZ

--------

AVCL~ 5

Capacitive Load
Capability

CL

nF

10

Rc = S60n
C c =47nF

NOISE TEST CIRCUIT (0.1 TO 10Hz)

O.lpF

100kO
100kn

+15V

2kO

BURN-IN CIRCUIT

OFFSET NULLING CIRCUIT
+18V

V+

10kn

OUTPUT

R,
100
krl
ALL RESISTORS ARE 1% METAL FILM.

V-

5·185

1/86, Rev. A

-------------t1fW

OP-50 HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
INPUT OFFSET VOLTAGE
vs TEMPERATURE

INPUT BIAS CURRENT
vs SUPPLY VOLTAGE
30

40

30r-----,-----~------~-----,

TA =2SOC

Vs = ±15V

30
20

20

........

-30

~ t-.....

1.0

........

05

.........

- --

f'

-40

-75

-50

-25

25

50

75

100

1.0

I-----+------t----=='T""-"""d

05~----~----_4------+-----~

±s

125

TEMPERATURE 1°C)

±lO

±lS

±20

SLEW RATE vs
TEMPERATURE

__

VOLTAGE (VOLTS)

SLEW RATE vs
TEMPERATURE
50

Vs = ±15V

AVCL '" 20

45

45f---t-+--t---I--+- ~~: ~l;:n

05

40

CC '" 1nF

04

~

/

03

"-

01

2.

......

40 f-__t-+-+---I--++SR

~C =~~~PF

35
."

~ 35~~::~~~~~~~~~~-J

~

J

"

02

-"""1'----

"rR

30f---t-+-+---I--+-~-f-~

V

-SR

'"

30

~

25

~

20

w
~

AV~l

= 10
Vs = ±15V
RC = 560n
Cc=47nF
Rl = lkH
Cl = 1000pF

15
10

25f---t-+-+---I--+-~-f-~

20.L-~

-75

-50

-25

25

50

75

100

125

-75

__~__~~__-L__~__~~

-50

-25

TEMPERATURE (OC)

25

50

150

-75

-50

-25

25

- '"

50

75

100

125

'" 1.5 f-H+++Ittt--++Hfttttt-t+tvlltt---+++t+tffi

110

0:

100

~

t-t+t+Hltt--+++t+tffi--l+f+ttIll-+f-ttHttl

~

120

w

O~~~~~~~~~~~WU

1000

140

~
z

0

130

r--

~

~

\

90

0:

80

1il

90

~

~

70
1

10

lOa

lk

FREQUENCY (Hz)

10k

.....

I\.
'\

lOOk

""-

r'\.

70
60

r"-

60

100

&:

TA = 25°C
Vs'" ±15V
A VCL '" 1000

...........

'" I"

120

;J
0:

UNCOMPENSATED

~

§ 110

80

t-t+t+Hitt-+++t-Hffi--l:+f+ttItt--+f-ttHttl

150

TA = 25°C
VCM = ±lOV '
AVCL = 1000

130

."

10
lOa
CAPACITIVE LOAD (nF)

PSRR vs
FREQUENCY

VSI~I±'5V

140

AVCL'" 10
RL = lkn
RC = 560n
Cc = 47nF

.2Of-+-++++Itff-++HHitlf-t+t+Hltt--+++t+tffi

1

o

125

CMRR vs
FREQUENCY

Vs'" ±1SV

0.1

100

_....

TEMPERATURE (OC)

SLEW RATE vs
CAPACITIVE LOAD

!\

75

-

r¥

05

o

0.5

____

10

COMMON~MODE

06

~

____

-5

SUPPLY VOLTAGE (VOLTS)

INPUT OFFSET CURRENT
vs TEMPERATURE

-. 1.0

____

-10

±25

~

-20

15~----+----__t------+-----~

~

~

~

1.5

"-

~

i""""oo ~

~

i<

........

-10

~

r"""Ioo.

~

10

>

tu
~
e

Vs = ±15V
25~--==~_.~_4------+-----~

25

o

~w

""~
e

INPUT BIAS CURRENT
vs COMMON-MODE VOLTAGE

"
"-

-PSRR

+PSRR

50

1

10

100

1k

10k

lOOk

1M

FREQUENCY (Hz)

NOTE:
The symbol ±Vsis used to indicate the supply voltages when the main amplifier
and the output stage are being operated at the same voltages.

5-186

1/86, Rev. A

-------------l~ OP·50 HIGH·OUTPUT·CURRENT OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
PSRR VB
FREQUENCY

SUPPLY CURRENT
VB TEMPERATURE

SUPPLY CURRENT

vs SUPPLY VOLTAGE

150
COMPENSATED

".:s 130 ~ ~
140

'~"

g

'\

~ 110

i

100

a:

90

~

80

Vs = ±15V
AVCL = 1000

1\. ~

120

Vs'" ±15V

35~-+--~~~~--+--+--4-----l

TA = 25°C

"

'\.
\

i

Cc = lnF
RC=33kil

~

I

::i~

~

">
~

\-PS~R

~

+PSRR\ \

~

70

so

10

1

100

lk

20~_t--+-_t--_r--r-_t--+-_4

1 5 1-_t--+-_t--_r--r-_t--+-_4

100k

-75

1M

-50

25

50

75

100

±5

125

±10

:1:15

±20

±25

CURRENT NOISE DENSITY
VB FREQUENCY

VOLTAGE NOISE DENSITY
VB FREQUENCY

20

...
......

,

80

40

-25

SHORT-CIRCUIT CURRENT
VB TEMPERATURE

~ 100

~

2t---+--+---I----+---1

SUPPLY VOLTAGE (VOLTS)

120

60

~

TEMPERATURE (CC)

ffi

~

~

10t--+--1-~1--t--+_-+--1--1

1.8

~

12

1,4

~

10

I

~

~

.....

-

r--..

~

.......

25

-25

50

75

100

,

1,2

!ll
5
z

ffi

~

\

~ 10
08

04

"

0.2

o

126

TEMPERATURE (OCI

'"

!il
'"~

1\

8

10

6

i'
4

100

1k

10k

o

10

1

FREOUENCY (Hz)

TOTAL HARMONIC
DISTORTION
VB FREQUENCY

"'

AVCL = 1000
VS "'±15V

!:;
o
> 2

~

1

1~~=25'e'

C

!!!

0.6

§

o
-60

16

a:

20

-75

14

TA - 26~~"
Vs· ±15V-

Vs - j:15V

~

a

o~~--~~~~--~~--~~

180

at::

4 t---_t----+-----j----_r------l

ill

a:

FREQUENCY (Hz)

140

i

t<

05t--+--1-~1--t--+_-+--1--1

~

10k

li;;;;;;~;;;~;t;::;;+:~::4::::j
251--+--+_-I--+--+--+--+_---1
30

100

lk

10k

FREQUENCY (Hz)

TOTAL HARMONIC
DISTORTION
VB LOAD RESISTANCE

o TO 1Hz
NOISE VOLTAGE DENSITY

006

Vs· ±15V

TA = 25°C

RL = 5000

t---I-t-1H-+t-ttt---t Vs = ±15V

Vo '" 20Vp _p

AVCl"" 100
~--~HI-++t-ttt---+f '" 1kHz
Vo = 20V p _p

Ave L- 3OO Ave L = 100
Ave L -10 I",\~
Ave L=5,

~

gaol
o

100

,/

o

OH,

""

1k

lH,

800/J.V FULL SCALE
10 O/J.V/DIV VERTICAL
AVCL = 10,000

io"""'" ....-!I::::
FREQUENCY (Hz)

0061--t~~H+t-tt1't--tI---H+t-H+I

10k

~00L---~~~~~1~k--~~~LU~'0k
LOAD RESISTANCE (m

5·187

1/86, Rev. A

II

-----------l~ OP-50 HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
GAIN, PHASE SHIFT
vs FREQUENCY
'40

... IIIII
I~Hds~

'20

70

90

60

I'

;;

15

i'o.

80

50

170

20

o

T

AI~I~I~QC
'00

'0

'90

20

210

'0

'Ok

lOOk

'M

'k

'0
-SWING

Vs = ±10V

~

+SW1NG

I='"

w

~

8

~

TA =25°C

6

~

-SWING

::>
1

iSWl'NGI

III'""'"
3
'00

'k

,k

'00

'OM

'Ok

MAXIMUM OUTPUT SWING
vs FREQUENCY

~/II

11111 II
11111 II
I'A~CL =5

ill

II III

30
TA

=2SOC

AVCL = 20

Vs = ±15V

Vs = ±15V

~

TA = 25°C

25

~

AC = 3.3k.l1
Cc = lnF

30

0
::>

C L = 100pF

20

"

AVCL = 20

~

i1
~

15

~

o

~

"

"'"

0'

'Ok

Rl = lkn

w 20

'5
'0

Vs = ±5V

o

TA = 2SOC
8

~ 2.

7

5

1

II

SMALL-SIGNAL OVERSHOOT
vs CAPACITIVE LOAD

35

~

~

5

I

LOAD RESISTANCE In)

40

Z

~

o

V'"+SWING

FREQUENCY (Hz)

45

"
~

10

::>

'M

lOOk

'Ok

50

o

~

Vs = ±15V

'"''''1k

OUTPUT VOLTAGE SWING
vs LOAD RESISTANCE

9

~
~

~

FREQUENCY (Hz)

~

"~ '1

30

TION

-SW'~~

w

150

NO
VS= ±15V

13

~ '2

GAIN

40

14

o
2:.

"

~

40
60

~

z

130

z

"

OUTPUT VOLTAGE SWING
vs LOAD RESISTANCE

110

'00

~

CLOSED-LOOP GAIN
vs FREQUENCY
70

ill1l11illil

IlllL-H'lJlJJ
'0

'00
CAPACITIVE LOAD (nFj

LOAD RESISTANCE (nJ

'0

~
'"~

AVCL '" 1000

1000

0
'00

1\
'k

100k

'Ok

,M

FREQUENCY (Hz)

OUTPUT IMPEDANCE
vs FREQUENCY
'00
TA " 25°C

. ~S

10

i ~'I~V
III

'.0

t=:=IA~C~1 ~ '000

O.

,

"

00

,

000

,
'0

'00

AVe l.·l00

'k

'Ok

lOOk

'M

FREQUENCY (Hz)

5-188

1/86, Rev. A

------------11fMD

OP-50 HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

overdriving of the long-tailed transistor pair and stop saturation of the output transistor. Power supply voltage is set to
±5V to lower the quiescent power dissipation and minimize
thermal feedback due to output stage dissipation. Operating
from ±5V supplies also reduces the OP-50 rise and fall times
as the output slews over a reduced voltage range. This, in
turn, reduces the output response time

TCVos TEST CIRCUIT
100kfl 1%*
+15V

1001%*

It IS common practice with voltage comparators to ground
one Input terminal and to use a single-ended input. The
historic reason is poor common-mode rejection on the input
stage. In contrast, the OP-50 has very high common-mode
rejection and is capable of detecting microvolt level differences in the presence of large common-mode signals.

-15V

The comparator IS not fast, but It is very sensitive and can
detect signal differences as low as O.3p.V. With large input
overdrives, the circuit responds in approximately 31'S. If
sharp transitions are needed, the use of a TTL Schmit!trigger input is recommended. A table of Response Time vs.
Input Overdrive is shown below.

"VISHAV TYPE S102K RESISTORS

APPLICATIONS INFORMATION
HIGH-SENSITIVITY VOLTAGE COMPARATOR
A comparator capable of resolving a submicrovolt difference
signal is shown in Figure 1. The OP-50, operating without
feedback, drives a second gain stage which generates a TTLcompatible output signal. Schottky-clamp diodes prevent

INPUT OVERDRIVE

Positive Output Delay
Negative Output Delay

100mV

10mV

1mV

100"V

10"V

3.2,,5
1.8,,5

5,,5
5,,5

40,,5
50,,5

340"s
380,,5

2.4ms
4.5ms

. -......- - - -.....- - - -......,....._----o+5V

*,on,

22kfl

*'

10nF

CERAMIC

15kO

2X
1N8263

820.0

10nF
~-+--+-~--_+-----+--oW

'--+-______________-+_____-o_5V

FIGURE 1: HIGH-SENSITIVITY VOLTAGE COMPARATOR

1/86, Rev. A

5-189

--

-------

- -------------------

--

----------I~ OP-50 HIGH-OUTPUT-CURRENT OPERATIONAL AMPLIFIER

INTEGRATOR AND UNITY-GAIN BUFFER
Figure 2 shows a method of obtaining unity-gain in a buffer
configuration. The R1 and C1 network provides input compensation to circumvent the minimum gain requirement
Figure 3 shows the same technique applied in the inverting
mode to form a high precision integrator.

l00kn ±O 1%

lkO ±1%

+15V

10k!l±Ol%

VIN(OIFF)

lkn

[~-,\N'v-_4
...

lOm±Ol%

>''---......~IM,.-..,...-o 'OUT

10M2 ±O 1%

100kn±O 1%

2kn
RourTRIM
HOWLAND CURRENT PUMP

VOLTAGE COMPLIANCE ±11 OV@'OUT=20mA
OUTPUT RESISTANCE, ROUT (AT lOUT = 20mA AND RL .;;;; SOOn) E!!2Mn
STABLE WITH ALL VALUES OF CAPACITIVE LOAD

FIGURE 2: UNITY GAIN BUFFER

FIGURE 4: 20mA CURRENT SOURCE
C

DRIVING 500 LOADS
The OP-50can provide up to SOmA into a500 load and up to
26mA into a 5000 load. The output is stable driving capacitive
loads of up to 10nF.

V,No-......Mr-+---.!..j
Cl

o lJ.lF

660n

FIGURE 3: INTEGRATOR
20m A CURRENT SOURCE
The 20m A current source exploits the high output current
and high linearity capabilities of the OP-50. Five precision
resistors and a trim potentiometer are required in this circuit
configuration, known as the Howland Current Pump. The
trim potentiometer is used to balance the resistive feedback
dividers. This maximizes the current-source output impedance. Compensation is selected for a voltage gain of 10.

Applications that make use of the high output current capability of the OP-50 will cause increased power dissipation in
the amplifier. To reduce internal dissipation in these applications, external voltage dropping resistors can be connected
in series with the output-stage power supply pins. As shown
in Figure 5,1300 resistors can be attached to pin 7 (-Vop) and
to pin 10 (+Vop). To maintain stability and specified performance levels, 0.0471'F decoupling capacitors should be used
as Indicated from pin 7 and pin 10 to ground.
,-------._-o+15V

-IN

>''----",,-0 Va
+IN

Compliance is better than ±11 V at an output current of 20mA
and the trimmed output resistance is typically 2MO with
RL:5 5000. The transfer function is given by:

son
C3

LOAD

'~~n o04L{C4*-:
O.047iJ f

lOUT =

VIN (DIFF) X 10.1
101
Amps

' - - - -.......----o-16V
NOTE
RESISTORS Rl AND R2 REDUCE IC POWER DISSIPATIDN

VIN (DIFF) is the differential input voltage. For the resistor
values shown in Figure 4, the maximum VIN (DIFF) is 200mV.

FIGURE 5: DRIVING 500 LOADS

5-190

1/86, Rev. A

OP-77
NEXT GENERATION OP-07
(ULTRA-LOW OFFSET VOLTAGE OPERATIONAL AMPLIFIER)
Precision Monolithics Inc.

PRELIMINARY
FEATURES
•
•
•
•
•
•
•
•

Low initial Vos drift and rapid stabilization time, combined
with only 50mW power consumption, are significant improvements over previous designs. These characteristics, plus the
exceptional TCVos of 0.3/l-V/oC maximum and the low Vos of
25/l-V maximum, eliminates the need for Vos adjustment and
increases system accuracy over temperature.

Outstanding Gain Linearity
Ultra High Gain ........................ SOOOV/mV Min
Low Vos ................................... 2S/l-V Max
Excellent TCVos ...................... " 0.3/l-V/oC Max
High PSRR ................................ 3/l-V/V Max
High CMRR .............................. 1.0/l-V/V Max
Low Power Consumption .................. 60mW Max
Fits 0 P- 07, 725, 108A/308A, 741 Sockets

PSRR of 3/l-V/V (110dB) and CMRR of 1.0/l-V/V maximum
virtually eliminate errors caused by power supply drifts and
common-mode Signals. This combination of outstanding
characteristics makes the OP-77 ideally suited for highresolution instrumentation and other tight error budget
systems.

ORDERING INFORMATIONt
PACKAGE
HERMETIC
DIP
8-PIN

TO-99
8-PIN
OP77AJ'
OP77EJ
OP77BJ'
OP77FJ

OP77AZ'
OP77EZ
OP77BZ'
OP77FZ
OP77GZ

PLASTIC
DIP
8-PIN
OP77EP
OP77FP
OP77GP

This product is available in five standard grades and three
standard packages: the TO-99 can and the 8-pin mini-dip in
ceramic or epoxy.

OPERATING
TEMPERATURE
RANGE

The OP-77 is a direct or upgrade replacement for the OP-07.
OP-05, 725, or 108A op amps. 741-types can be replaced by
eliminating the Ves adjust pot

MIL
INO
MIL
INO
COM

PIN CONNECTIONS

'For devices processed in total compliance to M IL-STO-883, add /883 after
part number. Consult factory for 883 data sheet.
tAil commercial and industrial temperature range parts are available with
burn-in, For ordering information see 1986 Data Book, Section 2

Vas

GENERAL DESCRIPTION

TRIEM":R~:
vas

-IN 2

The OP-77 significantly advances the state-of-the-art in
precision op amps. The OP-77's outstanding gain of
10,000,000 or more is maintained over the full ±10V output
range. This exceptional gain-linearity eliminates incorrectable system nonlinearities common in previous monolithic
op amps, and provides superior performance in high closedloop-gain applications.

+IN 3

60Ul

5 N,C.
4 V- (CASEI

TO-99 (J-Suffix)

EPOXY MINI-DIP (P-Suffix)
8-PIN HERMETIC DIP
(Z-Suffix)

SIMPLIFIED SCHEMATIC
V+07----~----------~----------~------~--~----~~----~------~~~~--_.

R7
*NOTE
R2A AND R2B ARE
ELECTRONICALLY
ADJUSTED ON CHIP
AT FACTORY

NONINVERTING
INPUT

INVERTING
INPUT

4

v-o---~----------------------~--~----~---+--------~~-+--~

This preliminary product information Is based on testing of a limited number of devices. Final specilicatlons may vary. Please contact local sales
ollice or distributor for final data sheet.

5-191

1/86, Rev. A

-----------I~ OP-77 NEXT GENERATION OP-07 - PRELIMINARY

ABSOLUTE MAXIMUM RATINGS

(Note 2)
NOTES:
1. See table for maximum ambient temperature rating and derating factor.

Supply Voltage •••••••••••••••••••••••••••••••••••• ±22V
Internal Power Dissipation (Note 1) ••••••••••••••• 500mW
Differential Input Voltage ••••••••••••••••••••••••• , ±30V

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

Input Voltage (Note 3) •••••••••••••••••••••••••••.• ±22V
Output Short-Circuit Duration •••••••••••••••••• Indefinite

PACKAGE TYPE

Storage Temperature Range
J and

Z Packages

•••••••••••••••••••• -65° C to + 150° C

P Package ••••••••••••••••••••••••••• -65° C to + 125° C
Operating Temperature Range
OP-77A, OP-77B ••••••••••••••••••••• -55°C to +125°C
OP-77E, OP-77F •••••••••••••••••••••• -25° C to +85° C
OP-77G •••••••••••••••••••••••••••••••••• O°C to 70°C
Lead Temperature Range (Soldering, 60 sec) •••••••• 300° C

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO·99 (J)

BO'C

7.1mW/'C

B-Pin Hermetic DIP (Z)

75'C

6.7mW/'C

B-Pin Plastic DIP (P)

36'C

5.6mW/'C

2.

Absolute maximum ratings apply to both packaged parts and DICE, unless

3.

otherwise noted.
For supply voltages less than ±22V, the absolute maximum input voltage is
equal to the supply voltage.

DICE Junction Temperature (Tj) •••••••• -65° C to + 150° C

ELECTRICAL CHARACTERISTICS

at Vs

PARAMETER

SYMBOL

CONDITIONS

I nput Offset Voltage

Vas

Long-Term Input Offset
Voltage Stability

aVos/Time

= ±15V, TA = 25°C, unless otherwise noted.

OP-77A

Input Offset CUrrent

los

Input Bias Current

I.

Input Noise Voltage

8 np _p

Input Noise Voltage Density
Input Noise Current

MIN

(Note 1)

OP-77B

TYP

MAX

10

25

0.2
1.5

1.2

2.0

O.IHz to 10Hz (Note 2)

0.35

en

fa = 10Hz (Note 2)
fa = 100Hz (Note 2)
fa = 1000Hz (Note 2)

10.3
10.0
9.6

i np_p

O.IHz to 10Hz (Note 2)

Input NOise Current Density

In

fa = 10Hz (Note 2)
fa = 100Hz (Note 2)
fa = 1000Hz (Note 2)

Input Resistance Differential-Mode

R'N

(Note 3)

Input ResistanceCommon-Mode

R'NCM

Input Voltage Range

IVR

Common-Mode Rejection Ratio

CMRR

VCM = ±13V

Power Supply Rejection Ratio

PSRR

Vs = ±3V to ±18V

Large-Signal Voltage Gain

Ava

R l ,,2kll, Vo=±10V

Output Voltage Swing

Va

Rl " 10kll
Rl " 2kll
Rl "lkll

Slew Rate

SR

Rl " 2kll (Note 2)

Closed-Loop Bandwidth

BW

Avel =

Open-Loop Output Resistance

Ro

Vo=O,lo=O

60

Pd

Vs = ±15V, No Load
Vs = ±3V, No Load

50
3.5

Rp = 20kll

±3

Offset Adjustment Range

-02

26

NOTES:
1. Long-Term Input Offset Voltage Stability refers to the averaged trend line of
Vas vs Time over extended periods after the first 30 days of operation
ExcludIng the initIal hour of operation, changes in Vas dunng the first 30
operatrng days are tYPically 2.5~V

60

UNITS
~V
~V/Mo

2.B

1.2

2.B

nA

0.6

0.35

0.6

~Vp.p

lB.O
13.0
11.0

10.3
10.0
9.6

lB.O
13.0
11.0

nV/JHZ

nA

14

30

14

30

pA p_p

0.32
0.14
0.12

O.BO
0.23
0.17

0.32
0.14
0.12

O.BO
0.23
0.17

pA/JHZ

45

±14
0.1

18.5

45

Mil

200

Gil

±13

±14

1.0

1.0

+ 1 (Note 2)

MAX

20

0.1
-0.2

200
±13

TYP

0.2

0.1

Power Consumption

MIN

V

0.1

1.6

1.0

3

~VIV
~VIV

5000

12000

2000

BOOO

VlmV

±13.5
±12.5
±12.0

±14.0
±13.0
±125

±13.5
±12.5
±12.0

±14.0
±13.0
±12.5

V

0.1

0.3

0.1

0.3

V/~s

0.4

0.6

0.4

0.6

MHz

60
60
4.5

50
3.5
±3

Il

60
4.5

mW
mV

2. Sample tested.
3. Guaranteed by design.

5·192

1/86, Rev. A

------------I1fHD

OP-77 NEXT GENERATION OP-07 - PRELIMINARY

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, -55° C:s TA:S +125°C, unless otherwise noted.
PARAMETER

SYMBOL

Input Offset Voltage

Vas

Average Input Offset
Voltage Drift

TCVos

Input Offset Current

los

Average Input Offset Current
Onft

TClos

Input Bias Current

Ie

Average I nput Bias Current
Drift

TCI B

Input Voltage Range

IVR

Common-Mode Rejection Ratio

CMRR

CONDITIONS

MIN

OP-77A
TYP MAX

(Note 1)

(Note 1)
-02

MIN

60

45

120

~V

01

03

02

06

~V/'C

01

22

01

45

nA

05

25

05

50

pAI'C

(Note 1)

-02
25

±13

Power Supply Rejection Ratio

PSRR

Vs = ±3V to ±18V

Large-Signal Voltage Gain

Ava

RL ,,2kll, Vo =±10V

Output Voltage Swing

Va

RL " 2kll

Power Consumption

Pd

Vs = ±15V, No Load

UNITS

25

24

01

24

nA

15

±135

VCM = ±13V

OP-77B
TYP MAX

±13
10

35

pA/'C

±135

V

0.1

~VIV

~V!V

2000

6000

1000

4000

V/mV

±12

±125

±12

±12.5

V

60

75

60

75

mW

Sample tested.
Guaranteed by design

TYPICAL LOW-FREQUENCY NOISE TEST CIRCUIT

TYPICAL OFFSET VOLTAGE TEST CIRCUIT

~
......

.....:l
~

1

~
~

~

33k.l1

loon

Vo
Vas =

4~~O

r

INPUT REFERRED NOISE =

OUTPUT

47J.1.F

v-

OPTIONAL OFFSET NULLING CIRCUIT

-=

(""'10Hz FILTER)

2~~OO

BURN-IN CIRCUIT
100kf2

+1av

>-....- - - - - - o v +

+

T 10J.l.F
\J

1051

7 6

> - ' - - - - - 0 OUTPUT

'C2

* 1 PER BOARD

10,u.F~
-18V

5-193

~

0......

0

loon

=-

INPUT

......

~

25Mn

200kn

50n

~
~

~.....:l

NOTES:
1.
2.

II

1/86, Rev. A

~ OP·77 NEXT GENERATION OP·07 -

PRELIMINARY

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, TA = 25° C, unless otherwise noted.

OP·77E
PARAMETER

SYMBOL

Input Offset Voltage

Vos

Long· Term Vos
Stability

Vos/Tlme

Input Offset Current

los

Input Bias Current

Ie

Input Noise Voltage

e np _p

Input Noise
Voltage Density

MIN

CONDITIONS

(Note 1)

OP·77F/G

TYP

MAX

10

25

MIN

0.3

MAX

UNITS

20

60

/lV

0.4

0.1

1.5

1.2

2.0

O.IHz to 10Hz
(Note 2)

0.35

en

fo= 10Hz
fo = 100Hz (Note 2)
fo= 1000Hz

Input Noise Current

I np_p

O.IHz to 10Hz
(Note 2)

Input Noise
Current DensIty

In

fo= 10Hz
f 0 = 100Hz (Note 2)
fo= 1000Hz

Input ResistanceDifferential-Mode

R'N

(Note 3)

-0.2

TVP

/lV/Mo

0.1

2.8

nA

1.2

2.8

nA

0.6

0.38

0.65

/lV p_p

10.3
10.0
9.6

18.0
13.0
11.0

10.5
10.2
9.8

20.0
13.5
11.5

nV/VHZ

14

30

15

35

pA p_p

0.32
0.14
0.12

0.80
0.23
0.17

0.35
0.15
0.13

0.90
027
0.18

pAlVHZ

26

45

±13

±14

-0.2

18.5

45

Mil

200

Gil

±13

±14

V

Input ReslstanceCommon-Mode

R 'NCM

Input Voltage Range

IVR

Common-Mode
RejectIon RatIO

CMRR

VCM =±13V

0.1

1.0

0.1

16

/lViV

Power Supply
Rejection Ratio

PSRR

Vs = ±3V to ±18V

1.0

3.0

1.0

3.0

/lVIV

Large-SIgnal
Voltage Gain

Avo

R L 2:2kll,
Vc= ±10V

5000

12000

2000

6000

V/mV

Output Voltage
SWIng

Vo

RL2: lOkI!
RL 2: 2kfl
RL2: Ikll

±13.5
±12.5
±12.0

±14.0
±13.0
±12.5

±13.5
±125
±12.0

±14.0
±13.0
±12.5

V

Slew Rate

200

SR

RL2: 2kfl (Note 2)

0.1

0.3

0.1

0.3

VI/ls

Closed-Loop
Bandwidth

BW

AVCL = +1
(Note 2)

0.4

0.6

0.4

0.6

MHz

Open-Loop Output
Resistance

Ro

Vo=O,lo=O

60

60

II

Power Consumption

Pd

Vs = ±15V, No Load
Vs = ±3V, No Load

50
35

Rp= 20kl!

±3

Offset Adjustment
Range

60
45

50
3.5
±3

60
4.5

mW
mV

NOTES:
1. Long-Term Input Offset Voltage Stability refers to the averaged trend line
of Vos vs. TIme over extended periods after the first 30 days of operation.
Excluding the initIal hour of operation, changes In Vos during the fIrst 30
operating days are typically 2.5/lV.
2. Sample tested
3. Guaranteed by deSIgn.

5·194

1/86, Rev. A

------------I~ OP-77 NEXT GENERATION OP-07 - PRELIMINARY

ELECTRICAL CHARACTERISTICS at Vs

=

±15V, -25°C

~ TA ~

+85°e for OP-77E/F, ooe

~ TA ~

+lOoe for OP-77G,

unless otherwise noted.
PARAMETER

SYMBOL

Inpul Offset Voltage

Vos

Average Input Offset
Voltage Dnft

TCVos

Input Offset Current

los

Average Input Offset

(Note 1)

(Note 1)

TClos

Current Drift

MIN

CONDITIONS

-02

OP-77E
TYP MAX

OP-77F/G
MIN
TYP MAX

UNITS

10

45

20

100

MV

01

03

02

06

MvrC

01

22

01

45

nA

05

85

pArC

24

60

nA

15

60

pArC

0.5

40

24

4.0

-02

Input Bias Current

la

Average Input Bias
Current Drift

TCl a

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

VCM~±13V

0.1

10

01

30

MV/V

PSRR

Vs ~ ±3V to ±18V

10

30

10

50

MVIV

Large-Signal
Voltage Gal n

Avo

RL" 2kO
Vo ~ ±10V

Output Voltage
SWing

Vo

RL" 2kO

Power Consumption

Pd

Vs ~ ±15V. No Load

(Note 1)

40
±13.0 ±13.5

±130 ±135

V

Z2
~

Power Supply
Rejection Ratio

2000

6000

1000

4000

±12

±13.0

±12

±130

60

75

60

V/mV

V
75

mW

~

::sp...
~
......:l
~

~

~

OPEN-LOOP GAIN LINEARITY
10M2

OP-77

Vy

TYPICAL
PRECISION OP AMP

Vy
10k!)

.......

0
.......

NOTES:
1. Sample tested.
2. Guaranteed by design.

1M"
~

Vx
l-l0V

.,..

II

OV

l+l0V

V,

AVO"" 650V/mV
RL =2k.\l

NOTES
1 GAIN NOT CONSTANT CAUSES NONLINEAR ERRORS
2 AVO SPEC IS ONLY PART OF THE SOLUTION
3 CHECK YOUR QP AMP PERFORMANCE, ESPECIALLY AT TEMPERATURES

AVO"" 10,CODV/mV
RL '" 2M2

This is the output gain linearity trace for the new OP-77. The
output trace is virtually horizontal at all pOints, assuring
extremely high gain accuracy. The average open-loop gain is
truly impressive - approximately 10,000,000.

Actual open-loop voltage can vary greatly at various output
voltages. All automated testers use end-point testing and
therefore only show the average gain. This causes errors in
high closed-loop gain circuits. Since this is so difficult for
manufacturers to test, you should make your own evaluation.
This simple test Circuit makes it easy. An ideal op amp would
show a horizontal scope trace.

5-195

1/86, Rev. A

0

------------I~ OP-77 NEXT GENERATION OP-07 - PRELIMINARY

DICE CHARACTERISTICS

1.
2.
3.
4.
6.
7.
8.
DIE SIZE 0.100 x 0.061 Inch, 6100 sq. mils
(2.54 X 1.55 mm, 3.935 sq. mm)

BALANCE
INVERTING INPUT
NONINVERTING INPUT
VOUTPUT
V+
BALANCE

For additional DICE Information refer to
1986 Dala Book, Section 2.

WAFER TEST LIMITS at Vs = ± 15V, TA = 25° C for OP-77N/G devices.
OP-77N

OP-77G

LIMIT

LIMIT

UNITS

Vos

40

75

JIoVMAX

Input Offset Current

los

2.0

2.8

nAMAX

Input Bias Current

Is

±2

±2.8

nAMAX

Input Resistance
Differential-Mode

R'N

26

17

MOMIN

Input Voltage Range

IVR

±13

±13

VMIN

16

JIoVIV MAX

3

3

JIoVIV MAX

PARAMETER

SYMBOL

Input Offset Voltage

CONDITIONS

(Notel)

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio

PSRR

Vs = ±3V to ±18V

Output Voltage Swing

Vo

RL = 10kO
RL =2kO
RL = lkO

±13.5
±12.5
±12.0

±13.5
±12.5
±120

VMIN

Large-Signal
Voltage Gain

Avo

RL =2kO
Vo =±10V

2000

1000

VlmV MIN

±30

±30

V MAX

60

60

mWMAX

Differential Input
Voltage
Power Consumption

Pd

NOTES:
1. Guaranteed by deSign.
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ±15V, TA = +25°C, unless otherwise noted.
OP-77N

OP-77G

TYPICAL

TYPICAL

UNITS

Rs= 500

0.1

0.2

JIoVl'C

Rs = 500, Rp = 20kO

01

0.2

p.VI'C

05

0.5

pArC

RL " 2kn

03

0.3

V/p.s

AVCL = +1

0.6

0.6

MHz

PARAMETER

SYMBOL

CONDITIONS

Average Input Offset
Voltage Drift

TCVos

Nulled Input Offset
Voltage Drift

TCVOSn

Average I n put Offset
Current Drift

TClos

Slew Rate

SR

Closed-Loop
Bandwidth

BW

5-196

1/86, Rev. A

OP-90
LOW-VOLTAGE MICROPOWER
OPERATIONAL AMPLIFIER
Precision M0l101ithics Inc.

ADVANCEPRODUCTINFO~TION
FEATURES

The OP-90 draws only 12j.1A of quiescent supply current. while
able to deliver over SmA of output current to a load. The extremely
low input offset voltage of less than 50j.1V virtually eliminates any
need for external Vas nulling. It has an open-loop gain of over 2
million. a CMRR above 110dB, and a PSRR that IS under 10j.1VN.
The OP-90 features a noise voltage density of 35nV/,j'RZ at
30Hz, a very low value in comparison to other micropower
opamps.

• Single/Dual Supply Operation .•.•...• + 1.6V to + 36V
..•..... :!:O.8V to:!: l8V
• Input and Output Swing to Ground in Single
Supply Operation
• Low Supply Current .•••..•.....•...•••.• l8j.1A Max
• High Output Drive .•..•...•.•••.••••.••••. SmA Min
• Low Input Offset Voltage •.•..•.•••.•••••• SOj.1V Max
• High Open-Loop Gain .••....•..•...• 2000VlmV Min

With its low offset voltage and high gain, the OP-90 brings
preCision performance to micropower applications. The low
supply voltage requirements of the OP-90 combined with its
minimal power consumption enables it to be powered by batteries
or solar cells and is ideal for applications in portable Instruments,
remote sensors, and satellites.

ORDERING INFORMATIONt
PACKAGE
TA = 2S·C
VosMax
(j.1V)

HERMETIC
DIP
8-PIN

50
50
100
100

OP90AZ*
OP90EZ
OP90BZ*
OP90FZ

PLASTIC
DIP
8-PIN

OPERATING
TEMPERATURE
RANGE

OP90EP
OP90FP

PIN CONNECTIONS

MIL
INO
MIL
INO

• For deVices processed In total compliance to MIL·STD-883. add 1883 after part
number Consult factory for 883 data sheet
t All commercial and Industnal temperature range parts are available with burn-In
For ordering Information see 1986 Data Book. Section 2.
GENERAL DESCRIPTION

8-PIN HERMETIC DIP
(Z-Sufflx)

The OP-90 is a high-performance micropower op amp operable
from a single supply of + 1.6V to +36V or from dual supplies of
:!:O.SV to :!:1SV. Input voltage range extends to the negative rail
allowing the OP-90 to accommodate input signals down to zero
volts in single supply operation. The OP-90's output will swing to
ground when operating from a single supply. enabling "zero inzero out" operation.
ELECTRICAL CHARACTERISTICS at Vs
PARAMETER
Input Oftset
Voltage
Input Oftset
Current

SYMSOL
Vos

CONDITIONS
Vs~

8-PIN EPOXY MINI-DIP
(P-Suffix)

= :!:15V, TA = 25°C, unless otherwise noted.
OP-90A/E
MIN TYP MAX

±15V

OP-90B
MIN TYP

150
50

v+ ~ +5V,V-~-5V

18

Input NOise
Vo~age DenSity

10

en

fo~30Hz

Input Voltage
Range

IVR

Vs = ±15V
V+ = +5V. V-~OV

Common-Mode
Rejection RatiO

CMRR

VCM~

Power Supply
Rejection RatiO

PSRR

Vs ~ ±1Vto ±15V

±14V

OP-90F
MIN TYP

MAX

UNITS

250
100

ILV

2

25

nA

15

20

nA

250
100

los

InputSlas
Current

MAX

35

40

35

nV/$z

-15/+14.2
0/+42

-15/+140
0/+40

-15/+140
0/+40

V

110

107

104

dB

10

15

15

ILVN

This advance product Informat/on describes a product In development at the time of this printing. Finalspecllications may vary. Please contact
local sales ollice or distributor for IInal data sheet.

5-197

._--- -

-~~~~~

~~~-

1/86, Rev. A

~

OP-90 LOW-VOLTAGE MICROPOWER OPERATIONAL AMPLIFIER - ADVANCE INFORMATION

ELECTRICAL CHARACTERISTICS at Vs = ±15V, TA = 25°C, unless otherwise noted. (Continued)
OP-90B

OP-90A/E
PARAMETER

SYMBOL CONDITIONS

Large,Slgnal
Voltage Gain

Ava

RL ,,100kil

Output Voltage
SWing

Va

RL ,,100kil

Gain Bandwidth
Product

GBW

Supply Current

ISY

Slew Rate

SR

MIN

Vs~

V+

~

±15V
+5V, V-~ OV

TYP

MAX

MIN

TYP

OP-90F

MAX

MIN

TYP

MAX

UNITS

2000

1000

1000

V/mV

±142
0/+42

±140
0/+40

±140
0/+40

V

25
No Load

25
18

8

5-198

18
8

kHz

25
20
8

!LA
V/ms

1/86, Rev. A

OP-207
DUAL ULTRA-LOW Vas

MATCHED OPERATIONAL AMPLIFIER
PreCiSion MOf)ohthlcs Inc.

FEATURES
•
•
•
•
•
•
•
•

parameters is provided between the channels of this dual
operational amplifier.

Low VOS ....................•••••......... 100~V Max
Offlet Voltage Match ....•.•...•..........•. 90~V Max
Offsat Voltage Match VI. Temp•.•....•.. 1.0~V/oC Max
Common-Mode Rejection Match ..........• 103dB Min
Blal Current Match ....................... 3.SnA Max
Low Noise .............................. O.8~Vp_p Max
Low Blal Current ......................... 3.0nA Max
High Channel Separation ........••. . . . . . .• 126dB Min

The excellent specifications of the individual amplifiers
combined with the tight matching and temperature tracking
between channels provide high performance in instrumentation amplifier designs. The individual amplifiers feature
very low input offset voltage, low offset voltage drift, low
noise voltage, and low bias current. Each amplifier is fully
compensated and protected.
Matching between channels is provided on all critical parameters including offset voltage, tracking of offset voltage vs.
temperature, non inverting bias currents, and commonmode rejection.

ORDERING INFORMATIONt
TA=25°C
VosMAX
(!IV)

HERMETIC
DIP
14-PIN

OPERATING
TEMPERATURE
RANGE

100
100
200
200

OP207AY'
OP207EY
OP207BY'
OP207FY

MIL
COM
MIL
COM

PIN CONNECTIONS
NOTES:
1. Device may be operated even
If Insertion IS reversed; this Is
due to Inherent symmetry
of pin locations of amplifiers
A and B.
2 V-IA) and V-IB) are Internally
connected VIS substrate
resistance.

• For devices processed In total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet.
tAli commercial and Indust"al temperature range parts are available with
burn-In. For ordering informallon see 1986 Data Book, Section 2

GENERAL DESCRIPTION
The OP-207 series of dual matched operational amplifiers
consists of two independent OP-07 high performance operational amplifiers in a single 14-pin dual-in-line package.
Exceptionally low offset voltage and tight matching of critical

14-PIN HERMETIC DIP
(V-Suffix)

SIMPLIFIED SCHEMATIC (1/2 OP-207)
v+o-----~--------~----~----~------~--~----~~----~--------~--~---

R9
OUTPUT

Rl0

+IN

-IN D-''M~I----''''----+--'''''------c

v-o---------------~----------

____________

~--~----~--+_------~~--~~

5-199

~~

--~~~

---------------

1/86, Rev. A

II

---------I1fMD

OP-207 DUAL ULTRA-LOW VOS MATCHED OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

Lead Temperature (Soldering, 60 sec) •••••••••••• 300·C

Supply Voltage ••••••••••••••••••••••••••••••••••• ±22V
Internal Power Dissipation (Note 1) •••••••••••••• 500mW
Differential Input Voltage •••••••••••••••••••••••••• ±30V
Input Voltage (Note 2) •••••••••••••••••••••••••••• ±22V
Output Short-Circuit Duration ••••••••••••••••• Indefinite
Storage Temperature Range ••••••••••• -65·C to +150·C
Operating Temperature Range
OP-207A,OP-207B ••••••••••••••••• -55·C to +125·C
OP-207E, OP-207F ••••••••••••••••••••• O· C to + 70· C

NOTES:
1. See table for maximum ambient temperature rating and derating lactor.

MATCHING CHARACTERISTICS at Vs = ±

15V, TA

PACKAGE TYPE

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

106°C

1'.3mW/oC

14-Pln Hermetic DIP

2. For supply voltages less than ±22V, the absolute maximum Input voltage Is
equal to the supply voltage.

= 25· C, unless otherwise noted.
OP-207A1E

PARAMETER

SYMBOL

CONDITIONS

Input Offset VOltage Match

AVos

Rs=loo0

Average Nonlnvertlng
Bias Current

le+

MIN

OP-207B/F

TYP

MAX

TYP

MAX

30

90

50

280

"V

±1.5

±3.5

±1.5

±6.0

nA

MIN

UNITS

Nonlnverting Ollset Current

los+

±0.7

±3.5

±1.0

±6.0

nA

Inverting OllS9t Current

los-

±0.7

±3.5

+1.0

±6.0

nA

Common-Mode Rejection
Ratio Match

ACMRR

VcM =±13V

Power Supply Rejection
Ratio Match

APSRR

Vs = ±3V to ±18V

103

96
32

Channel Separation

126

MATCHING CHARACTERISTICS at Vs =

120

140

114

dB

10
126

51

140

"VIV
dB

± 15V, -55· C :5 TA:5 125· C, unless otherwise noted.

OP-207A

OP-207B

TVP

MAX

TVP

MAX

UNITS

Rs= 1000

70

180

180

450

"V

(Note 1)
Rp=20kO (Note I)

0.5
0.3

1.0
1.0

0.9
0.4

1.5
1.3

"vrc

±6

±3

±12

nA

PARAMETER

SYMBOL

CONDITIONS

Input Ollset Voltage Match

AVos

Input Ollset Voltage
Tracking
WIthout External Trom
With External Trim

TCAVos
TCAVosn

MIN

Average Nonlnvertlng
Bias Current

le+

±2

Average Drift 01 Noninverting Bias Current

TCl e+

10

Nonlnverting Ollset
Current

los+

Average Drift of NonInverting Ollset
Current

TCl os+

2

MIN

12
6.5

3

12
2

Inverting Offset Current

los-

Common-Mode Rejection
Ratio Match

ACMRR

VcM =±13V

Power Supply Rejection
Ratio Match

APSRR

Vs = ±3V to ±18V

100

12

6.5

3
94

51

12

114
16

nA

pAloC

15

117
10

pArC

nA
dB

100

"VIV

NOTE:
1. Sample tested.

5-200

1/86, Rev. A

---------tlfMD

OP-207 DUAL ULTRA-LOW VOS MATCHED OPERATIONAL AMPLIFIER

MATCHING CHARACTERISTICS at VS= ±15V, D·C S TAS 7D·C, unless otherwise noted.
OP-207E
PARAMETER

MAX

TYP

MAX

UNITS

I!.vos

Rs= 1000

80

150

120

350

p.V

TC.l.Vos
TC.l.VoSn

(Note 1)
Rp = 20kO (Note 1)

0.5
0.3

1.0
1.0

0.9
0.4

1.5
1.3

p'vrc

±2

±5

±3

±10

nA

CONDITIONS

Input Offset Voltage Match
Input Offset Voltage
Tracking
Without External Tri m
With External Trim
Average Noninverting
Bias Current

le+

Average Drift of Noninverting Bias Current

TCl e+

Noninverting Offset
Current

los+

Average Drift of NonInverting Offset
Current

TClos+

MIN

OP-207F

TVP

SYMBOL

Inverting Ollsst Current

los-

Common-Mode Rejection
Ratio Match

.l.CMRR

VcM =±13V

Power Supply Rejection
Ratio Match

.l.PSRR

Vs =±3Vto±18V

MIN

10

12

2

3

12

15
5

100

3

117
10

94

pArC
10

pA/'C
10

114

51

16

nA

nA
dB

100

p.VIV

NOTE:
1. Sample tested.

~
~

~

~

Po<

~

~

-~
0

OFFSET NULLING CIRCUIT

BURN-IN CIRCUIT

~

~
2

3

>--+"'-----oN c

4

+18V

1

20k!!

14

-')..

;r
SIDE

IN PUT

~

v+

13

OUT (A)

12

v-

5

v-

OP-207

-18V

11

11

>--I"-----oN c

IN PUT

10

10

~
_.;,.9

•
~-

-

5-201

6

OUT (B)

7

20kH

V+

1/86, Rev. A

IfMD

OP-207 DUAL ULTRA-LOW VOS MATCHED OPERATIONAL AMPLIFIER

INDIVIDUAL AMPLIFIER CHARACTERISTICS at Vs = ±15V, TA = 25°C, unless otherwise noted.
OP-207A1E
MAX

TYP

MAX

UNITS

Rs= 1000

35

100

60

200

p.V

(Note 1)

0.3

I.S

0.4

2.0

p.VlMo
nA

SYMBOL

CONDITIONS

Input Ollset Voltage

Vas

Input Ollset Voltage
StabIlity

<1VosfTlme

MIN

OP-207B/F

TYP

PARAMETER

MIN

Input Oflset Current

los

0.9

28

15

60

Input Bias Current

Is

±1

±3

±2

±7

nA

I nput Noise Voltage

8ne:1!

O.IHz to 10Hz (Note 2)

0.35

0.6

0.35

0.6

p.Vp-p

Input Noise Voltage
Density

en

fa = 10Hz (Note 2)
fa = 100Hz (Note 2)
10 = 1000Hz (Note 2)

10.3
100
9.6

18.0
130

10.3
10.0
9.6

18.0
13.0

nV/y"Hz"

Input Noise Current

Inf!.E!

O.IHz to 10Hz (Note 2)

Input Noise Current
Density

in

fa = 10Hz (Note 2)
fo= 100Hz (Note 2)
fo= 1000Hz (Note 2)

I nput Resistance Differential Mode

R'N

(Note 3)

Input Resistanca Common-Mode

R,NC...

20

14

30

14

30

pAp_~

0.32
0.14
0.12

0.80
0.23

032
0.14
0.12

0.80
0.23

pNy"Hz"

60

8

200

30

MO

120

GO

±13

±14

:1'13'

±14

V

106

123

100

120

dB

Input Voltage Range

IVR

Common-Mode
Rejecloon Ratio

CMRR

Vc ... = ±13V

Power Supply Rejection
Ratio

PSRR

Vs= ±3V to ±18V

Large-Signal Voltage
Gain

Avo

RL2: 2kO, Vo = ±10V

Output Voltage Swing

Va

RL2: 10kO
RL2: 2kO
R L 2:1kO

Slew Rate

SR

RL 2:2kO

Closed-Loop Bandwidth

BW

AvCL =+1

Open-Loop Output
Resistance

Ro

Vo=O,lo=O

Power Consumption

Pd

No Load, Both AmplifIers

180

Rp= 20kO

±4

±4

mV

8

8

pF

Offset Adjustment Range
I nput Capacitance

20

C 'N

32

p.VN

200

500

150

400

VlmV

±12.S
±120
±100

±13.0
±128
±120

±12.S
±120
±10.0

±13.0
±128
±120

V

02

02

VII's

06

0.6

MHz

80

60

0

240

200

300

mW

NOTES:
Long-Term Input Offset Voltage StabIlity refers to the averaged trend line
of Vos vs Tome over extended periods alter the fIrst 30 days of operatIon
ExcludIng the Initial hour of operatIon, changes on Vos dUring the fIrst 30
operatIng days are tYPIcally 2 Sp.V Parameter IS sample tested
2. Sample tested
3 Guaranteed by deSIgn

5-202

1/86, Rev. A

~ OP-207 DUAL ULTRA-LOW VOS MATCHED OPERATIONAL AMPLIFIER

INDIVIDUAL AMPLIFIER CHARACTERISTICS at Vs = ±15V, -55· C:5 TA:5 125· C, unless otherwise noted.
OP-207A
MAX

TVP

MAX

UNITS

Rs= lOOn

75

230

100

400

p.V

Rp=20kll (Notes 1, 21

04
04

13

07
07

1.8

18

56

30

120

SVMBOL

CONDITIONS

Input Ollset Voltage

Vos

Average Input Ollset
Voltage Drift
WIthout External Trim
With External Trim

TCVos
TCVOsn

Input OIIset Current

los

Average Input Ollset
Current Drift

TClos

MIN

OP-207B

TVP

PARAMETER

MIN

10

Input BIas Current

I.

Average Input BIas
Current Drift

TCI.

Input Voltage Range

IVR

Common-Mode
Relection Ratio

CMRR

VcM =±13V

Power Supply
RejectIon RatIo

PSRR

Vs = ±3V to ±18V

Large-Signal Voltage
Gain

Ava

RL "'2kn. Vo =±10V

Output Voltage Swing

Va

RL ",2kn

12

±30

±5.6

±40

12

p.V/'C

nA
pAl'C

±140

nA

18

pA/'C

±13

±135

±13

±135

V

103

120

97

117

dB

~
~
......
~

32

10

51

p.VIV

150

400

120

350

VlmV

±120

±12.8

±12.0

±12.8

V

OP-207E
SYMBOL

MIN

CONDITIONS

TVP

OP-207F

MAX

MIN

TVP

MAX

Input OIIset Voltage

Vas

Rs= 100n

60

200

90

350

Average Input Offset
Voltage Drift
WIthout External Trim
WIth External Trim

TCVos
TCVOSn

Rp = 20kn (Notes 1, 21

04
04

13

07
07

1.8

14

5

25

10

Input Ollset Current

los

Average Input Ollset
Current Drift

TClos

Input Bias Current

I.

Average Input Bias

10
±2

±3

12

TCI.

Current Drift

12
±5

Input Voltage Range

IVR

Common-Mode
RejectIon Ratio

CMRR

VcM =±13V

Power Supply
RejectIon Ratio

PSRR

Vs = ±3V to ±18V

Large-Signal Voltage
Gain

Ava

RL ",2kn, Vo =±10V

Output Voltage Swing

Va

RL ", 2kn

UNITS
p.V

p.V/'C

nA
pAl'C

±11

18

nA
pAl'C

±13

±135

±13

±135

V

103

120

97

117

dB

7

32

10

51

p.VIV

150

400

120

350

VlmV

±120

±128

±120

±12.8

V

NOTES:
1. Exclude first hour of operation to allow for stabilization of external
CircUitry.

2. Sample tested

5-203

-------- --- - - - - -

~...:l
~

0
......

INDIVIDUAL AMPLIFIER CHARACTERISTICS at Vs = ±15V, Q·C:5 TA:5 7Q·C, unless otherwise noted.

PARAMETER

::s
p..

1/86, Rev. A

~

~

C5

---------I~ OP-207 DUAL ULTRA-LOW VOS MATCHED OPERATIONAL AMPLIFIER

APPLICATION OF DUAL MATCHED OPERATIONAL
AMPLIFIERS
ADVANTAGES OF DUAL MATCHED OPERATIONAL
AMPLIFIERS
Dual matched operational amplifiers provide a powerful tool
for the solution of some difficult circuit design problems.
Circuits include true instrumentation amplifiers, extremely
low drift, high common-mode rejection DC amplifiers, low
DC drift active filters, dual tracking voltage references, and
many other demanding applications. These designs all
require good matching between two operational amplifiers.

The circuit below, a differential-in, differential-out amplifier,
shows how errors can be reduced. Assuming the resistors
used are matched, the gain of each side will be identical; if the
offset voltage of each amplifier is matched, then the net
differential voltage at the amplifiers output will be zero. Note
that the output offset error of this amplifier is not a function of
the offset voltage of the individual amplifiers, but only a
function of the difference between the amplifiers' offset
voltages. This error-cancellation principle holds fora number
of input-referred error parameters - offset voltage, offset
voltage drift, inverting and noninverting bias currents,

R3

r20kn
~p--ov,

,

2
R,

~

3

'3

V
OP-207

"
R2

'0

~

V

POWER SUPPLIES
The V+ supply terminals are completely independent and
may be powered by separate supplies if desired. However,
this approach would sacrifice the advantages of the powersupply-rejection-ratio matching. The V- supply terminals are
both connected to the common substrate and must be tied to
the same voltage.
OFFSET TRIMMING
Offset voltage trimming is provided for each amplifier.
Guaranteed performance over temperature is obtained by trimming one side (side A) to match the offset of the other. A net
differential offset of zero results. This procedure is used
during factory testing of the devices. The same results are
obtained by trimming side S to match side A or by nulling
each side individually.

The OP-207 is designed to provide best drift performance
when trimmed with a 20kO potentiometer; this value provides
about ±4mV of adj ustment range which is adequate for most
applications. Trimming resolution can be increased by use of
the circuit shown below.

- .....
510:---

4

IN PUT

*

'4

common-mode and power supply rejection ratios. Note also
that the impedances of each input, both common-mode and
differential-mode, are extremely high, an important feature
not possible with single operational amplifier circuits.
Common-mode rejection can be made exceptionally high;
this is very important in instrumentation amplifiers where
errors due to large common-mode voltages can be far greater
than errors due to noise or drift with temperature. For example, consider the case of two op amps, each with BOdS
(100I'VIV) CMRR. If the CMRR of one device is +100I'VlV
CMRR and the other is-100I'VlV, then the net CMRR will be
200l'VlV, a 6dS degradation. The matching of CMRR
increases the effective CMRR when used as an instrumentation input stage.

OUTPUT

•

y:.c'___~""".,--_-.:.:.:;N't,LL
---J;...

N't,;:;:LL_ _-",~.,....

R4

5-204

RB

1/86, Rev. A

OP-215
DUAL PRECISION JFET-INPUT
OPERATIONAL AMPLIFIER
Precision Monolithlcs Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•
•

Low input offset voltages. low input currents. and low drift are
featured in these high-speed amplifiers.

High Slew Rate .......................... 10V/,..s Min
Fast Settling Time .................. 0.9,..s to 0.1% Typ
Low Input Offset Voltage Drift ........... 10,..V/oC Max
Wide Bandwidth ......................... 3.5MHz Min
Temperature-Compensated Input Bias Currents
Guaranteed Input Bias Current ..... 18nA Max (125° C)
Bias Current Specified Warmed-Up Over Temperature
Low Input Noise Current ........... 0.01pAlv"HZTyp
High Common-Mode Rejection Ratio. . . . . . .. 86dB Min
Pin Compatible With Standard Dual Pinouts
125°C Temperature Tested DICE
Models With MIL-STD-883 Class B Processing Available

On-chip zener-zap trimming is used to achieve 10wVoswhile
a bias-current compensation scheme gives a low input bias
current at elevated temperatures. Thus the OP-215 features
an input bias current of 18nA at 125°C ambient (not junction)
temperature which greatly extends the application usefulness of this device.
Applications include high-speed amplifiers for current output
DACs. active filters. sample-and-hold buffers. and photocell
amplifiers. For additional preCision JFET op amps. see the
OP-15/16/17 data sheet.
PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE
TA = 2SoC
VosMAX
(mV)

TO-99
a-PIN

1.0
1.0
2.0
2.0
4.0
6.0

OP215AJ'
OP215EJ
OP215BJ'
OP215FJ
OP215CJ'
OP215GJ

v+

HERMETIC HERMETIC OPERATING
DIP
TEMPERATURE
DIP
a-PIN
14-PIN
RANGE
OP215AZ'
OP215EZ
OP215BZ'
OP215FZ
OP215CZ'
OP215GZ

OP215AY'
OP215EY
OP215BY'
OP215FY
OP215CY'
OP215GY

8

OUT (AI@_++_70UT(B)

MIL
COM
MIL
COM
MIL
COM

-IN (A) 2

6 -IN (BI

+IN (A) 3

5 +IN (81

4

v-

TO-99
(J-Suffix)

• For deVices processed In total compliance to MIL-STD-883, add /883 after
part number Consult factory for 883 data sheet
tAli commerCial and industrial temperature range parts are available with
burn-In For ordering information see 1986 Data Book, Section 2

14-PIN HERMETIC DIP
(V-Suffix)
'V+ IAI & v+ I BIINTERNALLY CONNECTED

GENERAL DESCRIPTION
The OP-215 offers the proven JFET-input performance advantages of high speed and low input bias current with the
tracking and convenience advantages of a dual op-amp
configuration.

8-PIN HERMETIC DIP
(Z-Suffix)

SIMPLIFIED SCHEMATIC (1/2 OP-21S)
v+

NONINV
INPUT+

r----+---~--O OUTPUT
Al0

'--+----.. Q12

A12

v-o--~~~~-----~--+-+---

__~-+---~---+-+----+-~-~
S·20S

1/86, Rev. A

----------I~ OP-215 DUAL PRECISION JFET-INPUT OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

(Unless otherwise specified the absolute maximum negative input voltage is equal to the negative power supply
voltage.)
Output Short-Circuit Duration ••••••••••••••••• Indefinite
Storage Temperature Range •••••.•••• -65·C to +150·C
Lead Temperature (Soldering, 60 sec) •••••••••••• 300·C
DICE Junction Temperature (T,) ••••••• -65·C to +150·C

(Note 2)

Supply Voltage
OP-215A, OP-2158, OP-215E, OP-215F
(All DICE except GR) ••••••••••••••••••••••••••• ±22V
OP-215C. OP-215G (GR DICE only) •••.••••.••••• ±18V
Internal Power Dissipation (Note 1) ••••.••••••••• 500mW
Operating Temperature Range
OP-215A, OP-2158, OP-215C •.•••••• -55·C to +125·C
OP-215E, OP-215F. OP-215G •••••.••.••• O·C to +70·C
Maximum Junction Temperature (T,) ••••••••••••• +150·C
Differential Input Voltage
OP-215A, OP-2158,
OP-215E, OP-215F,

NOTES:
I See table for maximum ambient temperature rating and derating factor

PACKAGE TYPE

AI DICE
GR)
+40V
(I
except
•••••• -

(All DICE except

GR

TO-99

+20V
) •••••• -

2

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Rs= 500
'G'Grade

Input Offset Current

los

Input BIBS Current

18

Input Resistance

R'N

large-Signal Voltage
Gain

Output Voltage
SWing

MIN

OP-21SB/F

TYP

MAX

02

10

MIN

MAX

TYP

MAX

UNITS

08

20

20
2.5

4.0
60

mV

50
100

3

100
200

pA

±200
±400

±15
±18

±300
±600

pA

50
100

T, = 25"C (Note 11
Device Operating

T, = 25"C (Note 1)

±15

Device Operatmg

±18

±100
±300

±15
±18

1012

Ava

150

500

75

220

Va

RL = 10ka
RL =2kO

±12
±11

±13
±127

±12

±13
±12.7

±11

85

60

Slew Rate

SR

AveL =+1

10

18

7.5

18

GBW

(Note 3)

35

57

35

57

CLBW

AveL = +1

13

Settling Time

ts

to001%
to 0 05% (Note 2)
toO 10%

23
11
09

Input Voltage Range

IVR

BandWidth

Common~Mode

Rejection RatiO

Power Supply
RejectIon RatiO

Input NOI.e Voltage
DenSity

Input NOise Current
DenSity

Input Capacitance

MIN

10 12

RL ,,2kO
Vo=±10V

60

OP-21SC/G

TYP

'G' Grade

Closed-Loop

6.7mW'·C

Vs = ± 15V, TA = 25· C, unless otherwise noted.

ISY

Product

7ImW'·C

7S·C

Absolute maximum ratings apply to both packaged parts and DICE. unless

Supply Current

Gain BandWidth

10.OmW'·C

BO·C

otherwise noted

OP-21SAlE
PARAMETER

lOO·C

(J)

B-Pin Hermetic DIP (Z)

OP-215C, OP-215G (GR DICE only) •••••••••••••• ±16V

ELECTRICAL CHARACTERISTICS at

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

14-Pin Hermetic DIP IY)

OP-215C, OP-215G (GR DICE only) ••••••••••••.• ±30V
Input Voltage
OP-215A, OP-2158,
OP-215E, OP-215F,

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

1012

0

50

200

VlmV

±12
±11

±13
±12.7

85

7.0
7.0

V
10.0
12.0

mA

15

VIps

54

MHz

13

12

MHz

2.3
1.1
0.9

24
12
1.0

ps

V

30

+10.2
-102

+148
-11 5

+10.2
-102

+148
-115

+10.1
-10.1

+148
-11.5

86
82

100
100

86

100
100

82
80

96

CMRR

V = ±IVR A, B, C Grades
eM
E, F, G Grades

PSRR

Vs = ±10V to ±18V
Vs= ±10V to ±15V

10

en

fo= 100Hz
fo= 1000Hz

20
15

20
15

20
15

nVl.JHZ

In

fO= 100Hz
fO= 1000Hz

001
001

0.01
001

001
001

pA/.JHZ

3

3

3

pF

C'N

82
51

10

80

16

5-206

dB

96

100

pVN

1/86, Rev. A

IfHD

OP-215 DUAL PRECISION JFET-INPUT OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ±15V, -55°C ~ TA ~ +125°C, unless otherwise noted.
OP-215A
PARAMETER

SYMBOL

CONDITIONS

MIN

Input Offset Voltage

Vas

Rs= 50n

TCVos
TCVosn

{Note 31
Rp= 100kn

los

TA = + 125°C, Device Operating

0.8
1.2

TI = +125'C
TA = +125°C, Device Operating

±1.5
±22

OP-2158

TYP

MAX

0.5

2.0

MIN

OP-215C

TYP

MAX

1.5

3.0

MIN

TYP

MAX

UNITS

3.0

60

mV

Average Input Offset
Voltage Dnft

Without External Trim
WIth External Tnm
Input Offset Current

(Note 1)
Input Bias Current

(Note 1)
Input Voltage Range
Common-Mode
Rejection Ratio
Power Supply

Rejection Ratio

I.

Voltage Gain
Output Voltage
Swing

TI =+125'C

IVR

CMRR

VcM=±IVR

PSRR

Vs = ±10V to ±16V
Vs = ±10V to ±15V

Large-Signal

Ava
Va

10

RL2: 2kn

Vo =±10V
A L =:: 10kO

10

14
±1Q

±18

~VI'C

0.8
12

14

1.0
15

12
22

nA

±15
±2.2

±10
±18

±1.8
±2.7

±15
±28

nA

+102
-102

+14.6
-11.3

+102
-10.2

+146
-113

+101
-101

+14.6
-113

V

82

97

82

97

80

93

dB

10

100

15

100
23

126

~VIV

30

110

30

110

25

100

VlmV

±12

±13

±12

±13

±12

±13

V

CONDITIONS

Input Offset Voltage

Vas

Rs=50n

TCVos
TCVOSn

{Note 31
Rp= 100kn

los

T, = +70°C
TA = + 70° C, Device Operating

MIN

TYP

MAX

0.4

165

MIN

OP-215G

TYP

MAX

14

2.65

MIN

TYP

MAX

UNITS

3.5

8.0

mV

Voltage Drift
With External Trim
Input Offset Current

{Note 11
Input Bias Current
(Note11

I.

Input Voltage Range

IVR

Common-Mode
Rejection Ratio
Power Supply
Rejection RatiO

Output Voltage
SWing

15

TJ = +70'C
TA = +70°C, Device Operating

15

~V/'C

006
0.08

045
080

0.06
0.08

0.45
0.80

0.08
0.10

0.65
1.2

±0.12
±0.16

±0.70
±140

±0.12
±O 16

±0.70
±140

±0.14
±0.19

±1.8

±0.9

nA

nA

+102
-10.2

+14.7
-11.4

+102
-102

+147
-11.4

+10.1
-10 1

+147
-113

V

80

98

80

98

76

94

dB

CMRR

VCM=±IVR

PSRR

Vs = ±10V to ±16V
Vs = ±10V to ±15V

Ava

RL " 2kn
Va = ±10V

50

180

50

180

35

130

V/mV

Va

RL " 10kn

±12

±13

±12

±13

±12

±13

V

Large-Signal
Voltage Gain

~

~
~

0

Average Input Offset
Without External Tnm

~.....:I

0
......

OP-215F

OP-215E
SYMBOL

:::s
~

~

ELECTRICAL CHARACTERISTICS at Vs = ±15V, DOC ~ TA~ +7DoC, unless otherwise noted.
PARAMETER

~

......
....

13

100

13

100
20

NOTES:
1. Input bias current is specified for two different conditions. The TI = 25' C
specification is with the junction at ambient temperature; the Device
Operating specification is with the device operating in a warmed-up
condition at 25° C ambient. The warmed-up bias current value is correlated to the junction temperature value via the curves of lavs. T Jand IBvs.
T A. PMI has a bias current compensation circuit which gives improved
bias current and bias current over temperature vs. standard JFET input op
amps. Is and los are measured at VCM = O.

2.

3.

5-207

159

~VIV

Settling time isdefined here for a unity gain inverter connection using 2kO
resistors. It is the time required for the error voltage (the voltage at the
Inverting input pin on the amplifier) to sellie to within a specified percent of
its final value from the time a 10V step input is applied to the inverter. See
settling time test circuit.
Sample tested.

1/86, Rev. A

---------f1fMD

OP-215 DUAL PRECISION JFET-INPUT OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS (125°C TESTED DICE AVAILABLE)
8.
9.
10.
11.
12.
13.
14.

1. INVERTING INPUT (A)
2. NONINVERTING INPUT (A)
3. NULL(A)

4. V5. NULL(B)
6. NONINVERTING INPUT (B)
7. INVERTING INPUT (B)

NULL (B)
V+
Vo (B)
V+
Vo (A)
V+
NULL (A)

ALL v+ PADS ARE INTERNALLY CONNECTED

DIE SIZEO.059 X 0.093 Inch,5487 sq. mlls(1.50X 2.36 mm,3.54 sq. mm)
For additional DICE Information refer to 1986 Data Book, Section 2.

WAFER TEST LIMITS at Vs= ±15V, TA = 25° C for OP-215N,
OP-215GT devices, unless otherwise noted.
PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

Input Bias Current

Is

Input Offset Current

los

Large-Signal
Voltage Gain

Avo

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

Rs=500

Vo =±lOV,
RL =2kO

OP-21SNT
LIMIT

OP-215G

OP-21SN
LIMIT

and OP-215GR devices; TA =

125° C

for OP-215NT and

OP-21SGT
LIMIT

OP-21SG
LIMIT

OP-21SGR
LIMIT

UNITS

2

3

2

6

mVMAX

±18

±18

nAMAX

14

14

nAMAX

30

150

30

75

50

VlmVMIN

±10.2

±10.2

±10.2

±10.2

±10.1

VMIN

82

dBMIN

CMRR

VcM=±IVR

82

86

82

86

Power Supply
Rejection Ratio

PSRR

Vs =±lOto±16V
Vs= ±10 to ±15V

100

51

100

60

Output Voltage Swing

Vo

RL = 10kO
RL =2kO

±12

Supply Current

ISY

100
±12
±11

±12

8.5

p.VlV MAX

±12
±11

±12
±11

VMIN

8.5

12.0

mAMAX

NOTES:
For 25° C characteristics of NT & GT devices, see N & G characteristics respectively.
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging Is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs =
PARAMETER
Average Input
Offset Voltage Drift
Average Input
Offset Voltage Drift

±15V, TA = +25°C,

unless otherwise noted.

OP-21SNT
TYPICAL

OP-21SN
TYPICAL

OP-21SGT
TYPICAL

OP-21SG
TYPICAL

OP-21SGR
TYPICAL

UNITS

TCVos

Unnulled
Rp= 100kO

2

2

3

3

4

"VI"C

TCVosn

Nulled
Rp= 100kO

0.5

0.5

2

"VloC
pA

SYMBOL CONDITIONS

Input Offset Current

los

Input Bias Current

Is

Slew Rate

SR

AVCL =+1

Settling Time

ts

to 0.01%
to 0.05%
to 0.10%

Gain Bandwidth
Product

GBW

Closed-Loop
Bandwidth

CLBW

Input Noise
Voltage Density

3

3

3

3

3

±15

±15

±15

±15

±15

pA

17

17

16

16

15

VI"s

2.2
1.1
0.9

22
1.1
0.9

2.3
1.1
0.9

2.3
1.1
0.9

2.4
1.2
10

"s

6.0

6.0

5.7

5.7

5.4

MHz

AVCL = +1

14

14

13

13

12

MHz

en

fo= 100Hz
fo= 1000Hz

20
15

20
15

20
15

20
15

20
15

nVl0iZ

Input Noise
Current Density

in

fo= 100Hz
fo= 1000Hz

0.01

0.01

0.01

0.01

0.01

pA/0iZ

Input Capacitance

C

3

pF

'N

3

3

5-208

3

1/86, Rev. A

----------tlfMD

OP-215 DUAL PRECISION JFET-INPUT OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
SETTLING TIME
LARGE-SIGNAL
TRANSIENT RESPONSE

,0r----r--~T7--~----~~~

SMALL-SIGNAL
TRANSIENT RESPONSE

-10~--~--~~--~----~~~

o

SETTLING TIME (ps)

CLOSED-LOOP
BANDWIDTH AND
PHASE SHIFT VI FREQUENCY

,.
18

~

,

14

PHASE MARGIN - 66°C

'\.

12
10

I\.

-

AV>10

f-

,
\

-2

-.
-4

Vs = :!:15V
TA=2S"C

-8

-10

1M

24

130

9

140

t:

~

~
II:

Ei'"

'80
'90

i

~

......

'6

10

~

'2

":ilz ........

10M

o

100M

~0

2

~

"

~

~

""
~
~

"~

24

-,

20

,.
12

\

-25

o

1001<

25
50
75
TEMPERATURE ("Cl

SLEW RATE

~

60

"

40

~

20

VB

100

20

10

100

lk

10k

tOOk

FREQUENCY (Hz)

1M
FREQUENCY (Hz)

1M

10M 100M

COMMON-MODE REJECTION
RATIO VI FREQUENCY

TEMPERATURE
100

lV, = .1,5V

""" \.

Vs = ±15V

r- -..

NEGATIVE

r- ~ -..

TA '" 25"C

~

60

~

~
40

.....

\.

POSITIVE

20

........

"

~

1

125

80

I'

Vs = ±15V

TA;;; 2SoC

r\.

AV"+1

50

1'1..

,

-20

-60

60

8
4

r-

70
VS· :!:15V
TA" 25"C
AV-+ t

I-

60

~
I
i"""-o ~:THAv=+l

PRODUCT

FREQUENCY (Hz)

28

"

CLo,lo-LJ. ._

w

~

II

MAXIMUM OUTPUT SWING
VI FREQUENCY

z
;;:

GAIN BANDWIDTH

AV=+1

100

:!!

.±5V ",VS <±20V IS <5%

20

200

"'\1

Oi

BANDWIDTH VARIATION FROM

ffi

'70

'20
Vs'" ±15V

120

'50

'"",

28

90
'00
110

160

\.

OPEN-LOOP
FREQUENCY RESPONSE

BANDWIDTH vs TEMPERATURE

10

10M

o

-50

-25

26

60

75

AMBIENT TEMPERATURE (OCI

5-209

100

125

o

1

10

100

lk

tOle

tOOk

1M

10M 100M

FREQUENCY (Hz)

1/86, Rev. A

---------l~ OP-215 DUAL PRECISION JFET-INPUT OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
VOLTAGE NOISE DENSITY
VI FREQUENCY

OUTPUT IMPEDANCE
vs FREQUENCY

POWER SUPPLY REJECTION
VI FREQUENCY
120

'40

r-

~ 100
90

z
>=

0

~

80

60

g,

50

~
a:

40

~

.:

'1

" "'" ,"-

r-- SUPPLY
NEGATIV~

70

a:
~

VS=I±J,'W

TA = 25°C

110

~

20

o

10

100

lk

10k

1M

100

":Jj

"',"' "'
"'
"'
100k

~

~

~UPPLY

0

80
60

z

w

"~

40

>

20

0

I'..

10

120

~

POSITIvE_

I'...

30

TA = 25°C

~

10M

Tk

lOOk

'Ok

FREQUENCY (Hz)

'M

r~/fCORNER

tiil'illilNCi
I1111111

o

10M

1

FREOUENCY (Hz!

,k

100

'0

I
10k

FREQUENCY (Hz)

BASIC CONNECTIONS
SETTLING TIME TEST CIRCUIT
2k.l1 01%

+15Vo------t-----1---+------,

:OJ

-'2k"'' 0.,..'%......--=-I

0 -.......

0
•

5k"

01%

-15Vo-+--+---,=="""---+-+---,\M-+---"-----0

-OV

OUT

IAI

+IN

-5V
NOTE.
Vas CAN BE TRIMMED WITH POTENTIOMETERS RANGING FROM
10kn TO lMn FDA MOST UNITS Tevos WILL BE MINIMUM WHEN
Vas IS ADJUSTED WITH A 100kn POTENTIOMETER.

(PINOUT FOR "J" AND

"z" PACKAGES ONLY)

("Y" PACKAGE ONLY)

5-210

1/86, Rev. A

------------l~ OP-215 DUAL PRECISION JFET-INPUT OPERATIONAL AMPLIFIER

BASIC CONNECTIONS

APPLICATIONS INFORMATION
DYNAMIC OPERATING CONSIDERATIONS

TYPICAL BURN-IN CIRCUIT

As with most amplifiers, care should be taken with lead dress,
component placement and supply decoupling in order to
ensure stability. For example, resistors from the output to an
input should be placed with the body close to the input to
minimize "pick-up" and maximize the frequency of the
feedback pole by minimizing the capacitance from the input
to ground.

+15V

100kn

A feedback pole is created when the feedback around any
amplifier is resistive. The parallel resistance and capacitance
from the input of the device (usually the inverting input) to
AC ground sets the frequency of the pole. In many instances,
the frequency of this pole is much greater than the expected
3dB frequency of the closed-loop gain and consequently
there is negligible effect on stability margin. However, if the
feedback pole is less than approximately six times the
expected 3dB frequency, a lead capaCitor should be placed
from the output to the negative input of the op amp. The value
of the added capaCitor should be such that the RC time
constant of this capacitor and the resistance it parallels is
greater than, or equal to, the original feedback-pole time
constant.

20051

lOOkS?;

NOTES

1

TA ""125°CTO+150°C

2

RESISTORS ARE TYPE
RN55D, ±1%

-15V

("J" AND .. z" PACKAGES ONLY)

5-211

1/86, Rev. A

OP-220
DUAL MICROPOWER OPERATIONAL AMPLIFIER
(SINGLE OR DUAL SUPPLY)
Precision Monolithics Inc.

FEATURES
Excellent TCVos Match ................. 2p.VfO C Max
Low Input Offset Voltage .................. 150p.V Max
Low Supply Current. . . . . . . . . . . . . . . . . . . . . . . . . .. 100p.A
Single-Supply Operation ................ +5V to +30V
Low Input Offset Voltage Drift ............. O.75p.VfOC
High Open-Loop Gain.. . . . . . . . . . . . . . . . . . .. 2000V/mV
High PSRR ................................... 3p.VIV
Low Input Bias Current. . . . . . . . . . . . . . . . . . . . . . . .. 12nA
Wide Common-Mode Voltage
Range ...................... V- to within 1.5V of V+
• Pin Compatible with 1458, LM158, LM2904

•
•
•
•
•
•
•
•
•

ORDERING INFORMATIONt

offset voltage, and input offset voltage tracking as low as
1.0p.V/o C, make this the first micropower precision dual
operational amplifier.
The excellent specifications of the individual amplifiers combined with the tight matching and temperature tracking
between channels provides high performance in instrumentation amplifier designs. The individual amplifiers feature
extremely low input offset voltage, low offset voltage drift,
low noise voltage, and low bias current. They are fully
compensated and protected.
Matching between channels is provided on all critical
parameters including input offset voltage, tracking of offset
voltage vs. temperature, non-inverting bias currents, and
common-mode rejection ratios.

PACKAGE
TA = 25°C
VosMAX
!IN)

150
150
300
300
750
750

HERMETIC
TO-99
8-PIN

HERMETIC
DIP
8-PIN

OPERATING
TEMPERATURE
RANGE

OP220AJ'
OP220EJ
OP220BJ'
OP220FJ
OP220CJ'
OP220GJ

OP220AZ'
OP220EZ
OP220BZ'
OP220FZ
OP220CZ'
OP220GZ

MIL
INO
MIL
INO
MIL
INO

PIN CONNECTIONS

8 V+

OUT IAI@.70UTIBI
A

- +

B

+ -

-IN (A) 2

6 -IN (B)

+IN (AI 3

,. For devices processed In total compliance to MIL-STO-883, add 1883 after
part number Consult factory for 883 data sheet
t All commercial and industrial temperature range parts are available with
burn-In For ordenng mformatlon see 1986 Data Book, Section 2

5 +IN (B)

4V-

TO-99
(J-Suffix)

8-PIN HERMETIC DIP
(Z-Suffix)

GENERAL DESCRIPTION
The OP-220 is a monolithic dual operational amplifier that
can be used either in single or dual supply operation. The low

SIMPLIFIED SCHEMATIC (Each Amplifier)
r-----~------~----------~--~~--~--~--------~------~--~--~--~V+

OUTPUT

_IN
+IN

0----+------+-----1-------+---'

O---~--~--~--~------~------~-------4~------

__------~-----4~--~--~~V-

*ACCESSIBLE IN CHIP FORM ONLY

5-212

1/86, Rev. A

IfMD
ABSOLUTE MAXIMUM RATINGS

OP-220 DUAL MICROPOWER OPERATIONAL AMPLIFIER

(Note 1)

Supply Voltage •••.•..••••••••••••••••••.•.••••••• ±18V
Power Dissipation •.•••••.•.••.•.••••••••..••.•. 500mW

DICE Junction Temperature (TJ) •••.•••

Differential Input Voltage •••.•••... 30V or Supply Voltage
Input Voltage ••••••• , •••.••.••••••.•••••. Supply Voltage

NOTE:

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300°C

Absolute rallngs apply to both DJCE and packaged parts. unless otherwIse
noted

I

Output Short-Circuit Duration ••..•.••...•..•.. Indefinite
Storage Temperature Range •••.••••.•. -65°C to +150°C

-65°C to +150°C

Operating Temperature Range
OP-220A,
OP-220E,

B, C ••••••••.•.•.•••••••.• -55°C to +125°C
F, G ....................... -25°C to +85°C

ELECTRICAL CHARACTERISTICS

at Vs = ±2.5V to ± 15V, TA = +25° C, unless otherwise noted.

OP-220AlE
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Vs = ±2.5V to ± 15V

Input Offset Current

MIN

OP-220B/F

TYP

MAX

120

MIN

OP-220C/G

TYP

MAX

TVP

MAX

UNITS

150

250

300

500

750

p.V

MIN

II
~

los

VCM=O

015

15

02

2

0.2

3.5

nA

Input BIas Current

I.

VCM=O

12

20

13

25

14

30

nA

~
......

Input Voltage Range

IVR

V+=5V, V-=OV,
Vs= ±15V

V

::l
Po.

Common-Mode

Rejecllon RatIo

CMRR

V+=5V, V-=OV,
OV"VCM,,35V
Vs =±15V,
-15V" VCM " 13 5V

Power Supply

PSRR

Vs =±25Vto±15V
V-= OV. V+ = 5V to 30V

large-SIgnal
Voltage Gain

Avo

V+= 5V, V-= OV, RL = lOOkn
tv" Vo ,,3 5V
Vs = ± 15V, RL = 25kn
Vo=±IOV

Output Vollage
SWing

Vo

Rejection Ratio

V+=5V, V-=OV,
RL = 10kn
Vs = ± 15V. RL = 25kn

0/35
-15/135

0/35
-15/135

0/35
-15/135

90

100

85

90

75

85

95

100

90

95

80

90

3
6

10
18

10
18

32
57

32
57

500

1000

500

800

300

500

1000

2000

1000

2000

800

1600

07/4

07/4

08/4

±14

±14

±14

SR

RL = 25kn, (Note I)

0.05

005

005

AVCL = +1, RL = 25kn

200

200

200

ISY

Vs =.t2 5V, No Load
Vs = ± 15V, No Load

100
140

liS
ISO

liS
170

125
190

125
205

ELECTRICAL CHARACTERISTICS at Vs= ±2,5V to ±15V. -55°C:5 TA:5 +125°C for OP-220A,
-25°C:5 TA:5 +85°C for OP-220 E, F, and G. unless otherwise noted.
OP-220AlE
Average Input Offset TCV
Voltage Drift (Note I )
os

CONDITIONS

Vos

Input Offset Current

los

VCM=O

Inpul BIas Current

I.

VCM=O

Input Voltage Range

IVR

V+=5V, V-=OV,
Vs =±15V

Common-Mode
RejectIon Ratio

CMRR

V+=5V, V-=OV,
OV"VCM "3 2V
Vs =±15V
-15V" VCM " 13.2V

Power Supply
Rejection Ratio

PSRR

MIN

Vs =±15V

Input Offset Voltage

Vs = ±2.5V to ± 15V
V-=OV, V+=5Vt030V

"VIV

~

0......

~

OP-220B/F

TVP

MAX

075

B, and

V/p.s
kHz
135
220

,.A

C,

OP-220C/G

TYP

MAX

IS

12

2

200

300

400

500

1000

1300

p.V

05

2

06

25

08

5

nA

12

25

13

30

14

40

nA

0/32
-15/132

MIN

0/32
-15/132

MIN

TYP

MAX

UNITS

3

p.V/oC

0/3.2
-15/13.2

V

85

90

80

85

70

80

90

95

85

90

75

85

dB

6
10

18
32

5·213

-."--------~~--~

18
32

57
100

57
100

~

~
V

BW

SYMBOL

100
180

V/mV

BandWIdth

PARAMETER

~

dB

Slew Rate

Supply Current
(Both AmplifIers)

~

180
320

p.VIV

1/86, Rev. A

.

_.-

~

---~.-.~

IfMD

OP-220 DUAL MICROPOWER OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ±2.5V to ± 15V. -55° C ~ TA ~ + 125° C for OP-220A.
-25°C ~ TA ~ +85° C for OP-220 E, F, and G, unless otherwise noted. (Continued)
OP-220A/E
PARAMETER

SYMBOL

Large-Signal
Voltage Gain

Avo

CONDITIONS
Vs=±15V. RL =50kn
Vo =±10V
V+= 5V. V-= OV.

Output Voltage
SWing

Vo

Supply Current
(Both Amplifiers)

Isv

RL = 20kn
Vs = ± 15V. RL = 50kn

MIN

TYP

500

OP-220B/F

MAX

MIN

1000

500

TYP

e, and C.
OP-220C/G

MAX

800

MIN
400

09/38

0.9/38

1/38

±138

±138

±13.8

135
190

Vs = ±2 5V. No Load
Vs = ±15V. No Load

TYP

170

155

185

250

200

280

MAX

500

UNITS
V/mV

V

170
275

210
330

~A

NOTE: 1 Sample tested

MATCHING CHARACTERISTICS at Vs = ± 15V, TA = 25°C, unless otherwise noted.
OP-220AlE
PARAMETER

SYMBOL

Input Offset
Voltage Match

avos

Average Nonlnvertlng
Bias Current

Nonlnvertlng
Offset Current

CONDITIONS

MIN

OP-220B/F

OP-220C/G

TYP

MAX

300

250

15

TYP

MAX

150

18+

VCM=O

10

20

los+

VCM=O

07

15



180
1M

10k ,ook

SLEW RATE vs
TEMPERATURE

'iii 32 ~VS·±'5V

GAIN

60

o

36

Tl2.le
VS-±1.V-

r\

~

MAXIMUM OUTPUT SWING
VB FREQUENCY

w

'!l

100

!i1

z

ffi

w

~
!;

i'

~ 010

::>

"
001

10
01

10

100

1

1000

1

11

100

1000

FREQUENCY (Hz)

FREQUENCY (Hz)

SMALL-SIGNAL TRANSIENT RESPONSE

LARGE-SIGNAL TRANSIENT RESPONSE

OUTPUT

OUTPUT

r

10kO

'00PF

26.n

5-217

--.~.~---

-----~-

----_.

1/86, Rev. A

------------t[fMD 01'-220 DUAL MICROPOWER OPERATIONAL AMPLIFIER
INSTRUMENTATION AMPLIFIER APPLICATIONS
Of THE OP-220

Dynamic range is limited by Al as well as A2; the output of Al
is:

TWO·OP·AMP CONFIGURATION
The excellent input characteristics of the OP-220 make it
ideal for use in instrumentation amplifier configurations
where low-level differential signals are to be amplified. The
low-noise.low input offsets. low drift. and high gain combined
with excellent CMRR provide the characteristics needed for
high-performance instrumentation amplifiers. In addition.
the power supply current drain is very low.

If the instrumentation amplifier were designed for a gain of 10
and maximum Vd of ± lV. then RN/Ro would need to be four
and Vo would be a maximum of ±10V. Amplifier Al would
have a maximum output of ±5V plus 2VCM. thus a limit of
±10Von the output of Al would imply a limit of±2.5Von VCM.

The circuit of Figure 1 is recommended for applications
where the common-mode input range is relatively low and
differentiai gain will be in the range of 10 to 1000. This twoop-amp instrumentation amplifier features independent
adjustment of common-mode rejection and differential gain.
Input impedance is very high since both inputs are applied to
noninverting op amp inputs.

A nominal value of 100kO for RN is suitable for most applications. A range of 2000 to 25kO for Ro will then provide a gain
range of 10 to 1000. The current through Ro is Vd/RO. so the
amplifiers must supply ±10mV/2000 when the gain is at the
maximum value of 1000 and Vd is at ± 10mV.
Rejecting common-mode inputs is most important in accurately amplifying low-level differential signals. Two factors
determine the CMR of this instrumentation amplifier configuration (assuming infinite gain):

RO

(1) CMRR of the op amps
(2) Matching of the resistor network (RalR4 = R2/R,)

GAIN

r-'IIV'v---"M-.........
R1

AOJ

In this instrumentation amplifier configuration. error due to
CMRR effect is directly proportional to the differential CMRR
of the op amps. Forthe OP-220A/E. this combined CMRR is a
minimum of 9BdB. A combined CMRR value of 100dB and
common-mode input range of ±2.5V indicates a peak inputreferred error of only ±25~V.

R2

Vl

R3

R4

{1121
OP220

VCM + 1/2V d <>'-+--------~

ReSistor matching IS the other factor affecting CMRR.
Defining Ad as the differential gain of the instrumentation
amplifier and assuming that R,. R2. R3 and R4 are approximately equal (RN will be the nominal value). then CMRR will
be approximately Ad divided by 4AR/RN' CMRR at differential gain of 100 would be BBdB with resistor matching of 0.1%.
Trimming R, to make the ratio R3/R4 equal to R2/R, will
directly raise the CMRR until it is limited by linearity and
resistor stability considerations.

Vo

R4[ 1 +2"
1 (R2
R3) + AD
R2+R3J Vd+ii3
R4 (R3
R2) VCM
iii+'R4
Ri- ifi

Vo=R'3

IF Rl· R2· R3· R4, THEN Vo" 2 (1

+i5,) Vd

Figure 1. Two-Op-Amp Instrumentlltlon Amplifier
Configuration

The high open-loop gain of the OP-220 is very important in
achieving high accuracy in the two-op-amp instrumentation
amplifier configuration. Gain error can be approximated by:

The input voltages are represented as a common-mode
input VCM plus a differential input Yd' The ratio R3/R4 is
made equal to the ratio R2/R, to reject the common-mode
input VCM. The differential signal Vd is then amplified according to:

Gain Error -

2 AAdA
0, 02

~1

where Ad is the instrumentation amplifier differential gain
and A02 is the open-loop gain of op amp A2. This analysis
assumes equal values of R,. R2. R3. and R4. For example.
consider an OP-220 with A02 of 700V/mV. If the differential
gain Ad were setto 700. the gain error would be 1/1.001 which
is approximately 0.1%.

Note that gain can be independently varied by adjusting Ro.
From considerations of dynamic range. resistor tempco
matching. and matching of amplifier response. it is generally
best to make R,. R2. R3. and R4 approximately equal. Designating R,. R2. R3. and R4 as RNaliows the output equation to
be further simplified:
Vo=2(1 +

+ .

1 +_d_
A02

Another effect of finite op amp gain is undesired feedthrough
of common-mode input. Defining Ao, as the open-loop gain
of op amp A1. then the common-mode error (CME) at the
output due to this effect will be approximately
CME-

~~)Vd' where RN= R, = R2= R3= R4
5·218

~
1 +Ad
Ao,

Ao,

VCM

1/86, Rev. A

-----------I~ OP-220 DUAL MICROPOWER OPERATIONAL AMPLIFIER

For AdI'Ao,,~1, this simplifies to (2 AdI'A01) xVcM.lftheopamp
gain is 700V/mV, VCM is 2.SV, and Ad is setto 700, then the error
at the output due to this effect will be approximately SmV.
The OP-220 offers a unique combination of excellent dc
performance, wide input range, and low supply current drain
that is particularly attractive for instrumentation amplifier
design.
THREE-OP-AMP CONFIGURATION
A three-op-amp instrumentation amplifier configuration
using the OP-220 and OP-22 is recommended for applications requiring high accuracy over a wide gain range. This

circuit provides excellent CMR over a wide input range. As
with the two-op-amp instrumentation amplifier circuits, tight
matching of the two op amps provides a real boost in performance. The OP-22 is a micropower op-amp featuring
programmable supply current.
A simplified schematic is shown in Figure 2. The input stage
(A 1 and A2) serves to amplify the differential input Vd without
amplifying the common-mode voltage VCM. The output stage
then rejects the common-mode input. With ideal op-amps
and no resistor matching errors, the outputs of each amplifier
will be:
V,

RI

= - (1 +2R'~Vd
- - + VCM
Ro 2

Vo

The differential gain Ad is 1 + 2R,/Roand the common-mode
input VCM is rejected.
VCM + 1I2Vd()..:...---I

R2

V-

This three-op-amp instrumentation amplifier configuration
using an OP-220 at the input and an OP-22 at the output
provides excellent performance over a wide gain range with
very low power consumption. A gain range of 1 to 2000 is
practical and CMR of over 120dB is readily achievable.

Figure 2. Three-Op-Amp Instrumentation Amplifier
Using OP-220 and OP-22

5-219

1/86, Rev. A

OP-221
DUAL LOW-POWER OPERATIONAL AMPLIFIER
(SINGLE OR DUAL SUPPLY)
PrecisIon MOllolithics Inc.

FEATURES
•
•
•
•
•
•
•
•

Excellent TCVos Match .............•... 2p.V/oC Max
Low Input Offset Voltage •................. 150p.V Max
Low Supply Current . . . . . . . . . . . . . . . . . . . . .. 550p.A Max
Single Supply Operation ................ +5V to +30V
Low Input Offset Voltage Drift .......•..... O.75p.VI"C
High Open-Loop Gain ......... . . . . . . .. 1500VlmV Min
High PSRR ................................... 3p.V/V
Wide Common-Mode Voltage
Range ...................... V-to within 1.5V of V+
• Pin Compatible with 1458. LM158. LM2904
ORDERING INFORMATIONt
PACKAGE
TA=25°C
VosMAX

(p.V)
150
150
300
300
500
500

HERMETIC
TO-99
a-PIN

HERMETIC
DIP
a-PIN

OPERATING
TEMPERATURE
RANGE

OP221AJ*
OP221EJ
OP221BJ*
OP221FJ
OP221CJ*
OP221GJ

OP221AZ*
OP221EZ
OP221BZ*
OP221FZ
OP221CZ*
OP221GZ

MIL
INO
MIL
INO
MIL
INO

wide supply voltage range. wide input voltage range, and low
supply current drain of the OP-221 make it well-suited for
operation from batteries or unregulated power supplies.
The excellent specifications of the individual amplifiers combined with the tight matching and temperature tracking
between channels provide high performance in instrumentation amplifier designs. The individual amplifiers feature
very low input offset voltage, low offset voltage drift, low
noise voltage, and low bias current. They are fully compensated and protected.
Matching between channels is provided on all critical
parameters including input offset voltage, tracking of offset
voltage vs. temperature, non-inverting bias currents, and
common-mode rejection.
PIN CONNECTIONS

B V+

.. For devices processed In total complJance to MIL-STD-883, add /883 after
part number Consult factory for 883 data sheet
tAli commerCial and industrial temperature range parts are available With
burn-m For ordering information see 1986 Data Book. SectIOn 2

• v-

TO-99 (J-Suffix)

8-PIN HERMETIC DIP
(Z-Sufflx)

GENERAL DESCRIPTION
The OP-221 is a monolithic dual operational amplifier that
can be used either in single or dual supply operation. The
SIMPLIFIED SCHEMATIC (Each Amplifier)
r-----~------~----------~~-4~--~--~--------~------~--~--~--~V+

-IN
+IN

OUTPUT

0-----+------+-----+-------+---

~--~--~--~--~------~----~~------~~-------+------~----~~--*---~~V*ACCESSIBLE IN CHIP FORM ONLY

5-220

1/86, Rev. A

------------1~ OP-221 DUAL LOW-POWER OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS (Note 2)

Lead Temperature (Soldering, 60 sec.) ............ 300°C
DICE Junction Temperature (TJ) ••••••• -65°C to +150°C

Supply Voltage ................................... ±18V
Power Dissipation (Note 1) ...................... 500mW
Differential Input Voltage .......... 30V or Supply Voltage
Input Voltage ............................ Supply Voltage
Output Short-Circuit Duration ................. Indefinite
Storage Temperature Range ........... -65°C to +150°C
Operating Temperature Range
OP-221A, B, C ...................... -55°Cto+125°C
OP-221E, F, G ....................... -25°C to +85°C

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

PACKAGE TYPE

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (J)

80°C

71mW/OC

8-Pin Hermetic DIP (Z)

75° C

6 7mW/o C

NOTES:
1 See table for maximum ambient temperature ratmg and derating factor
2 Absolute rallngs apply to both DICE and packaged parts, unless otherwise
noted

ELECTRICAL CHARACTERISTICS at Vs = ± 2.5Vto ± 15V, TA = 25° C, unless otherwise noted.
OP-221B/F

OP-221AlE
TYP

MAX

Vos

75

150

Input Offset Current

los

05

3

Input Bias Current

18

50

80

Input Voltage Range

IVR

V+ = 5V, V-=OV
Vs =±15V

Common-Mode
ReleCtlon RatiO

CMRR

V+=5V. V-=OV
OV$VcM $35V
Vs =±15V
-15V $ VcM $13 5V

Power Supply
Rejection RatiO

PSRR

Vs= ±2 5V to ±15V
V-=OV. V+=5Vt03OV

Large-Signal
Voltage Gain

AvO

Vs = ± 15V, RL = 10kO
Vo =±10V

Output Voltage
SWing

Vo

V+=5V. V-=OV.
RL = 10kO
Vs =±15V. RL = 10kO

Slew Rate

SR

Bandwidth

BW

PARAMETER

SYMBOL

Input Offset Voltage

Supply Current
(Both Amplifiers)

ISY

CONDITIONS

RL = 10kO. (Note 1)

MIN

MIN

OP-221C/G

TYP

MAX

TYP

MAX

150

300

250

500

5

15

60

100

70

0/35
-15/135

0/35
-15/135

MIN

UNITS

nA
120

0/35
-15/135

nA

v

90

100

85

90

75

85

95

100

90

95

80

90

dB

10
18

3
6

10
18

32
57

32
57

1500

1000

07/4.1

07/4.1

0.8/4

±138

±138

±13.5

0.2

03

02

450
600

Vs = ±2 5V. No Load
Vs = ± 15V. No Load

800

03

02

600

800
550
800

100
180

500
800

V/mV

V

0.3
kHz

600
600
850

550
850

650
900

ELECTRICAL CHARACTERISTICS at Vs=±2.5V to± 15V, -55° CS: TAS:+ 125° Cfor OP-221 A, Band C;-25° Cs: TAS:+85° C
for OP-221 E, F and G, unless otherwise noted.
OP-221B/F

OP-221A1E
PARAMETER

SYMBOL

CONDITIONS

MIN

Average Input Offset TCIl
Voltage Drift (Note 1)
os
Input Offset Voltage

Vos

Input Offset Current

los

VCM=O

Input Bias Current

Ie

VCM=O

IVR

V+ = 5V. V- = OV
Vs =±15V

CMRA

V+ = 5V, V- = OV
OV$VcM $32V
Vs =±15V
-15V$ VcM $13 2V

Input Voltage Range

Common-Mode
Relection RatiO

MAX

075

1.5

12

150

300

250

450

400

700

I'V

5

15

7

2

10

nA

65

120

80

140

nA

55

0/32
-15/132

-----

100

TYP

0/32
-15/132

MAX

MIN

TYP

0/32
-15/132

MAX

UNITS

3

I'VioC

V

85

90

80

85

70

80

90

95

85

90

75

85

dB

5-221

~-----

MIN

OP-221C/G

TYP

1/86, Rev. A

-----------I~ OP·221 DUAL LOW·POWER OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at VS= ±2.5V to ±

15V, -55° C~ TA~+ 125° C for OP-221 A, Band C;-25° C~ TA~+85° C

for OP-221 E, F and G, unless otherwise noted. (Continued)

PARAMETER

SYMBOL

CONDITIONS

Power Supply
Rejection Ratio

PSRR

Vs =±25Vto±15V
V-= OV. V+= 5V to 30V

Large-Signal
Voltage Gain

Ava

Vs =±15V, RL = 10kn
Vo =±10V

Va

V+ = 5V, V-=OV,
RL = 10kn
Vs = ±15V, RL = 10kn

Output Voltage
SWing

Supply Current
(Both Amplifiers)

OP-221A1E
TYP
MAX
MIN
6
10

18
32

500
700

±135
650
900

TYP

MAX

UNITS

57
100

180
320

JlVIV

600

0.8/38

±14

OP-221C/G
MIN

57
100

800

08/3.8

Vs = ±2.5V, No Load
Vs = ± 15V, No Load

ISY

18
32

1000

±135

OP-221B/F
MIN
TYP
MAX

V/mV

09/37

±14
550
900

V

±132
700
950

600
950

750
1000

JlA

NOTE:
1 Sample tested.

MATCHING CHARACTERISTICS

at Vs = ± 15V, TA = 25° C, unless otherwise noted.

CONDITIONS

OP-221AlE
MIN
TYP
MAX

PARAMETER

SYMBOL

Input Offset
Voltage Match

tJ.Vos

Average Nonmvertmg
BIBS Current

'6+

80

Nonlnvertmg Input
Offset Current

105+

5

Common-Mode
Rejection Ratio
Match (Note 11

tJ.CMRR

VCM = -15V to +13 5V

Power Supply
Rejection Ratio
Match (Note 21

tJ.PSRR

Vs = ±2.5V to ± 15V

50

MATCHING CHARACTERISTICS

OP-221B/F
MIN
TYP
MAX

200

150

400

TYP

MAX

UNITS

250

600

JlV

120

nA

10

nA

100
4

87

92

OP·221C/G
MIN

72

14

dB

44

140

JlV/V

at Vs = ±15V, -55°C ~ TA ~ +125°C for OP-221A, Band C; -25°C ~ TA ~ +85°C for

OP-221 E, F and G, unless otherwise noted. Grades E, F, and G are sample tested.

PARAMETER

SYMBOL

CONDITIONS

OP·221A1E
TYP
MAX
MIN

OP·221B/F
MIN
TYP
MAX

OP-221C/G
MIN

TYP

MAX

UNITS

400

800

JlV

140

nA

Input Offset
Voltage Match

tJ.Vos

Average Nonmverting
BIBS Current

'6+

VCM=O

100

120

Input Offset
Voltage Tracking

TCtJ.Vos

(Note 3)

2

3

3

Nonmvertmg Input
Offset Current

105+

VCM=O

7

6

Common-Mode
Rejection Ratio
Match (Note 11

tJ.CMRR

VcM =-15Vto+132V

Power Supply
Rejection Ratio
Match (Note 2)

tJ.PSRR

100

400

250

3

87

90

82

26

600

85

72

78

JlV/'C

12

80

nA

dB

250

JlV/V

NOTE:
tJ.CMRR 1520 log10 VcMI tJ.CME, where VCM IS the voltage applied to both
nonmvertmg inputs and .:lCME is the difference In common-mode
Input-referred error.
tJ.PSRR IS Input-Referred Differential Error
tJ.Vs
3. Sample tested

5·222

1/86, Rev. A

-----------l1fMD

OP-221 DUAL LOW-POWER OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS
1. INVERTING INPUT (A)
2. NONINVERTING INPUT (A)
3. BALANCE (A)

4. V5.
6.
7.
•.
9.
10.
11.
12.
13.
14.

DIE SIZE 0.096 X 0.061 Inch, 5856 sq. mils
(2.44 X 1.55 mm, 3.78 sq. mm)
NOTE: All V+ PADS ARE INTERNALLY CONNECTED

BALA~PE (B)
NONIN\tERTING INPUT (B)
INVERTING INPUT (B)
BALANCE (B)
V+
OUT (B)
V+
OUT (A)
V+
BALANCE (A)

For additional DICE intormation refer to
1986 Dala Book, Section 2.

~
~
......
;:..t.,

~

~

WAFER TEST LIMITS at Vs= ± 2.SV to ± 1SV, TA = 2S"C, unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Input Offset Current

los

Input BIas Current

Ie

VCM=O

Input Voltage Range

IVR

V+=5V, V-=OV
Vs =±15V

Common-Mode
RejectIon RatIO

CMRR

V- = OV, V+ = 5V, OV" VCM " 3 5V
Vs = ±15V, -15V" VCM " 13 5V

Power Supply
RejectIon RatIO

PSRR

Vs =±2.5Vto±15V
V-= OV, V+ = 5V to 30V

Large-SIgnal
Voltage Gain

Ava

Vs =±15V
RL = 10kO

Output Voltage SWing

Va

V+ = 5V, V- = OV, RL = 10kO
Vs =±15V, RL = 10kO

Supply Current
(Both Amphliers)

ISY

Vs = ±2 5V, No Load
Vs = ± 15V, No Load

VCM=O

OP-221N

OP-221G

OP-221GR

LIMIT

LIMIT

LIMIT

UNITS

200

350

500

p.VMAX

35

5.5

7

nAMAX

85

105

120

nAMAX

0/35
-15/13.5

0/3.5
-15113.5

0/35
-15/135

V MIN/MAX
VMIN

88
93

83
88

75
80

dBMIN

12.5
22.5

40
70

100
180

p.VIV MAX

1500

1000

800

V/mV MIN

07/41
±138

07/41
±138

0.8/4
±13.5

V MIN/MAX
VMIN

560
810

610
860

650
900

p.AMAX

NOTE:
Electrical tests are performed at wafer probe to the lImIts shown Due to variatIons In assembly methods and normal YIeld los., YIeld alter packaging IS not
guaranteed for standard product dIce Consult factory to negotIate specIficatIons based on dIce lot qualifIcatIon through sample lot assembly and testIng

5-223

1/86, Rev. A

~

9

~
~

~

------------l~ OP·221 DUAL LOW·POWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
NORMALIZED INPUT OFFSET
VOLTAGE VI TEMPERATURE

OFFSET VOLTAGE SHIFT DUE
TO THERMAL SHOCK

INPUT OFFSET VOLTAGE
vs SUPPLY VOLTAGE
400

100

40

DEVICE IMMERSED

Vs = ±16V

75

!

50
25

...

0

g

.......

~

"-

~

5

~ 250

t;;

"- ,~

~

"

-75
-100
-50

-26

0

25

50

TEMPERATURE

75

100

150

~

100

o

126

re)

...~
a

65
80

65

~

60

~

5&

!

/ r"""-- r-....

/

70

V

50

./

V

45

I-..

L

/

.....

V

r---.

-

50

~

TEMPERATURE

75

SlF

V

AlE

50
45

""
~

ii!

+~oc

..... 1---"

500
400

~

300

200

........

o

.t2 5

:1:5

--

:1:75

~

........
.....-

C/G

.,::rc

...- ~

Loc

:1:10 :1:125 :1:15 :1:17,6 :1:20

POWER SUPPLY VOLTAGE (VOL lSI

~

I-

1.0

AlE

V

0

'fC/G

15

"IF

'/ t/"
~

05

'""""
r--

50

~

75

~

1~

~

o

700

500
400
300

f..-" V
V
i..--" V

--

0

25

,/

50

75

"IF

If

AlE

,/

100 126 160

TEMPERATURE (DC)

SUPPLY CURRENT vs
TEMPERATURE AT
Vs = ± 15V AND ±2.5V

INITIAL OFFSET VOLTAGE
DRIFT VI TIME

I
V =:t15V..,

..

;;....-

--- ---

.... ...
::::=-

--

I--

./ ~ ~
,/

800

.00

--

io'

TEMPERATURE (Cle)

900

I-""

-

i.-

/ V

rei

600

V

INPUT OFFSET CURRENT
VI TEMPERATURE

-75 -50 -26

1000

700

10 20 30 40 50 60 70 80 90 100
TIME (SEC)

100 126 150

900

600

-20
-20 -10 0

20

v. ", l....-" r-I-..

55

I'...

1000

"..~
.

:1:17.5

25

60

SUPPLY CURRENT VI SUPPLY
VOLTAGE FOR OP·221A1E

.s

:1:15

3.0

35_ _ _

26

:1:126

75

40

0

:1:10

.5

35

-76 -50 -25

:1:76

VS"':l:2SV

c/G

/'

40

:1:5

70

/

lL

:1:2.5

-10

INPUT BIAS CURRENT
VI TEMPERATURE

80
VS=:t1SV

--

.......... AlE

POWER SUPPLY VOLTAGE (VOL lSI

85

75

o

I

,... 8/F

V

~

........

C/G

V

....V

r--....

50

INPUT BIAS CURRENT
VI TEMPERATURE

!

\
f\.
,

200

o

.".

1\

g

-60

-75

--J

300

!i1

:5-25

!

-~~

;;

"~
~

IN 700C OIL BATH

350

-

~ ;;..-

C/G

"IF

AlE
C/G

"IF
AlE

\ \

U.VS'>25V
200
100
-75 -50 -25

0

25

50

75

TEMPERATURE (DC)

5·224

100 125 150

-2 0 L-J.........J--L---l_L--J......-L-L---J:--J
o 1 2 3 4 5 6 7 8 9 10
TIME (MINUTES)

1/86, Rev. A

------------1~ OP-221 DUAL LOW-POWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
OPEN-LOOP GAIN AT ±15V
VB TEMPERATURE

OPEN-LOOP GAIN AT ± 5V
VB TEMPERATURE

140

OPEN-LOOP GAIN
SUPPLY VOLTAGE

VB

140

140

~

do
do

120

;;;

'
VB FREQUENCY
100

10

80

~
~
~

70

50
40

3

z

:!:

§
g

~

60

30

20

~

1\

~

3z

10

~
a'"

........

10

"

01
10

100

lk

10

FREQUENCY (Hz)

100

lk

FREQUENCY (Hz)

NONINVERTING STEP RESPONSE
INPUT
OUTPUT

INVERTING STEP RESPONSE
10kfl
INPUT 10k.Q
OUTPUT

5-226

1/86, Rev. A

-----------I1fW

OP-221 DUAL LOW-POWER OPERATIONAL AMPLIFIER

SPECIAL NOTES ON THE APPLICATION OF
DUAL MATCHED OPERATIONAL AMPLIFIERS

The high open-loop gain of the OP-221 is very important in
achieving good CMRR in this configuration. Finite open-loop
gain of A1 (Ao,) causes undesired feedthrough of the
common-mode input. For Ad/ Ao, «1, the common-mode
error (CME) at the output due to this effect is approximately
(2 Ad/A o,) x VCM. This circuit features independent adjustment of CMRR and differential gain.

ADVANTAGES OF DUAL MONOLITHIC
OPERATIONAL AMPLIFIERS
Dual matched operational amplifiers provide the engineer
with a powerful tool fordesigning instrumentation amplifiers
and many other differential-input circuits. These designs are
based on the principle that careful matching between two
operational amplifiers can minimize the effect of DC errors in
the individual amplifiers.

Three-Op-Amp Configuration
The three-op-amp circuit (Figure3). has increased commonmode voltage range because the common-mode voltage is
not amplified as it is In Figure 2. The CMR of this amplifier is
directly proportional to the match of the CMR of the input op
amps. CMRR can be raised even further by trimming the
output stage resistors.

Reference to the circuit shown in Figure 1, a differential-in,
differential-out amplifier, shows how the reductions in error
can be accomplished. Assuming the resistors used are
ideally matched, the gain of each side will be identical. If the
offset voltages of each amplifier are perfectly matched, then
the net differential voltage at the amplifier's output will be
zero. Note that the output offset error of thiS amplifier IS not a
function of the offset voltage of the individual amplifiers, but
only a function of the difference (degree of matching)
between the amplifiers' offset voltages. This error-cancellation principle holds for a considerable number of input
referred error parameters - offset voltage, offset voltage
drift, inverting and noninverting bias currents, commonmode and power supply rejection ratios. Note also that the
impedances of each input, both common-mode and differential-mode, are high and tightly matched, an important
feature not practical with single operational amplifier circuits.

RD
GAIN
r-'Vv.---J\Nv----<~ ADJ
R1

Vl

V

+

INPUT

OP·221

R2

-=1=-

~

~

(1/2)

OP·221

VCM + 1/2Vd
Va"

of

R4

R3

R3

R'

R2

R4 [ 1 + 2"
, (R2
R3) + R2+R3]
R3
FIT + R4
~

Figure 2

R4 (R3
R4

Vd + Ra

R2)
- AT

IF R1 = R2= R3=R4, THENVO=2 (1

VCM

+~)Vd

+

R'

OUTPU

-

V

VCM -1/2Vd

RD

R4

Vo

Figure 1

INSTRUMENTATION AMPLIFIER APPLICATIONS
Two-Op-Amp Configuration
The two-op-amp circuit (Figure 2). is recommended where
the common-mode input voltage range is relatively limited;
the common-mode and differential voltage both appear at V1.

R2

v-

Figure 3

5-227

1/86, Rev. A

II

OP-227

JPMI)

DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION
OPERATIONAL AMPLIFIER
Prccisiol1 Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•

Excellent Individual Amplifier Parameters
Low Vos ................................... 80!'V Max
Offset Voltage Match ....................... 80!'V Max
Offset Voltage Match vs Temperature ..... l!'V/oC Max
Stable Vos vs Time ...................... l!'V/Mo Max
Low Voltage Noise ................. 3.9nV/y'HZ Max
Fast ..................................... 2.8V/!,s Typ
High Gain ............................ 1.8 Million Typ
High Channel Separation ............... . .. 154dB Typ

ORDERING INFORMATIONt
TA=25°C
YosMAX
(!'Y)

HERMETIC
DIP
14-PIN

OPERATING
TEMPERATURE
RANGE

80
80
120
120
180
180

OP227AY'
OP227EY
OP227BY'
OP227FY
OP227CY'
OP227GY

MIL
INO
MIL
INO
MIL
INO

• For devices processed In total compliance to MIL-STD-883. add /883 after
part number Consult factory for 883 data sheet
tAli commercial and Industnal temperature range parts are available with
burn-in. For ordering information see 1986 Data Book, Section 2

When used in a three-op-amp instrumentation amplifier configuration, the OP-227 can achieve a CMRR in excess of
100dB at 10kHz. In addition, this device has an open-loop
gain of t5M typical with a 1kO load. The OP-227 also features
an 18 of ± 10nA typical, an los of 7nA typical, and guaranteed
matching of input currents between amplifiers. These outstanding input current specifications are realized through the
use of a unique input current-cancellation-circUit which
typically holds 18 and los to ±20nA and 15nA respectively
over the full military temperature range.
Other sources of input-referred errors, such as PSRR and
CMRR, are reduced by factors In excess of 120dB for the
individual amplifiers. D.C. stability is assured by a long-term
drift specification of tOllY/month.
Matching between channels is provided on all critical parameters including offset voltage, tracking of offset voltage vs.
temperature, noninverting bias current, CMRR, and power
supply rejection ratio. This unique dual amplifier allows the
elimination of external components for offset nulling and
frequency compensation.
The OP-227 is pin compatible with the OP-l0 and OP-207.

PIN CONNECTIONS

GENERAL DESCRIPTION

14-PIN CERAMIC DIP
(V-Suffix)

The OP-227 is the first dual amplifier to offer a combination of
low offset, low noise, high speed and guaranteed amplifier
matching characteristics in one device. The OP-227 with a
Vos match of 25!'V typical, a TCVos match of O.3!,VfDC
typical, and a l/f corner of only 2.7Hz is an excellent choice
for precision low nOise designs. These D.C. characteristics,
coupled with a slew rate of 2.SV/!,s typical and a small-signal
bandwidth of SMHz typical, allow the designer to achieve AC
performance previously unattainable with op-amp-based
instrumentation designs.

NOTES:
1 Device may be operated even If
insertion IS reversed, this IS
due to Inherent symmetry of
pm locations of amplifiers A
and 8
V-IAI and V-181 are Internally
connected via substrate
resistance

SIMPLIFIED SCHEMATIC (1/2 OP-227)
r-------~----------~--------~--~~--~-------.------~~~--~~~~v,

OUTPUT

NON
INVERTING
INPUT I,,'_I-+-~~""'-+----E----t:
INVERTING
INPUT H
~4-~~~~------+-------~

L---------------~----~------~----------~~~~--~_v-

5-228

1/86, Rev. A

_ _ _ _ _--j~

OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS
NOTES:

Supply Voltage •••••••••••••••••••••••••••••.••••• ±22V

1. See table for maximum ambient temperature rating and derating factor.

Internal Power Dissipation (Note 1) •••••••••••••• 500mW
Input Voltage (Note 3) •••••••••••••••••••••••••••• ±22V
Output Short-Circuit Duration ••••.•••••.••••.. Indefinite
Differential Input Voltage (Note 2)

•••••••.•••.••••

±0.7V

Differential Input Current (Note 2) ••••••••••••••• ±25mA
Storage Temperature Range •••••••••••

PACKAGE

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

14-Pin (V)

106'C

11.3mW/'C

-65°C to +150°C

Operating Temperature
OP-227 A, OP-227B, OP-227C ••••••• -55° C to + 125° C
OP-227E, OP-227F, OP-227G ••••••••• -25°C to +85°C

3

Lead Temperature Range (Soldering, 60 sec) •••••• 300°C

The OP-227's Inputs are protected by back-to-back diodes. Current
limiting resistors are not used In order to achieve low noise. If differential
input voltage exceeds ±0.7V, the input current should be limited to 25mA.
For supply voltages less than ±22V, the absolute maximum input voltage is
equal to the supply voltage.

INDIVIDUAL AMPLIFIER CHARACTERISTICS

at Vs

PARAMETER

SYMBOL CONDITIONS

OP-227A/E
MIN
TVP
MAX

Input Offset Voltage

Vas

Long· Term Vas
Stability

Vas/Time (Notes 2,4)

Input Offset Current

los

Input Bias Current

18

= ±15V, TA = 25°C, unless otherwise noted.

(Note 1)

OP-227B/F
TVP
MAX
MIN

OP-227C/G
MIN
TVP
MAX

80

40

120

60

180

~V

0.2

1.0

0.3

1.5

0.4

2.0

~V/Mo

50

12

75

nA

35
±IO

±40

±12

±55

±15

±BO

nA

O.OB

0.20

O.OB

0.20

0.09

0.2B

~Vp.p

9.0
5.9
4.6

nV/v"HZ

Input Noise Voltage

8 np _p

0.1 Hz to 10Hz
(Notes 3, 5)

Input Noise
Voltage Density

en

fo = 10Hz (Note 3)
fo=30Hz (Note 3)
fo = 1000Hz (Note 3)

3.5
3.1
3.0

6.0
4.7
3.9

3.5
3.1
3.0

6.0
4.7
3.9

3.B
3.3
32

Input Noise
Current Density

in

fo= 10Hz (Notes 3, 6)
fo = 30Hz (Notes 3, 6)
fa = 1000Hz (Notes 3, 6)

1.7
1.0
04

4.5
2.5
0.7

17
1.0
04

45
2.5
0.7

1.7
1.0
04

Input ResistanceDifferential-Mode

A'N

Input Resistance Common-Mode

R 1NCM

(Note 7)

IVA
CMAA

VCM = ±11V

Power Supply
Rejection RatiO

PSAA

Vs = ±4V to ±IBV

Ava

A L <:2kn, Vo=±IOV
AL<: 600n, Vo= ±IOV

Va

AL<: 2kn
AL<: 600n

Output Voltage Swing

Mn
Gn

±123

±110

±123

±11.0

±12.3

V

114

126

106

123

100

120

dB

10

10

20

1800
1500

1000
800

1800
1500

700
600

1500
1500

V/mV

±12.0
±10.0

±13.8
±115

±12.0
±IO.O

±13.8
±115

±11.5
±IO.O

±13.5
±11.5

V

28

1.7

2.8

17

2.8

V/~s

5

8

5

SA

AL <: 2kn (Note 4)

GBW

(Note 4)

Open-Loop Output
Resistance

Ao

Vo=O,lo=O

70

Power Consumption

Pd

Each Amplifier

90

1.7

Ap= 10kn

±4

5

NOTES:
Input offset voltage measurements are performed by automated test
equipment approximately 05 seconds after application of power
AlE grade specifications are guaranteed fully warmed up
Long-Term Input Offset Voltage Stability refers tothe average trend line of
Vos vs Time over extended periods after the first 30 days of operation
Excluding the Initial hour of operation, changes In Vos dUring the first 30

70
140

90

70
140

100

±4

4
5.
7

MHz

--

----~---

n
170

±4

mW
mV

days are tYPically 2 5JlV - refer to tYPical performance curve.
Sample tested
Parameter IS guaranteed by design
See test circuit and frequency response curve for 0 1Hz to 10Hz tester.
See test circuit for current nOise measurement
Guaranteed by Input bias current

1/86, Rev. A

5-229

- - - - - ---------

~VIV

1000
800

Slew Aate

--------

------

---"------

------------

~

~
......
>L.
......

.....:I
p..

~

~

0
......

~

~

p..

0

±IIO

Gain Bandwidth Prod.

Offset Adjustment
Aange

0.7

25

Input Voltage Aange

Voltage Gain

PA/v"HZ

0.7

094

13

Common-Mode
Rejection RatiO

Large-Signal

UNITS

20

------JIEMD

OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

INDIVIDUAL AMPLIFIER CHARACTERISTICS for Vs = ± 15V, -55° C ~ TA ~ + 125° C, unless otherwise noted.
OP-227A
PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

(Note 1)

Average Input
Offset Drift

TCVos
TCVOsn

(Notes 2, 3)

Input Offset Current

los

Input Bias Current

Ie

Input Voltage Range

IVR

Common-Mode

MIN

CMRR

VCM

Power Supply
Rejection Ratio

PSRR

Vs ~ ±4.5V to ±18V

Large-Signal
Voltage Gain

AvO

RL;e 2kll, Vo

Output Voltage SWing

Vo

Rejection Ratio

~

±10V

~

±10V

OP-227B

TYP

MAX

60
0.3

MIN

MAX

TYP

MAX

UNITS

180

80

270

110

350

!'V

10

0.4

1.5

0.5

1.8

15

50

22

85

30

135

nA

±60

±28

±95

±35

±150

nA

±10.0

±115

±10.0

±115

±10.0

±11.5

V

108

122

100

119

94

116

dB

16

20

Input Offset Voltage

Vos

Average Input

TCVOS
TCVOS n

Offset Drift
Input Offset Current

500

1000

300

800

V

±11.5

±13.5

±110

±132

±105

±130

V

~

TA ~ 85° C, unless otherwise noted.
OP-227F

TYP

MAX

(Note 1)

40

(Note 2)

0.5

MIN

los

Common-Mode
Rejection Ratio

Power Supply
Rejection RatiO

IVR
CMRR

PSRR

Large-Signal
Voltage Gain

AVO

Output Voltage Swing

Vo

Vs

~

MIN

OP-227G

TYP

MAX

TYP

MAX

140

60

200

85

280

10

04

15

0.5

18

MIN

UNITS

10

50

14

85

20

135

nA

±14

±60

±18

±95

±25

±150

nA

± 10.0

± 11.8

±10.0

±11.8

110

124

102

121

±10.0

±11.8

V

96

118

dB

16

15

±4.5V to ±18V

RL;e 2kll

51

1200

Input Bias Current
Input Voltage Range

4

600

OP-227E
SYMBOL CONDITIONS

MIN

±20

INDIVIDUAL AMPLIFIER CHARACTERISTICS for Vs = ±15V, -25°C
PARAMETER

OP-227C

TYP

32

750

1500

700

1300

450

1000

V/mV

±11.7

±13.6

±11.4

±13.5

±11.0

±133

V

NOTES:
Input Offset Voltage measurements are performed by automated test
equipment approximately 0.5 seconds after application of power.
2. The TeVos performance is within the specifications unnulled or when

nulled with Rp = 8kO to 20kO, optimum performance is obtained with

Rp ~ 8kll.
Sample tested.

MATCHING CHARACTERISTICS for Vs = ± 15V, TA = 25° C, unless otherwise noted.
OP-227A/E
PARAMETER

SYMBOL CONDITIONS

MIN

Input Offset Voltage
Match
Average Nonmverting
Bias Current
Noninverttng Offset

Current

los+

Inverting Offset Current
Common-Mode

110

Rejection Ratio Match

Power Supply
Rejection RatiO Match

IlPSRR

Vs ~ ±4V to ±18V

Channel Separation

CS

(Note 1)

OP-227B/F

TYP

MAX

25

MIN

MAX

80

35

±10

±40

±12
±12

TYP

MAX

150

55

300

±12

±55

±15

±90

nA

±60

±15

±80

±20

±130

nA

±60

±15

±80

±20

±130

nA

123

103

120

10
126

OP-227C/G

TYP

154

5-230

MIN

97

117

10
126

154

UNITS

dB

20
126

154

dB

1/86, Rev. A

-----

~

------

OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

MATCHING CHARACTERISTICS for Vs= ±15V, TA = -55°C to +125°C, unless otherwise noted.
OP-227A
PARAMETER

SYMBOL CONDITIONS

Input Ollset Voltage
Match

t!.Vos

Input Ollset Voltage
Tracking

TCt!.Vos

Average Noninverting
Bias Current

la+

Average Drift of NonInverting Bias Current

TCl a+

Nonlnvertlng Ollset
Current

los+

MIN

Nulled or
Unnulled (Note 2)
I

_la+A+la+a
a + - - -2- -

OP-227B

TYP

MAX

55

MIN

MAX

180

75

0.3

1.0

±20

±60

100

Average Drift of NonTC1 08+
Inverting Ollset Current

±90

Inverting Ollset Current

I os-

108-= I a-A-I a-a

t!.CMRR

VCM=±10V

Power Supply
Rejection Ratio Match

t!.PSRR

V8 = ±4.5V to ±18V

MAX

UNITS

300

100

480

p.V

0.4

1.5

0.5

1.8

p.V/'C

±28

±95

±35

±170

nA

200
±140

±45

200

±25
105

TYP

±35

130

Common-Mode
Rejection Ratio Match

MIN

160

±25

los+ = I a+A-1a+a

OP-227C

TYP

±35

±90

118

97
16

pA/'C
±250

250
±140
90

3

pA/'C

±45

114

±250

110

20

nA

nA
dB

4

51

p.VIV

TYP

MAX

UNITS

OP-227F

OP-227E
SYMBOL CONDITIONS

Input Ollset Voltage
Match

t!.Vos

Input Ollset Voltage
Tracking

TCt!.Vos

Average No",nvertlng
Bias Current

la+

TYP

MAX

40
NUlled or
Unnulled (Note 1 )
_la+A+la+a
a+--2- -

Average Drift of NonInvertl ng Bla. Cu rrent

TCl a+

Noninverting Ollset
Current

los+

I

MIN

MIN

210

90

400

p.V

~

0.3

1.0

0.4

1.5

05

1.8

p.V/·C

C5

±14

±60

±18

±95

±25

±170

nA

±20

140
±90

±25

±90

±25

130
los- = la-A-I a- a
VcM =±10V

Power Supply
Rejection Rat,o Match

t!.PSRR

Vs = ±4.5V to ± 18V

0......

65

MIN

ILl

Average Drift of NonTClos+
inverting Ollset Current
los-

OP-227G

140

±20

t!.CMRR

~

MAX

los+ = I a+A-1a+a

Common-Mode
Rejection Ratio Match

2

TYP

80

Inverting Ollset Current

~

~

MATCHING CHARACTERISTICS for Vs = ±15V, TA = -25°C to +85°C, unless otherwise noted.
PARAMETER

ffi

......

106

180
±140

±35

±140

±35

200

98

120
15

pA/'C
±250

250

117

90

pA/'C
±250

112

16

3

nA

nA
dB

32

p.VIV

NOTES:
1. Sample tested.
2. Guaranteed by de.'gn.

1/86, Rev. A

5-231

~~~~~~~-------

-~--~-~--~-----

-~---

-

-------~-~-

------II£MD

OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
VOLTAGE NOISE TEST CIRCUIT (0.1 Hz-TO-10Hzp _p)

LOW-FREQUENCY NOISE
120

~
100

80

~
0
z

40

~

-40

."

236pfF
SCOPE

-80

>

XI
RIN"",M,Q

-120

110kn

o 1Hz TO 10Hz PEAK-TO-PEAK NOISE
NOTE: OBSERVATION TIME MUST BE LIMITED TO 10 SECONDS
TO ENSURE 0.1 Hz CUTOFF.

VOLTAGE NOISE DENSITY
va FREQUENCY

,a

'00

9
8
1

INPUT WIDEBAND NOISE
va BANDWIDTH (0.1 Hz TO
FREQUENCY INDICATED)

COMPARISON OF OP-AMP
VOLTAGE NOISE SPECTRA
t---!i;;

10

141

TA - 25"C
Vs • +15V

TA '" 2SoC
Vs '" ±15V

~
\

\

I/f CORNER

'" 2 7Hz

-

,

......

,

Ilf CORNER
Ilf CORNER
27Hz

~~

I

~~

lOW NOISE
AUDIO
QP AMP

IIIII

11111

INSTRUMENTATION

RANGE, TO DC

1~,~-L~~lO~-L~W,~oo~-L~WJ'k

10

1

IIIII

AUDIO RANGE
TO 20kHz

00

100

lk

TOTAL NOISE va
SOURCE RESISTANCE

VOLTAGE NOISE DENSITY
va TEMPERATURE

~

lOOk

~

I-----I-jl-+++++fj--

RS "" 2R1

.

~

r-~'O~_
~

10k

'00

1
12

lk

CURRENT NOISE DENSITY
va FREQUENCY

==~­

5Vt-+1+H+--_
I-I--_VS-t-=-j±1r-

100

BANDWIDTH (Hz)

100 F===f===f9''FffffF=====t'I
TA=2"C

,

FREQUENCY (Hz)

FREQUENCY (Hz)

F

........ 1-"

I/f CORNER

-- -

OP-227

AT 10Hz

~'kHz

Ilf CORNER

I-- - =
0'
100

'k

SOURCE RESISTANCE (.Q)

10k

-50

-25

25
50
75
TEMPERATURE (OC)

5-232

100

125

10

140Hz

IIIIIIII I
100
lk
FREQUENCY (Hz)

10k

1/86, Rev. A

------I~ OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
SUPPLY CURRENT

OFFSET VOLTAGE DRIFT
OF REPRESENTATIVE UNITS

VI SUPPLY VOLTAGE

OFFSET VOLTAGE
STABILITY WITH TIME

120

10

OP-227C

100
80

-

I--TA = +12~

/ / --:::::: /

'~

~

V

TA • +25°C,

t;:: ---:::. ~

"'"

f-"""

>
Ii;

~

~

'0

......

......

.;><

I'-..

-60

-

W

5

~

ffi

~

~

~

~

~

-75 -66 -36 -15 6

Tl

=

;;

+,J

V

/

~

-

OP-'~7B1F

.

I

;::
;::

~

lILV-

OP-227AjE

V~=~,.)-

I I

--I

1

2

3

4

5

1-- B~
1--

7

0
-20

50

~

110

VS .. ±15V -

m 90

I

10

OP-227B

4J.J

§"

o
-75 -50

--

-25

"o
~

OP-221B

i--

0

UI

60

80

'"

50

75

125

50

75

"

'\

125

150

SLEW RATE, GAIN-BANDWIDTH
PRODUCT, PHASE
MARGIN VI TEMPERATURE
1

=~M
±1SV-

-GBW

I'\.

'\

50

-

II'\.

50

100

Vs

"

'\..

30

-

I'\.

r-

-SLEW

I'\.

"

-"I'"

.......

10

100
lk
10k 100k
FREQUENCY (Hz)

5-233

- ----------------

25

80

-10

100

0

I'\.

10

I
25

-25

TEMPERATURE (oC)

OP-227A

TEMPERATURE (oC)

-50

100

70

z

,I'\.

12

OP-227 A

r-...

:;: 70

,~ ~ ~,..,

40

20

:!!
OP-227C

10 11

OP-227C

r-..r-., i'r-., ~rr--

I

OPEN-LOOP GAIN
VI FREQUENCY

II

~

9

IJU

TIME (SEC)

130

t-

20

SHOCK

'If I

w

=

III-- V

OFFSET VOLTAGE CHANGE
DUE TO THERMAL SHOCK

~ 25

OP-227C/G

3

j!:

O.'.V/~ ~

TIME (MONTHS)

'"~

t;

4

'"'"

TEMPERATURE (GCI

3~

r- Vs =±15V

~

iii

...... OP- 27.

-100

TOTAL SUPPLY VOLTAGE (VOLTS)

WARM-UP DRIFT

.....

r-..

-so

,

"" OP-f'7A

:.-::: -.?' I'-..

--10

.I

./

..... iX r-...""'"

f'

-,0

~
0

VrA'" -5S O C

-t-

60
~

V ,..... OP-~.

1M

10M 100M

-75

-50

-25

0

25

50

TEMPERATURE

75

100

125

fe)

1/86, Rev. A

-------1~

OP·227 DUAL LOW·NOISE LOW·OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
GAIN, PHASE SHIFT
VI FREQUENCY
25
20

15

~

~10

\

Z

~

80

Tl J2~~1

~LJ

20 ;--TA

140

1M

10M

10

r--- poslmv1e

Rl= lkH

l-

;...-

SWING

I"
NEGATIVE
SWING

//
V

//

200

'\

220
100M

I--

Vs '" ±15V

60

80



>-

5

~ 30

Vs = ±15V

U

VIN =lOOmV
AV= +1

a:

"I":

-

-:s~1
' SC (+)

~

o
i:!i

TA '" +25 0
Vs = ±1SV

20

10

500

10M

FREQUENCY (Hz)

SMALL-SIGNAL
TRANSIENT RESPONSE

1000
1500
2000
CAPACITIVE LOAD (pFl

a

2500

LARGE-SIGNAL
TRANSIENT RESPONSE

TIME FROM OUTPUT SHORTED TO GROUND (MINUTES)

MATCHING CHARACTERISTIC
CMRR MATCH VI FREQUENCY
140

+50mV

+5V

OV

OV

120

'"

100

-50mV

-5V
80
AVCL = +1, CL:; 15pF
Vs = ±15V

AVCL" +1

TA"25 C

TA = 25°C

Q

Vs

=±15V

60
1k

10k

lOOk

1M

10M

FREQUENCY (Hz)

5·234

1/86, Rev. A

-----t1EHD

OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
COMMON-MODE INPUT RANGE
vs SUPPLY VOLTAGE

OPEN-LOOP VOLTAGE GAIN
VB LOAD RESISTANCE

140

~AI.12~!~II+-+++t+Htt--+H++HH

:: r-

121---+--

• ±15V"t--t-t:J;;I#!!+--++++++1H

120

218t-~-K~r,~-H~-f--+-HH+HH

's
::Sl00

~ 20 t-- Vs

c

"

/

a:
a:

~'6t-~-K~~~-H~-f--+-HH+HH

~e 80
l!

"~14t-~-K~t-~-H~-f--+-HH+HH

g'2t-~-K~t-~-H~-f--~HH+HH

-4~-~~~-~---~---4

-16

a:
a:

~10t-~~~t-~-H~-f--~HH+HH

II

80

~


~

l5

i'

~

-80
~O

~

'

~...

.....

l- f- r- O·-f27'
,... I:>oc f""'.t--.
~
r--. I'"
..... r-.,
V

iiia:

"'" OP-227A

-

......

k
~

~

..... ~ F- OP-227A

~

"

-100

ffi

\

40

\

\

-

~0'-227S

......

-I20

Z

;> ,-227S

\

-56

-35 -15

5

"-

""- ""

0•.227?'10

TEMPERATURE (OC)

25

45

65

85 105

-55

125

-35

TEMPERATURE (OC)

MATCHING CHARACTERISTIC;
CMRR MATCH
VB TEMPERATURE

~

\

\

r-.,

r

o

~

\ \

j P-22 A

Z

25 45 65 85 105 125 145 165

30

>
........

0

' \ OP-227C

9

\
\OP-227C

i'\OP-227C

'\. "-

>
.,.Z 10

I

I
5

20

">=z

I
" ' OP-2278

/'

-120
-75 -55 -35 -15

""

OP-2278

r-- r-.,

30

~

80

a:

OP-~27C

~k

MATCHING CHARACTERISTIC;
p.,
AVERAGE OFFSET
~
CURRENTvsTEMPERATURE
~
(INVERTING OR NONINVERTING) ~

40

I

./~

~

MATCHING CHARACTERISTIC;
AVERAGE NONINVERTING
BIAS CURRENT
VB TEMPERATURE

-15

......

5

25

45

65

85

105

125

TEMPERATURE (OC)

CHANNEL SEPARATION
vs FREQUENCY

125

160
OP-227A

d•.22;s

V

120

L'
115

V

",.....
Ol-227l

.......

r......

r--...

",.....

V

V
110

'"

140

'\

120

""\

100
80

105
-55 -35

I'

60
-15

5

25

46

65

85

105

100

125

TEMPERATURE ("C)

10k

1k

lOOk

1M

10M

FREQUENCV (Hz)

1/86, Rev. A

5-235

~--------------.------~----

-- -

-----.--

~~~-

------I~ OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

BASIC CONNECTIONS

INSTRUMENTATION AMPLIFIER APPLICATIONS OF THE
OP-227
The excellent input characteristics of the OP-227 make it
ideal for use in instrumentation amplifier configurations
where low-level differential signals are to be amplified. The
low-noise, low input offsets, low drift, and high gain combined with excellent CMR provides the characteristics
needed for high-performance instrumentation amplifiers. In
addition, CMR vs. frequency is very good due to the wide
gain-bandwidth of these op amps.

OFFSET NULLING CIRCUIT
V+(A)
10kO

'4

(-)0-4--1

The circuit of Figure 1 is recommended for applications
where the common-mode input range is relatively low and
differential gain will be in the range of 10 to 1000. This twoop-amp instrumentation amplifier features independent
adjustment of common-mode rejection and differential gain.
Input impedance is very high since both inputs are applied to
non-inverting op amp inputs.

>-_-+'.:.3-oOUTIA)

INPUTS

1+10---"-If--I
'--_ _-+'"'-2--0 V-fA)
OP....227

.-----+=---o V-IBI
1+lo-...!'.!.j'f--I
INPUTS

>--+"---oOUTIBI

FIGURE 1: Two-Op-Amp Instrumentation Amplifier Configuration

I-Io-.:.::'O+---I

RO
Rl

R2

V+(B)

V1
VCM - 1/2Vd

APPLICATIONS INFORMATION

0-----1

R4

R3

NOISE MEASUREMENTS
Vo

To measure the 80nV peak-to-peak noise specification of the
OP-227 in the 0.1 Hz to 10Hz range, the following precautions
must be observed:

VCM + 1/2Vd 0 - - - - - - - - - - - 1

-m:r-

R4 [ 1 + '2
'(R2
R3) + R2+R3J Vd
Vo='R3
if'j+Fi4

(1) The device has to be warmed-up for at least five minutes.
As shown in the warm-up drift curve, the offset voltage
typically changes 4p.V due to increasing chip temperature after power-up. In the 10-second measurement
interval these temperature-induced effects can exceed
tens-of-nanovolts.

Vo= R4 (1 + R3 + R2+ R3) Vd, where R3_ R2
R3
R4
Ro
R4 -R;'

(3) Sudden motion in the vicinity of the device can also

"feedthrough" to Increase the observed noise.

VCM

The output voltage Vo, assuming ideal op amps, IS given in
Fig. 1. The input voltages are represented as a commonmode input VCM plus a differential input Vd. The ratio R3/R4is
made equal to the ratio R2/Rl to reject the common-mode
input VCM. The differential signal Vd is then amplified
according to:

(2) For similar reasons, the device has to be well-shielded
from air currents. Shielding minimizes thermocouple
effects.

(4) The test time to measure 0.1 Hz to 10Hz noise should not
exceed 10 seconds. As shown in the noise-tester frequency-response curve, the 0.1Hz corner is defined by
only one zero. The test time of 10 seconds acts as an
additional zero to eliminate noise contributions from the
frequency band below 0.1 Hz.

R4 (R3
R2)
+ii3
R4 - R1

Note that gain can be independently varied by adjusting Ro.
From considerations of dynamic range, resistor tempco
matching, and matching of amplifier response, it is generally
best to make R10 R2, R3, and R4 approximately equal. Designating Rl, R2, R3, and R4 as RN allows the output equation to
be further simplified:

(5) A noise-voltage-density test is recommended when

Vo=2(1 +

measuring noise on a large number of units. A 10Hz
noise-voltage-density measurement will correlate well
with a 0.1 Hz-Io-10Hz peak-Io-peak nOise reading, since
both results are determined by the white noise and the
location of the 1/f corner frequency.

:~)Vd' where RN= Rl = R2= R3= R4

Dynamic range is limited by A1 as well as A2; the output of A1
is:

5-236

1/86, Rev_ A

_ _ _ _ _--/~ OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

If the instrumentation amplifier were designed for a gain of 10
and maximum Vd of ±1V, then RN/Ro would need to be four
and Vo would be a maximum of ±10V. Amplifier A1 would
have a maximum output of ±SV plus 2VCM, thus a limit of
±10V on theoutputof A1 would imply a limitof±2.SVon VCM.
A nominal value of 10kO for RN is suitable for most applications. A range of 200 to 2.SkO for Ro will then provide a gain
range of 10 to 1000. The current through Ro is Vd/Ro, so the
amplifiers must supply ±10mV/200 (or ±O.SmA) when the
gain is at the maximum value of 1000 and Vd is at ± 10mV.

A complete instrumentation amplifier designed for a gain of
100 is shown in Figure 2. It has provision fortrimming of input
offset voltage, CMR, and gain. Performance is excellent due
to the high gain, high CMR, and low noise of the individual
amplifiers combined with the tight matching characteristics
of the OP-227 dual.
FIGURE 2: Two-Op-Amp Instrumentation Amplifier Using
OP-227 Dual

Rejecting common-mode inputs is important in accurately
amplifying low-level differential signals. Two factors determine the CMR in this instrumentation amplifier configuration
(assuming infinite gain):

995kQ

3

,e

10
VCM + 1/2Vd

~l[
~

"

~

13

i-I-l

......

~

12

::s
~

v-

~

"

K

~

6

5

v+
Va

=

100Vd

v-

~

a......

~

10kn, 0 1%
10kn, 0 1%

i-I-l

C;
A three-op-amp instrumentation amplifier configuration
using the OP-227 and OP-27 is recommended for applications requiring high accuracy over a wide gain range. This
circuit provides excellent CMR over a wide frequency range.
As with the two-op-amp instrumentation amplifier circuits,
the tight matching of the two op-amps within the OP-227
package provides a real boost in performance. Also, the lownoise, low offset, and high gain of the individual op-amps
minimize errors.

where Ad is the instrumentation amplifier differential gain
and A02 is the open-loop gain of op amp A2. This analysis
assumes equal values of R" R2, R3, and R4. For example,
consider an OP-227 with A02 of 700V/mV. If the differential
gain Ad were set to 700, then the gain error would be 1/1.001
which is approximately 0.1%.

A simplified schematic is shown in Figure 3. The input stage
(A 1 and A2) serves to amplify the differential input Vd without
amplifying the common-mode voltage VCM. The output
stage then rejects the common-mode input. With ideal opamps and no resistor matching errors, then the outputs of
each amplifier will be:

Another effect offinite op amp gain is undesired feedthrough
of common-mode input. Defining Ao, as the open-loop gain
of op amp A1, then the common-mode error (CME) at the
output due to this effect will be approximately

1 +&L.

1

OP-227

19m

1 +_d_
A02

1
A01

2

7

--~--

~

25n

GAIN

The high open-loop gain of the OP-227 is very important to
achieving high accuracy in the two op-amp instrumentation
amplifier configuration. Gain error can be approximated by

CME-

•

VCM - 1/2Vd

II

~
":"

In this instrumentation amplifier configuration, error due to
CMR effect is directly proportional to the CMR match of the
op amps. Forthe OP-227 this .:lCMR isa minimum of97dBfor
the "G" and 110dB for the "E" grade. A .:lCMR value of 100dB
and common-mode input range of ±2.SV indicates a peak
input-referred error of only ±2S!'V. Resistor matching is the
other factor affecting CMR. Defining Ad as the differential
gain of the instrumentation amplifier and assuming that R"
R2, R3and R4are approximately equal (RNWili bethe nominal
value), then CMR for this instrumentation amplifier configuration will be approximately Ad divided by 4.:lR/R N. CMR at
differential gain of 100 would be aadB with resistor matching
of 0.1%. Trimming R, to make the ratio R3/R4 equal to R2/R,
will raise the CMR until limited by linearity and resistor
stability considerations.

v+

e--

son

(1) CMR of the op amps
(2) Matching of the resistor network ratios (R3/R4 = R2/R,)

Gain Error -

ADJUST

OFFSET
eMR

10kn

01%

2R,)Vd
(
V,=-1+--+VCM
Ro 2

VCM

Ao,

VO=V2-V,=(1

For Ad/Ao,,~1, this simplifies to (2 AJ Ao,) x VCM.lftheop amp
gain is 700VlmV, VCM is 2.SV, and Ad is set to 700, then the error
at the output due to this effect will be approximately SmV.

+~:')Vd

VO=AdVd

5-237

1/86, Rev. A

------I~

OP-227 DUAL LOW-NOISE LOW-OFFSET INSTRUMENTATION OPERATIONAL AMPLIFIER

The differential gain Ad is 1 +2R1/Roand the common-mode
input VCM is rejected.

FIGURE 3: Three-Op-Amp Instrumentation Amplifier Using
OP-227 and OP-27

While output error due to input offsets and noise are easily
determined. the effects of finite gain and common-mode
rejection are more subtle. CMR of the complete instrumentation amplifier is directly proportioned to the match in CMR of
the input op-amps. This match varies from 97dB to 110dB
minimum for the OP-227. Using 100dB. then the output
response to a common-mode input VCM would be:

R1

Vo

[VolcM = Ad VCM X 10-5

=

(1 + 2RRa1 ) Vd

RO

Va

R1

CMRR of the instrumentation amplifier. which is defined as
20 log lOAd/ACM. is simply equal to the aCMRR of the OP-227.
While this aCMRR is already high. overall CMRR of the complete amplifier can be raised by trimming the output stage
resistor network.

V2 R2
VCM + 1/2V d

0----1

R2

Finite gain of the input op-amps causes a scale factor error
and a small degradation in CMR. Designating the open-loop
gain of op-amp A1 as A01. and op-amp A2 as A02. then the
following equation approximates the output:

[AdVd+ R;
2R1 ( 1 1)
1
A01- A 02 VCM

This can be simplified by defining Ao as the nominal openloop gain and aAo as the differential open-loop gain. Then

The unity-gain output stage contributes negligible error to
the overall amplifier. However. matching of the four-resistor
R2-network is critical to achieving high CMR. Consider a
worst-case situation where each R2 resistor has an error of
±aR2. If the resistor ratio is high on one side and low on the
other. then the common-mode gain will be 2aR 2/R2. Since
the output stage gain is unity. CMRR will then be R2/2aR2. It
is common practice to trim the R2 resistor connected to
ground to maximize overall CMRR for the total instrumentation amplifier circuit.

The high open-loop gain of each amplifier within the OP-227
(700.000 minimum at 25° C into RL ~ 2k) assures good gain
accuracy even at high values of Ad. The effect of finite openloop gain on CMR can be approximated by:

This three-op-amp instrumentation amplifier configuration
provides excellent performance over a wide gain range. A
gain range of 1 to 2000 is practical and CMR of over l20dB
is achievable.

Vo -

1
1+!3.!U_+_l_)
Ro \A01

A02

CMRR_ A 02
!1Ao
If aAo/ Ao were 6% and Ao were 600.000. then the CMRR due
to finite gain of the input op-amps would be approximately
l40dB.

5·238

1/86, Rev. A

OP-400
QUAD LOW-OFFSET LOW-POWER
OPERATIONAL AMPLIFIER
Precision Monolithics Inc.

ADVANCEPRODUCTINFO~TION
FEATURES
•
•
•
•
•
•
•

GENERAL DESCRIPTION

Low Input Offset Voltage ••••••••.•••.•.• 100llV Max
Low Input Offset Voltage Drift ••.••••.•• O.5IlVloC Max
High Open-Loop Gain ••.•••••••••••• 2000VlmV Min
Low Input Bias Current ••••.•••••••••••.••• 1nA Max
Low Noise Voltage Oensity ...•••.• 12nVI JHZat 1kHz
Low Supply Current (per Amplifier) ••••••• 700llA Max
Pin Compatible to LM148, LM348, RM4156, and HA4741
with Improved Performance

ORDERING INFORMATIONt
PACKAGE
TA = 25°C
VosMax
(IlV)

HERMETIC
DIP
14·PIN

100
100
300
300

OP400AY'
OP400EY
OP400BY'
OP400FY

PLASTIC
DIP
14·PIN

OPERATING
TEMPERATURE
RANGE
MIL
INO
MIL
INO

OP400EP
OP400FP

The OP-400 is a monolithic quad operational amplifier that
combines precision performance and low power consumption.
This precision is guaranteed by an input offset voltage of less
than 100llV with a drift of under 0.5IlVloC. an input offset current
below 0.2nA, and an open-loop gain of over 2 million into a 2k!l
load. In addition, the OP-400 has an input bias current of less
than 1nA and a CMRR and PSRR of over 115dB. The OP-400
meets low power requirements by drawing less than 700llA of
supply current per amplifier. Voltage noise density of the OP-400
is a low 12nV/JRZat 1kHz.
The OP-400 is pin compatible with the LM148, LM348, RM4156,
and HA4741 op amps and can be used to improve systems using
these devices. It is an ideal choice for applications where multiple
op amps are required and precision performance and low power
consumption is critical.
PIN CONNECTIONS

• For devices processed In total compliance to MIL·STD-883. add 1883 after part
number Consult factory for 883 data sheet
t All commercial and Industrial temperature range parts are available With burn-In
For ordering Information see 1986 Data Book. Seellon 2

14·PIN HERMETIC DIP
(V·Suffix)
14·PIN PLASTIC DIP
(P·Suffix)

ELECTRICAL CHARACTERISTICS at Vs = ±15V. TA = 25°C. unless otherwise noted.
OP·400A/E
MIN TYP MAX

Op·400B/F
MIN TYP MAX

UNITS

vas

100

300

",V

InputOOset
Current

los

02

0.5

nA

Input Bias
Current

Ie

3

nA

Input Noise
Voltage DenSity

en

Input Voltage
Range

IVR

PARAMETER

SYMBOL

Input Offset
Voltage

Common-Mode
Relectlon RatiO
Power Supply
Relectlon RatiO
Large-Signal
Vo~ageGaln

CONDITIONS

fo~

12

1kHz

12

nV/JHz

±12

±12

V

CMRR

VCM~

±12V

115

115

dB

PSRR

Vs ~ ±5Vto ±18V

18

18

",VN

Ava

RL ", 10kO
RL "'2kO

5000
2000

2500
1000

V/mV

This advance productlnlormatlon describes a product In development at the time 01 this printing. Final specilications may vary. Please contact
local •• Ie. ollice or distributor lor IInal dat. sheet.
5-239
1/86, Rev. A

II

------I1fMD

OP-400 QUAD LOW-OFFSET, LOW-POWER OPERATIONAL AMPLIFIER -

ADVANCE INFORMATION

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, TA = 25°C, unless otherwise noted. (Continued)
OP-400A/E
PARAMETER

SYMBOL

CONDITIONS

MIN

Output Voltage
Swing

Vo

RL~ 10k!}

±12

Gain Bandwidth
Product

GBW

Supply Current
(All Amplifiers)

ISY

Slew Rate

SR

TYP

OP-400B/F

MAX

MIN

TYP

MAX

±12

UNITS
V

MHz

No Load

2
0.3

5-240

2.8

2.8
0.3

mA
VI",s

1/86, Rev. A

OP-420
QUAD MICROPOWER
OPERATIONAL AMPLIFIER
Precision MOI1olithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•

power single operational amplifier. A Darlington PNP input
stage allows the input common-mode voltage to include V-.
The wide input range combined with low power-supply drain
(-40/-lA/section at 5Vj, provides a unique solution for designs
requiring high functional density and portable operation.
Applications include two-wire transmitters for process
control loops, battery-operated remote-line filters, signal
preconditioning amplifiers, and a variety of multiple-gain
block arrays.

Low Supply Current ........... 200/-IA Max@Vs=+5V
Single-Supply Operation ................ + 5V to + 30V
Dual-Supply Operation ................ ±2.5V to ±15V
Low Input Offset Voltage .................. 500/-IV Typ
Low Input Offset Voltage Drift ............ 5/-1V/oC Typ
High Common-Mode Input Range ... V- to (V + - 1.5V)
High CMRR .............................. 100dB Typ
High Open-Loop Gain ................. 1100V/mV Typ
LM 148 Pinout

For micropower applications requiring offset nulling, see the
OP-20, OP-21 and OP-22 data sheets.

ORDERING INFORMATIONt
PACKAGE
TA = 25°C
Vos MAX
(mV)

HERMETIC
DIP
14-PIN

2.5
2.5
4.0
4.0
6.0

OP420BY'
OP420FY
OP420CY'
OP420GY
OP420HY

PIN CONNECTIONS
LCC

OPERATING
TEMPERATURE
RANGE

OP420CRC/883

MIL
IND
MIL
IND
COM

• For devices processed in total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet.
t All commercial and industnal temperature range parts are available with
burn·m. For ordering information see 1986 Data Book, Section 2.

14-PIN HERMETIC DIP
(Y-Sufflx)

GENERAL DESCRIPTION
The OP-420 quad micropower operational amplifier is a
single-chip quad patterned after the OP-20 precision micro-

OP-420CRC/883
20-LEAD LCC
(RC-Sulflx)

SIMPLIFIED SCHEMATIC (1/4 Shown)

-IN
+IN

0--+---+----1_----1---_..1

5-241

1/86, Rev. A

-----------I1fHD

OP-420 QUAD MICROPOWER OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS (Note 2)
OP-420HY ............................. O°C to +70°C
DICE Junction Temperature (TJ) ••••••• -65°C to +150°C

Supply Voltage ................................... ± 18V
Internal Power Dissipation (Note 1) .............. 500mW
Differential Input Voltage .......................... ±30V
Input Voltage ............................ Supply Voltage
Output Short-Circuit Duration ................ Continuous
(One Amplifier Only)
Storage Temperature Range ........... -65°C to + 150°C
Lead Temperature Range (Soldering. 60 sec) ...... 300°C
Operating Temperature Range
OP-420BY. OP-420CY ............... -55°C to +125°C
OP-420FY. OP-420GY ................ -25°C to +85°C

NOTES:
1 See table for maximum ambient temperature rating and derating factor.

PACKAGE TYPE

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

100·C

10.0mW/·C

14-PIn Hermetic DIP (V)

2. Absolute maximum ratings apply to both packaged parts and DICE. unless
otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs = ±15V. TA = +25°C. unless otherwise noted.
OP-420B
OP-420F
MAX

Vs = ±2.5V to ± 15V

0.5

25

los

Vs= ±2.5V to ±15V

0.5

15

IB

Vs= ±2.5V to ±15V

9

20

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Input Offset Current
(Note 1)
Input Bias Current
(Note 1)

MIN

OP-420C
OP-420G

TYP

PARAMETER

MIN

TYP

OP-420H

MAX

MIN

TYP

MAX

UNITS

4

2

6

mV

0.8

25

1.2

6

nA

12

30

18

40

nA

Input NOise Voltage
Density

fo= 10Hz
fo= 100Hz

50

50

50

Input Noise Current
Density

fO= 10Hz
fo= 100Hz

0.12

0.12

0.12

V+=+5V. V-=OV
Vs =±15V

Input Voltage Aange

IVA

Common-Mode
Aejection Aatio

CMAA

Power Supply
Aejection Aatio

PSAA

Vs = ±2.5V to ± 15V; &
V-= OV. V+= 5V t030V

Large-Signal
Voltage Gain

Avo

AL = 25kO.
Vo= ±1OV

Slew Aate

SA

V+=+5V. V-=OV

Closed-Loop
Bandwidth

BW

ISY

Vs = ±2.5V. No Load
Vs = ±15V. No Load

0/3.5
-15/13.5

v

83

100

80

96

76

90

83

100

80

96

76

90

dB

10
600

AVCL = +1.0
AL = 10kO
V + = 5V. V- = OV.
A L = 10kO
Vs= ±15V.
AL = 25kO

Output Voltage
Swing
Supply Current
(Four Amplifiers)

OV'" VCM '" 3.5V
Vs= ±15V
-15V", VCM '" 13.5V

0/3.5
-15/13.5

0/3.5
-15/13.5

nVlVHZ

20

30

1100

400

50

900

30
200

80

JIoVIV

800

VlmV

0.05

0.05

0.05

VII's

150

150

150

kHz

0.7/4.1

0.8/4.0

0.9/3.8

±14.0

±14.0

±13.8

V

140
330

200
360

170
360

300
460

200
390

400
600

JIoA

NOTE:
1. IBandlosaremeasuredatVCM=O

5-242

1/86, Rev. A

~

OP-420 QUAD MICROPOWER OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs= ±15V, -55°C $ TA$+125° C for OP-420B and
for OP-420F and OP-420G, and 0° C $ TA $ + 70° C for OP-420H, unless otherwise noted.
OP-420B
OP-420F
PARAMETER

SYMBOL

CONDITIONS

Average Input Offset
TCVos
Voltage Drift (Note 1 )

Unnulled

Input Offset Voltage

Vas

Vs = ±2.5V to ± t5V

Input Offset CUrrent
(Note 2)

los

Vs =±2.5Vto±15V

Input Bias Current
(Note 2)

IB

Vs = ±2.5V to ± 15V

IVR

V+=+5V. V-=OV
Vs =±t5V

CMRR

V+ = +5V. V- = OV.
OV,; VCM'; 3.2V
Vs =±15V.
-15V'; VCM $ 13.2V

Input Voltage Range

Common-Mode
Rejection Ratio

Power Supply
Rejection Ratio

PSRR

Vs = ±2.5V to ±15V
and V-=OV. V+=5V
t030V

Large-Signal
Voltage Gain

Ava

Vs = ±15V. RL = 50kO.
Va = ±10V

Output Voltage
Swing

Va

V+=5V.V-=OV.
RL = 20kO
Vs=±15V.
RL = 50kO

Supply Current
(Four Amplifiers)

ISY

Vs = ±2 5V. No Load
Vs = ±15V. No Load

MIN

TYP

OP-420C, -25°C$ TA$+85°C

OP-420C
OP-420G
MAX

MIN

TYP

OP-420H

MAX

to

15

3.5

5.5

MIN

TYP

MAX

UNITS

15

25

J1.V/'C

75

mV
nA

40

30

0/32
-15/13.2

0/3.2
-15/13.2

60

0/32
-15/13.2

nA
V

76

96

73

92

73

86

76

96

73

92

73

86

dB

II
~
~
.......

~

15

300

25

50

200

800

80

100

650

1.0/38

0.9/3.9

40

100

400

J1.VIV

V/mV

11/3.6
V

±138

±13.8
170
390

300
500

±13.6
210
420

400
640

250
500

600
800

J1.A

::sPo.
~

~

0.......

~
~

Po.

0

NOTES:
1. Sample tested.
2. lsand los are measured at VCM = O.

5-243

1/86, Rev. A

----------l~ OP·420 QUAD MICROPOWER OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS
1. OUTPUT 1
2. INVERTING INPUT 1
3. NONINVERTING INPUT 1

4. V+
5. NONINVERTING INPUT 2
6. INVERTING INPUT 2
7. OUTPUT 2
8. OUTPUT 3
9. INVERTING INPUT 3
10. NONINVERTING INPUT 3
11. V12. NONINVERTING INPUT 4
13. INVERTING INPUT 4
14. OUTPUT 4
For addilional DICE informalion refer 10
1986 Dala Book, Section 2.

DIE SIZE 0.092 X 0.086 Inch, 7912 sq. mils
(2.34 X 2.18 mm, 5.10 sq. mm)

WAFER TEST LIMITS at VS= ±15V,

TA

= 25°C, unless otherwise noted.
OP·420N

OP·420G

OP·420GR

LIMIT

LIMIT

LIMIT

UNITS

Vs = ±2 5V to ±15V

25

4

6

mVMAX

los

Vs = ±2.5V to ±15V, (Note 1)

1.5

25

18

Vs = ±2.5V to ±15V, (Note 1)

20

30

40

nAMAX

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Input Offset Current
Input Bias Current
Input Voltage Range

IVR

Common·Mode Rejection Ratio

CMRR

V+=+5V, V-=OV
OV 5 VCM 5 3.5V
Vs = ±15V, -15V 5 VCM 513 5V

Power Supply Rejection Ratio

PSRR

Vs = ±2.5V to ±15V
V- = OV, V+ = +5V to +30V

Large·Signal Voltage Gain

Avo

RL = 25kll, Vo =±10V

Vo

V+=+5V, V-=OV
RL = 10kll
Vs= ±15V
RL = 25kll

ISY

No Load, (Four Amplifiers)

Output Voltage Swing

Supply Current

nAMAX

-15/13.5

-15/135

-15/13.5

V MIN

83
83

80
80

76
76

dBMIN

30

50

80

",V/v MAX

600

400

200

VlmV MIN

0.7/4.1

08/4.0

09/3.8

VMAX

±14.0

±14.0

±13.8

VMIN

360

460

600

",A MAX

NOTES:
1. 18 and lOS are measured at VCM = O.
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging IS not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

5·244

1/86, Rev. A

------------1~ OP-420 QUAD MICROPOWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
CLOSED-LOOP GAIN
vs FREQUENCY

INPUT OFFSET CURRENT
vs TEMPERATURE

120

100
600

;

IIII IIIII
AVCL = 100

~40

c:: 400

il

E

300

r-.

AVCL = 1

o

111111

vV

;;
100

I

15

15

150

15k

15k

150k

-100

-50

50

100

150

-100

TEMPERATURE (OC)

FREQUENCY (Hz)

MAXIMUM OUTPUT VOLTAGE
vs LOAD RESISTANCE
17

r----i'-

L

~ 200

r-.

i

-

~S' ;,5)

/

>-

IIII I I

-20
015

Ii

~

I~~~~ .I,~I

20

,........
500

~

AVCL = 1000

Z

12

<

TA = 2SOC
80

~60

14

v~. ,,~v

1111111
Vs = ±15V
RL '" 25kD

100

INPUT BIAS CURRENT
vs TEMPERATURE

-so

50

100

150

TEMPERATURE (CC)

INPUT CURRENT NOISE
DENSITY (in) vs FREQUENCY

INPUT VOLTAGE NOISE
DENSITY (en) vs FREQUENCY
10

1 1.1

JJ,I,H

TA = 2!fc

~
~
~

iii , 00

~
o

"-

w

~

V

is
z

>z

V

/

1k

il
>i<

II III

V

II III

/'
10k

100k

LOAD RESISTANCE tn)

t--

~ 010

VS =+5,OV

;;

10 L-LUJ.lill'--LUJ..UJJL..LUJ.lillL..LUJ.lillJ
o,
10
100
1000
FREQUENCY (Hz)

SMALL-SIGNAL TRANSIENT RESPONSE

001
01

10

10

100

1000

FREQUENCY (Hz)

LARGE-SIGNAL TRANSIENT RESPONSE

OUTPUT

5-245

1/86, Rev. A

OP-421
QUAD LOW-POWER
OPERATIONAL AMPLIFIER
PreciS10n MOl1olithics Inc.

FEATURES

GENERAL DESCRIPTION

•
•
•
•
•
•
•
•
•
•

The OP-421 quad low-power operational amplifier is a singlechip quad patterned after the OP-21 single operational
amplifier. The PNP input stage allows the input commonmode vOltage to include V-. Featuring a low power-supply
current (150/,A/section typical at 5V), the OP-421 offers a
unique solution for designs requiring a combination of high
function density, wide bandwidth, and low-power operation.
Applications for the OP-421 include low-power active filters,
battery-operated remote line filters, and signal preconditioning amplifiers. In addition, the ever-present problem of
crossover distortion in low-power devices is eliminated by a
unique double-buffered output section.

Low Supply Current .. ...................... 1mA Max
Slew Rate .............................. O.25V//,s Min
Single Supply Operation ................ +5V to +30V
Low Input Offset Voltage .................. 500/,V Typ
Low Input Offset Voltage Drift .......... 10/,V/oC Max
High Common-Mode Input Range ... V- to V+ (-1.5V)
High CMRR .............................. 100dB Typ
High Open-Loop Gain .. ................ 400V/mV Typ
Single-Chip Monolithic Construction
Pin Compatible With LM124, LM148, and OP-11

ORDERING INFORMATIONt
TA = 25°C
VosMAX
(mV)

HERMETIC
DIP
14-PIN

OPERATING
TEMPERATURE
RANGE

2.5
2.5
4
4

OP421 BY'
OP421FY
OP421CY'
OP421GY
OP421HY

MIL
IND
MIL
IND
COM

6

PIN CONNECTIONS

• For devices processed In total compliance to MIL'STD-883, add /883 after
part number Consult factory for 883 data sheet.
tAil commercial and Industrial temperature range parts are available with
burn-In For ordering mformatlon see 1986 Data Book, Section 2

14-PIN HERMETIC DIP
(V-Suffix)

SIMPLIFIED SCHEMATIC (1/4 Shown)

v'

-IN
"N

O---+----+-----+--------1f---'

5-246

1/86, Rev. A

-~

---

IfM!) OP-421 QUAD LOW-POWER OPERATIONAL AMPLIFIER
ABSOLUTE MAXIMUM RATINGS

(Note 2)

Supply Voltage •••••••••••••••••••••••.••••••••••• ± 18V

DICE Junction Temperature (TJ)

.......

-6S·C to +1S0·C

Internal Power Dissipation (Note 1) •••••••••••••• SOOmW
Differential Input Voltage •••••••••••••••••••••••••• ±30V

NOTES:
1. See tabJe for maximum ambient temperature rating and derating factor

Input Voltage •••••••••••••••••••••••••••• Supply Voltage
Output Short-Circuit Duration •••••.••••••••• Continuous
(One Amplifi!!r Only)
-6S·C to +1S0·C

Storage Temperature Range •••••••••••

PACKAGE TYPE

Lead Temperature Range (Soldering, 60 sec) •••••• 300·C
-SS· C to +12S·C
OP-421 FY, OP-421 GY •••••••••••••••• -2S·Cto+8S·C
OP-421HY • • • • • • • • • • • • • • • • • • • • • • • • • • • •• O· C to + 70· C

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Input Offset Current
Input Bias Current
Input NOise Voltage
Density

l00'C

10.0mW/'C

2. Absolute maximum ratings apply to both packagad parts and DICE. unless
otherWise noted

±1SV, TA = +2S·C, unless otherwise noted.

OP-421B
OP-421F
PARAMETER

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

14-Pln Hermetic DIP (V)

Operating Temperature Range
OP-421 BY, OP-421 CY •••••••••••••••

ELECTRICAL CHARACTERISTICS at Vs =

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

MIN

OP-421C
OP-421G
MAX

Vs =±2.5Vto±15V

0.5

2.5

los

Vs = ±2.5V to ± 15V

0.6

50

20

10

Ie

Vs= ±2 5V to ±15V

20

50

50

80

en

fo= 10Hz (Note 1)
fo= 100Hz (Note 1)

20
15

40
30

20
15

40
30

Input Noise Current
Density

in

fO= 10Hz (Note 1)
fo= 100Hz (Note 1)

0.3
02

0.6
0.4

0.3
02

06
0.4

Input Voltage Range

IVR

V+=+5V. V-=OV
Vs =±15V

Common-Mode
Rejection Ratio

CMRR

V+=+5V. V-=OV.
OVSVcM S+3.5V
Vs=±15V.
-15VSV CM S+13.5V

Power Supply
Rejection Ratio

PSRR

Vs = ±2.5V to ±15V; &
V-=OV. V+= 5Vto30V

Large-Signal
Voltage Gai n

Avo

Vo=±10V
RL = 10kO

Output Voltage
Swing

Vo

V+ = 5V. V-= OV
RL = 5kO
Vs=±15V.
RL = 10kO

Closed-Loop
Bandwidth (Note 2)

BW

AveL =+1.0.
RL = 10kO

Supply Current
(Four Amplifiers)

ISY

Vs = ±2.5V. No Load
Vs = ± 15V. No Load

0/3.5
-15/13.5

MIN

TVP

OP-421H

TYP

MAX

MIN

4

0/3.5
-15/13.5

MAX

UNITS

......

2

6

mV

50

20

nA

:::l
~

100

150

nA

20
15

40
30

nV/..jHZ

0.3
0.2

0.6
0.4

pA/..jHZ

0/3.5
-15/13.5

V

83

100

80

96

76

90

83

100

80

96

76

90

200

30

20
100

400

50

200

30
100

07/4.0

0.8/39

0.9/38

±14

±139

±138

80

,.V/v
VlmV

200

V

10

1.9
0.6
1.2

1.0
10
18

1.9
0.7
14

Slew Rate

SR

(Note 1)

0.25

05

025

Channel Separation

CS

(Note 1)

100

120

100

1.0
1.5
2.3

1.9
0.9
18

MHz
2.0
30

mA

05

0.25

05

VI,.s

120

100

120

dB

NOTES:
1. Sample tested.
2. Guaranteed by design.

5·247

-

.

__ . - - - - - - - . _

..

_ - - - - - - - _ . - ..

---

~

~

~

0....,

~
~
~

dB

10

ffi

TYP

1/86, Rev. A

0

IEHD

OP-421 QUAD LOW-POWER OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, -55 0 C:o; TA:O; + 125 0 C for OP-421 Band OP-421 C, -25 0 C:O; TA:O; +85 0
for OP-421 F and OP-421 G, and 00 C:O; TA:O; + 70 0 C for OP-421 H, unless otherwise noted.
OP-421B
OP-421F
PARAMETER

SYMBOL

MIN

CONDITIONS

TYP

OP-421C
OP-421G
MAX

Average Input Offset
TCVos
Voltage Drift (Note 1 )
Input Offset Voltage

Vos

Input Offset Current

los

Input Bias Current

18

I nput Voltage Range

IVR

Common-Made
Rejection Ratio

CMRR

Power Supply
Rejection RatiO

PSRR

Large-Signal
Voltage Gain

Avo

Output Voltage
Swing

Vo

V+=+5V. V-=OV
Vs = ±15V

1.6
70

25

MAX
15

1.8

5.5

3.0
60

= +5V, V- = OV.

OV" VCM " +3 2V
Vs =±15V.
-15V"VCM ,,+13.2V

RL = 10kO

UNITS

10

15

p.VI'C

3

7.5

mV

15

6.0

30

nA

125

140

230

nA

0/3.2
-15/13.2

V

96

74

94

73

86

78

96

74

94

73

86

dB

15
15
100

RL = 20kO

TYP

78

Vs= ±2 5V to ±15V; &
V-=OV. V+=5Vt030V
Vo= 10V

MIN

MAX

0/3.2
-15/13.2

0/3.2
-15/13.2

V+=5V. V-=OV

50
50

25
25
50

200

80
80

100

40
40
50

0.8/3.9

09/3.8

1.0/3.7

±13.8

±137

±13.7

100
100

100

p.VIV
VlmV

V

Vs= ±15V.
RL = 20kO

ISY

3.5

OP-421H

TYP

10

= ±2.5V to ±15V
Vs = ±2.5V to ± 15V
Vs = ±2.5V to ± 15V
Vs

V+

Supply Current
(Four Amplifiers)

MIN

C

Vs = ±2.5V, No Load
Vs = ± 15V, No Load

12
2.0

0.68

1.5
2.5

0.68

1.5
2.5

2.0
3.2

0.68

2.0
3.2

3.0
4.0

mA

NOTE:
1 Sample tested.

BURN-IN CIRCUIT
-18

v

13

12

11

10

v-

v+

GND

+18 V

5-248

1/86, Rev. A

-----------I~ OP-421 QUAD LOW-POWER OPERATIONAL AMPLIFIER

DICE CHARACTERISTICS

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
DIE SIZE 0.086 X 0.092 Inch, 7912 sq. mils
(2.34 X 2.18 mm, 5.10 sq. mm)

OUTPUT 1
INVERTING INPUT 1
NON INVERTING INPUT 1
V+
NON INVERTING INPUT 2
INVERTING INPUT 2
OUTPUT 2
OUTPUT 3
INVERTING INPUT 3
NONINVERTING INPUT 3

II

VNON INVERTING INPUT 4
INVERTING INPUT 4
OUTPUT 4

i:2

For addilional DICE informalion refer to
1986 Data Book, Section 2.

~
......
~

WAFER TEST LIMITS at Vs= ±15V,

TA =

25°C, unless otherwise noted.
OP-421N
LIMIT

OP-421G
LIMIT

OP-421GR
LIMIT

UNITS

2.5

4

6

mV MAX

10

20

nAMAX

50

SO

150

nAMAX

-15113.5

-15113.5

-15113.5

VMIN

CMAA

V+=+5V, V-=OV
OV:5 VCM :5 +3.5V
Vs =±15V,
-15V:5VCM :5+13.5V

S3

SO

76

dBMIN

PSAA

Vs= ±2.5V to ±15V; and
V- = OV, V+ = +5V to 30V

30

50

SO

!'VIV MAX

Ava

Vo =±10V
AL = 20kO

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Vs = ±2.5V to ± 15V

Input Offset Current

los

Vs = ±2.5V to ±15V

Input Bias Current

Ie

Vs = ±2.5V to ±15V

Input Voltage Aange

IVA

Common-Mode
Aeiection Aatio
Power Supply
Rejection Ratio

Large-Signal
Voltage Gain

::sp.,
~

Output Voltage Swing

Va

V+ = +5V, V- = OV, AL = 5kO
Vs = ± 15V, AL = 10kO

Supply Current
(Four Amplifiers)

ISY

Vs = ± 2.5V, No Load
Vs = ± 15V, No Load

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V,

Input Noise Voltage
Density
Closed-Loop
Bandwidth

SYMBOL

CONDITIONS

en

fo= 10Hz
fo= 100Hz

BW

AVCL =+1.0
AL = 10kO

~
~

p.,

0

200

200

100

VlmV MIN

0.714.0
±14

0.SI3.9
±13.9

0.913.S
±13.S

VMIN

1.0
I.S

1.5
2.3

2.0
3.0

mAMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

PARAMETER

~

0
......

TA =

+25° C, unless otherwise noted.

OP-421N
TYPICAL

OP-421G
TYPICAL

OP-421GR
TYPICAL

20
15

20
15

20
15

nVIJHZ

19

1.9

1.9

MHz

0.5

0.5

Vi!,s

120

120

dB

Slew Aate

SA

0.5

Channel Separation

CS

120

5-249

UNITS

1/86, Rev. A

-----------I1fMD

OP·421 QUAD LOW·POWER OPERATIONAL AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
OPEN·LOOP
FREQUENCY RESPONSE

-..,

'"

60

""

RL = 2kn

10

'\
100

'\

20

1\
10

1 - - - - + - - - - i -"R'L-=.:-;2"'Ok:c",---!

vl.~ lL

15

40

10

120

20

100

lk
10k
lOOk
FREQUENCY (Hz)

80 L -_ _ _

1M

10M

15

POSITIVE

Wj1llllc

SWING

___

___

10k

1M

lOOk

FREQUENCY (Hz)

SUPPLY CURRENT vs
SUPPLY VOLTAGE

GAIN BANDWIDTH vs
SUPPLY VOLTAGE
26

1.]l.I,VS-',5V
NEGATIJ~II

IV

1\

i'-lk

30

10
20
TOTAL SUPPLY VOLTAGE (VOLTS)

OUTPUT SWING vs
OUTPUT LOAD

\

1----+----\-----1

~

80

r----..,-----,-------,

~

100

140

~

120

OUTPUT SWING
vs FREQUENCY

OPEN·LOOP GAIN
vs POWER SUPPLY VOLTAGE

24

SWING

10

22

I

20

V

.

,:

o
10

100

lk

10k

_55°C

m~r

-

r-

18

I-:"

16

'"

~

V

-

14

lOOk

1M

10

LOAD RESISTANCE 1m

20

30

±5

POWER SUPPLY REJECTION
RATIO vs FREQUENCY

±75

±1O

±125

SUPPL Y VOLTAGE (VOLTS)

SUPPLY VOLTAGE (VOLTS)

COMMON·MODE REJECTION
RATIO vs FREQUENCY
120

100

it'
80

11'\

60

40
40

H-tttttlt-+Httltlt-t-H-tttffiH-ttfI11lk:-H-tHtIH
20

I
10

100

lk

10k

lOOk

o lk

lk

10k

lOOk

1M

FREQUENCY (Hz)

FREQUENCY (Hd

5·250

1/86, Rev. A

-----------l{fMD OP-421 QUAD LOW-POWER OPERATIONAL AMPLIFIER
TYPICAL PERFORMANCE CHARACTERISTICS

VOLTAGE FOLLOWER
LARGE-SIGNAL RESPONSE

VOLTAGE FOLLOWER
SMALL-SIGNAL RESPONSE
OUTPUT
OUTPUT

5OmV/DIV

, VOLT/OIV

2~s/DIV

5,us/OIV

INPUT

INPUT

Vs = av, +15V

CL = 50pt

Vs" av, +15V
RL'" lOkn
TA = 25°C

CL = 50pf

RL = 10kfl.

TA '" 2soC

NOISE CHARACTERISTICS

INPUT NOISE VOLTAGE
DENSITY vs FREQUENCY

INPUT NOISE CURRENT
DENSITY vs FREQUENCY

1000

10

@
-"'

.~100 r-...
>

~c
:Ji
0
Z

w

10

"!:;
~

"-

0

'"

~

1

'"

r---..

!$
00 1
01

10
01

10

100

lk

10k

10

100

lk

10k

FREQUENCY (Hz)

FREQUENCY (Hz)

1/86, Rev. A

5-251

~~~~---~

---~-

~

~-

~

~~-

PM.108A/PM.2108A
LOW·INPUT·CURRENT OPERATIONAL AMPLIFIERS
PM·I08A/PM·208A/PM·308A/PM·I08/pM·208/pM·308/pM·2108A/PM·2108
PrecIsion Mo]]o]ithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

directly interchangeable with industry-standard types,
Precision Monolithics' advanced processing provides the
PM-108A series with a significant improvement in input noise
voltage. Low supply current drain over a wide power-supply
range makes the PM-108A attractive in battery operated and
other low-power applications. The low bias current provides
excellent performance with piezoelectric and capacitive
transducers and in such high-impedance circuits as longperiod integrators and sample-and-holds. For improved
performance see OP-08, OP-12, OP-20, OP-21 , and OP-22.

Low Offset Current ...................... 200pA Max
Low Bias Current ........................... 2nA Max
Low Power Consumption •......... 18mW Max @ ± 15V
Wide Supply Range ..................... ± 3V to ± 20V
High Power-Supply Rejection Ratio ......... 96dB Min
Low Offset Voltage Drift ................. 5p.V/oC Max
High Common-Mode Input Range .•...... ± 13.5V Min
High Common-Mode Rejection Ratio . . . . . . .. 98dB Min
MIL-STD-883 Processing Models Available
Silicon-Nitride Paulvatlon

The PM-2108A contains two superbeta, PM-108A op amps
in a single 16-pin DIP. Compared to the Single PM-108A
types, this model offers higher packaging density, closer
thermal tracking between the two amplifiers, and reduced
insertion cost.

GENERAL DESCRIPTION
The PM-108A series of precision operational amplifiers
feature very low input offset and bias currents. Although
ORDERING INFORMATIONt
PACKAGE
HERMETIC
TA = 25°C
VosMAX
(mV)

TO·99
a·PIN

a·PIN

16-PIN

0.5
0.5
05
2.0
2.0
7.5

PM108AJ'
PM208AJ
PM308AJ
PM108J'
PM208J
PM308J

PM108AZ'
PM208AZ
PM308AZ
PM108Z'
PM208Z
PM308Z

PM2108AO'

PLASTIC
DIP
a·PIN

DIP

OPERATING
TEMPERATURE
RANGE

LCC
PM108ARC/883

MIL
INO
COM
MIL
INO
COM

PM308AP
PM21080'

'For devices processed In total compliance to MIL·STD·883, add 1883 after
part number. Consult factory for 883 data sheet

tAli commercial and industrial temperature range parts are available with
burn·,n For ordering information see 1986 Data Book. Section 2.

SIMPLIFIED SCHEMATIC (Pin numbers for PM-108 only. Circuit Is 1/2 2108.)
111

caMP

181

COMP

r------_--1r--+---+-_--.......--------_---o V+
171

R5

L..--I--"'--0161

OUTPUT

131
+

R6

121

~------~-~~--------~--_4-_4-

5-252

__

~--__o~1

V-

1/86, Rev. A

_ _ _ _ _ _ _--I~PM-108A1PM-2108A LOW-INPUT-CURRENT OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS
Supply Voltage
PM-10BA, PM-1OS, PM-20BA, PM-20S,

NOTES:
1. Maximum package power diSSipation vs. ambient temperature

PM-210SA, PM-2108, PM-10SARC •..••••....••.•• ±20V
PM-30BA, PM-30S .............................. ± 1SV
Internal Power Dissipation (Note 1) ••.•••.•...••• 500mW
Differential Input Current (Note 2) ••••.•••••••.•. ±10mA
Input Voltage (Note 3) ............................ ±15V
Output Short-Circuit Duration ••••••..•••....•• Indefinite
Operating Temperature Range
PM-10BA, PM-1OS, PM-210SA,
PM-210S, PM-10SARC •••••••••.•••••• -55·C to +125·C
PM-20BA, PM-20S .................... -25·C to +S5·C
PM-30BA, PM-30S ...................... O·C to +70·C
Storage Temperature Range
(C-, J-, Z- or ARC-Package) •••.•••••. -65·C to +150·C
(P-Package) ......................... -65·C to +125·C
Lead Temperature Range (Soldering, 60 sec) ••.••. 300·C

ELECTRICAL CHARACTERISTICS at ±5V S

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (JI

80'C

7.1mW/'C

PlastIc 8-Pin DIP (PI

36'C

5.6mW/'C

HermetIc 8-PIn DIP (ZI

75'C

6.7mW/'C

HermetIc 16-Pln DIP (QI

100'C

10.0mW/'C

80'C

7.8mW/'C

PACKAGE TYPE

LCC (RCI
2

The Inputs are shunted with back-to-back dIodes for overvoltage protectIon Therefore, If a dIfferentIal input voltage In excess of tV is applied
between the Inputs, excessive current wIll flow, unless some limiting
resistance IS provided.
3. Forsupplyvoltages less than ±15V, the absolute maximum input voltage IS
equal to the supply voltage.

VsS ±20V and TA = 25·C, unless otherwise noted.

PM-108A1PM-2108A
PM-208A
PARAMETER

SYMBOL

Input OIIset Voltage

MAX

TYP

MAX

UNITS

Vos

0.3

0.5

0.7

2.0

mV

Input Offset Current

los

0.05

0.2

0.05

0.2

nA

Input Bias Current

Ie

0.8

2.0

0.8

2.0

nA

Input Resistance

RIN

(Note II

30

70

30

70

MO

Avo

Vs =±15V, VouT =±10V,
RL~ 10kO

80

300

50

300

V/mV

ISY

10UT=O, VOUT=O,
Each AmplIfIer

Supply Current

MIN

PM-108/PM-2108
PM-208

TYP

Large-Signal Voltage GaIn

CONDITIONS

03

MIN

06

0.3

0.6

mA

ELECTRICAL CHARACTERISTICS at ±5V S VsS±20V, -55· C S TAS + 125· C for PM-10SA, PM-10S, PM-210SA and PM-210S,
-25·C S TAS +S5·C for PM-20SA, PM-20S, unless otherwise noted.
PM-108A1PM-2108A
PM-208A
PARAMETER

SYMBOL

Input Offset Voltage

Vos

Average Input Offset
Voltage Drrft

TCVos

Input Offset Current

los

Average Input Offset
Current Drrft

TClos

Input Bias Current

Ie

CONDITIONS

MIN

TYP

MAX

TYP

MAX

04

1.0

10

3.0

mV

5

3

15

p.VI'C

0.1

04

01

04

nA

0.5

25

05

2.5

pA/'C

3

nA

(Note21

(Note 21

Large-Signal Voltage Gain

Avo

Vs =±15V. VouT =±10V.
RL~ 10kO

Output Voltage SWIng

Vo

Vs = ± 15V, RL = 10kO

Input Voltage Range

IVR

Vs =±15V

Common-Mode
RejectIon Rallo

CMRR

Vs= ±15V, VCM = ±13 5V

Power-Supply
RejectIon RatIO

PSRR

Vs = ±5V to ±20V

Supply Current

ISY

VOUT=O. TA = MAX,
Each AmplifIer

40

200

±13

±14

±135
96

NOTES:
1 Guaranteed by mput bias current.

2

5-253

PM-108/PM-2108
PM-208
MIN

25

200

VlmV

±13

±14

V

±135
110

UNITS

85

V
100

dB

3

15

15

100

p.VN

015

04

015

04

mA

Sample tested

1/86, Rev. A

_ _ _ _ _ _ _---l~PM-108A1PM-2108A LOW-INPUT-CURRENT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at ±5V:5 VS:5 ±2DV and TA = 25'C, unless otherwise noted.
PM-308A
PARAMETER

SYMBOL

Inpul Offset Voltage
Input Offset Current

MIN

PM-308

TYP

MAX

TYP

MAX

UNITS

Vas

0.3

0.5

2.0

7.5

mV

lOS

02

1.0

0.2

1.0

nA

1.5

70

1.5

7.0

nA

CONDITIONS

Input Bias Current

MIN

Input Resistance

(Note 1)

10

40

10

40

MO

Large-SIgnal Voltage Gain

Ava

Vs =±15V, VouT =±10V,
R L ", 10kO

80

300

25

300

VlmV

ISY

10UT=O, VOUT=O,
Each Amplifier

Supply Current

0.8

0.3

0.8

mA

TYP

MAX

UNITS

3.0

10.0

mV

6

30

p.V/'C

0.3

1.5

nA

10

pA/'C

10

nA

0.3

ELECTRICAL CHARACTERISTICS at ±5V:5 Vs:5 ±15V and D'C:5 TA:5 HD'C, unless otherwise noted.
PM-308A
PARAMETER

SYMBOL

Input Offset Voltage

Vas

Average Input Offset
Voltage Drift

TCVOS

Input Offset Current

los

Average Input Offset
Current Drift

TCIOS

Input Bias Current

18

MIN

CONDITIONS

PM-308

TYP

MAX

0.4

0.73

MIN

(Note 1)

0.3

1.5

(Note 1)

10

2

10

Large-Signal Voltage Gain

Ava

VS =±15V, VouT =±10V,
R L ", 10kO

Output Voltage Swing

Va

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

Power-Supply
Rejection Ratio

PSRR

Vs = ±5V to ±15V

Supply Current

ISY

VOUT=O, TA = MAX,
Each AmplIfier

60

200

15

100

V/mV

Vs = ±15V, RL = 10kO

±13

±14

±13

±14

V

Vs= ±15V

±14

96

±13

80

110

3

15

V
100

15

0.23

dB

100

0.23

p.VIV
mA

NOTE:
1.

Guaranteed by input bias current.

PIN CONNECTIONS
PM-108ARC/883

COMP
B

COMPe.
7v+
-IN 2

...

+IN 3

EPOXY MINI-DIP
(P-Sufflx)
AND
8-PIN HERMETIC DIP
(Z-Sufflx)

6 OUT

5 NC

4

v- (CASE)

16-PIN HERMETIC DIP
(O-Suffix)

TO-99
(J-Sufflx)

5-254

LCC PACKAGE
(RC-Suffix)

1/86, Rev_ A

----------1lmD

PM-108A1PM-2108A LOW-INPUT-CURRENT OPERATIONAL AMPLIFIERS

APPLICATIONS INFORMATION
The PM-10BA series has very low input offset and bias
currents; the user is cautioned that printed circuit board
leakages can produce significant errors, especially at high
board temperatures. Careful attention to board layout and
cleaning procedure is required to achieve the PM-10BA's

rated performance. It is suggested that board leakage be
minimized by encircling the input pins with a guard ring
maintained at a potential close to that of the inputs. The
guard ring should be driven by a low impedance source such
as an amplifier's output or ground.

COMPENSATION CIRCUITS
STANDARD

ALTERNATE

R2
R2

R1
R1

R3

R3

~"---+---o OUTPUT
R1
Cf;;;'R1+R2 Co

>-=-----{) OUTPUT
Cs = 100pF

Co = 30pF

(IMPROVES REJECTION OF POWER SUPPLY NOISE BY A FACTOR OF TEN)

5-255

1/86, Rev. A

II

I-~-146/I-~-246
PROGRAMMABLE QUAD
OPERATIONAL AMPLIFIER
Precision Monolithics Inc.

ADVANCE PRODUCT INFORMATION
CROSS REFERENCE

FEATURES
• Wide Supply Range ................ ± 1.2V to ± 22V
• Programmable Operation
Supply Current (per Amplifier) ..... 3.511A to 1.75mA
Gain-Bandwidth Product ........... 10kHz to 4MHz
• High Gain ............................. 100dB Min'
• Low Noise ........................ 2SnVlJAZTyp'
• High Power-Supply Rejection ............ SOdB Min'
• Excellent Channel Separation ............ 120dB Typ'
• At ISET

PMI

NATIONAL

EXAR

PM1460
PM2460

LM146J
LM246J

XR-146M
XR-246N

PIN CONNECTIONS

= to",A.

ORDERING INFORMATIONt
TA = 25°C
VosMAX
(mV)"

CERAMIC
DIP
PACKAGE

OPERATING
TEMPERATURE
RANGE

5

PM1460'
PM2460

MIL
IND

6

16-PIN CERAMIC DIP
(Q-Suffix)
l6-PIN EPOXY DIP
(P-Suffix)

• For devices processed in total comphance to MIL-STD-883, add 1883 after part
number. Consult factory for 883 data sheet.
.. At ISET = 10",A.
t All commercial and Industrial temperature range parts are available With burn-m.
For ordering Information see 1986 Data Book, Section 2.

SIMPLIFIED SCHEMATIC

r-------------------~~--------~-----------ov+

~--_+----+_--~_oOUT

r------------,
I

(EXTERNAL)

v+ v+

I
I

I

I
I

I
I

II

1

RSET

RSEl

ISET
L____
_

I
I
t

TO OTHER
OPAMPS

-.

I

ISET
____
.JI

v-

This advance product information describes a product in development at the time of this printing. Final specifications may vary_ Please contact
local sales ollice or distributor for final data sheet.
5-256

1/86, Rev. A

-----l~ PM-146/PM-246 PROGRAMMABLE QUAD OPERATIONAL AMPLIFIER -

ADVANCE INFORMATION

GENERAL DESCRIPTION
The PM-146 monolithic quad operational amplifier incorporates
four independent programmable low-power amplifiers. Using two
external resistors, the supply currents may be programmed for
the optimal combination of bias and offset currents, gainbandwidth product, slew rate, input noise, and output currents.
By means of this capability, the user can achieve the required
level of speed and performance while maintaining the minimum
possible power consumption.

The PM-146 operates with a supply voltage down to ±1.5V,
which makes the amplifier well-suited to applications utilizing
battery or solar-cell supplies. Since speed is controlled by the
programming current, the slew rate and gain-bandwidth of the
PM-146 remain relatively unchanged with varying supply
voltages. The extreme versatility afforded by the PM-146's
programmability allows the device to be tailored to a wide variety
of applications, including active filters, oscillators, and generalpurpose amplifiers.

Gain-bandwidth products of 2MHz with slew rates of 1V/",s, at a
supply current drain of 1mA per amplifier, are achieved using a
programming ("set") current of 30",A. At the other end of the
spectrum, bias currents of 1nA at a supply drain of only 3.5",A per
amplifier are obtained with a set current of 0.1 ",A for applications
where speed is not important.

The pin configuration of the PM-146 matches the OP-11 pin-out,
with the addition of the two programming pins on one end. These
pins are internally protected to survive a momentary short to the
power supply rails.

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, ISET = 10",A, TA = 25°C, unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Rs s 50D
(Note 1)

Input Offset Current

los

(Note1)

Input Bias Current

18

(Note 1)

Supply Current
(4 Op Amps)

ISY

Open Loop
Voltage Gain

Ava

RL ~ 10kD,
dVOUT = ±10V

PM-146Q
MIN
TYP MAX

PM-246Q
MIN
TYP MAX

05

5

05

6

mV

20

2

100

nA

50

100

50

250

nA

14

20

14

25

mA

UNITS

~

100

1000

50

1000

V/mV

±135

±14

±135

±14

V

Input Voltage Range

IVR

(Note 2)

Common· Mode
Rejection

CMR

VCM ~ ",13 5V,
Rss 10kD

80

100

70

100

dB

PSR

Rss 10kD

80

100

74

100

dB

Va

RL", 10kD

±12

±14

±12

±14

5

20

08

12

Power· Supply
Rejection

Output Voltage SWing
Short-CirCUit
Output Current

Ise

Gain-Bandwidth
Product

GBW

Phase Margin

<1>0

Slew Rate

SR

Input NOise

1kHz

en

f~

Channel Separation

CS

RL ~ 10kD,
t.vOUT ~ OVto "'12V

Input Resistance

R'N
C'N

Voltage Density

Input Capacitance

~
~
.......
.....
::l

35

20
05

V
35

mA

12

MHz

60
04

60

Degrees

04

V/fLS

28

28

nV/VHZ

120

120

dB

10

10

MD

20

20

pF

NOTES:
1. VeM ~ OV.
2 Guaranteed by CMR test

5-257

1/86, Rev. A

~

~

0.......

~

~
~

0

-----I~ PM-146/PM-246 PROGRAMMABLE QUAD OPERATIONAL AMPLIFIER -

ELECTRICAL CHARACTERISTICS at Vs = ±15V,

ISET

= 10/-LA, -55°c:5 TA :5

ADVANCE INFORMATION

+ 125°C for PM-146Q, -25°c:5 TA :5 +85°c for

PM-246Q, unless otherwise noted.

PM-146Q
PARAMETER
Input Offset Voltage

SYMBOL

CONDITIONS

Vas

Rs$50n
(Note 1)

Input Offset Current

los

(Notel)

Input Bias Current

18

(Note 1)

Supply Current
(40pAmps)

Isv

Ava

RL = 10kn,
ilVOUT = ±10V

IVR

(Note2)

Open Loop
Voltage Gain
Input Voltage Range
Common-Mode
Rejection
Power-Supply
Rejection
Output Voltage Swing

MIN

PM-246Q

TYP

MAX

05

6
25

50

100

15

20

1.5

MIN

TYP

MAX

UNITS

05

7.5

mV

2

100

nA

50

250

nA

25

mA

50

1000

25

1000

V/mV

±13.5

±14

±13.5

±14

V

CMR

Rs$50n
VCM = ±13.5V

70

100

70

100

dB

PSR

Rs$50n

76

100

74

100

dB

Va

RL ", lOkn

±12

±14

±12

±14

V

NOTES:
1. VOM = OV.
2 Guaranteed by CMR test.

ELECTRICAL CHARACTERISTICS at Vs

= ± 15V, ISET = 1 /-LA, TA = 25°C, unless otherwise noted.

PM-146Q
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

PM-246Q

TYP

MAX

Vas

Rs$50n
(Notel)

0.5

(Note 1)

Input Bias Current

18

Supply Current
(4 Op Amps)

Isv

Gain-Bandwidth
Product

GBW

MIN

80

MIN

TYP

MAX

UNITS

5

0.5

7

mV

75

20

75

100

nA

140

250

140

300

fJ.A

100

50

100

kHz

NOTE:
1. VOM

= OV

ELECTRICAL CHARACTERISTICS at Vs

= ±1.5V, ISET = 10/-LA, TA = 25°C, unless otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Rs$50n
(Note 1)

Input Voltage Range

PM-146Q

IVR

(Note 2)

Common-Mode
Rejection

CMR

RS $ son
VOM = ±0.7V

Output Voltage Swing

Va

RL ", 10kn

MIN

PM-246Q

TYP

MAX

0.5

5

±O 7

MIN

0.5
±0.7

80
±0.6

TYP

UNITS
mV
V

80
±O 6

MAX

dB
V

NOTES:
1. VOM = OV.
2 Guaranteed by CMR test.

5-258

1/86, Rev. A

]E»~-155Jl/]E»~-156Jl/]E»~-157Jl
MONOLITHIC JFET- INPUT
OPERATIONAL AMPLIFIERS
Precision Monolithics Inc.

LOW SUPPLY CURRENT - PM-155A/PM-355A/pM-155
GENERAL PURPOSE - PM-156A/PM-356AjPM-156
WIDE-BANDWIDTH - PM-157A/pM-357A/PM-157

FEATURES

GENERAL DESCRIPTION

All Devices
• Low Input Bias and Offset Currents
• Low Input Offset Voltage ...................... 1.0mV
• Low Input Offset Voltage Drift .............. 3.0/lV/ o C
• Low Input Noise Current ............... O.01pAly'"HZ
• High Common-Mode Rejection Ratio ........... 100dB

The PM JFET-input series provides low input current, high
slew rate, and direct interchangeability with LF155, 156, and
157 types. These operational amplifiers use a new process
which allows fabrication of matched JFET transistors and
standard bipolar transistors on the same chip. High accuracy
and low cost make the PM JFET-input series useful in new
designs and as replacements for modular and hybrid types.
Unlike many designs, nulling the input offset voltage does
not degrade common-mode rejection ratio or input offset
voltage drift. Low input voltage noise and current noise plus a
low 1/f noise corner frequency allow these amplifiers to be
used in a variety of low noise, wide-bandwidth applications.

•

PM-155 (Only) ................... LF155 Replacement
Low Supply Current . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 2mA

•
•

PM-156 (Only) ................... LF156 Replacement
High Slew Rate ............................ 12V//lsec
Fast Settling to ±O.01% ....................... 4.0/lsec

Dynamic specifications for the PM-155 include a slew rate of
5V1 /lS, a 2.5MHz gain bandwidth product, and settling time to
within ±0.01% of final value in 5.0,..s. The PM-156 has a slew
rate of 12V/,..s and a settling time of 4.0/ls to ±0.01% of final
value.

PM-157 (Only) ................... LF157 Replacement
• Wide-Bandwidth Decompensated (AVCL = 5 Min) ... 20MHz
• High Slew Rate ............................ 45V//lsec
• Fast Settling to ±O.01% ....................... 4.0/lsec

The PM-157 is a very fast decompensated device. This results
in a 45V/,..s slew rate, a 20MHz gain bandwidth product, and a
settling time of 4.0,..s. Decompensation requires a minimum
closed-loop gain of five because of stability considerations.
For improved performance, see the OP-15/0P-16/0P-17 data
sheet. For duals, see the OP-215 data sheet.

SIMPLIFIED SCHEMATIC

5-259

1/86, Rev. A

II

--------1~

PM-155A/PM-156A/PM-157A MONOLITHIC JFET-INPUT OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS
Supply Voltage
PM-155A, PM-156A, PM-157A, PM-155, PM-156, PM-157,
PM-355A, PM-356A, PM-357A ................. ±22V
Internal Power Dissipation
PM-155A, PM-156A, PM-157, PM-155, PM-156,
PM-157 .................................... 670mW
PM-355A, PM-356A, PM-357 A ................... 500mW
(Derate based on a thermal resistance of 150° C/W
junction to ambient or 45° C/W junction to case.)
Operating Temperature Range
PM-155A, PM-156A, PM-157 A, PM-155, PM-156,
PM-157 ....•.•.•....•.....•...... -55°C to +125°C
PM-355A, PM-356A, PM-357 A .•......... 0° C to + 70° C
Maximum Junction Temperature (Tj)
PM-155A, PM-156A, PM-157A, PM-155, PM-156,
PM-157 ................................•... +150°C
PM-355A, PM-356A, PM-357 A ...•..•••........ -+ 100° C

Differential Input Voltage
PM-155A, PM-156A, PM-157A, PM-155, PM-156, PM-157,
PM-355A, PM-356A, PM-357A ................. ±40V
Input Voltage
PM-155A, PM-156A, PM-157A, PM-155, PM-156, PM-157,
PM-355A, PM-356A, PM-357A ...............•. ±20V
NOTE:
The absolute maximum negative input voltage is equal to the negative power
supply voltage.

Output Short-Circuit Duration ................. Indefinite
Storage Temperature Range. . • . . . . . . .. -65° C to + 150° C
Lead Temperature Range (Soldering, 60 sec) .... +300°C

ELECTRICAL CHARACTERISTICS at ±15V::; Vs::; ±20V, -55°C::; TA ::; +125°C and THIGH = +125°C for PM-155A,
PM-156A and PM-157A, 0° C::; TA::; +70° C and THIGH= +70°C for PM-355A, PM-356A and PM-357 A, unless otherwise noted.
PM-1SSAI
PM-1S6A1
PM·1S7A

PM-3SSAI
PM·3S6A1
PM·3S7A

PARAMETER

SYMBOL

CONDITIONS

TYP

MAX

Input Offset Voltage

Vas

Rs = 500

1.4

2.5

Input Offset Voltage Drift

TCVos

Rs=500

3

( aTCVos)
avos

Rs= 500

05

Change in Input Offset
Drift with Vos Adjust

MIN

MIN

TYP

MAX

1.2

2.3

UNITS
mV
p.VloC
p.V/oC per mV

0.5

Input Offset Current

los

T,,; THIGH (Note 1)

4.0

10

0.4

1.0

nA

Input Bias Current

Ie

T,'; THIGH (Note 1)

±10

±25

±2

±5

nA

Large-Signal Voltage Gain

Ava

Vs =±15V, Vo =±10V,
RL = 2kO

25

75

25

75

VlmV

Output Voltage Swing

Va

Vs = ± 15V, RL = 10kO
Vs = ± 15V, RL = 2kO

±12
±10

±13
±12

±12
±10

±13
±12

V

Input Voltage Range

IVR

Vs =±15V

±10.4

+151
-12.0

±104

+15.1
-120

V

CMRR

VcM=±IVR

85

100

85

100

dB

PSRR

(Note 2)

Common~Mode

Rejection

RatIO

Power Supply Rejection
Ratio

10

57

10

57

p.VIV

NOTES:
1 PMI has a bias current compensation cIrcuit which gives Improved bias
current overthe standard JFET Inputop amps. 'Band los are measured at
VCM=O.
2. Power supply rejection ratio is measured for both supply magnitudes
Increasing or decreasing simultaneously, In accordance With common
practice.

5-260

1/86, Rev. A

---------t~ PM-155A/PM-156A1PM-157A MONOLITHIC JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at ±15V:5 VS:5 ±20V. TA = 25°C. unless otherwise noted.
PM·155A1
PM·156A1
PM·157A
PARAMETER

SYMBOL

CONDITIONS

Input Ollset Voltage

Vos

Rs = 500

Input Offset Current

los

TI = 25'C (Note 1)

3

10

Input Bias Current

Ie

TI = 25'C (Note 1)

±30

±50

Input Resistance

R'N

Large-Signal Voltage Gain
Supply Current

Slew Rate

Avo
ISY

SR

MIN

GBW

ts

Vs =±15V, Vo =±10V,
RL = 2kO

en

Vs =±ISV
AVCL = +1, Vs= ±ISV

AVCL =+1. Vs =±ISV

Vs = ±15V (Note 2)

Input Noise Current

In

Input Cspacitance

C 'N

Rs =
Rs =
Rs =
Rs =

1000, I
1000, I
1000, I
1000, I

50
PM-15S
PM-15S/PM-1S7

Vs =±15V(Note3)

Input NOise Voltage

MIN

10'2

AVCL =+5, Vs =±ISV
Selliing Time (to ±0.01%)

MAX

TYP

MAX

UNITS

2

mV

3

10

pA

±30

±50

pA

2

AVCL =+5, Vs =±15V
Gain Bandwidth Product

TYP

PM-355A1
PM·356A1
PM·357A

=
=
=
=

100Hz
1000Hz
100Hz
1000Hz

PM-ISS
PM-1S6
PM-I 57
PM-15S
PM-156
PM-I 57
PM-ISS
PM-ISS
PM-I 57
PM-ISS
PM-156/PM-1S7

I = 100Hz, Vs = ± ISV
1= 1000Hz. Vs =±ISV

10'2

200
2

50
4
7

12
45

40
15

2.5
45
20

200
2
5

3

3
10
40

0
VlmV
4
7

mA

10
40

5
12
45

VII's

4.0
15

2.5
4.5
20

MHz

50
40
40

40
1.5
1.5

/'S

25
20
15
12

25
20
15
12

nVl..,fHZ

001
0,01

0.Q1
0.01

pN..,fHZ

3

3

pF

~
......

NOTES:
1. PMI has a bias current compensation circuit which gives improved bias
current over the standard JFET input opamps.IBand losare measured at
VCM=O.
2. Selliing time is dell ned here lor a unity gain inverter connection uSing 2kO
resistors. It IS the time required lor the error voltage (the voltage at the
Inverting Input Pin on the ampliller) to sellle to within 0.01% 01 its final
value Irom the time a 10V step input is applied to the inverter. See selliing
time test circuit.
3. Sallling time IS dellned herelora Av=-S connection with R F= 2kO.ltls
the time required lor the error voltage (the voltage at the Inverting Input
pin on the amplifier) to sellie to within 0.01% 01 Its Ilnal value Irom the time
a 2V step input is applied to the inverter. See selliing time test circuit.

5-261

ge

1/86, Rev. A

~

::s
I=l-<

~

~

0
......

~
~

--------1~ PM-155A/PM-156A/PM-157A MONOLITHIC JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at TA = +25 0 C, ± 15V:S Vs:S ±20Vfor PM-155, PM-156 and PM-157, unless otherwise noted.
PM-155
PM-156
PM-157
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs~50n

TYP

MAX

Input Offset Current

los

T,~

25°C (Note 1)

3

20

Input Bias Current

Ie

T,~

pA

25°C (Note 1)

±30

±100

Input Resistance

R'N

pA

Large-Signal Voltage GaIn

Avo

±15V.
Rl ~ 2kn

Supply Current

ISY

Vs~±15V

Slew Rate

SR

Vs~

AVCl ~+1,

MIN

mV

Vo~

±10V,

50

PM-155
PM-156
PM-157

Vs~±15V

Settling Time (to ±O,Ol%)

Input Noise Voltage

GBW

ts

en

Input Noise Current

in

Input Capacitance

C'N

AVCl ~+1,

Vs~±15V

AVCl ~+5,

Vs~±15V

Vs~±15V

(Note 2)

Vs~±15V

(Note 3)

Rs ~
Rs ~
Rs ~
Rs ~
f
f

~

~

lOOn,
100n,
100n,
lOOn,

f
f
f
f

~
~

~
~

100Hz
1000Hz
100Hz
1000Hz

n

200

VlmV

7.5
30

12
40

4

rnA

VII's

PM-155
PM-156
PM-157

25
5
20

MHz

PM-155
PM-156
PM-157

5
4
4

p.s

25
20
15
12

nV/jHZ

0,01

pAljHZ

3

pF

PM-155
PM-156/PM-157

100Hz, Vs ~ ± 15V
1000Hz, Vs ~ ±15V

10'2

2
5

PM-155
PM-156/PM-157

Avcl ~ +5, Vs ~ ±15V
Gain Bandw,dth Product

UNITS

NOTES:
1. PMI has a bias current compensation circuit which gives improved bias
current over the standard J FET input op amps, leand I osare measured at
VCM~ O.
2. Settling time is defined here for a unity gain inverter connection using 2kn
resistors. It is the time required for the error voltage (the voltage at the
inverting input pin on the amplifier) to settle to within 0.01% of its final
value from the time a 10V step input is applied to the inverter. See settling
time test circuit.
3. Settling time is defined here for a Av~-5 connection with RF~ 2kn.1t is the
time required for the error voltage (the voltage at the Inverting input pin on
the amplifier) to settle to within 0.01'10 of its final value from the time a 2V
step input is applied to the inverter. See settling time test circuit.

5-262

1/86, Rev. A

-------l~ PM-155A/PM-156A/PM-157A MONOLITHIC JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at ± 15V:O; Vs:O;±20V and -55 0 C:O; TA:O; + 1250 C and THIGH= +125 0 C for PM-155, PM-156
and PM-157, unless otherwise noted.
PM-155
PM-156
PM-157
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs= 500

Input Offset Voltage
Drift

TCVos

Rs = 500

( aTCVos)
avos

Rs= 500

Change In Input Offset
Drift With Vas Adjust.

MIN

TYP

MAX

UNITS

4

mV

p.V/'C
p.V/'C
per mV

0.5
20

nA

±50

nA

Input Offset Current

los

T,'; THIGH (Note I)

Input Bias Current

18

T,'; THIGH (Note I)

Large-Signal Voltage
Gain

AyO

Vs= ±15V, Vo= ±10V
RL = 2kO

25

75

V/mV

Output Voltage Swing

Va

Vs= ±15V, AL = 10kO
Vs = ±15V, RL = 2kO

±12
±10

±13
±12

V

±10.4

+151
-12.0

V

100

dB

Input Voltage Range

IVR

±2

Vs = ±15V

Common-Mode
Rejection Ratio

CMRR

VcM=±IVR

Power Supply
Rejection Ratio

PSRR

(Note 2)

85

10

~......
~

~

p.,

57

p.VIV

NOTES:
1. PMI has a bias current compensation circuit which gives improved bias
current overthe standard JFET input op amps. IBand IOSara measured at
VCM = 0, T, = +125'C.

2. Power supply rejection ratio is measured for both supply magnitudes
increasing or decreasing simultaneously, In accordance with common
practice.

ORDERING INFORMATIONt

PIN CONNECTIONS

~
.....:l

~

0......

~

~

C5

PACKAGE
TA = 2S'C
VosMAX
(mV)

TO-99
8-PIN

8-PIN
HERMETIC
DIP

2.0

PMI55AJ'
PMI56AJ'
PMI57AJ'

PMI55AZ'
PMI56AZ'
PMI57AZ'

2.0

PM355AJ
PM356AJ
PM357AJ

PM355AZ
PM356AZ
PM357AZ

PMI55J'
PMI56J'
PMI57J'

PM155Z
PM156Z
PM157Z

5.0

LCe

OPERATING
TEMPERATURE
RANGE

PM155ARC/883
PM156ARC/883

MIL

NC
8

_,::0

V
:OUT

"SC7.AL

COM

8-PIN HERMETIC DIP
(Z-Sufflx)

v-

4

(CASE)

TO-99 (J-Sufllx)

MIL

• For deYlces processed In total compl,ance to MIL-STD-883, add 1883 after
part number Consult factory for 883 data sheet.
tAil commercial and industrial temperature range parts are available with
burn-In. For ordering Information see 1986 Data Book, Section 2.

PM-155ARC/883,
PM-156ARC/883
20-LEAD LCC
(RC-Suffix)

5·263

1/86, Rev. A

_ _ _ _ _---IIfMDPM-155A1PM-156A1PM-157A MONOLITHIC JFET-INPUT OPERATIONAL AMPLIFIERS

BASIC CONNECTIONS

If both inputs exceed the negative common-mode voltage
limit, the amplifier will be forced to a high positive output. If
only one input exceeds the negative common-mode voltage
limit, a phase reversal takes place forcing the output to the
corresponding high or low state. In either of the above
conditions, normal operation will return when both inputs are
returned to within the specified common-mode voltage range.

SETTLING-TIME TEST CIRCUIT
2kn 0 1%
+15V

:oy

o-_'2-J1"A'0,..,'_%......"-1
5k"
01%

_15V

VOUT

POWER SUPPLY CONSIDERATIONS
Power supply polarity reversal can result in a destroyed unit.

Av=-l

2N4416

Exceeding the positive common-mode limit on a single input
will not change the phase of the output. However, if both
inputs exceed the limit, the output of the amplifier will be
forced to a high state.

+1SV

.. 400.11 ±O 1% FOR PM·157A
** lk.l1 ±O 1% FDA PM·157A

INPUT OFFSET VOLTAGE NULLING

NOTE
FOR POTENTIOMETERS WITH A TEMPERATURE COEFFICIENT";;; 10Qppm/"C,
THE ADDED TCVos WITH NULLING IS"" 0 5/NloC/mV Of ADJUSTMENT

DYNAMIC OPERATING CONSIDERATIONS
As with most amplifiers, care should be taken with lead
dress, component placement, and supply decoupling in
order to ensure stability. For example, resistors from the
output to an input should be placed with the body close to the
input. This minimizes "pick-up" and increases the frequency
of the feedback pole by minimizing the capacitance from
input to ground.

A feedback pole is created when the feedback around any
amplifier is resistive. The parallel resistance and capacitance
from the input of the device to AC ground sets the frequency
of the pole. In many instances, the frequency of this pole is
much greater than the expected 3dB frequency of the
closed-loop gain. Consequently, the pole has negligible
effect on stability margin. However, if the feedback pole is
less than approximately six times the expected 3dB frequency, a lead capacitor should be placed from the output to
the inverting input of the op amp. The capacitor value should
be such that the RC time constant of the capacitor and
feedback resistor is greater than, or equal to, the original
feedback-pole time constant.

APPLICATIONS INFORMATION
INPUT VOLTAGE CONSIDERATIONS
The PM series JFET input stages can accommodate large
input differential voltages without external clamping as long
as neither input exceeds the negative powersupply. An input
voltage which is more negative than V- can result in a
destroyed unit.

5-264

1/86, Rev. A

PM-725
INSTRUMENTATION
OPERATIONAL AMPLIFIER
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•

addition, Precision Monolithics' exclusive Silicon-Nitride
"Triple Passivation" process minimizes "popcorn noise" and
provides maximum reliability and long-term stability. For
improved specifications, refer to the OP-06 series data sheet.
For devices with internal frequency compensation, refer to
the OP-05 instrumentation amplifier and OP-07 ultra-low offset
voltage operational amplifier data sheets.

Extremely High Gain ......................... 3M Typ
Low Offset Voltage and Offset Current
Low Drift with Temperature
High Common-Mode ReJecllon ............ 110dB Min
High Power Supply ReJecllon ............. 10p.VIV Max
Silicon-Nitride Passlvallon
Dlfferenllal-Input Overvoltage Protecllon

II

PIN CONNECTIONS
ORDERING INFORMATIONt
PACKAGE
HERMETIC
TA=2S"C
VosMAX
(mV)

1.0
2.5

TO·99
a·PIN

DIP
a·PIN

PM725J"
PM725CJ

PM725Z"
PM725CZ

vos
PLASTIC
DIP
a·PIN
PM725CP

OPERATING
TEMPERATURE
RANGE

-IN 2

MIL
COM

e
rn
+

+IN 3

7V+

6 OUT

5 COMP

4
v-leASE)

* For devices processed in total compliance to MIL-STD-S83, add 1883 after

part number. Consult factory for 883 data sheet.
commercial and industrial temperature range parts are available with
burn·,n. For ordering information see 1986 Data Book, Section 2

TO-99
(J-Sufflx)

t All

GENERAL DESCRIPTION

8-PIN HERMETIC DIP
(Z-Sufflx)
&
EPOXY MINI-DIP
(P·Sufflx)

The PM-725 series of monolithic instrumentation operational
amplifiers provide industry-standard 725 specifications. In

SIMPLIFIED SCHEMATIC
v+

A1'
OUTPUT

A20

v-

5·265

1/86, Rev. A

------------1~ PM-725 INSTRUMENTATION OPERATIONAL AMPLIFIER

PM-725C .•.•••••••.••••••••..•••..••••• 0° C to + 70° C
Lead Temperature Range (Soldering, 60 sec) ••••.. 300° C

ABSOLUTE MAXIMUM RATINGS
Supply Voltage ••••••••••••••••••••.•.•••..••••..• ±22V
Internal Power Dissipation (see note) ••••••..•.•• 500mW
Differential Input Voltage ••••••••••••..•.•...•••... ±30V
Input Voltage ••••••••••••••••••••••.•.••• Supply Voltage
Output Short-Circuit Duration ••••••••••••...•• Indefinite
Storage Temperature Range
J and Z Packages • • . • . • . • • • • • • . • • . .. -65° C to + 150° C
P Package ••••..••..•.••.•••.•..••• -65°C to +125°C
Operating Temperature Range
PM-725 .•••••••••••...•.••••.••.••. -55° C to + 125° C

ELECTRICAL CHARACTERISTICS at Vs = ±

15V, TA

NOTE:
1, See table for maximum ambient temperature rating and derating factor,

PACKAGE TYPE

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (JI

BO°C

7,lmW/oC

B-Pin Plastic DIP (PI

36°C

5,6mW/oC

B-Pin Hermetic DIP (ZI

75°C

6,7mW/oC

= 25° C, unless otherwise noted.
PM-725

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs 510kO

Input Offset Current

los

Input Bias Current

Is

Input Noise Voltage

en

MIN

MAX

TYP

MAX

UNITS

0,5

1,0

0,5

25

mV

20

2

35

nA

100

42

125

nA

42
fo = 10Hz
fo= 100Hz
fo= 1000Hz

Input Resistance

R'N

Input Voltage Range

IVR

Large-Signal Voltage
Gain

Avo

RL ",2kO, Vo =±10V

Common-Mode
Rejection Ratio

CMRR

Rs 510kO, VcM =±13,5V

Power Supply
Reiection Ratio

PSRR

Rs 510kO, Vs =±5Vto±15V

Output Voltage Swing

Vo

RL ",10kO
RL ",2kO

Output Resistance

Ro

Vo=O,lo=O

Power Consumption

Pd

No Load

PM-725C

TYP

MIN

15
9
B
1.5

nV/y'HZ

1,5

MO

±13.5

±14

±13,5

±14

V

1,000

3,000

250

3,000

VlmV

110

120

94

120

dB

10
±12,0
±10,0

±13,5
±13.5

BO

2
±12.0
±10,0

150

5·266

15
9
B

35

±13,S
±13,5

V

150
105

BO

p.V/v

0
150

mW

1/86, Rev. A

-----------l~ PM-725 INSTRUMENTATION OPERATIONAL AMPLIFIER

COMPENSATION CIRCUIT

OFFSET VOLTAGE NULL CIRCUIT

R3"

Rl

Cl

C2

R2

.V-OR~
GND

-:-

.USE R3 '" 510 WHEN THE AMPLIFIER
IS OPERATED WITH CAPACITIVE LOAD

PINOUTS FOR J, Z, AND P PACKAGES,

COMPENSATION COMPONENT VALUES

",

C,
CuF)

"2

C2

1m

(p.F)

00015

AV

In)

10,000

'Ok
470
47

50pF

27

005

270

'0

0,05

3.

1,000

'00
'0

II

0001
0.01
0.02

-FOR MAXIMUM PSRR VS FREQUENCY
COMPENSATION NETWORK SHOULD
BE RETURNED TO v-

PINOUTS FOR J, Z, AND P PACKAGES,

ELECTRICAL CHARACTERISTICS at Vs=±15V, -55° C:5TA:5+125°Clor PM-725, O°C:5TA:5+700 Clor PM-725C, unless
otherwise noted,

PM-725
MIN

TYP

PM-725C
MAX

MIN

TYP

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs510kO

Average Input Offset
Voltage Drift

TCVos

Rs = 500, Unnulled (Note 1)

Average Input Offset
Voltage Drift

TCVOSn

Rs = 500, Nulled

0.6

Input Offset Current

los

TA=MAX
TA=MIN

1.2
7.5

20
40

1.2
4.0

Average Input Offset
Current Drift

TClos

(Note 1)

35

150

10

Input Bias Current

18

TA = MAX
TA = MIN

20
80

100
200

30
100

Large-Signal Voltage
Gain

Avo

RL <:2kO, TA = MAX
RL <:2kO, TA=MIN

Common-Mode
Rejection Ratio

CMRR

Rs 510kO, VCM =±13.5V

Power Supply
Rejection Ratio

PSRR

Rs 5 10kO, Vs = ±5V to ± 15V

Output Voltage Swing

Vo

RL <: 2kO

1.5

3.5

100
20
±10

±10

UNITS

mV

2

".V/oC

0.6

I'V/oC

125
125

1,000
250

MAX

35
50

nA
pA/oC

125
250

nA

VlmV
115

dB

20

I'V/v
V

NOTE: 1. Sample tested.

5-267

1/86, Rev. A

PM-741
COMPENSATED
OPERATIONAL AMPLIFIER
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

•
•
•
•
•

The PM-741 series of internally-compensated operational
amplifiers provide industry-standard 741 specifications. In
addition, Precision Monolithics' exclusive Silicon-Nitride
"Triple Passivation" process provides high reliability and
long-term stability of parameters. For higher performance
general purpose op amps, refer to the OP-02 data sheet. See
the OP-04/0P-14 data sheet for duals.

Industry Standard 741 Specifications
Internal Frequency Compensation
Continuous Short·Clrcult Protection
Silicon-Nitride Passivation
Low Noise

ORDERING INFORMATIONt
TA = 25°C

PACKAGE

OPERATING

VosMAX

TO-99
I-PIN

TEMPERATURE
RANGE

PM741J
PM741CJ

MIL
COM

(mV)
5.0
6.0

PIN CONNECTIONS
8Ne

BAL~7V'
TO-99

t All

commercial and industnal temperature range parts are available with
burn-in. For ordenng mformatlOn see 1986 Data Book, Section 2.

-IN 2

-

+IN3

6 OUT

(J-Sufflx)

SBAL
4

v- (CASE)

SIMPLIFIED SCHEMATIC

~--",,-------.£"Q20

R"

OUTPUT

,----+--0
R8

ABSOLUTE MAXIMUM RATINGS

Rl0

Operating Temperature Range
PM-741 ............................ -55°C to +125°C
PM-741C ............................... O°C to +70°C

Supply Voltage
PM-741 ........................................ ±22V
PM-741C ....................................... ±18V
Internal Power Dissipation (Note 1) .............. 500mW
Differential Input Voltage .......................... ±30V
Input Voltage ........................... , Supply Voltage
Output Short-Circuit Duration ................. Indefinite
Storage Temperature Range ........... -65°C to +150°C
Lead Temperature Range (Soldering, 60 sec) ...... 300°C

NOTE:
1. See table for maximum ambient temperature rating and derating factor.

PACKAGE TYPE
TO-99 (J)

5-268

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

eo°c

7.1mW/oC

1/86, Rev. A

------------I1fMD PM-741 COMPENSATED OPERATIONAL AMPLIFIER
ELECTRICAL CHARACTERISTICS at

TA

= 25° C, Vs =

± 15V, unless otherwise noted.
PM-741

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Rs " 10kfl

Input Offset Current

MIN

TYP

PM-741C
MAX

MIN

MAX

UNITS

5.0

TYP

6.0

mV

los

200

200

nA

Input Bias Current

18

SOO

500

Input Resistance

R'N

(Note II

Large-Signal
Voltage Gain

Avo

R L " 2kfl. Vo= ±IOV

Supply Current

ISY

VOUT= 0

0.3

Mfl

50,000

25,000

V/v

2.8

2.8

ELECTRICAL CHARACTERISTICS at -55° C::; TA::; + 125° C for PM741, 0° C::; TA::; +70° C for PM741 C, Vs =
otherwise noted.
PM-741
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs" 10kn

Input Offset Current

los

Input Bias Current

18

Large-Signal
Voltage Gain

Avo

Output Voltage Swing

nA

0.3

MIN

TYP

mA

±15V, unless

PM-741C
MAX

MIN

TYP

MAX

UNITS

6.0

7.5

mV

500

300

nA

15

08

JIoA

25.000

15.000

V/v

Vo

±12
±10

±12
±10

V

±12

±12

V

70

70

dB

Input Voltage Range

IVR

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio

PSRR

R L " 2kfl. Vo=±IOV

142

Rs" 10kfl

142

JIoV/v

NOTE:
I. Guaranteed by design

TYPICAL OFFSET NULLING CIRCUIT

TYPICAL BURN-IN CIRCUIT

v-

~20V

5-269

1/86, Rev. A

PM-747
DUAL COMPENSATED
OPERATIONAL AMPLIFIER
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

•

The PMI series of internally-compensated operational amplifiers provides industry-standard 747 specifications. In addition,
Precision Monolithics' exclusive Silicon-Nitride "Triple Passivation" process provides maximum reliability and long-term
stability of parameters for lowest overall system operating cost.

•
•
•
•

Dual PM-741 Internally-Compensated Operallonal
Amplifier
Internal Frequency Compensation
Low Power Consumption
Continuous Short-Circuit Protection
Silicon-Nitride Passivation

PIN CONNECTIONS
ORDERING INFORMATIONt
TA = 25°C
VosMAX
(mY)

PACKAGE
HERMETIC
DIP
14-PIN

OPERATING
TEMPERATURE
RANGE

5.0
6.0

PM747Y
PM747CY

MIL
COM

14-PIN DIP
(Y-Sufflx)

*V+ (A) AND v+ (B)
INTERNALLY CONNECTED

tAli commercial and industrial temperature range parts are available with
burn~in, For ordering information see 1986 Data Book, Section 2.

SIMPLIFIED SCHEMATIC (1/2 of Circuit Shown)

R9
OUTPUT

,----1'--C
Rl0

5·270

1/86, Rev. A

-----------I~ PM-747 DUAL COMPENSATED OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

Operating Temperature Range

Supply Voltage
PM-747 •.•.•..•••...•••.•••.•.••••••.•..•.•••.. ±22V
PM-747C ••.••.•••.•..•••••••••.•••••.•.•••••••• ±18V
Internal Power Dissipation (Note 1)

Y Package ••..•.....•.••.•••••.•...•••.....•. 670mW

PM-747

............................

NOTE:
1. See table for maximum ambient temperature rating and derating factor.

Differential Input Voltage .......................... ±30V
Input Voltage ••.••.••.•••.•••••••••••.••• Supply Voltage
Output Short-Circuit Duration •.•.••••••...•••. Indefinite

PACKAGE TYPE

Storage Temperature Range ••••••••••• -65°C to +150°C
Lead Temperature Range (Soldering. 60 sec) ••••.• 300°C

14-Pin Hermetic DIP (Y)

ELECTRICAL CHARACTERISTICS

-55°C to +125°C

PM-747C ............................... O°C to +70·C

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

83· C

10.0mW/·C

at TA = 25°C. Vs= ±15V. unless otherwise noted.

PM-747
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs '; 20kO

MIN

PM-747C

TYP

MAX

1.0

5.0

MIN

TYP

MAX

UNITS

1.0

6.0

mV

Input Offset Current

los

20

200

20

200

nA

Input Bias Current

18

80

500

80

500

nA

Input Resistance

R'N

Input Capacitance

C 'N

(Note 1)

022

Offset Voltage
Adiustment Range
Large-Signal
Voltage Gain

Avo

RL "2kO. Vo =±10V

Output Voltage Swing

Vo

RL" 10kO
RL " 2kO

Output Resistance

Ro
Isc

Output Short-Circuit
Current

Supply Current

ISY

Input Voltage Range

IVR

Common~Mode

Rejection Ratio

Power Supply
Rejection Ratio

Rs'; 20kO. VCM = ±10V

PSRR

Vs = ±5V to ±20V
Vs =±5Vto±18V

2.0

MO

14

0.3

1.4

pF

±15

±15

mV

50

200

25

200

V/mV

±12
±10

±14
±13

±12
±10

±14
±13

V

75

75

0

25

25

rnA

Per Amplifier. No Load

CMRR

2.0

1.7

2.8

1.7

2.8

rnA

±12

±13

±12

±13

V

70

90

70

90

dB

30

150
30

150

50

85

!'VIV

Power Consumption

Pd

Per Amplifier. No Load

50

Transient Response.

Risetime
Overshoot

Y,N = 20mV. RL = 2kO
C L ,; 100pF

0.3
5

0.3
5

!,s

Slew Rate

SR

RL ,,2kO

0.7

0.7

V/!'s

Channel Separation

CS

120

120

dB

Unity Gain

85

mW
%

NOTE:
1. Guaranteed by input bias current.

5·271

1/86, Rev. A

I
ffi

......
j:.l..,

~

~

~

0
......

~
~

C5

~

PM-747 DUAL COMPENSATED OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs=±15V, -55· C:5TA:5+125· Cfor PM-747, O·C:5TA:5+70·Cfor PM-747C, unless
otherwise noted.

PM-747
MAX

TYP

MAX

UNITS

Rs:5 20kn

1.0

6.0

1.0

7.5

mV

los

TA = MAX
TA = MIN

7
85

200
500

7

30

200
300

nA

18

TA = MAX
TA = MIN

0.03
0.3

0.5
15

0.03
0.10

0.5
0.8

!loA

Output Voltage Swing

Va

RL--+--------If---'

I
I

I

r-.J

I

I
I
I
I

I
I

OUTPUT

I

I
IL ____ ...lI
vNOTE: BIAS CIRCUIT SHOWN IN DASHED BOX IS COMMON TO ALL DEVICES.

5-279

1/86, Rev. A

----------I~ JM-38510/11004 JAN QUAD 741-TYPE OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

The differential input voltage range shall not exceed the supply voltage
range.

Supply Voltage Range (Note 1) •.•••.••.•••••••.•..••• ±22V

4

Input Voltage Range (Note 2) ••.•...••.•••••••.•.•••• ±22V

Short circuit may be to ground orelther supply. Rating applies to +125'C
case temperature or +75° C ambient temperature

Differential Input Voltage Range (Note 3) ..•••..•••..•• ±30V
Input Current Range ..••..•.••.•..••..•..•.••• 10 to 0.1 mA

For short-term test (In the specific burn-in and lile-test configuration
where required and up to 168 hours maximum)

1j =

2750 C.

Storage Temperature Range •••.•....•••.• -65°C to +150°C
Output Short-Circuit Duration (Note 4) ........... Unlimited

RECOMMENDED OPERATING CONDITIONS

Lead Temperature (Soldering, 60 sec) ••..••••.••..••• 300°C
Junction Temperature

Supply Voltage Range •••••••.•.•••••••••••••• ±5V to ±20V
Ambient Temperature Range •.••.••.••.•.•.• -55° to +125° C

(Tj) (Note 5) ..•••.••..•...•..• 175°C

NOTES:
1 Voltages In excess of these may be applied for short-term tests If voltage
difference does not exceed 44 volts

2

For supply voltages less than ±20V. the absolute maximum input voltage IS
equal to the supply voltage

ELECTRICAL CHARACTERISTICS

at ±5V:S Vcc:S ±20V and -55°C:S TA:S 125°C, Rs = 50n, unless otherwise noted.

04 LIMITS
PARAMETER

Input Offset Voltage
Input Offset Voltage
Temperature Sensitivity

SYMBOL

CONDITIONS

MIN

MAX

UNITS

V,O

TA =25'C
-55'C ST. S 125'C
(Note 1)

-5
-6

5
6

mV

t.V,o/t.T

-55'C S T. S 125'C

-25

25

pVl'C

-75

75

-150

150

-55'C ST. S 25'C

-1000

1000

25'C ST. S 125'C

-500

500

+118

Rs =20kO,
25'C S T. S 125'C
TA =-55'C
(Note 1)

-250
-400

-1
-1

-lIB

Rs =20kO,
25'C ST. S 125'C
TA =-55'C
(Note 1)

-250
-400

-1
-1

+PSRR

+Vee

-100

100

-PSRR

+Vee

-100

100

25' C S TA S 125' C, Rs
(Note 1)
Input Offset Current

Input Offset Current
Temperature Sensitivity

1,0

RatiO

Input Voltage Common-Mode
Rejection

Output Short C"CUlt
Current

=20kO

nA

.l1,o/t.T

Input Bras Current

Power Supply Rejection

T. =-55' C, Rs
(Note 1)

=20kO

CMR

'08(+)

pAI'C

=10V, -Vee =-20V
=20V, -Vee =-10V
Common-Mode Range =30V
(Note 2)
±Vee =±15V, 25'C ST. S 125'C
(Note 3)

nA

pVIV

76

dB

-80
mA

Output Short C"cUlt
Current

Supply Current

Output Voltage SWing

loS I_I

±Vee = ±15V, T. = -55'C
(Note 3)

80

Icc

T. = -55'C
Vee = ±15V
TA = 25'C
(Note 4)
TA = 125'C

13
11
11

+Vop

_
RL = 10kO
Vee - ±20V, RL = 2kO

-Vop

_
RL = 10kO
Vee - ±20V, RL = 2kO

mA

+16
+15
V

(MaXimum)

5-280

-16
-15

1/86, Rev. A

-----------I1fMD

JM-38510/11004 JAN QUAD 741-TYPE OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at ±5V :5,Vcc :5,±20V and -55 0 C :5,TA :5,125° C, Rs = 500, unless otherwise noted. (Continued)
04 LIMITS
PARAMETER

Open-Loop Voltage Gain
(Single Ended)

SYMBOL

CONDITIONS

A VS (+)

RL = 10kO, ±Vo = ±15V.
TA = 25°C
-55°C S TA S 125°C

50
25

AVSH

RL = 2kO, ±Vo = ±15V.
TA =25°C
-55°C STA S 125°C

50
25

RL = 10kO, TA = 25°C
RL = 2kO, ±Vcc = ±5V,

10
10

Avs

MIN

MAX

UNITS

V/mV

-55°C ::;TA ::; 125°C
Transient Response
Rise Time
Transient Response
Overshoot

Slew Rate
NOise
(Broadband)

Noise
(Popcorn)
Channel Separation

TR ltr )

±Vcc = ±20V, Av = 1

TRIOS)

±Vcc = ±20V

SR(+)

±Vcc = ±20V, Av = 1

N,(BB)

TA = 25° C, ±Vcc = ±20V,
Rs = 500

N,(PC)

TA = 25° C, ±Vcc = ±20V,
Rs = 20kO

CS

TA =25°C

OFFSET VOLTAGE
CONDITION

!is

50

!is

06

JiV rms

dB

80

NOTES:
1. Tested at VCM = 0, +15V and -15V with ±Vcc = ±20V, and at VCM = OV and
-2.5V with ±Vcc = ±5V
2. CMR is determined by measuring Input offset voltage as follows

03

3

Only one amplifier shorted to ground atone time, 0 :5 t :5 25ms Continuous
limits will be considerably lower and apply for -55°C :5 TA ::;: 25° C
Icc limits are the total for all four amplifiers at no load, connected as
followers with the nonlnvertlng inputs grounded

.Vee

+Vee
35V

-5V

15V

5V

-35V

-15V

BURN-IN CIRCUIT

6800

680n

J"'.'

680n

680n

,on

300n

-,v

J

O ,.,

+15V

aoon

1_
+10V

o".''--_~--'-IL-_ _ _~===~I-''--------'
,on

1_

O.1j.1F

NOTES
1 RESISTORS ARE ±5%
2 VOLTAGE TOLERANCE -OV. +1.6V

-15V

5-281

1/86, Rev. A

II

J~38510/11401/11402/11403/

11404/11405/11406
JAN JFET-INPUT
OPERATIONAL AMPLIFIERS
Precision MOllolithics Inc.

GENERAL DESCRIPTION
This data sheet covers the electrical requirements for a
monolithic, low-power, internally-compensated JFET-input
operational amplifier as specified in MIL-M-38510/114 for
device types 01 to 06. Devices supplied to this data sheet are
manufactured and tested at PMl's MIL-M-38510 certified
facility and are listed in QPL-38510.

not have identical operational performance characteristics
across the military temperature range or reliability factors
equivalent to the MIL-M-38510 device.

Military Device Type

Generic-Industry Type

01
04

LF-155
LF-155A
LF-156
LF-156A
LF-157
LF-157A

Complete device requirements will be found in MIL-M-38510
and MIL-M-38510/114 for Class B processed devices.

02
05
03
06

GENERIC CROSS-REFERENCE INFORMATION
This cross-reference information is presented for the convenience of the user. The generic-industry types listed may

SIMPLIFIED SCHEMATIC

{2}

o---+......+-'--f--+----+-+-----+---.:+-++---~

NOTE: For values of C1, C2, RS, RB see the
following table:

5-282

0104
0205

030B

C1

7pF

1.7pF

C2

7pF

1.7pF

RS

7.2kn

3.Bkn

RB

7.2kn

3.Bkn

1/86, Rev. A

-----I~ JM38510111401/11402111403/11404/11405/11406 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS
Supply Voltage Range ............................ ±22V
Input Voltage Range (Note 1) ...................... ±20V
Differential Input Voltage Range ••••••••••••••••••• ±40V
Lead Temperature (Soldering, 60 sec) ••••••••••••• 300·C
Junction Temperature ••••••••••••••• Tj = 17S· C (Note 3)
Storage Temperature Range ••••••••••• -6S·C to +1S0·C
Output Short-Circuit Duration •••••••• Unlimited (Note 2)
NOTES:
1. The absolute maximum negative Input voltage is equal to the negative
power supply voltage.

2. Short circuit may be to ground to either supply. Rating applies to + 125°C
case temperature or + 750 C ambient temperature.
3. For short-term test (In the specific burn-In and hfe test configuration when
reqUired and up to 168 hours maximum), TJ= 275°C

RECOMMENDED OPERATING CONDITIONS
Supply Voltage Range ................... ±S to ±20 VDC
Ambient Temperature Range ••.••••••• -SS·C to +12S·C

ELECTRICAL CHARACTERISTICS at Vcc from ±SV to ±20V; source resistance = SO ohm; ambient temperature range =
-SS·C to +12S·C and figure 1, unless otherwise noted.
PARAMETER

Input Offset
Voltage

Input Offset Voltage
Temperature Sensitivity
Input Offset Current

SYMBOL

CONDITIONS

V,O

±vec = ±5V, VCM = OV
TA = 25°C
±Vee =±20V
VcM =±15V,OV
-55°CSTA S+125°C

av,o

"3T
1,0

+I JB
Input Bias Current
(Note 1)
(Note 2)
(Note 3)

-liB

01 LIMITS
MIN
MAX
-5

5

04 LIMITS
MIN
MAX
-2

UNITS

-2.5

25

30

-10

10

p.V/oC

~

±Vcc = ±20V, VCM = OV,
TJ=25°C
T 1= 125°C

-20
-20

20
20

-20
-20

20
20

pA
nA

0
......

-100
-10

3500
60

-100
-10

3500
60

pA
nA

±VCC = ±20V, VCM = +15V
TJ=25°C
tS25ms
TJ=125°C
±Vee = ±15V, VCM = +10V
TJ= 25°C
tS25ms
TJ= 125°C
±Vcc = ±20V, -15V S VCM S OV
TJ= 25°C
ts 25ms
T1= 125°C

-100
-10

300
50

-100
-10

300
50

pA
nA

-100
-10

100
50

-100
-10

100
50

pA
nA

+ Vce = 10V, -Vec = -20V
+ Vee = 20V, -Vee = -10V

85

85

dB

Input Voltage Common-Mode
Rejection (Note 4)

CMR

±Vee= ±20V
V,N =±15V

85

85

dB

Adjustment for
I n put Offset Voltage

V,oADJ(+)
V,oADJ(-)

±Vec= ±20V
±Vcc=±20V

Output Short-Circuit Current
(for Positive Output) (Note 5)

108 [+)

±Vcc =±15V
tS25ms
(Short Circuit to Ground)

Output Short-Circuit Current
(for Negative Output)
(Note 5)

108[-)

±Vec= ±15V
ts 25ms
(Short CirCUit to Ground)

Supply Current

Icc

TA = -55°C
±Vee = ±15V, TA = +25°C
TA =+125°C

Output Voltage Swing
(Maximum)

VOP

±Vce = ±20V, RL = 10kfl
±Vee = ±20V, RL = 2kfl

Open-Loop Voltage Gain
(Single Ended) (Note 6)

AV8

~
p.,

-30

+PSRR
-PSRR

AV8 [+)
AV8 [_)

~

±Vce=±20V
VCM=OV

Power Supply
Rejection Rallo

Open-Loop Voltage Gain
(Single Ended) (Note 6)

~
~
......

2
mV

-7

II

+8

+8
-8

±Vce=±20V, Vo uT =±15V
RL = 2kfl, TA = 25°C
-55°C S TAS +125°C
±Vee =±5V
RL =2kfl
VouT =±2V

-50

~
~

C5

mV

-8
-50

mA

50

50

mA

6
4
4

6
4
4

mA

±16
±15

±16
±15

V

50
25

50
25

V/mV

10

10

VlmV

1/86, Rev. A

5-263

~----

~

~----

-

-

._--_._--

---~--

------1~

JM38510/11401/11402I11403/11404/11405/11406 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at Vcc from ±5V to ±20V; source resistance = 50 ohm; ambient temperature range =
-55°C to +125°C and figure 1. unless otherwise noted. (Continued)
01 LIMITS
PARAMETER

MIN

SYMBOL

CONDITIONS

Transient Response
Rise Time

TRltr)

±Vee= ±ISV. RL =2kfi. A y = I
CL = 100pF. See Figure 2
V,N= SOmV

Transient Response
Overshoot

TRIOS)

±Vee = ±ISV. RL = 2kfi. Av= I
CL = 100pF. See Figure 2
V,N=SOmV

Slew Rate

SR(+)
and
SR(-)

V,N = ±SV. ±Vee = ±ISV
Av= I. See Figure 2
TA =2S·C
TA = -SS·C. +12S·C

Sellilng Time

ts(+)
and
ts(-)

±Vec = ± ISV (0.1% error)
TA = 2S·C. Av =-1
See Figure 3

NOise (Referred to Input)
Broadband

N,(BB)
N,(PC)

NOise (Referred to Input)
Popcorn

04 LIMITS

MAX

MIN

MAX

UNITS

150

150

ns

40

40

%

2

Vlp.s

3
1.5
4000

4000

ns

±Vcc = ±20V. TA = 2S·C
Bandwidth = 5kHz

10

10

I'V rms

±Vec=±20V. TA =2S·C
Bandwidth = 5kHz

80

80

P.Vpk

NOTES:
I Bias currents are actually Junction leakage currents which double
(approximately) for each 10·C Increase In Junction temperature T,.
Measurement of bias current IS specified at TJrather than TA• since normal
warm-up thermal transients will affect the bias currents. The measurements
for bias currents must be made within 2Sms or Sloop time constants after
power is first applied to the device for test Measurement at TA= -55· C IS
not necessary since expected values are too small for tYPical test systems.
2. Bias current IS sensitive to power supply Yoltage, common-mode voltage
and temperature as shown by the following typical curves:

IIB(nAl

,.+ /
____-'_-'-_':_1-'--"-=---1-'

_-'-1_-"1_ _ _ _ _ TA(OCI

-50

-25

26

50

75

100

3

Negative lie minimum limits reflect the characteristics of deVice with bias
current compensation
4 CMR IS calculated from V'c measurements at Vet.! = +ISV and -ISV
5. Continuous limits shall be conSiderably lower. Protection for shorts to
either supply eXists prOViding that TJ(max):;; 17S·C
6. Because of thermal feedback effects from output to Input. open-lOOp gain
IS not guaranteed to be linear or positive over the operallng range. These
'equlrements. if needed. should be specified by the user In additional
procurement documents.

IIBlpA)

tVee .. t20V

400
200

..

----.~~:=---15/-10

-6

~=o:r::--'--_y'mlyl
10

15

POWER AND THERMAL CHARACTERISTICS
Package

CASE OUTLINE
Per MIL-M-38510. Appendix C. Case Outline A-1 (8 Lead
Can). Package Type Designator "G".

8 Lead Can
(TO-GG)

Case outline

G

Maximum allowable

Maximum

Maximum

power dlsalpatlon

8J-C

8J-A

330mW at TA = 12S·C

40·CIW

ISO·CIW

PIN CONNECTIONS AND ORDERING INFORMATION

N.C.

Jan Device
JM38510/11401 BGC
JM38510/11404BGC
JM38510/11402BGC
JM38510/11405BGC
JM38510/11403BGC
JM38510/11408BGC

S

o:.~~: C\+::T

+IN~~~~t"T
4

PMI Device Type
PM155J1/38510
PM155AJ1/38510
PM156J1/38510
PM156AJ1/38510
PM157J1/38510
PM157AJ1/38510

-Vee
NOTE: Lead Flnlsh·Gold Plate
Check With factory for other qualified lead finishes

(PIN 4 CONNECTED TO CASE)

5-284

1/86, Rev. A

-----l1fMD

JM38510/11401/11402l11403/11404/11405/11406 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS
Supply Voltage Range

............................ ±22V

Short CIrcuit may be to ground to either supply. Rating applies to + 125°C
case temperature or +7S Q C ambient temperature
3. For short-term test (In the specific burn-In and life test configuration when
required and up to 168 hours maximum), T, = 275°C.

Input Voltage Range (Note 1) ...................... ±20V
Differential Input Voltage Range •.•.•••••...•••..•• ±40V
Lead Temperature (Soldering. 60 sec) ..••••••••••. 300°C
Junction Temperature

............... Tj = 175° C (Note 3)

Storage Temperature Range ••.••••••.•
Output Short-Circuit Duration.. ••••••

RECOMMENDED OPERATING CONDITIONS

-65°C to +150°C
Unlimited (Note 2)

Supply Voltage Range

ELECTRICAL CHARACTERISTICS at Vcc from ±5V to ±20V; source resistance
-55°C to +125°C and figure 1. unless otherwise noted.

Input Offset
Voltage

SYMBOL

CONDtTIONS

V,o

±Vee = ±5V, VeM = OV
TA = 25°C
±Vee= ±20V
VeM = ±15V, OV
-55°C:S TA:S + 125°C

LIMITS
MtN
MAX
-5

05

LIMITS
MIN
MAX

UNITS

-2

±Vcc= ±20V

AT

VeM = OV

Input Offset Current

1'0

±Vee = ±20V. VeM = OV.
T, = 25°C
T, = 125°C
±Vee = ±20V. VeM = +15V
T, = 25°C
t:s 25ms
T, = 125°C
±Vee = ±15V, VeM = +10V
T, = 25°C
t:S25ms
T, = 125°C
±Vee = ±20V,-15V:S VeM:S OV
T, = 25°C
tS25ms
T, = 125°C

~

30

-10

10

/,V/oC

-20
-20

20
20

-20
-20

20
20

pA
nA

0
......

-100
-10

3500
60

-100
-10

3500
60

pA
nA

-2.5

2.5

I=l-.

-100
-10

300
50

-100
-10

300
50

pA
nA

-100
-10

100
50

-100
-10

100
50

pA
nA

Input Bias Current
(Note 1)
(Note 2)
(Note3)

-I'B

Power Supply
Rejection Ratio

+PSRR
-PSRR

+Vee= 10V, -Vee = -20V
+Vee = 20V, -Vee = -10V

85

85

dB

Input Voltage Common-Mode
Rejection (Note 4)

CMR

±Vee =±20V
V'N= ±15V

85

85

dB

Adiustment for
Input Offset Voltage

V,oADJ(+)
V,oADJ(-)

±Vee=±20V
±Vee= ±20V

Output Short-Circuit Current
(for Positive Output) (Note 5)

105 (+)

±Vee= ±15V
t:s 25ms
(Short CircUit to Ground)

Output Short-CirCUit Current
(for Negative Output)
(Note 5)

1051 - 1

Supply Current

Icc

TA =-55°C
±Vee =±15V, TA =+25°C
TA = +125°C

Output Voltage Swing
(Maximum)

VOP

±Vee = ±20V, RL = 10kfl
±Vee = ±20V, RL = 2kfl

AVS(+I

±Vee = ±20V, VOUT = ±15V
RL = 2kfl, TA = 25°C
-55°C:5TA:5+125°C

Open-Loop Voltage Gain
(Single Ended) (Note 6)
Open-Loop Voltage Gain
(Single Ended) (Note 6)

+8

+8
-8

-50

±Vee= ±15V
t:s 25ms

-8
-50

mV

rnA

50

50

rnA

11

11
7

rnA

(Short Circuit to Ground)

AV51 - 1
AV5

±Vee= ±5V
RL = 2kfl
VouT =±2V

5-285

ffi......

-30

-7

~

II
::s
~

mV

Input Offset Voltage
Temperature SensItivity

+I'B

-55°C to +125°C

= 50 ohm; ambient temperature range =

02

PARAMETER

................... ±5 to ±20 VDC

Ambient Temperature Range ••••••••••

NOTES:
1. The absolute maximum negative Input voltage IS equal to the negative
power supply voltage.

±16
±15

±16
±15

V

50
25

50
25

V/mV

10

10

V/mV

1/86, Rev. A

~

~
>.t.l

C5

------l!EMD

JM38510/11401/11402l11403/11404111405/11408 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at Vcc from ±5V to ±20V; source resistance = 50 ohm; ambient temperature range =
-55°C to +125°C and figure 1, unless otherwise noted. (Continued)
02 LIMITS
PARAMETER

SYMBOL

CONDITIONS

Transient Response
Rise Time

TR ltr )

±Vcc= ±15V. RL = 2kO, Av= 1
C L = 100pF, See Figure 2
V'N=50mV

Transient Response
Overshoot

TRIOS)

±Vcc= ±15V, RL = 2kO, Av= 1
C L = 100pF, See Figure 2
V,N =50mV

Slew Rate

SR(+)
and
SR(-)

V,N = ±5V, ±Vcc= ±15V
Av = I, See Figure 2
TA = 25°C
TA= -55°C, +125°C

Settling Time

ts(+)
and
ts(-)

±Vcc= ±15V (0 1% error)
TA=25°C. Av=-1
See Figure 3

Noise (Relerred to Input)
Broadband

N,(BB)

Noise (Relerred to Input)
Popcorn

N,(PC)

MIN

MAX

UNITS

100

100

ns

40

40

%

7.5

VI"s

10

5
1500

1500

ns

±Vcc= ±20V, TA = 25'C
Bandwidth = 5kHz

10

10

",V rms

±Vcc =±20V, TA =25°C
Bandwidth = 5kHz

80

80

"Vpk

NOTES:
Bias currents are actually Junction leakage currents which double
(approximately) lor each 10'C increase In junction temperature TJ.
Measurement 01 bias current specilled at TJrather than TA, since normal
warm-up thermal transients will allect the bias currents. The measurements
lor bias currents must be made within 25ms or 5 loop time constants alter
power IS lirst applied to the device for test. Measurement at TA =-55°C IS
not necessary since expected values are too small lor typical test systems.
2. Bias current Is sensitive to power supply voltage, common-mode voltage
and temperature as shown by thelollowlng typical curves:

IS

_____

-' _'- _: _:"-'t_nA_I~,-/~I~ I- 'I~- - l"

-60

-25

26

50

75

TAI'el

100

3. Negative 1,9 minimum limits reflect the characteristics of device with bias
current compensation.
4. CMR is calculated from V,O measurements at VCM = +15V and -15V.
5. Continuous limits shall be considerably lower. Protection lor shorts to
either supply exists providing that TJ(max) S 175°C
Because of thermal feedback ellects from output to input, open-loop gain
not guaranteed to be linear or positive over the operating range. These
reqUirements, II needed, should be speCified by the user in additional
procurement documents.

IIB(pAI

400
200

-15/-'0

OS LIMITS

MAX

MIN

IS

-5

5-286

1/86, Rev. A

_ _ _ _ _-l~JM38510/11401/11402/11403/11404/11405/11406 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS
Supply Voltage Range ............................ ±22V
Input Voltage Range (Note 1) ...................... ±20V
Differential Input Voltage Range ................... ±40V
Lead Temperature (Soldering, 60 sec) ............. 300°C
Junction Temperature ............... Tj = 175°C (Note 3)
Storage Temperature Range ........... -65°C to +150°C
Output Short-Circuit Duration........ Unlimited (Note 2)

3.

Short circuit may be to ground to e,thersupply. Rating applies to +125'C
case temperature or +75°C ambient temperature.
For short-term test (In the specific burn Min and life test configuration when
required and up to 168 hours maXimum), TJ = 275°C

RECOMMENDED OPERATING CONDITIONS
Supply Voltage Range ................... ±5 to ±20 VDC
Ambient Temperature Range .......... -55° C to +125°C

NOTES:
The absolute maximum negative mput voltage IS equal to the negative
power supply voltage.

ELECTRICAL CHARACTERISTICS at Vcc from ±5V to ±20V; source resistance
-55°C to +125°C and figure 1, unless otherwise noted.

= 50 ohm; ambient temperature range =

03 LIMITS
PARAMETER

Input Offset
Voltage

SYMBOL

V,o

MIN

CONDITIONS
±Vee ~ ±5V. VeM
TA ~ 25'C
±Vee~ ±20V
VeM ~ ±15V, OV

~

OV

-5

Input Offset Current

tiV,o
tiT

1'0

+I'B

±20V
OV

±Vee ~ ±20V, VeM ~ OV.
T, ~ 25'C
T, ~ 125'C

10

30

-10

-20
-20

20
20

-20
-20

20
20

pA
nA

300
50

-100
-10

300
50

pA
nA

±Vee ~ ±20V. -15V" VeM " OV
T, ~ 25'C
t::;25ms
T, ~ 125'C

-100
-10

100
50

-100
-10

100
50

pA
nA

Adjustment for
Input Offset Voltage

V,oADJ(+)
V,oADJ(-)

±Vcc= ±20V
±Vee ~ ±20V

Output Short-Circuit Current
(for Pos,tive Output) (Note 5)

1081 +1

±Vee~ ±15V
t:525ms
(Short Circuit to Ground)

1051 -)

t-:;25ms

±Vee ~ ±20V
V'N~

±15V

±Vee~

85

85

dB

85

85

dB

+8

+8
-8

-50

-8

-50

mV

mA

±15V
50

50

11

11

mA

(Short Circuit to Ground)
TA~-55'C

Icc

±Vee~±15V, TA~+25'C

7
7

TA ~ +125'C
Output Voltage Swing
(Maximum)

VoP

±Vee ~ ±20V, RL ~ 10kO
±Vee ~ ±20V, RL ~ 2kO

Open-Loop Voltage Gain
(Single Ended) (Note 6)

A VS (+)

±Vee ~ ±20V. VOUT ~ ± 15V
RL ~ 2kO, TA ~ 25'C
-55'C"TA ,,+125'C

Open-Loop Voltage Gain
(Single Ended) (Note 6)

/JV/'C

-100
-10

+Vee ~ 10V, -Vee ~ -20V
+Vee ~ 20V, -Vee ~ -10V

Avsl -)

mA

±16
±15

±16
±15

V

50
25

50
25

V/mV

10

10

V/mV

±Vee~

Avs

±5V
RL ~ 2kO
VOUT~ ±2V

5-287

p...

~
~

-30

pA
nA

CMR

Supply Current

2.5

3500
60

Input Voltage Common-Mode
Rejection (Note 4)

(Note 5)

-25

-100
-10

+PSRR
-PSRR

.......

~

3500
60

Power Supply
Rejection Ratio

~
;:..t.:I
~

-100
-10

-I'B

(for Negative Output)

UNITS

±Vee ~ ±20V, VeM ~ +15V
T, ~ 25'C
t,,25ms
T, ~ 125'C
±Vee ~ ±15V, VeM ~ +10V
T, ~ 25'C
t,,25ms
T, ~ 125'C

Input Bias Current
(Note 1)
(Note 2)
(Note3)

Output Short-Circuit Current

MAX

-2

-7

±Vee~
VeM~

06 LIMITS
MIN

mV

-55'C" TA" +125'C
Input Offset Voltage
Temperature SensitiVity

MAX

1/86, Rev. A

~

0
.......

~

;:..t.:I

p...

0

_ _ _ _-t[fMDJM38510111401/11402111403/11404/11405/11406 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICS at Vcc from ±5V to ±20V; source resistance = 50 ohm; ambient temperature range =
-55°C to +125°C and figure 1, unless otherwise noted. (Continued)
03 LIMITS
PARAMETER

MIN

SYMBOL

CONDITIONS

Transient Response
RlaeTlme

TR 'trl

±Vcc= ±15V. Rl = 2kfl. Av= 5
C l = l00pf. See Figure 2
V'N=50mV

Transient Response
Overshoot

TR 'OSI

±Vcc= ±15V. Rl = 2kfl. Av= 5
C l = 100pF. See Figure 2
V'N=50mV

Slew Rate

SR(+)
and
SR(-)

V'N=±1V. ±Vcc=±15V
Av = 5. See Figure 2
TA =25'C
TA =-55'C. +125'C

Settling Time

18(+)
and
18(-)

±Vcc= ±15V (0.1% error)
TA =25'C. Av=-5
See Figure 3

Noise (Referred to Input)
Broadband

N,(BB)

Noise (Referred to Input)
Popcorn

N,\PC)

450

na

25

25

%

VI"s

40
25

30
20

±Vcc=±20V. TA =25'C
Bandwidth = 5kHz

10

10

p.V rms

±Vcc=±20V. TA =25'C
Bandwidth = 5kHz

80

80

"Vpk

'nA I

____-'-_.L'_::_J...L_ ...1,..:/=--,L--'-,
_.L'___.... TAI'CI
-50

-26

25

50

75

100

Negat,ve I'B minimum limits reflect the characte"stlcs of device with bias
current compensation.
4. CMR Is calculated from V,c measurements at VCM = +15V and -15V.
5 Conllnuous limits shall be considerably lower. Protection for shorts to
either supply exists providing that TJ(max) ,; 175' C.
6. Because of thermal feedback ellects from output to Input. open-lOOp gain
IS not guaranteed to be linear or positive over the operating range. These
requrrements, If needed, should be specified by the user in additional
procurement documents.

:tVee '" :l:2QV

-,,,e~;::z=---....!!!!!!~=:e:::-L--_Vcm(V)
10

450

ns

400

-5

UNITS

800

200

-16/-10

MAX

MIN

800

NOTES:
1. Bias currents are actually Junction leakage currents which double
(approximately) for each 10'C increaae in junction temperature TI'
Measurement of bias current is specified at TJratherthan TA• since normal
warm-up thermal transients will allect the bias currents. The measurements
for bias currents must be made within 25ms or 5 loop time constants alter
power islirst applied to the device for test. Measurement at TA = -55' C is
not necessary since expected values are too small for typical test systems.
2 Bias current is sensitive to power supply voltage. common-mode voltage
and temperature as shown by the following typical curves:

".,oAl

06 LIMITS

MAX

16

5-288

1/86, Rev. A

-----I~ JM38510111401/11402111403/11404/11405/11406 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

(SEE NOTE 3)

150pF

-- -If- -

K5b

560pF

K5a

!----A
499kO
(SEE NOTE 3)

--If----

II

NO

CONNECTION

102kn

(SEE NOTE 1)

499n

499Mn

499n
102kn

(SEE NOTE 3)

=*=
I
I

49.9kn

I

±1%

-=E:-

NOTES:
1. All resistors are ±O.I% tolerance and all capacitors are ±10%
tolerance, unless otherwise specified.
2. Precautions shall be taken to prevent damage to the D.U.T. during
insertion into socket and change of state of relays (I.e. disable
voltage supplies. current limit ±Vcc. etc.).
3. Compensation capacitors should be added as required for test
circuit stability Two general methods for stability compensation
exist. One method is with a capacitor for nulling amp feedback. The
other method is with a capacitor in parallel with the 49.9kn closedloop feedback resistor. Both methods should not be used simultaneously. Proper wiring procedures shall be followed to prevent
unwanted coupling and oscillations, etc. Loop response and

settling time shall be consistent with the test rate such that any
value has settled for at least five loop time constants before the
value is measured.
4. Adequatesettling time should be allowed such that each parameter
has settled to within 5% of Its final value.
5.

All relays are shown in the normal de-energized state.

S

The nulling amplifier shall bea M38510/10101XXX. Saturation of the
nulling amplifier IS not allowed on tests where the E (Pin 5) value IS
measured.
7. The load resistors 2050nand 11.1 kn yield effective load resistances
of 2kn and 10kn respectively.
8. Any oscillation greater than 300mV in amplitude (peak-to-peak)
shall be cause for device failure.

Figure 1. Test Circuit for Static Tests

5-289

1/86, Rev. A

~ JM38510111401/11402/11403/11404/11405/11406

10pF (SEE NOTE 2)

I~~~
o.1MF

JAN JFET-INPUT OPERATIONAL AMPLIFIERS

DEVICE TYPES

DEVICE TYPES

~

~

K9a

:;
5

'50

-

-

225

-

-45

50

i

~VO

-

w

'"~
0

>

Vo

~

=>

i=
=>

0

'5-

TIME (ns)

NOTES:
1. Resistors are ± 1.0% tolerance and capacitors are ± 10% tolerance
2 This capacitance includes the actual measured value with stray and
wire capacitance.
3 PrecautIOns shall be taken to prevent damage to the D U T during
insertIOn into socket and In applying power

+5

~
w

+25

~

!!
~

~
5

-25
-5
'----W;;;A-;";V7.Eo;oF;;COR"'M='- - - - . . . L - - :W
:::cA""V;;:E-::cFO"'R;;M:-:3;---TIME (us)
{POSITIVE SLEW RATE)

PARAMETER
SYMBOL

TR

(tr)

TR 10si
SR (+J

SR (-I

DEVICE
TYPE

INPUT PULSE
SIGNAL AT
tr ~ SOns

OUTPUT
PULSE
SIGNAL

EQUATION

ALL

+50mV

WAVEFORM 1

ALL

+50mV

WAVEFORM 1

TA (OS! "" 100 (.lVOIVO! %

-5V to +5V STEP
-1Vto+1VSTEP

WAVEFOAM2
WAVEFORM 2

SA

+5V to -5V STEP
-lV to +lV STEP

WAVEFORM 3
WAVEFORM 3

SR (-) = ..lVOI-)/.lt(-l

01,02,04, 05
03,06
01, 02, 04, 05

03,06

(NEGATIVE SLEW RATE)

TR ItrJ-.lt

(+) "" .1VO (+)/.lt(+)

Figure 2. Test Circuit for Transient Response and Slew Rate.

5-290

1/86, Rev. A

-----I1fMD

JM38510/11401111402/11403/11404/11405/11406 JAN JFET-INPUT OPERATIONAL AMPLIFIERS

II
NOTES:
Resistors are ± 1.0% and capacitors are ± 10%, unless otherwise
specified.
2 Precaution shall be taken to prevent damage to the D.U.T during
insertion Into socket and In applying power
3. For device types 01,02,04 and OS, 51 is open, A v =-1 and V,N = 10V.
4. For device types 03 and 06, 51 is closed, Av = -5 and Y,N = 2V.
5. Settling time, ts measured on Pm 5, IS the Interval dunng which the
summing node is not nulled within the specified accuracy referred
to the output.

Figure 3. Test Circuit for Settling Time
BURN-IN
Devices supplied by PMI have been subjected to burn-in per
Method 1015 of MIL-STD-883 using test condition C with
circuit shown on Figure 4 or test condition F using circuit
shown on Figure 5.

+40V

47kD.

OPEN

47k.\1

Figure 4. Test Circuit, Burn-In (Steady-State Power and
Reverse Bias) and Operating Life Test

NOTE
ALL RESISTORS ARE ±20%

Figure 5. Accelerated Burn-In and Life Test Circuit

5-291

1/86, Rev. A

J~38510/13501/13502
ULTRA-LOW OFFSET VOLTAGE
OPERATIONAL AMPLIFIERS
Precision Monolithics Inc.

FEATURES
•
•
•
•

Low Vos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .• 25!N
Low Vos Drift ............................. 0.61'V/oC
Low Noise ................................ 0.61' Vpop
Wide Supply Voltage Range ............. ±4.5V to ±20V

MILITARY DEVICE TYPE

GENERIC-INDUSTRY TYPE

01

OPO?A

02

OPO?

For an 833-processed device with improved electrical specifications, review the OP-07 data sheet.

ORDERING INFORMATION
JAN SLASH SHEET

PMI DEVICE

JM38510/13501 BPB
JM38510/13502BPB

OPO? AZ2/38510
OPO?Z2/38510

JM38510/13501 BGC
JM38510/13501 BGA

OPO? AJ1/38510
OPO? AJ3/38510

JM38510/13502BGC

OPO? J1/38510

PIN CONNECTIONS

TR'EM
Vas :R~:
S

Vas

-IN 2

GOUT

+IN 3

5NC
4

GENERAL DESCRIPTION
8-PIN HERMETIC DIP
TIN-REFLOW LEAD TYPE
(Z2-Suffix)

This data sheet covers the electrical requirements for a monolithic, low offset voltage, internally-compensated operational
amplifier as specified in MIL-M-38510!135 for device type 01 and
02. Devices supplied to this data sheet are manufactured and
tested at PMI's MIL-M-38510 certified facility and are listed in
QPL-38510.

v-

(CASE)

TO-99
GOLD-PLATE LEAD TYPE
(J1-Sufflx)
SOLDER-DIPPED LEAD TYPE
(J3-Suffix)

Complete device requirements will be found in MIL-M-38510
and MIL-M-38510!135 for Class B processed devices.
POWER AND THERMAL CHARACTERISTICS
Case
Oulline

GENERIC CROSS-REFERENCE INFORMATION
This cross-reference information is presented for the convenience of the user. The generic-industry types listed may not
have identical operational performance characteristics across
the military temperature range or reliability factors equivalent to
the MIL-M-38510 device.

Package

Maximum Allowable Maximum Maximum
Power Dissipation
°JC
°JA

P

Dual-In-Line 208mW @ TA

= 125'C

50'C/W

120°C/W

G

8-Lead CAN 16?mW @ TA

= 125'C

60'C/W

150'C/W

SIMPLIFIED SCHEMATIC
V+o---~----------~----------.-----~~-.----~~----~------~-+--+---~

7

R7
*NOTE
R2A AND RlB ARE
ELECTRONICALLY

ADJUSTED ON CHIP
AT FACTORY,
R9

OUTPUT
6
NONINVERTING
INPUT

R,O

INVERTING
INPUT

4

v-o---~--------------------~--~----~--~--------~--~--~

5-292

1/86, Rev. A

~~~~~~~-t~ JM38510/13501/13502 ULTRA-LOW OFFSET VOLTAGE OPERATIONAL AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS

RECOMMENDED OPERATING CONDITIONS

Supply Voltage (Vce) ••.••.••.•.•.•..•.•....••..• ±22V
Input Voltage Range (V,NI .......•.•.••••.•..•...• ±VCC
Differential Input Voltage Range •.•..•.•.•.•..••..• ±30V
Output Short-Circuit Duration (Note 1 I
Lead Temperature (Soldering, 60 sec) ..•.•.•••... +300°C
Storage Temperature Range .••...•.•..•. -65°C to +150°C
Junction Temperature (TJI ..•..•.•....•.•.•...•. +150°C
Maximum Power Dissipation (Pol (Note 2) ••.••.•• 500mW

Supply Voltage Range •••..••.••..••...••. ±4.5V to ±20V
Ambient Temperature Range .•••..••..•• -55°C to +125°C

NOTES:
1 Output may be shorted to ground Indefinitely at Vs = ±15V, TA = 25°C
Temperature and/or supply voltages must be limited to ensure diSsipation

rating

IS

not exceeded

Maximum power diSSipation versus ambient temperature

ELECTRICAL CHARACTERISTICSat±4.5V:S Vcc:S±20Vand -55°C:STA:S 125°C, Rs=50n unnulled, unless otherwise noted.
01 LIMITS
PARAMETER

SYMBOL

Input Offset Voltage

V,o

Input Offset Voltage

:,v,ol.n

Temperature Sensitivity

+1 18

CONDITIONS

TA = 25°C

TA = 25°C

MIN

-liB

Input Offset Current

Power Supply Rejection Ratio

"0

TA = 25°C

02 LIMITS
MIN

MAX

UNITS

-75
-200

75
200

~V

-06

06

-13

13

~VloC

(Note 1)

-2
-4

-3
-6

3

4

(Note 1)

-2
-4

4

-3
-6

3
6

(Note 1)

-2
-4

4

-28
-56

28
56

+PSRR

10

10

-PSRR

+Vee = 15V, -Vee = -20V to -5V
TA = 25°C

10

10

+PSRR

+Vee = 20V to 5V. -Vee = -15V

20

20

-PSRR

+Vee = 15V. -Vee = -20V to -5V

20

20

PSRR

Vee = ±4 5V to ±20V
TA = 25°C

10

10

Vee = ±4 5V to ±20V

20

20

nA

nA

~VIV

Tested at VeM = 0, Vee = ±15V
Due to the Inherent warm-up drift, testing shall occur no sooner than three

(3)

minutes after applicatIOn of power

5-293

.....:l
p..

~

~

0
......

~

~

C5

NOTES:
2

~
......

......

25
60

+Vee = 20V to 5V. -Vee = -15V
TA = 25°C

ge
~

-25
-60

(Notes 1, 2)

Input Bias Current

TA = 25°C

MAX

1/86, Rev. A

--------I~ JM38510/13501/13502 ULTRA-LOW OFFSET VOLTAGE OPERATIONAL AMPLIFIERS

ELECTRICAL CHARACTERISTICSat±4.5V5 Vcc 5±20Vand -55°C5TA 5125°C, Rs=500unnulled, unless otherwise noted.
(Continued)
01 LIMITS
PARAMETER
Common-Mode Rejection Ratio

SYMBOL

CONDITIONS

VCM

~

±13V, TA

VCM

~

MIN
~

25'C. Vcc

~

±15V

MAX

02 LIMITS
MIN

110

110

±13V

106

106

Via Adj 1+1

TA = 25°C I Note 11

05

Via Adj 1-1

TA = 25°C I Note 11

MAX

CMAA

dB
05

Adjustment for Input Offset

Output Short-Circuit Current

mV

'051+1

t:s; 25ms INote 1, 3)

1081 - 1

t~

Icc

Output Voltage SWing (Mlnlmuml

Vap

(Single-Ended 1
Slew Aate

Input NOise Voltage DenSity

Avs

~

Voltage

enp~p

-65
65

25'C

65

(Note 1)

5

AL

~

1kll, INote 11

-10

10

-10

10

RL

~

2kll. I Note 11

-12

12

-12

12

TA

~

25'C

300
200

rnA

V

fa
fa

~

~
~

fa~

Low Frequency Input NOise

-65

-0.5
rnA

SA(+I. SA(-I V,N

e,

-05

25ms (Note 1, 3)

TA

Supply Current

Open Loop Voltage Gain

UNITS

f

~

10V. TA

10Hz
100Hz
1kHz

(Note 21
~

25'C. I Note 11

TA = 25°C, I Note 1)

0 1Hz to 10Hz, TA

~

25'C, INote 11

200
150

08

VlmV

08

VlMS

18
14
12

18
14
12

nV/VHZ

06

0.6

MVp_p

NOTES:
Tested at VCM ~ 0, Vee ~ ±15V
VauT ~ 0 to +10V for Avsl+1 and VauT ~ 0 to -10V for AVS'-' AL ~ 2,0001l
3 Continuous short-Circuit lImits are considerably less than the indicated test lImits, since maximum power diSSipation cannot be exceeded.

5-294

1/86, Rev. A

JM38510/13503
LOW-NOISE PRECISION
OPERATIONAL AMPLIFIER
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•

Low VOS' ...•.••.....................•...... 25!-'V
Low VOS Drllt ............................. 0.6!-,VloC
High Speed ••.....•.........•........•.... 1.7V1!-,s
Low Noise .••..•..•.........•..•..•..•... 0.18!-'Vp_p
High Gain ..•...•..•...•................ 1.0 Million
Wide Supply Voltage Range .•••.•••••.• ±4.5V to ± 18V

MILITARY DEVICE TYPE

GENERIC-INDUSTRY TYPE

03

OP27A

For an 833-processed device with improved electrical specifications, review the OP-27 data sheet.

PIN CONNECTIONS

ORDERING INFORMATION
JAN SLASH SHEET

.~:.:~:~

PMI DEVICE

J M3851 0/13503B PC

OP27 AZ2/38510

J M3851 0113503BGC

OP27 AJ1/38510

+INV,'NC
4 V- (CASE)

GENERAL DESCRIPTION
This data sheet covers the electrical requirements for a monolithic, low offset vOltage, internally-compensated operational
amplifier as specified in MIL-M-38510/135 for device type 03.
Devices supplied to this data sheet are manufactured and
tested at PMI's MIL-M-38510 certified facility and are listed in
QPL-38510.

8-PIN HERMETIC DIP
TIN REFLOW LEAD TYPE

TO-99
GOLD PLATE LEAD TYPE

(Z2-Sullix)

(J1-Suffix)

POWER AND THERMAL CHARACTERISTICS
Case
Outline

Complete device requirements will be found in MIL-M-38510
and MIL-M-38510/135 for Class B processed devices

GENERIC CROSS-REFERENCE INFORMATION

Package

Maximum Allowable Maximum Maximum
Power Dissipation
BJC
BJA

P

Dual-In-Line 20BmW @ TA ; 125°C

50°C/W

120°C/W

G

8-Lead CAN 167mW@TA ; 125°C 60°C/W

150°C/W

This cross-reference information is presented for the convenience of the user. The generic-industry types listed may not
have identical operational performance characteristics across
the military temperature range or reliability factors equivalent to
the MIL-M-38510 device.

SIMPLIFIED SCHEMATIC
r-------~--------~--------~--~----~------._------~_.--~~~v+
R3

R4

Vos Ad)

R"
OUTPUT
NONINVERTING

INPUT (o:+I-+-<~~-+---fi::,--~
INVERTING
INPUT

t-J

~4-~~~

________

~

______

~

* R1 &. R2 ARE PERMANENTL V ADJUSTED
AT WAFER reST
L---------------~--

5-295

__ ______ __________
~

~

~~--~~~~v-

1/86, Rev. A

---------I!EM:D

JM38510/13503 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

ABSOLUTE MAXIMUM RATINGS

RECOMMENDED OPERATING CONDITIONS

Supply Voltage (VCC) •.•.•.••••••..•..••••...•..• ±22V
Input Voltage Range (VIN) ...•.....•....•.•....•.• ±Vcc
Differential Input Voltage Range ....•..••.•..•.•... ±0.7V
Output Short-Circuit Duration (Note 1)
Lead Temperature (Soldering. 60 sec) ...•........ +300·C
Storage Temperature Range .....•.•..... -65·C to + 150·C
Junction Temperature (TJ ) •••••••••••••••••••••• +150·C
Maximum Power Dissipation (Po) (Note 2) .....•.. 500mW

Supply Voltage Range .••••...•..•......•. ±4.5V to ±18V
Ambient Temperature Range .•.•.••..... -55·C to +125·C

NOTES:
1 Output may be shorted to ground ,ndeflnitely at Vs = ±15V. TA = 25'C
Temperature and/or supply voltages must be limited to ensure dissipation
rating Is not exceeded.
2. Maximum power dissipation versus ambient temperature.

ELECTRICAL CHARACTERISTICS at ±4.5V:5 VCc:5 ±20V and -55·C:5 TA:5125·C. Rs = 50n unnulled. unless otherwise noted.
03 LIMITS
PARAMETER

SYMBOL

Input Ollset Voltage

v,o

Input Ollset Voltage
Tempersture Sensitivity

_W,o/·H

+I'B

CONDITIONS

MIN

MAX

UNITS

TA = 25'C

-25
-60

25
60

p.V

-06

0.6

p.VI'C

(Note 1)

-40
-60

40
60

(Note I)

-40
-60

40
60

(Note 11

-35
-50

35
50

TA = 25'C

(Notes 1. 2)

nA

Input Bias Current
-I'B
Input Ollset Current

Power Supply AeJecllon Aatlo

TA = 25'C

TA = 25'C
1'0
+PSRA

+Vee = 18V to 5V. -Vee = -15V
TA = 25'C

10

-PSAA

+Vee = 15V, -Vee = -18V to -5V
TA = 25'C

10

+PSAA

+Vee = 18V to 5V, -Vee =-15V

16

-PSAA

+Vee = 15V, -Vee = -18V to -5V

16

Vee = ±4 5V to ±18V
TA = 25'C

10

PSAA

Vee = ±4 5V to ±18V

16

nA

p.V/v

NOTES:
1. Tested at VeM = 0, Vee = ±15V
2. Due to the inherent warm-up drift, testing shall occur no sooner than three (3) minutes after appllcallon of power

5-296

1/86, Rev. A

----------I~ JM38510/13503 LOW-NOISE PRECISION OPERATIONAL AMPLIFIER

ELECTRICAL CHARACTERISTICS at ±4.5V::; Vee::; ±20V and -55°C::; TA ::; 125°C, Rs = 500 unnulled, unless otherwise noted.
(Continued)

03 LIMITS
PARAMETER
Common-Mode Rejection Ratio

SYMBOL

114

Via Ad] (-)
1081 + 1

t,; 25ms, (Note 1,3)

-50

1081 - 1

t,; 25ms, (Note 1,3)

eMRR

Via Adj (+)

Supply Current

MIN

= ±11V, TA = 25'e, Vcc = ±15V
VCM = ±11V, VCC = ±15V
TA = 25'e, (Note 1)
TA = 25'e, (Note 1)

Adjustment for Input Offset

Output Short-CircUit Current

CONDITIONS

Icc

VCM

VoP

TA

= 25'e

mV
-05

4

= 500!!. (Note 11
RL = 2k!!. (Note 11
= 25'e

Slew Rate

SR(+I, SR(-I Y,N

Input NOIse Voltage DenSity

en

fa
fa
fa

e np _p

f = 0 1Hz to 10Hz
TA = 25'e, (Note 11

In

fa
fa
fa

Input NOise Current Density

0.5

(Note 11

Avs

Voltage

dB
108

rnA

Open Loop Voltage Gain
(Single-Ended I

Low Frequency Input NOise

UNITS

70

-10

RL

Output Voltage Swing (Minimum)

MAX

TA

-115

(Note 2)

= 10Hz
= 100Hz
= 1kHz

TA

TA

10

V

= 10V, TA = 25'e, (Note 11

= 10Hz
= 100Hz
= 1kHz

rnA

115

1000
500

V/mV

17

V/~s

5.5
40
3.8

nV/VHZ

018

J.l.Vp.p

40
15
08

pA/VHZ

= 25'e, (Note 11

= 25'e, (Note

11

II
~
~
......
.....

~
p..,

~

~

0
......

~
~

p..,

NOTES:

0

1.
2

Tested at VCM = 0, Vcc = ±15V
Your = a to +10V for Avs (+) and Your

3.

Continuous short-circuit limits are considerably less than the Indicated test limits, Since maximum power diSSipation cannot be exceeded

= 0 to -10V for AvsH

RL

= 2,000!!

5-297

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

INSTRUMENTATION
AMPLIFIER
Prcclsion MOllolithics Inc.

6-3

Introduction

6-3

Definitions

6-5

AMP-Ol
Low-Noise Precision
Instrumentation Amplifier

6-27

AMP-OS
Fast-Settling JFET
Instrumentation Amplifier

8-2

INSTRUMENTATION
AMPLIFIER
P . . ecIsion MouoiIthics Inc.

INTRODUCTION
An instrumentation amplifier is a committed
gain block that amplifies a differential input voltage by a precisely set gain. Voltages common
to both inputs are rejected. Differential gain is
set by one or two external resistors, usually over
a range of 1 to 1000. Instrumentation amplifiers
are designed to have very high input impedance;
this assures that the gain will not be affected by
signal-source impedances (Rs). Input bias current must be low to minimize input offset voltages due to 18 X Rs. In the output stage, low
output impedance keeps the output voltage from
being affected by the load impedance. Instrumentation amplifiers employ heavy negative
feedback which provides excellent gain linearity
even at high gains.
Most instrumentation amplifiers have an output
sense pin (SENSE) and a reference input pin
(REFERENCE). As shown in Figure 1, the load is
usually connected between SENSE and REFERENCE points. The amplifier will make VOUT =
GVIN despite voltage drops between OUTPUT
and LOAD, or from REFERENCE to GROUND.
The SENSE and REFERENCE inputs are particularly useful for driving remote loads with high
currents. The essential characteristics of
instrumentation amplifiers-high input impedance, low output impedance, low offset, high
linearity, stable gain, and ability to reject
common-mode inputs-make them very useful
for amplifying low-level transducer outputs.
Transducers such as thermocouples, straingage bridges, biological probes, and current
shunts produce small differential signals superimposed on common-mode bias voltages. In
addition, common-mode ground noise is usually
prevelant. Instrumentation amplifiers are gain
blocks that have been optimized for preamplifying low-level transducer signals in the presence
of common-mode noise.
The PMI AMP-01 instrumentation amplifier has
all the features needed for use in high-accuracy
data-acquisition systems and high-performance
instruments:
• Wide Gain Range (0.1:5 G :510,000)
• High Input Impedance

• Low Input Bias Current
• Low Offsets
• Excellent Linearity
Unlike conventional instrumentation amplifiers,
the AMP-01 has high output drive capability; it
can supply ±10V at ±50mA. This enhanced output drive capability enables the AMP-01 to drive
unusually large capacitive loads without encountering stability problems.
Gain of the AMP-01 is set by the ratio of two
external resistors according to:
- (20 X RSCALE)
VOUTR
VIN
GAIN

Output sense and reference pOints are provided.
The AMP-01 is unusually versatile, and can be
connected as a precision current source or
high-performance op amp as well as a conventional instrumentation amplifier.
The AMP-05 JFET instrumentation amplifier
supports high-speed applications, such as
analog-multiplexed data acquisition and fast
analog signal proceSSing. The design offers a
1010'S maximum settling-time to 12 bits at gains
up to 1000, and with 14-bit linearity. The AMP-05
also provides on-board circuits for guard
driving, which maximizes input signal speed,
and a precision current source for transducer or
reference excitation.

AMP-05 features include:
• Input bias current, 25pA max.
• High gain-bandwidth product, 200M Hz
• 6V/Io's output slew-rate with a 1000pF load
• High common-mode rejection, 110dB min.

DEFINITIONS
Offset at the output of an
instrumentation amplifier consists of two terms,
a gain-dependent input-offset-voltage and a
gain-independent output-offset-voltage. Total
offset is the sum of the unity-gain-output-offset
(Voos) plus input-offset (VIOS) multiplied by the
gain (Output Offset = Voos + GVIOS). At high
gain, the input offset term dominates. For the
AMP-01 and AMP-05, both input and output
offsets can be trimmed externally if desired .
Voltage Offsets -

~
~
~

::s
~
:..(

Z
0

~

f-i

§
~

~

(-<
v:l

Z
t--I

INSTRUMENTATION
AMPLIFIER
Precision Monolithics Inc.

Power Supply Rejection - Offset changes with
variations in the power supply voltages. The
ability of the instrumentation amplifier to reject
fluctuations in power supply voltage is referred
to as "power supply rejection". It varies with
gain and is different for the positive and negative supplies. The offset change referred-toinput (RTI) is usually specified in dB form. For
example, a PSR of 100dB at a gain of 1000 would
imply an input-offset-voltage change of lO#-,Vper-volt of power supply change. The output
offset change-per-volt of power-supply change
would be 10mV. PSR in the specification tables
is measured at DC.

to reject common-mode inputs. The ratio of
change in output voltage to a change in
common-mode input voltage is the commonmode gain (/1VO//::",vCM). The ratio of differential gain (G) to common-mode gain (ACM) is
defined as common-mode rejection ratio
(CMRR). The CMR is conventionally specified in
log form; CMR = 20 10glO CMRR.

Input Bias Current - The input bias currents
are currents flowing into (or out of) the two
inputs of the amplifier. The value given in the
specification table is the maximum current into
either input. Input offset current is the difference between the two input bias currents.

As an example, consider a CMR of 120dB at a
gain of 1000 with a common-mode input range
of ±10V. The 120dB of CMR implies a CMRR of
1000/ACM = 1,000,000, or a common-mode gain
of 1/1000. A ± 10V common-mode input will
cause an output change of ± 10mV for this
example (CMR = 120dB, G = 1000).

Since instrumentation amplifiers are designed to
amplify differential signals while rejecting
common-mode inputs, common-mode gain
stays essentially independent of gain setting.
Therefore, CMRR increases almost directly with
the gain setting.

Input Voltage Range - The linear operating
range of the amplifier is referred to as the "input
voltage range". When operating at high gains
with small differential inputs, this input range is
the common-mode input voltage range.

Gain Equation Accuracy - Differential gain is
given as a function of the two external resistors.
For the AMP-01 and AMP-OS, the relationship is
ideally 20 x Rs/RG. The specified accuracy
limits indicate the accuracy of the amplifier
given an exact ratio of Rs/RG.

Common-Mode Rejection - Common-mode
rejection (CMR) specifies the amplifiers ability

INSTRUMENTATION AMPLIFIER FUNCTIONAL DIAGRAM

SIGNAL SOURCE
SENSE
I

OUTPUT

VCM

RSCALE

":" GROUND

6-4

AMP. 01
LOW-NOISE
PRECISION INSTRUMENTATION AMPLIFIER
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•

PIN CONNECTIONS

Low Offset Voltage ......................... 50/LV Max
Very Low Offset Voltage Drift ........... 0.3/LV/oC Max
Low Noise. . . . . . . . . . . . . . . . .. 0.12/LVp _p (0.1 Hz to 10Hz)
Excellent Output Drive ............... ±10V at ±50mA
Capacitive Load Stability ...................... to 1/LF
Gain Range ............................ 0.1 to 10,000
Excellent Linearity ................. 16-Bit at G = 1000
High CMR ...................... 125dB Min (G = 1000)
Low Bias Current ........................... 3nA Max
May be Configured as a Precision Op-Amp
Output-Stage Thermal Shutdown

18-PIN HERMETIC DIP
(X-Suffix)

I

'Make no electrical connection to these pms

ORDERING INFORMATIONt
PACKAGE
CERDIP
18-PIN
AMP01AX'
AMP01BX'
AMP01EX
AMP01FX

LCC

z
'i'

OPERATING
TEMPERATURE
RANGE

AMP01BTC/883

MIL
MIL
IND
IND

Voos NULL

6

NC

,

AMP-01 BTC/883
28-LEAD LCC
(TC-Suffix)

"For devices processed In total compliance to MIL-STD-BS3, add /883 after
part number Consult factory for 883 data sheet
t All commercial and industrial temperature range parts are available with
burn-In For ordering Information see 1986 Data Book, Section 2

SIMPLIFIED SCHEMATIC
r-----------~~-------------------------Ov+

r----------O+voP

>-------0 OUTPUT
-IN

o----"o/W----------t-------------j-------,

+IN

O---¥,Iy--.---[,

'-----------0 -VOP

R3

47Skn
I--~-~M.--O SENSE

R4

25kn

'---------~------~----------~_----_ov-

Manufactured under the following U S. patents 4,471.321 and 4,503,381

6-5

1/86, Rev. A

---------;[fM!) AMP-G1 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER
c. The current-feedback design is immune to CMR degradation when series resistance is added to the reference
input. A small (trimmable) offset change results from
added resistance, e.g. a printed circuit track.

GENERAL DESCRIPTION
The AMP-01 is a monolithic instrumentation amplifier
designed for high-precision data acquisition and instrumentation applications. The design combines the conventional
features of an instrumentation amplifier with a high-current
output stage. The output remains stable with high capacitance loads (1I'F), a unique ability for an instrumentation
amplifier. Consequently, the AMP-01 can amplify low-level
signals for transmission through long cables without requiring an output buffer. The output stage may be configured as a
voltage or current generator.
.

The AMP-01 utilizes low-drift thin-film resistors to minimize
output offset temperature drift. A feedback voltage-to-current
converter is employed having high linearity and low noise,
particularly at low frequencies. Parameter shifts during
packaging are eliminated by a post-assembly trimming
technique which electronically adjusts the output offset
voltage.

Input offset voltage is very low (20I'V) which generally
eliminates the external null potentiometer. Temperature
changes have minimal effect on offset; TCVlos is typically
0.15I'VI·C. Excellent low-frequency noise performance is
achieved with a minimal compromise on input protection.
Bias current is very low, less than 10nA over the military
temperature range. High common-mode rejection of 130dB,
16-bit linearity at a gain of 1000, and 50mA peak output
current are achievable simultaneously. This combination
takes the instrumentation amplifier one step further towards
the ideal amplifier.

The AMP-01 input transistors Q1 and Q2 feed active loads,
yielding stage gain in excess of 4000 (see simplified
schematic) The output amplifier, A1, is a two-stage design
having a gain of about 50,000 driving a 1000 load. Overall
gain of 2 x 108 yields excellent linearity, even at high
closed-loop gains.

AC performance complements the superb DC specifications.
The AMP-01 slews at 4.5V1l's into capacitive loads of up to
15nF, settles in 50l's to 0.01% at a gain of 1000, and boasts a
healthy 26MHz gain-bandwidth product. These features
make the AMP-01 ideal for high-speed data-acquisition
systems.

Superbeta transistors use a new transistor geometry resulting in an input noise of only 5nVly"HZ at G = 1000. Noise
includes contributions from the gain-setting resistor and
internal overload-protection resistor. The input stage
achieves an offset voltage drift of less than 0.3I'VI·C
(E Grade).

Gain is set by the ratio of two external resistors over a range
of 0.1 to 10,000. A very low galn-temperature-coefficlent of
10ppml"C is achievable over the whole gain range. Output
voltage swing is guaranteed with three load resistances; 500,
5000, and 2kO. Loaded with 5000, the output delivers ± 13.0V
minimum. A thermal shutdown circuit prevents destruction
of the output transistors during overload conditions.

The AMP-01 uses a unique two-pole compensation scheme
where the load capacitance is incorporated into the dominate
pole. Stable operation results even with high capaCitance
loads. The high output current capability (90mA peak) allows
the 4.5V1l's slew-rate to be maintained with load capacitance
as high as 15nF.

The AMP-01 can also be configured as a high-performance
operational amplifier. In many applications, the AMP-01 can
be used in place of op-amp/power-buffer combinations.

ABSOLUTE MAXIMUM RATINGS (Note 2)

Low bias current is achieved by using lon-implanted superbeta transistors combined with a new bias-current cancellation system, patents applied for. Input bias current remains
below 10nA over the military temperature range, -55·C
to +125·C.

Supply Voltage ................................... ± 18V
Internal Power Dissipation (Note 1) ...•.......... 500mW
Common-Mode Input Voltage ...•...•.... Supply Voltage
Differential Input Voltage, RG2: 2kO •....••...•..... ±20V
RG< 2kO ................ ±10V
Output Short-Circuit Duration •............••.• Indefinite
Storage Temperature Range. • . . . • . . . .. -65· C to + 150· C
Operating Temperature Range
AMP-01A, B ........................ -55·Cto+125·C
AMP-D1E, F ......................... -25·Cto+85·C
Lead Temperature (Soldering, 60 sec) .•......•..•. 300· C
DICE Junction Temperature (Ti) .•.•... -65· C to + 150· C

THEORY OF OPERATION
An instrumentation amplifier, unlike an op amp, requires
precise internal feedback. The two techniques presently in
use are resistive and current feedback.
The AMP-01 employs the current feedback approach which
has significant advantages over resistive feedback. Advantages of current-feedback are:
a. The technique yields a very high common-mode rejection ratio. The AMP-01 CMR is in excess of 130dB at a
gain of 1000.
b. The gain of the current feedback design is set by the
ratio of two external resistors. Using external resistors
allows any practical gain to be set with high precision
and very low gain temperature coefficient.

PACKAGE TYPE

18-Pln Hermetic DIP (X)

MAXIMUM AMBIENT
TEMPERATURE FOR
RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

100'0

10mW/'C

NOTES:

1. See table for maximum ambient temperature rating and derating factor.
2. Absolute ratings apply to both DICE and packaged parts. unless otherwise
noted.

6-6

1/86, Rev. A

[fMDAMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, Rs = 10k!l, RL = 2k!l,

TA

= 25° C, unless otherwise noted.

AMP-01A
PARAMETER

AMP-01B

TYP

MAX

TA = 25'C
-55'C" TA" +125'C

20
40

SYMBOL

CONDITIONS

V,OS

MIN

MIN

TYP

MAX

UNITS

50
80

40
60

100
150

0.3

0.3

1.0

/"V/'C

OFFSET VOLTAGE

I nput Offset Voltage

/"V

TCV ,OS

-55'C" TA " +125'C

0.15

Output Offset Voltage

Voos

TA =25'C
-55'C" TA" +125'C

1
3

3
6

2
6

6
10

mV

Output Offset Voltage Drift

TCVoos

RG = c (Note 3)
-55'C" TA " + 125'C

20

50

50

120

/"VI'C

Input Offset Voltage Drift

Offset Referred to Input
vs. Positive Supply
V+ = +5V to +15V

PSR

Offset Referred to Input
vs. Negative Supply
V- = -5V to -15V

G = 1000
G=100
G= 10
G=1

120
110
95
75

130
130
110
90

110
100
90
70

120
120
100
80

dB

-55'C" TA " +125'C
G = 1000
G= 100
G= 10
G= 1

120
110
95
75

130
130
110
90

110
100
90
70

120
120
100
80

dB

G=
G=
G=
G=

110
95
75
55

125
105
85
65

105
90
70
50

115
95
75
60

dB

110
95
75
55

125
105
85
65

105
90
70
50

115
95
75

1000
100
10
1

~
~
......

dB

60

Input Offset Voltage Trim
Range

Vs = ±4.5V to ±18V
(Note 1)

±6

±6

mV

Output Offset Voltage Trim
Range

Vs= ±4.5V to ±18V
(Note 1)

±100

±100

mV

1
4

la

TA =25'C
-55'C" TA " +125'C

Input Bias Current Drift

TCl a

-55'C" TA" +125'C

40

TA = 25'C
-55'C" TA ,,+125'C

0.2
0.5

TClos

-55'C"TA ,,+125'C

3

Input ReSIstance

R'N

DifferentIal, G = 1000
Differential, G" 100
Common-Mode, G = 1000

Input Voltage Range

IVR

TA = 25' C (Note 2)
-55'C" TA" + 125'C

Input Offset Current

los

Input Offset Current Drift

3
10

6

1.0
3.0

0.5
1.0

6
15

50

nA
pAl'C

2.0
6.0

nA
pAl'C

INPUT

Common-Mode.

CMR

Rejection

10
20
±10.5
±10.0

±10.5
±10.0

V

125
120
100
85

130
130
120
100

115
110
95
75

125
125
110
90

dB

-55°C S TA~ +125°C
G=I000
G=100
G=IO
G=I

120
115
95
80

125
125
115
95

110
105
90
75

120
120
105
90

dB

2.

Refer to section on common-mode rejection

AMP-01AX specIficatIon changed from lOOI"V/'C to 50/"V/'C, effective
January, 1985

respectively.

6-7

.--

Gfl

VCM = ±10V, lkfl
source imbalance
G = 1000
G = 100
G= 10
G=1

NOTES:
I. Vms and Voos nulling has minimal affect on TCV,DS and TCVDDS,

~----

10
20

-~--~~~--.----

--------

~
p..

Z

0

~

~
~

:E

~

INPUT CURRENT

Input Bias Current

~

~

PSR
-55'C" TA" +125'C
G = 1000
G= 100
G= 10
G=1

II

1/86, Rev. A

G;
Z
......

[fM!)

AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs= ±1SV, Rs= 10kO, RL = 2kO,

TA =

2SoC, unless otherwise noted.

AMP-01E
PARAMETER

MIN

SYMBOL

CONDITIONS

Input Offset Voltage

V,OS

TA =2S"C
-2S"CSTA S+85"C

Input Offset Voltage Drift

TCV ,OS

-2S"CSTA S+85"C. (Note 21

Voos

TA = 2S"C
-25"CS TAS +85"C

TCVoos

Ra = ... (Note 2)
-2S"CSTA S+85°C

AMP-01F

TYP

MAX

20
40

MIN

TYP

MAX

50
80

40
60

100
150

0.15

03

0.3

1.0

~VI"C

3

3
6

2
6

6
10

mV

20

100

50

120

~VI"C

UNITS

OFFSET VOLTAGE

Output Offset Voltage
Output Offset Voltage Drift

Offset Referred to Input
vs Positive Supply
V+ = +SV to +lSV

Offset Referred to Input
vs Negative Supply
V- = -SV to -15V

PSR

PSR

G= 1000
G= 100
G=10
G=l

120
110
95
75

130
130
110
90

110
100
90
70

120
120
100
80

-2s oCSTA S+8SoC
G = 1000
G=100
G=10
G=l

120
110
95
75

130
130
110
90

110
100
90
70

120
120
100

G = 1000
G=100
G=10
G=l

110
95
75
55

125
105
85
65

105

115
95
75

-2s oCSTA S+85°C
G = 1000
G= 100
G=10
G=l

110
95
75
55

125
105
85
65

105
90
70
50

~V

dB

dB

80

90
70
50

dB

60
115
95
75
60

dB

Input Offset Voltage Trim
Range

Vs = ±4.5V to ± 18V
(Note 1)

±6

±6

mV

Output Offset Voltage Tnm
Range

Vs =±45Vto±18V
(Note 1)

±100

±100

mV

INPUT CURRENT
Input Bias Current

Ie

TA =25"C
-25°CSTA S+SSoC

Input Bias Current Dnft

4

3
10

2
6

1.0
3.0

0.5
1.0

6
15

nA

TCl e

-25°CSTA S+85°C

40

Input Offset Current

los

TA = 25°C
-25°C 5 TAS +85°C

0.2
0.5

Input Offset Current Drift

TClos

-2s oCSTAS+85°C

3

5

pN°C

Input ReSistance

R'N

Differential. G = 1000
Differential, G 5 100
Common-Mode, G = 1000

10
20

1
10
20

GO

Input Voltage Range

IVR

TA = 25°C (Note 3)
-25°C:5 TA:5 + 85°C

50

pArC
2.0
6.0

nA

INPUT

Common-Mode
Rejection

CMR

±10.5
±10.0

±10.5
±10.0

V

VCM = ±10V, lkO
source Imbalance
G = 1000
G=l00
G=10
G=l

125
120
100
85

130
130
120
100

115
110
95
75

125
125
110

-25°C 5 TAS +85°C
G = 1000
G=l00
G=10
G=l

120
115
95
80

125
125
115
95

110
105

90

120
120
105

75

90

NOTES:
1. V,OS and Voos nulhng has minimal affect on TCV,os and TCVoos,
respectively.

dB

90

dB

Sample tested
3

6-8

Refer to section on common-mode rejection

1/86, Rev. A

IEWAMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

ELECTRICAL CHARACTERISTICS at

Vs

=

± 15V,

Rs

= 10kO, RL = 2kO, TA = 25° C, unless otherwise noted.
AMP-01A1E

PARAMETER

SYMBOL

MIN

CONDITIONS

AMP-01B/F

TYP

MAX

03

0.6

0.0007

MIN

TYP

MAX

UNITS

0.5

0.8

%

10k

VIV

0.005
0.005
0.005
0.010

00007

0005
0005
0.007
0.015

%

10

5

15

ppm/oC

GAIN

G ~ 20 x Rs
RG

Gam Equation

Accuracy

Accuracy Measured

from G ~ 1 to 1000
Gain Range

01

G
G ~ 1000
G~ 100
G~10 (Note1)

Nonlinearity

10k

G~1

Temperature Coefficient

1 S G S 1000
(Notes 1, 2)

GTe

5

01

OUTPUT RATING

II

±130
±13.0
±25

±13.8
±13.5
±40

±13.0
±13.0
±2.5

±138
±13.5
±4.0

V

Over Temp.
(Note3)

±120
±12.0

±138
±13.5

±12.0
±12.0

±13.8
±13.5

V

~
......

Positive Current limit

Output-to-Ground Short

60

100

120

60

100

120

rnA

Negative Current Limit

Output-to-Ground Short

60

90

120

60

90

120

rnA

::sp..

Capacitive Load Stability

1SGS1000
No Oscillations, (Note 1)

01

Output Voltage
Swing

RL ~ 2kO
RL ~ 5000
RL ~ 500

V OUT

RL ~ 2kO
RL ~5OOO

Thermal Shutdown

!,F
165

°C

G~100
G~

10

G~1

NOise Current Density, RTI

fO~

In

1kHz, G ~ 1000

10
59
540

10
59
540

nV/yHZ

0.15

0.15

pAlyHZ

G ~ 1000
G~ 100

9 np _p

012
016
14
13

G~10
G~1

Input Noise Current

i

0.12
016
1.4
13

o 1Hz to 10Hz, G ~ 1000

ne-e

2

570
100
82
26

G~1

BW

Slew Rate

SR

G~10

ts

To 0 01 %, 20V step
G=1
G=10
G~ 100
G = 1000

Settling Time

e

pAp-

G=10
G= 100
G = 1000
35

4.5
12
13
15
50

3.0

570
100
82
26

kHz

4.5

VI!,s

12
13
15
50

!,s

NOTES:

1.
2.
3.

Guaranteed by design.
Gain tempec does not include the effects of gam and scale reSistor
tempec match.

-55°C STA S + 125°C for AlB grades, -25°C STA S + 85°C for ElF grades

6-9

-

-

----,.~-

-.

-

-

-----_.-----

:E

~

E-<

Z
......
!,Vp.p

DYNAMIC RESPONSE

Smail-Signal
Bandwidth (-3dB)

Z
~

rJ)

o 1Hz to 10Hz
Input NOise Voltage

0

f-'

1kHz
G ~ 1000

fo~

en

~
~

NOISE

Voltage Density, RTI

~

Z

165

Junction Temperature

Temperature

01

ge

----------

1/86, Rev. A

_ _ _ _ _ _ _ _-I~AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

ELECTRICAL CHARACTERISTICS at Vs= ±15V. Rs= 10kO. RL = 2kO. TA = 25°C. unless otherwise noted.
AMP-01A1E
PARAMETER

SYMBOL

CONDITIONS

MIN

AMP-018/F

TYP

MAX

50

65

MIN

TYP

MAX

UNITS

SENSE INPUT
Input Resistance

R'N

Input Current

"N

Voltage Range

35
Relerenced to V-

35

280
-10.5

(Note I)

50

65

280
+15

-10.5

65

35

kO

"A
+15

V

65

kO

REFERENCE INPUT
Input Resistance

R'N

Input Current

liN

Voltage Range

35

50
280

Relerenced to V(Note I)

-10.5

50
280

+15

-10.5

"A
+15

V

VlV

Gain to Output
POWER SUPPLY -25'CS TA S+65'C lor ElF Grades. -55'C S TAS +125'C lor AlB Grades
Supply Voltage Range

Vs

+V linked to +Vop
-V linked to -Vop

Quiescent Current

10

+V linked to +Vop
-V linked to-Vop

±IS

±4.5
3.0
3.4

4.8
4.8

±4.5
3.0
3.4

±18

V

4.8
4.8

rnA

NOTE:
I. Guarantesd by design.

6-10

1/86, Rev. A

---------I~ AMP·en LOW·NOISE PRECISION INSTRUMENTATION AMPLIFIER

DICE CHARACTERISTICS

1. RG
2. RG
3. -INPUT

4. VOOS NULL
5. VoosNULL
6.
7.
8.
9.

TEST PIN"
SENSE
REFERENCE
OUTPUT

10.
11.
12.
13.
14.
15.
16.
17.
18.

V- (OUTPUT)

VV+
V+ (OUTPUT)
Rs
Rs
VIOS NULL
VIOS NULL
+INPUT

"Make no electrical connection

II

For additional DICE information refer to
1986 Dala Book, Section 2.

DIE SIZE 0.110 X 0.148 Inch, 16,280 sq. mils
(2.79 X 3.76 mm, 10.50 sq. mm)

WAFER TEST LIMITS at VS= ±15V, Rs= 10kO, RL = 2kO, TA= 25°C, unless otherwise noted.
PARAMETER

SYMBOL

Input Offset Voltage

VIOS

Output Offset Voltage

Voos

Offset Referred to Input
vs. Positive Supply

PSR

CONDITIONS

V+ ~ +5V to +15V
G ~ 1000
G~ 100
G~10
G~1

Offset Referred to Input
vs. Negative Supply

PSR

Input Bias Current

Ie

Input Offset Current

los

Input Voltage Range

IVR

V- ~ -5V to -15V
G ~ 1000
G ~ 100
G~10
G~

1

AMp·01NBC

AMp·01GBC

LIMIT

LIMIT

UNITS

60

120

I'VMAX

S

mVMAX

120
110
95
75

110
100
90
70

110
95
75
55

105
90
70
50

4

Guaranteed by CMR Tests

dBMIN

dBMIN

nAMAX
3

nAMAX

±10

±10

VMIN

125
120
100
85

115
110
95
75

06

OS

% MAX

±13
±13
±2.5

±13
±13
±2.5

VMIN

VCM~

Common-Mode
Rejection

G~
CMR

±10V
1000

G~100
G~

10

G~1

Gain Equation
Accuracy

Output Voltage Swing

G ~ 20 X Rs
RG
VOUT

RL ~2kn
RL ~ 500n
RL ~ 50n

dB MIN

Output·Current Limit

Output·to-Ground Short

±60

±60

mAMIN

Output-Current Limit

Output-to-Ground Short

±120

±120

rnA MAX

4.S

4.S

rnA MAX

Quiescent Current

10

+ V Linked to + Vop
-V Linked to -Vop

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specificatiOns based on dice lot qualification through sample lot assembly and testing.
6·11

1/86, Rev. A

_ _ _ _ _ _ _ _-lOOAMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

TYPICAL ELECTRICAL CHARACTERISTICS at Vs= ±15V, Rs= 10kO, RL = 2kO,
PARAMETER

SYMBOL

Input Offset
Voltage Dnft

TCV,os

Output Offset
Voltage Drift

TCVoos

Input Bias
Current Drift

Input Offset
Current Drift

CONDITIONS

RG=ao

TCI.

TA =

25°C, unless otherwise noted.

AMP-01NBC

AMP-D1GBC

TYPICAL

TYPICAL

UNITS

0.15

0.30

p.vrc

20

50

p.VloC

40

50

pArC

5

pArC

0.0007

0.0007

%

TClos

Nonlinearity

G = 1000

Voltage Noise Density

en

G = 1000
fO= 1kHz

5

Current NOIse Density

in

G = 1000
fO= 1kHz

0.15

0.15

pAlv'HZ

Voltage Noise

e np _p

G = 1000
o 1Hz to 10Hz

0.12

012

p.Vp_p

Current NOise

inp_p

G = 1000
0.1Hz to 10Hz

2

pA p_p

Small-Signal
Bandwidth (-3dB)

BW

G = 1000

26

26

kHz

Slew Rate

SR

G=10

4.5

4.5

VII'S

Settling Time

ts

To 0.01%. 20V Step
G = 1000

50

50

p.s

6-12

nVlv'HZ

1/86, Rev. A

------------I~ AMP-Ol LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
INPUT OFFSET VOLTAGE
VB TEMPERATURE
50

rv~=±,L

40

~

30

w
"

20

o~
>

-10

~

-20

>-

-

_ TA

,..,.-

10

~
o

INPUT OFFSET VOLTAGE
VB SUPPLY VOLTAGE

V

/ V-

I'"

!25

0

t-v~="L

C

~

1-

I""':

r--..

"-

..............

-1

3-

-2

4

-3

........ r-....

r--...

-4

-40
-75 -50 -25

-6
0

25

50

75

lOa 125

o

150

TEMPERATURE (Oe)

\

10
05
00

::20

±15

:,:25

1

25

50

75

±15

±lQ

'jblgb

G

v~~"= 2V p p
Vs == ±15V

120

130

T~I~}5°C
G'" 100

100

V

02
120

±20

COMMON-MODE REJECTION
VB FREQUENCY
140

~5'='~;~~

04

V

±5

POWER SUPPLY VOLTAGE (VOLTS)

TA == 2S"C

I"'-.. r-....

o

100 125 150

COMMON-MODE REJECTION
VI VOLTAGE GAIN
140

"~V

-15

0

TEMPERATURE (oC)

06

V-

-

-10

-2
-75 -50 -25

INPUT OFFSET CURRENT
VI TEMPERATURE
V5 =

0

-05

POWER SUPPL Y VOLTAGE (VOL lS)

08

1~

_T)25 C

05

1

±10

==

~

V

r'-...

1

~

10

/

\

25

INPUT BIAS CURRENT
VB SUPPLY VOLTAGE

J
\

0

TEMPERATURE (OC)

15

-05

o

-5_ _ -25

20

1\

15

±20

±15

V~ = "~V

TA ='25°C

-10

±lQ

INPUT BIAS CURRENT
VI TEMPERATURE

25
20

±5

POWER SUPPLY VOLTAGE (VOLTS)

OUTPUT OFFSET VOLTAGE
CHANGE VB SUPPLY VOLTAGE

-02

r-....

2

-4

-30

00

r-....

UNIT No

..........

-2

OUTPUT OFFSET VOLTAGE
VB TEMPERATURE

80
G

60

==

10

~
G == 1

40

~

110
20

-04

-06
-75 -50 -25

100
0

25

50

75

TEMPERATURE (OC)

100 125 150

/
1

0
10

100
VOL TAGE GAIN, G

6-13

1k

10k

1

10

100

1k

10k

lOOk

FREQUENCY (Hz)

1/86, Rev_ A

II

---------;m

AMP·01 LOW·NOISE PRECISION INSTRUMENTATION AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
COMMON·MODE VOLTAGE
RANGE VB TEMPERATURE
16

POSITIVE PSR
FREQUENCY

140

V~M .10

J±,L .....

14
12
10

11111

I I

100

Vl",L

80

I I

60

80
60

I

20

20

25

50

o
75

100

10

1

125 150

1k

100

+1V

I'

G=1

lOOk

100

10

1k

10k

lOOk

CLOSED·LOOP OUTPUT
IMPEDANCE VB FREQUENCY

I~S.' ±\J~I
RL = 2kl1

'"

14
12

0,

FREOUENCY (Hz)

30

V~.I ±I,~~

25

"~

10k

MAXIMUM OUTPUT SWING
vs FREQUENCY

20

/

10

Vs = :t15V

TA = 2SoC

avs

G=1

MAXIMUM OUTPUT VOLTAGE
VB LOAD RESISTANCE

w

~ ~'ig'

~

FREQUENCY (Hz)

16

100

40

0

=1000

G - 100

1\

TEMPERATURE (OC)

18

~

G
120

ll.VS = :tlV

J±5l
o

~0

140

TA = 2SoC

40

I

-50 -25

IIIII I

G = 100

G=1

VB

1e~'~ ±;J~I

111111

0'1000

120

I I

~75

NEGATIVE PSR
FREQUENCY

VB

15

0

>
~

"
"0

10

~

o

10

100

1K

10K

o

100

LOAD RESISTANCE In)

60

TOTAL HARMONIC
DISTORTION
VB LOAD RESISTANCE
0.02

I

I 11111111
007 _ Vs'" !15V
RL =600n
006 :- Your = 20V p-p

W~II,J

20

1M
FREQUENCY (Hz)

008

W~~,JJ

40

100k

TOTAL HARMONIC
DISTORTION
VB FREQUENCY

V~I~I ±1SV
TA = 2SOC

W~~,JJo

10k
FREQUENCY (Hz)

CLOSED·LOOP VOLTAGE
GAIN VB FREQUENCY
80

1k

1\

GJoJo
I

004

001

V

003

0-1

~

G ;"10,
002

1/ tl'dl~~

-20
001
-40

10

,

005

W~I,

1

vl. Lt III
G= 100
f = 1KHz
VOUT = 20Vp-p

100

lk

10k

FREQUENCY (Hz)

lOOk

1M

o

.......

~

10

100
FREQUENCY (Hz)

6·14

1K

10K

o

100

1K

10K

LOAD RESISTANCE (n)

1/86, Rev. A

---------I~ AMP-G1 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

SLEW RATE VB
LOAD CAPACITANCE

SLEW RATE VB
VOLTAGE GAIN

SETTLING TIME TO 0.01%
VB VOLTAGE GAIN
70

~sl.lll!~~

Js!'~W
r- 20V
STEP

60
50

4

-"
w
Ii

40

"

30

:Ji

20

>-

o

2

~

,

100p

VOLTAGE GAIN, G

VOLTAGE NOISE DENSITY
VB FREQUENCY

"

lOOn

'0"

,"

o

\
1--++ttItttI--t-tttttttt-+tttllllll,oU

,

1000

'00

VOL lAGE GAIN, G

RTI VOLTAGE NOISE
DENSITY VB GAIN

POSITIVE SUPPLY CURRENT
VB SUPPLY VOLTAGE

_TA=2~OC

Vs = +15V
t= 1KHz

~

~,oo._.

'0

:Ji
0
z

--- -

:Ji

o2

w

W

:1

~ 'O~I;I~I
~

C;
0
>

~
0

'0

LOAD CAPACITANCE (F)

".~s:J

15

~';

,"

i.-"

'0

,

'0

'00

,.

'0'

,L-~~llllL-~~llllL-~-W~

1

10

FREQUENCY (Hz)

NEGATIVE SUPPLY CURRENT
VB SUPPLY VOLTAGE
-s

r--

7

TA =

100

lK

o

o

±5

±10

±15

±20

VOLTAGE GAIN, G

POWER SUPPLY VOLTAGE (VOLTS)

POSITIVE SUPPLY CURRENT
VB TEMPERATURE

NEGATIVE SUPPLY CURRENT
VB TEMPERATURE

v~."L

2~OC

6
5
-4

-3
-2

,
o

o

-

.......-'

,
o

±5

±10

±15

POWER SUPPLY VOLTAGE (VOLTS)

±20

-75 -50 -25

0

25

50

75

TEMPERATURE (Oe)

6-15

100 125

150

~'::75~-5~0---!2::-5~0-2:!::5"'-:5!:0--:7'::5--:-!'O::-0-':-':2:-5-7.'50
TEMPERATURE (OC)

1/86, Rev. A

---------I~ AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

INPUT AND OUTPUT OFFSET VOLTAGES

GAIN

Instrumentation amplifiers have independent offset voltages
associated with the input and output stages. While the initial
offsets may be adjusted to zero, temperature variations will
cause shifts in offsets. Systems with auto-zero can correct
for offset errors, so initial adjustment would be unnecessary.
However, many high-gain applications don't have auto zero.
For these applications, both offsets can be nulled, which has
minimal effect on TCVlos and TCVoos.

The AMP-01 uses two external resistors for setting voltage
gain over the range 0.1 to 10,000. The magnitudes of the
scale resistor, Rs, and gain-set resistor, RG, are related by the
formula: G =20X RglR G, where G is the selected voltage gain
(Refer to Figure 1).

The input offset component is directly multiplied by the
amplifier gain, whereas output offset is independent of gain.
Therefore, at low gain, output-offset-errors dominate, while
at high gain, input-offset-errors dominate. Overall offset
voltage, Vos, referred to the output (RTO) is calculated as
follows;

v.

"N

Vos(RTO)= (VIOSX G) + Voos.. ······ .. ········ .. ·· .. ·· .. ······· .. ··········.. · (1)
where VIOS and Voos are the input and output offset voltage
specifications and G is the amplifier gain. Input offset nulling
alone is recommended with amplifiers having fixed gain
above 50. Output offset nulling alone is recommended when
gain is fixed at 50 or below.

-IN

In applications requiring both initial offsets to be nulled, the
input offset is nulled first by short-circuiting RG, then the
output offset is nulled with the short removed.

VOL TAGE GAIN, G

The overall offset voltage drift TCVos, referred to the output,
is a combination of input and output drift specifications.
Input offset voltage drift is multiplied by the amplifier gain, G,
and summed with the output offset drift;

RS
( 20R'G )

v-

Figure 1. Basic AMP-Ol connections for gains 0.1 to 10,000.

The magnitude of Rs affects linearity and output referred
errors. Circuit performance is characterized using Rs = 10kO
when operating on ±15 volt supplies and driving a ±1 0 volt
output. Rs may be reduced to 5kfl in many applications
particularly when operating on ±5 volt supplies or if the
output voltage swing is limited to ±5 volts. Bandwidth is
Improved with Rs = 5kfl and this also increases commonmode rejection by approximately 6dB at low gain. Lowering
the value below 5kfl can cause Instability in some circuit
configurations and usually has no advantage. High voltage
gains between two and ten thousand would require very low
values of RG. For Rs = 10kfl and Av=2000we get RG = 1000;
this value is the practical lower limit for RG. Below 1000,
mismatch of wlrebond and resistor temperature coefficients
will introduce significant gain tempco errors. Therefore, for
gains above 2,000, RG should be kept constant at 1000 and
Rs increased. The maximum gain of 10,000 is obtained with
Rs set to 50kO.

TCVos (RTO) = (TCVlosx G) + TCVoos·············· .. ·· .. ·········· .. (2)
where TCVlos is the input offset voltage drift, and TCVoos is
the output offset voltage specification. Frequently, the
amplifier drift is referred back to the input (RTI) which is then
equivalent to an input signal change;
TCVos(RTI)=TCVlos+ TCVoos···········································(3)
G
For example, the maximum input-re'terred drift of an
AMP-01 EX set to G = 1000 becomes;
TCVos (RTI) = 0.3JlV/oC + 100JlVrC = O.4JlV/oC max.
1000

INPUT BIAS AND OFFSET CURRENTS
Input transistor bias currents are additional error sources
which can degrade the input Signal. Bias currents flowing
through the signal source resistance appear as an additional
offset voltage. Equal source resistance on both inputs of an
IA will minimize offset changes dueto bias current variations
with signal voltage and temperature. However, the difference
between the two bias currents, the input offset current,
produces a non-trimmable error. The magnitude of the error
is the offset current times the source resistance.

Metal-film orwirewound resistors are recommended for best
results. The absolute values and TC's are not too important,
only the ratio metric parameters.
AC amplifiers require good gain stability with temperature
and time, but DC performance is unimportant. Therefore, low
cost metal-film types with TC's of 50ppm/oC are usually
adequate for Rs and RG. Realizing the full potential of the
AMP-01 's offset voltage and gain stability requires precision
metal-film or wirewound resistors. Achieving a 15ppm/oC
gain tempco at all gains requires Rs and RG temperature
coefficient matching to 5ppm/o C or better.

Acurrent path must always be provided between the differential inputs and analog ground to ensure correct amplifier
operation. Floating inputs, such as thermocouples, should
be grounded close to the signal source for best commonmode rejection.

6-16

1/86, Rev. A

------------I~ AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

vOltage gain. For example, at 25° C, IVR is specified as ±10.5
volt minimum with ±15 volt supplies. Using a ±10 volt
maximum swing output and substituting the figures in (4)
simplifies the formula to:

1M

Vs - +15V

§
100

k~

CMVR = ±
AS

I"

10k

100
1

10

100
VOL TAGE GAIN

. . . . (5)

ACTIVE GUARD DRIVE

'"

1k

Rejection of common-mode noise and line pick-up can be
improved by using shielded cable between the signal source
and the IA. Shielding reduces pick-up, but increases input
capacitance, which in turn degrades the settling-time for
signal changes. Further, any imbalance in the source resistance between the inverting and noninverting inputs, when
capacitively loaded, converts the common-mode voltage into
a differential voltage. This effect reduces the benefits of
shielding. AC common-mode rejection is improved by"bootstrapping" the input cable capacitance to the input signal, a
technique called "guard driving". This technique effectively
reduces the input capacitance. A Single guard-driving signal
is adequate at gains above 100 and should be the average
value of the two inputs. The value of external gain resistor RG
is split between two resistors RG1 and RG2; the center tap
provides the required signal to drive the buffer amplifier
(Figure 2).

10k

RG AND Rs SELECTION
Gain accuracy is determined by the ratio accuracy of Rs and
RG combined with the gain equation error of the AMP-01
(0.6% max for AlE grades).
All instrumentation amplifiers require attention to layout so
thermocouple effects are minimized. Thermocouples formed
between copper and dissimilar metals can easily destroy the
TCVos performance of the AMP-01 which is typically
O.15p.V/0 C. Resistors themselves can generate thermoelectric
EMF's when mounted parallel to a thermal gradient. "Vis hay"
resistors are recommended because a maximum value for
thermoelectric generation is specified. However, where
thermal gradients are low and gain TC's of 20-50ppm are
sufficient, general-purpose metal-film resistors can be used
for RG and Rs.

GROUNDING
The majority of instruments and data acquisition systems
have separate grounds for analog and digital Signals. Analog
ground may also be divided into two or more grounds which
will be tied together at one point, usually the analog powersupply ground. In addition, the digital and analog grounds
may be joined, normally at the analog ground pin on the
A-to-D converter. Following this basic grounding practice is
essential for good circuit performance (Figure 3).

COMMON-MODE REJECTION
Ideally, an instrumentation amplifier responds only to the
difference between the two input signals and rejects
common-mode voltages and noise. In practice, there is a
small change in output vOltage when both inputs experience
the same common-mode voltage change; the ratio of these
vOltages is called the common-mode gain. Common-mode
rejection (CMR) is the logarithm of the ratio of differentialmode gain to common-mode gain, expressed in dB. CMR
specifications are normally measured with a full-range input
voltage change and a specified source resistance unbalance.

Mixing grounds causes interactions between digital circuits
and the analog signals. Since the ground returns have finite
resistance and inductance, hundreds of millivolts can be
developed between the system ground and the data acquisition components. Using separate ground returns minimizes
the current flow in the sensitive analog return path to the
system ground point. Consequently, noisy ground currents
from logic gates do not interact with the analog signals.

The current-feedback design used in the AMP-01 inherently
yields high common-mode rejection. Unlike resistive feedback designs, typified by the three-op-amp lA, the CMR is not
degraded by small resistances in series with the reference
input. A slight, but trimmable, output offset voltage change
results from resistance in series with the reference input.

Inevitably, two or more circuits will be joined together with
their grounds at differential potentials. In these situations,
the differential input of an instrumentation amplifier, with its
high CMR, can accurately transfer analog information from
one circuit to another.

The common-mode input voltage range, CMVR, for linear
operation may be calculated from the formula'
CMVR = ± (IVR

5

For all gains greater than or equal to 10, CMVR is ±10 volt
minimum; at gains below 10, CMVR is reduced.

' " AG

1k

(10 -~)

SENSE AND REFERENCE TERMINALS
The sense terminal completes the feedback path for the
instrumentation amplifier output stage and is normally
connected directly to the output. The output signal is
specified with respect to the reference terminal, which is
normally connected to analog ground.

JV~~TI) ..... (4)

IVR is the data sheet specification for input voltage range;
VOUT is the maximum output signal; and G is the chosen
6-17

1/86, Rev. A

II

----------t~ AMP-01

LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

+15V

• C5

AV

=

+---____T..J lOJ.iF

R4

VOLTAGE GAIN, G" (20 x RS)'
RGt

NC

500 WITH COMPONENTS SHOWN

SENSE

.,NQ-- " - - - - - - - - - - - < ; > VOUT • ±3V MAX
5011
LOAD

-IN 0------"-1

C2

o 04711F
VOLTAGE GAIN, G =

CO:GRs)

-1---'
o
J...

~'IV\"""'-----l~1

047,uF

RESISTORS R, AND R2 REDUCE Ie DISSIPATION

'-----------------e--------o-15V

Figure 5. Driving son loads.
6-19

1/86, Rev. A

----------l~ AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

Protection can also be achieved by connecting back-to-back
9.1V zener diodes across the differential inputs. This technique does not affect the input noise level and can be used
down to a gain of 2 with minimal increase in input current.
Although voltage-clamping elements look like short circuits
at the limiting voltage, the majority of signal sources provide
less than 50mA, producing power levels that are easily
handled by low-power zeners.

to several hundred milliwatts (for example, the 4-20mA
current transmitter application, Figure 8). Excessive dissipation will cause thermal shutdown of the output stage thus
protecting the device from damage. A heatsink is recommended in power applications to reduce the die temperature.
Several appropriate heatsinks are available; the Thermalloy
6010B is especially easy to use and is inexpensive. Intended
for dual-in-line packages, the heatsink may be attached with
a cyanoacrylate adhesive. This heatsink reduces the thermal
resistance between the junction and ambient environment to
approximately 80° C/W. Junction (die) temperature can then
be calculated by using the relationship:

Simultaneous connection of the differential inputs to a lowimpedance signal above 10V during normal circuit operation
is unlikely. However, additional protection involves adding
1000 current-limiting resistors in each signal path priorto the
voltage clamp; the resistors increase the input noise level to
just 5.4nVlyHZ (refer to Figure 6).

Tj-T.

Pd = - - OJ.

Input components, be they multiplexers or resistors, should
be carefully selected to prevent the formation of thermocouple junctions which would degrade the input signal.

where T J and T. are the junction and ambient temperatures
respectively, OJ' is the thermal resistance from junction to
ambient, and Pd is the device's internal dissipation.

OVERVOLTAGE PROTECTION
Instrumentation amplifiers invariably sit at the front end of
instrumentation systems where there is a high probability of
exposure to overloads. Voltage transients, failure of a transducer, or removal of the amplifier power supply while the
signal source is connected may destroy or degrade the
performance of an unprotected amplifier. Although it is
impractical to protect an IC internally against connection to
power lines, it is relatively easy to provide protection against
typical system overloads.

*OPTIONAl PROTECTION
RESISTORS, SEE TEXT

+15V

LINEAR INPUT RANGE,
±5VMAXIMUM
DIFFERENTIAL PROTECTION
TO ±30V

lOon lW*

"N<>--'\NI.-.---j

91V lW

Your

ZENERS

loon lW*
-IN <>--'\NI,....---j

The AMP-01 is internally protected against overloads for
gains of up to 100. At higher gains, the protection is reduced
and some external measures may be required. Limited
internal overload protection is used so that noise performance would not be significantly degraded.

-15V

AMP-01 noise level approaches the theoretical noise floor of
the Input stage which would be 4nVlyHZat 1kHz when the
gain is set at 1000. Noise is the result of shot noise in the input
devices and Johnson noise in the resistors. Resistor noise is
calculated from the values of RG (2000 at a gain of 1000) and
the input protection resistors (2500). Active loads for the
input transistors contribute less than 1nVlyHZof noise. The
measured noise level is typically 5nV/yHZ.

Figure 6. Input overvoltage protection for gains 2 to 10,000.

POWER SUPPLY CONSIDERATIONS
Achieving the rated performance of precision amplifiers in a
practical circuit requires careful attention to external influences. Forexample, supply noise and changes in the nominal
voltage directly affectthe input offset voltage. A PSR of 80dB
means that a change of 100mV on the supply, not an
uncommon value, will produce a 10j.LV input offset change.
Consequently, care should be taken in choosing a power unit
that has a low output noise level, good line and load
regulation, and good temperature stability.

Diodes across the input transistor's base-emitter junctions,
combined with 2500 input resistors and RG, protect against
difrerential inputs of up to ±20V for gains of up to 100. The
diodes also prevent avalanche breakdown that would degrade
the 18 and los specifications. Decreasing the value of RG for
gains above 100 limits the maximum input overload protection to ± 10V. External series resistors could be added to
guard against higher voltage levels at the input, but resistors
alone increase the input noise and degrade the signal-tonoise ratio, especially at high gains.

6-20

1/86, Rev. A

-----------l~ AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

APPLICATIONS INFORMATION

+1SV

COMPLIANCE, TYPICALLY !10V
LINEARITY - 0 01%

r 0 047 ,1.1F

OUTPUT RESISTANCE AT 20mA - 5Mn
POWER BANDWIDTH {-3d8} - 60kHz
I NTO
LOAD

soon

-IN

(20' RS)

0------''-1

lOUT = VIN AG x A,

A, = loon FDA IOUT=!20mA
V 1N =±100mV FOR ±20mA FULL SCALE

-15V

Figure 7. High-compliance bipolar current source with 13-blt linearity.

r------~------------------~-----o;~~ov
ALL RESISTORS 1% METAL FILM
Rs , 2 Okn

+INo--------"'1
ROUT TRIM

RG
275kn

v-

REF-02

'0

RS ' 221kn

11
-IN

ZERO TRIM

A,

0--------='1

R,

loon

lOon

L---------------4--------~~~120mA
L - -_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___o-5V
COMPLIANCE OF lOUT, +20V WITH +30V SUPPLY (OUTPUT w r

t

OV}

DIFFERENTIAL INPUT Of 100mV FOR l6mA SPAN
OUTPUT RESISTANCE

~

5MH AT lOUT

20mA

LINEARITY 0 01% OF SPAN

Figure 8. 13-bllllnear 4-20mA transmitter constructed by adding a voltage reference. Thermocouple signals can be accepted
without preampliflcation.

6-21

1/86, Rev. A

_ _ _ _ _ _ _ _-I~AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

, - - - - - -.....- - - - - - - - . - - - - - - o + 1 5 V

10kn

+IN

o-----'''"l

SENSE

10051

~...:c.-+-¥.IV-_4--~O---O~~l~~ INTO lOn)

''''
-IN

0-----'-1
VOLTAGE GAIN, G '= 100
POWER BANDWIDTH (-3dB), 60KH2
QUIESCENT CURRENT, 4mA
LINEARITy-a 01%@FULL

l

OUTPUT INTO 10H

L - - - - -.....- - - - - - - - - * - - - - - - o - 1 5 V

Figure 9. Adding two transistors Increases output current to ±1A without affecllng the quiescent current of 4mA. Power bandwidth Is 60kHz.

+15V
Q1, Q2
Q3, Q4, 05.
lel

Jl10
Jl07

AS·l0kn

. CMP-04

~

::~O---------------------------~1:~:G
IC'

0P15GZ

200kn

20kn

2kn

196n

t 0047"F

1'4 ...

-I

~
RS" 2

RG

47kn

'13

~-("'>':"""'t--oOUT

AMP-01

47kn

v~ 10

l

47M2

Voas

f

13

5

H

LlNEARITy ..... a 005%, G'" 10 & 100
....... 002%. G = 1 & 1000
GAIN ACCURACY, UNTRIMMED-05%
SETTLING TIME TO 01%, ALL GAINS,
LESS THAN 75ps

IC1

8'

+15V o_-4~'I2117k1l"~+_--".::.°I_-_I_---I---_+--.J

a

12

GAIN SWITCHING TIME, LESS THAN lOOps

"

27kn

REFERENCE

t-----OGND

~~

/

~

t-_-O.6 1-_+-_---'

~11

7~

:?~4
14

D

V~~S:j

~

47kn

-15V
TTL COMPATIBLE INPUTS

Figure 10. The AMP-01 makes an excellent programmable-gain Instrumentation amplifier. Combined gain-switching and
settling lime to 13-bits falls below 1001'S. Linearity Is beUer than 12-blts over a gain range 1 to 1000.

6·22

1/86, Rev. A

_ _ _ _ _ _ _ _--I!fMDAMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

r---~~---------------------------------------t--------------------O+'5V
AS_ 10kfl
*MATCHEDTOO.l%
+IN C>--------"'-j

SENSE

15kn

*Skn

REFERENCE
-IN C>---------'-j

T0047MF

~OV

(20 , RS)

-+______

L-____________________

VOLTAGE GAIN, G'" - -

~------+_-------+--------------------O-'5V

RG
MAXIMUM OUTPUT, 20Vp-p INTO 600.11
THO 00l%@1KHz,20Vp-pINT0600n, G=10
DIFFERENTIAL
OUTPUT

OUTPUT
COMMON-MODE
REFERENCE
(±5V MAX)

Figure 11. A differential Input Instrumentation amplifier with differential output replaces a transformer In many applications. The
output will drive a 6000 load at low distortion, (0.01%).

+15V

POWER BANDWIDTH (-3dB)-150KHz
TOTAL HARMONIC DISTORTION-a 006%
@lKHz, 20Vp-p INTO 500.11 111000pF

1

R,

~~----------------+-------~~------~~OVOUT

390n

R,
495kn

NC

NC

CLOSED-LOOP VOL lAGE GAIN MUST BE
GREATER THAN 50 FOR STABLE OPERATION

-15V

VOL TAGE GAIN, G = (1 +

~)

Figure 12. Configuring the AMP-01 as a nonlnvertlng operational amplifier provides exceptional performance. The output
handles low load Impedances at very low distortion, 0.006%.

6-23

1/86, Rev. A

----------l[fMI> AMP·ln LOW·NOISE PRECISION INSTRUMENTATION AMPLIFIER
NC

NC

":>-"--_+---0 Your

47kn

R R2
1 - GAIN (G)

+15V

20V p-p INTO 500fllll000pF.

R3 = Rl II R2
R4

-15V

TOTAL HARMONIC DISTORTION:

= 1.5kfl @ G = 1
1.2kfl @ G = 10
120fl @G = 100 & 1000

<0.005% @ 1kHz, VOUT

20V p.p,

=

G = 1 TO 1000

Figure 13. The Inverting operational amplifier configuration has excellent linearity over the gain range 1 to 1000, typically
0.005%. Offset voltage drift at unity gain Is Improved over the drift in the Instrumentation amplifier configuration.

+15V

I

+

-=

lO F
/./

POWER BANDWIDTH

(-3dB)~60KHz

TOTAL HARMONIC DISTORTION~O.OOl%
@ 1 KHz, 20Vp·p INTO 500flll 1000pF

1

~~-------~----~------~--oVOUT

"2

47k.l1

NC

NC

-15V

Figure 14. Stability with large capacitive loads combined with high output current capability make the AMP-D1 Ideal for line
driving applications. Offset voltage drift approaches the TCVIOS limit, (0.3IlYlo C).

6·24

1/86, Rev. A

_ _ _ _ _ _ _ _--I~AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

NOISE TEST CIRCUIT (O.lHz to 10Hz)

162kn

v'

V-

">-~-4---o OUTPUT
11

200kH

20k~1

2kH

1 82kH

200n

162MD.

en (G = 1, 10, 100) =

en (G = 1000)

=

V'

V-

909kH

'O~~~~

1~~~~
loon

lkH

SETTLING-TIME TEST CIRCUIT

200n lOT

V'N
20V p-p

191knOl%

2 x HSCH-l001
10kH
01%

2kH
01%
10kUO 1%

11kD.
01%

102n
01%

lOn
01%

200kn
01%

6-25

1/86, Rev. A

----------l!EMD

AMP-01 LOW-NOISE PRECISION INSTRUMENTATION AMPLIFIER

INSTRUMENTATION AMPLIFIER WITH AUTO-ZERO
r-------------------------~~------------~------------~---------------------------o+15V

RS ' TOkH

VOL TAGE GAIN, G =(20 x RS)
RG

VOUT

75kS1

,.

1SkH

14

13
±lmA
OAe-OB

15

R,
lOOH

7 Skll

001jJF

TTL INPUT
"OFFSET"

°v
TTL INPUT

"ZERO"
-15V

BURN-IN CIRCUIT
+18V

10kn

::::~----+--< VOUT

lOkn

~ 0 047,uF

-lav

6-26

1/86, Rev. A

AMP-OS
FAST-SETTLING JFET
INSTRUMENTATION AMPLIFIER
Precision Monolithics Inc.

PRELIMINARY
FEATURES

PIN CONNECTIONS

Settling-Time to 12-Bit Accuracy, G ~ 1000 .. 12fLS Max
14-Bit Gain Linearity at G ~ 1000
On-Board Dual Guard Drivers
On-Board lOOfLA Precision Current Source
Low Bias Current ....•........... 50pA Max @ 25°C
. • . . . • . . . . . . . .. 20nA Max @ 125°C
• Temperature Stable CMR
• . . . . • • . . • . . . • . .• 105dB Min Over -55°C to + 125°C
• High Slew-Rate with 500pF Load ......••..• 5V1fLS Min
•
•
•
•
•

-GUARD

DRIVE
Voos NULL 4
Voos NULL 5

II

14 VIOS NULL

13 ~g~=~~T

ORDERING INFORMATIONt
CERDIP
l8-PIN
PACKAGE

OPERATING
TEMPERATURE
RANGE

AMP05AX'
AMP05BX'
AMP05EX
AMP05FX

MIL
MIL
IND
IND

REFERENCE 8

SENSE 9

18-PIN HERMETIC DIP
(X-Suffix)

• For devices processed In total compliance to MIL-STD-883, add /883 after part
number. Consult factory for 883 data sheet.
t AU commercial and industnal temperature range parts are available with burn-In
For ordenng information see 1986 Data Book, Section 2

SIMPLIFIED SCHEMATIC
r---------------~----~------~----~----------------------_4r_---Ov+

OUT

-GUARD

DRIVE
REFERENCE

47 SkU
475kn

1---.....---'lNV-----+----oSENSE

25kO

25HZ

1V

L---------------*-----*-------~----~----------------------~~--_Ov-

This preliminary producllnformatlon Is based on lesllng of a IImiled number of devices. Final specifications may vary. Please conlacllocal sales
office or dlslrlbulor for final dala sheet.

6-27

1/86, Rev. A

---------I~ AMP-OS FAST -SETILING JFET INSTRUMENTATION AMPLIFIER -

GENERAL DESCRIPTION

PRELIMINARY

resistance. Internal input protection allows a 30V differential
overload at all gain settings. AMP-05 voltage gain is set by the
ratio of two external resistors over the range 0.1 to 2000 and a low
gain temperature-coefficient of 20ppm/oC is achievable in the
range 1 to 1000.

The AMP-05 is a fast JFET instrumentation amplifier designed
for high-speed analog signal-processing and analog-multiplexed data acquisition systems. Settling-time to 12-bits is 12J-Ls
maximum, with better than 14-bit linearity at all gains up to 1000.
Two functions are added to the instrumentation amplifier that reduce external component count in many applications. On-board
dual guard drivers maintain good settling-time and commonmode rejection performance when shielded cable connects the
input signal to the AMP-05. A precision 100J-LA current source is
also provided for transducer excitation, powering a low-current
voltage reference, and other functions.

The AMP-05's outputs can all drive large capacitive loads without
oscillation. The amplifier output is guaranteed stable with loads
up to 2,000pF and the guard drivers can tolerate up to 10,000pF
without oscillation.
Sense and reference pins complete the output feedback-loop
and provide an output ground reference, respectively. The reference pin may be used for zeroing system offsets, where autozero hardware is employed. ReSistance, in series with the reference terminal, does not degrade common-mode rejection on
PMl's AMP-05, which is a significant problem with instrumentation amplifiers employing the three op-amp configuration.

The AMP-05 employs a current-feedback technique which provides a high and stable common-mode rejection, 105dB minimum over the military temperature range. JFET inputs reduce
bias current to 50pA maximum at 25°C and only 20nA maximum
at 125°C; low bias current reduces errors due to signal-source

6-28

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
,

Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

VOLTAGE
FOLLOWERS/BUFFERS
Precision Monolithics Inc.

7-3

7-4

Introduction

BUF.03
High-Speed Voltage Follower/Buffer

7-2

VOLTAGE
FOLLOWERS/BUFFERS
Precision M0l101ithics Inc.

INTRODUCTION
feedback can provide better frequency
response. In addition, output current can be
increased substantially beyond that of conventional Ie operational amplifiers.
The BUF-03 is a high-speed, unity-gain Ie that
is optimized for the buffer function. A FET input
provides high input impedance. On-chip zenerzap trimming is used to reduce the offset voltage. The output stage is designed to supply
approximately 70mA of peak current. These features combine to make the BUF-03 an Ie analog
buffer of unique capability.

The function of a unity-gain buffer is to accurately reproduce the input signal under widelyvarying load conditions. To do this, buffers must
have high input impedance, wide bandwidth,
and high output drive. Offsets and gain error
need to be minimized.
The buffer function can be implemented by use
of general-purpose operational amplifiers connected as unity-gain voltage followers, but
higher performance can be obtained by optimizing a circuit specifically for buffering. A design
dedicated to unity-gain buffering and using no

7-3

BUF-03
HIGH-SPEED VOLTAGE
FOLLOWER/BUFFER
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

in an open-loop circuit employing source followers and
emitter followers, the BUF-03 utilizes a quasi-quad FET input
structure to optimize both speed and D.C. input characteristics. On-chip zener-zap trimming is used to achieve low
offset voltage while careful biasing throughout results in
excellent gain linearity over the full input voltage range,

Very High Slew Rate .................. 220V//Lsec Min
Wide Bandwidth ............................. 63MHz
Load Drive Current .. . . . . . . . . . . . . . . . . . . . .. 70mA Peak
Easily Drives Large Capacitive Loads Without Oscillation
High Input Resistance .................... " 5 x 10"0
Low Output Resistance . . . . . . . . . . . . . . . . . . . . . . . . . .. 20
very Low Bias Current (Warmed-Up) ...... 400pA Max
Low Offset Voltage ......................... 6mV Max
Unity Gain ................................. 0.997V/V
Excellent Gain Linearity ....................... 0.015%

Applications for which the BUF-03 is well-suited include
high-speed line drivers, isolation amplifiers for driving
reactive loads, and high-speed sample-hold circuits.

PIN CONNECTIONS
ORDERING INFORMATIONt
TA = 2S"C
VosMAX
(mV)

6
6

'5
15

N.C .

PACKAGE
TO-99
8-PIN

OPERATING
TEMPERATURE
RANGE

BUF03AJ'
BUF03EJ
BUF03BJ'
BUF03FJ

MIL
COM
MIL
COM

•

NUL&,'7V+
N.C.2

6 OUTPUT

INPUT 3

TO-99 (J-Sufflx)

5 NULL

4
v-CeASE)

'For devices processed ,n total compl,ance to MIL·STD·883, add /883 after
part number. Consult factory for 883 data sheet.
t All commercial and Industrial temperature range parts are available with
burn-In. For ordenng information see 1986 Data Book, Section 2.

OPTIONAL OFFSET NULLING CIRCUIT
:>...~~---ov+

GENERAL DESCRIPTION
The BUF-03 is the first very high-speed monolithic voltage
follower. Featuring performance previously unobtainable in a
monolithic unit, it offers a combination of both exceptional
speed and excellent input/output specifications. Implemented

INPUT

0-----',

>-"----0 OUTPUT

SIMPLIFIED SCHEMATIC
,--~--.----~-~--------~r-ov+

OUTPUT

INPUT

L-----~----~--~---~-~~v-

7-4

1/86, Rev, A

-----------l~ BUF-03 HIGH-SPEED VOLTAGE FOLLOWER/BUFFER

ABSOLUTE MAXIMUM RATINGS (Note)
Maximum Junction Temperature (Tj) .............. 175° C
Storage Temperature Range. . .. . . . . . .. -65° C to + 175° C
Operating Temperature Range
BUF-03A, BUF-03B ................. -55°Cto+125°C
BUF-03E, BUF-03F ..................... 0°Cto+70°C
Lead Temperature (Soldering, 60 sec) ............. 300°C
DICE Junction Temperature (Tj) ....... -65°C to +175°C

Supply Voltage (V+ to V-) .......................... 36V
Internal Power Dissipation (P d ) (see curves)
in still air, no heat sink ......................... 1.05W
with heat sink, ()JA = 90° C/W ................... 1.40W
Input Voltage (for Vs < ± 1BV, maximum input
voltage is equal to supply) ...................... ±1BV
Continuous Output Current ....................... 70mA
Peak Output Current ............................. 100mA
Short-Circuit Protection (Maximum Pd or TJ
not to be exceeded) ................ Indefinite at BOmA

NOTE: Absolute ratings apply to both DICE and packaged parts. unless
otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, TA = 25° C, TCHIP = 75° C, device fully warmed-up, unless
otherwise noted. (Note 1)
8UF-03A/E
SYMBOL

PARAMETER

CONDITIONS

MIN

TYP

220

250

8UF-038/F

MAX

MIN

TYP

MAX

UNITS

180

250

VI!'sec

AC SPECIFICATIONS
Slew Rate

SR

RL" 2kO. C L = 50pF

Power Bandwidth

PBW

V ,N = 10Vp_p' R L ,,2kO

BandWidth

BW

aV ,N ='; 2Vp_p

63

50

MHz

90

100

nsec

7

nsec

Settling Time

ts

To 0.1 %. ± 10V step

Capacitive Load Capability

C LOAD

No Oscillations

Propagatton Delay

td

Step Input

Rise Time

t,

av =0.5V

Wide Band Input Noise Voltage

Vn

DC to 50MHz

Input Noise Voltage Density

en

f= 10kHz

Input Offset Voltage

Vos

Rs '; 20kO

Input Bias Current

I.

9

MHz

!'F

7

nsec

350

400

!'VRMS

50

60

nV/y'HZ

DC SPECIFICATIONS

Input Resistance

150

6

4

15

mV

400

180

700

pA

_ 4X10 11

_5X1Q11

R'N

09960
09945
09925

Voltage Gain (V,N = ± 10V)

Avo

R L " 10kO
RL" 2kO
RL" 1kO

Nonlinearity (Note 2)

NL

V,N =±10V. R L ,,2kO
V,N =±7V. RL" 1kO

Maximum Output Error

OUTerror

Y,N = + 10V. OV. -10V
Rs =Ot020kO
R L ;::: 2kO in all combinations

Power Supply Rejection Ratio

PSRR

Vs = ±6V to ±18V

Supply Current

ISY

No Load

Peak Load Current

IL(PKI

Output ResIstance

Ro

Offset Voltage Nulling Range

avos

Input Voltage Range
(Reduced Accuracy)

IVR

09975
0.9960
09945

09940
09930
0.9905

09970
0.9950
09930

0015
0013

0023
0.023

0017
0015

003
0.03

%FS.

40

60

50

85

mV

010

071

0.15

1.42

mV/v

19

25

19

25

mA

70

70

mA

±80

±80

mV

±115

±115

V

2

If no heat Sinking IS used, the chip temperature (m still air) may exceed
1050 C The effect of this elevated temperature Will be to mcrease the Input
bias current by a factor of eight, increase the Vos specification by
TCVos X 30' C, and ,educe device speed by 10%.
Nonlinearity IS computed using linear regression techniques with data
from five POtnts (e g. -10V. -5V. OV. +5V, and +10V for ±10V full-scale
linearity)

1/86, Rev. A

7-5

...- - - - - - -

V/V

0

Rp" 1kO

NOTES:
1. The BUF-03 package thermal resistance. In stili air. IS 145°C/W (45'C/W
Junction-ta-case, 10QoC/W case-te-amblent). The chip temperature of 75° C
IS achieved by redUCing the case-to-amblent thermal resistance to 45° C/W
An Inexpensive heat Sink, such as the Thermalloy 2271 B or 6203, IS
recommended for use In thiS application. In addition, if the deVice IS
operated In a forced-air enVironment, or is attached to a PC board which
has good thermal conductivity, the chip temperature may be further
reduced.

.~-----~-~--

0

-----

..

_-

-------------,-

-----_.

~

~
>l.o

~

~
~

~
....:l
....:l

0>l.o
f.I.l

~
b
~

~

BUF-03 HIGH-SPEED VOLTAGE FOLLOWER/BUFFER

ELECTRICAL CHARACTERISTICS at Vs =

± 15V, -55 0 e:s TA:S + 125 0 e, T CHIP( MAX) = + 165 0 e, device fUlly warmed-up,

unless otherwise noted. (Note 1)

BUF-03A
PARAMETER

SYMBOL

CONDITIONS

Slew Rate

SR

R L ", 2kll, C L = 50pF

Input Offset Voltage

Vos

Rs" 2kll

Average Input Offset
Voltage Dnft

TCVos

Rs" 2kll, (Note 2)

Input Bias Current

18

TA =+125°C

Voltage Gam

Avo

R L ",

Power Supply Relectlon RatiO

PSRR

Vs = ±7V to ±15V

Supply Current

ISY

TA=+125°C

MIN

TYP

BUF-03B
MAX

MIN

220

2kll, Y,N = ± 10V

09920

TYP

MAX

220

UNITS
VllJ.sec

6

20

10

35

mV

50

100

90

170

pVloC

25

75

30

90

0.9955

09902

0.9942

nA
VIV

Gain Drift with Temperature

ppmfOC
0.15

126

020

2.24

mVIV

18

24

18

24

mA

ELECTRICAL CHARACTERISTICS at Vs=±15V, ooe :STA:S+70oe, TCHIP(MAX) =+120 o e, device fully warmed-up, unless
otherwise noted.

BUF-03E
PARAMETER

SYMBOL

CONDITIONS

Slew Rate

SR

R L ",2kll

Input Offset Voltage

Vos

Rs" 2kll, C L = 50pF

Average Input Olfset
Voltage Drift

TCVos

Rs" 2kfl, (Note 2)

Input Bias Current

18

TA =+70°C

Voltage Gain (V ,N = ± 10V)

Avo

R L ", 2kll

Power Supply Reiecllon Ratio

PSRR

Vs= ±7V to ±15V

Supply Current

ISY

TA =+70°C

MIN

TYP

BUF-03F
MAX

MIN

240

TYP

14
40

80

90

15
0.9935

MAX

0.9958

09918

UNITS

Vlpsec

240
28

mV

150

pVloC

1.8

nA

0.9946

VIV

ppm/oC

Gain Drift with Temperature
0.12
19

25

016

178

mVIV

19

25

mA

NOTES:
1. In order to operate the device at an ambient temperature of +125° C, more
extensive heat sinking must be used to ensure that the chip temperature
never exceeds the absolute maximum of +1750 C. Thechip temperature of
+ 1650 C is achieved by reducing the case-te-ambient thermal resistance
to 30° C/W (e g., Thermalloy 2227)
2. Guaranteed by design.

7-6

1/86, Rev. A

-------------1~ BUF-03 HIGH-SPEED VOLTAGE FOLLOWER/BUFFER

DICE CHARACTERISTICS

1.
3.
4.
5.
6.
7.

DIE SIZE 0.070 X 0.048 Inch, 3360 sq. mils
(1.78 X 1.22 mm, 2.17 sq. mm)

NULL
INPUT
NEGATIVE SUPPLY
NULL
OUTPUT
POSITIVE SUPPLY

For additional DICE information refer to
1986 Data Book, Section 2.

WAFER TEST LIMITS at Vs = ± 15V, Tj =25°C, unless otherwise noted.
SYMBOL

CONDITIONS

Inpul Offsel Vollage

Vos

Rs" 20kll

Slew Rale I Nole 1 )

SR

RL ;" 2kll, C L = 50pF

Vollage Gain

Avo

R L ;" 10kll, V,N = ±10V

Power Supply Rejection Ratio

PSRR

Vs =±6Vlo±18V

Supply Currenl

ISY

No Load

PARAMETER

aUF-03N

BUF-03G

LIMIT

LIMIT

UNITS

6

15

mVMAX
VI~sec

220

180

09960

0.9940

VIV MIN

MIN

071

142

mVIV MAX

25

25

rnA MAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualificatIOn through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V, TJ = 25° C, unless otherwise noted.
aUF-03N

aUF-03G

TYPICAL

TYPICAL

UNITS

ILIPK)

70

70

rnA

I npul Bias Currenl

18

40

60

pA

Input Resistance

R'N

5 X 10"

5 X 10"

Il

Oulpul Reslslance

Ro

Offsel Vollage Nulling Range

avos

±80

±80

mV

Inpul Vollage Range
I Reduced Accuracy)

IVR

±115

±115

V

MHz

PARAMETER

SYMBOL

Peak Load Current

CONDITIONS

Il

Rp;" 1kll

Power Bandwidth

PBW

V,N = 10Vp_p, RL ;" 2kll

Bandwidlh

BW

av,N " 2Vp_p

63

55

Seltling Time

Is

To 01%, ±10V slep

90

100

Capacitive Load Capacily

C LOAD

No Oscilialions

9

MHz

ns
~F

Propagalion Delay

Id

Slep Inpul

ns

Rise Time

I,

av ,N = 0.5V

ns

Wide Band Inpul Noise Vollage

Vn

DC 10 50MHz

en

f = 10kHz

Inpul Noise Vollage Densily

350

400

#lVRMS

50

60

nV/VHZ

NOTE:
1. Sample lesled.

7-7

1/86, Rev. A

-----------l1fHD

BUF-03 HIGH-SPEED VOLTAGE FOLLOWER/BUFFER

TYPICAL PERFORMANCE CHARACTERISTICS
GAIN AND PHASE RESPONSE
vs FREQUENCY

SLEW RATEvs
CAPACITIVE LOAD
500

300
250

~

~

~
W

I-

;i
1il
til

IIII

TA '" 2sOe
Vs '" ±15V

400

-

TA'25°C

LARGE-SIGNAL
FREQUENCY RESPONSE

240

GAIN
Cl = 50pF

28

~0

Ii

I I I:

200

GAIN
CL'" 10pF

/

IlJ.I.

~

---

-2

II
IIII

lOD

-4

PHASE
CL'" 10pF

70

-6

50
100

-8

150

:--

"z
~
":=""
"0

~

l'lj

V

~

120 [:B

~

ew
~

60

1/

~

5:

200
300 400 500
700
CAPACITIVE LOAD (pF)

14

\.

12

. IIII

~

1.0

~

08

z
o

\

-

006
~

\ 1\

2

04

·1

02

30°C!W CASE TO
AMBIENT HEAT
SINK
8JA '" 7SoC/W

1

1

1

40

~

20

f5

~

gj 140

- r--

~;~I~~:~?w

I

I

"~
~

120

i<

~

(Dr

Vs '" ±15V
TA=25"C
ElJA = 90°clW

r-- V
-10

-15

5

10

15

:,....-

GAIN ERROR
vs TEMPERATURE

310

V i--"""

-0.70

Vs = ±15V
ElJA = 900 C!W

~

'"
~
w

!;t

-0.50

-

i'- ~

NEGATIVE

250

...........

~

~

cil

-0.60

r-- ...........

270

230

~

210

,

-2

0

INPUT VOLTAGE (VOL IS)

10

190
-50

-25

0

RL = 1kfl

"""

1-1-"""
RL

'ii
;

"" '"
'-..,

100

-0.40

iii

POSITIVE

25
50
75
TEMPERATURE (OC)

7-8

45

AMBIENT TEMPERATURE (OC)

..........

-6

25

SLEW RATE
vs TEMPERATURE

290

_ _L__~_~
125
65
85

001L-_~_~_~

-5

OUTPUT VOLTAGE {VOLTS)

WARMED UP
TA '" 25°C
Vs '" ±15V

,....

.........-

y

-80

100

80
-10

'--'T""--r---r--.-~T""-"

SAFE OPERATING
AREA

-60

50
75
100 125 150 175 200
AMBIENT TEMPERATURE (OC)

,....

100

V

~

8JA'" 90°C/W

!160

60

INPUT BIAS CURRENT vs
TEMPERATURE (WARMED-UP)
100

""
"":= -20
"0_40

-

180 -

40

/

INPUT BIAS CURRENT
VB INPUT VOLTAGE
200

4
6
10
20
FREQUENCY (MHz)

100

~

NO HEAT SINK

I
25

-

~

45°C!W CASE TO
AMBIENT HEAT SINK
8JA '" 90"C/W

w

60

80

'" '"X I\'

ili

40

OUTPUT CURRENT
vs OUTPUT VOLTAGE
60

\

"

~

-40

4
6 8 10
20
FREQUENCY (MH.:1

\

12

~

0

MAXIMUM POWER
DISSIPATION vs
AMBIENT TEMPERATURE

"

16

6
I

1000

20

~

I I ILl.....•........

150

TA '" 2sOe
SR = 250V / tlseC

~

180

-_
-

vsl •• !svl I I

24

125

ffi2

2'"

I

1-"""

RIL .110:"

-0.30

<1

" -0.20

Vs - ±15V
VIN= ±10V

-0.1 0

ElJA= 90 0 C/W

I

o
-50

-25

25

50

75

I
100

125

TEMPERATURE (Oc)

1/86, Rev. A

-------------t~ BUF-03 HIGH-SPEED VOLTAGE FOLLOWER/BUFFER

TYPICAL PERFORMANCE CHARACTERISTICS
GAIN ERROR VI
INPUT VOLTAGE

NONLINEARITY VI
FULL-SCALE VOLTAGE
50

0.05
TA '" 25°C
Vs = ±15V

_

-

0.03

./
1/

V

/

...........

000
±5

£,

~

RLi?

10

i

>-- I- ,..... ---+-1'::
RL"""a·-

001

z

19

./

-2026

I

10

""- ~

18

15

i2l

15
-50

±24

I

-25

I

I

I

I

-6

-10

Vs
TA=25"C _

I

I I

-2

0

2

10

INPUT VOLTAGE (VOLTS)

OUTPUT SHORT-CIRCUIT
CURRENT VI TIME
-

~

TA = 25"C
:VS=±15V
eJA" g(f'c/W

;: 80

1£

r---....

...........

""::> \'r-...
~
" 70 -r-VIN
Ii
"
<3
"~60
I-

...........

---

0

25

50

75

100

125

TEMPERATURE (OC)

SLEW RATEvs
SUPPLY VOLTAGE

VIN"' -tOV ~

r--_

+10V

-

I-

r-

(SHORT CIRCUIT TO GROUND)-

o

20

50

SUPPLY VOLTAGE (±VOLTSJ

I

,I ±15~-

/
RL = 2000

l

Ri' 5 0a

/'

-100

'"

~

/

90

16

±18

r---

I-"

I

17

±15

/

o

Vs = ±15V
eJA = 90°C/W

iO -

t12

~
i<
~

/

SUPPLY CURRENT
VI TEMPERATURE

I- TA 125.c

:1;9

"V

~ ~RL~ 10kn,_

I-

RL '" 50n

100

-300

21

••

~

/

INPUT VOLTAGE (VOLTS)

21

./'

)'( V

, RL' 200a

,I

200

V

6

SUPPLY CURRENT

,. / '

V V"

/

VI SUPPLY VOLTAGE

./"

300

-200

-10

±10

,;

/

FULL-SCALE (VOLTS)

20

RL = lkn

TA" :ZS-c

V
0.02

~S' ~'5VI

_

RL"'6000 /

GAIN ERROR VI
INPUT VOLTAGE

I I I I I I I I
40

60

80

--

100

120

TIME FROM SHORT (SEC)

SLEW RATE

300

---

TA=25"C

250

I-"'"

¥

~

C

200
RISE

w

~

"
~ 150
oj
100

50
±5

/
FALL

/

V
/

...........

/V /
V

V

±7

±9

.t11

±13

±15

SUPPLY VOLTAGE (VOL T8)

7-9

1/86, Rev. A

-----------I~ BUF-03 HIGH-SPEED VOLTAGE FOLLOWER/BUFFER

HIGH-SPEED 6-BIT AID BUFFER

APPLICATIONS INFORMATION

+15V

OPERATING THE BUF-03 AT REDUCED
POWER SUPPLIES

24~;:------'.!..._--------'''l

OUTPUT

0-----..------..,--_.-----,
Rl

2kn

12k,Q

SAMPLE +15V
LOGIC

r~--~+136V

IN

R2
12k!2
SAMPLE,;;; 0

av

Vl
HOLD -7V

DIGITAL _
GROUND -

R4

{

150n

~

soon TO 6V
TTL GATE LOAD
fAN OUT = 4
NO LOAD

-0.10

0.06

60

-60

120

90

TTL GATE LOAD

FANOUT=4
NO LOAD

60
90
120
RESPONSE TIME (nsec)

OFFSET VOLTAGE
TEMPERATURE

==
=TA=25"C
Vs

10

j°"fN

>

iu

~
o

r--

TYPICAL

""'\!)I I II

010

10

100

':;UI~V

YI

25

50

" ~;"'('
1

Vs = ±15V

,RS = 50n

75

750

~s- ~ 0 -

100

2

"- ~ \.!.l."

""
r---:.

a:
a:
[j300

tfu tx.:

:l
1O

"- t-.
"- t--

~ 200

INPUT BIAS CURRENT
V8 DIFFERENTIAL
INPUT VOLTAGE
r-,-,---r-...,......,-,r-T""'"'-""''''''''''''''''''

-

~

~ ~o 1-+-++--+-l~-l-+-~+~__l100 ffi
a:
a:

a~

Vs+ = 15V

500

7nA CHANGE

200

a

300

:l

1O 400

~

m

~ 300

Vs+= 5V

400

~

t--

~

eMP·01
eMP·01E

~

Z

fooC TO 7cf'C)
2

CMP.Q1C

-25

25

75

TEMPERATURE (OC)

125

Vs = ±lSV

100

---------~~~~~-

f-

o

500

~

~

700

-5~,L2~-_~8~L-_~4~--~L-~4~--~l-J,2
DIFFERENTIAL INPUT VOLTAGE (VOLTS)

5
~

600

TA = 25"C

8-11

-

'----

I

(O"C TO 70"C)
CMP Ole

-25

5V _

_VS+=5V_

25

I

75

125

INPUT VOLTAGE RANGE
VB TEMPERATURE

~ 200
1

CMP-Ol
CMP·01E

Vs+ '" 15V

--vs+

TEMPERATURE (OC)

1 700 I-+-++--+-l~-l-+-..J.-..-+--I--I

I
L

~...........

-75

125

I-

~ 400

Vs+ '" 15V

~

I

~

-25

"" ~ 'lzl

eMP·01
eMp,01E
(O"C TO 70"C)
eMP·Ole

TEMPERATURE (OC)

INPUT BIAS CURRENT
V8 TEMPERATURE
,I5V "

I

001
-50

MATCHED SOURCE RESISTANCE (k,n)

BOO

=2

!!

eMP·01
eMp·01E
eMP·Ole

2

I

1

~=

~

1

I III

!

II

I

I

~

UNNULLED

W

~o

-tl

600

I

,>-

-

1
V:..-

SE

01
10

INPUT OFFSET CURRENT
V8 TEMPERATURE

VB

-15V" Vs-" 0

-W

-g:{

(Vos+ OVERDRIVE)
(Vos- OVERDRIVE)

\,. RISE

150

'0

+15V

,........

-75

2
3.

VIN~FALL

-60

INPUT OFFSET ERROR V8
SOURCE RESISTANCE

100

soon TO SV

~

160

RESPONSE TIME (nsec)

!!

1

! -005

-015

10

CL '" 12pF INCLUDING PROBE
AND JIG CAPACITANCE

~ ~\ CD

>

W

-005

RESPONSE TIME TEST CIRCUIT

~

V+
V+ -05

CC'CI=E::E±:J

--r--

1: : : ]

~V+-l01-__+-+-+_~~~~±-~r-__l

~ V+ -1 5 I-~-+--I-__+-+-;W

~ V+ -2 0 I---+-+-l--l-+----'-l-~f___l

~
W

~V-+201---+-+-+--4-~----'+-~I---l
~V-+15~~:::F~~~--~~~~~__j

.

~ v- +1

0

I---l.-+--I-~-

V-+05~-T-~f-+-_l-+-+__lI___l
-25

75
25
TEMPERATURE I"CI

125

1/86, Rev. A

--------------i~ CMP-01 FAST PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS
RESPONSE TIME
FOR 100mV STEP AND VARIOUS INPUT OVERDRIVES
~

f-

NJ LOA~

VS=±15V
TA=25°C

RS=50n

/

....- -;::? :::.V
/:: / ' /
!/ V /

JV

~

I

?

w

w

.,

§:

"~ -005 r--- r- ,' lmV
'mV
5mV
3
~
20mV

I-

-010

~

1
2
3
4

005

.,

f----' - 4

240

360

10

1125'~

V

~

V

V

V

lmV

120

~~~~ r---

~

L

I-::

2

3
VS= +15V
TA = 25°C

20mV

V

V

llill

01

'0

r---

I

120 , - ; - , - , - , , - - , - - , - ,

f---+-+---j-Vs '" ±15V. Vo

I

f""""

TA=2SoC

V

OUTlpUT LOW

/

v-::

t-- t--

10
20
30
40
OUTPUT SINK CURRENT (rnA)

I

o

50

o

~

8

~~

5V, Vi = ov, Vo = LOW

TEMPERATURE (ee)

RESPONSE TIME
FOR SV STEP AND SmV OVERDRIVE

V

/
~

5

-1

~

-2

~

STROBE

=

~7~5r_~5~0-_~~=--~~0-~2~5r60=--=75r~100~~125

±20

±5
±10
±15
±SUPPLY VOLTAGE (VOLTS)

/

33kn

t---j----j---i---r--t---t---r---j

~ 20 t---j--t-- Vs

NEGATIVE SUPPLY

2kf!
~_---oV+

>:-'~--O OUTPUT

LOW_

~60t---j---j--i--r--t--t--r---j

X

OFFSET TRIMMING AND
STROBE CIRCUIT

=

I--+=I--+=I==t==t=j=:::j

E

V

OUTPUT H , / '

100

+_1
Z80t==J:::~~~;vS~O='~'i5V~'~V~O=O=H~IG~H==l­

'"

'xJOSITIVE SUPPLY

-5SoC

100

POWER CONSUMPTION
VI TEMPERATURE

SUPPLY IS +45V

V

10

SOURCE RESISTANCE (k.l1)

r- MIINIMU~ posiTivE

I--- I--

~

360

240

SUPPLY CURRENT VB
SUPPLY VOLTAGE

25'~

+sv,;;; Vs+,,;;; +15V
-15V";; Vs- -;;;; ov

L

25

,.,
>

OUTPUT

V

~

~

~~=L~~5~-

~ -4

TA = 25°C
RS =50.11-

~

'I

-5

100

200

RESPONSE TIME (nsec)

8·12

4"'
0
~

;;

/
INPUT

~

3l'J

-3

o

>
v-

~.

=>

0

RESPONSE TIME (nsec)

SATURATION VOLTAGE
VI SINK CURRENT

~

RS MATCHED TO 1 0%

1200

12
120

420

RESPONSE TIME (nsec)

A

"~

~ -0 05

1

120

V
V V
......
~~

5mV OVERDRIVE
20mV OVERDRIVE
3
lOamV STEP
NO LOAD

3 "

2 0
>
>1

2mV OVERDRIVE

1

w

>-

1

~ -015

0 10

,

5

Vs = t15V_ 4
T A =25°C
~

\ 1\
1 \
b('> cD 18
l 1 _\

~o

J

12k

~

NO LOAD

RS=50.11_

}~

r;5¢ I¢

B

.......

RESPONSE TIME VI
SOURCE RESISTANCE

1
300

=>
0 0

~5

NJ LOAi -

0

\

~4

"" 3
~

~

INPUT

\

OUTPUT

>-2
=>
~

is

Vs = ±15V

TA =2SoCAS = 50.11

II

1

\

0

:10
~

~

-'.
100

200

300

RESPONSE TIME (nsec)

1/86, Rev. A

--------------1~ CMP-01 FAST PRECISION COMPARATOR

wideband circuits, it is recommended that the supplies be
bypassed near the socket of the device.

APPLICATIONS INFORMATION
The CMP-01 provides fast response times even with small
overdrives; to achieve this performance requires very high
gain at high frequencies. The CMP-01 is completely free of
oscillations; however, small values of stray capacitance from
output to input when combined with high-source resistances
can cause an unstable condition. DC characteristics are not
affected, but when the input is within a few microvolts of the
transition level, certain conditions can create an oscillation
region. The width of this oscillatory region and the size of
source resistance where oscillations begin is a strong function of the stray coupling present. The following suggestions
are offered as a guide towards minimizing the conditions for
oscillation: matched source resistors, minimized stray capacitances (e.g., a ground plane between output and input), or
capacitive output loading (CLl. The capacitive loading techniques will eliminate the oscillations, but result in slower
response time. Matched bypass capacitors across the input
resistors also can eliminate the instability,

MINIMIZING OSCILLATION

BURN-IN CIRCUIT
V+

and if C 2: 20 F (m~~imUm step s~ze)
s
P minimum overdrive
the response time will approximate the response time for low
values of Rs. It should be noted that the offset nulling
terminals do not require bypassing for stability. As with all

-=

V-leASE)
MIL-STD·883. METHOD 1015, CONDITION B

a-BIT TRACKING AID CONVERTER

r - - - - - - - - - - + - - - - - - - - - - - < P - - - - - - - - - - < > CLOCK

IN
MAXIMUM CLOCK RATE'" 3 OMHz

~r~
£\~
1'
_

_

-

--15V

• c

•

,

C

UfO
Ufo

Q

8284
UP/DOWN
COUNTER

,.

9

'3

12 11

6

6

~

•

UP/DOWN

COUNTER

OUT

2 CARRY

,.

3

+5V TRACK

W

.284

9

••

•

7

if

'1/4740.

24.n

~
CLOCK
OUT

C

FOR CLOCK RATE'" 30MHz

C = 470pf

~

,.

""'l~ ,N9'4 ~

.2;:..

'7

......

+15V

::st

DAC-l00cCQ3

\'----+.-::-----r.:-----r:,...-----I,.

R1N!i!!E 4.8kn

~Np~~OG

r-......---.---o+5V

• I. T•
LSB

lN914

'--_____...-_--+__~3r - -.......VIN - 0 TO +10V

9

MSB

10-BIT D/A CONVERTER

::s~

••an

'6

I"

o------,r----t--------'
MAXIMUM FULL SCALE
SINE WAVE INPUT
IS 4000Hz

2

2000

FULL SCALE
ADJUST L - -.....------~-,.V

8·13

1/86, Rev. A

CMP-02
LOW-INPUT- CURRENT
PRECISION COMPARATOR
Precision MOTlohthics Inc

FEATURES

GENERAL DESCRIPTION

•
•
•
•
•
•
•
•
•
•
•
•
•
•
•

The CMP-02 is a monolithic low input current comparator
using an advanced NPN-Schottky Barrier Diode process. It
features superior input characteristics with extremely low
offset voltage, offset current, bias current and temperature
drift. High common-mode and power supply rejection plus
good response time contribute to excellent performance in
the most demanding applications. The balanced offset
nulling, large output drive, and wired-OR capability combined with internal pull-up maximize application convenience. The CMP-02 is capable of operating over a wide range
of supply voltages, including Single plus 5 volt supply
operation, and is pin-compatible to earlier 111,106, and 710
types. For applications requiring faster response time, please
refer to the CMP-01 fast preCision comparator data sheef.

Low Offset Voltage .....•..•... 0.3mV Typ, 0.8mV Max
Low Offset Current ....•......... 0.3nA Typ, 3nA Max
Low Bias Current .......•.••••.. 28nA Typ, SOnA Max
Low Offset Drift .•.•...............••• 1I'VloC,4pA/oC
High Gain ..........•..•...•............. 200,000 Min
High CMRR .....•...........••. 110dB Typ, 94dB Min
High Input Impedance .•..•.••.............•.. _18MO
Fast Response Time •....••.... 190ns Typ, 270ns Max
Standard Power Supplies ..•...... +SVor ±SV to ±18V
Guaranteed Operation from Single +SV
No Pull-Up Resistor Required for TTL Drive
Wired OR Capability
Fits 111, 106, 710 Sockets
Easy Offset Nulling .....•..• Single 2kO Potentiometer
Easy to Use ................... Free from Oscillations

PIN CONNECTIONS
ORDERING INFORMATIONt
v+

PACKAGE
HERMETIC
+25·C
Vos
(mV)

0.8
2.8

B

TO-.
a·PIN

DIP
a·PIN

PLASTIC
DIP
a·PIN

CMP02J'
CMP02EJ
CMP02CJ

CMP02Z'
CMP02EZ
CMP02CZ

CMP02EP
CMP02CP

OPERATING
TEMPERATURE
RANGE
MIL
COM
COM

GNBo
70UT
+IN 2 :

6 BALANCE

-IN 3

5 BALANCE

4
v-teASE)

'For devices processed In total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet.
tAli commercial and industrial temperature range parts are available With
burn-In. For ordering Information see 1986 Data Book, Section 2

8-PIN HERMETIC MINI-DIP
(Z-Sufflx)

TO-99
(J-Sufflx)

EPOXY MINI-DIP
(P-Sufflx)

SIMPLIFIED SCHEMATIC

w~

____

~

____

~

__

~~

____

~

____

~

____

~

____

~

____

~

____

~

____

~

INPUT

INVERTING

v-~----*-----*---~

8-14

1/86, Rev, A

----------I1fMD

CMP-02 LOW-INPUT-CURRENT PRECISION COMPARATOR

ABSOLUTE MAXIMUM RATINGS (Note 2)
Storage Temperature Range ••••••••••••• -65·C to +150·C
P-Suffix ............................. -65·C to +125·C
Lead Temperature (Soldering, 60 sec) •••••••••••••• 300·C
Output Short-Circuit Duration - to ground ••••••• Indefinite
to V+ ........... 1 Minute
NOTES:
1. Maximum package power dissipation va. ambient temperature.

Total Supply Voltage, V+ to V- ..................... 36V
Output to Ground ••••••••••••••••••••••••• -5V to +32V
Output to Negative Supply Voltage. • • • • • • • • • • • • • • • •• 50V
Ground to Negative Supply Voltage •••••••••••.•.••• 30V
Positive Supply Voltage to Ground .••••••••••••••••• 30V
Positive Supply Voltage to Offset Null ••••••••••.• 0 to 2V
Power Dissipation (See Note 1) ••••••••••••••.••• 500mW
Differential Input Voltage .......................... ±llV
Input Voltage (Vs = ± 15V) ••••••••••••••••••••••••• ± 15V
Output Sink Current (Continuous Operation) ••••••• 75mA
Operating Temperature Range
CMP-02 ............................ -55· C to + 125· C
CMP-02E, CMP-02C .................... 0·Cto+70·C
DICE Junction Temperature (Tj) ••••••• -65· C to + 150· C

ELECTRICAL CHARACTERISTICS at Vs

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

PACKAGE TYPE
TO-99 (J)

SO"C

7.1mW'"C

Mini-DIP (P)

36"C

5.6mW'"C

Hermetic Mini-DIP (Z)

75"C

6.7mW'"C

2. Absolute ratings apply to both DICE and packaged parts, unless otherwise
noted

= ± 15V, TA = 25· C, unless otherwise noted.
CMP-02
CMP-02E

PARAMETER

SYMBOL CONDITIONS

Input Ollset Voltage

Vos

Input OIIoet Voltage

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

CMp·02C

TYP

MAX

TYP

MAX

UNITS

RsSSkO. (Note 1)

0.3

08

04

28

mV

Vos

Rs S 50kO. (Note 1)

0.3

0.9

0.4

3

mV

Input OIIset Current

los

(Note 1)

03

30

0.4

15

nA

Input Bias Current

18

28

50

35

100

nA

Vl

MO

0

Differential Input

MIN

Resistance

R'N

(Note 2)

17

Voltage Gain

Av

Vo=0.4Vto 2.4V, (Notes 1. 2)

200

tr

100mV step, 5mV Overdrive
No Load (No Pull-Up)
5kO to 5V (Pull-Up)
TTL Fan-Out = 4. No Pull-Up

Response Time
(Note 3)

MIN

09
500
190
190
190

100
270

500
190
190
190

VlmV

~

~
I=l-<

270
ns

~

U
15

Input Slew Rete

15

VI".s

Input Voltage Range

CMVR

±12.5

±13.0

±125

±13.0

V

Common-Mode
Rejection Ratio

CMRR

94

110

90

110

dB

Power Supply
Rejection Ratio

PSRR

5V S Vs+ S 18V,
-18VSVs_SOV

SO

100

74

98

dB

Positive Output
Voltage

Y,N '" 3mV, 10 = 320".A
V,N ",3mV, 10= 24O".A
V,N ",3mV,1 0 =OmA

24

3.2

VOH

4.8

2.4
24

34
48

V

24

Saturation Voltage

VOL

Y,N S -10mV. ISink = OmA
Y,N S -10mV, ISink S 6.4mA
Y,N S -10mV. Isink S 12mA (CMP-{)2 only)

016
03
038

040
0.45
05

016
031

0.40
0.45

V

Output Leakage Current

ILEAK

V,N'" 10mV. Vo =+30V

003

2.0

0.05

8.0

".A

Positive Supply Current

1+

V,N S-l0mV

5.5

80

5.6

8.5

mA

Negative Supply Current

1-

V,N S-l0mV

1.1

2.2

1.2

2.2

mA

Pd

V,N S-l0mV

99

153

102

161

mW

Nulling Pot'" 2kO

±5

Power Dissipation
Offset Voltage
Adjustment Range

NOTES:
1. These Parameters are specified as the maximum values required to drive
the output between the logiC levels ofO.4Vand 2.4V With a 1kO load tied to
+5V; thus, these paremeters define an error band which takes Into

±5

mV

account the worst case effects of voltage gain and Input impedance.
2. Guaranteed by deSign.
3 Sample tested

8-15

1/86, Rev. A

Pol

~
tj
~

IfMD

CMP-G2 LOW-INPUT-CURRENT PRECISION COMPARATOR

ELECTRICAL CHARACTERISTICS at Vs = 5V,

VS-

= OV, TA = 25

0

C, unless otherwise noted.
CMP-G2
CMP-02E

PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vas

Rs S5kO. (Note 1)

Input Ollset Current

los

(Note 1)

Input Bias Currenl

Is

Voltage Gain

Av

Vo=0.4V to 2.4V, (Notes 1. 2)

Response Time

tr

l00mV Step, 5mV Overdrive
5kO to 5V (Pull-Up)
TTL Fan-out = 4, 5kO to 5V

Input Voltage Range

MIN

MAX

TVP

MAX

UNITS

0.4

1.5

0.5

3.5

mV

0.25

3

0.35

14

nA

24

45

30

90

50

250

250

250

250

1.8-3.5 1.7-3.8

CMVR
VOL

0.3

Positive Supply Current

1+

VIN S-l0mV

2.2

Power Dissipation

Pd

VIN S-l0mV

11

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, -55 0 C ::; TA::;

1250

MIN

50

VIN S -3.5mV, I oink S 6.4mA

Saturation Voltage

CMP-G2C

TYP

na

1.8-3.5 1.7-3.8

0.45

nA
VlmV

V

0.3

0.45

3

2.3

3.6

mA

15

11.5

18

mW

V

C, unless otherwise noted.
CMP-G2

PARAMETER

SYMBOL CONDITIONS

MIN

TVP

MAX

UNITS

0.4

1.6
2.8

mV

Input Offset Voltage

Vas

RsS 5kO, (Note 1)
Vs+=5V, Vs-=OV, (Note 1)

Average Input Offset
Voltage Drift
Without External Trim
With External Trim

TCVos
TCVoSn

Rs=500
Rs=500

1.5
1.0

Input Offset Current

los

TA= +125'C, (Note 1)
TA =-55'C, (Note 1)

0.3
0.4

Average Input Offset
Current Drift

TClos

+25'Cs TAS +125'C
-55'CSTA S+25'C

2
4

Input Bias Current

Is

TA=+125'C
TA =-55'C

Voltage Gam

Av

Vo= O.4V to 2.4V, (Notes I, 2)

tr

l00mV Step, 5mV Overdrive
TA = +125'C, No Load
TA = -55' C, No Load

Response Time

0.5

25
45
100

pvrc

4
12

nA
pArC

50
120

500

nA
VlmV

310

ns

155

Input Voltage Range

CMVR

±12

±13

V

Common-Mode
Rejection Ratio

CMRR

88

106

dB

Power Supply
Relection Ratio

PSRR

5VS Vs+S 15V, -15V S Vs_S OV

75

96

dB

Positive Output Voltage

VOH

VIN~4mV,lo=200pA

VOL

VINS-l0mV,lsonk=OmA
VIN S-l0mV, Isonk = 6.4mA

Saturation Voltage

2.4

3
0.20
0.32

V
0.4
0.5

V

NOTES:
1. These Parameters are specified as the maximum values required to drive
the output between the logic levels of 0.4 V and 2.4V with a 1kO load tied to
+5V; thus, these parameters define an error band which takes Into
account the worst case effects of voltage gain and Input Impedance.
2. Guaranteed by design.
3. Sample tested.

8-16

1/86, Rev. A

----------I~ CMP-GZ LOW-INPUT-CURRENT PRECISION COMPARATOR

ELECTRICAL CHARACTERISTICS at Vs = ±15V, O·C S

TAS

70·C, unless otherwise noted.
CMP-02E

PARAMETER

MIN

SYMBOL CONDITIONS

MAX

0.4
0.5

1.4
2.4

Input Offset Voltage

Vos

Rs S 5kO. (Note 1)
Vs+=5V. Vs-=OV, (Note 1)

Average Input Offset
Voltage Drill
Without External Trim
With External Trim

TCVos
TCVoSn

Rs= SOO
Rs= SOO

1.5

Input Offset Current

los

TA = +70·C, (Note 1)
TA=O·C, (Note 1)

0.3
0.4

Average Input Offset
Current Drill

TClos

+25·CSTA S+70·C
0·CSTA S+2S·C

2
4

I nput Bias Current

Ie

TA =+70·C
TA=O·C

Voltage Gain

Av

Vo = 0.4V to 2.4V, (Notes I, 2)

Response Time

t,

l00mV Step, 5mV Overdrive
TA = +70·C, No Load
TA = O·C, No Load

26
34
100

CMP-02C

TVP

MIN

TVP

MAX

UNITS

0.5
0.6

3.5
4.3

mV

1.8
1.2
0.4
0.5

3
6

I'VI·C
15
25

3
5
SO
60

33
42
70

500
225
160

nA
pAl·C

100
160

nA

500

VlmV

225
160

ns

Input Voltage Range

CMVR

±12.0

±13

±12

±13

V

Common-Mode
Rejection Ratio

CMRR

90

loe

66

loe

dB

Power Supply
Rejection Ratio

PSRR

5V S Vs+ S 1SV, -15V S Vs-S OV

77

98

70

88

dB

Positive Output Voltage

VOH

Y,N 2: 4mV, 10 = 200l'A

2.4

3.2

2.4

32

VOL

Y,N S -10mV, Is,nk = 0
V'N S -10mV, Is,nk = 6.4mA

Saturation Voltage

0.17
0.30

0.17
0.31

0.4
05

V
0.4
0.5

V

~

0

~

NOTES:
1. These Parameters are specified as the maximum values ,equired to drive
the output between the logic levels of O.4V and 2.4V with a 1kO load tied to
+5V; thus, these parameters define an error band which takes into
account the worst case effects of voltage gain and input impedance.
2. Guaranteed by design.
3. Sample tested.

1=1-<

~

U

~

~
tj
~

1/86, Rev. A

S-17

~~~---.--.-.

'-"'--.-.-.

-. .

-'~'-'---'-'---'

-.-

----------I~ CMPo02 LOW-INPUT-CURRENT PRECISION COMPARATOR

DICE CHARACTERISTICS

1.
2.
3.
4.
5.
8.
7.
8.

DIE SIZE 0.085 X 0.042 Inch, 2730 sq. mill
(1.851 X 1.089 mm, 1.781 Iq. mm)

WAFER TEST LIMITS at

Vs= ±1SV, TA=2S"C.

PARAMETER

SYMBOL

CONDITIONS

Vos

Rs :55kO
Rs:5SOkO

Input Offset Voltage
Input Offset Current

los

Input Bias Current

Ie

Differential Input Resistance

RIN

Input Voltage Range

CMVR

Common-Mode Rejection Ratio

CMRR

Power Supply Rejection RatiO

GROUND
NONINVERTING INPUT
INVERTING INPUT
NEGATIVE SUPPLY (SUBSTRATE)
BALANCE
BALANCE
OUTPUT
POSITIVE SUPPLY

For additional DICE inlormallon refer to
1986 Data Book, Section 2.

CMP-02N

CMP-02GR

LIMIT

LIMIT

UNITS

0.8
0.9

2.8
3

mVMAX

3

15

nAMAX

50

100

nAMAX
MOMIN

1.7

0.9

±12.5

±12.5

VMIN

VCM=±CMVR

94

90

dBMIN

PSRR

5V:5 Vs+:5 18V
-18V:5 Vs -:5 OV

80

74

dBMIN

Positive Output Voltage

VOH

VIN ",3mV, 10= 320l'A
V,N ", 3mV, 10 = 240I'A

Saturation Voltage

VOL

I.,nk = S.4mA

Output Leakage Current

I LEAK

VIN'" 10mV, Vo = 30V

2.4
2.4

VMIN

0.45

0.45

V MAX

2

8

I'AMAX
mAMAX

Positive Supply Current

1+

V1N :5-10mV

8

8.5

Negative Supply Current

1-

V1N :5-10mV

2.2

2.2

mAMAX

Power Consumption

Pd

VIN:5-10mV

153

161

mWMAX

WAFER TEST LIMITS at

Vs + = SV

CMP-G2N

CMP-G2GR

PARAMETER

SYMBOL

CONDITIONS

LIMIT

LIMIT

UNITS

Input Offset Voltage

Vos

Rs :55kO

1.5

3.5

mVMAX

Input Offset Current

los

3

14

nAMAX

and Vs -

=

av, TA = 2S"C.

NOTE:
Electrtcal tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging Is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs =
PARAMETER

SYMBOL

Average Input Offset
Voltage Drtft

TCVos

Average Input Offset
Current Drill

TClos

Response Time

tr

±15V.

CONDITIONS
Rs=SOO

l00mV Step, 5mV Overdrtve
No Load (No Pull-Up), TA = 25°C

8-18

CMP-02N

CMP-02GR

TYPICAL

TYPICAL

UNITS

1.5

1.8

I'VI"C

4

5

pArC

190

190

ns

1/86, Rev. A

-------------I~ CMP-02 LOW-INPUT-CURRENT PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS
RESPONSE TIME,
100mV STEP, 5mV OVERDRIVE, VARIOUS LOADS

RESPONSE TIME TEST CIRCUIT

5~

CL = 1ZpF INCLUDING PROBE

~",,":-+---l----+---4 4 0

AND JIG CAPACITANCE

~
w

3

~

+-t------+-----l'~_l'<_A~---12 ~

t;

~o

~

~

~

~ -005
~ -010
~

::>

o

010

w

w

it

, :=

+-+---+--j Vs
TA
+_+---+_-j RS

~

= i15V
= 2SOC

=

0.05

~

son

>-

it'!: -005

-0.15
-60

0

1:lQ

'80

'50

INPUT OFFSET ERROR VB
SOURCE RESISTANCE

OFFSET VOLTAGE
TEMPERATURE

r

2. CMP·02C

10

>-

~o

1=
F--

>-

itz
- '0

~

/

IIIIII
WORST CASE

o

>

2

~

-I
1

TYPICAL

0001

0010

.... r-

~~2

==

TA

2S"C

~YI
-25

25

MATCHED SOURCE RESISTANCE (MD.)

,l\..
"'-

40

30

""
~
.........

"U);

~

\2..i r-

20

t-

1
2

1
-75

I
-25

I

=

75

lOa

125

25
TEMPERATURE (OC)

75

-75

V+
V+ -0 5

,

125

-25

0

~

V+ -1 0

~

75

V+-l 5

---

1
~
: Vs+

1

w

~
3
Vs = ±15V
TA = 2SoC

-4

8-19

~

+1 5

V-+l 0
V-+05

57

12

DIFFERENTIAL INPUT VOLTAGE (VOLTS)

V- +2 0

g v;:

-f--

III
-8

125

:-1 Vs+ = 5V

w

;2
30

-12

Vs+ i5V

25

~ V+ -2 0

'0

r- -15V';;; VS-.;;; OV

~~

Vs+ i15V -

INPUT VOLTAGE RANGE
VB TEMPERATURE

20

1 I
J I

I.-.,..;..IVS+=SV -

TEMPERATURE (OC)

40

t:-i+

"

Vs+= 15V

-(i)

IRS," 50"

50

CMp·02
CMP·02E

1

\

w: :::::::

±15V

VS+=5V -

g~=:g~E
(O°c TO 70°C)
CMP·02C
-

I

50

Vs+ = 5V

'0 ~

Vs

- -

60

VS+=lSV _

1--1

~ 200

INPUT BIAS CURRENT
vs DIFFERENTIAL
INPUT VOLTAGE

I I
1 I

Vs+ = 15V

>-

TEMPERATURE (Oel

INPUT BIAS CURRENT
VB TEMPERATURE
50

o

2

2 CMp·02C

~ I-.."=:

~

CMP-02C

;;:

00'
-50

~400

WCT070°C)

I\0'-UIL~~

100

II

1 CMP·02
CMP·02E
(DOC TO 70°C)

1\.'\
'\

13

1

;;:

100

~

II

CMP·02

If-

;;:
±15V

1111'

0100

010

r\.

;: 600

CMP·OZE

>-

Vs

,

f--- f---

i

21""

o

1

2

I IIIIII

0'

INPUT OFFSET CURRENT
VB TEMPERATURE

VB

6-1-(, .::.-

UNNULLED

"~

;;

~

{Vas + OVERDRIVE I
(Vas - OVERDRIVE)

800

:;.§

~

rFALL
VIN-I.. RISE

'0

~' g~=:~~E

a:

210

'80

RESPONSE TIME (nsec)

o

ffi

'50

RESPONSE TIME (nsec)

'00

~a:

120

-60

210

=

15V

:1

1

~

I

1

~---I---l--l-----/--+-----''-I-----~--l
1---1===t=:::1=-1-~~d::::-I---j

--L- -15V"';;; VS-';;; OV
+5V;;;; VS+';;;; +15V

V- '--'-_---'-_l-----'_....L_..L_L.........J
-75
-25
0
25
75
125
TEMPERATURE re)

1/86, Rev. A

----------l~ CMP-02 LOW-INPUT-CURRENT PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS

SUPPLY CURRENT VI
SUPPLY VOLTAGE

SATURATION VOLTAGE
VI SINK CURRENT
'0

1125o~

V

V

.:.,.

V
V V
,.....
~

V

l-- SUPPLY

25°~

V -55 C
,.......
IQ

10

20

30

40

OUTPUT SINK CURRENT

'00 f--f_+--:C;;..:,VS::.=::.':,:';:5V:;.,.:V;;,O.:=..:L::;OW;:"';;;;j

i.--

10<-

....... Ve"OSITIVE SUPPLY

/

o

50

lmAI

Y

V

V":

- f-- f-

~~~~~~:~~1~~

I
l?

OUr

-

r---r-..-----r-..---I--r-..-----r~

I

I

TA'" 2SoC

V

i...--::i ~

If'

'20

M'~'MU~ISPOS'~'VE
+45V

V

POWER CONSUMPTION
VI TEMPERATURE

OUTPUT HIGH

NEGATIVE SUPPLY

!

o

±5

±10

±15

-75

±20

'20

<"

T A '" 25°C
Vs = ±15V

100

S

ilia:

80

\ t-...

a:

"">-

:;

"a:
Ii
>a:

f.-..

Ti-!-

......

I'.

60

ill

V-leASE}

OUTPUT

~t= 10 SEe

~M~

40

33kn
STROBE

I

20

MIL-STD-883, METHOD 1015, CONDITION B

l0

>--+-----0 V+

t = 1 SEC

0

I.

I-

furUI TSIHOrTyD fT I' = 10

'0

0

'5

OUTPUT VOL rAGE (VOLTS)

RESPONSE TIME VI
SOURCE RESISTANCE

RESPONSE TIME,
100mV STEP AND VARIOUS INPUT OVERDRIVES
2
3
4

I
20mV

V

~

12,000

;:::

/
/
V
(/) rJ- J$) i¢ ,
/V/

/

~

-/

.....--::

V

/

w

~~ =L~~5~-

~ -0.10

TA = 25"C

~

-015
-60

120

TA=25°C

As = son

~0

'~"

c'4; ~

\

\

0,10

0.05

i<

4~

\

3~
2g

0,' ~

>-

,.

100mV STEP
NO LOAD
RS MATCHED TO 1.0%

~

r\

,0

0

I I
RESPONSE TIME {nsec}

\

\

\

\

"'
>

360

\

NO LOAD \
Vs = ±15V

AS=50n -

240

\

1\

2:

'~" -005
i<

'25

I-

I

I'

>-

'mV
2mV
5mV

I
75

I I

.....

V+

I

I
25

OFFSET TRIMMING AND
STROBE CIRCUITS

OUTPUT SHORT-CIRCUIT
CURRENT VI OUTPUT VOLTAGE

STANDARD BURN-IN CIRCUIT

,

-25

± SUPPLY VOLTAGE (VOLTS)

0
0"

2
3.

'mV
2mV
5mV

4.

20mV

I,.

1200

w

;::
w

k1

~a:

1=',
~~

420

'2

-005

-60

120

240

RESPONSE TIME (nsec)

8-20

360

420

2l.,... 3

'20

>~

./

I-

!Ii

2mV OVERDRIVE
5mV OVERDRIVE
20mV OVERDRIVE

TTTTIIlf Tn
0'

Vs = :t15V

TA I=

10
10
SOURCE RESISTANCE (knJ

~~CI
'00

1/86, Rev. A

----------I~ CMP-02 LOW-INPUT-CURRENT PRECISION COMPARATOR

APPLICATIONS INFORMATION
The CMP-02 provides fast response times even with small
overdrives; to achieve this performance requires very high
gain at high frequencies. The CMP-02 is completely free of
oscillations; however, small values of stray capacitance from
output to input when combined with high-source resistances
can cause an unstable condition. DC characteristics are not
affected, but when the input is within a few microvolts of the
transition level, certain conditions can create an oscillation
region. The width of this oscillatory region and the size of
source resistance where oscillations begin is a strong function ofthe stray coupling present. The following suggestions
are offered as a guide towards minimizing the conditions for
oscillation: matched source resistors, minimized stray capacitances (e.g., a ground plane between output and input), or
capacitive output loading (CLl. The capacitive loading techniques will eliminate the oscillations, but result in slower

response time. Matched bypass capacitors across the input
resistors also can eliminate the instability,
and if C :::: 20 F (m~~imUm step s~ze)
S
p
minimum overdrive
the response time will approximate the response time for low
values of Rs. It should be noted that the offset nulling
terminals do not require bypassing for stability. As with all
wideband circuits, it is recommended that the supplies be
bypassed near the socket of the device.

PRECISION, DUAL LIMIT, GO/NO GO TESTER
+15V

V L '" 32V
IL < 75mA

UPPER
LIMIT

MINIMIZING OSCILLATION

INPUT

LAMP

-15V

OUTPUT

•

LOWER
LIMIT
-15V

8-21

1/86, Rev. A

CMP-04
QUAD lDW-POWER
PRECISION COMPARATOR
PrecisIon Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

•
•
•
•
•
•
•
•
•

Four preCision independent comparators comprise the CMP04. Performance highlights include a very low offset voltage,
low output saturation voltage and high gain in a single supply
design. The input voltage range includes ground for single
supply operation and V- for split supplies. A low power
supply current of 2mA, which is independent of supply
voltage, makes this the preferred comparator for precision
applications requiring minimal power consumption. Maximum logic interface flexibility is offered by the open-collector
TTL output.

High Gain ............................. 200VlmV Typ
Single or Dual Supply Operation
Input Voltage Range Includes Ground
low Power Consumption (1.5mW/Comparator)
low Input Bias Current .................... 100nA Max
low Input Offset Current ................... 10nA Max
low Offset Voltage ......................... 1mV Max
low Output Saturation Voltage ...•.... 250mV @ 4mA
logic Output Compatible with TTL, DTl, ECl, MOS and
CMOS
• Directly Replaces lM139/239/339 Comparators

PIN CONNECTIONS
ORDERING INFORMATIONt
2S"C
Vos
(mV)

DIP PACKAGE
HERMETIC
14-PIN

PLASTIC
14-PIN

OPERATING
TEMPERATURE
RANGE

CMP04FP

MIL
INO
COM

CMP04BY'
CMP04FY

14-PIN HERMETIC DIP
(V-Suffix)
14-PIN EPOXV DIP
(P-SuHlx)

• For devices processed In total compliance to MIL-STD-883, add /883 after
part number Consult factory for 883 data sheet
tAli commerCial and mdustrial temperature range parts are available with
burn-In, For ordering information see 1986 Data Book, Section 2

SIMPLIFIED SCHEMATIC (1/4 CMP-04)

TYPICAL INTERFACE
Driving CMOS
50

v.
100kn

OUTPUT

r ......-f--+<>-INPUT

Driving TTL
50

*SUBSTRATE DIODES

8-22

1/86, Rev_A

----------I~ CMP-04 QUAD LOW-POWER PRECISION COMPARATOR

ABSOLUTE MAXIMUM RATINGS

(Note 2)

Input Current (VIN

< -3.0V) •••••••••••••••••••••••• 50mA

36V or±18V

Output Short-Circuit to GND •••••••••••••••• Continuous

Differential Input Voltage •••.••••••••••••••••••••• 36VDC

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300· C
NOTES:
1 See table for maximum ambient temperature rating and derating factor.
2. Absolute maximum ratings apply to both packaged parts and DICE, unless
otherwise noted.

Supply Voltage •••.••••.•••..•••••••••.••••

Input Voltage ••••••.••..••••..••••••.••••• -0.3V to +36V
Power Dissi pation (Note 1) ••••••••.•••••••••...• 500mW
Operating Temperature Range
CMP-04FY ••••••••••••••••••••••••••

-25° C to +85° C
to + 70· C

PACKAGE TVPE

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

DICE Junction Temperature (Tj) •••••••
Storage Temperature Range. • • • • • • • . ••

-65°Cto+150°C
-65° C to + 150° C

Hermetic DIP IV)

100'C

10mW/'C

P-Suffix • • • • • • • • • • • • • • • • • • • • • • • • • • ••

-65° C to + 125· C

50'C

6mW/'C

CMP-04BY •••••••••••••••.•••••••••

-55° C to + 125° C

CMP-04FP •••••••••••••••••••••••••••••

O· C

Plastic DIP IP)

ELECTRICAL CHARACTERISTICS at V+ =

+5V, TA

PARAMETER

SVMBOL

CONOITIONS

Input Offset Voltage

Vas

Rs=OO, RL =5.1kO
Vo = 1.4V, INote 1)

Input Offset Current

los

I,NI+) -I'NI-)
RL =5.1kO
Vo= 1.4V

= 25· C, unless otherwise noted.

CMP-04B/F
MIN

Input Bias Current

Ie

I,NI+) or I'NI-)

Av

RL " 15kO, V+ = 15V, INote 5)

Large-Signal Response Time

t,

Y'N = TTL Logic SWing
VAEF = 1 4V, INote4)
VAL = 5V, RL = 5.1kO

Small-Signal Response Time

t,

Y,N = 100mV Step, (Note 4)
5mV Overdrive
VAL = 5V, RL = 5.1kO

Input Voltage Range

CMVR

INote2)

CMRR

INotes 3,5)

Power Supply Rejection Ratio

PSRR

Saturation Voltage

VOL

MAX

0.4

25
80

100

V+ = +5V to 18V, INote 5)

nA
nA

200

VlmV

300

ns

C/.l

~

1.3

I'S

V

80

100

dB

80

100

dB

= 0,

2S0

400

mV

'SINK:54mA

Output Sink Current

ISINK

V,NI-) " 1V,
V,NI+) = 0, V0 51 SV

Output Leakage Current

ILEAK

V,N (+) ,,1V,
V,NI-) = 0, Vo = 30V

01

100

nA

1+

RL =~, All Camps
V+ = 30V

08

20

mA

Supply Current

6

16

mA

NOTES:
1. At output SWitch pOint, Vo = 1 4V, Rs = 00 with V+ from SV; and over the
full input common-mode range 10V to V+ -1.SV).
2. The Input common-mode voltage or either Input signal voltage should not
be .IIowed to go negative by more than 0.3V. The upper end of the
common-mode voltage range IS V+-1.5V, but either or both Inputs can go
to +30V without damage.
3. RL " 1SkO, V+ = 15V, VCM = 1.SV to 13.5V.
4. Sample tested.
5, Guaranteed by design,

8-23

•
0

V+-1.5

v,NI-) " 1V, V,NI+)

UNITS
mV

10

Voltage Gain

Common-Mode Rejection Ratio

TYP

1/86, Rev_ A

~

~

~

U

>LI

~
tj
~

-----------f!fMD

CMP-04 QUAD LOW·POWER PRECISION COMPARATOR

ELECTRICAL CHARACTERISTICS at V+ = +5V. For CMP-04BY, -55 0 C::; TA::; 1250 C. For CMP-04FY, -25 0 C::; TA::; 85 0 C.
For CMP-04FP, 00 C ::; TA::; 70 0 C, unless otherwjse noted.
CMP-04B/F

(Nole 3)
PARAMETER

MIN

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs = on. RL = 5.1kn
vo= 1.4V, (Note 1)

'nput Offset Current

'os

"N(+) -"N(-)
RL = 5.1kn
Vo = l.4V

Input Bias Current

'8

"N(+) or "N(-)

Vo'tage Gai n

Av

RL " 15kn. V+ = 15V, (Note 5)

Large-Signa' Response Time

t,

V'N = TTL Log'c Swing
VREF = 1.4V. (Note 4)
VRL = 5V, RL = 5.1kn

Small-Signal Response Time

t,

V'N = 100mV Step. I Note 4)
SmV Overdrive
VRL = 5V. RL = 5.1kn

TYP

MAX

UNITS
mV

70

4

20

40

200

CMVR

INote2)

CMRR

INotes 3. 5)

60

Power Supp'y Rejection Ratio

80

nA

125

VlmV

300

ns

1.3

I'S

0

'nput Voltage Range
Common-Mode Rejection Ratio

nA

V+-1.5
100

V
dB

PSRR

V+ = +5V to 18V

Saturation Voltage

VOL

V'N(-)" 1V, V'NI+) = O.
IS'NK,,4mA

100

Output Sink Current

'SINK

V'NI-)" 1V.
V'NI+) =0, Vo ,,1.5V

Output Leakage Current

'LEAK

V'NI+)" 1V.
V'N'-) = 0, Vo = 30V

0.1

200

nA

Supply Current

1+

RL =~. All Comps
V+= 30V

1.2

3.0

rnA

250

5

dB
700

16

mV

rnA

NOTES:
1. At output switch pOint. Vo=l.4V. Rs=Onwith V+from5V; andoverthefull
input common-mode range (OV to V+ -1.5V)
2. The input common-mode voltage or either Input signal voltage should not
be allowed to go negative by more than 03V. The upper end of the
common-mode voltage range is V+-1.SV, but either or both inputs can go
to +30V without damage.
3. R L " 15kn, V+ = 15V. VCM = 1.5V to 13.5V.
4. Sample tested.
5. Guaranteed by design

8-24

1/86, Rev. A

------------t~ CMP-04 QUAD LOW-POWER PRECISION COMPARATOR

DICE CHARACTERISTICS

1.
2.
3.
4.
5.
6.
7.

OUTPUT (2)
OUTPUT (1)
POSITIVE SUPPLY
INVERTING INPUT (1)
NONINVERTING INPUT (1)
INVERTING INPUT (2)
NONINVERTING INPUT (2)

8.
9.
10.
11.
12.
13.
14.

INVERTING INPUT (3)
NONINVERTING INPUT (3)
INVERTING INPUT (4)
NONINVERTING INPUT (4)
GROUND (SUBSTRATE)
OUTPUT (4)
OUTPUT (3)

For additional DICE information refer to
1986 Data Book, Section 2.
DIE SIZE 0.052 x 0.056 Inch, 2912 sq. mils
(1.32 x 1.42 mm, 1.88 sq. mm)

WAFER TEST LIMITS at V+ = +5V, TA = 25°C, unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Rs=OO, RL =5.1kO
Va = 1.4V, INote 1)

Input Offset Current

los

I'NI+) -I'NI-)
R L =5.1kO
Vo=1.4V

Input Bias Current

18

I'NI+) or
I'NI-), INote 1)

CMP-04N

CMP-04G

LIMIT

LIMIT

UNITS
mVMAX

Voltage Gain

Av

R L " 15kO, V+ = 15V, (Note 3)

Input Voltage Range

CMVR

INotes 2, 3)

Common-Mode Rejection Ratio

CMRR

Power Supply Rejection Rat,o

PSRR

Saturation Voltage

VOL

10

25

nAMAX

100

100

nAMAX

80

50

V+-1.5

V+-15

VMAX

INote4)

80

80

dBMIN

V+ = 5V to +18V

80

80

dBMIN

~

400

400

mVMAX

0

6

mAMIN

100

nA MAX

2

mAMAX

V'NI-)" 1V, V'N(+) = 0,
I SINK ::54mA

Output Sink Current

ISINK

V'NI-)" 1V,
V'NI+) =0, VaS 1.5V

Output Leakage Current

ILEAK

V'NI+) ,,1V,
V'NI-) = 0, Va = 30V

Supply Current

1+

RL =~, All Comps
V+ = 30V

100

VlmVMIN

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to vanatlons In assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testmg

TYPICAL ELECTRICAL CHARACTERISTICS at V+ = +5V, unless otherwise noted.
PARAMETER

~

p:;

0

CMP-04N

CMP-04G

TYPICAL

TYPICAL

UNITS

Ir

V'N = TTL LogiC Swing
VREF = 14V, I Note 5)
VRL =5V, RL =51kO

600

600

ns

Ir

V'N = 100mV Step, (Note 5)
SmV Overdrive
VRL =5V, RL =5.1kO

1.3

1.3

I'S

SYMBOL

CONDITIONS

Large-Signal Response Time

Small-S'gnal Response Time

common-mode voltage range IS V+-1.5V, but either or both Inputs can go
to +30V without damage

NOTES:
1. At output switch pOint, Vo= 1.4V, Rs=OO with V+ from 5V; and overthe
full input common-mode range 10V 10 V+ -1.5V)
2. The input common-mode voltage orelther input signal voltage should not
be allowed to go negative by more than 0.3V. The upper end of Ihe

3. Guaranteed by design.
4. RL " 15kO. VCM = 1.5V to 13.5V
5 Sample tested.

8-25

1/86, Rev. A

Po<

~

U
r..t.l

~
b
§2

_ _ _ _ _ _ _ _ _---;~ CMP-04 QUAD LOW-POWER PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS
OFFSET VOLTAGE
VI TEMPERATURE

+0.2

......

~ +0.1

~
g

~

""'

~ -0.1

r-

t\..

I

't\..

-0.2

I

~ 140
~130

10

100 120 140

~ 100
I

09

"

-

07

~

J

16

20

26

30

36

-3.0

40

'to...

~
o

-40 -20

...-

«> 80

40

80

-

80 100 120 140

OUTPUT VOLTAGE VI
OUTPUT CURRENT AND
TEMPERATURE

- -TA =

,..

-t.Jc-

o"c

...-

TA = +70"C

.....

.

03

20

I--'"

-55'5,..-

...

,

05

I

90

-

~

TEMPERATURE (OC)

SUPPLY CURRENT
VB SUPPLY VOLTAGE

l\.

1/

~ 110

r--1.0

V+ - SUPPLY VOLTAGE Noel

TA

'\.

~
~

0

Jl-2.0

11

I

~ 120

g

r-..

~

1.0

a:

""

..... 1--

0 20 40 80 80
TEMPERATURE (OCI

160

2.0

I

VOLTAGE GAIN
VI TEMPERATURE
150

TA • 2"Cl7lfC
"-

TA"lfC

-0.3

-40 -20

~
~
a:

~ ~,.;,

1\

:g
>

3.0

80

+0.3

~

INPUT OFFSET CURRENT
VI TEMPERATURE

INPUT BIAS CURRENT
VI V+ AND TEMPERATURE

~,2.·C

70
80
--40 -20

0
20 40 60 80
TEMPERATURE ("el

100 120 140

10

20

6.0

~

'...

g~

i~

4.0

3.0

~o 2.0

0>

~, ~ '~"

w

~gs '"

5.0

"~

4.0

\

r'~

I

- > -100

'f I

1.•

-

..

f- I- TA"2"C .
I I I

I.

TA -25"C
1.0

VIN

I

I I

i~-60

5.1K

~

I I

5mV

II

0> 1.0

0

I

I

3.0

~~2.0

V OUT

,~

0••

w

g~
56

IN

...

100

INPUT OVERDRIVE = lOOmV

100mV

1.0

,

••0

+5Voc

. '

10

RESPONSE TIME FOR
VARIOUS INPUT
OVERDRIVES - POSITIVE
TRANSITION

I

5.0mV"VERDRIVE

6.0

1.0

10 - OUTPUT SINK CURRENT (rnA)

(Voel

RESPONSE TIME FOR
VARIOUS INPUT
OVERDRIVES - NEGATIVE
TRANSITION
w

0.1

40

30

SUPPLY VOLTAGE

I

I

2.0

TIME

8-26

I
1.0

0••

TIME (LIsee)

Vour

0:-

1.•

2.0

(~secl

1/86, Rev, A

-------------l~ CMP-04 QUAD LOW-POWER PRECISION COMPARATOR

TYPICAL APPLICATIONS
OUTPUT STROBING

INVERTING COMPARATOR WITH HYSTERESIS
V+
V+

+VIN

0---------1

V+

0----./11'01'--.---1

3kn

1Mn

Va

1Mn
*OR LOGIC WITHOUT
PULLUP RESISTOR

LIMIT COMPARATOR

SQUAREWAVE OSCILLATOR

V+

LED

+VREF HI

o--.NV'---i
f == 186kHz

Va

~

o

~

lOOk.l1
+VREF LOW

•
Po<

O--.NV'---i

~

U

~

o
COMPARING INPUT VOLTAGES OF OPPOSITE POLARITY

NON INVERTING COMPARATOR WITH HYSTERESIS

V+

V+

100kn
+VREF

0--------1

VINl

3kn

o--Mf----,

51k.l1

lOOkn

10kn

Va

8-27

1/86, Rev. A

~

~

~

-------------I~ CMP-G4 QUAD LOW-POWER PRECISION COMPARATOR

ONE-SHOT MULTIVIBRATOR

PULSE GENERATOR
V+

V+l
Rl

0-

tM!)

to

,M!)

'Ok!)

100pF

+VIN

15M2

0'
lN914

_ _ r:1-:'-- v+

o--j 1--....- - . -......--1

lm:=JPW~o

to

t1

Vo

to
lN914

t,

t2

lM.Q

,Mn
FOR LARGE RATIOS OF Rl/R2
D1 CAN BE OMITTED.

AND GATE

OR GATE
V+

39kn

200kn

3kn

3kn

10Ok.l1

A o------",'VV---,
+Q,Q75V

+0 375V

100kn

"0"

"1"

100kn
"0"

l00kn

lkn

"1"

100k.l1

lkn

lkn

1M2
l=A+B+C

ONE-SHOT MULTIVIBRATOR WITH INPUT LOCK OUT
V+

rE

~

+4V

,.

lMil

10Mn

lMn

15kn

560M2

100kn

+VIN

o---.vW'--.---......----+--I

o
Vo
10MO

to

t,

T = 03mSEC

100ld1

240kn

62k!l

8-28

1/86, Rev. A

------------i1fMD

CMP-04 QUAD LOW-POWER PRECISION COMPARATOR

TYPICAL APPLICATIONS
TIME DELAY GENERATOR
V+

,Okll

200k1l

'5k1l

3.0kIl
V+

10k.{l

o
V03

IO

r-

13

V3

6lkS)

3.OkIl
10MO

'0""

a1kO

-.-

10

VO.

V.

Ve,
I

o=r-••
V+

3.Okn

'0""

v+

.=r.,
10

VOl
V,

V;=r-L 10kn
'0

..

INPUT GATING SIGNAL

J

-=

e,

51kn

O.OO'_F

-::-

+VIN

BURN-IN CIRCUIT
GND
(OV)

GND
tOY)

~

,.

3.Ok

3.6k
'3

,.

~
....3 .......

1

+39V

3.Ok 30k

,"::~ZENER

~6.8VT06'V

ONE EACH

PER BOARD

1 WATT

6SOCKETS
~b~J~NT

S

'0

8

, .AZ-• ,;ot±--

, ~
•
•
3

'OOk

"

11
6

•

7

+36V
470k

MIL-STo 883, METHOD 1016, CONDITION B

8-29

----

-- ---------------

1/86, Rev. A

CMP-05
HIGH -SPEED PRECISION COMPARATOR
(WITH LATCH CIRCUIT)
Precision Monolithics Inc.

FEATURES
•

•
•
•
•
•

accuracy along with high speed. An exceptionally fast
response time of 60nsec is possible with only 1/2 LSB
overdrive (12-bit, 10-volt system).

Precision Input Stage
Input Offset Voltage. . . . . . . . . . . . . . . . . . . . . . . .. 150!'V
Input Offset Current ......................... 15nA
Fast Response Time (5mV Overdrive) ............ 38ns
High Voltage Gain ......................... 16,OOOV/V
Latch Function with TTL Compatible Input
TTL Compatible Output
Available In Hermetic Mini-DIP Package

The CMP-05 design makes itthe ideal component in systems
requiring high speed with excellent lOW-level analog signal
resolution. High-speed 12-bit successive approximation AID
converters, zero crossing detectors and logic threshold
detectors are typical system applications.

PIN CONNECTIONS

ORDERING INFORMATIONt

Vs+

•

PACKAGE
25°C

vos
(~v)

600
600
1000
1000
1000

HERMETIC

PLASTIC

TO-99
8-PIN

DIP
8-PIN

CMP05BJ'
CMP05FJ
CMP05CJ'
CMP05GJ

CMP05BZ'
CMP05FZ
CMP05CZ'
CMP05GZ

DIP
8-PIN

OPERATING
TEMPERATURE
RANGE

CMP05GP

MIL
INO
MIL
INO
COM

DIGITAL@7DUT

GROUND

+

+IN 2

-

6

-IN 3

8-PIN HERMETIC MINI-DIP
(Z-Suffix)

Vs- (CASE)

EPOXY MINI-DIP
(P-Suffix)

TO-99
(J-Suffix)

"For devices processed In total compliance to M IL-STO-883, add /883 after
part number Consult factory for 883 data sheet.
tAil commercial and Industrial temperature range parts are available with
burn-In. For ordenng information see 1986 Data Book, Section 2

~~li~E

,5 N C.

4

LOGIC TABLE
LATCH ENABLE

o or NC

GENERAL DESCRIPTION

1

The CMP-05's very high speed and precision input specifications make it the ideal comparator in systems needing 12-bit

OUT
Comparing
Latched

SIMPLIFIED SCHEMATIC DIAGRAM

L-~~--~

______-+________

~

____

8-30

~~

______

~

__

~

____________

~vs_

1/86, Rev. A1

------------I~ CMP-05 HIGH-SPEEO PRECISION COMPARATOR

ABSOLUTE MAXIMUM RATINGS (Note 2)
Positive Supply Voltage ............................. +6V
Negative Supply Voltage ........................... -18V
Power Dissipation (Note 1) ...................... 500mW
Differential Input Voltage ........................... ±5V
Latch Enable Input Voltage ........... -0.5V to V+ Supply
Operating Temperature Range
CMP-05B/C (J or Z Package)
(Note 3) ............................ -55·C to +125·C
CMP-05F/G (J or Z Package) .......... -25·C to +85·C
CMP-05G (P Package) .................. O·C to +70·C
DICE Junction Temperature (Ti) ......... -65·C to +150·C
Storage Temperature Range ............ -65· C to + 150· C
P-Suffix ............................. -65·C to +125·C
Lead Temperature (Soldering. 60 sec) ............. 300·C

Output Short-Circuit Duration - to ground .... Indefinite
- to V+ = 5.0V ... 1 Minute
MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (JI

eo'c

7.1mW/'C

Epoxy Mini-DIP (PI

36'C

S.6mW/'C

Hermetic Mini-DIP IZI

7S'C

6.7mW/'C

PACKAGE TYPE

NOTES:
1. See table for maximum ambient temperature rating and derating factor.
2. Absolute maximum rallngsapply to both packaged parts and DICE, unless
otherwise noted
3. Latch IS functional for -55°C ~ TA ::::: +85 0 C

-------------1[fMD CMP-05 HIGH-SPEED PRECISION COMPARATOR
ELECTRICAL CHARACTERISTICS at Vs+ =5.0V, Vs-=-5.0V, and Latch Enable grounded. ForCMP-05B/C, -55°C :5TA:5
125°C. For CMP-05F/G, -25°C:5 TA:5 85°C (J, Z Packages) and O°C:5 TA:5 70°C (P Package), unless otherwise noted.
CMP-05B/F
PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs = 50n

Input Offset Voltage Drift

TCVos

MIN

CMP-05C/G

TYP

MAX

MIN

TYP

MAX

0.3

1.5

0.55

2.0

mV

1.5

7.5

2.5

15

I'VI'C

nA

UNITS

Input Offset Current

los

40

250

70

400

Input 81as Current

Ie

1.1

2.5

1.5

3.8

Voltage Gain

Avo

(Note 1)

6

11

10

Vim V

Input Voltage Range

CMVR

(Note 1)

±2.9

±3.2

±2.9

±3.2

V

Common-Mode Rejection
Ratio

CMRR

VCM = ±2.9V, (Note 1)

83

90

80

88

d8

Power Supply Rejection
Ratio

PSRR

±4.75V $ Vs $ ±5.25V
P Package

Output High Voltage

VOH

V,N " 10mV. 10 = OI'A
V,N " 10mV, 10 = 240l'A
V,N " 10mV, 10 = 160l'A

Saturation Voltage

VSAT

63

178

2.4
2.4

80
150

252
400

I'VIV

2.4

V
2.4
0.18
0.2

Y,N $ -10mV, ISINK = OmA
Y,N $ -10mV, ISINK = 9.6mA
Y,N $ -10mV, ISINK = 6.4mA

0.40
0.40

0.20

0.40

0.30

0.40

V

Positive Supply Current

Is+

Vo $ 0.4V

11

16

12

17

Negative Supply Current

1.--

Vo $ 0.4V

12

17

13

19

mA

Power DIssipation

Pd

Vo $ 0.4V

115

165

125

180

mW

Latch Input Current
Logic 1
Logic 0

'LH
'LL

VLH =3V, (Notes 1, 4)
VLL = 0.8V. (Notes 1. 4)

18
10

90

18
10

90

50

tpd +

Voo= 1.2mV, (Notes 1, 2)
Voo = 5.0mV. (Notes 2. 5)

125
92

125
92

ns

tpd-

Voo = 1.2mV, (Notes 1, 2)
Voo = 5.0mV, (Notes 2, 5)

115
88

115
88

ns

t LPO +

(Notes 2, 4, 5)

56
30

56
30

ns

Input to Output High
Response Time
Input to Output Low
Response Time
Latch Disable Time

tLPoNOTES:
1. Guaranteed by design.
2. Times are for 100mV step inputs. See switching time waveforms.
3. A high on the latch enable input will cause the latch to assume the state

Minimum Input Timing Requirements·
LATCH

3V

1.4V- -

Parameter
-

-

-

--- -

T- - - - -

VIN

Minimum
Limit

Units

_. - -

·'~t:Jr- ~:~
OIFF,~~~~TlAL vos~t. -~+4 ,...-------\1
-1--------_\J __
-J.-'\.
VOLTAGE

50

of the comparator and not follow subsequent inputs.
4. Latch is functional for -55'C $ TA $ +85'C.
5. Sample tested.

SWITCHING TIME WAVEFORMS
E~ATBCL~

mA

ts
th
tw

Setup Time
Hold Time
Latch Pulse Width

35
10
25

ns

'ts, th' tw are tested with Y,N = 100mV and Voo= 5mV.

- - - - --

00

tLPO_-j

t pd

VOH

OUTPUT t.4V- -

-

-

-

VSAT _ _ _ _ _

8-32

1/86, Rev. A1

------------t/fMD CMP-Cl5 HIGH·SPEED PRECISION COMPARATOR
DICE CHARACTERISTICS
1.
2.
3.
4.
6.

DIGITAL GROUND
NON INVERTING INPUT
INVERTING INPUT
NEGATIVE SUPPLY (SUBSTRATE)
LATCH ENABLE
7. OUTPUT
6. POSITIVE SUPPLY
For addilional DICE Informallon rafar 10
1984 Dala Book, Sacllon 2.

DIE SIZE 0.051 X 0.045 Inch, 2295 sq. mila
(1.295 X 1.143mm, 1.481 sq. mm)

WAFER TEST LIMITS at Vs = ±5V, TA = 25° C, unless otherwise noted.

CMP·05G
LIMIT

UNITS

tooo

IlVMAX

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vos

Rs= son

Input Offset Current

los

150

nAMAX

Input Bias Current

Ie

1.8

IlAMAX

Voltage Gain

Avo

(Note 1)

Input Voltage Range

CMVR

(Note 1)

Common·Mode Rejection
Ratio

CMRR

VCM = ±2.9V
(Note 1)

Power Supply Rejection Ratio

PSRR

Positive Output Voltage
Saturation Voltage
Positive Supply Current

VlmV MIN
±3.0

VMIN

80

dBMIN

±4.75 :s Va:S ±S.25
Vs+ = SV, Vs- = -SV to -15V

178
63

IlV/V MAX

VOH

V,N ~ 10mV, 10 = OIlA

2.4

VMIN

VSAT

V,N:S 10mV, 10 = OIlA

0.4

V MAX

1+

Vo :SO.4V

16

mAMAX

Negative Supply Current

1-

Vo :SO.4V

18

mAMAX

Negative Supply Current

I-

V- = -ISV, Vo:S 0.4V

20

mAMAX

Latch Input Voltage
Logic 1
Logic 0

VlH
Vll

Latch Enabled
Latch Disabled

2.0
0.8

VMIN
V MAX

Latch Input Current
Logic 1
Logic 0

IlH
III

VlH = 3.0V, (Notes 1,4)
Vll = 0.8V, (Notes 1, 4)

45
2S

IlAMAX

Input to Output High
Response Time

tpd+

Voo = S.OmV, (Notes 1,2)

60

nsMAX

Input to Output Low
Response Ti me

tpd-

Voo = S.OmV, (Notes 1, 2)

60

nsMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to vanations in assembly methods and normal yield loss, Yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ±5V, TA = 25°C, unless otherwise noted.

CMp·05G
PARAMETER

SYMBOL

CONDITIONS

Input to Output High
Response Ti me

TYPICAL

UNITS

tpd+

Voo = 1.2mV, (Note 2)

41

ns

Input to Output Low
Response Time

tpd-

Voo = 1.2mV, (Note 2)

37

ns

Latch Disable Time

tlPO

(Notes 3,4)

so

ns

NOTES:
1. Guaranteed by design.
2. Times are for 100mV step inputs.

3. See switching time waveforms.
4. Latch IS functional for -5SoC:S TA S 8SoC.

8·33

1/86, Rev. A1

~

~

~
~

U

~

~
tj
~

------------l~ CMP-05 HIGH-SPEED PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS
INPUT BIAS CURRENT
vs DIFFERENTIAL
INPUT VOLTAGE

VOLTAGE GAIN
VB FREQUENCY
100

I I II

90

VS" ±5V

1200

VSAT vs LOAD CURRENT

,...._~-,..---,---r--:;o_.....,

0.600

./

0.450

TA "" 2SOC

TA = +2SOC ........

1000 ( - - + - - - J . - - + - P - - - - I - - _ _ _ j
0.400

.....

800 (--t---r----\-+f--r---r-___j

ii)

:;

~

600

TA = +125°C

0350

I~

0300

~V

~

g

0250

....... ~

400 f - - + - - + - - + + - \ - - + - _ f - _ _ _ j
0200

40

200f--+--~-+--~--_f--___j

0'00

30
0.1

10

100

FREQUENCY (MHz)

-150

1200

600

60

40

-

'\ "\ t'<;j'B

1"-

-75 -50

0

25

10

50

Llw
,.:.~~

/
/

75

100

0
125

/

70

40

t-- 1 2mV OVERDRIVE

~

t--

V

+

ILOAD

OUT

8

2

10

12

14

....-

L

V

16

VOL

18

20

/ /

Vs" ±5V

I-- lOOmV INPUT STEP

50

. . , TA' _55°C

200

90

60
20

N

-25

o

..,

V

RESPONSE TIME vs BALANCED
SOURCE RESISTANCE

I

100

30

o

v~. '5V'

80
30

"- b--- 'OS

200

120
110

50

1,\

400

V

~

~

ILOAD (rnA)

RESPONSE TIME vs
TEMPERATURE

Vs = ±5V

800

150

./.

DIFFERENTIAL INPUT VOLTAGE (mY)

INPUT CURRENTS vs
TEMPERATURE

toaD

0.150

i"'"

~

~ /'

/
E
w
~

;::

"""""SmV OVERDRIVE
,...... ........-

125

w

~

~

V

- ,oor

TA '" 25°C

_ 150

/

/

175

-

/ /
Lv OLDRIVV~ / /
IN'UIT STEP

100

CC 75
50
25

./" ~

/ Vsmv

OVERDRIVE

,/

::..-o

BALANCED SOURCE RESISTANCE {knj

TEMPERATURE (OC)

8-34

1/86, Rev. A1

------------I~ CMP-05 HIGH-SPEED PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS
RESPONSE TIME
TO SmV AND 1.2mV
(= 1/2 LSB) OVERDRIVES

RESPONSE TO 10MHz
SINE WAVE
+10mV

INPUT

14

INPUT

-

,

I'

I

,

+2V
OUTPUT

+1V

I

OV

I

r

r... r':.'=. ::1

r

i
+3V

,

OUTPUT

1

Iii
.!!!!!!!
10ns/DIV

HO~mV
J,
L 1.c:50mv;l;
-VOD

10mV

Ir 2~mV

r- ~oD'25mv",,-

'0 ~ V /
'51 ff fd- 1";:- ~ --- -J"

rlsmv

'\ ~"I 1\ IX /

l00m~omv

I I

:t-

Ipd_RESPONSE TIME

Ipd+ RESPONSE TIME

---

12mV
OVERDRIVE

--1- llii

20ns/DIV

14

5mV
OVERDRIVE

i
,

~

+10mV

-1. . .·1~

14

-- --- -\-

\ \

1'-5mV-'·fmv

,12mV

+-h ~f v- --- --\

\ \

' - 10my
+100

+100

t
t

Vos
-100

10

t

INPUT OVERDRIVE (VOO)

VOS

t

-100

20

30

40

50

60

70

INPUT OVERDRIVE (VOO)

10

20

30

40

50

60

70

RESPONSE TIME (ns)

RESPONSE TIME (ns)

CMP-OS NEGATIVE COMMONMODE INPUT RANGE vs
NEGATIVE SUPPLY

RESPONSE TIME vs
OVERDRIVE VOLTAGE

-17
CMP-05

-15

VIN = 100mV + VOO
TA = 2SoC

V+=+50V
TA = 25°C

100

-13

/V

-11

80

-9

60

-7

V

/
40

::::::: :::::.....

-5

20

tpr

-3

-2

-4

-6

-8

-10

-12

-14

7

-16

tpd+

10

30

70100

V oo (mV)

NEGATIVE SUPPLY VOL rAGE (VOLTS)

8-35

1/86, Rev. A1

•

-------------l~ CMP-05 HIGH-SPEED PRECISION COMPARATOR

APPLICATION INFORMATION

Fortunately, in high-speed circuitry the comparator inputs
will be driven at a fast rate, in which case no transition region
oscillations will occur. As the minimum slew rate versus
source resistance curve indicates, if the input is driven at a
rate exceeding 6mVll'sec, no oscillations will occur with
source resistors of less than 1kO. Examples of "clean" transitions can be observed in the photographs of the response
time with SmV and 1.2mVoverdrives, and the response to the
10 and 2SMHz input signals.

The CMP-OS is a very accurate device providing fast response
time even with small-Microvolt level-overdrives. To achieve
this performance requires high gain at high frequencies. As
shown in the voltage gain versus frequency curve, the gainbandwidth product of the CMP-OS is 1.S X 1011 Hz. It maintains its full gain to approximately8MHz and rolls off at a very
fast rate beyond that frequency due to the fact that five poles
occur in the 30 to 60MHz range. At 30MHz the gain of the
comparator is still 2000. Therefore, in the transition region
small values of source lead inductance and stray feedback
capacitance can cause an oscillatory condition.
Forexample (in the figure below) with L=0.1I'H, Cs=0.1SpF,
the closed-loop gain of the circuit at 30MHz is:

In order to not degrade its speed the CMP-OS's inputs are not
internally clamped. If large differential voltages are present it
is recommended that the inputs be clamped with high speed,
low capacitance diodes such as the H.P. S082-283S, which is
a Schottky Diode.

A - _1_ v- LCsw2 -10

As in all high-speed devices, it is to the user's advantage to
keep the source impedances low and matched.

7

1
X 0.1S X 10-12 X (271' X 30 X 10 6 )2

880
1

LATCH
The CMP-05 has a latch feature which functions over -55° C
to +85° C. When the latch is enabled, the output stays in its
existing logic state regardless of the input signal. The input
timing requirements of the latch are presented in the Switching Time Waveforms. The latch opens up a broader applications area at no sacrifice in total system speed. Effectively,
the latch allows high speed sampling of comparison decisions. This is important in automatic test equipment limit
comparators, in measuring pods used in logic analyzers and
other similar synchronous measurement circuitry needing
fast clocking frequencies. The latch pulse width tw allows
sampling of input signals to take place in 2Snsec.

POTENTIAL FEEDBACK SOURCES
Cs

VOUT

With the open-loop gain at 2000 oscillation will occur since
the phase shift exceeds 180°C. To minimize these problems
powersupplies should be decoupled, lead lengths should be
kept as short as possible, and a ground plane should be used
to reduce the stray feedback capacitance. In addition, a
ground plane substantially diminishes the possibility of the
output current spike coupling back to the inputs through the
ground lead. Keeping a separate digital ground (pin 1) and
analog ground (to which the inputs are referenced) also
reduces the magnitude of the problem.

The latch prevents self oscillation (due to positive feedback)
from taking place when slowly-moving high-sourceimpedance signals pass thru the linear amplification region
of the comparator. This is successfully accomplished by
rapidly strobing the comparator near its minimum tw time
which prevents self oscillation from making a complete cycle
since tw is shorter than the total response time tpd through
the comparator.

12-BIT FAST AID CONVERTER
SERIAL
DATA OUT

CONVERSION TIME vs ACCURACY

1.25

\

\
100

CLOCK

(WOI\ST CASE)\

+15V

DAC-312

LS8

CMP..()5

075

050

r--

(TVr

REF-Ol

025

000
100

,\

DAC-3~

CMP..()5

200

300

\

'"

400

500

600

700

BOO

CONVERSION TIME PEA TRIAL (n5)

CONVERSION
TIME (nl)
SAR
CMP-oS
NOTE

TYP
33
92

WORST
CASE
55

TOTAL

375ns

125
680n8

x 13

49J.1s

88,us

DEVICE(S) CONNECTED TO ANALOG INPUT MUST BE CAPABLE OF SOURCING 4.0mA
A BUFFER (eg aUF·OJ) MAY BE REQUIRED

8-36

1/86, Rev. A1

CMP-07
VERY HIGH -SPEED COMPARATOR
(WITH LATCH CIRCUIT)
Precision MOll0lithics Inc.

PRELIMINARY
FEATURES
•
•
•
•
•
•
•

latch timing is also faster than the Am685, resulting in less uncertainty when used as a sampling comparator.

Input-to-Output Propagation
Delay at 5mV Overdrive ..............• 6.5ns Max
latch-to-Output Propagation
Delay at 5mV Overdrive ............... 5.5ns Max
latch Pulse Width •....•.•................ 2ns Min
low Power Dissipation ................. 200mW Max
ECl Compatible Complementary Outputs and latch Input
50 Ohm Line Driving Capability
Pin Compatible with Am685

The CMP-07 is specified for operation with a standard -5.2V ECl
power supply and a +6V supply. The circuit operates with a +5V
supply with some reduction in input voltage range. The output
stage directly drives 50n transmission lines.
The CMP-07 electrical characteristics are ideally suited for use in
data conversion systems, logic analyzer probes, automatic test
equipment, pulse detectors, and simple flash converters.

PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE
TO-100

DIP

OPERATING
TEMPERATURE
RANGE

CMP07BK'
CMP07FK

CMP07BO'
CMP07FO

MIL
IND

16-PIN CERDIP
(Q-Suffix)

TO-100
(K-Suffix)
GND1

v@."o
9 GN02

• For devices processed In total ocmpliance to MIL-STD-883. add 1883 after part
number. Consult factory for 883 data sheet.
tAil ocmmerClal and industrial temperature range parts are available with burn-In
For ordering information see 1986 Data Book, Section 2.

+IN2

+

-IN 3

-

LE 4

saOUT

7QOUT

,

6 N.C.

v-

NOTEPin 5 IS connected to package

GENERAL DESCRIPTION

LOGIC TABLE

The CMP-07 is a very high-speed comparator with complementary EeL output logic compatibility. It offers the same propagation
delay specifications as the Am685, at 66% of the Am685's power
consumption, reducing heat generation and improving reliability.

LATCH ENABLE

OUT

LOW

LATCHED

HIGH

COMPARING

BLOCK DIAGRAM

INPUT

+IN 0 - I - -

-IN10-

r>....

1--1;:/

LEVELSHIFTI

2NDSTAGEI
LATCH

STAGE

~

Q

0

>

OUTPUT STAGE

ii

Q

LE

V

LE

>

Q

RL

RL

LATCH ENABLE

BUFFER
LE

~

VT

>
IV

--O OUTPUT

50n
-VREF

4.7k

0-.,....--01
50H

S011

-5.2V
-2.0V

(PINS SHOWN FOR DUAL-IN-LINE PACKAGE)
OUTPUT PULSEWIDTH "" 0.75 C (Rl

+ R2)

8-39

1/86, Rev_ A

•
:

CMP-404

®IPMI)

QUAD LOW-POWER
PRECISION COMPARATOR
Precision Monolithjcs Inc.

FEATURES
•
•
•
•
•
•
•

The low input-offset-voltage makes the CMP-404 an ideal
companion to CMOS logic when the stability and accuracy of
a bipolar technology is needed along with low power consumption. The open-collector outputs with pull-up resistors
provide CMOS interface with excellent noise immunity.
Improved isolation between comparators was achieved by
use of an independent bias circuit for each comparator. This
is especially important when one comparator is detecting
low-level signals while an adjacent comparator is being
driven by a high-level signal. In single-supply operation, the
inputs can operate at ground. The CMP-404 can operate from
5 to 30 volts single supply or ±2.5 to ±15 volts dual supply.

Very Low Power Consumption .............. 1.5mW Max
Low Input Offset Voltage .................. , .. 1mV Max
Very Low Drift ............................ 3/lV/o C Typ
High Output-Drive Current .................. 25mA Typ
Single or Dual Supply Operation
Ideal for CMOS Logic Interface
LM139 Pinout

ORDERING INFORMATIONt
25°C
Vos(mV)

HERMETIC
DIP PACKAGE

1
2
1

CMP404AY'
CMP404BY'
CMP404EY
CMP404FY

2

OPERATING
TEMPERATURE
RANGE

Window comparators, limit comparators, multivibrators, one
shots, voltage-controlled oscillators, and set-point detectors
are common applications.

MIL
MIL
INO
INO

PIN CONNECTIONS

• For devices processed in total compliance to MIL-STD-883. add 1883 after
part number. Consult factory for 883 data sheet.

t All commercial and industrial temperature range parts are available with
burn-in. For ordering information see 1986 Data Book, Section 2.

GENERAL DESCRIPTION
Four precision-input comparators provide excellent speed
with low power consumption through use of a novel Schottkyclamped design. These open-collector output comparators
only consume 365 microwatts each, yet they make accurate
5mV decisions in only four microseconds. In addition, they
can drive load currents of 25mA. This output stage is ideal for
driving relays, lamps, and LEDs.

14-PIN HERMETIC DIP
(V-Suffix)

SIMPLIFIED SCHEMATIC (1/4 OF CMP-404)

V+

......-o OUT

+--t---r>l--I~>I-+INPUT

-'NPUT

o-_--+----jf-+-----+----I

8-40

1/86, Rev. A

-----------I~ CMP-404 QUAD LOW-POWER PRECISION COMPARATOR

ABSOLUTE MAXIMUM RATINGS (Note 1)
Supply Voltage ............................ 36V or

± 18V

Storage Temperature Range ............. -6S0Cto + 1S0°C

Input Voltage ............................... -0.3V to V+

Input Current (Note 2) ............................. 20mA

Output Voltage •.•••••••••••••••.•.•••.•••• -0.3V to 36V

Output Short-Circuit to V+ (Note 3) ••.... , •..•.•..•. SOmA

Power Dissipation

•••.••...••••....••••••.•.•.•• SOOmW

Lead Temperature (Soldering, 60 sec) ••............ 300° C

10mW/oC
...•.......•........... 100°C/W

NOTE:
1. Absolute maximum ratings apply to both packaged parts and DICE, unless
otherwise noted.
2. Limit for input current that flows when input voltage signals exceed V+ or
GND forward biasing internal junctions.
3. Short circuits to V+ can cause excessive heating and eventual destruction.
The maximum output current is SOmA.

Derate Above 100°C by .......................
Thermal Resistance (I1JA)

Operating Temperature Range
CMP-404EY/FY ....................... -2SoC to + 8SoC
CMP-404AY/BY ...................... -SsoC to + 12SoC
DICE Junction Temperature (Tj) .......•. -6So C to+ 1S0° C

ELECTRICAL CHARACTERISTICS at V+= SV, RL =

S.1 kOand-SSo C$TA$12S0C forCMP-404AY/BY; -2SoC $TA $8S0C

for CMP-404 EY/FY, unless otherwise noted.

PARAMETER
Input Offset Voltage

SYMBOL
Vas

Average Input Offset
Voltage Drift

TCVos

Input Offset Current

los

Input Bias Current

I.

CMP-404A/E
MIN
TYP
MAX

CONDITIONS

mV

Rs = 5011, Full Temp

mV

IINI+I-IINI-I' TA=25'C

10

25

nA

IINI+I-IINI-I' Full Temp

50

100

nA

10

liN 1+1 or liN I-I' TA = 25'C

Av

RL = 15kll

tr

Voo=5mV, V STEP =100mV
RL =5.1kll, TA=25'C(Note4)

Large-Signal
Response Time

tr

VIN = TTL Logic Swing
V RE ,= 1.4V, RL =5.1kll

Input Voltage Range

CMVR

Saturation Voltage

VOL

50

Output Leakage Current

ISINK

10

3.5

3.5

~s

0.8

0.8

~s

0
75

TA = 25'C, (Note 2)

50

V+ -1.5

V+ -15

V

V+ -2

V+ -2

V

85
0.32

75
0.4

85
0.32

0.5

VIN (_I = W
VIN (+I=OV, Vo=2V, (Note 5)

10

25
001

TA = 25'C, (Note 3)

nA
nA

Vim V

TA = 25'C

RL = 15kll, (Note 6)

100
200

400

400

Full Temp, (Note 2)
Output Sink
Current

50
100

TA = Full Temp
CMRR

~V/'C

3

Rs = 5011

Small-Signal
Response Time

Common-Mode
Rejection Ratio

UNITS

Rs = 5011, TA = 25'C

liN 1+1 or liN I_I' Full Temp
Voltage Gain

CMP-404B/F
MIN
TYP
MAX

10
0.1

dB
0.4

V

0.5

V

25
001

mA
0.1

~A

0.4

~A

ILEAK

Full Temp, (Note 3)
Power Supply
Rejection Ratio

PSRR

V+ = 5V to 30V, RL = 15kll

Supply Current

1+

RL =co

04
75

100
220

65
300

100
220

dB
350

~A

NOTES:
1. TYPical values are reported for TA = 25'C.
2. ISINK = 1mA, VIN H = W, VIN (+1 = OV
3. VIN (_I = OV, VIN 1+1 = W, Va = 30V
4. Guaranteed by design. See response-time test circuIt.
5. Output Sink Current should be limited to SOmA by external resistance
6. Applies over the CMVR range

8-41

1/86, Rev. A

~

0

~

p.j

:E

0

U
~

~
b
~

_ _ _ _ _ _ _ _ _ _~~ CMP-404 QUAD LOW-POWER PRECISION COMPARATOR

DICE CHARACTERISTICS

1.
2.
3.
4.
5.
6.
7.

OUTPUT (2)
OUTPUT (1)
POSITIVE SUPPLY
INVERTING INPUT (1)
NONINVERTING INPUT (1)
INVERTING INPUT (2)
NONINVERTING INPUT (2)

8.
9.
10.
11.
12.
13.
14.

INVERTING INPUT (3)
NON INVERTING INPUT (3)
INVERTING INPUT (4)
NON INVERTING INPUT (4)
GROUND
OUTPUT (4)
OUTPUT (3)

For additional DICE information refer to
1986 Data Book, Section 2.
DIE SIZE 0.068 X 0.076 inch, 5168 sq. mils
(1.727 X 1.93 mm, 3.33 sq. mm)

WAFER TEST LIMITS at V+

=

5V, RL = 5.1 kfl., TA = 25°C, unless otherwise noted.
CMP-404G

PARAMETER

SYMBOL

CONDITIONS

Input Offset Voltage

Vas

Rs ~ 50ll

Input Offset Current

los

IIN(+) -liN (-)

Input Bias Current

18

IIN(+)

Voltage Gain

Av

RL

~

15kll

LIMIT

25

or IIN(-l

Input Voltage Range

CMVR

Common-Mode Rejection RatiO

CMRR

RL

~

15kll

Power Supply Rejection RatiO

PSRR

V+

~

5V to 30V. RL

Saturation Voltage

VOL

'SINK = 1mA

Output Sink Current

'SINK

V,N I_I ~ 1V
V,N 1+1 ~ OV. Va ~ 2V

Output Leakage Current

'LEAK

V,N I_I ~ OV
V,N 1+1 ~ 1V, Va

Supply Current

1+

RL

=

UNITS
mVMAX

~

~

15kll

nAMAX

50

V/mV MIN

V+ -15

VMAX

75

dBMIN

65

dBMIN

0.4

VMAX

10

mAMIN

0.1

30V

300

!Xl

nAMAX

100

~A

MAX

~AMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to vanatlons in assembly methods and normal Yield loss, Yield after packaging is not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

TYPICAL ELECTRICAL CHARACTERISTICS at V+

=

5V, unless otherwise noted.
CMP-404G

PARAMETER

SYMBOL

Large-Signal
Response Time

tr

Smail-Signal
Response Time

tr

CONDITIONS
V,N ~ TTL LogiC SWing
VREF~
Voa~

1 4V, RL ~ 51kll
5mV. VSTEP ~ 100mV

RL~51kll

8-42

TYPICAL

UNITS

0.8

~s

35

~s

1/86, Rev. A

-----------t1fMD

CMP·404 QUAD LOW·POWER PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS
SUPPLY CURRENT
VI TEMPERATURE
50

VeIM'
400

~
~
a:

300

a:

a
~

200

~

~~
~3V

~

-~

-V

100

"

o
~

~

.........

30

20
/25, 12
10

_

0

~

~

~

o

=

~

o

o
16

20

10
VIN=lV

01

06

........ ........

-03

-50

-25

........ ........

32

-75

36

75

125

800
700

~0

'1'1..\

500
400

?
0
>

300

,

200

-75

:>
-so

-25

25

50

TEMPERATURE (OC)

75

100

125

100

125

OUTPUT CURRENT
OUTPUT VOLTAGE

II

_55°C

III

r1

III

15r---+-~fL~~~-+---+---1

I[j

'/

10~--+-~~~~---+---+--~

I

I/, rj III

& e:::;;. ;....-V
o

01

02

03

04

05

OUTPUT VOLTAGE (VOL TS)

OUTPUT VOL TAGE (VOL TS)

NEGATIVE RESPONSE TIME
vs OVERDRIVE

POSITIVE RESPONSE TIME
vs OVERDRIVE

60

60

50

50

"

40
V OD = 50mV

20mV

30

lOmV

~
0

5mV

?
0
>

20

V+= 5V

TA I= 250 C (

l

(

40 rVOD

10mV

=50mV

5mV

30
20mV

20
10

'>

'>

2

o

75

I\,
E

100

50

VB

10

1\

25

TEMPERATURE (OC)

AL'" lSk.l1

.,.,.

-25

I

J

........
100

OPEN-LOOP GAIN
VI TEMPERATURE

600

-50

~

II

_25°e,-..:

02

o
50

28

~~:~-..:

---- -

04

25

24

J II

JJ

08 - - - I T A ' 125"e

02

-04
-75

l

12

V+ =5V

r-

-55 o e

OUTPUT CURRENT
OUTPUT VOLTAGE

V+ =5V
V CM = OV

~

C

VB

04

-02

r

SUPPL Y VOLT AGE (VOLTS)

OFFSET VOLTAGE
VI TEMPERATURE

-01

........

r- r-

~ ~e

TEMPERATURE (OC)

03

V+ = 5V

40

.-- ""'" .....

i"'"

ol

I-

"..,

v+=;"- ~

INPUT OFFSET CURRENT
VI TEMPERATURE

INPUT BIAS CURRENT
VB SUPPLY VOLTAGE

E

-50

2

V+=5V

-100

TA = 25°C
10

20
TIME (us)

8-43

30

40

100
50

:>
10

2,0

30

4.0

TIME (/.Is)

1/86, Rev. A

-----------I~ CMP·404 QUAD LOW·POWER PRECISION COMPARATOR

TYPICAL PERFORMANCE CHARACTERISTICS
POSITIVE RESPONSE TIME
VB TEMPERATURE

NEGATIVE RESPONSE TIME
VI TEMPERATURE

RESPONSE TIME
VI OVERDRIVE
40

sv

V+ =

V+! 5V

I-- I

r-...

54

OVERDRIVE

--

~~

Lmv
I'-.

w

"
~

iji 3

z

~

0:

2

+2OmV

+rmv

...

...........

--

-50

-25

25

50

75

100

125

-75

TA = 25°C

\

30

i"""--

g
w

i'-

"

25

1\

20

\

~

r-

~

l!1

~

15

10
50mV

-25

0

25

50

75

100

TEMPERATURE (OC)

TEMPERATURE lOCI

RESPONSE TIME

OUTPUT LEAKAGE
VI TEMPERATURE

VI COMMON-MODE VOLTAGE

r.... r--.

~

0.'

o
-50

\'NEGATIVE OVERDRIVE

\~

r- POSIT~
OViRDRI~VE

0:

-20mV

o
-75

V+ '" 5V

3.

oVELRJ
t----;;;;;;;:; ~V

125

o

5

10

15

20

25

30

35

40

45

50

OVERDRIVE (mY)

INPUT-OUTPUT
TRANSFER CHARACTERISTIC

120
V+

=5V

v+ = 30V

TA '" 2S Q C

100

]
w
"~ 4
w

~

Z

~

-5mV OVERDRIVE

.0

+5m) OVERbRIVE

60

3

I
II
I

40

20

o
05

10

15

20

25

30

35

-75

- ""

/

~

-50

-25

25

50

75

100

125

-4

TEMPERATURE (OC)

COMMON-MODE VOL T AGE IVOL TS)

BURN-IN CIRCUIT

-3

-2

-1

0

INPUT VOLTAGE (mV)

RESPONSE-TIME TEST CIRCUIT

~~~o-~------------~--,-----~--------t--o~~~

•• V

+36V

o-.,+--+--.....~~......~---II------+---+----VH OR VS- 10kn

-

-.-.--.-

------------I~ CMP-404 QUAD LOW-POWER PRECISION COMPARATOR

TTL-COMPATIBLE ANALOG SWITCH

2N4391

TTL
CONTROL
INPUT
LeVEL SHIFTER
SWITCH DRIVER

-15V

LOW-FREQUENCY OPERATIONAL AMPLIFIER

OUTPUT STROBING

AV = 1000V/V
+5V

+5V

15kn

>-.......--

2.0

~
g

i

~

z

~ 0.2

-

o

-25

0

25

50

75

100

05

-55

w

"

-20,0

>-

-17.5

(,)

.
".

'":!

S

0:
0:

""

-15.0

'\

-12.5
25

50

75

100

125

-55

o

-25

25

50

75

100

COMMON-MODE LIMITS
vs TEMPERATURE
y+

VCM= OV
Vs=±15V

1
>ffi

.......

"

(,)

:--.....

~
o

...... r--....

>-

>-

~

10

12

-8

INPUT

-4

+4

COMMON~MODE

+8

+12

+16

-03

./

-04
-05
-55

f-Of--

4

_

~

~

~

~

~

m

TEMPERATURE (OC)

SATURATION VOLTAGE
vs SINK CURRENT
14r----;-----;-----r----~--_,

V

........

1.21----t---+--+--+---1

",..

~ 10~-~--_+--~--+_-~

~

a 08
.

1---+~~~--4---~~-1

~
~~

0.6

1---+-""-"+-""'~"""'-~""''''''''..-1

0.4

1---+....c--+:,.".'5'''I'''=----+---1

>

OUTPUT
RL == 600n

'"

0.2

0.21--""=---+--4---+---1

I
-05

_

125

Vs=±15V
ISINK = SOmA

0.8

1\
I \

-10

100

- --~

o. 2

F~~L6W~R 1\

0

0.4

y-

....... I-'"

0

::E

"8

0

VS= 30V

OUTPUT
Ri"10kfl

-1.5

SATURATION VOLTAGE
vs TEMPERATURE

T~ ~ 2~oC

_ COL~~~~~~----/I
0 - _I

8

;

TEMPERATURE (0C)

TRANSFER FUNCTION
-

~ -1.0

"
:;

J

0255075

-25

VOLTAGE (V)

50

/

r--

VOLTAGE
REFERRED TO SUPPLY VOLTAGES

E

/

~ -0.2

bOM~ON-MODE

125

-0.5

/
-0.1

!§

......... r.....

f--

",..

0:

20

o
-1.

"- ~

INPUT OFFSET CURRENT
vs TEMPERATURE

.....

>-

~

~

1

"15 30

"""

INPUT BIAS CURRENT

40

0

-22.5

Z

0:
0:

VCM == OY
Vs=±15V

TEMPERATURE (QC)

TA == 25°C

~

1>-

I'....

TEMPERATURE (OC)

50

.."

o

-25

-25.0

TEMPERATURE (OC)

vs COMMON-MODE VOLTAGE

'"~

/

---- -

o

125

VV

1.5

... 1.0

~

V

.§.

a

-55

V

VCM == OV

,. 08

INPUT BIAS CURRENT
vs TEMPERATURE
-275

ov

Vs == ±15V

"~

INPUT OFFSET VOLTAGE
vs TEMPERATURE

05

DIFFERENTIAL INPUT VOLTAGE (mV)

10

0
-55

-25

0255075
TEMPERATURE (
>E

~

\

\~

\

"

100

TA = 25°C

50

Vs= ±lSV

t--

~
i!:

1
o

240

05

10

15

20

25

30

3.5

TIME (n8)

TIME (/-t8)

COMMON-MODE REJECTION
RATIO vs TEMPERATURE

vs TOTAL SUPPLY VOLTAGE

TIME (n8)

RESPONSE TIME FOR
VARIOUS INPUT OVERDRIVES

180

120

\1\

20m~~ r\
~~~-./ 1\

-15

Vs = ±15Y
60

-5

~-."

~ ,,\

10

-10

TA = 25°C

200

e"

~€
:>

rD :/

,-

15

w

t:>
0

TA == 25°C

150

100

N J. rrl

/. r-J
II 1/'1

, \1\

L\ 1\

RESPONSE TIME FOR
VARIOUS INPUT OVERDRIVES

RESPONSE TIME FOR
VARIOUS INPUT OVERDRIVES

40

TOTAL SUPPLY CURRENT

•

120 ...--;-..,....-r--,--,.---,--,
20mV""""\.

5

::~~ I

i1. J

1/

II V

~
"

L

-

-10
-15

100~~=t::t::::t=t=~~

1

1 "I

50~-+--4---~-+--+-~--~

'-

601---+--+--1---+---+--t----1

~

401---+--+---1---+---+--t----1

~

i!

g~

i!:

20~-+--4---~-+--+-~--~

-50
TA= 25°C
ys~ ±1'5V

~- -100

o

o~~-~-~-L-~~-~

0.5

1.0

1.5

2.0

2.5

3.0

35

~.

VS=±15V
4.0

-55

-25

25

TIME (J.ls)

50

75

100

125

1

o

TA,= 25°C

I I
10

YIN = 10mV

15

20

25

30

35

TOTAL SUPPLY VOLTAGE (V)

TEMPERATURE (Oel

SUPPLY CURRENT
vs TEMPERATURE
4.5
40
3.5
3.0

Ys '" ±15V

.........

..........

2.5

NO LOAD

"

~

'~

~

t"-.....
..........

2.0

~
..........

1.5
1.0
-55

""

........

["..

-25

0

25

50

75

100

125

TEMPERATURE (0C)

8-53

1/86, Rev. A

J»1Vl-119/J»1Vl-219/J»1Vl-319
DUAL
COMPARATORS
Precision MOllolithics Inc.

ADVANCEPRODUCTINFO~TION
FEATURES
•
•
•
•
•
•

allowing output swings of up to +35V. High output drive capability facilitates RTL, DTL, and TTL interfacing, as well as relay and
lamp driving at currents up to 25mA. Typical response time of
80ns with ±15V power supplies makes the PM-119 ideal for
application in fast AID converters, level shifters, oscillators,
and multivibrators.

Fast Response Time ..•••....••••..•.•••• 80ns Typ
High Output Drive Current •...•.••...•.•••••• 2SmA
Single or Dual Supply Operation •••••.•. + SV to ± 1SV
Open Collector Outputs ••••..••••..•.•• Up to + 3SV
Inputs and Outputs Can Be Isolated from System Ground
Two Independent Comparators

PIN CONNECTIONS
ORDERING INFORMATIONt
PACKAGE

v+

HERMETIC
DIP
14-PIN

TO-100
10-PIN
PM119K'
PM219K

PM119Y'
PM219Y
PM319Y

EPOXY
DIP
14-PIN

OPERATING
TEMPERATURE
RANGE

PM319P

MIL
IND
COM

OUTA~'0

9 -IN9

GNOA 2

A

8 +INB

+ - - +
+INA 3

B

-INA 4

7 GNDB
6 aUTe

5

v, For devices processed In total compliance to MIL-STD-883, add 1883 after part
number. Consult factory for 883 data sheet.

NOTE'
Pin 5 (V-) is connected to case.

t All commercial and industnal temperature range parts are available with burnMln
For ordering information see 1986 Data Book, Section 2.

14-PIN EPOXV DIP
(P-Suffix)

TO-100
(K-Suffix)

14-PIN HERMETIC DIP
(V-Suffix)

GENERAL DESCRIPTION
The PM-119 is adual high-speed voltage comparator designed to
operate from a single +5V supply up to ±15V dual supplies.
Open-collector outputs are provided for logic interface flexibility,

SIMPLIFIED SCHEMATIC (Each Comparator)
r---4r~~------~--~--~--------------~--~~--~----------4r-----OV+
A1
A6

3Sk!!
R2
4kU

3k0

R7
3k0

R3
4kn

~--~-----4r

~

__

~

____________

-{~Q8

Q4

R1.

13kU

+IN

A15

300n
OUTPUT

Q14
Q16
R16

6000

A17
Q17

30
GND

v-

This advance product Information describes a product in development at the time of this printing. Final speCifications may vary. Please contact
local sales office or distributor for final data sheet.

8-54

1/86, Rev. A

--------t~ PM-119/PM-219/PM-319 DUAL COMPARATOR -

ELECTRICAL CHARACTERISTICS at Vs =

ADVANCE INFORMATION

± 15V, ground pin at ground and TA = 25°C, unless otherwise noted.

PM-119/PM-219
PARAMETER

SYMBOL

CONDITIONS

Vos

Rs '" 5kU (Note 1)

los

I.

Input Offset
Voltage
Input Offset
Current
Input BIas

Current
VottageGain
(Collector)
Response TIme

'Vc

t,

PosItIve Supply
Current
PosItIve Supply

Current
NegatIve Supply

Current

PM-319
MIN

TYP

MAX

TYP

MAX

UNITS

07

40

2.0

80

mV

(Nole1)

30

75

80

200

nA

(Note 1)

150

500

250

1000

nA

(Note 2)

MIN

to

V ,N = 100mV Step
Voo =5mV(Note3)

40

V/mV

80

80

ns

43

4.3

mA

40

8

ISY

Vs =+5V,OV

ISY+

Vs= ±15V

8

11.5

8

11.5

mA

ISY-

Vs= ±15V

3

45

3

4.5

mA

NOTES:
1. The offset vottage, offset current, and bIas current gIven are the maxImum values
required to drive the collector output to wIthIn 1V of the supplies with a 7 5kU

II

load. These parameters defIne an error band and take Into account the worst
case effects of voaage gam and Input Impedance.
2. Average of 'V+ and 'V- over a ± 10V output range measured at the emItter
3. The respcnse lime specihed IS for a 100mV Input step wIth a 5mV overdrive and
IS the time reqUired for the slowest edge.

1/86, Rev. A

8-55

-----------------

----------

I»~-139/I»~-139jl/I»~-339jl
QUAD LOW-POWER
VOLTAGE COMPARATORS
Precision MOl1o]ithics Inc.

FEATURES

are offered in a design that features single power supply
operation. The input voltage range includes ground for
convenient single supply operation. The 2mA power supply
current, independent of supply voltage - coupled with the
single supply operation, makes this comparator ideal for low
power applications. Open collector outputs allow maximum
applications flexibility.

• Single or Dual Supply Operation
• Input Voltage Range Includes Ground
• Low Power Consumption (2mW/Comparator)
• Low Input Bias Current . . . . . . . . . . . . . . . . . . . . . . . .. 25nA
• Low Input Olfset Current ....................... ±5nA
• Low Olfset Voltage ............................ ±2mV
• Low Output Saturation Voltage (250m V @ 4mA)
• Logic Outputs Compatible with TTL, DTL, ECL, MOS
and CMOS
• Directly replaces LM139 and LM139A/339A Comparators

PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE

+25°C
Vos
(mV)

14-PIN
HERMETIC DIP

±2"
±5"
±2

PM139AY"
PM139Y"
PM339AY

OPERATING
TEMPERATURE
RANGE

LeC
PM139ARC/883

MIL
MIL
COM

14-PIN HERMETIC DIP
(Y-Sufflx)

"For devices processed In total compliance to MIL-STD-883, add 1883 after
part number. Consult factory for 883 data sheet
tAil commercial and industrial temperature range parts are available with
burn-In. For ordering information see 1986 Data Book, Sechon 2.

PM-139ARC/883
LCC PACKAGE
(RC-Suffix)

TYPICAL INTERFACE

GENERAL DESCRIPTION
The PM-139 has four independent voltage comparators, each
with precision DC specifications. Low offset voltage, bias
current, power consumption and output saturation voltage

DRIVING CMOS

50V

SIMPLIFIED SCHEMATIC (ONE COMPARATOR)

10Ok.n

v+

OUTPUT

DRIVING TTL
50V

,..........- i - -......-o-INPUT
I
I
I

,

I

I

~*

~*

I

I

....L

....L..

*SUBSTRATE DIODES

8-56

1/86, Rev. A

---------l(fHD PM-139/PM-139A1PM-339A QUAD LOW-POWER VOLTAGE COMPARATORS
ABSOLUTE MAXIMUM RATINGS
Supply Voltage, V+ ........................ 36V or ± 18V
Differential Input Voltage. • • • • • • • •• • • • • . • . • • • • • • • • •• 36V
Input Voltage ••.••.•••••••••••••.••••••..• -O.3V to +36V
Power Dissipation Hermetic DIP ••••.•.••.••..••• 500mW
Derate Above 100°C .••...•.••••••.......••••• 10mW/oC
Output Short-Circuit to Ground •••...•...••• Continuous

ELECTRICAL CHARACTERISTICS at V+ =

Input Current (V IN < -O.3V) ........................ 50mA
Operating Temperature Range
PM-339A •••••.•....•.....•.•...•••.•••• O°C to +70°C
PM-139A/139/139ARC ...........•..•. -55°C to +125°C
Storage Temperature Range ..•••..•.•. -65°C to +150°C
Lead Temperature (Soldering, 60 sec) ••••••••••••• 300°C

+5V, TA = 25°C, unless otherwise noted.

PM-339A

PM-139A
PARAMETER

SYMBOL

CONOITIONS

Input Offset Voltage

Vos

I Note 1)

Input Bias Current

18

I'NI+) or I'N(-) with
Oulput in Linear Range

Input Offset Current

los

I'NI+) or I'NI-)

CMVR

(Notes 2, 5, 61

Is

RL = GO on all Comparators
V+=30V

Avo

RL2! 15kO, V+ = 15V
Ito support large
Vo swingl (Note 5)

t,

V'N = TTL Logic Swing,
VREF = I.4V, VRL = 5V,
RL = 5.1kO, (Note 41

Response Time

t,

VRL = 5V, RL = 5.1kO
(Notes 3. 41

Output Sink Current

'SINK

V'N(-I2! IV, V,N(+I = 0,
V0 51.5V

Saturation Voltage

VOL

V'NI-) 2! IV, V,N(+I = 0,
IS , NK 54mA

250

ILEAK

V'N(+) 2! IV, V'NI-I = 0,
Vo= SOV

0.1

Input Common-Mode

Voltage Range
Supply Current

Voltage Gain

Large-Signal
Response Time

Output Leakage
Current

ELECTRICAL CHARACTERISTICS

MIN

TYP

MAX

MIN

TYP

PM-139
MAX

MIN

TYP

MAX

UNITS

mV
25

100

3

25

a

3.5

25

a

0.8

50

25

100

nA

50

3

25

nA

3.5

V

3.5

a

0.8

50

200

6

250

200

50

0.8

mA

200

V/mV

300

300

300

ns

1.3

1.3

1.3

I'S

16

mA

16

6
400

16

6

250

400

250

0.1

400

0.1

mV
nA

at V+ = +5V, -55°C:::; T A :::; +125°C for PM-139/139A and O°C:::; TA:::; +70°C for

PM-339A, unless otherwise noted.

PM-139A

PM-339A

PARAMETER

SYMBOL CONDITIONS

Input Offset Voltage

Vos

(Note 11

Input Offset Current

los

I'N(+) or I'N(-)

100

Input Bias Current

18

I'NI+) OR I'NI-I with
Output in Linear Range

300

Input Common-Mode
Voltage Range

CMVR

I Notes 3, 5)

Saturation Voltage

VOL

V'NI-) 2! IV, V'N(+) = 0,
IS ,NK 54mA

Output Leakage
Current

I LEAK

V'NI+) 2! IV, V'NI-) = 0,
Vo= 30V

Differential Input
Voltage

MIN

TYP

MAX

MIN

TYP

PM-139
MAX

MIN

MAX

UNITS

9

mV

150

100

nA

400

300

nA

V+-2

V

700

mV

4

a

V+-2
700

a

V+-2
700

a

TYP

I'A

36

Keep All V'N's 2! OV

NOTES:
1. Atoutputswitch point, Vo=I.4V, Rs= 00 with V+from 5V,and over the full
input common-mode range 10V to V+ -1.5V).
2. The input common-mode voltageoreitherinput voltage signal should not
be allowed to go negat,ve by more than 0.3V The upper end of the
common-mode voltage range is V+-l.5V, but either or both inputs can go
to +SOV without damage.

3.

36

36

V

The response time specified is for a 100mV input step with 5mV overdrive.
For larger overdrive signals 300ns can be obtained. See characteristics

section.
4. Sample tested.
5. Guaranteed by design.
6 Posrtrve CMVR lim,t equals V+ -1 5V fo, supply voltages other than 5V

8-57

1/86, Rev_ A

II

---------I~ PM-139/PM-139A1PM-339A QUAD LOW-POWER VOLTAGE COMPARATORS

TYPICAL APPLICATIONS
AND GATE

OR GATE
V+

200kn

39kn

3kn

+Q.07SV

1001<0

100kf2
Ao--JYVv-~

Ao--JYVv-~

100k.Q

100ldl

100kn

100kn
l=A+B+C

"0"

OIl"

"0" "1"

TIME DELAY GENERATOR
V+

10kn

15kn

200kn

30kn

v+
o

10kn

to

r-

t3

V03

30kf2

51kn

V+

10Mn

w

----------;~~

o=rto
t2

10kU

__

V02
Ve1

I

V2

51M"!

30kn

v,

v+

10Mn

oT
tt
to

lOkS}

VA'

V

+-r--1

o .--1

to

L-

J c,

10kn

INPUT GATING SIGNAL

S1kn

O.OOl,uF

14
":"

+VIN

8-58

1/86, Rev. A

----------l~ PM-139/PM-139A1PM-339A QUAD LOW-POWER VOLTAGE COMPARATORS

TYPICAL APPLICATIONS
ONE-SHOT MULTIVIBRATOR WITH INPUT LOCK-OUT

V+

n="'

~

15

lMn

0

lMn

100kn

lOMS}

~

15kn

560M2

~

+VIN

:=IEV+
-T
0

~

7

>-----

100k.\1

+
:;/'

to

Va

'1

T;; 03mSEC

10Mn

100pF

-=-

62kn

I

ONE-SHOT MULTIVIBRATOR

240k.l1

-=-

BURN-IN CIRCUIT

V+

GND

GND
(OV)

(OVI

V+l
0-

'0

3.•

kn
1.

kn

13

12

11

~

100pF

o------l H

3.

.......--+---l

~

10

9

8

'«

+VIN

2&=

~~
1
100

kn

+36V

3 .•

kn

"[ ~~''''"

2
3 .•

A58VT062V
lWATT

ONE EACH
PER BOARD

8-59

3

kn

I'

5

•

7

+36V

470kn

l, SOCKETS
~~J~i~NT

MIL STD 8B3, METHOD 1015, CONDITION B

1/86, Rev_ A

II

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

MATCHED
TRANSISTOR ARRAYS
PreClS10n Monolithlcs Inc.

9-3

Introduction

9-3

Definitions

9-4

Parameter Comparison Tables

9-5

MAT-Ol
Matched Monolithic Dual Transistor

9-11

MAT-02
Low-Noise, Matched Dual
Monolithic Transistor

9-23

MAT-04
Matched Monolithic Quad Transistor

9-2

MATCHED
TRANSISTOR ARRAYS
Precision Monolith ics Inc.

INTRODUCTION
Broadband Noise Voltage (enRMS) - The rootmean-square noise voltage referred to the input
over a specified bandwidth at a specified collector
voltage and current.

Monolithic dual and quad transistor arrays
feature inherently close matching of electrical
parameters and very low thermal differentials. In
addition, PMI arrays are specifically designed
for low offset voltage, low offset voltage drift,
low noise, and high gain, specified over a wide
range of collector currents. Monolithic duals are
optimized for amplifier input use and provide the
best possible input stage performance. The dual
and quad transistors are excellent for use in
high-performance audio systems, high-gain
instrumentation amplifiers, and precision current
mirrors.
Both the dual and quad transistors are also
designed for minimal base-to-emitter resistance
which makes log conformity excellent. For an
ideal transistor, the base-to-emitter voltage is
equal to (kT/q) In (Ie/ls). An added term, Ie rSE,
causes departure from this idealized logarithmic
relationship. The MAT-02 and new MAT-04 have
very low rSE over a wide range of collector
current. Circuits for squaring, RMS-to-DC
conversion, and logarithmic amplification can be
accurately implemented through use of these
10w-rSE products. The MAT-04 preliminary data
sheet is shown in the Advanced Products
Section.

Current Gain Match (~hFE) - The difference in
hFE between the transistors at a specified voltage
and current, expressed as a percentage of the
lower of the two hFE'S.

(1 -

hFE1) X 100
hFE2
Excess Emitter Resistance (rSE) - The
effective resistance between the base and
emitter terminals of each transistor.
Noise Voltage (enp-p) - The peak-to-peak noise
voltage referred to the input over a specified
bandwidth at a specified collector voltage and
current.
Noise Voltage Density (en) - The rms noise
voltage referred to the input in a 1Hz band
surrounding a specified frequency, measured
at a specified collector voltage and current.
Offset Current (los) - The difference between
the base currents at a specified collector voltage
and current.
Offset Current Change (~Ios/ ~VCB) - The ratio
of the change in offset current to the change in
collector-base voltage producing it.

The well-defined relationship between VSE and
collector current can also be used for temperature sensing or for generating bandgapreference voltages. The low noise, low offsets,
and high gain combined with a wide operating
range for collector current make these monolithic arrays very useful for a diverse range of
applications.

Offset Voltage (Vos) - The difference between
the base-emitter voltages (VSE1-VSE2) at a
specified collector voltage and current.

DEFINITIONS
Average Offset Current Drift (TClos) - The
ratio of the change in los to the change in temperature producing it.
Average Offset Voltage Drift (TCVos) - The ratio
of the change in Vos to the change in temperature producing it.
Bias Current (Is) - The average of the base
currents at a specified collector voltage and
current.
9-3

II
~

0
f-<
~

V~l:
@

::c

U

~

MATCHED
TRANSISTOR ARRAYS
Precision MOIlolithics Inc.

Parameter Comparison Table (Ie = 10J,LA) for MAT-02

Device
MAT-02A1E""
MAT-02B/F""
LM194
LM394
MAT-01AH
MAT-01GH
LM114A
LM114
LM115A
LM115
AD810"
AD811"
AD812"
AD813"
AD818*

BVCEO
Min
(V)

Vos
Max
(mV)

40
40
40
40
45
45
45
45
60
60
35
45
35
45
20

.05
.15
.05
.15
0.1
0.5
0.5
2.0
0.5
2.0
3.0
1.5
1.0
0.5
1.0

TCVos
Max
(p.V/OC)

hFE
Min

los
Max
(nA)

TClos
Max
(pAloC)

0.3
1
0.3
1
0.5
1.8
2.0
10
2.0
10
15
7.5
5.0
2.5
5.0

400
300
300
200
500
250
500
250
250
250
100
200
400
200
200

0.5
1.3
0.7
2.0
0.6
3.2
2.0
10
2.0
10
2.0
10
2.5
5
10

90
150
N.C.
N.C.
90
150

600
300
300
300
300

" Discontinued
""Temperature range for A-grade and B-grade is -55° C to + 125° C; temperature range for
E-grade and F-grade is -25° C to +85° C.

Parameter Comparison Table (Ie = 10J,LA) for MAT-01 to 2N-Types

Device
MAT-01GH
2N2639
2N2640
2N2642
2N2643
2N2915
2N2915A
2N2916
2N2916A
2N2917
2N2918
2N2919
2N2919A
2N2920
2N2920A
2N2060
2N2060A
2N2060B

BVCEO
Min
(V)

Vos
Max
(mV)

TCVos
Max
(p.V;oC)

hFE
Min

%hFE
Match
Max

los
Max
(nA)

TClos
Max
(pAl°C)

45
45
45
45
45
45
45
45
45
45
45
60
60
60
60
60
60
60

0.5
5.0
10
5.0
10
3.0
2.0
5.0
2.0
10
5.0
3.0
1.5
3.0
1.5
5.0
3.0
1.5

1.8
10
20
10
20
10
5.0
10
5.0
20
20
10
5.0
10
5.0
10
5.0
5.0

250
50
50
100
100
60
60
150
150
60
150
60
60
150
150
25
25
25

8
10
20
10
20
10
15
10
15
20
20
10
10
10
10
10
10
10

3.2
20
40
10
20
17
26
7
10

150
1000
2000
500
375
600
900
N.C.
300
1450
750
600
600
N.C.
300
N.C.
N.C.
N.C.

17
7
17
17
7
7
40
40
40

Notes: 1. TClos Max and los Max calculated from published data.
2. N.C. = Insufficient published data to calculate.
3. All of above are physically interchangeable pin-for-pin with MAT-01 and MAT-02 series.
9·4

MAT-Ol
MATCHED MONOLITHIC
DUAL TRANSISTOR
Precisioll Monolithics Inc.

FEATURES

ABSOLUTE MAXIMUM RATINGS (Note 4)

• Low VOS (VBE Match) ....................... 40~V Typ
100~V Max
• Low TCVos ............................ 0.5~ V,· C Max
• High hFE .................................... 500 Min
• Excellent hFE Linearity from 10nA to 10mA
• Low Noise Voltage ......... 0.23~Vp_p - 0.1 Hz to 10Hz
• High Breakdown ............................ 45V Min

Collector-Base Voltage (BVCBO)
MAT-01AH, GH, N ............................... 45V
Collector-Emitter Voltage (BVCEO)
MAT-01AH, GH, N ............................... 45V
Collector-Collector Voltage (BVcd
MAT-01AH, GH, N ............................... 45V
Emitter-Emitter Voltage (BVEE)
MAT-01AH, GH, N ............................... 45V
Emitter-Base Voltage (BVEBO) (Note 1) ............•... 5V
Collector Current (Ic) ............................. 25mA
Emitter Current (IE) ............................... 25mA
Total Power Dissipation
Case Temperature:540·C (Note 2) .............. 1.BW
Ambient Temperature:5 70·C (Note 3) ......... 500mW
Operating Ambient Temperature. . . . . .. -55· C to +125· C
Operating Junction Temperature ...... -55· C to +150· C
Storage Temperature ................. -65· C to +150· C
Lead Temperature (Soldering, 60 sec) ............. 300· C
DICE Junction Temperature. . . . . . . . . .. -65· C to +150· C

ORDERING INFORMATIONt
TA=2S·C

VosMAX
(mV)
0.1

0.5

PACKAGE

OPERATING
TEMPERATURE
RANGE

MAT01AH'
MAT01GH'

MIL
MIL

• For devices processed in total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet.
t All commercial and industrial temperature range parts are available with
burn-in. For ordering information see 1986 Data Book, Section 2.

NOTES:
1. Application of reverse bias voltages In excess of rating shown can result In
degradation of hFE and hFE matching characteristics. Do not attempt to
measure BVEBO greater than the 5V rating shown.
2. Rating applies to applications using heat sinking to control case temperature. Derate linearly at 16.4mW/' C for case temperatures above 40' C.
3. Rating applies to applications not using heat sinking; device in free air only.
Derate linearly at 6.3mW/' C for ambient temperatures above 70' C.
4. Absolute maximum ratings apply to both DICE and packaged devices.

GENERAL DESCRIPTION
The MAT-Ol is a monolithic dual NPN transistor. An exclusive Silicon Nitride "Triple-Passivation" process provides
excellent stability of critical parameters over both temperature and time. Matching characteristics include offset voltage
of 40~V, temperature drift of 0.15~VI· C, and hFE matching of
0.7%. Very high hFE is provided over a six decade range of
collector current, including an exceptional hFE of 590 at a
collector current of only 10nA. The high gain at low collector
current makes the MAT-Ol ideal for use in low-power,
low-level input stages.

PIN CONNECTIONS
(TOP VIEW)

81

C@."C2
2

E1 3

6

82

TO-7a
(H-Sufflx)

5 E2

NOTE:
Substrate

IS

connected to case.

9-5

1/86, Rev. A

I

~

MAT-01 MATCHED MONOLITHIC DUAL TRANSISTOR

ELECTRICAL CHARACTERISTICS at VCB = 15V,

Ic

= 10J'A, TA = 25

C, unless otherwise noted.

0

MAT-G1AH
SYMBOL

CONDITIONS

Breakdown Voltage

BVeEo

le= IOOI'A

Offset Voltage

Vos

Offset Voltage Stability
First Month
Long-Term

Vos/Time

Offset Current

los

0.1

0.6

0.2

3.2

nA

Bias Current

la

13

20

18

40

nA

45

(Note 1)
(Note 2)

Current Gain

hFE

Current Gain Match

ah FE

le= IOI'A
100nA Sic SlOmA

enp~p

Broadband NOise
Voltage
Noise Voltage
Density

Voltage

MAX

MIN

0.1

2.0
0.2

MAX

UNITS

0.10

0.5

mV

V

2.0
0.2

590
770
840

500

TYP

45
0.04

le= 10nA
le= IOI'A
le= 10mA

Low Frequency Noise

MIN

TYP

MAT-01GH

PARAMETER

250

I'ViMo

430
560
610

0.7
0.8

3.0

1.0
1.2

8.0

O.IHz to 10Hz (Note 3)

0.23

0.4

0.23

0.4

enRMS

1Hz to 10kHz

0.60

en

fo= 10Hz (Note 3)
fo= 100Hz (Note 3)
fo= 1000Hz (Note 3)

7.0
6.1
6.0

9.0
7.6
7.5

7.0
6.1
6.0

9.0
76
7.5

0.5

3.0

0.8

8.0

I'VN

15

3

70

pAN

0.60

%
I'Vp.p
I'VRMS

nV/y'HZ

Offset Voltage Change

avOS/aVeB

OS VeaS 30V

Offset Current Change

aIOS/avCB

OS Vea S30V

Collector-Base
Leakage Current

leBO

VeB = 30V, IE=O
(Note 4)

15

50

25

200

pA

Collector-Emitter
Leakage Current

ICES

VCE = 30V, VaE = 0
(Notes 4, 6)

50

200

90

400

pA

Collector-Collector
Leakage Current

Icc

Vee = 30V, (Note 6)

20

200

30

400

pA

Collector Saturation
Voltage

IB= O.lmA, le= lmA
IB= lmA, Ic= 10mA

0.12
0.8

0.20

0.12
0.8

0.25

VeEISATi

V

Gain-Bandwidth Product

IT

VCE = 10V, Ic = 10mA

450

450

MHz

Output Capacitance

COb

VeB =15V,I E =0

2.8

2.8

pF

Collector-Collector
Capacitance

C cc

Vee=O

8.5

8.5

pF

ELECTRICAL CHARACTERISTICS at

VCB

= 15V, Ic = 10J'A, -55

0

C::;

TA::;

+125

0

C, unless otherwise noted.

MAT-01AH
PARAMETER

SYMBOL

Offset Voltage

Vos

Average Offset
Voltage Drift

TCVos

Offset Current

los

Average Offset
Current Drift

TClos

Bias Current

IB

Current Gain

hFE

Collector-Base

CONDITIONS

MIN

(Note 7)

(NoteS)

167

MAT-01GH

TYP

MAX

TYP

MAX

UNITS

0.06

0.15

0.14

0.70

mV

0.15

0.50

0.35

1.8

I'vrc

0.9

8.0

1.5

15.0

nA

10

90

15

150

pArC

28

60

36

130

nA

400

MIN

77

300

Iceo

TA =125°C, Vee=30V,
IE = 0 (Note 4)

15

80

25

200

nA

Collector-Emitter
Leakage Current

ICES

TA = 125°C, VeE = 30V,
VeE = 0 (Notes 4, 6)

50

300

90

400

nA

Collector-Collector
Leakage Current

Icc

TA = 125°C, Vee = 30V
(Note 6)

30

200

50

400

nA

Leakage Current

9-6

1/86, Rev. A

------------l~ MAT-01 MATCHED MONOLITHIC DUAL TRANSISTOR

DICE CHARACTERISTICS
1.
2.
3.
5.
6.
7.

DIE SIZE 0.035 X 0.025 Inch, 875 sq. mils
(0.89 X 0.64 mm, 0.58 sq. mm)

COLLECTOR (1)
BASE (1)
EMITTER (1)
EMITTER (2)
BASE (2)
COLLECTOR (2)

For additional DICE information refer to
1986 Data Book, Section 2.

WAFER TEST LIMITS at VCB = 15V and Ic = 10!,A, TA = 25°C, unless otherwise noted.
MAT-01N
PARAMETER

SYMBOL

Breakdown Voltage

BVCEO

Offset Voltage

CONDITIONS

LIMITS

UNITS

45

VMIN

Vos

05

mVMAX

Offset Current

los

3.2

nAMAX

Bias Current

Is

40

nAMAX

Cu rrent Gai n

hFE

250

MIN

Current Gain Match

~hFE

8.0

% MAX

Offset Voltage Change

~Vos/~Vcs

Offset Current Change

~los/~Vcs

Collector Saturation Voltage

VCEISATI

0" v cs " 30V

I B =O.lmA, Ic= lmA

8.0

".VN MAX

70

pAN MAX

0.25

VMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to vanations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consun factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at VCB = 15V and Ic = 10l'A, TA = 25°C, unless otherwise noted.
MAT-01N
TYPICAL

PARAMETER

SYMBOL

CONDITIONS

Average Offset Voltage Drift

TCVos

0.35

".VloC

Average Offset Current Drift

TClos

15

pArC

Collector-Emitter-Leakage Current

ICES

VCE = 30V, VSE = 0

90

pA

Collector-Base-Leakage Current

leBo

VcB =30V,I E=0

25

pA

Gain Bandwidth Product

fT

VCE = 10V, Ie = lOrnA

450

MHz

Offset Voltage Stability

~Vos/T

First Month (Note 1 )
Long-Term (Note 2)

2.0
02

".V/Mo

NOTES:
1. Exclude first hour of operation to allow for stabilization
2 Parameter describes long-term average dnft after first month 01 operation
3 Sample tested.
The collector-base (I CBO ) and collector-emitter (ICES) leakage currents
may be reduced by a factor of two to ten times by connecting the substrate
(package) to a potential which IS lower than either collector voltage.

5.
6.

9-7

UNITS

Guaranteed by los test limits over temperature
Icc and ICES are guaranteed by measurement of I CBO
Vos
Guaranteed by Vos test (TCVos = for Vos '" VSE) T = 298°K for TA
= 250C
T

1/86, Rev. A

-----------I~ MAT·01 MATCHED MONOLITHIC DUAL TRANSISTOR

TYPICAL PERFORMANCE CHARACTERISTICS
OFFSET VOLTAGE
VI TEMPERATURE

BASE-EMITTER VOLTAGE
COLLECTOR CURRENT

OFFSET VOLTAGE VI TIME

200

~

OV.;.vCB';;;30V

VI
10mA

'0

~

'50

;;
3
w

""~

'00

§;

~

V

0

~

-

- -

50

I--

o

o

-50

-26

~

g

w

-2

"
"" .....
z

MAT-OM

~\~~

-

100

126

j-s

'DOpA

-'0

'OpA

160

o

1

2

3

V

l'• 800

8

7

8

9

10 11 12 13

CURRENT GAIN
TEMPERATURE

V

--

~

~'0.0

z

~ 800 I--+--+--;I'-'!--!-=>""!-+--+----I
~

a:
a:

f= :M~;:~;

~

g

2OO.~~

-75

~', = '25'C.

III

/

1.0

II~A I ~~"C

0:

0,1

-50

__'---L~__~-..l__~-.J
-26

50

75

100

0.01
0.01

125

Ilil

I

0.'

'.0

I

10.0

100.0

COLLECTOR CURRENT (rnA)

NOISE VOLTAGE

NOISE CURRENT DENSITY

r---,--...,--,...--,....-.....,

VI

1000 , . . . . - . . . . . , . - - . . , . . - - , - - - , - - . . . . . ,
NOISE CURRENT
DENSITY T A'" 25cC

~

~

.
.i!lili

>
;;

i.

GAIN·BANDWIDTH
COLLECTOR CURRENT

'000
500

g 200
g

MAT-OI,

~

~

"III"

I

60

z

'0

~

/
1/

20

/

Z

z

"./:'

O·'O..., - _....._-,l.O--,.J.OO---'l.k--'.JOk

9·8

VeE'" 10V

/

/
10pA

FREQUENCY (Hz)

TA = 26°C_

J

~

i5

/

/

g:100

'00

z

FREQUENCY (Hz)

lilll

z
o

COLLECTOR CURRENT

Ie" 300J..(A TYPICAL

aoo

700

," '0,'.

I

Ie = 10f,lA TYPICAL

600 80D

\"~'C

S

J

Ie .. 1o~ WORST CASE

400

~

a400 I---A----.;F---i-

1nA 10nA l00nA 'J.lA 10~ 10Cl,uA lmA 10rnA tCOmA

'000

300

w

I
I

o

200

SATURATION VOLTAGE
COLLECTOR CURRENT

800r--+--~--+--4--~~4

TA=25°C VeB'" 15V

200

100

tt

100.0

I

a:
a:

a

10

STRAIGHT Ll

.7

VI

MAT~

i---'""

t. VBE = 0
TA = 125'C

50

50

nA

Icc

Vec = VMAX
TA = 125'C

30

40

nA

Offset Voltage

Collector-Emitter

Leakage Current
Collector-Collector
Leakage Current

80

40

0.08
0.03

90

40

60
275
225
175
150

0.3

"V/'C

15

nA

150

pArC

70

nA

250
200
150
100

CJ)

~

0

~
....

~r-<
Cl
~

NOTES:
V
1. Guaranteed by Vos test (TeVos ~ ~ for Vos 
e-

O;

w

"~

+125°C

01
+25°C

111111
001

COLLECTOR CURRENT (rnA)

NOISE VOLTAGE
DENSITY VB FREQUENCY

15c

001
0001

100

10

COLLECTOR CURRENT (rnA)

VB

~
f-

01

COLLECTOR CURRENT (rnA)

IC=lJ.'A-

...... 1-0.

10

Ie .

\O"A~

0

>

~

I'..

'"~

iT11111
10

01

001
0001

COLLECTOR CURRENT (mA)

~

10

8

&:

04

100

I

100

w
Z

in

"e-~

10

SMALL-SIGNAL OUTPUT
CONDUCTANCE VB
COLLECTOR CURRENT

I

10

"

I--'

01

1000

w
Z

~

001

COLLECTOR CURRENT (rnA)

SMALL-SIGNAL INPUT
RESISTANCE VB COLLECTOR
CURRENT

VeE"" 5V

10

01
0001

175

TEMPERATURE (oC)

WI~I+'5U

001

~

~

100

05

IL

10

'1

;:

(EXCLUDES ICBO)

100

01

06

0001

100

0

~

BASE-EMITTER-ONVOLTAGE VB COLLECTOR
CURRENT

I

e-

'00

COLLECTOR CURRENT (mA)

07

L

I

V v,;;=I'IlA
./

/
/

GAIN BANDWIDTH
COLLECTOR CURRENT

1000

V

700

~

1I111

....

600

;;

w

VB

Ic=~mA .,1

800

J 1,I,Iimreitt

".

~
~

z

900

1111

I
800

CURRENT GAIN

vs TEMPERATURE

le= lmA

I

01
01

10

100

lk

FREQUENCY (Hz)

9-15

10k

lOOk

---------I1ffiD

MAT-02 LOW NOISE, MATCHED DUAL MONOLITHIC TRANSISTOR

TYPICAL PERFORMANCE CHARACTERISTICS

TOTAL NOISE VB
COLLECTOR CURRENT

NOISE CURRENT
DENSITY VB FREQUENCY
100

100

_ T A "'2SoC
- V c e - SV

1

~

r---

~

~

1.0

a

IC=10,uA

w

0

z

01

1w

Ic=lmA

~

0

~

le-1OmA-

i

~

80

10

.......

'"

==

f = 1kHz

60

...........

RS

z

=

~

10

if

lOOkn

./

Til

40

"
g

§

100

t--.

20

III

R~ - ~~kn

....

10

100

lk

10k

AS'" lkn
001

0001

lOOk

01

0.01

100

40

/'

~
w

~

";:;z

/'

40

TA '" 25°C

~

1\

;:;"
Z

~
,,""

"

30

w

\.

I

":\:

/

,

TA ='25 0 C

8

13

I

01

30

125

COLLECTOR-BASE
CAPACITANCE VB
REVERSE BIAS VOLTAGE

50

10

100

75

TEMPERATURE (OC)

COLLECTOR-TO-COLLECTOR
CAPACITANCE VB COLLECTOR-TOSUBSTRATE VOLTAGE

COLLECTOR-TO-COLLECTOR
LEAKAGE VB TEMPERATURE

""
-""

50

25

COLLECTOR CURRENT {mAl

FREOUENCY (Hz)

::;
g;

V

01

o

001
01

1

I II

0

~

IC=1JA

IIIII

COLLECTOR-TO-BASE
LEAKAGEwTEMPERATURE

~

20

20

I

\

\
~

10

10

-

-

001
25

50

75

125

100

10

20

30

40

COLLECTOR-TO-COLLECTOR
CAPACITANCE VB
REVERSE BIAS VOLTAGE

10

30

40

50

8'
TA

1.5

20

REVERSE BIAS VOLTAGE (VOL IS)

EMITTER-BASE CAPACITANCE
VB REVERSE BIAS VOLTAGE

25

20

10

50

COLLECTOR-TO-SUBSTRATE VOL TAGE (VOLTS)

TEMPERATURE ( C)

84

=1250C

~

1\

\

w

;:;"

82
80

z

78

13

::
""

r--.....

I

~

76
74

" ."

I
TA=25°C

"-

-

.......

"""" ........

72
70

"""" r.......

68

""""

66
10

20

30

40

01

50

0.2

03

04

0.5

REVERSE BIAS VOLTAGE (VOL TS)

REVERSE BIAS VOLTAGE (VOL IS)

9-16

1/86, Rev. A

---------I1fHD

MAT-02 LOW NOISE, MATCHED DUAL MONOLITHIC TRANSISTOR

LOG CONFORMANCE TEST CIRCUIT

R1

1kn
1V-l0V
RAMP
GENERATOR

R2

10kS1

I
LOG CONFORMANCE TESTING

An error term must be added to this equation to allow for the
bulk resistance (rBE) of the transistor. Errordue to the op amp
input current is limited by use of the OP-15 BIFET-input op
amp. The resulting AMP-01 input is:

The log conformance of the MAT-02 is tested using the
circuit shown above. The circuit employs a dual transdiode
logarithmic converter operating at a fixed ratio of collector
currents that are swept over a 10:1 range. The output of each
transdiode converter is the VBE of the transistor plus an error
term which is the product of the collector current and rBE, the
bulk emitter resistance. The difference of the VBEisamplified
at a gain of x100 by the AMP-01 instrumentation amplifier.
The differential emitter-base voltage (AVBE) consists of a
temperature-dependent DC level plus an AC error voltage
which is the deviation from true log conformity as the collector
currents vary.

IC1
kT
AVBE = - In q
IC2

q

k=
q=
T=
Is =
Ic =

s

(2)

A ramp function which sweeps from 1V to 10V is converted by
the op amps to a collector current ramp through each
transistor. Because IC1 is made equal to 10 IC2' and assuming
TA = 25 0 C, the previous equation becomes:
AVBE = 59mV + 0.9 IC1 rBE (ArBE - 0)
As viewed on an oscilloscope, the change in AV BE for a 10:1
change in Ie is then displayed as shown below:

The output of the transdiode logarithmic converter comes
from the idealized intrinsic transistor equation (for silicon):
Ic
kT
V BE = - In -I where

+ IC1 rBE1 - IC2 rBE2

(1)

_ _ _ _ )
VOUT {

Boltzmann's Constant (1.38062 x 10- 23 J/oK)
Unit Electron Charge (1.60219 x 10- 19 C)
Absolute Temperature, 0 K (= 0 C + 273.2)
Extrapolated Current for VBE-O
Collector Current

-

-

-

-

-

-

-

-

~VO"90IC1'BE

)

59V

COLLECTOR CURRENT
1rnA

9-17

I

10mA

1/86, Rev_ A

_ _ _ _ _ _ _ _--IIfMDMAT-02 LOW NOISE, MATCHED DUAL MONOLITHIC TRANSISTOR

With the oscilloscope AC coupled. the temperature dependent term becomes a DC offset and the trace represents the
deviation from true log conformity. The bulk resistance can
be calculated from the voltage deviation avo and the change
in collector current (9mA):
avo
1
rBE = 9mA x 100

easily accomodated by various offsetting techniques. Protective diodes across each base-to-emitter junction would
normally be needed. but these diodes are built into the
MAT -02. External protection diodesare therefore not needed.
For the circuit shown in Figure 1. the operational amplifiers
make 11 = VX/R1. 12 = Vy/R2. 13 = VZ/R3. and 10 = Vo/Ro.The
output voltage for this one-quadrant. log-antilog multiplier/
divider is ideally:

(3)

This procedure finds rBEfor Side A. Switching R1 and R2 will
provide the rBEfor Side B. Differential rBEis found by making
R1 = R2.

R3 Ro Vx Vy
Vo=---R1 R2
Vz

(Vx. VY. Vz>O)

(4)

If all the resistors (Ro. R1• R2• R3) are made equal. then Vo=
VxVy/Vz. Resistorvalues of 50kO to 100kOare recommended
assuming an input range of 0.1V to +10V.

APPLICATIONS: NONLINEAR FUNCTIONS
MULTIPLIER/DIVIDER CIRCUIT
The excellent log conformity of the MAT -02 over a very wide
range of collector current makes it ideal for use in log-antilog
circuits. Such nonlinear functions as multiplying. dividing.
squaring. and square-rooting are accurately and easily
implemented with a log-antilog circuit using two MAT-02
pairs (see Figure 1). The transistor circuit accepts three input
currents (110 12, and 13) and provides an output current 10
according to 10 = 1112/13. All four currents must be positive in
the log-antilog circuit. but negative input voltages can be

ERROR ANALYSIS
The base-to-emitter voltage of the MAT-02 in its forwardactive operation is:
kT Ic
VBE=-InI s +rBElc.VCB-O
q

(5)

The first term comes from the idealized intrinsic transistor
equation previously discussed (see equation (1)).

Co

R3

Vz
'1~

R1

Vx

-=3.3krt

-'0

RO

R2

Vy

Vo
R3RO VXVy

vo= R,R 2

-=-

--vz

-=-

Figure 1. One-Quadrant Multiplier/Divider

9-18

1/86, Rev. A

---------l~ MAT-02 LOW NOISE, MATCHED DUAL MONOLITHIC TRANSISTOR

approximately 26m V and the error due to an rBEleterm will be
rBElcl26mV. Using an rBE of 0.40 for the MAT-02 and
assuming a collector-current range of up to 200!,A, then a
peak error of 0.3% could be expected for an rBEle error term
when using the MAT-02. Total error is dependent on the
specific application configuration (multiply, divide, square,
etc.) and the required dynamic range. An obvious way to
reduce ICrBE error is to reduce the maximum collector
current, but then op amp offsets and leakage currents
become a limiting factor at low input levels. A design range of
no greater than 10!,A to 1mA is generally recommended for
most nonlinear function circuits.

,-----________---.'1
v,o---~4A~1_----~---_k

3kD.

A powerful technique for reducing error due to lerBEis shown
in Figure 2. A small voltage equal to ICrBE is applied to the
transistor base. For this circuit:

Figure 2. Compensation of Bulk Resistance Error

Extrinsic resistive terms and the Early effect cause departure
from the ideal logarithmic relationship. For small VeB , all of
these effects can be lumped together as a total effective bulk
resistance rBE. The rBEle term causes departure from the
desired logarithmic relationship. The rBEterm for the MAT -02
is less than 0.50and ArBEbetween the two sides is negligible.

(10)
The error from rBEle is cancelled if Rc/R2 is made equal to
rBE/R 1. Since the MAT -02 bulk resistance is approximately
0.390, an Rc of 3.90 and R2 of 10R 1 will give good error
cancellation.

Returning to the multiplier/divider circuit of Figure 1 and
using Equation (4):

In more complex circuits, such as the circuit in Figure 1, it
may be inconvenient to apply a compensation voltage to
each individual base. A better approach is to sum all
compensation to the bases of Ql. The "A" side needs a base
voltage of (Vo/Ro+ Vz/R3) rBEand the "S" side needs a base
voltage of (Vx/Rl + Vy/R2) rBE. Linearity of better than ±0.1 %
is readily achievable with this compensation technique.

VBE1A + VBE2A - VBE2B - VBE1B + (11 + 12- 10 - 13) rBE = 0
If the transistor pairs are held to the same temperature, then:
11 12
kT
ISlA IS2A
kT
-In - - = -In I
I
+ (1 1 + 12 - 10 - 13) rBE
q
13 10
q
51B 52B

(6)

Operational amplifier offsets are another source of error. In
Figure 2, the input offset voltage and input bias current will
cause an error in collector current of (Vos/R1) + lB. A low
offset op amp, such as the OP-07 with less than 75!'V of Vos
and IB of less than ±3nA, is recommended. The OP-22/32, a
programmable micropower op amp, should be considered if
low power consumption or single-supply operation is
needed. The value of frequency-compensating capaCitor
(Co) is dependent on the op amp frequency response and
peak collector current. Typical values for Co range from
30pF to 300pF.

If all the terms on the right-hand side were zero, then we
would have In (1 1 12/1310) equal to zero which would lead
directly to the desired result:
10 =

11 12

1;-' where 11, 12, 13, 10> 0

(7)

Note that this relationship is temperature independent. The
right-hand side of Equation (6) is near zero and the output
current 10 will be approximately 11 12/13' To estimate error,
define I$> as the right-hand side terms of Equation (6):
ISlA IS2A
q
1$>= In I
I
+kT(ll+ 12- lo- 13)rBE
SlB S2B

FOUR-QUADRANT MULTIPLIER
A simplified schematic for a four-quadrant log/antilog multiplier is shown in Figure 3. As with the previously discussed
one-quadrant multiplier, the circuit makes 10 = 11 12113. The
two input currents, 11 and 12, are each offset in the positive
direction. This positive offset is then subtracted out at the
output stage. Assuming ideal op amps, the currents are:

(8)

For the MAT-02, In (ISA/lsB) and lerBE are very small. For
small 1$>, .1$> - 1 + I$> and therefore:

(9)

(11 )
Vx
Vy
VR
Vo
VR
1 0 = - + - + - + - 13=Rl
R1
R2
Ro'
R2

The In (ISA/lsB) terms in I$> cause a fixed gain error of less than
±0.6% from each pair when using the MAT-02, and this gain
error is easily trimmed out by varying Ro. The lerBEterms are
more troublesome because they vary with signal levels and
are multiplied by absolute temperature. At 25° C, kT/q is

From 10= 11 12/13, the output voltage will be:
Ro R2
VO="""R;2

9-19

Vx Vy
VR

(12)

1/86, Rev. A

II

------------I~ MAT·02 LOW NOISE, MATCHED DUAL MONOLITHIC TRANSISTOR

Collector-current range is the key design decision. The
inherently low rSE of the MAT-02 allows the use of a relatively
high collector current. For input scaling of ± 10V fUll-scale
and using a 10V reference, we have a COllector-current range
for I, and 12 of:
( -10

R,

+1Q..)
:5lc:5 (1Q..+~)
R2
R,
R2

MULTIFUNCTION CONVERTER
The multifunction converter circuit provides an accurate
means of squaring, square rooting, and of raising ratios to
arbitrary powers. The excellent log conformity of the MAT-02
allows a wide range of exponents. The general transfer
function is:

(13)

Vo = Vv
Practical values for R, and R2 would range from SOkO to
100kO. Choosing an R, of 82kO and R2 of 62kO provides a
collector-current range of approximately 391-'A to 2831-'A. An
Ro of 108kO will then make the output scale factor 1/10 and
Vo = Vx Vv/10. The output, as well as both inputs, are scaled
for ± 10V fUll-scale.

Vz
Vo= 10 ( 10

)m

(16)

As with the multiplier/divider circuits, assume that the
transistor pairs have excellent matching and are at the same
temperature. The In ISA/lss will then be zero. In the circuit of
Figure 4, the voltage drops across the base-emitter junctions
of 01 provide:

(I, + 12-13-10) rSE+ pVo= 0
The currents are known from the previous discussion, and
the relationship needed is simply:
Ro

(1S)

Vx , Vv, and Vz are input voltages and the exponent "m" has a
practical range of approximatelyO.2 to S. Inputs Vxand Vvare
often taken from a fixed reference voltage. With a REF-01
providing a precision + 10V to both Vx and Vv, the transfer
function would simplify to:

Linear error for this circuit is substantially improved by the
small correction voltage applied to the base of 01 as shown
in Figure 3. Assuming an equal bulk emitter resistance for
each MAT-02 transistor, then the error is nulled if:

Vo=~Vo

(vz)m
Vx

(14)

Iz
Rs
kT
Rs + KRA VA =
In T,;

(17)

q

The output voltage is attenuated by a factor of rseiRo and
applied to the base of 01 to cancel the summation of voltage
drops due to rSEIC terms. This will make In (I, 12/1310) more
nearly zero which will thereby make 10 = I, 12/13 a more
accurate relationship. Linearity of better than 0.1% is readily
achievable with this circuit if the MAT -02 pairs are carefully
kept at the same temperature.

Iz is Vz/R, and Ix is Vx/R,. Similarly, the relationship for 02 is:
Rs
Rs+(1·K)RA

kT
10
In q
Iv

(18)

-::--..:=--:-:-::- VA = -

10 is VoiRo and Iv is Vv/R,. These equations for 01 and 02
can then be combined.
In ~ = In
Rs + KRA
Rs + (1-K)RA
Ix
Iv

.!sL

(19)

+15V

RO=10Bkn

R,=82kn
R2 =62kn

"2

Vx

",

Vy

",

"2

"2
"2

Figure 3. Four-Quadrant Multiplier
9·20

1/86, Rev. A

----------IM

MAT-02 LOW NOISE, MATCHED DUAL MONOLITHIC TRANSISTOR

Substituting in the voltage relationships and simplifying
leads to:

better than 1% are readily achievable with this circuit configuration and can be better than ±0.1% over a limited operating
range.

Vz)m,where
Ro
( Vx
Vo=R';'"Vy
FAST LOGARITHMIC AMPLIFIER
The circuit of Figure 5 is a modification of a standard
logarithmic amplifier configuration. Running the MAT-02 at
2.5mA per side (full-scale) allows a fast response with wide
dynamic range. The ci rcuit has a 7 decade current range. a 5'
decade voltage range. and is capable of 2.5l'sec settling time
to 1% with a 1 to 10V step.

(20)
m = Re+ KRA
Re + (1-K) RA
The factor "K" is a potentiometer position and varies from
zero to 1.0, so "m" ranges from ReI(R A+ Re) to (Re+ RA)/R e.
Practical values are 1250 for Reand 5000 for RA; these values
will provide an adjustment range of 0.2 to 5.0. A value of
100kO is recommended for the R1 resistors assuming a fullscale input range of 10V. As with the one-quadrant multiplier/
divider circuit previously discussed. the Vx. Vy. and Vzinputs
must all be positive.

The output follows the equation:

R3 + R2 kT
VREF
Vo=----In - R2
q
Vin

(21 )

The output is inverted with respect to the input. and is nominally -1 Vldecade using the component values indicated.

The op amps should have the lowest possible input offsets.
The OP-07 is recommended for most applications. although
such programmable micropower op amps as the OP-22 or
OP-32 offer advantages in low-power or single-supply
circuits. The micropower op amps also have very low inputbias-current drift. an important advantage in log/antilog
circuits. External offset nulling may be needed. particularly
for applications requi ring a wide dynamic range. Frequencycompensating capacitors. on the order of 50pF. may be
required for A2 and A3. Amplifier A1 is likely to need a larger
capacitor. typically 0.0047I'F. to assure stability.
Accuracy is limited at the higher input levels by bulk emitter
resistance. but this is much lower for the MAT-02 than for
other transistor pairs. Accuracy at the lower signal levels
primarily depends on the op amp offsets. Accuracies of

LOW-NOISE x1000 AMPLIFIER
The MAT-02 noise voltage is exceptionally low. only
1nVlV"HZ at 10Hz when operated over a collector-current
range of 1 to 4mA. A single-ended x1000 amplifier that takes
advantage of this low MAT-02 noise level isshown in FigureS.
In addition to low noise. the amplifier has very low drift and
high CMRR. An OP-32 programmable low-power op amp is
used for the second stage to obtain good speed with minimal
power consumption. Small-signal bandwidth is 1MHz. slewrate is 2.4V1l's, and total supply current is approximately
2.8mA.

1

RO (Vz m
Vo=-VyRl
Vx

R1=100kn
Ro~100kn

CO -O.0047MF

R A=500n
Rs =125n
VyO-~AA~----------------------~

Figure 4. Multifunction Converter
9-21

1/86, Rev. A

------------J~ MAT-D2 LOW NOISE, MATCHED DUAL MONOLITHIC TRANSISTOR

Transistors 02 and 03 form a 2mA current source
(0.65V/3300 - 2mA). Each collector of 01 operates at 1mAo
The OP-32 inputs are 3V below the positive supply voltage
(RLle - 3V). The OP-32's low input offset current, typically
less than 1nA. and low offset voltage of 1mV cause negligible
error when referred to the amplifier input. Input stage gain is
gmRL. which is approximately 100 when operating at Ie of
1ma with RL of 3kO. Since the OP-32 has a minimum openloop gain of 500.000. total open-loop gain for the composite
amplifier is over 50 million. Even at closed-loop gain of 1000.
the gain error due to finite open-loop gain will be negligible.
The OP-32 features excellent symmetry of slew-rate and
very linear gain. Signal distortion is minimal.

The IS ET• using ±15V supplies and an RSET of 549kO. is
approximately 52p.A which will result in supply current of
784p.A.
Dynamic range of this amplifier is excellent; the OP-32 has an
output voltage swing of ±14V with a ±15V supply.
Input characteristics are outstanding. The MAT-02B/F has
offset voltage of less than 150p.V at 25°C and a maximum
offset drift of 1p.V/oC. NUlling the offset will further reduce
offset drift. This can be accomplished by slightly unbalancing
the collector load resistors. This adjustment will reduce the
drift to less than 0.1p.V/oC.
Input bias current is relatively low due to the high current
gain of the MAT-02. The minimum f3 of 400 at 1mA for the
MAT-02B/F implies an input bias current of approximately
2.5p.A. This circuit should be used with signals having
relatively low source impedance. A high source impedance
will degrade offset and noise performance.

Frequency compensation is very easy with this circuit; just
vary the set-resistor Rs for the desired frequency response.
Gain-bandwidth of the OP-32 varies directly with the supply
current. A set resistor of 549kO was found to provide the best
step response for this circuit. The resultant supply current is
found from:
RSET

(V+) - (V-) - (2VBE) I - 15 I
.SYSET
ISET

This circuit configuration provides exceptionally low input
noise voltage and low drift. Noise can be reduced even
further by raising the COllector currents from 1mA to 3mA.
but power consumption is then increased.

(22)

+'5V

o------""'tn;"'"T'M,,---,

4kn
V,N
(0 to lOV)

".
20kn

V ,N

"3
75kn

+15V

V REF

'OV

"2

4kn

..on

",
R2" TEL LABS Q81E (+0 35%/-C)

-15V
-15V

+15V

Figure 5. FaIt Logarithmic Amplifier

Figure 6. Low-Noise, Single-Ended X1000 Amplifier

9-22

1/86, Rev. A

MAT-04
MATCHED MONOLITHIC
QUAD TRANSISTOR
Precision Monolithics Inc.

PRELIMINARY
FEATURES
•
•
•
•
•
•

PIN CONNECTIONS

Low Offset Voltage ..................... 150fLV Max
High Current Gain .........••..•••........ 400 Min
Excellent Current Gain Match ...•.....•.... 2% Max
Low Noise Voltage at 100Hz, 1mA .•..• 2.5nV/vHzMax
Excellent Log Conformance ....•... rBE
O.SO Max
Matching Guaranteed for All Transistors

=

14-PIN HERMETIC DIP
(V-Suffix)
14-PIN EPOXV DIP
(P-Suffix)

ORDERING INFORMATIONt
PACKAGE
TA = 25°C
VosMAX
lfN)

HERMETIC
DIP
14-PIN

150fLV
150fLV
300fLV
300fLV

MAT04AY'
MAT04EY
MAT04BY'
MAT04FY

EPOXY"
DIP
14-PIN

OPERATING
TEMPERATURE
RANGE
MIL
IND
MIL
IND

MAT04EP
MAT04FP

ABSOLUTE MAXIMUM RATINGS (Note 1)
Collector-Base Voltage (BVCBO) . . . . . . . . . . . . . . . . . . . 40V
Collector-Emitter Voltage (BVCEO ) . . . . . . . . . . . . . . . . .. 40V
Collector-Collector Voltage (BVcc) ................. 40V
Emitter-Emitter Voltage (BVEE ) . . . . . . . . . . . . . . . . . . . . 40V
Collector Current .............................. 30mA
Emitter Current ............................... 30mA
Substrate (Pin-4 to Pin-ll) Current ................ 30mA
Total Power Dissipation (Note 2) ................ 500mW
Operating Temperature Range
MAT-04AY, BY ..................... -55 D C to + 125D C
MAT-04EY, FY, EP, FP ............... -25 D C to +85 D C
Storage Temperature
Y Package ........................ -65 D C to + 150D C
P Package
.. -65 D C to + 125D C
Lead Temperature (Soldering, 60 sec) .............. 300D C

• For devices processed In total compliance to MIL-STD-883, add 1883 after part
number. Consult factory for 883 data sheet.
•• Epoxy packaged devices available in 1st Quarter, 1986.
t All commercial and industnal temperature range parts are available with burnMln.
For ordenng Information see 1986 Data Book, Section 2

GENERAL DESCRIPTION
The MAT -04 is a quad monolithic NPN transistor that offers
excellent parametric matching for precision amplifier and nonlinear circuit applications. Performance characteristics of the
MAT-04 include high gain (400 minimum) over a wide range
of collector current, low noise (2.5nV/vHz maximum at 100Hz,
Ic = 1mA) and excellent logarithmic conformance. The MAT-04
also features a low offset voltage of 150fLV and tight current gain
matching, to within 2%. Each transistor of the MAT-04 is
individually tested to data sheet specifications. For matching
parameters (offset voltage, input offset current, and gain match),
each of the dual transistor combinations are verified to meet
stated limits. Device performance is guaranteed at 25 C and over
the industrial and military temperature ranges.

NOTES:
Absolute maximum ratings apply to both DICE and packaged devices

The table below lists maximum ambient temperature ratings and derating
factors

D

PACKAGETYPE

The long-term stability of matching parameters is guaranteed by
the protection diodes across the base-emitter junction of each
transistor. These diodes prevent degradation of beta and matching characteristics due to reverse bias base-emitter current.

14-Pln
Hermetic DIP (Y)
14-Pln
Epoxy DIP (P)

The superior logarithmic conformance and accurate matching
characteristics of the MAT -04 makes it an excellent choice for use
in log and antilog circuits. The MAT-04 is an ideal choice in
applications where low noise and high gain are required.

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

100·C

10mW/·C

42°C

6mW/·C

This preliminary product Information Is based on testing of a limited number of devices. Final specifications may vary. Please contact local sales
office or distributor for final data sheet.
9-23

1/86, Rev. A

II

~

MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR - PRELIMINARY

ELECTRICAL CHARACTERISTICS at TA = 25°C unless otherwise noted. Each transistor is individually tested. For matching
parameters (Vos. los. Ilh FE ) each dual transistor combination is verified to meet stated limits. All tests made at endpoints unless otherwise noted.
MAT-04A/E
PARAMETER
Current Gain

Current Gain
Match

Offset Voltage

SYMBOL

CONDITIONS

MIN

TYP

hFe

10",Asle S 1mA
OVsVee s30V
(Note 1)

400

800

ahFe

Ie = 100",A
OVsVee s 30V
(Note 2)

Vos

aVos/ale

Offset Voltage
Changevs
Collector Current
Offset Voltage

Bulk Emitter Resistance

ree

Input Bias Current

Ie

Input Offset Current

los

Breakdown Voltage

300

",V

10",Asle s1mA
Vee=OV

5

25

10

50

",V

50

100

100

200

",V

0.4

0.6

0.4

0.6

!l

125

250

165

330

nA

0.6

5

13

nA

OVsVee s 30V
10",As Ie s 1mA
Vee=OV
le= 100",A
OVsVee s 30V
le= 100",A
Vee=OV

Collector-Base
Leakage Current

leBO

Vee = 40V

5

NOIse Voltage
DenSity

en

Vee=OV
le=1mA
(Note 3)

Gain Bandwidth
Product

fT

Output Capacitance

C OBO

Vee = 15V
f=1MHz

le=O

Input CapaCitance

C eeo

Vee=OV
f=1MHz

le=O

2
1.8
1.8

fo= 10Hz
fo= 100Hz
fo=1kHz

le=1mA
Vee = 10V

V

40

40
0.03

Sample tested.

%

100

Vee(SAT)

3

4

150

le= 10",A

10~,

800

UNITS

50

le= 100",A
le=1mA

Current gain match IS defined as:

300

MAX

10",Asle s1mA
OVS Vee s30V

BVeeo

2.

TYP

2

Collector Saturation
Voltage

NOTES:
Current gam measured at Ie =

MIN

0.5

1Of.loA:s;:lc:S;1mA

avos/avee

Change vs Vee

MAT-04B/F

MAX

0.06

0.03

0.06

5
3
2.5
2.5

2
1.8
1.8

V

pA
4
3
3

nV/$z

300

300

MHz

10

10

pF

40

40

pF

100j.LA and 1mA.
100 (alB) (hFe min)

ah Fe =

Ie

9-24

1/86, Rev. A

-----------I~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

ELECTRICAL CHARACTERISTICS

at

-25°C

:5 TA :5

85°C

PRELIMINARY

unless otherwise noted. Each transistor is individually tested. For

matching parameters (Vos. los) each dual transistor combination is verified to meet stated limits. All tests made at endpoints unless
otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

MIN

Current Gain

hFE

10.,As Ics 1mA
OVSVcB s 30V
(Note 1)

225

Offset Voltage

Vos

10.,As Ics 1mA
OVsVcB s30V

Average Offset
VoltageDnft

TCVos

VCB~OV

Input Bias Current

IB

Input Offset Current

los

MAT-04E
TYP
MAX
625

60

MIN
200

210

MAT-04F
TYP
MAX

UNITS

500

120

420

!'-V

0.4

2

!,-V/'C

445

200

500

nA

20

8

40

nA

Ic ~ tOO!,-A
0.2

(Note 2)

Average Offset
Current Drift

TClos

Ic ~ 100!,-A
oVsVcB s 30V
Ic~

160

100!,-A

VCB~OV

Ic ~ 100!,-A

Breakdown Voltage

BVCEO

Ic~

Collector-Base
Leakage Current

ICBO

VCB~40V

Collector-Emitter
Leakage Current

ICES

VCE~40V

Collector-Substrate
Leakage Current

Ics

Vcs~40V

ELECTRICAL CHARACTERISTICS

at

10!,-A

-55°C

:5 TA :5

100

50

VCB~OV

40

pA/'C

40
0.5

V
05

nA
nA

0.7

125°C

07

nA

unless otherwise noted. Each transistor is individually tested. For

matching parameters (VOS. los) each dual transistor combination is verified to meet stated limits. All tests made at endpoints unless
otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

Current Gain

hFE

10.,As Ic s 1mA
OV s VcB :s30V
(Note 1)

Offset Voltage

Vos

TCVos

Input Bias Current

IB

Input Offset Current

los

Average Offset
Current Drift

MAT-04B
MIN
TYP
MAX

175

lOjJ-As les 1mA

TClos

UNITS

475

70

125

250

425

140

500

!'-V

02

0.4

2

!,-V/'C

210

570

235

800

nA

6

30

12

60

nA

100!,-A

VCB~OV

Ic ~ 100!,-A
OV

ICBO

VCB~4OV

Collector-Emitter
Leakage Current

ICES

VCE~40V

Collector-Substrate
Leakage Current

Ics

VCS~

100

50

VCB~
Ic~

10!,-A

40

40

pA/'C
V

5

nA

100

100

nA

7

7

nA

40V

NOTES:
1. Current gain measured at Ic ~ 10!,-A, 100!,-A and 1mAo
2. Guaranteed by Vos test (TCVos ,,; Vos/T for Vos < < VBE ) T ~ 298'K for
TA ~ 25'C.

1/86, Rev. A

9-25

--------

::r::

U

100!,-A
OVsVcB s30V
Ic~

BVCEO

~E-<
~

VCB~OV

Breakdown Voltage

~
.....

Cl

100!,-A

Collector-Base
Leakage Current

~

0

(Note 2)

Ic~

V)

V)

OVsVcB s 30V
Ic~

Average Offset
Voltage Drift

MAT-04A
MIN
TYP
MAX

II

----

--------

~

----------1~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

PRELIMINARY

DICE CHARACTERISTICS
1.
2.
3.
4.
5.
6.

7.
8.
9.
10.
11.
12.
13.
14.
DIE SIZE 0.059 x 0.059 inch, 3481 sq. mils
(1.50 x 1.50 mm, 2.25 sq. mm)

Q1 COLLECTOR
Q1 BASE
Q1 EMITTER
SUBSTRATE
Q2EMITTER
Q2BASE
Q2 COLLECTOR
Q3 COLLECTOR
Q3BASE
Q3EMITTER
SUBSTRATE
Q4EMITTER
Q4BASE
Q4 COLLECTOR

For additional DICE information refer to
1986 Data Book, Section 2.

WAFER TEST LIMITS at TA = 25°C unless otherwise noted. Each transistor is individually tested. For matching parameters (Vas,
los, t.h FE ) each dual transistor combination is verified to meet stated limits. All tests made at endpoints unless otherwise noted.
MAT-04N
PARAMETER
Current Gain
Current Gain Match

Offset Voltage

SYMBOL
hFE
AhFE
Vos

Offset Voltage
Changevs

AVos/Alc

Collector Current
Offset Voltage
Change vs VCB
Bulk Emitter Resistance

AVoS/AVCB

Ic~ 100"A

OV <; VCB <; 30V
Ic ~ 100"A, VCB ~ OV

10,.,..A.::.;l c .:s::1mA
OV <; VCB <; 30V
10,.,..As les 1mA
VCB ~ OV
10"A <; Ic <; 1rnA
OV" VCB <; 30V

1O/-1A51cs 1mA
rBE

Collector Saturation
Voltage

CONDITIONS

VCE(SAT)

Input Bias Current

IB

Input Offset Current

los

Breakdown Voltage

BVCEO

VCB~OV

IB~ 100"A
Ic~lrnA

Ic ~ 100"A

OV=:::;Vcs ::s:30V
Ic~ 100"A
VCB~ OV

Ic~ 10"A

LIMITS

UNITS

300

MIN

4

%MAX

300

"V MAX

50

"V MAX

200

"V MAX

0.6

flMAX

0.06

VMAX

330

nAMAX

13

nAMAX

40

VMIN

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to vanatlons in assembly methods and normal Yield loss, Yield after packaging IS not guaranteed for
standard product dice. Consult factory to negotiate based on dice lot quahflcatlon through sample lot assembly and testing

9-26

1/86, Rev. A

----------l~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

PRELIMINARY

TYPICAL PERFORMANCE CHARACTERISTICS
CURRENT GAIN
vs COLLECTOR CURRENT
1200

2000

+125oclll

~

1100

GAIN BANDWIDTH vs
COLLECTOR CURRENT

CURRENT GAIN
vs TEMPERATURE
1800
VCB=3o/

11111

1000

1600

vc~ ~'ci~'

900
800

II

700

+d5.hl

!

1400

~

1200

a:

:::)

500

"

500

-5r1

400

10f!.

100jJ.

1m

/

800

/

600
400

10m

COLLECTOR CURRENT (A)

BASE-EMITTER-ON-VOLTAGE
vs COLLECTOR CURRENT
0.65

s
~

m

0.55

a:

~
!

.

~

-25

25

75

125

"iii
~I

lOlL

1001'-

1m

10m

COLLECTOR CURRENT (A)

SATURATION VOLTAGE vs
COLLECTOR CURRENT

100f,!.

1m

10m

100m

SMALL SIGNAL OUTPUT
CONDUCTANCE vs
COLLECTOR CURRENT

"

1M

,

.;!;

,

1lz

~

1'.
O~

,

"

100fl.

V

01

I

2

II
10f.l

v

0

"

lOOk

I

10

"

10k
1f.l

10f!.

SMALL SIGNAL INPUT
RESISTANCE (hie) VS
COLLECTOR CURRENT

1

0.35

I
1f.l

COLLECTOR CURRENT (A)

-'

040

01

175

TEMPERATURE (Oe)

2l

t/

0.45

V

Ie

-'

~

... V
....... V

100

i

/

0 SO

V

10M

~

/
en

le= 1rnA

Vcs=ov"""

~

0.60

I

, /V

200
-75

300

"

/

V"

"i i 1000
a:

I

1/

V

001
1m

,
11-'

10f.!.

100j.l

1m

COLLECTOR CURRENT (A)

COLLECTOR CURRENT (A)

NOISE VOLTAGE DENSITY
vs FREQUENCY

NOISE VOLTAGE DENSITY
vs COLLECTOR CURRENT

10

'-'0HZ

"- r--...

I

100Hz

.........
+125°C

r--

.-""

+ 25°C

-55·C..,J,.

1

I

10m
10,

100j.L

1m

COLLECTOR CURRENT (A)

10m

0.1 L-_L-_L._L._.J...._.J....---l
100m
10
100
1k
10k
lOOk
FREQUENCY (Hz)

9-27

o

o

3m

6m

9m

12m

COLLECTOR CURAENT (A)

1/86, Rev. A

----------1~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

PRELIMINARY

TYPICAL PERFORMANCE CHARACTERISTICS
COLLECTOR-TO-BASE
CAPACITANCE VB COLLECTORTO-BASE VOLTAGE

TOTAL NOISE VB
COLLECTOR CURRENT
200

160

r~1

As

20

! 1Jo~h

",,'

~
w

~

..

50

l'....1'-

~ 120

40

~~

o

'"

-

10f.(,

Rs

As

\

-

80

;!
~

COLLECTOR-TO-SUBSTRATE
CAPACITANCE VB COLLECTORTO-SUBSTRATE VOLTAGE

\

........

1'--0..

= 10kU

1,Lll
o

1m

100f.l

COLLECTOR CURRENT (AI

o

10

15

20

25

COLLECTOR-TO-BASE VOLTAGE (VOLTS)

APPLICATION NOTES

30

o
o

ro

~

~

~

COLLECTOA· TO·SUBSTAATE VOLTAGE (VOLTS)

FIGURE 2: Current Mirror, lOUT = 2 (IREF)

It is recommended that one of the substrate pins (Pins 4 and 11)
be tied to the most negative circuit potential to minimize coupling
between devices. Pins 4 and 11 are internally connected.

lOUT

Q1

APPLICATIONS
CURRENT SOURCES
The MAT-04 can be used to implement a variety of high impedance current mirrors as shown in Figures 1,2, and 3. These current mirrors can be used as biasing elements and load devices for
amplifier stages.

Q4

V-

FIGURE 1: Unity Gain Current Mirror, lOUT = IREF

lOUT

FIGURE 3: Current Mirror, lOUT = 112 (lREF)

02

lOUT

Q1

04
Q4

V-

V-

The unity-gain current mirror of Figure 1, using a MAT-04AY, has
an accuracy of better than 1% and an output impedance of over
100MO at 100f1A. Figures 2 and 3 show modified current mirrors
designed for a current gain of two, and one-half respectively. The
accuracy of these mirrors is reduced from that of the unity-gain
source due to base current errors but is still better than 2%.

Figure 4 is a temperature independent current sink that has an
accuracy of better than 1% over the military temperature range at
an output current of 100f1A to 1mA. The Schottky diode acts as a
clamp to insure correct circuit start - up at power on. The resistors
used in this circuit should be 1% metal-film type.

9-28

1/86, Rev. A

---------I~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

PRELIMINARY

FIGURE 4: Temperature Independent Current Sink, lOUT = 10V/Rn
+15V

lOUT

lOUT

lOUT

-----,

100pF

I

14

MAT.04AY

12

I
I
I

_ _ _ _ _ --.J

R

R

-15V

NONLINEAR FUNCTIONS
An application where precision matched-transistors are a powerful tool is in the generation of nonlinear functions. These circuits
are based on the transistor's logarithmic property which takes the
following idealized form:

This circuit uses two MAT-04AYs and maintains an accuracy of
better than 0.5% over an input range of 10mV to 10V. The layout
of the MAT-04s reduces errors due to matching and temperature
differences between the two precision quad matched-transistors.
Op amps A 1 and A2 translate the input voltages into logarithmic
valued currents (IA and IB in Figure 5) that flow through transistor
0 3 and Os. These currents are summed by transistor 0 4
(10 = IA + IB = ,jI,2 + 122) which feeds the current-to-voltage

Ie
kT
VBE= - I n q
Is

ro,~rt" ron"";"

The MAT-04, with its excellent logarithmic conformance, maintains this idealized function over many decades of collector current. This, in addition to the stringent parametric matching of the
MAT-04, enables the implementation of extremely accurate logl
antilog circuits.

of ""

~p A3. To m.""," -"'Y, 1%

metal-film resistors should be used.

~

0

f-<

;:a
~

~

@
:r:
u

The circuit of Figure 5 is a vector summer that adds and subtracts
logged inputs to generate the following transfer function:
VOUT =

II
~

~ ,jVA2 + VB2

9-29

1/86, Rev. A

---------tlfMD

MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR - PRELIMINARY

FIGURE 5: Vector Summer
C3

.OOpF
A
233kH

~~~-------------------oyOOT

I-----;;A:;:-~

L _____ J

L __ ~ ___ ,
...---+1-ie•

/

I

8

I

C2

L ________ .....l
~

MAT-04

10

I
I
I
~--+--+I~~

A,No---J\I\_---<~=-I
33kO

L __________

LOW NOISE, HIGH SPEED INSTRUMENTATION AMPLIFIER
The circuit of Figure 6 is a very low noise, high speed amplifier,
ideal for use in precision transducer and professional audio applications. The performance of the amplifier is summarized in
Table I. Figure 7 shows the input referred spot noise over the
0-25kHz bandwidth to be flat at 1.2nVI JHz: Figure 8 highlights
the low llf noise corner at 2Hz.

(

R10R11 )
R10 + R11

R3

F---.JV\/'v--OB'N
33kn

~

TABLE I: Instrumentation Amplifier Characteristics

The circuit uses a high speed opamp, the OP-17, preceded by an
input amplifier. This consists of a precision dual matched-transistor, the MAT-02, and a feedback V-to-I converter, the MAT-04.
The arrangement of the MAT-04 is known as a "linearized cross
quad" which performs the voltage-to-current conversion. The
OP-17 acts as an overall nulling amplifier to complete the feedback loop. Resistors Rl, R2, and R3, R4 form voltage dividers
that attenuate the output voltage swing since the "cross quad" arrangement has a limited input range. Biasing for the input stage is
set by zener diode Zl. At low currents the effective zener voltage
is about 3.3V due to the soft knee characteristic of the zener
diode. This results in a bias current of 5301lA per side forthe input
stage. The gain equation for this amplifier is:

VOUT

~

r

I
I
I
~~w~+I--~--~

v+

R.

I

.oooPFI

L _________________ ________

----------,

--;-,?

Q6'3"<9H

r---~--~

I

l'OOOpF

/r--l

v-

L~~---l

Input Noise
Voltage Density

G= 1000
G=100
G=10

Bandwidth

G=500
G=100
G=10

Slew Rate

1.2nV/v'HZ
3.6nV/v'HZ
30nVlv'HZ
400kHz
1MHz
1.2MHz

4OV/".s

Common-Mode
Rejection

G= 1000

130dB

Distortion

G=100
f = 20Hz to 20kHz

0.03%

Settling Time

G=1000

Power Consumption

10".s
350mW

(R4+R3)
R3

----=-=------=.-=---=------'~

Y,N

RG

and for the values shown in Figure 6 this equation becomes:
VOUT = 33000

Y,N

RG

9-30

1/86, Rev_ A

---------I~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

PRELIMINARY

FIGURE 6: Low Noise, High Speed Instrumentation Amplifier
r---------------~----------------------t_---------------------------o+1SV

R.

RS
68kn

6.8kn
C1

R7

500pF

1.Sk!l

:>"--.....---oVOUT

-15V

+IN

R4

10kH

Z1
3.9V
IN748A

RG = 33k!l
RG = 33kn
RG = 330n
RG = 33n

GAIN
GAIN
GAIN
GAIN

=
=
=
=

1
10
100
1000

R10

R.
39kn

~----~~----------

____

~

__________________________________________________-o_1SV

FIGURE 7: Spot Noise of the Instrumentation Amplifier from
0-25kHz at a Gain of 1000

FIGURE 8: Low Frequency Noise Spectrum Showing Low 2Hz
Noise Corner. Gain = 1000.

NORMALIZED VERTICAL AXIS = 2.6nV/-vHZ/DIVISION
REFERENCED TO INPUT.
HORIZONTAL AXIS == 0 TO 25kHz.

HORIZONTAL AXIS = 0 TO 5Hz

9-31

1/86, Rev. A

II

---------I~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

PRELIMINARY

FIGURE 9: Voltage-Controlled Attenuator
r-----~------------------------~----~----------------------ov+

R3

R4
3DkO

aDkO

R7
30k!!

R6
30kn

I-

R8
30kn

v+
VOUT

MAT·04

I
I

C2

~ 100,F

R13

R14

4.7kfl

330n

V CONTROL

v+

Rl0
22kO

Rll
10kO

2N2222
Q5

R2

R5

30kU

3Dkn

~----------------------------+-------------------~----o~

VOLTAGE-CONTROLLED ATTENUATOR
The voltage-controlled attenuator (VCA) of Figure 9, widely used
in professional audio circles, can easily be implemented using a
MAT-04. The excellent matching characteristics of the MAT-04
enables the VCA to have a distortion level of under 0.03% over a
wide range of control voltages. The VCA accepts a 3V RMS input
and easily handles the full 20Hz-20kHz audio bandwidth as
shown in Figure 10. Noise level for the VCA is more than 110dB
below maximum output.

The ideal transfer function for the voltage-controlled attenuator is:
VourlVlN =

1 + exp

Where k
T

Boltzmann constant 1.38 x 10-23 J/oK
temperature in OK
electronic charge = 1.602 x 10-19C

q

(

(-V control)

(2

R14
) / ( kT ) )
R13 + R14
q

From the transfer function it can be seen that the maximum gain
of the circuit is 2 (6dB).

In the voltage-controlled attenuator, the input signal modulates
the stage current of each differential pair. Op amps A2 and A3 in
conjunction with transistors 05 and 06 form voltage-to-current
converters that transform a single input voltage into differential
currents which form the stage currents of each differential pair.
The control voltage shifts the current between each side of the
two differential pairs, regulating the signal level reaching the output stage which consists of op amp A 1. Figure 11 shows the increase in signal attenuation as the control voltage becomes more
negative.

To insure best performance, resistors R2 through R7 should be
1% metal film resistors. Since capacitor C2 can see small
amounts of reverse bias when the control voltage is positive, it
may be prudent to use a nonpolarized tantalum capacitor.

9-32

1/86, Rev. A

----------1~ MAT-04 MATCHED MONOLITHIC QUAD TRANSISTOR -

PRELIMINARY

FIGURE 11: Voltage-Controlled Attenuator,
Attenuation vs Control Voltage

FIGURE 10: Voltage-Controlled Attenuator,
Attenuation vs Frequency

10

)CU~l~L Joll

/'

-10

!

-5

,IV

~ -20

/

iii

-10

i
V CONTROL

-15

~

= -1V

-20

10

-30

"IV
-40
-50

-25
100

,.

---

/

-60
10.

-3

100k

FREQUENCY (Hz)

-2

-1

CONTROL VOLTAGE (VOLTS)

II

9-33

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9

Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and -Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

VOLTAGE
REFERENCES
PrecisioJl MOllolithics Inc.

10-3

Introduction

10-3

Definitions

10-4

REF-Ol
+ lOV Precision Voltage Reference

10-11

REF-02
+ 5V Precision Voltage
Reference/Temperature Transducer

10-19

REF-03
+2.5V Precision Voltage Reference

10-21

REF-OS
+ 5V Precision Voltage Reference

10-22 REF-lO
+ lOV Precision Voltage Reference

10-2

VOLTAGE
REFERENCES
Precision MOlloilthlcs Inc

INTRODUCTION
Voltage references provide a constant output
voltage irrespective of changes in input voltage,
output current, or temperature. References are
needed in such diverse equipment as power
supplies, panel meters, calibration standards,
precision current sources, data conversion systems, and control set-point circuits.
Line regulation, load regulation (output impedance), and temperature coefficient specifications indicate how close a reference will be to an
ideal voltage source. Line regulation specifies
reference-output-voltage vs. input-voltage
changes. Output voltage changes due to load
current variations are reflected by load regulation specifications. Temperature coefficient
specifications indicate output voltage variation
over temperature.
PMI references use the bandgap principle which
sums voltages with negative and positive
temperature coefficients to yield a stable output
voltage over temperature. A transistor baseemitter-junction voltage (VBE) exhibits a negative temperature coefficient. Two transistors
operating with unequal current densities will
have different VBES and the difference, ~VBE,
exhibits a positive temperature coefficient.
When ~VBE is amplified and added to VBE, a
near-zero temperature coefficient results if the
sum equals 1.23V. The 1.23V level is then amplified to provide stable output voltages of +S.OOV
or + 10.00V. The bandgap technique has the
advantages of low power consumption, low noise,
and excellent long-term stability.
PMl's zener-zapping technique allows for trimming of the ~VBE amplification factor to ensure
low output voltage temperature coefficients.
Additional zapping trims the output's absolute
value to within specified limits.
The REF-01 and REF-02 are stable + 10.00V and
+S.OOV monolithic bandgap voltage references.
Output voltages are adjustable with small effect
on output-voltage temperature coefficients. The
REF-02 provides an additional output voltage
that has a linear temperature dependence.
The REF-OS and REF-10 are premium versions
of the REF-01 and REF-02 that have guaranteed
long-term stability and MIL-STO-883 process-

ing. Extensive testing over a long period of time,
combined with tight control of processing, has
enabled PMI to specify limits on output change
with time.

DEFINITIONS
Line Regulation - The ratio of the change in
output voltage to the change in input (line) voltage producing it. It includes the effects of
self-heating.
Load Regulation - The ratio of the change in
output voltage to the change in load current. It
includes the effects of self-heating.
Output Change With Temperature (~VOT) - The
absolute difference between the maximum output voltage and the minimum output voltage
over the specified temperature range expressed
as a percentage of the typical output voltage.

~VOT = IVMAX- VMINlx 100
Vo (Typical)

Output Temperature Coefficient (TCVo) - The
ratio of output change with temperature variation to the specified temperature range
expressed in ppm/o C. For example, TCVo is
defined as ~VOT divided by the temperature
range; i.e.,
TCV (O°C to +700C) = ~VOT(O°C to +70°C)
o
700C
and TCVo(-SSOC to +12S0C)
~VOT(-SSOC to +12S0C)
180°C

Output Turn-On Settling Time (tON) - The time
required for the output voltage to reach its final
value within a specified error band after application of VIN.
Output Voltage Noise (enp_p) - The peak-topeak output noise voltage within a specified frequency band.
Quiescent Supply Current (ISY) - The current
required from the supply to operate the device
with no load.
10-3

- - -- .. --- - - - -

=

I
~

S:2

A
-

S2

I

REF-Ol

+lOV PRECISION
VOLTAGE REFERENCE
Prcci sion MOllO} ithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

drain of 1mA, and excellenttemperature stability are achieved
with an improved bandgap design. Low cost, low noise, and
low power make the REF-01 an excellent choice whenever a
stable voltage reference Is required. Applications include
DIA and AID converters, portable instrumentation, and digital voltmeters. Full military temperature range devices with
screening to MIL-STD-883 are available. For guaranteed
long-term drift see the REF-10 data sheet.

10 Volt Output ........................... ±0.3% Max
Adjustment Range .......................... ±3% Min
Excellent Temperature Stability ....... 8.5ppml" C Max
Low Noise ......................... . . . .. 30~Vp_p Max
Low Supply Current ....................... 1.4mA Max
Wide Input Voltage Range ................. 13V to 33V
High Load-Driving Capability ................... 20m A
No External Components
Short-Circuit Proof
MIL-STD-883 Screening Available

PIN CONNECTIONS

NC•
NCO'N'C

ORDERING INFORMATIONt
PACKAGE
TA =25'C
avo MAX
(mV)

TO-SS
'-PIN

± 30
± 30
± 50
± 50
±100

REF01AJ'
REF01EJ
REF01J'
REF01HJ
REF01CJ

HERMETIC PLASTIC
DIP
DIP
'-PIN
'-PIN
REFOIAZ'
REFOIEZ
REF01Z'
REF01HZ
REF01CZ

LCC

REFOI RC/883
REF01HP
REF01CP

OPERATING
TEMPERATURE
RANGE

VIN 2

6 Your

NC3

MIL
COM
MIL
COM
COM

5 TRIM

4
GROUND

ICASE)

TO-99

REF-01RC/883
LCC
(RC-Sufflx)

(J-Sufflx)

• For devices processed in total compliance to M I L-STD-883. add /883 after
part number. Consult factory for 883 data sheet.
tAli commerCial and industrial temperature range parts are available with
burn-In. For ordering Information see 1986 Data Book, Section 2

EPOXY MINI-DIP
(P-Sufflx)

GENERAL DESCRIPTION

8-PIN HERMETIC DIP
(Z-Sufflx)

The REF-01 precision voltage reference provides a stable
can be adjusted over a ±3% range with
minimal effect on temperature stability. Single-supply operation over an input voltage range of 12V to 40V. low current

+10V output which

SIMPLIFIED SCHEMATIC
,----~--.,------1r_-------_----

;:: 0015

100

z

II1IIIIIIIIII

~

II
v/

~ 0.005
~

75"C

7

0.D10

~

'O'~O~~'~OO~UW~'k-W~~'O-k~lli'~O~OkLlllliW'M

DEVICE IMMERSED
IN 75"C OIL BATH

o
-10

FREQUENCY (Hz)

FREQUENCY (Hz)

TA = TA =

- - ,.1/

0.020

o

100
1M

tOOk

VIN = 15V

0025

~

~

z

::;

1111111

16

:l::>
:.la:
w

I

!5

w

0

031

~

-"

z

[

§?

r

I

~

!~ l000~11~11'1111'1

0031 )

0,1

!I

~ 0.030

o

I

1

I

36

_ 0035
~

~ ~N"~:'i.V

001 ~

56

46

10,000

OUTPUT CHANGE DUE TO
THERMAL SHOCK

10

30

20

40

50

60

TIME (SEC)

3.

NORMALIZED
LINE REGULATION
vs TEMPERATURE

NORMALIZED LOAD
REGULATION (diL = 10mA)
vs TEMPERATURE

MAXIMUM LOAD CURRENT
vs INPUT VOLTAGE
14

1.4

13

1.3

SHORT CIRCUIT PROTECTION

30

/' .00mwMIAX~
I
/

/

07

10

,.

~ 11

V

~

V
VIN

=

15V

30

-60 -40 -20

0.7

0

20

40

60

0,6
-60 -40 -20

80 100 120 140

0

'"

ffi

1'-

....... 1---"

a

10

~

09

~

i

1i

V

0

20

40

60

60

80

100 120 140

v

V

"

08

VI~

o
-60--40 -20

V

11

~

i'-r-.

,. 10

40

QUIESCENT CURRENT
vs TEMPERATURE

11.2

VIN = 15V

20

TEMPERATURE (Oel

13

1'- r-...

V

w 0.8

f'--...

"~

::>

/"

TEMPERATURE (oG)

a:
~ 20

,.9 15

.... P'

V

z

30

a;

O.g

7

::;

MAXIMUM LOAD CURRENT
vs TEMPERATURE

1
>- 2.

E
illa:

""

INPUT VOLTAGE (VOLTS)

/'

w
~ 1,0

06

2.

20

t!

/

a:
~ 08

g

TA '" 25"C

:/

g 1.2

/

"~ 1.1
"'" 10
~
~
ffi 0.9

DISSIPATION

I

/

u

~ 12

07
-60 -40 -20

80 100 120 140

TEMPERATURE (OC)

0

20

40

60

"

'~V

80 100 120 140

TEMPERATURE (OCI

10-8

1/86, Rev. A

------------I~ REF-01 +10V PRECISION VOLTAGE REFERENCE

TYPICAL APPLICATIONS

AID CONVERTER REFERENCE

D/A CONVERTER REFERENCE

ANALOG

+15V

INPUT

o TO +10V

+15V

10kO
+lSV

MSB

-=

LSB

-1SV

-lSV

+1SV

B'
B5
BS
B7
B8

82 83 84 85 86 87 88

81

POS. FULL-SCALE -1 LSB

SERIAL
OUTPUT

E

NEG. FULL-SCALE + 1 LSB

0

NEG. FULL-SCALE

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

CONNECT "START" TO
"CONVERSION COMPLETE"
FOR CONTINUOUS CONVERSIONS

START

+4.960
0

ZERO-SCALE

-15V

0.000
CONVERSION
COMPLETE

-4.960
0

TTL CLOCK
INPUT 225MHz

-5000

±10V REFERENCE
PRECISION CALIBRATION STANDARD

+15V

+10V

II

m
~l'A

1
{
-

REF-Ol

10kil

12
V'N

10kn

GND

+

S
Va

+15V

9V

REF-Ol

1

rv

10000V

5
TRIM
GND

lOOka

t

-lOV

-15V

CURRENT SOURCE

!2+

CURRENT SINK

VOLTAGE COMPLIANCE

1SV

-25V TO +3V

~~OUT

V'N

V'N

VO!---

Va

REF-al
TRIM

~

VOLTAGE COMPLIANCE

-3V TO +25V

~

REF·O'

R

lOUT '" 10 ~v +lmA

TRIM

GND

~

R

lOUT = 10~v +lmA

GND

j'

l'

+'OUT

6-15V

10-9

1/86, Rev. A

------------I~ REF-01 +10V PRECISION VOLTAGE REFERENCE

REFERENCE STACK WITH EXCELLENT LINE
REGULATION

PRECISION CURRENT SOURCE
A current source with 25V output compliance and excellent
output impedance can be obtained using this circuit. REF-01
®keeps the line voltage and power dissipation constant in
the only important error consideration at room
device
temperature is the negative supply rejection of the op amp.
The typical 3JJ,V/v PSRR of the OP-02E will create an 8ppm
change (31JV/v x 25V/10V) in output current over a 25V range.
For example, a 10mA current source can be built (R = 1kO)
with 300MO output impedance.

Three REF-01's can be stacked to yield 10.000, 20.000, and
30.000V outputs. An additional advantage is near-perfect line
regulation of the 10.0V and 20.0Voutput. A 32V to 60V input
change produces an output change which is less than the
noise voltage of the devices. A load bypass resistor (RB)
provides a path for the supply current (Isy) of the 20.000V
regulator.
In general, any number of REF-01's can be stacked this way.
For example, ten devices will yield outputs of 10, 20, 30 ...
100V. The line voltage can range from 105Vto 130V. However,
care must be taken to ensure that the total load currents do
not exceed the maximum usable current (typically 21mA).

CD;

25V
Ro=----8 X 1Q-6 x 10mA

t

12

50V

V ,N

32V to 60V

V ,N

6 Va

Va

TRIMM ED
QUTPU TS
30.00OV

6

REF-01

REF-O'

®

TRIM
GND

GND

4

2
V ,N
Va

10kn

4
2
V ,N

6

Va

REF-01

: =c

CD

6

20.0 OOV

R
(TRIM FOR

REF-01

CALIBRATION)
-----<

GND

4

6

R

TRIM 5
GND

2
V ,N

~:J

Va

4

10.00OV

REF-01

~~
6

6

10kn

QP-02E

+

3

TRIM
GND

Vo 
780
.§

~

15V

/

~ 730

V

~

6 680
~

/

630

/

g 580
~

~

530

450

V

/

430
-60 -40 -20

0

20

40

60

80

to +12.SV,

REFERENCE STACK WITH EXCELLENT LINE
REGULATION
Two AEF-01's and one AEF·02 can be stacked to yield 5.000V,
15.000V and 25.000Voutputs. An additional advantage ofthis
circuit is near·perfect line regulation of the 5.0V and 15.0V
outputs. A 27V to 55V input change produces an output
change which is less than the noise voltage of the devices. A
load bypass resistor (AB) provides a path for the supply
current (Isy) of the 15.000V regulator.
In general, any number of AEF·01's and AEF-02's can be
stacked this way. For example, ten devices will yield ten
outputs in 5Vor 10V steps. The line voltage can range from 100V
to 130V. However, care must be taken to ensure that the total
load currents do not exceed the maximum usable current
(typically 21mA).

sso
830

SWing

100 120 140

TEMPERATURE 1°C)

TEMPERATURE CONTROLLER
. l227V TO 55V

v+

V ,N

(12 TO 32V)

Va

,------t---------_--,

c---- ---

I
I

I
I

~27kn
TRIM
GND

;.

c,. R6

I

V~
HEATING
ELEMENT

5

3 \

10kn

4

2

Vo 1-'-c1-'VI."v-~t-1

REF·02
GND

.1

RI
(92kn)

250 OOV

REF-01

.1,. R7

r-----------,

•

V ,N

1 ~~n

Va

6

1500OV

T I"-;-""'V--+-=-i
REF-01

IL

I

NOTE
4 _ .J
_ _ _ _(SEE
__
_II
__

R2

15kn

R5

TRIM 5
GND

2

22kn

V,N
Va

R4
27k.\1

6

4

10kn

500OV

REF-02
NOTES

1 REF-02 SHOULD BE THERMALLY CONNECTED
TO SUBSTANCE BEING HEATED
2 NUMBERS IN PARENTHESES ARE FOR A
SETPOINT TEMPERATURE OF 60°C
3 R3= Rl11R211R6

TRIM
OND

6

10kn

RB
68kn

4

10·17

1/86, Rev. A

II

-------l1fMD

REF-02 +5V PRECISION VOLTAGE REFERENCE/TEMPERATURE TRANSDUCER

PRECISION CURRENT SOURCE
A current source with 3SV output compliance and excellent
output impedance can be obtained using this circuit. REF-02
®keeps the line voltage and power dissipation constant in
device G) ; the only important error consideration at room
temperature is the negative supply rejection of the op amp.
The typical 3p.VlV PSRR of the OP-02E will create a 20ppm
change (3p.VIV x 3SVlSV) in output current over a 3SV range.
For example, a SmA current source can be built (R = 1kO)
with 3S0Mll output impedance.

CURRENT SINK

12

10UT

V,N

Vo ~

REF-02

3

TRIM

~

R

lOUT '" 6.~V +lmA

GND

l'
J-15V
VOLTAGE COMPLIANCE. -8V TO 25V

+SOV

3SV
Ro = 20 X 10- 6 X SmA

.-------':..rVo

D/A CONVERTER REFERENCE

REF-02

5,DkS1

GNO

MSB

+'6V

LSB

V,N
VO~----~----~~--'
REF-02

R
(TRIM FOR
CALIBRATION)

GND

±3V REFERENCE

Vo=O TO 35V

+7.6V

RC -'.-&sEC

VIN

vour"'6'--..------r-----<> Vo(+}
REF-02HJ

=

+3.OV

R1
2Ok!!

OND

CURRENT SOURCE

R2
13.3kO

12+1SV
V,N

Vo ~

REF-02

3

TRIM

!......o

R

lOUT. 6.~ +lmA

GND

>---1~--o

VoH • -3 OV

l'

t

lOUT

VOLTAGE COMPLIANCE' -25V TO +8V

10-18

1/86, Rev. A

REF-03
+2.SV PRECISION
VOLTAGE REFERENCE
Precision Monolithics Inc.

PRELIMINARY
FEATURES

GENERAL DESCRIPTION

•
•
•
•
•
•
•
•
•
•

The REF-03 preCision voltage reference provides a stable
+2.5V output which can be adjusted over a ±3% range with
minimal effect on temperature stability. Single-supply operation over an input voltage range of 4.5V to 33V, low current
drain of 1mA, and excellent temperature stability are achieved
with an improved bandgap design. Low cost, low noise, and
low power make the REF-03 an excellent choice whenever a
stable voltage reference is required. Applications include DIA
and AID converters, portable instrumentation, and digital
voltmeters. Full military temperature range devices with
screening to MIL-STD-883 are available.

2.5 Volt Output ............................... ±O.3%
Adjustment Range .............................. ±3%
Excellent Temperature Stability ............. 3ppm/oC
Low Noise .................................... 5/LVp_p
Low Supply Current ....................... 1.4mA Max
Excellent In Single +5V Systems
Wide Input Voltage Range ................ 4.5V to 33V
High Load-Driving Capability ..•................ 20m A
No External Components
Short-Circuit Proof

ORDERING INFORMATIONt
PIN CONNECTIONS

PACKAGE
TO-99
8-PIN

HERMETIC
DIP
8-PIN

REF03AJ'
REF03EJ
REF03BJREF03FJ

REF03AZREF03EZ
REF03BZREF03FZ

PLASTIC
DIP
8-PIN
REF03EP
REF03FP

N
N.CO·7N.C

OPERATING
TEMPERATURE
RANGE

C.

8

MIL
INO
MIL
INO

VIN 2

6 VOUT

N.C, 3

• For devices processed in total compliance to MIL-STO-883. add /883 after
part number. Consult factory for 883 data sheet.
t All commercial and industrial temperature range parts are available with
burn-in For ordering Information see 1986 Data Book, Section 2

5 TRIM

a-PIN EPOXY MINI-DIP
(P-Suffix)

4
GROUND
(CASE)

a-PIN HERMETIC DIP
(Z-Sufflx)

TO-99
(J-Sufflx)

SIMPLIFIED SCHEMATIC
.......r----...,...--------~---o

r-----~-

INPUT

2

R15

019

OUTPUT

6
A12'" 2kil
R9"'" 6kil

R11"", 2kn

R2
4

k---------~-------------~----~--~GROUNO

This preliminary product Information Is based on testing of a limited numberof devices. Final specifications may vary. Please contact local sales
office or distributor for final data sheet.

10-19

1/86, Rev. A

II

----------I~ REF-03 +2.SV PRECISION VOLTAGE REFERENCE -

ABSOLUTE MAXIMUM RATINGS

(Note 2)

DICE Junction Temperature (Tj) ••...••

••••.•.•••••..•...•....•.•....•........•.

40V

Power Dissipation (Note 1) ........•.....•.....•• 500mW
Output Short-Circuit Duration

NOTES:
1 See table for maximum ambient temperature rating and derating factor

(to Ground or VIN) •••.••••••••••.•••••.••••• Indefinite

PACKAGE TYPE

Storage Temperature Range
J, and

Z Packages .••...•••..•••..•• -65°C to +150°C

P Package

-65°C to +150°C

Lead Temperature (Soldering, 60 sec.) •••...••••.• 300°C

Input Voltage
REF-03

PRELIMINARY

........••....••.••••.•••

Operating Temperature Range

ELECTRICAL CHARACTERISTICS
SYMBOL

CONDITIONS

Output Voltage

Vo

IL

Output Adlustment Range

.6.Vtnm

Re

Output Voltage NOise

8

ne -e

71mW;oC
67mW;oC
56mW/"C

2 Absolute maximum ratings apply to both packaged parts and DICE, unless
otherwise noted.

~

MIN

0

~

10kO

REF-03A/E
MAX
TYP

2493

2500

±30

±33

0.1Hz to 10Hz (Note 5)
V,N

Load Regulation (Note 4)

80"C
75"C
36"C

at V IN = +15V, TA = 25°C, unless otherwise noted.

PARAMETER

Line Regulation (Note 4)

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

TO-99 (JI
8-PIn Hermetic DIP (ZI
8-PIn Plastic DIP (P)

-65°C to +125°C

REF-03A,REF-03B ................. -55°Cto+125°C
REF-03E, REF-03F ....... .. • .. . . • .. •• -25° C to +85° C

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

IL

~

2508

MIN

REF-038/F
TYP
MAX

2485

2500

±30

±3.3

UNITS

2.515

V

%

7.5

7.5

MV

e-e

4 5V to 33V

0006

0010

0.006

0.010

%/V

10m A

0005

0008

0006

0.010

%/mA

10

14

10

1.4

mA

~Oto

Turn-on Settling Time

ton

To ±0.1% of Final Value

QUiescent Supply Current

ISY

No Load

MS

Load Current

IL

10

21

10

21

mA

Sink Current

Is

-03

-05

-03

-0.5

mA

30

mA

Short-Circuit Current

Ise

ELECTRICAL CHARACTERISTICS
PARAMETER

SYMBOL

Temperature (Notes 1. 2)

= + 15V, -55° C :s:: TA :s:: + 125° C and IL = OmA, unless otherwise noted.
MIN

CONDITIONS

(Note 31

TCVo

Temperature Coefficient

at V IN

-25"Co> TA 0> +85"C
-55°C::;TA::;+125°C

Ouput Voltage Change With

Output Voltage

30

Vo~O

REF-03A1E
TYP
MAX

MIN

REF-038/F
TYP
MAX

UNITS

003
006

009
015

011
018

028
045

%

30

85

100

25.0

ppm/"C

Change in Vo Temperature

Coefficient With Output

07

07

ppm/%

Adjustment
Line Regulation
(VIN ~ 4'5V to 33V) (Note 4)

Load Regulation
(lL ~ a to 8mA) (Note 4)

-25° C:::; TA ::; +85° C
-55"C 0> TA 0> +125"C

0007
0009

0012
0015

0007
0009

0012
0.015

%/V

-25°C:::; TA ::; +85°C
-55"Co>TA o>+125"C

0006
0007

0010
0012

0.007
0009

0012
0015

%/mA

NOTES:
1

6.VOT IS defined as the absolute difference between the maximum
output voltage and the minimum output voltage over the specified
temperature range expressed as a percentage of 2 5V

AV

OT

3.

~IVMAX-VMINI
25V

TCVo (-25" to +85"CI

x 100

~

and TCVo (-55" to + 125" C)

..lVOT specification applies trimmed to +2 500V or untnmmed
TCVo IS defined as ~VOT diVided by the temperature range, Ie,

,WOT (-25" to +85"C)

110"C
~

-WOT (-55" to +125"CI

180"C

Line and Load Regulation specificatIOns Include the effect of self-heating.

5. Sample tested.

10-20

1/86, Rev. A

REF-OS
+5V PRECISION VOLTAGE REFERENCE
(GUARANTEED LONG-TERM STABILITY)
Precision Monolithics Inc.

FEATURES
• 5 Volt Output
• Guaranteed Long-Term Stability
. . . . . . . . . . . . . .. 100ppm/1000 Hrs Max
• Excellent Temperature Stability ....... 8.5ppm/ o C Max
• Low Noise. ... ..... ... .... ... ..... ... ... 15,.Np _p Max
• Low Supply Current.... .. ..... ..... .. .. .. 1.4mA Max
• Wide Input Voltage Range .................. 8V to 33V
• High Load-Driving Capability ................... 20mA
• Short-Circuit Proof
• Processed Per MIL-STD-883

effect on temperature stability. Long-term drift is guaranteed
at 100ppm/l000 hrs. maximum. Single-supply operation over
an input voltage range of 7V to 40V, low current drain of 1 mA,
and excellent temperature stability are achieved with an
improved bandgap design. Low cost, low noise, and low power
make the REF-05 an excellent choice whenever a stable
voltage reference is required. Applications include D/A and
AID converters, portable instrumentation, and digital voltmeters. The versatility of the REF-05 is enhanced by its use as
a monolithic temperature transducer. For +10V Precision
Voltage References see the REF-l0 data sheet.

GENERAL DESCRIPTION

ELECTRICAL CHARACTERISTICS/
ABSOLUTE MAXIMUM RATINGS
Contact factory for 883 data sheet with full electrical specifications. Typical performance characteristics and applications information are provided in the REF-02 data sheet.

The REF-05 precision voltage reference provides a stable +5V
output which can be adjusted over a ±6% range with minimal

LONG-TERM DRIFT PLOT (Average of 20 Devices)
PIN CONNECTIONS & ORDERING INFORMATION
0

NC
NC()'NC

0
0

8

0

,1\
0

"- tv' 1'...

REF-05

...... t-.....

o

VIN 2

t-

TEMP 3

-10 0

TO-99 (J-Sufflx)
REF-05AJ/883
REF-05BJ/883

6 VOUT

5 TRIM

4
GROUND
CASE

0
0

om_

•

•

_

m

_

~

_

DAYS ELAPSED

SIMPLIFIED SCHEMATIC

,-_____

-....,.---_~-------_---<>

~NPUT

R15

019

OUTPUT
6

R12 =6 lkH
R9 "'" 18kH

TRIM

5

3

TEMP

0-+----

Rll "'" 2kH

R2

L-_ _ _ _ _ _ _~~--_ _ _ _ _ _ _ _ _ _~---~~-_<~ROUND

10-21

1/86, Rev_ A

II

REF-tO

+lOV PRECISION VOLTAGE REFERENCE
(GUARANTEED LONG-TERM STABILITY)
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

•
•

The REF-10 precision voltage reference provides a stable
+10V output that can be adjusted over a ±3% range with
minimal effect on temperature stability. Long-term drift is
guaranteed at 50ppm/1000 hrs. maximum. Single-supply
operation over an input voltage range of 13V to 40V, low
current drain of 1mA, and excellent temperature stability are
achieved with an improved bandgap design. Low cost, low
noise, and low power make the REF-10 an excellent choice
whenever a stable voltage reference is required. Applications
include D/A and AID converters, portable instrumentation,
and digital voltmeters. For+5V precision voltage references,
see the REF-05 data sheet.

•
•
•
•
•
•
•

10 Volt Output
Guaranteed Long-Term Stability
...................... " SOppm/1000 Hrs Max
Excellent Temperature Stability ....... 8.Sppm/o C Max
Low Noise. . . . . . . . . . . . . . . . . . . . . . . . . . . ... 30,",Vp _p Max
Low Supply Current . . . . . . . . . . . . . . . . . . . . .. 1.4mA Max
Wide Input Voltage Range ................. 13V to 40V
High Load-Driving Capability ................... 20mA
Short-Circuit Proof
Processed Per MIL-STD-883

LONG-TERM DRIFT PLOT (Average of 20 DevIces)

ELECTRICAL CHARACTERISTICSI
ABSOLUTE MAXIMUM RATINGS
40

Contact factory for 883 data sheet with full electrical specifications. Typical performance characteristics and applications information are provided in the REF-01 data sheet.

20

-20

t;:

"~

~

....

-40

~

-......REF-l0

Q

-60
-80

PIN CONNECTIONS & ORDERING INFORMATION

NC
NC07NC

- r- -

8

-100

VIN 2

-120
-140

o

20

40

60

80

100

120

140

160

NC 3

180

TO-99 (J-Sufflx)
REF-10AJ/883
REF-10BJ/883

6 VOUT

5 TRIM

4

DAYS ELAPSED

GROUND
(CASE)

SIMPLIFIED SCHEMATIC

,--____-+_--.-____- - - - - - - - . _ - - 0 INPUT
2

R15

a19

OUTPUT
6
R3

A12 "" 16.7kn
R9 ""5OkO

R5

TRIM

5

a1Jr~~~~--_4~~-----~~-----4----_+
R1

A11 .... 2kn
R2

~_ _ _ _ _ _ _~~------------~---~~-_a~ROUND

10-22

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
)

Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

D/A
CONVERTERS
Precision MOJlolithlcs Inc.

11-4

Introduction

11-4

Definitions

11-8

Selection Guide

11-10

DAC-Ol

11-84

11-88

11-92

11-104 DAC-888
BYTEOAC® 8-Bit High-Speed
"Microprocessor Compatible"
Multiplying OJA Converter

DAC-06
Two's-Complement lO-Bit
Voltage-Output OJA Converter

11-23

11-116

DAC-08
8-Bit High-Speed Multiplying
OjA Converter

11-34

DAC-IO

11-131

DAC-20

11-136 DAC-8408
Quad 8-Bit Multiplying CMOS OJA
Converter With Memory

DAC-86
COMOAC® Companding
OJA Converter

11-58

11-140

11-148

PM-565A
Complete High-Speed 12-Bit
Monolithic OJA Converter

DAC-89
COMOAC® Companding
OjA Converter

11-76

PM-562
12-Bit Multiplying Current-Output
OjA Converter

DAC-88
COMOAC® Companding
OjA Converter

11-67

DAC-8212
Dual 12-Bit Buffered Multiplying
CMOS OjA Converter

2-0igit BCD High-Speed Multiplying
OJA Converter
11-50

DAC-1508A/1408A
8-Bit Multiplying OJA Converters

11-122 DAC-8012
CMOS 12-Bit Multiplying OJA
Converter "With Memory"

lO-Bit High-Speed Multiplying
OjA Converter
11-42

DAC-312
12-Bit High-Speed Multiplying
OjA Converter

DAC-02/DAC-03/DAC-05
lO-Bit-Plus-Sign Voltage-Output
OJA Converters

11-19

DAC-210
11-Bit Voltage-Output OjA Converter

6-Bit Voltage-Output OJA Converter
11-14

DAC-208
9-Bit Voltage-Output OJA Converter

11-149 PM-7226
Quad 8-Bit CMOS OJA Converter
With Voltage Output

DAC-IOO
lO-Bit Current-Output OJA Converter

11-2

n/A
CONVERTERS
Precision Monolithics Inc.

11-150 PM-7524
CMOS 8-Bit Buffered Multiplying
D/A Converter

11-196 PM-7543
12-Bit Serial-Input Multiplying CMOS
D/A Converter

11-159 PM-7528
Dual 8-Bit Buffered Multiplying CMOS
D/A Converter

11-197

11-175

PM-7545/PM-7645
12-Bit Buffered Multiplying CMOS
D/A Converters

11-208 PM-7548
12-Bit Multiplying CMOS
D/A Converter

PM-7533
CMOS Low Cost lO-Bit Multiplying
D/A Converter

11-185 PM-7541
CMOS 12-Bit Monolithic Multiplying
D/A Converter

11- 209 JM38510/11301/11302
JAN 8-Bit Multiplying D/A Converters

11-195 PM-7542
12-Bit Multiplying CMOS
D/A Converter

III

11-3

D/A
CONVERTERS
Precision Monolithics Inc.

INTRODUCTION

Figure 11.1 Gain and Offset Error Defined

A D/A converter accepts a digital input and
produces an analog output. The basic DAC
consists of a voltage or current reference, binaryweighted precision resistors, a set of electronic
switches, and a means of summing the weighted
currents.
Three important criteria for selecting a good
DAC are resolution, accuracy, and speed. Other
essential requirements to be considered are
temperature stability, input coding, output
format, reference requirements, and power
consumption.
Since introducing the first monolithic D/A
converter in 1970, PMI has continually
improved and updated its DAC product line.
Using both bipolar and CMOS technologies,
PMI offers complete selections of parametric
trade-offs available from these technologies.
Very high speed, internal references and
amplifiers are key features of the bipolar
technology DACs. The CMOS technology DACs
offer a much higher degree of logic interface
function, while maintaining absolute minimums
in power dissipation. A wide offering of
microprocessor-interfaceable DACs simplify
connection to 4,8, and 16-bit microprocessor
systems. The DAC with "memory" is another
first offered by PMI to simplify system selfdiagnosis of data path integrity.
The selection guides following the definitions
will aid you in quickly locating the appropriate
DAC for your application.

/'
/"

/

,,"

,,"

fJ';/

~"

OFFSET
,,/
ERROR o /'

T

""0-O-IG-IT-A-L-IN-P-U-T-

OIGITAL INPUT

A.C. Feedthrough - The ratio of the amplitude
of signal at the DAC output to the reference
input with all DAC switches off. This parameter
is expressed in dBs.
BCD - The abbreviation BCD stands for binarycoded decimal. It is a binary code used to
represent decimal numbers in which the digits 0
through 9 are coded, using the 4-bit binary
8-4-2-1 code.
Binary - A positive-weighted code in which a
number is represented by:
N = a020 + a1 2 1 + a2 22 + ... + a n2n
where each coefficient "aj" has a value of zero
or one. Data converters use this code in its
fractional form where:
N = a 12-1 + a 22- 2 + a32- 3 + ... + a n2- n
and N has a fractional value between zero
and one.
Bit - The unit of binary information. It can have
the value of zero or one.
Bipolar Output - When the analog signal range
includes both positive and negative values, the
output is said to be bipolar. The transfer
characteristic of an ideal 2-quadrant bipolaroutput DAC is shown in Figure 11.2.
Differential Nonlinearity (DNL) - Differential
nonlinearity is the worst case deviation of any
adjacent analog outputs from the ideal 1 LSB
step size. The deviation of the actual "step size"
from the ideal step size of 1 LSB is called
differential nonlinearity error or DNL. DACs
with DNL greater than ±1 LSB may be
non monotonic. Maximum DNL error is less than or
equal to twice the maximum INL. (See Figure 11.3)

DEFINITIONS LINEAR DIGITAL-TO-ANALOG
CONVERTERS
Absolute Accuracy - The absolute accuracy of
a DAC is the difference between the actual
unadjusted analog output and the ideal output
that is expected when a given digital code is
applied. Sources of error include full-scale error
(gain error), zero-scale error (offset error),
nonlinearity errors, and the drift of all of these.
Therefore, absolute accuracy includes all
deviations from the ideal. (See Figure 11.1)

11-4

n/A
CONVERTERS
Precision MOl1olithics Inc.

same relative accuracy (the output can change
absolutely).
Full Scale (FS) - The full-scale output of a
DAC is its maximum voltage or current. For a
binary DAC, the full-scale output occurs when
the digital inputs are all ones. The full-scale
value is one LSB less than the reference value.

Figure 11.2 Bipolar Output Converter
QUADRANT

1

+F8
ANALOG

/

/

/

ANALOG
OUTPUT

/

/

/

GAIN ~RROR

t~~T~~~~

/

Full-Scale Gain Error (GFSE) - See Gain Error.
Full-Scale Range (FSR) - The difference
between the maximum analog output and the
minimum analog output of a DAC.
Gain Drift (TCGFS) - The variation of the fullscale value (voltage or current) measured over
the operating temperature range is called gain
drift. This parameter has units of %FS, ppmFS,
or LSB. It may also be expressed % of FS/oC,
ppmFS/oC, etc.
Gain Error (GFSE) - The difference between the
actual and the ideal analog output range,
expressed as a percent of full-scale or in terms
of LSB value (see Figure 11.1). It is the deviation
in slope of the DAC transfer characteristic from
ideal.
Glitch - A glitch is a switching transient
appearing in the output during a code transition.
Its value is expressed as a product of voltage
(V x ns) or current (mA x ns) and time duration or
charge transferred (in Picocoulombs).
Integral Nonlinearity (INL) or Nonlinearity (NL)
- This is the single most important DAC
specification. PMI measures INL as the
maximum deviation of the analog output (from
the ideal) from a straight line drawn between the
end pOints, expressed as a percent of full-scale
range or in terms of LSBs. (See Figure 11.5)
For DACs, a specification of ±1/2 LSB INL
guarantees monotonicity and ±1 LSB maximum
differential nonlinearity.
Least Significant Bit (LSB) - The analog value
of the LSB is the smallest change that can occur
in the output of a DAC. It corresponds to a onebit change in the binary input. The analog value
will be either a voltage or current.
FSR
LSB (Analog Value) = - 2n
where FSR = Full-Scale Range
n = number of bits

DIGI;::~ +---=I'=-,z.¥...L..=':'::::'=:.!+ ~r:ITAL
/

/

/
/

/

IDEAL ANALOG OUTPUT

/

/

/

L ____ _

-F8
ANALOG

QUADRANT

3

Figure 11.3 Nonlinearity (NL) and Differential
Nonlinearity (DNL)

//;/
DNL = -1 1/2 LSB
.....
,.1

/
/

....... NON-MONOTONIC POINT

.DNL=+1 LSB

001

010

011

I

100

I

101

I

110

111

Dynamic Range (DR) - The dynamic range of a
DAC is the ratio of the largest output to the
smallest output (excluding zero) expressed in
decibels (dB). For linear DACs, this ratio is 2 n ,
where n = number of bits of resolution.
DR (in dB) = 20 L091Q2n '" 6n for linear DACs;
(COMDACs® are 66 or 72dB.)
Endpoint Linearity - See Integral Nonlinearity.
Functional Compliance - The functional
compliance of a DAC is the voltage range over
which the current output can be driven and for
which the DAC output current will maintain the

11-5

II

n/A
CONVERTERS
Precision Monolithics Inc.

Figure 11.5 Nonlinearity

Most Significant Bit (MSB) - The analog value
of the MSB is the largest incremental output
change obtainable by switching a single input
bit. The analog value will be either a voltage or
current.
FSR
MSB (Analog Value) = - 2
Monotoniclty - A DAC is monotonic if the
analog output either increases or remains the
same for an increasing digial input code. If the
DNL is less than or equal to ±1 LSB,
monotonicity is guaranteed. (See Figure 11.3)
Figure 11.4 DAC Transfer Curves
ANALOG

OUTPUT

Nonlinearity (NL) - See Integral Nonlinearity.
Offset Drift (TCVos, TClos) - The variation of
the offset (voltage or current) measured over the
operating temperature range. The offset drift is
divided by the temperature range over which it is
measured, and expressed in ppm per degree
centigrade or percent of full-scale range. This
parameter applies to DACs operating in the
bipolar output mode. See zero-scale drift for
DACs operating in the unipolar output mode.
Offset Error (VOSE, lOSE) - The offset error is
the error at analog zero for a data converter
operating in the bipolar mode.
Output Resistance (Ro) - Output resistance is
the equivalent internal resistance for a current
output 01 A converter as seen at its output. It is
measured as the change in output current ~I
with the change in output voltage ~V. It is a
direct measure of the true compliance.
Power Supply Sensitivity (Pss) - The change in
the output of the converter due to a change in
the power supply value. This may be expressed
as a percent of full-scale range per one percent
change in the power supply, or as a percent of
full scale per volt of power supply change.
Normally Pss is specified at DC; it is sometimes
specified over a given frequency range.

Multiplying DACs - The DAC multiplies an
analog reference by a digital word. Some DACs
can multiply only positive digital words by a
positive reference. This is known as single
quadrant operation (Quadrant I, see Figure 11.4).
Two quadrant operation (Quadrants I and III)
can be performed by a DAC that usually
operates in Quadrant I by configuring the
output for bipolar output operation. This is
accomplished by offsetting the output by a
negative MSB (1/2 of FSR), so that the MSB
becomes the sign bit. CMOS DACs provide four
quadrant operation by allowing the use of both
positive and negative references. (Quadrants
I, II, III, IV).

11·6

n/A
CONVERTERS
Precision MonoHthics Inc.

Relative Accuracy - See Integral Nonlinearity.
Resolution (n) - The resolution of a DAC is the
number of states (2n) that the FSR is divided (or
resolved) into, where n is equal to the number
of bits.
Settling Time - Settling time is the elapsed time
for the analog output to reach its final value
within a specified error band after a digital input
code change. It is usually specified for a fullscale change and measured from the 50% point
of the logic input change to the time the output
reaches its final value within the specified error
band. (See Figure 11.6)

Zero-Scale Drift (TCVzs. TClzs) - The variation
of zero scale measured over the operating
temperature. It is expressed in ppmFS/oC, or
%FS/OC, etc.
Zero-Scale Symmetry Error (VZSS) - This
definition applies only to sign-magnitude DACs.
It is the change in the analog output produced
by switching the sign bit with a zero-code input
to the magnitude bits. It is expressed in units of
voltage, current, or in fractions of an LSB.

DEFINITIONS - COMPANDING DACs
The companding (COMDAC®) DACs that PMI
manufactures are the DAC-86, DAC-88, and the
DAC-89. They are constructed such that the
more significant bits of the digital input have a
larger than binary relationship to the less
significant bits. This decreases the resolution
of the more significant bits, which increases the
analog signal range. The effect of this is to
compress more data into the more significant bits.
Chord - The mathematical formula, describing
the DAC transfer function, is implemented by
performing a piecewise linear approximation of
the function. The straight line segments used in
the approximation are called chords.
Chord Endpoints - The digital code corresponding to the maximum analog output for a
given chord is called the chord endpoint.

Figure 11.6 Settling Time Measurement
+6.Vo

I--.---'------;J_-+--..".-VO

I:~~~ "'SLEWING~
SETTLING TIME TO

FIN~NG
±Avo-----.-I

Three-State Outputs - A digital output circuit
that can be programmed to output a logic low,
logic high, or a high output impedance state.
These devices are generally connected to digital
buses.
True Compliance - The true compliance of a
DAC is the voltage range over which the current
output can vary while the DAC maintains an
absolute accuracy of ±1/2 LSB. The higher the
DAC output impedance, the better the voltage
compliance will be.
Unipolar Output - A DAC operates in the
unipolar output mode when the analog output
starts at zero, stopping at a full-scale positive or
negative value, while the digital inputs are
changed from zero to all-ones code. The analog
output occurs in one quadrant.
Zero-Scale Error (VZSE. IZSE) - The zero-scale
error is the error at analog zero for a data
converter operating in the unipolar mode.

Dynamic Range (DR) - The dynamic range of a
DAC is the ratio of the largest output to the
smallest output (excluding zero) expressed in
decibels (dB). For the COMDACs® this would be
output (117 15) divided by output (10 1)' This is then
converted to dB using the formula:

DR=20L091O(~)
10,1

(dB)

Encode Current - The encode current is the
difference between 10E(+) and 100 (+) or the
difference between 10E (_) and 100 (_) at any
code.
Full-Scale Symmetry Error - The full-scale
symmetry error of a DAC is the difference
between the maximum and the minimum analog
output values. For the COMDACs® this is the
difference between 100 (_) and 100 (+) or 10E (+)
and 10E (_).
11·7

II

n/A
CONVERTERS
Precision MOIlolithics Inc.

Output-Level Notation - Each output current
level may be designated by the digital input
code as Ie,s; where c = chord number and s =
step number. For example, 100 = zero scale
current; 101 = first step from zero; 10 ,15 =
endpoint of the first chord (Co); and 17,15 = fullscale current.
Steps - Each chord is divided into equal increments called steps.
Step Nonlinearity - This is the deviation of the
actual step size from the ideal step size within a
chord. In a linear DAC, it corresponds to
differential nonlinearity.

DIGITAL-TO-ANALOG CONVERTER
SELECTION GUIDE
PMI offers a complete line of digital-to-analog
converters (DACs), all of which are guaranteed to
be monotonic over their operating temperature
ranges, and some which have become industry
standards. The DACs have been arranged in a
matrix, which highlights their primary
characteristics so that the user can easily
narrow the selection, according to specific
requirements. More detailed specifications are
then tabulated in each product group.

Voltage Output

..

Current Output

'
E

..."'"9
5

0

0

N

0
N
0

<0

'"

•

6-Blts
8-Blts
10-Bits
12-Blts

•

0

0

••••

••

Input Coding
BCD
Binary
Complementary Binary

Sign Magnitude
Input Latches

~p

Compatible

Two's Complement
Complementary Current Outputs
Internal Reference
Logic Threshold Control
JAN Qualified
Operating Temperature Range
Military
Industrial
Commercial
'Compandlng

0

0
0

'"

<0
<0
<0

0

co .,;

'" '"
<0

~

...::;;.,;

'"5

"'"

'" ;:; ;:; tl1? '"::i; u u u '"~ ~ ~ "~ u ~
U
U
« « « « « '"
« C§ « ::;; ::;; ::;; ::;; C§ ::;;
0
0
0
0
0
11. 0
0
11. 11. 11. 11.
11.

'" U U U
U
U
« « « « «
0
0
0
0
0

Resolution

"'"~

'"<0

U
«

tl

«
<0

CMOS

0

0

0

••

co

••

<0

<0

••• •
•

<0

•
•••

•
• • ••• ••••••
•
•
••••
•
• •••
•
••• • • • • •
•
••••••
••• •• ••
•
••• ••• ••••• ••••••
• • ••••••••
•••••••••••••• ••••••

11-8

n/A
CONVERTERS
Precision Monoltthics Inc.

DIGITAL-TO-ANALOG CONVERTER SELECTION GUIDE
Six-Bit
DAC-01

6-Bit Voltage-Output D/A

31's settling, includes ref., 250mW power dissipation

8-Bit
8-Bit
8-Bit
8-Bit

85ns settling, 1LSB gain error, 0.1%INL
Includes ref., 0.1 %INL, 750ns settling
8-bit input latch, 0.1%INL, 400ns settling
0.19%INL, 250ns settling, 157mW power dissipation

Eight-Bit
DAC-08*
DAC-208
DAC-888
DAC-1508A1
1408A
PM-7524
PM-7528
JM385101
11301-11302

High-Speed Multiplying D/A
Plus Sign Voltage-Output D/A
High-Speed I'P-Compatible D/A
Multiplying D/A

'hLSB INL, 1mW power dissipation, 10ppm/oC
gain drift
1% matching, +5V to +15V single supply, 0.5LSB INL

CMOS 8-Bit Multiplying D/A with
Latches
CMOS Dual 8-Bit Buffered
Multiplying D/A
JAN Qualified DAC-08

MIL-M-38510/113 device types 01 and 02

Eight-Bit Companding
DAC-86
DAC-88
DAC-89

Companding D/A Converter
Companding D/A Converter
Companding D/A Converter ('A' Law)

1'255 law, meets D3 specs, encode and decode

Improved accuracy and speed over DAC-86
11-bit accuracy and resolution around zero

Ten-Bit and BCD
DAC-021
03/05
DAC-06
DAC-10
DAC-20
DAC-100
DAC-210
PM-7533

10-Bit-Plus-Sign Voltage-Out DIA

Includes ref., ±10V out, 21's settling

Two's Complement 10-Bit VoltageOut D/A
10-Bit High-Speed Multiplying D/A
2-Digit BCD High-Speed Multiplying D/A
10-Bit Current-Output D/A
10-Bit-Plus-Sign Voltage-Output D/A
CMOS Low-Cost 10-Bit Multiplying D/A

Includes ref., 1LSB INL, 1.51's settling
85ns settling, 10ppmfOC drift, 0.05% INL
85ns settling, 0.25LSB gain error, 0.25LSB INL
Includes ref., 375ns settling, 15ppm/oC drift
Includes ref., 0.05%INL, 1.51's settling
0.5LSB INL, 4-quadrant multiplying, 2mA Isy

Twelve-Bit
DAC-312
DAC-8012
PM-562
PM-7541
PM-75451
7645

0.012%DNL, 0.025%INL, 250ns settling
Data read back, 0.5LSB INL, 2ppm/oC TC

12-Bit High-Speed Multiplying D/A
CMOS 12-Bit Multiplying D/A with
Memory
12-Bit Multiplying D/A Converter
CMOS 12-Bit Multiplying D/A
CMOS 12-Bit Multiplying D/A with
Latches

0.25LSB INL, 1.51's settling, 0.5LSB DNL
0.5LSB INL, 4-quadrant multiplying, 2mA Isy
0.5LSB INL, 1LSB gain error, 2mA ISY

"IndIcates product avaIlable In LCC package (1986)

11·9

a

DAC-Ol
6-BIT VOLTAGE-OUTPUT

D/A CONVERTER
PreciSIOil MOIlO]ithics Inc.

FEATURES
•
•
•
•
•
•
•
•

voltage reference and fast summing op amp on one chip.
Monolithic construction provides low power consumption
and high reliability. Wide power supply range, three output
voltage options, and three input code options assure flexibility for a wide variety of applications. A seventh bit may also be
added for greater resolution. Introduced in 1970, the DAC-01
is still the fastest, lowest power, most accurate 6-bit complete
monolithic DAC available. The DAC-01 is ideal for CRT
deflection circuits, servo pOSitioning controls, digitally programmed power supplies and pulse generators, modem and
telephone system digitizing and demodulation circuits, digital filters, and 6-bit AID converters.

Fast........................... 3/o1s Settling Time Max
Complete ....... Includes Reference, Ladder, Op Amp
Low Power Consumption ................ 250mW Max
6-Blt Resolution .....•.•.............. 7-Blt Accuracy
3 Output Options .••.•.............. +10V, ±5V, ±10V
Standard Power Supplies .•..•.......... ±12V to ±18V
TTL - Compallble Logic Levels
MIL-STD-883 Cia.. B Processing Available From Stock

ORDERING INFORMATIONtt
14-PIN HERMETIC DIP
FULL TEMP.
N.L. LSB

MILITARY
TEMP.

±1/8

DAC01AY"
DACOW"
DAC01BY"
DAC01FY"t

±1/4

COMMERCIAL
TEMP.

PIN CONNECTIONS

DAC01CY
DAC01HYt
DAC01DY

±1I2

14

MSB B1

• For devices processed In total compliance to MIL-STD-883, add /883 after
part number Consult factory for 883 data sheet
tUmpolar only - all others unipolar or bipolar
trAil commercial and Industrial temperature range parts are available With
burn-In, For ordering information see 1986 Data Book, Section 2,

LSB 86

6

1....-_-1

GENERAL DESCRIPTION

FULL-SCALE TRIM

12

BIPOLAR/UNIPOLAR

11

SUM NODE

10

SCALE FACTOR

9

GROUND

8

ANALOG OUTPUT

14-PIN HERMETIC DIP
(V-Suffix)

The DAC-01 is a complete monolithic 6-bit digital-to-analog
converter. The device contains current steering logie, current sources, a diffused resistor ladder network, precision

SIMPLIFIED SCHEMATIC

,

OIGITAL LOGIC INPUTS
MSB

1

v+

LSB

2

3

4

5

7

6

SUM

11

MODE

--

10

SCALE
FACTOR

IFS = 2mA

I I

1

I I I

r-...

r-...

.......

J J J J J"" J

VREF

.......

.......

~
~

"r~

541<

~>--

@ @

541<

- .........

""

+/

•

ANA LOG
OUTPUT

I
b13v-

14J FULL
SCALE
TRIM

11-10

BIPOLAR/
UNIPOLAR

12

•

GROUND

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _---\~ DAC-01 6-BIT VOLTAGE-OUTPUT D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS (See Note 3)

Storage Temperature ................. -65°C to +150°C
Lead Temperature (Soldering, 60 sec) ............. 300°C
Output Short-Circuit Duration (Note 2) ........ Indefinite

Operating Temperature
DAC-01A, DAC-01, DAC-01B,
DAC-01F ......................... -55°C to +125°C
DAC-01C, DAC-01H, DAC-01D .......... O°C to +70°C
DICE Junction Temperature (Ti) ....... -65°C to +150°C
V+ Supply Voltage to Ground ................. 0 to +18V
V- Supply Voltage to Ground .................. 0 to -18V
Logic Input to Ground ...................... -0.7 to +6V
Internal Power Dissipation (Note 1) .............. 500mW

NOTES:
Rating applies to ambient temperatures of 100°C For temperatures above
100·C. derate linearly at 10mWI"C.
2. Short circuit may be to ground or either supply. Rating applies to + 125· C
case temperature or +75° C ambient temperature.
3 Absolute maximum ratings apply to both DICE and packaged parts. unless
otherWise noted.

ELECTRICAL CHARACTERISTICS at Vs = ±15Vand over the rated operating temperature range, unless otherwise noted.
PARAMETER

SYMBOL

Output Options

DAC-01A

DAC-01

DAC-01B

DAC-01F

DAC-01C

DAC-01H

DAC-01D

Unipolar
Bipolar

Unipolar
Bipolar

Unipolar
Bipolar

Unipolar

Unipolar
Bipolar

Umpolar

Unipolar
Bipolar

UNITS

Temperature Range

TA

-55/+125

-55/+t25

-55/+125

-55/+125

0/+70

0/+70

0/+70

·C

Nonlinearity 25° C/Maxlmum

NL

±0.20

±0.40

±0.40

±040

±040

±0.40

±0.78

%FS

Nonlinearity Over
Temperature - Maximum

NL

±0.30

±0.45

±0.45

±0.45

±0.45

±0.45

±0.78

%FS

Full-Scale TempcoMaximum

Tc

±80

±80

±120

±80

±160

±160

±160

ppm/·C

25

25

25

40

25

40

50

mV

Unipolar Zero-Scale Output
V
Voltage - Maximum (Notes 1,2) zs

ELECTRICAL CHARACTERISTICS for all DAC-01 grades, Vs = ±15V and over the rated operating temperature range
unless otherwise noted.
DAC-01
PARAMETER
Unipolar Full Range
Output Voltage I Note 3)

SYMBOL

CONDITIONS

VFR

2kO load, 10gic:O; 0.8V, short pm 13 to pin 14.
Short pin 12 to Ground and pin 10 to pin 11.

Bipolar Output Voltage INote 3)
±5 Volt Range
VFR+

VFR±10 Volt Range

VFA+
VFRBipolar Offset Voltage (Note 1)

MIN

2kO load, short pin 11 to pin 12.
Short pin 13 to pin 14, short pin 10 to pin 11
Logic Inputs :0; 0.8V
Logic Inputs 2: 2 OV
Open pm 10
Logic Inputs,; 0.8V
Logic Inputs 2: 2.0V

MAX

UNITS

+10.0

+11.75

V

+4.93
-5.94

+5.94
-4.93

V

+9.86
-11.89

+11.89
-9.86

V

±40
±80

±5 Volt Range
± 10 Volt Range

±112 II V FR+ I -I VFs-1 )

TYP

Resolution

±70
±140

6

Logic Input "0"

V 1NL

Logic Input "1"

V ,NH

Logic Input Current. Each Input

liN

Power Supply Sensitivity

Pss

±12V'; Vs'; ±18V VFS ~ 10.0V

Power Consumption

Pd

No Load

Supply Current

1+
1-

V+= +15V
V-= -15V Logic Inputs'; 0.8V

Setting Time to ±1/2 LSB
INote4)

ts

2.0V ,; Logic Level,; 0.8V TA = 25· C

mV
Bits

0.8

V

±2

±8

p.A

±0.01

±015

200

250

mW

7.3
9.3

mA

3

P.s

V

NOTES:
1. Zero-scale or bipolar offset voltage can be trimmed to zero Yolts or to the
exact one's or two's complement condition with an external resistor
network to pin 11.
2. Logic input voltage 2: 2.0V

3.

4.

11-11

15

Full-scale IS adjustable to precisely 10V for unipolar operation and 10V or
20V peak-to-peak bipolar operation with an external 5000 potenllometer
from pin 14 to V-.
Guaranteed by design

1/86, Rev. A

II

_ _ _ _ _ _ _ _ _--IIfMDDAC-01 8-BIT VOLTAGE-OUTPUT DlA CONVERTER

DICE CHARACTERISTICS
1.
2.
3.
4.
5.
8.

B1 (MSB)
B2
B3
B4
BS
B8 (LSB)

7. V+

8.
9.
10.
11.
12.
13.
14.

ANALOG OUTPUT
GROUND
SCALE FACTOR
SUM NODE
BIPOLAR/UNIPOLAR
VFULL-SCALE TRIM

DIE SIZE 0.092 X 0.054 Inch, 4988 sq. mill (2.34 X 1.37 mm, 3.21 Iq. mm)
For additional DICE Information ref.r to
1988 Data Book, Section 2.

WAFER TEST LIMITS at TA=25°C.
DAC-01N
BIPOLAR AND
UNIPOLAR

DAC-01G
BIPOLAR AND
UNIPOLAR

PARAMETER

SYMBOL

CONDITIONS

LIMIT

LIMIT

UNITS

Nonlinearity

NL

VS =±15V

114

1/2

L.S.B.MAX

Zero-Scale Voltage

VZS

VS =±15V

25

35

mVMAX

WAFER TEST LIMITS at Vs= ± 15V, TA= 25°C, unless otherwise noted.
DAC-D1
PARAMETER

SYMBOL

CONDITIONS

Unipolar Full-Scale Output
Voltage (All Models)

VFR

2kn Load. Logic S 0.8V. Short V- to Full-Scale Trim. Unipolar/
Bipolar to Ground. and Scale Factor to Sum Node

VFRt
VFR-

2kn Load. Short Sum Node to Unipolar/Bipolar.
Short V- to Full-Scale Trim and Scale Factor to Sum Node.
Logic Inputs S O.BV
Logic Inputs ~ 2.0V
Open-Scale Factor
Logic Inputs S 0.8V

Bipolar Output Voltage
±5 Volt Range
± 10 Volt Range

VFRt
VFR-

Logic Inputs ~ 2.0V

± 5 Volt Range
±to Volt Range

Bipolar Offset Voltage
±1I2 (IVFRtl-IVFR-I)

LIMIT

UNITS

10.00
11.75

VMIN
V MAX

+4.93
-5.94

VMIN
V MAX

+9.78

VMIN

-11.89

V MAX

±1/2

LSBMAX

6

Bits MAX

Logic Input "0"

V1NL

0.8

V MAX

Logic Input "I"

V1NH

2

VMIN

LogiC Input Current. Each Input

Vov

±8

pAMAX

Power Supply Rejection

PSR

±12V S Vs S ±18V. Vs = 10.OV

0.15

'IIoFSNMAX

Power Consumption

Pd

No Load

250

mWMAX

Resolution

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In asesmbly methods and normal yield loss, yield after packaging Is not
guaranteed for standard product dice. Consult factory to negotiate specifications baesd on dice lot qualification through sample lot asesmbly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at 25° C.
DAC-01N

DAC-01G

TYPICAL

TYPICAL

To ±1/2 LSB

1.5

1.5

,.s

VS=±15V

eo

90

ppmfOC

PARAMETER

SYMBOL

CONDITIONS

Settling Time

ts

Full-Scale Tempco

11-12

UNITS

1/86, Rev. A

-----------l~ DAC-01 a-BIT VOLTAGE-OUTPUT D/A CONVERTER

BASIC CIRCUIT CONNECTIONS

APPLICATIONS INFORMATION

FULL-SCALE ADJUSTMENT TECHNIQUE

INPUT CODES
The DAC-01 uses standard complementary binary coding for
unipolar operation (all inputs logic high produces zero output voltage). One's complement coding may be Implemented
by shorting pin 11 to pin 12 and inverting the MSB (all other
bits are not inverted). Complementary offset binary coding
may be implemented by shorting pin 11 to pin 12, and Injecting approximately 5/JA into pin 11 (which Is at ground potential) by using the "optional Zero-Scale or bipolar offset
adjustment" circuit. Two's complement code is achieved
when the MSB for complementary offset binary Is Inverted.
FULL-SCALE ADJUST
A 500n pot from pin 14 to V- can be used to adjust the
Full-Scale output voltage to exactly 10 volts in unipolar mode
or 10 to 20 volts peak-to-peak in bipolar mode. If no pot is
used, connect pin 14 to V-.

~15V

OPTIONAL ZERO-SCALE OR
BIPOLAR OFFSET ADJUSTMENT

SCALE FACTOR
For+ 10 volts or±5 volt outputs, short pin 10 to pin 11 (adjusts
the feedback resistor around the output amplifier). For ±10
volt output, leave pin 10 open. Intermediate output voltages
may be obtained by placing a pot between pin 10 and pin 11.
This will, however, seriously degrade the Full-Scale temperature coefficient due to the mismatch between the
+1150ppmfOC tempco of the diffused resistors and the pot
tempco.

100kn

v- O----I'o/V'----{) v+
4701d1

I

"
OAe-01

CAPACITIVE LOADS
When driving capacitive loads greater than 50pF in Unipolar
mode or 30pF in Bipolar mode a 100pF capacitor may be
placed from pin 11 to ground for added stability.

ADDITION OF 7th BIT
-15Vo-~-------,

100kS)

IN4148

LOWER RESOLUTION APPLICATIONS
When less than 6 bits of resolution is required, connect
unused bits to a voltage level greater than +2.0 volts. The +5
volt logic supply is adequate.

27MH

I
MSB

~ ~T AL
INPUT

o---r------'

LSB

11-13

1/86, Rev. A

III

])1lC-02/])1lC-03/])1lC-05
lO-BIT-PLUS-SIGN VOLTAGE-OUTPUT
D/A CONVERTERS
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•

tage reference, current steering logic, current sources, R-2R
resistor network, logic-controlled polarity switch, and high
speed internally-compensated output op amp. Monotonicity
guaranteed over the entire temperature range is achieved
using an untrimmed diffused R-2R resistor network. The buffered reference input is capable of tracking over a wide range
of voltages, increasing application flexibility. The wide power
supply range, low power consumption, wide logic input
compatibility and sign-magnitude coding assures utility in a
wide range of applications including CRT displays, data
acquisition systems, A/D converters, servo positioning controls, and audio digitizing/reconstruction systems.

Complete ........... Includes Reference and Op Amp
Compact .................. Single 18-Pln DIP Package
Bipolar Output . . . . . . .. (± 10V) Sign-Magnitude Coding
DAC-03 - Unipolar Only; ............... +5V or +10V
Monotonlclty Guaranteed
Nonlinearity ................................. ± 1 LSB
Fast .............................. 2.0JLs Settling Time
Stable.................. Full-Scale Tempco 60ppm/o C
Low Power Consumption ................ 300mW Max
TTL, CMOS Compatible Inputs
MIL-STD-883 Class B Processing Available on DAC-05

ORDERING INFORMATION
MONOTONOCITY
BITS

10
9
8
7

The DAC-03 is similar in construction to the DAC-02/DAC-05
except for a unipolar only output. This device is intended for
low cost, limited temperature range applications, with the
same general specifications as its premium counterparts.

t

PACKAGE: la-PIN HERMETIC DIP
MILITARY
TEMp·
DAC05AX
DAC05CX

COMMERCIAL TEMP
DAC02ACX
DAC02BCX
DAC02CCX
DAC02DDX

PIN CONNECTIONS

DAC03ADX DAC05EX
DAC03BDX
DAC03CDX DAC05GX
DAC03DDX

• For devices processed in total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet.
t All commercial and Industrral temperature range parts are available with
burn-in. For ordering Information see 1986 Data Book, Section 2,

18-PIN DIP
(X-Suffix)

GENERAL DESCRIPTION
The DAC-02 and DAC-05 are complete 10-bit plus sign D/A
converters on a single monolithic chip. All elements of a
complete sign-magnitude DAC are included; precision vol-

SIMPLIFIED SCHEMATIC
~---------OIGITAL LOGICINPUTS--------~

BIT 2

~~~~~iN.~,~,~r<:ID-IT--t---t--f__-+_-+_-+-+--+-___j--t_-_t_-~-~

> __

ANALOG
t-t--<,OUTPUT

ANALOG GROUND

11-14

1/86, Rev. A

-------------

--

_ _ _ _ _ _-I~ DAC-02IDAC-03/DAC-05 10-BIT-PLUS-SIGN VOLTAGE-OUTPUT D/A CONVERTER
ABSOLUTE MAXIMUM RATINGS

(Note)

Lead Temperature (Soldering, 60 sec) ............ 300·C
Output Short Circuit Duration ••••••••••••••••• Indefinite
(Short circuit may be to ground or either supply.)
NOTE: For ambIent temperatures above 100' C derate 100mW/' C.

Operating Temperature Range
DAC-05A, C •••••••••••••••••••••••• -55· C to + 125· C
DAC-02 and DAC-03, All
DAC-05E, G •••••••••••••••••••••.•••••• O· C to + 70· C
Storage Temperature Range
-65·C to +150·C
V+ Supply to Analog Ground •.••••••••••••••• Oto+18V
V- Supply to Analog Ground ••••••••••••••••••
to -18V
Analog Ground to Digital Ground
Oto ±0.5V

..........

............

OUTPUT VOLTAGE RANGE SELECTION TABLE

PRODUCT

OUTPUT
VOLTAGE
RANGE

ADD AS
SUFFIX TO
PART NO.

DAC02
DAC03
DAC03
DAC05

±10V
01o+10V
010 +5V
±10V

1
1
2
1

o

Logic Inputs to Digital Ground •••••••• -5V to (V + -0.7V)
Internal Reference Output Current •••••••••••••.• 300l'A
Reference Input Voltage •••••••••••••••••••••• Oto +10V
Internal Power Dissipation •••••••••••••••••••••• 500mW

=

ELECTRICAL CHARACTERISTICS at Vs ± 15V, 0:,; TA :,; + 70· C for DAC-02 and DAC-05E & G, TA
and -55 :,; TA:'; + 125· C for DAC-05A & C, unless otherwise noted.
PARAMETER

SYMBOL CONDITIONS

DAC-02

DAC-D3

DAC-OS

AC
BC
CC
DO

AD
BD
CD
DO

AlE

AC/BC
CC
DO

AD/BD
CD
DO

Monotonoclty

Nonlinearity

NL

C/G

MIN

= 25· C

AC/BC/CC
Full-Scale Tempco

EtG
C
DO
ALL

EXT REF
Settling Time
Full Range Output
Voltage (Note 1)

Zero-Scale Offset

ts

To 112 LSB. 10V Step (Note 4)

ALL

VFR

VFR+ (SB High)
VFA- (SB Low)
DAC-03+10V
+5V

ALL
ALL

Vzs

Full Range BIpolar Symmetry

Vzss

VFRS

ALL

±01
±02
±04
±0.5

% FS

±60
±60
±45 ±100
±60 ±120
±150

ppm/'C

±30
±40

ppm/'C

ALL
ALL

SB High. All other logIc
Inputs low. TA = 25'C

±5
±10
±10

mV

ALL

±2

±10

mV

±5
±10
±10

mV

ALL

±1
±1
±4
±30
±30

±60
±80

mV

±20
±10

±70
±50

mV
mV

ALL
ALL

(Note 2)
VFR+-IVFR-I
(Note 3)

AC/BC/CC
DO
AC/BC/CC
DO

N/A

N/A
ALL
ALL

TA = Min or Max

TA=25'C
Reference Input Bias Current

±1
±1
±5

ALL

I.

ALL

ALL

ALL

100

nA

Reference Input Impedance

Z,N

ALL

ALL

ALL

200

MO

Reference Input Slew Rate

SR

ALL

ALL

EtG
AlC

1.5

V/p.$

Reference Output Voltage

VREF

ALL

ALL

ALL

6.7

Volts

11-15

U
()

0

~
~

......
......
Q

Volts

-

0

C)

+11.5
-10
+115
+5.75

ALL

+10
-11.5
+10
+5.00

~
~

0
f;<

,."

ALL

ALL
ALL

TA = Min or Max

Zero-Scale Symmetry

ALL
ALL

UNITS

BIts

AlE

A

Tc

MAX

10
9
8

ALL

for DAC-03

Z

TYP

C/G
INTREF

~
~

1/86, Rev. A

I

-------I!EMD

DA~2IDAC_o3IDAC_o510-BIT-PLUS-SIGN VOLTAGE-OUTPUT DlA CONVERTER

ELECTRICAL CHARACTERISTICS at Vs =± 15V, 0 ~ TA ~ + 70° C for DAC-02 and DAC-05E & G, TA = 25° C for DAC-03
and -55 ~ TA~ +125°C for DAC-05A & C, unless otherwise noted. (Continued)
PARAMETER

SYMBOL CONDITIONS
liN

Logic Input 0
Logic Input 1

V,NL
V,NH

Positive Supply Current

1+

SB High. All other
logiC Inputs low.

Negative Supply Current

1-

SB High. All other
logic Inputs low
Vs= ±12 to ±18V

Power Supply Sensitivity

Pss

Power DisSipation

Pd

DAC-02

DAC-03

DAC-D5

ALL

ALL

ALL

ALL
ALL

ALL
ALL

ALL
ALL

AC/BC/CC
DD

ALL

Each input
-5V to IV+ - 0.7)V

Logic Input Current

AC/BC/CC
DD

10

ALL
ALL

Guaranteed by VFR test

ALL

0.8
2

-10
-9
-9 -11.6

ALL

TA = Min to Max

Output Drive Current

/loA

ALL

ALL

T A =25·C

NOTES:
1. Reference output termonal connected directly to reference Inputterminal.
RL = 2kO for 10V devices. RL = lkO for 5V devices. all logic inputs"2.0V
2. Zero-scale symmetry Is the change in the output voltage produced by
swltchong the sign-bit with all logic bits low IVzs+ - Vzs-).

UNITS

±10

+7
+10
+7 +11.6

ALL
ALL
AC/BC/CC
DD

MAX

±1

Volts

mA

mA

-10 -11.6

TA = Min to Max
T A =25·C

10UT= 0

TVP

ALL

ALL
AC/BClCC
DD

MIN

ALL

ALL

- ±0.Q15 ±0.05
-±0.015 ±01
±0.05 ±01
± 02 ±.05
225
225
200
250

% VFsIV

300
350
300
350

mW

5

mA

3. Full-scale bipolar symmetry is the magnitude of the difference between
VFR+and IVFR-I.

4. Guaranteed by design.

BURN·IN TEST CIRCUIT

-IBV±O.5V

+IBV±O.5V

10

NOTE:

PINS 14. 15. III 17 MUST NOT BE CONNECTED TO
A COMMON BUS BETWEEN DEVICES.

11-16

1/86, Rev. A

-------tlfMI>

DAC-02/DAC-03/DAC·0510·BIT·PLUS·SIGN VOLTAGE·OUTPUT D/A CONVERTER

DICE CHARACTERISTICS

1.
2.
3.
4.
5.
6.
7.
8.
9.

BIT 1·MSB
BIT2
BIT3
BIT4
BITS
BIT6
BIT7
BIT8
BIT9

10.
11.
12.
13.
14.
15.
16.
17.
18.

BIT 10
DIGITAL GROUND
VANALOG GROUND
ANALOG OUTPUT
REF IN
V+
REF OUT
SIGN BIT

For additional DICE information refer to
1986 Data Book, Section 2.
NOTE:
Voltage output range programmable by connecting
'(10VI to analog output for 10 volt range. Jumps from
"(5VI to analog output for 5 volt range t Bits 11 & 12 (not
normally used)

DIE SIZE 0.163 X 0.090 Inch; 14,670 sq. mils
(4.14 X 2.286 mm, 9.464 sq. mm)

WAFER TEST LIMITS at Vs = ± 15V, TA = 25° C and +10V full-scale output, unless otherwise noted.
PARAMETER

CONDITIONS

Resolution
(Bits 11 and 12
Not Normally Used I

Bipolar Output
Unipolar Output

Monotonicity
Nonlinearity

DAC-02-N

DAC·02-G

LIMIT

LIMIT

UNITS

13
12

13
12

Bits MAX

9

S

Bits MIN

±01

±02

% FS MAX

±10

±10

mVMAX
mVMAX

Zero-Scale Offset

Sign Bit High, All Other
Inputs Low

Zero-Scale Symmetry

± 10V Full-Scale

±5

±5

Full-Scale Bipolar Symmetry

± 10V Full-Scale

±60

±60

mVMAX

Power Supply Rejection

Vs =±12Vto±1SV

005

005

'I, VFriV MAX

Power Dissipation

300

300

mWMAX

Lagle Input "0"

OS

OS

V MAX

±11.5
±10

±11.5
±10

V MAX
VMIN

Lagle Input "1"
Full Range
Output Voltage

VMIN
Sign-Bit
High or Low

NOTE:
Electncal tests are performed at wafer probe to the limits shown. Due to vanatlons In assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice Consult factory to negotiate speCifications based on dice lot qualification through sample lot assembly and testing

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V and + 10V full-scale output, unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

Full-Scale Tempeo

Internai Reference

Settling Time ITA = 25° CI

To ±1/2 LSB 10 Volt Step

DAC-02-N

DAC·02·G

TYPICAL

TYPICAL

UNITS

60

60

ppmfOC

2

I'S

Logic Input Current
NOTE:
When ordering DICE In this series, use DAC-02 numbers and grades above

11·17

1/86, Rev. A

I

_ _ _ _ _ _--I~ DAC-ii2iDAC-03iDAC-iiS iO-BIT-PLUS-SIGN VOLTAGE-OUTPUT D/A CONVERTER

TYPICAL APPLICATIONS

will be obtained if a low tempco resistor is used or if pot and
resistor tempcos match. Alternatively, a single pot of ::;72kO
may be used.

The DAC-02's, DAC-03's and DAC-OS's logic input stages
require about 1J.LA and are capable of operation with inputs
between -S volts and V+ less 0.7 volt. This wide input voltage
range allows direct CMOS interfacing in most applications,
the exception being where the CMOS logic and D/A converter
must use the same positive power supply.

REFERENCE INPUT BYPASS
Lowest noise and fastest settling operation will be obtained
by bypassing the reference input to analog ground with a
0.01 J.LF disk capacitor.

In this special case, a diode should be placed in series with
the CMOS driving device's Voolead as shown in Figure 1. The
diode limits Vo to V+ less 0.7 volt - since the output from the
CMOS device cannot exceed this value, the DAC's maximum
input voltage rule is satisfied. Summarizing: in all applications, the DAC-02, DAC-03 and DAC-OS require either no
interfacing components, or at most a single inexpensive
diode for full CMOS compatibility.

GROUNDING
For optimum noise rejection, separate digital and analog
grounds have been brought out. Best results will be obtained
if these grounds are connected together at one point only,
preferably near the DAC-02, DAC-03 and DAC-OS package,
so that the large digital currents do not flow through the
analog ground path.

CMOS LOGIC INTERFACE CIRCUIT

APPLICATIONS INFORMATION
LOWER RESOLUTION APPLICATIONS
For applications not requiring full 1O-bit resolution, unused
logic inputs should be tied to ground.

v-

v+

UNIPOLAR OPERATION
Operation as a 10-bit straight binary converter may be
implemented by permanently tying the sign-bit to +SV (for
positive full-scale output) or to ground (for negative fullscale output). In the DAC-03 only, Pin 18 unipolar enable is
tied to Pin 17.

VOLTAGE
OUTPUT

INPUTS,;;;; V+ LESS

IN4148

07 VOLTS

POWER SUPPLIES
The DAC-02, DAC-03 and DAC-OS will operate within specifications for power supplies ranging from ± 12Vto ± 18V. Power
supplies should be bypassed near the package with a 0.1J.LF
disk capacitor.

CMOS DR IVING

DEVICE

vSS

CAPACITIVE LOADING
The output operational amplifier provides stable operation
with capacitive loads up to 100pF.

CONNECTION INFORMATION

REFERENCE OUTPUT
For best results, reference output current should not exceed
100J.LA.

FULL-SCALE ADJUSTMENT CIRCUIT
r---

DIGITAL INPUTS

MS'

FULL
SCALE
ADJUST
POT

~

USE WITH EXTERNAL REFERENCES
Positive-polarity external reference voltages referred to
analog ground may be applied to the reference input terminal to improve full-scale tempco, to provide tracking to other
system elements, or to slave a number of DAC-02's, DAC-03's
and DAC-OS's to the reference output of anyone of them.
This reference voltage should be between +SV to +7V for
optimum performance.

SIGN*
LSB BIT V+

~'-O.=,,+-+---f'.

SIGN PLUS MAGNITUDE CODING TABLE
(DAC-02, DAC-03 and DAC-05)
LSB

SIGN-BIT MSB
+ FULL SCALE
+ HALF-SCALE
ZERO-SCALE (+)
ZERO-SCALE (-)
- HALF-SCALE
FULL-SCALE

FULL-SCALE ADJUSTMENT
Full-scale output voltage may be trimmed by use of a potentiometer and series resistor as shown; however, best results
11-18

1
0
0
0

1
0
0 0
0 0
0

1
0
0
0
0

1
0
0
0
0

1
0
0
0
0

1
0
0
0
0

0
0
0
0

0
0
0
0

0
0
0
0

0
0
0
0

1/86, Rev. A

DAC-06
TWO'S-COMPLEMENT lO-BIT VOLTAGE-OUTPUT
D/A CONVERTER
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•

are included - precision voltage reference, current steering
logic, current sources, R-2R resistor network, bipolar offset
circuit and high speed internally compensated output op
amp. Monotonicity guaranteed over the entire operating
temperature range is achieved using an untrimmed diffused
R-2R resistor network. The buffered reference Input is
capable of tracking over a wide range of voltages, increasing
application flexibility. The user may also easily implement
one's complement, straight offset binary, or unipolar operation. The ±12V to ±18V power supply range, low power
consumption, TTL and CMOS compatibility, wide logic input
compatibility and adaptable logic coding capability assure
utility in a wide range of applications.

Complete ............. Includes Reference and Op Amp
Compact ...............•... Single 18-Pln DIP Package
Bipolar Output ..........•.. Two's Complement Coding
Monotoniclty Guaranteed
Nonlinearity.................................. ± 1 LSB
Fast •..............•........•..... 1.51's Settling Time
Low Power Consumption ................• 300mW Max
TTL, CMOS Compatible Inputs

ORDERING INFORMATIONt
MONOTONICITY

BITS
10
9
B

PACKAGE 18-PIN HERMETIC DIP
MILITARY
TEMP

COMMERCIAL
TEMP

PIN CONNECTIONS

DAC-OBEX
CAC-06BX'
CAC-06CX'

CAC-06FX
CAC-06GX

18

• For devices processed In total compliance to MIL-STD-BB3, add IBB3 after
part number Consult factory for BB3 data sheet.
tAli commercial and Industrial temperature range parts are available with
burn·1n For ordertng information see 1986 Data Book, Section 2
BIT 7

7

BIPOLAR ADJUST

17

REFERENCE OUTPUT

16

POSITIVE POWER SUPPLY

16

REFERENCE INPUT

14

ANALOG OUTPUT

13

ANALOG GROUND

12

NEGATIVE POWER SUPPLY

11

DIGITAL GROUND

GENERAL DESCRIPTION
The DAC-06 is a complete 10-bit two's complement DlA
converter on a Single 90 x 163 mil monolithic chip. All elements of a complete bipolar output two's complement DAC

18-PIN DIP (X-SUFFIX)

I

SIMPLIFIED SCHEMATIC
, " - - - - - - - O I G I T A L LOGIC INPUTS -------~\

SIGN
BIT
1

REF

BIT
2
2

BIT

BIT

BIT

BIT

BIT

BIT

BIT

BIT

V+

BIPOLAR

,.

345678910ADJUST

3

4

5

6

7

8

9

10

18

+-__4-__~__-+____~__+-__4-__~__-+__~~__~~__~

OUTPU~T~~-{O~~~~~~~__

REF
INPUTo-+--+_~

,.

DIGITAL

GNO

~l1~~1---------+-~~~~~-+~-+~--~--~--~~-++--+~--h

ANALOG

OUTPUT

>--H"",.'"

13

12

ANALOG GROUND

V-

1/86, Rev. A

11-19

---~--

--------

----

--------I~ DAC-iiii TWO'S-COMPLEMENT 10-BIT VOLTAGE-OUTPUT D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
Operating Temperature Range
DAC-OSB, C ......................... -55°C to +125°C
DAC-OSE, F, G .......................... 0° C to + 70° C
DICE Junction Temperature (TJ) ••••••••• -S5°C to +150°C
Storage Temperature Range ••••••.•••••• -S5°C to +150°C
V+ Supply to Analog Ground ................... 0 to +18V
V- Supply to Analog Ground .................... 0 to -18V
Analog Ground to Digital Ground ••••••••••••••• 0 to ±0.5V

ELECTRICAL CHARACTERISTICS at
DAC-OSE, F & G, unless otherwise noted.
PARAMETER

SYMBOL

Vs

=

Logic Inputs to Digital Ground •• , •••••• " -5V to (V+ -0.7)V
Internal Reference Output Current ••••••••••••••••• 3ool'A
Reference Input Voltage ... • • • • .. • • .. .. .. .. .. ... 0 to + 10V
Bipolar Offset Input Voltage .................... 0 to +10V
Internal Power DisSipation ••••••••••••••••••••••• 500mW
Lead Temperature (Soldering, SO sec) •••••••••••••• 300°C
Output Short-Circuit Duration ••••••••••••••••••• Indefinite
(Short circuit may be to ground or either supply)

±15V; -55°C S TA S +125°C for DAC-OSB

CONDITIONS

DAC-08

Resolution
Monotonlcity

Nonlinearity

Full-Scale Tempco

ts

10

E
BIF
CIG

10

TYP

MAX

UNITS
Bits
Bits

E
BIF
CIG

±O.I
±0.2
±0.4

% FS

TA = Full Temp.

E
BIF
CIG

±0.2
±0.3
±0.5

% FS

Total Internal
Ref Con nected
Zero Drift Ext
Ref Applied

Settling Time

MIN

All

and O°C S TA S +70°C for

TA= 25'C
NL

TC'JFS

& C;

B
ElF/G
C

±45
±45
±60

All

±30

To ± 112 LSB. 10V Step

All

1.5

Short Pin 18 to

TA =25'C

All

±I

Ground I Note I I

TA = Full Temp.

All

±90
±IOO
±120

ppml'C
ppml'C
p.s

±5

mV

Unipolar Zero-Scale Output

Vzs

±IO

mV

Bipolar Offset Voltage

BPOff

Connect Pins 15. 17 & 18 INote 31

All

-5

+2.5

% Range

Full Range Output Voltage

VFR

Connect Pin 15 to 17 I Note 21
RL = 2kll

All

10

11.5

V

±2

Reference Input Bias Current

I.

All

100

nA

Reference Input Impedance

Z,N

All

200

Mil

Reference Input Slew Rate

SR

All

1.5

VII's

Reference Output Voltage

VREF

All

6.7

Logic I nput Current

liN

Logic Input "0"
Logic I nput "I"
Power Supply Sensitivity

Pss

Supply Current

10

V,NL

All

0.8

V,NH

All

1+

V
V

±0.02

±0.05

% FSN

All

±0.02

±O.I

%FSN

10

mA

All

-9

-10

mA

TA = 25'C

All

250

300

mW

TA = Full Temp.

All

350

mW

All

TA=25'C

Po

,..A

All

Vs =±12Vto±18V TA=25'C
TA = Full Temp.

1Power DiSSipation

V

All

Each Input -5V to IV+ -0.7IV

NOTES:
1. May be operated in the 0 to + 10V unipolar mode by shorting Pin 18 to
Ground.
2. VFR = IVFR+I + IVFR-Iand istrimmabletoexactly 10Vrangewith the circuit
shown in typical applications.
3. Bipolar offset voltage is trimmable to exact two's or one's complement
condition with the circuit shown in typical applications.

11-20

1/86, Rev. A

--------l!EMD

DAC-Oe TWO'S-COMPLEMENT 10-BIT VOLTAGE-OUTPUT D/A CONVERTER

TYPICAL APPLICATIONS
ADJUSTING FOR TWO'S COMPLEMENT CODING
1. Connect Full-Scale Adjust and Bipolar Adjust Circuitry as
shown In figure.

REFERENCE OUTPUT
For best results, reference output current should not exceed
1001LA.

2. Turn all bits OFF (VFs--1 LSB) = 1000000000
3. Adjust Bipolar Pot for VFS at output ............ -5.000V

POWER SUPPLIES
The DAC-OS will operate within specifications for power
supplies ranging from ± 12V to ± 18V. Power supplies should
be bypassed near the package with a 0.11LF disk capacitor.
Chip users should connect the substrate to V-.

4. Turn all bits ON (VFR+) = 0111111111
5. Adjust Full-Scale Pot for desired
VFR+ value ................................... +4.990V
S. Check Zero-Scale Reading (Vzs) = 0000000000
If this reading Is outside desired Vzs range, readjust BIpolar Pot until the output reads O.OOOOV.

GROUNDING
For optimum noise rejection, separate digital and analog
grounds have been brought out. Best results will be obtained
if these grounds are connected together at one point only,
preferably at the DAC-OS package, so that large digital
currents do not flow through the analog ground path.

TWO'S COMPLEMENT CODING TABLE
INPUT
MSB

LSB

IDEAL
OUTPUT
+4.990V

VFS+ -1 LSB

0 1 1

1
0

VFS+-2LSB

0 1 1 1 1

+1LSB

0 0 0 0 0 0 0 0 0

Zero

0 0 0 0 0 0 0 0 0 0

CAPACITIVE LOADING
The output operational amplifier provides stable operation
with capacitive loads up to 100pF.

+4.980V
+0.010V

FULL-SCALE OUTPUT RANGE AND
BIPOLAR OFFSET ADJUSTMENT CIRCUIT

O.OOOV

-1LSB

1 1 1 1 1 1

-0.010V

VFs-+1LSB

0 0 0 0 0 0 0 0 1

-4.990V

VFS-

0 0 0 0 0 0 0 0 0

-5.000V

;----DIGITAL INPUTS ~
MSB

Lse

V+

ADJUSTING FOR ONE'S COMPLEMENT CODING
1. Connect Full-Scale Adjust and Bipolar Adjust Circuitry as
shown in above figure.
2. Turn all bits OFF (VFR-) = 1000000000

ANALOG

I

13

GNO*u

3. Adjust Bipolar Pot for VFR- at output .......... -5.0000V
4. Turn all bits ON (VFR+) = 0111111111
5. Adjust Full-Scale Pot for desired
VFR+value .................................. +5.0000V

·SEE APPLICATION NOTES FOR DETAILS

**Tle TO GROUND FOR UNIPOLAR OPERATION
"*GROUNDING - FOR OPTIMUM NOISE REJECTION, SEPARATE DIGITAL AND ANALOG GROUNDS HAVE BEEN BROUGHT
OUT BEST AESUL T5 WILL BE OBTAINED

ONE'S COMPLEMENT CODING TABLE
INPUT
MSB

LSB

IDEAL
OUTPUT

VFS+ -1 LSB

0

1

+5.000V

VFs+-2LSB

0

0

+4.990V

+0

0 0 0 0 0 0 0 0 0 0

+0.005V

-0

1 1 1 1 1 1 1 1 1

-0.005V

VFs-+2LSB

0 0 0 0 0 0 0 0 1

-4.990V

VFs-+1LSB

0 0 0 0 0 0 0 0 0

-5.000V

IF THESE GROUNDS ARE CONNECTED
TOGETHER AT ONE POINT ONLY, PREFERABLY NEAR OAe-OS PACKAGE, so THAT
THE LARGE DIGITAL CURRENTS 00 NOT
FLOW THROUGH THE ANALOG GROUND

PATH

EXTERNAL ADJUSTMENT NETWORK
Full-scale output range and bipolar offset may be adjusted
by using the circuit shown in the figure above. Best results
will be obtained when low tempco pots and resistors are
used, or if pot and resistor tempcos match.

Note that two zero states will straddle (±1/2 LSB) the true zero. Therefore the
DAC will give symmetrical outputs for both positive and negative full-scale.

11-21

1/86, Rev_ A

---------I~ DAC-06 TWO'S-COMPLEMENT 10-BIT VOLTAGE-OUTPUT D/A CONVERTER

CODE CONVERSION TO OFFSET BINARY
Offset binary coding is exactly the same as two's complement
coding except that the most significant bit occurs in true,
ratherthan inverted form and the output states are relabeled.
To convert the DAC-06 to offset binary code operation,
simply place a logic inverter in series with the MSB input
(Pin 1) and invert the MSB value shown in steps 2,4 and 6 of
the two's complement adjustment procedure shown above.

the CMOS logic and D/A converter must use the same
positive power supply.
In this special case, a diode should be placed in series with
the CMOS driving device's Voolead as shown in Figure 1. The
diode limits Vo to V+ less 0.7 volt - since the output from the
CMOS device cannot exceed this value, the DAC's maximum
input voltage rule is satisfied. Summarizing: in all applications, the DAC"06 requires either no interfacing components,
or at most a single inexpensive diode for full CMOS
compatibility.

OFFSET BINARY CODING TABLE
INPUT
MSB

LSB

IDEAL
OUTPUT

1

+4.990V

1 0

+4.980V

VFS+ -1LSB
VFs+-2LSB

1 1 1 1 1 1

ZERO

0 0 0 0 0 0 0 0 0

v-

v+

VOLTAGE

0.00

-1LSB

0 1

1 1 1

-0.005V

VFS-+1LSB

0 0 0 0 0 0 0 0 0

-4.990V

VFS-

0 0 0 0 0 0 0 0 0 0

-5.000V

1

CMOS LOGIC INTERFACE CIRCUIT

OUTPUT

INPUTS';;'; V+ LESS
07 VOLTS

IN4148

INTERFACING WITH CMOS LOGIC

CMOS DR IVING
DEVICE

The DAC-0610gic input stages require about 11'A and are
capable of operation with inputs between -5 volts and V+ less
0.7 volt. This wide input voltage range allows direct CMOS
interfacing in most applications, the exception being where

vss

11-22

1/86, Rev. A

DAC-08
8-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER
(UNIVERSAL DIGITAL LOGIC INTERFACE)
Pn:ClSlOll MOl1olithlcs Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•

interface to all popular logic families with full noise immunity
is provided by the high swing, adjustable threshold logic
input.

Fait Sattllng Output Current ..................... 85nl
Full-Scale Current Prematched to ± 1 LSB
Direct Interface to TTL, CMOS, ECL, HTL, PMOS
Nonlinearity to ,0.1% Maximum Over Temperature Range
High Output Impedance and
Compliance ....................••..•.. -10V to +18V
Complementary Current Outputl
Wide Range Multiplying Capability ..• 1MHz Bandwidth
Low FS Current Drift ......•.••..••.•...•• ±10ppm/oC
Wide Power Supply Range •••..•..•.•. ±4.5V to ±18V
Low Power Conlumptlon •...•.......... 33mW @ ±5V
Low COlt

High voltage compliance complementary current outputs are
provided, increasing versatility and enabling differential
operation to effectively double the peak-to-peak output
swing. In many applications, the outputs can be directly
converted to voltage without the need for an external op amp.
All DAC-08 series models guarantee full8-bit monotonicity,
and nonlinearities as tight as ±0.1 % over the entire operating
temperature range are available. Device performance is
essentially unchanged over the ±4.5 to ± 18V power supply
range, with 33mW power consumption attainable at ±5V
supplies.

GENERAL DESCRIPTION
The DAC-08 series of 8-bit monolithic digital-to-analog
converters provide very high-speed performance coupled
with low cost and outstanding applications flexibility.

The compact size and low power consumption make the
DAC-08 attractive for portable and military/aerospace applications; devices processed to MIL-STD-883, Level Bare
available.

Advanced circuit design achieves 85ns settling times with
very low "glitch" energy and at low power consumption.
Monotonic multiplying performance is attained over a wide
20 to 1 reference current range. Matching to within 1 LSB
between reference and full-scale currents eliminates the
need for full-scale trimming in most applications. Direct

DAC-08 applications include 8-bit, 1~s AID converters, servo
motor and pen drivers, waveform generators, audio encoders
and attenuators, analog meter drivers, programmable power
supplies, CRT display drivers, high-speed modems and other
applications where low cost, high speed and complete input!
output versatility are required.

EQUIVALENT CIRCUIT

I
4'" lOUT
2

... -

lOUT

VREF {o-+)--'-'14'+-_.--+-------,

1st-_--II .......
VREF {-o-)--'-'

11-23

----

- ------------

1/86, Rev. A

-----------tlfMD

DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
Operating Temperature

Logic Inputs ••••••••••••••••••••••••• V- to V- plus 36V
VLC ••••••••••••••••••••••.•••••••••••••••••••• V-to V+
Analog Current Outputs (at Vs- = 15V) ••••••••••• 4.25mA

° ......................
°

DAC-OBAO,
-55°C to +125°C
DAC-OBHO, EO, CO, HP, EP, CP •••••••• O°C to +70°C
DICE Junction Temperature (Ti) ••••••• -65°C to +150°C
Storage Temperature
Package •••••••• -65° C to 150° C
Storage Temperature P Package ••••••• -65° C to + 125° C
Power Dissipation
• • • • • • • • • • • • • • • • • • • • • • • • • . • •• 500mW
Derate above 100°C •••••••••••••••••••••••• 10mW/oC
Lead Temperature (Soldering, 60 sec) ••••••••••••• 300°C
V+ Supply to V- Supply.. • • • • • • • • • • • • ••• •• . • •• • • • •• 36V

ELECTRICAL CHARACTERISTICS at Vs =
+ 70° C for DAC-OBC, E

Reference Input (V14 to VIS) •••••••••••••••••••• V- to V+
Reference Input Differential Voltage
(V14 to VIS) ••••••••••••••••••••••••••••••••••••• ±1BV
Reference Input Current (114) ••••••••••••••••••••• 5.0mA
NOTE: Absolute ratings apply to both DICE and packaged parts, unless
otherwise noted.

±15V, IREF= 2.0mA, -55°C::> TA::> +125°C for DAC-OB/OBA, O°C::> TA::>

& H, unless otherwise noted. Output characteristics refer to both lOUT and lOUT.

DAC-08A/H
PARAMETER

SYMBOL

CONDITIONS

MIN

Resolution

8

Monotonlclty

8

TVP

MIN

TVP

MIN

TVP

MAX

UNITS

Bits
±0.1

ts

DAC-08C

MAX

Bits

Nonlinearity
Settling Time

DAC-08/E

MAX

±0.19

±0.39

%FS

To ± 1/2 LSB. all bits switched
ON or OFF. TA=25"C. (Note)

85

135

85

150

85

150

ns

TA =25"C
(Note)

35
35

80
80

35
35

80
80

35
35

80
80

ns

±10

±50

±10

±80
±50

±10

±80

Propagation Delay

Each bit
All bits swltchsd
Full-Scale Tempco
(Note)

tpLH

tpHL
TCI FS

Output Voltage Compliance
(True Compliance)

DAC-D8E
Full-Scale current change

Vee

< 1/2

'FR4

VREF = 10.000V
R'4' R,s = 5.0ookO
TA = +25"C

Full Range Symmetry

'FRS

IFR4 -I FR2

Zero-Scale Current

Izs

Full Range Current

lOR,
Output Current Range

IOR 2

Output Current Noise

LSB. ROUT> 2OMO typical

-10

1.984

+18

-10

1992

2.000

1.94

±0.5

±4

01
R'4' R,s =5.0001<0
VREF=+15.0V. V-=-10V
VREF = +25 OV, V-=-12V

+18

-10

1.99

2.04

1.94

±1

±8

02

+18

Volts

1.99

2.04

mA

±2

±16

pA
pA

0.2

21

2.1

2.1

4.2

4.2

4.2

ppm/"C

mA
25

'REF =2mA

25

25

nA

Logic Input Levels

Logic "0"
LogiC Input "1"

VIL
V,L

0.6

VLC=OV

LogiC "a"
Logic Input "I"

'lL
'IH

VLC=OV
VIN = -10V to +0.8V
Y,N = 2.0V to 18V

Logic Input Current

-2
0.002

0.8

-10
10

-2
0.002

0.8

-10
10

-2
0.002

VOlts

-10
10

pA

LogiC Input Swing

VIS

V-=-15V

-10

+16

-10

+18

-10

+18

Volts

LogiC Threshold Range

VTHR

Vs = ±15V. (Note)

-10

+13.5

-10

+135

-10

+135

Volts

Reference Bias Current

I,.

-3

pA

Reference Input Slew Rate

Power Supply Sensitivity

dl/dt

PSSI FS+
PSSIFS-

-1
REa =2oo0
RL = 1000
Cc=OpF

-3

-1

-3

-1

See fa.t pulsed
4

ref. info.

mAl",s

follOWing (Note)

V+ = 4.5V to 18V
V- = -4 5V to -IBV
'REF= lamA

-±0.0003
- ±0002

±0.01
±0.01

-±0.0003
- ±0.OO2

±001
±OOI

-±O 0003
- ±O002

±0.01 %<1la/%<1V+
±O.OI %41a1'1b<1V-

NOTE: Guaranteed by design

11·24

1/86, Rev. A

----------I~ DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

ELECTRICAL CHARACTERISTICS at Vs = ±15V, IREF= 2.0mA, -55°C:::; TA:::; +125°C for DAC-08/0BA, O°C:::; T A :::;
+ 70° C for DAC-OBC, E & H, unless otherwise noted. Output characteristics refer to both lOUT and lOUT. (Continued)
DAC-OSA/H
PARAMETER

Power Supply Current

MIN

MAX

MAX

23
-4.3

TYP

MAX

3.8
-58

23
-4.3

38
-58

23
-43

38
-58

+5V. -15V. 'REF ~ 2 OmA

24
-64

3.8
-7.8

24
-6.4

3.8
-78

2.4
-6.4

3.8
-78

±15V, IREF ~ 2 OmA

2.5
-6.5

38
-78

25
-65

3.8
-78

2.5
-6.5

38
-78

= 10mA

33
lOB

48

33

4B

136

33
103

48

+5V. -15V, IREF ~ 2 OmA

136

±15V, IREF = 2.0mA

135

174

135

174

108
135

136
174

1+
1-

Vs = ±5V, IREF = 1 OmA

1+
11+

Vs

~

Vs

~

±5V, IREF

Pd

MIN

DAC-OSC

TYP

CONDITIONS

1-

Power DISSipation

DAC-OS/E

TYP

SYMBOL

MIN

UNITS

mA

mW

NOTE: Guaranteed by deSign.

ORDERING INFORMATIONt

PIN CONNECTIONS

16-PIN DUAL-IN-LINE PACKAGE
NL

OPERATING
TEMPERATURE
RANGE

HERMETIC

PLASTIC

01%

DACOSAO'
DACOSHO

DACOSHP

019%

DACOSO'
DACOSEO

DACOSEP

COM

039%

DACOSCO

DACOSCP

COM

LCC

MIL

COM
DACOSRC/SS3

MIL

For deVices processed In total compliance to M IL-STO-883, add /883 after
part number Consult factory for 883 data sheet
tAil commercial and industrial temperature range parts are available with
burn-In For ordenng information see 1986 Data Book, Section 2
*

16-PIN DUAL-IN-LiNE
PACKAGE

DAC-OBRC/SB3
20-LEAD LCC
(Re-Suffix)

I

11-25

1/86, Rev. A

-----------l~ D.C-08 8-BIT HIGH-SPEED MULTIPLYING OJ. CONVERTER

DICE CHARACTERISTICS

(125°C TESTED DICE AVAILABLE)

1. VLC

9.
10.
11.
12.
13.
14.
15.
16.

2. lOUT
3. V4.
5.
6.
7.
8.

DIE SIZE 0.085 X 0.062 Inch, 5,270 sq. mils
(2.159 X 1.575 mm, 3.4 sq. mm)

lOUT
BIT 1 (MSB)
BIT2
BIT3
BIT4

BIT5
BIT6
BIT 7
BIT 8 (lSB)
V+

VREF (+)
V REF (-)
COMP

For additional DICE information refer to
1986 Dala Book, Section 2.

WAFER TEST LIMITS at Vs = ±15V. IREF = 2.0mA, TA = 125° C for DAC-OSNT, DAC-OSGT devices; TA = 25°C for
DAC-OSG and DAC-OSGR devices, unless otherwise noted. Output characteristics apply to both lOUT and lOUT.
PARAMETER

DAC-OSNT
LIMIT

DAC-OSN
LIMIT

DAC-OSGT
LIMIT

8

8

8

8

8

8

8

Bits MIN

±Ol

±O.I

±0.19

±0.19

±0.39

%FSMAX

Full-Scale Current
Change < 112 LSB

+18
-10

+18
-10

+18
-10

+18
-10

+18
-10

Volts MAX
Volts MIN

VREF = 10.000V
R'4' R,s= 5.000kO

204
1.94

2.04
194

204
1.94

2.04
1.94

2.04
1.94

mAMAX
mAMIN

±8

±8

±8

±8

±16

p.AMAX

4

4

4

p.AMAX

SYMBOL CONDITIONS

Resolution
Monotonicity
Nonlinearity
Output Voltage
Compliance
Full-Scale Current

Vee
'FS4 or
'FS2

Full-Scale Symmetry

I FSS

Zero-Scale Current

I zs

2

DAC-OSG
LIMIT

DAC-OSN.

DAC-OSGR
LIMIT

UNITS
Bits MIN

V-=-IOV,
'FS1 0r

Output Current Range

V AEF

= +15V

V-=-12V,
VREF = +25V
'FS2

R'4' R,s = 5.000kO

2.1

21

2.1

21

2.1

mAMIN

42

42

4.2

4.2

4.2

mAMIN

08

08

08

0.8

V MAX

LogiC Input "0"

V,L

0.8

Logic Input "1"

V,H

2

I'H

VLe = OV
Y,N = -IOV to +0.8V
Y,N = 2.0V to 18V

V's

V-=-15V

Logic Input Current

LogiC "0"
Logic "1"
Logic Input SWing

I,L

Power Supply
Sensitivity

±IO
±10

±10
±IO

±IO
±IO

±IO
±IO

p.AMAX

+18
-10

+18
-10

+18
-10

+18
-10

+18
-10

V MAX
VMIN

-3

-3

-3

-3

-3

p.AMAX

0.01

0.01

001

om

0.01 %FS/%V MAX

Power Supply Current

Vs= ±15V
I REF ,; 2 OmA

3.8
-7.8

3.8
-7.8

3.8
-7.8

38
-78

3.8
-7.8

mAMAX

Power Dissipation

Vs= ±15V
I REF ,; 2.0mA

174

174

174

174

174

mWMAX

1'5
PSSI FS +
PSSI FS _
1+

Pd

VMIN

±10
±IO

Reference Bias
Current

2

V+ = 4.5V to 18V
V-=-45Vto-18V
IREF = I OmA

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

11-26

1/86, Rev. A

-----------l~ DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

TYPICAL ELECTRICAL CHARACTERISTICS

at

Vs = ±15V,

and IREF =

2.0mA,

unless otherwise noted. Output

characteristics apply to both lOUT and lOUT.

ALL GRADES
PARAMETER

SYMBOL CONDITIONS

Reference Input
Slew Rate

dlldt

Propagation Delay

t pLH , tpHL TA = 25"C, Any Bit

Settling Time

TYPICAL

UNITS
mA/p.s

To ± 1/2 LSB, All Bits
SWitched ON or OFF,
TA =25"C

ts

35

ns

85

ns

NOTE:
For DAC08NT & GT 25"C characteristics, see DAC08N & G characteristics
respectively.

VJ

~
~

~

~

BURN-IN CIRCUIT

PULSED REFERENCE OPERATION

0

\?+VREF

U

C2

l

(OPTIONAL RESISTOR
:::RREF FOR OFFSET INPUTS

RIN

U
0

+18V

1

.....l

REa"" 14

~

......

200n

ovJL..

Rp

TYPICAL VALUES
RIN = SkD.

0
~

+VIN = 10V

OAC·OS

~......

C)

......

FAST PULSED
REFERENCE OPERATION

Q

C3

.0

2SV--

-lav MIN

Rl = 9kD.
Cl =OOOl .. F
C2, C3 = 0 01!1F

a 5V-05mAlOUT

-2 SmA-

REa"" 200n
RL = lOon

Cc

=

200NSEC/DIV1SION

a

TRUE AND COMPLEMENTARY
OUTPUT OPERATION

.

FULL-SCALE SETTLING TIME
LSB SWITCHING

ALL BITS SWITCHED ON

.."

24V-

24VLOGIC INPUT

lOUT

OmA -

1 OOlA'

04VOV-

lOrnA'

8pA0-

I

04V-

I

OUTPUT -1/2LSBSETTLING
0-

!
I
I

lOUT

+1/2LSB-

I
\
~

l(}oW.l

(000010000)

IREF = 2mA

(111111111)

5ONSEC/OIVISION

SETTLING TIME FIXTURE
IFS = ZmA, RL '" lkfl
1/2LSB

11-27

~

..

~.-

~------

50NSEC/OIVISION

= 4j.tA

1/86, Rev. A

-----------11fM!} DAC-08 8-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER
TYPICAL PERFORMANCE CHARACTERISTICS
FULL-SCALE CURRENT VI
REFERENCE CURRENT
6.0

LSB PROPAGATION
DELAYVIIFS
500

LIMIT fOR

V-C>'6V

TA" TMIN TO TMAX
ALL BITS "HIGH"

'0

8

5.0

I

o
~

is

I

I

0.05
002
0'
0.6
20
10
0.01
0.05
0.2
1,0
5.0
IFS. OUTPUT FULL SCALE CURRENT (mAl

~

cc·

1.
-8

~

1\,

1SpF, VIN ·2.OVp-p

,
~

CENTERED AT +1,OV LARGE
LARGE SIGNAL
CC'" 16pF. VIN '" 60mVp-p
CENTERED AT +200mV
SMALL SIGNAL

0.2

0.5

1,0

~

2.0

5,0

10

FREQUENCY (MHz)

LOGIC INPUT CURRENT VI
INPUT VOLTAGE

ALL BITS ON

~2

0

VTH - VLC VI TEMPERATURE

,00

2.0

80

'.8

TA = TMIN TO TMAX

3.6

:li

I

ov

-2

-'0 2.
-'2
-'4 01

'LSB-mlli

4.0

Cj

~
a:

1LS8-7.~

1\

1,0
2.0
3.0
4.0
'REF, REFERENCE CURRENT (mA)

VRtS"

~-4

t-

REFERENCE AMP
COMMON-MODE RANGE

12.8
.

2

~

LIMIT FOR

3.2

•

~

I!:

v- - -5V

/

iii

:s

1/
1/

R'4= R'6='';'

RL '" 6000
ALL BITS "ON"

400

V

'.0

REFERENCE INPUT
FREQUENCY RESPONSE

t-- Noie.

pcislTlV~ COJMON ~ODE1_

RANGE IS ALWAYS (V+I -'.5V

I I

2.4

V-= -f6V

2.0

I

1.6

II

.""'"

'REF'" 2mA

!

J .1J

'.2
0.8

,

~

20

-14

-10

-8

-2

10

14

18

""5. REFERENCE COMMON MOOE VOLTAGE (VOLTS)

OUTPUT CURRENT VI
OUTPUT VOLTAGE (OUTPUT
VOLTAGE COMPLIANCE)
4.0

ALL SITS ON

3.6
T A'" TMIN TO TMAX

1.2

~

............

r--..

>"

Il

'REF - O~imA

0.4

............

~O.8

40

§'"

,

'REF - 1rnA

I

~0

~ 60

V+ .. +1SV

V---6V

<"

'ffi"

-12.0 -8,0

-40

l

0.4

4.0

80

120

o

'60

LOGIC INPUT VOLTAGE (VOLTS)

OUTPUT VOLTAGE COMPLIANCE
V8 TEMPERATURE

+50
+100
TEMPERATURE (OCI

+'60

BIT TRANSFER
CHARACTERISTICS

+28.0 I " ' " " ' T " - - T " " - - - r - - " ' T " - - T " " '

'.8

NOTE B1 THROUGH lSI

+24.01--+--+_---+--+--+_--1

'.S

CHARACTERISTICS SITS ARE FULLY SWITCHED, WITH
LESS THAN 1/2 LIB ERROR, AT LESS THAN ±100mV
FROM ACTUAL THRESHOLD THESE SWITCHING

HAV~

IDENTICAL TRANSFER

POINTS ARE GUARANTEED TO LIE BETWEEN 0.8 AND

32

20 VOLTS OVER THE OPERATING TEMPERATURE

RANGE IVLe .. OOVI

v- '" -16V

I

II

I~EF

B'

IRE! - 2.0!.,.

'REF" 2mA

V- '" -5V

FOR OTHER V- OR 'REF, SEe
OUTPUT CURRENT VS OUTPUT

l'mA

82

r---r-

VOLTAGE CURVE

I

I

04
-14

-...---1,...,....-+--1

J

0.2

'REF" O.2mA
-10

-6

-2

2

6

10

14

18

v
V

-j

--6V

'sv

-12.0 -8.0 --4.0

B4

85

B3

\
0

4.0

8.0

12.0 16.0

lOGIC INPUT VOLTAGE (VOLTS)

DUPUT VOLTAGE (VOLTS)

11-28

1/86, Rev. A

----------1~ DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

TYPICAL PERFORMANCE CHARACTERISTICS
POWER SUPPLY
CURRENT VB V+
'00

POWER SUPPLY
CURRENT VB V10.0

AL~ Blis "H:GH'·I OR \o~"

9.0

1

8. 0

~ 7.0

r-

1-

~

3.0

~

1+ -

;:

a:

I I I I

'.0

'0

o

o
2.0 4.0 6 a 8.0 10012.0 140 16.0 18.0 20,0
V+, POSITIVE POWER SUPPLY (Vdc)

r- r-- v_I

~ 4.0

a:

~ 3.0

-40

-2.0

-6.0

--80
-12.0
-16.0
-20,0
-10.0
-14.0
-lB.O

r,_- r-

1+-

r-

I
I

'.0
o

+--

r- f---- V+ ~ +15V

.: 2.0

1+

-'5V

I

il:

1- WITH IREF = O.2mA _

=

IREF = 2.0mA

"

:: 5.0

1- WITH 'REF = lmA

~ 30
.: 20

A~L BITS "~'GH" OR I"LOW"
I
I

~ 6.0

4,0

2 20

o

t~ 7.0

1- WITH IREF = 2mA -

~ 5.0

a.. 4.0

18.0

II

7.0

a

ia 5.0

9.0

I I

8. 0

~ 6.0

~ 6.0

i>l
ffi

ffi

10,0

BI~S Mlv B~ "HI~H" ~R "~ow.!

9.0

1

POWER SUPPLY CURRENT
VB TEMPERATURE

o

-50

+50
+100
TEMPERATURE ('Ie)

+150

V-, NEGATIVE POWER SUPPLY (Vdc)

BASIC CONNECTIONS

BASIC POSITIVE REFERENCE OPERATION

ACCOMODATING BIPOLAR REFERENCES
MSB
LSB
81 B2B3 B4 B5 B6 B7 B8
+VREF

RREF

~

VIN-%-

~ 'REF

,.

RIN

RREF
(R14j

R'5

'5
VREF = +10.000V
RREF = 5.000k
R15 = RREF
Cc = O.01,uF
VLC = OV (GROUND)

'REF ~ PEAK NEGATIVE SWING OF liN
+VREF

RREF "" R15

~_

O.'~F

,.

RREF

I

R'5
(OPTIONAL)

VIN~____

_:':Y.B..E..E

15

v-

255

J

V+

VLC

'FR - RREF x""2OG'
10 + TO = IFR FOR

1-_ _ _ _ _ _ _- '

HIGH INPUT

0 l~F

IMPEDANCE

ALL LOGIC STATES

+VREF MUST BE ABOVE PEAK POSITIVE SWING OF VIN

BASIC UNIPOLAR NEGATIVE OPERATION

MSB

Bl

LSB

81 82 8384 85 86 87 88

IREF =
2.000mA

EO

,.
10

2

B2

B3

B4

B5

B6

B7

B8 lemA IOmA
1.992

FULL RANGE
HALF-SCALE +LSB

0

HALF-SCALE

0

HALF-SCALE -LSB

0

ZERO-SCALE + LSB

0

0

ZERO-SCALE

0

0

11-29

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

Eo

Eo

0.000

-9.960 - .000

1.008

0.984

-5.040 -4.920

1.000

0.992

-5.000 -4.960

0.992

1.000

-4.960 -5.000

0.008

1.984

-0.040 -9.920

0.000

1992

0.000 -9.960

1/86, Rev. A

II

-----------1~ DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

BASIC CONNECTIONS
BASIC BIPOLAR OUTPUT OPERATION

11 12 13 14 15 B8 17 18

+10.00OV
MSB

LSB

POS. FULL RANGE -LSB

81 B2 B3 B4 B6 B8 B7 B8

10.000kO

IREF(+I =

10.

,.

2.01lOmA

1O.OOOkn

2

RECOMMENDED FULL-SCALE ADJUSTMENT CIRCUIT

Eo

0 - 9.840 + 9.920

ZERO-SCALE +LSB

0

0

0

0

0

0

1 - 0.080 + 0.180

ZERO-SCALE

0

0

0

0

0

0

0

0

1

1

1

1

1

1

+ 0.080

NEG. FULL-SCALE +LSB

0

0

0

0

0

0

0

+ 9.920 - 9.840

NEG. FULL-SCALE

0

0

0

0

0

0

0

0 + 10.000 - 9.920

ZERO-SCALE -LSB

i0

Eo

1 - 9.920 +10.000

POS. FULL RANGE

0.000 + 0.080
0.000

BASIC NEGATIVE REFERENCE OPERATION

LOWT.C.
VREF

4.6kIl

"ov
39kn

.,"

RREF

.L.- "

,-----A/IIV---t ' •

IREF(+) "'=' 2mA
~1V

10k!!

APP~OX

POT

.".

-: -VREF

,.

R'6

,.
NOTE. RREF SETS IFS: R16 IS FOR
BIAS CURRENT CANCELLATION

OFFSET BINARY OPERATION
10ka

MSB
LSB
81 B2 B3 B4 B5 B8 87 B8

."""
11 12 13 14 15 18 17 18

6000kn

POS. FULL RANGE
ZERO-SCALE

• Ok!!

.".

.".

+15V -16V

1

1

1

1

1
0

0

0

0

0

0

0

NEG. FULL-SCALE +1 LSB

0

0

0

0

0

0

0

NEG. FULL-SCALE

0

0

0

0

0

0

0

Eo
+4960
0.000
-4.980

0

-5.000

-15V

11-30

1/86, Rev. A

----------I~ DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

BASIC CONNECTIONS
NEGATIVE LOW IMPEDANCE OUTPUT OPERATION

POSITIVE LOW IMPEDANCE OUTPUT OPERATION

IFR '"

~

tREF

FOR COMPLEMENTARY OUTPUT (OPERATION AS A NEGATIVE LOGIC OAC),
CONNECT INVERTING INPUT OF OP-AMP TO
(PIN 2); CONNECT 10 (PIN 4)
TO GROUND.

TO

FOR COMPLEMENTARY OUTPUT {OPERATION AS A NEGATIVE LOGIC OAC},
CONNECT NON INVERTING INPUT OF OP-AMP TO 10 (PIN 2),CQNNECT 10 (PIN
4) TO GROUND

INTERFACING WITH VARIOUS LOGIC FAMILIES
CMOS, HTl, NMOS

Eel
VTH = VLC +1 4V
+15V CMOS
VTH=+76V

TTL, DTL

V+

+1SV

VTH '" +1.4V

20M2

13k"

9.1kn

Vee

3k"

62k.\1
TO PIN 1

'----+-<: Vee

20kn

TO PIN 1

'----t---o Vee
R3

62kn

II

-52V
TEMPERATURE COMPENSATING VLC CIRCUITS

APPLICATIONS INFORMATION

15. The voltage at pin 14 is equal to and tracks the voltage at
pin 15 due to the high gain of the internal reference amplifier.
R 15 (nominally equal to R 14 ) is used to cancel bias current
errors; R15 may be eliminated with only a minor increase in
error.

REFERENCE AMPLIFIER SET-UP
The DAC-08 is a multiplying D/A converter in which the
output current is the product of a digital number and the
input reference current. The reference current may be fixed
or may vary from nearly zero to +4.0mA. The full-scale output
current is a linear function of the reference current and is
given by:

Bipolar references may be accomodated by offsetting VREF
or pin 15. The negative common-mode range of the reference
amplifier is given by: VCM- = V- plus (lREF x 1kO) plus 2.5V.
The positive common-mode range is V+ less 1.5V.

255
I FR = 256 x I REF, where I REF = 114.

When a DC reference is used, a reference bypass capacitor is
recommended. A 5.0V TTL logic supply is not recommended
as a reference. if a regulated power supply is used as a
reference, R14 should be split into two resistors with the
junction bypassed to ground with a O.1I'F capacitor.

In positive reference applications, an external positive reference voltage forces current through R14 into the VREF(+) terminal (pin 14) of the reference amplifier. Alternatively, a negative reference may be applied to VREF(-) at pin 15; reference
current flows from ground through R14 into VREF(+) as in the
positive reference case. This negative reference connection
has the advantage of a very high impedance presented at pin

For most applications the tight relationship between IREFand
IFs will eliminate the need for trimming I REF. If required,
full-scale trimming may be accomplished by adjusting the
value of R 14, or by using a potentiometer for R 14. An improved

11-31

1/86, Rev. A

-----------l~ DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

ANALOG OUTPUT CURRENTS
Both true and complemented output sink currents are provided where 10 + fO = IFS. Current appears at the "true" (10)
output when a "1" (logic high) is applied to each logic input.
As the binary count increases, the sink current at pin 4
increases proportionally, in the fashion of a "positive logic"
DI A converter. When a "0" is applied to any input bit, that
current is turned off at pin 4 and turned on at pin 2. A
decreasing logic count increases fQ as in a negative or
inverted logic D/A converter. Both outputs may be used
simultaneously. If one of the outputs is not required it must
be connected to ground orto a point capable of sourCing ' FS;
do not leave an unused output pin open.

method of full-scale trimming which eliminates potentiometer T.C. effects is shown in the recommended full-scale
adjustment circuit.
Using lower values of reference current reduces negative
power supply current and increases reference amplifier negative common-mode range. The recommended range for
operation with a DC reference current is +0.2mA to +4.0mA.
REFERENCE AMPLIFIER COMPENSATION FOR
MULTIPLYING APPLICATIONS
AC reference applications will require the reference amplifier
to be compensated using a capacitor from pin 16 to V-. The
value of this capacitor depends on the impedance presented
to pin 14: for R14 values of 1.0, 2.S and S.OkO, minimum values
of Cc are 1S, 37, and 7SpF. larger values of R14 require
proportionately increased values of Cc for proper phase
margin, such that the ratio of Cc(pF) to R14 (kO) = 1S.

Both outputs have an extremely wide voltage compliance
enabling fast direct current-to-voltage conversion through a
resistor tied to ground or other voltage source. Positive
compliance is 36V above V- and is independent of the
positive supply. Negative compliance is given by V- plus
(I REF X 1kO) plus 2.SV.

For fastest response to a pulse, low values of R14 enabling
small Cc values should be used. If pin 14 is driven by a high
impedance such as a transistor current source, none of the
above values will suffice and the amplifier must be heavily
compensated which will decrease overall bandwidth and
slew rate. For R14 = 1kO and Cc = 1SpF, the reference amplifier slews at 4mAl/ls enabling a transition from 'REF = 0 to
I REF = 2mA in SOOns.

The dual outputs enable double the usual peak-to-peak load
swing when driving loads in quasi-differential fashion. This
feature is especially useful in cable driving, CRT deflection
and in other balanced applications such as driving centertapped coils and transformers.
POWER SUPPLIES
The DAC-08 operates over a wide range of power supply
voltages from a total supply of 9V to 36V. When operating at
supplies of ±SV or less, IREF s; 1mA is recommended. low
reference current operation decreases power consumption
and increases negative compliance, reference amplifier negative common-mode range, negative logic input range, and
negative logic threshold range; consult the various figures
for guidance. For example, operation at -4.SV with IREF =
2mA is not recommended because negative output compliance would be reduced to near zero. Operation from lower
supplies is pOSSible, however at least 8V total must be applied
to insure turn-on of the internal bias network.

Operation with pulse inputs to the reference amplifier may be
accomodated by an alternate compensation scheme. This
technique provides lowest full-scale transition times. An
internal clamp allows quick recovery of the reference amplifier from a cutoff (lREF = 0) condition. Full-scale transition
(Oto 2mA) occurs in 120ns when the equivalent impedance at
pin 14 is 2000 and Cc=O. This yields a reference slew rate of
16mAl/ls which is relatively independent of RIN and VIN
values.
LOGIC INPUTS

The DAC-08 design incorporates a unique logic input circuit
which enables direct interface to all popular logic families
and provides maximum noise immunity. This feature is made
possible by the large input swing capability, 2/lA logic input
current and completely adjustable logic threshold voltage.
For V-= -15V, the logic inputs may swing between -10V and
+ 18V. This enables direct interface with + 1SV CMOS logic,
even when the DAC-08 is powered from a +SV supply. Minimum input logic swing and minimum logic threshold voltage
are given by: V- plus (IREF x 1kO) plus 2.SV. The logic threshold may be adjusted over a wide range by placing an
appropriate voltage at the logic threshold control pin (pin 1,
VLC ). The appropriate graph shows the relationship between
VLC and VTH over the temperature range, with VTH nominally
1.4 above VLC. For TTL and DTl interface, simply ground pin
1. When interfacing ECl, an IREF= 1mA is recommended. For
interfacing other logic families, see preceding page. For general set-up of the logic control circuit, it should be noted that
pin 1 will source 100/lA typical; external circuitry should be
designed to accommodate this current.
Fastest settling times are obtained when pin 1 sees a low
impedance. If pin 1 is connected to a 1kO divider, for example, it should be bypassed to ground by a 0.01 /IF capaCitor.

Symmetrical supplies are not required, as the DAC-08 is
quite insensitive to variations in supply voltage. Battery
operation is feasible as no ground connection is required:
however, an artificial ground may be used to insure logic
swings, etc. remain between acceptable limits.
Power consumption may be calculated as follows:
Pd = (1+) (V+) + (1-) (V-). A useful feature of the DAC-08
design is that supply current is constant and independent of
input logic states; this is useful in cryptographic applications
and further serves to reduce the size of the power supply
bypass capacitors.
TEMPERATURE PERFORMANCE
The nonlinearity and monotonicity specifications of the
DAC-08 are guaranteed to apply over the entire rated operating temperature range. Full-scale output current drift is low,
typically ± 10ppm/o C, with zero-scale output current and
drift essentially negligible compared to 1/2 lSB.

The temperature coefficient of the reference resistor R14
should match and track that of the output resistor for min11-32

1/86, Rev. A

----------l!mD

DAc-oa a-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

imum overall full-scale drift. Settling times of the DAC-08
decrease approximately 10%at-55° C; at +1250 C an increase
of about 15% is typical.

to logic input amplitude and rise and fall times, due to the
high gain of the logic switches. Settling time also remains
essentially constant for IREF values. The principal advantage
of higher I REF values lies in the ability to attain a given output
level with lower load resistors, thus reducing the output RC
time constant.

The reference amplifier must be compensated by using a
capacitor from pin 16 to V-. For fixed reference operation, a
0.01 ",F capacitor is recommended. For variable reference
applications, see previous section entitled "Reference Amplifier Compensation for Multiplying Applications".

Measurement of settling time requires the ability to accurately resolve ±4",A, therefore a lkO load is needed to
provide adequate drive for most oscilloscopes. The settling
time fixture shown in schematic labelled "Settling Time
Measurement" uses a cascode design to permit driving a 1kO
load with less than 5pF of parasitic capacitance at the
measurement node. At IREF values of less than 1.0mA,
excessive RC damping of the output is difficult to prevent
while maintaining adequate sensitivity. However, the major
carry from 01111111 to 10000000 provides an accurate
indicator of settling time. This code change does not require
the normal 6.2 time constants to settle to within ±0.2% of the
final value, and thus settling times may be observed at lower
values of I REF.

MULTIPLYING OPERATION
The DAC-08 provides excellent multiplying performance
with an extremely linear relationship between I FS and IREF
over a range of 4mA to 4",A. Monotonic operation is maintained over a typical range of I REF from 100",A to 4.0mA.
SETTLING TIME
The DAC-08 is capable of extremely fast settling times,
typically 85ns at IREF = 2.0mA. Judicious circuit design and
careful board layout must be employed to obtain full performance potential during testing and application. The logic
switch design enables propagation delays of only 35ns for
each ofthe8 bits. Settling timeto within 1/2 LSB of the LSB is
therefore 35ns, with each progressively larger bit taking
successively longer. The MSB settles in 85ns, thus determining the overall settling time of 85ns. Settling to 6-bit
accuracy requires about 65 to 70ns. The output capacitance
of the DAC-08 including the package is approximately 15pF,
therefore the output RC time constant dominates settling
time if RL > 5000.

DAC-08 switching transients or "glitches" are very low and
may be further reduced by small capacitive loads at the
output at a minor sacrifice in settling time.
Fastest operation can be obtained by using short leads,
minimizing output capacitance and load resistor values, and
by adequate bypassing at the supply, reference and VLc
terminals. Supplies do not require large electrolytic bypass
capacitors as the supply current drain is independent of input
logic states; 0.1 ",F capacitors at the supply pins provide full
transient protection.

Settling time and propagation delay are relatively insensitive
SETTLING TIME MEASUREMENT

FOR TURN-ON, VL '" 2 7V
fOR TURN-OfF, VL = 07V

VL

II

'5V

lkH

+04V
VOUT lX

~

PROBE

OV

---I"""
-04V
100kn

R1S

2kn

15kr:!

15

T01JJF

-15V

11-33

-=-

1

1/86, Rev. A

DAC-I0
lO-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER
(UNIVERSAL DIGITAL LOGIC INTERFACE)
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•

All DAC-10 series models guarantee full 10-bit monotonicity,
and nonlinearities as tight as ±O.05% over the entire operating temperature range are available. Device performance is
essentially unchanged over the ± 18V power supply range,
with 85mW power consumption attainable at lower supplies.

Fast Settling .................................... 85ns
low Full-Scale Drift ....................... 10ppm/o C
Nonlinearity to 0.05% Max Over Temp Range
Complementary Current Outputs ............ 0 to 4mA
Wide Range Multiplying Capability ... 1MHz Bandwidth
Wide Power Supply Range .. +5, -7.5 Min to ±18V Max
Direct Interface to TTL, CMOS, ECl, PMOS, NMOS

A highly stable, unique trim method is used, which selectively
shorts zener diodes, to provide 1/2 LSB full-scale accuracy
without the need for laser trimming.
Single-chip reliability coupled with low cost and outstanding
flexibility make the DAC-10 device an ideal building block for
AID converters, Data Acquisition systems, CRT display,
programmable test equipment, and other applications where
low power consumption, input/output versatility, and longterm stability are required.

ORDERING INFORMATlONt
18-PIN HERMETIC
DUAL-IN-LINE PACKAGE
MILITARY TEMP."
COMMERCIAL TEMP.

NL
LSB
±1/2
±1

DAC10BX'
DAC10CX'

DAC10FX
DAC10GX

PIN CONNECTIONS

* For devices processed In total compliance to Ml L-STD-883, add 1883 after
part number Consult factory for 883 data sheet
tAil commercial and Industrral temperature range parts are available with
burn-In For ordenng Information see 1986 Data Book, Section 2

GENERAL DESCRIPTION

18-PIN HERMETIC

The DAC-10 series of 10-bit monolithic multiplying digitalto-analog converters provide high-speed performance and
full-scale accuracy,

DUAl-IN-lINE PACKAGE
(X-Suffix)

Advanced circuit design achieves 85ns settling times with
very low 'glitch' energy and low power consumption, Direct
interface to all-popular logic families with full noise immunity
is provided by the high-swing, adjustable-threshold logiC
inputs.

SIMPLIFIED SCHEMATIC

VREFI-)

17

0--+---1

CQMP

v-

Manufactured under one or more of the following patents

4,055,770,4,056,740,4,092,639

11-34

1/86, Rev. A

----------I~ DAC-10 10-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
Logic Inputs •••••••••••••••••.••••••• V- to V- plus 36V
VLC ••••••••••••••••••••••••••••••••••••••••••• V-to V+
Analog Current Outputs •••••••••••••••••• + 18V to -18V
Reference Inputs (V 16 to V17) ••.•••••••••••••••• V- to V+
Reference Input Differential Voltage
(V16 to V17) •••••••••••••••••••••••••••••••••••.• ±18V
Reference Input Current (1 16) ••••••••••••••••••••• 2.SmA

Operating Temperature
DAC-10BX, CX ..................... -SS·C to +12S·C
DAC-10FX, GX ••••••••••••••••••••••••• O· C to + 70· C
DICE Junction Temperature (TI ) .•••••. -6S·C to +1S0·C
Storage Temperature ••••••••••••••••• -6S· C to + 1S0· C
Power Dissipation • • • • • • • • • • • • • . • • • • • • • • • • • • • •• SOOmW
Derate above 100· C •••••••••••••••••••••••••• 10mW/· C
Lead Temperature (Soldering, 60 sec) ••••••••••••• 300· C
V+ Supply to V- Supply. • • • • •• • • • • • • • • • • • • • •• • • • • •• 36V

NOTE: Absolute rallngs apply to both DICE and packaged parts, unless
otherwise noted

ELECTRICAL CHARACTERISTICS at Vs= ±1SV; IREF= 2mA; -SS·C:5 TA:512S·C for DAC-10B and DAC-10C,
O·C:5 TA:5 70·C for DAC-10F and G, unless otherwise noted. Output characteristics apply to both lOUT and lOUT.
DAC-10B/F
PARAMETER

SYMBOL

CONDITIONS

MIN

Monotonicity

TYP

MAX

10

Nonlinearity

NL

0.3

DNL

0.3

Settling Time

ts

Output Capacitance

Co

Propagation Delay

tpLH
tpHL

All Bits Switched

Output Voltage
ComplIance

All Bits Switched ON or OFF
Settle to 0.05% of FS (See Note)

MIN

TVP

MAX

10

Differential Nonlinearity

65

~
~

DAC-10C/G

0.5

135

UNITS
Bits

0.6

LSB

07

LSB
150

65

ns

18

18

VOC

Full-Scale Current
<1 LSB
Change

50
50
-5.5
+10

50
50
-55
+10

Gain Tempco

TCI F•

(See Note)

±10

±25

±10

±50

ppm/DC

Full-Scale Symmetry

I FSS

IFA-I FA

0.1

4

0.1

4

"A

Zero-Scale Current

Izs

0.01

0.5

Full-Scale Current

IFA
DI/dt

3.996

4.072

"A
rnA

Reference Input Slew Rate
Reference Bias Current
Power Supply Sensitivity

Power Supply Current

Ie
PSSI F.+
PSSI F....
1+
11+
I-

Power Dissipation

Pd

Logic Input Levels

V,L
V,H

Logic Input Currents

I,L
I'H

(See Nota)

rouT.

3.960

0.01

0.5

3.996

4.032

3.920

6

SYMBOL

0.001
0.0012
2.3
-9
18
-59
231
85

4.5V " V+" 18V
-18V" V-" -10V
V.=±15V,IAEF=2mA
V. = +5V1-7.5V; IAEF= lmA
V.=+5V/-7.5V; 'AEF= lmA

0.01
0.01
4
-15
4
-9
276
107

VLC=O

2
10

VLC-O; 5V" V,N " 0.8V
2V"V ,N ,,18V
at Vs = ±1SV; IREF = 2m A; TA

-5
0.001

Differential Nonlinearity
Output Voltage
Compliance

DNL
VOC

Full-Scale Current

CONDITIONS

0.01
0.01

%~IFsI%~V

2.3
-9
1.8
-59
231
85

4
-15
4
-9
276
107

TVP

10

mW
V
"A

DAC-10G

MAX

MIN

3.978

TYP

MAX

3.996

0.6

+10

-5 -6/+15

LSB

0.7

4.014

3.956

UNITS
Bits

0.5

0.3
-5 -6/+18

Izs

10

10
0.3

Full-Scale Current <1 LSB
Change
VAEF = 10.000V,
R,. = R,s = 5.000kO

5
0.001

rnA

= 2S·C, unless otherwise noted. Output characteristics
MIN

Full-Scale Symmetry
Zero-Scale Current

"A

0.001
0.0012

0.8
2
10

10
NL

mAl"s
-3

0.8

Monotonicity
Nonlinearity

V

6

DAC-10B/C/F
PARAMETER

ns

3

ELECTRICAL CHARACTERISTICS
apply to both lOUT and

RL -5kO
RL =0

pF

3.996

LSB
+10

V

4.036

rnA

0.1

4

0.1

4

"A

0.01

0.5

0.01

0.5

pA

NOTE: Guaranteed by design

11-35

1/86, Rev. A

~

~

0

U

0
0

~
0

~

~

0
.....
Cl

II

----------I~ DAC·10 10·BIT HIGH·SPEED MULTIPLYING D/A CONVERTER

DICE CHARACTERISTICS

1. VLC (LOGIC)
THRESHOLD CONTROL

2. iO
3. V4. 10

5. B1 (MSB)
B2
B3
B4
B5

6.
7.
8.
9.

10.
11.
12.
13.
14.
15.
16.

B6
B7
B8
B9
B10 (LSB)

V+
VREF (+)

17. VREF (-)
18. COMPENSATION

For additional DICE Information refer 10
1986 Dala Book, Section 2.
DIE SIZE 0.086 X 0.090 Inch, 7740 sq. mils
(2.184 X 2.286 mm, 4.993 sq. mm)

WAFER TEST LIMITS at

Vs = ± 15V, IREF = 2m A, TA = 25°C, unless otherwise noted. Output characteristics refer to both

lOUT and lOUT·

PARAMETER

SYMBOL

CONDITIONS

Resolution

DAC·10G

LIMIT

LIMIT

UNITS

10

10

Bits MIN

10

10

Bits MIN

±0.5

±1

LSBMAX

True 1 LSB

+10
-5

+10
-5

V MAX
VMIN

IFS ±3.996 MA

±18

±40

/JAMAX

0.5

0.5

/JAMAX

Monotonicity
Nonlinearlly

NL

Output Voltage Compliance

Voe

Zero-Scale Current

Izs

All Bits OFF

Logic Input "1"

Output Current Range

DAC·10N

VIH

IIN= 100nA

Logic Input "0·'

VIL

VLe@Ground
IIN=-100/JA

Posilive Supply Current

1+

V+= 15V

Negative Supply Current

1-

V-=-15V

2

2

VMIN

0.8

0.8

V MAX

4

4

mAMAX

-15

-15

mAMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly melhods and normal yield loss, yield after packaging is nol
guaranteed for slandard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample 101 assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS

at Vs = ±15V, and IREF = 2mA, unless otherwise noted. Output

characteristics refer to both lOUT and lOUT·

PARAMETER
Settling Time
Gain Temperature
Coefficient (TCI

SYMBOL

CONDITIONS

ts

To ±1/2 LSB When Output is
Switched from 0 to FS
VAEF Tempco Excluded

DAC·10N

DAC·10G

TYPICAL

TYPICAL

UNITS

85

85

ns

±10

±10

ppm FS/oC

Output CapaCitance

18

18

pF

Output Resistance

10

10

MO

11·36

1/86, Rev. A

-----------l[fffi) DAC-10 10-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER
TYPICAL PERFORMANCE CHARACTERISTICS
OUTPUT CURRENT VI
OUTPUT VOLTAGE
(OUTPUT VOLTAGE COMPLIANCE)

TRUE AND COMPLEMENTARY
OUTPUT OPERATIONS

OUTPUT VOLTAGE
COMPLAINCE VI TEMPERATURE
+280r-,-----~--_,----~----_r_,

80
ALL BITS ON

+240 f-+---+---I~--+--+-I

72
OmA

TA" TMIN TO TMAX

lOUT

6.4

~

;t

5.6
E
;: 48

1.0mA

illa:

2.0mA

v-

= -15V

v-

= -10V

C.)

t~EF _1'mA

I-

~ 2.4
:>
01.6

(00000000)

("111111)

rJ::~ii'i~iili~+~

+120

IREF = 2mA

r

4.0

":> 3.2

lOUT

+200
+160

~

IR~F

0.6

-40
= Ol2mA

-80f-+--+---I~--+--+-I

IREF = 2mA
-120~~

-14 -10

-6

10

-2

14

____

-50

18

100
90

"<

E

I

1

80

BO

~ 70

~ 60

~ 60

1l

:>

~

50

~ 50

8::

40

~ 40

" 30

" 30

iil

~ 20

1.0
o

2.0

4,0

I

lmd_

o

-40
-20

V+, POSITIVE POWER SUPPLY (Vdc)

-80

I

I

-60

-laO

-160

-140

~~

+150

i:il

40

"~

30

~

20

1-

I
I
ALL BITS "HIGH" OR "LOW"

I
V+ =

l,sv

I.

I
I

10

1J I

-120

80

~ 60
u
~ 50
~

WITH IREF = 02mA

10

60 80 10012014,016018020,0

T

____

IREF = 20mA

~ 70

I I I

~

o

IREF =

r-

I-

1- WITH IREF = 02mA

2 20

'j- -

;;
!

I I I
t-i'- WITH
I I

~

+100

v- = -15V

90

1- WITH IREF = 2mA

~ 7.0

____

+50

100

J

BITS MAY BE "HIGH" OR "LOW"

ALL BITS "HIGH" OR "LOW"

90

~

POWER SUPPLY CURRENT
VI TEMPERATURE

POWER SUPPLY CURRENT VI V-

10.0

__

TEMPERATURE (Cel

OUTPUT VOLTAGE (VOLTS)

POWER SUPPLY CURRENT VI V+

~

-200

-50

-180

+50

+100

+150

TEMPERATURE (OCJ

V-, NEGATIVE POWER SUPPL V (Vdc)

BASIC CONNECTIONS
ACCOMMODATING BIPOLAR REFERENCES

BASIC POSITIVE REFERENCE OPERATION

+VREF
IFR"

+~:~~

It

~g~! It 2

10 + iO = IFR FOR
ALL LOGIC STATES

RREF
MSB
LSB
B1 B2 B3B4 B5 B6 B7 B8 B9 Bl0

~

VIN~

~ IREF
16

RIN
DAC-10

10

17

IREF---'
R17
IREF

RREF"'" R17

V+

VLC

VREF " +10 ODOV
RREF " 5DDOk
R15 = RREF
CC" O.D1.uF
VLC '" OV (GROUND)

~

~

PEAK NEGATIVE SWING OF liN

+VREF o--A,RR"E,,'_-,,'6'-j

...

R17
(OPTIONAL)

VIN~__

HIGH INPUT

OAC-l0
17

L-______________J

IMPEDANCE
+VREF MUST BE ABOVE PEAK POSITIVE SWING OF VIN

11-37

1/86, Rev_ A

II

-----------1~ DAC-10 10-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

BASIC NEGATIVE REFERENCE OPERATION

RECOMMENDED FULL-SCALE ADJUSTMENT CIRCUIT

LQWTC

~l~V o-- +VRE'
I
...

jOPTIONAL RESISTOR
OFFSET INPUTS

~RREF lFOR

,

,.

RL

REQ 5000.

Both outputs have an extremely wide voltage compliance
enabling fast direct current-to-voltage conversion through a
resistor tied to ground or other voltage source. Positive
compliance is 36V above V- and is independent of the
positive supply. Negative compliance is + 10V above V-.
The dual outputs enable double the usual peak-to-peak load
swing when driving loads in quasi-differential fashion. This
feature is especially useful in cable driving, CRT deflection
and in other balanced applications such as driving centertapped coils and transformers.

Settling time and propagation delay are relatively insensitive
to logiC input amplitude and rise and fall times, due to the
high gain of the logic switches. Settling time also remains
essentially constant for IREF values down to 1mA, with
gradual increases for lower IREF values lies in the ability to
attain a given output level with lower load resistors, thus
reducing the output RC time constant.

POWER SUPPLIES
The DAC-1O operates over a wide range of power supply
voltages from a total supply of 9V to 36V. When operating with
V-supplies of-10Vor less, IREF~ 1mA is recommended. Low
reference current operation decreases power consumption
and increases negative compliance, reference amplifier
negative common-mode range, negative logiC input range,
and negative logic threshold range; consult the various
figures for guidance. For example, operation at -9V with
IREF = 2mA is not recommended because negative output
compliance would be reduced to near zero. Operation from
lower supplies is possible, however at least 8V total must be
applied to insure turn-on of the internal bias network.

Measurement of settling time requires the ability to accurately resolve ± 2/o1A, therefore a 4kO load is needed to provide
adequate drive for most oscilloscopes. The settling time
fixture of schematic titled "Settling Time Measurement" uses
a cascode design to permit driving a 4kO load with less than
5pF of parasitic capaCitance at the measurement node. At
IREF values of less than 1mA, excessive RC damping of the
output is difficult to prevent while maintaining adequate
sensitivity. However, the major carry from 0111111111 to
1000000000 provides an accurate indicator of settling time.
This code change does not require the normal 6.2 time
constants to settle to within ±0.2% of the final value, and thus
settling times may be observed at lower values of IREF.

Symmetrical supplies are not required, as the DAC-10 isquite
insensitive to variations in supply voltage. Battery operation
is feasible as no ground connection is required; however, an
artificial ground may be used to insure logic swings, etc.
remain within acceptable limits.

DAC-10 switching transients or "glitches" are very low and
may be further reduced by small capacitive loads at the
output at a minor sacrifice in settling time.
Fastest operation can be obtained by using short leads,
minimizing output capacitance and load resistor values, and
by adequate bypassing at the supply, reference and VLC
terminals. Supplies do not require large electrolytiC bypass
capaCitors as the supply current drain is independent of input
logic states; 0.1/o1F capaCitors at the supply pins provide full
transient protection.

TEMPERATURE PERFORMANCE
The nonlinearity and monotonicity specifications of the
DAC-1O are guaranteed to apply over the entire rated operating temperature range. Full-scale output current drift is
tight, typically ±10ppm/oC, with zero-scale output current
and drift essentially negligible compared to 1/2 LSB.

11-41

-----_ _..

-----.-

1/86, Rev. A

~
~

~

ou
C)

o

~

~
,....;j

is
~

~
~

DAC-20
2-DIGIT BCD HIGH-SPEED MULTIPLYING D/A CONVERTER
(UNIVERSAL DIGITAL LOGIC INTERFACEI
Precision Monolithics Inc.

between reference and full-scale currents eliminates the
need for full scale trimming in most applications. Direct interface to all popular logic families with full noise immunity is
provided by the high swing, adjustable threshold logic
inputs.

FEATURES
•
•
•
•
•
•
•
•
•
•
•

Fast Settling Output Current ..................... 85n8
Full-Scale Current Prematched to ±O.3 lSB
Direct Interface to TTL, CMOS, ECl, PMOS, NMOS
Nonlinearity to ±1/2 lSB Maximum Over Temp.
High Output Impedance and Compliance -10V to +18V
Complementary Current Outputs
Wide Range Multiplying Capability ... 1MHz Bandwidth
low FS Current Drift .................... ± 10ppml~C
Wide Po_r Supply Range ............. ±4.5V to ±18V
low Po_r Consumption ............... 37mW @ ±5V
low Cost

Complementary current outputs with -10V to + 18V voltage
compliance enable resistive termination, a voltage output
without an external op amp.
Both DAC-20 models guarantee full 2-digit monotonicity,
some have nonlinearity as tight as ± 112 LSB over the entire
operating temperature range. Nonlinearity is unchanged
over the ±4.5V to ± 18V power supply range, with 37mW
power consumption attainable at ±5V supplies.

ORDERING INFORMATIONt
NL
LSB

±1/2

The compact size and low power consumption make the
DAC-20 attractive for portable applications.

16-PIN DUAL-IN-LiNE PACKAGE
COMMERCIAL TEMPERATURE RANGE
HERMETIC

PLASTIC

DAC20CQ

DAC20CP

DAC-20 applications include AID converters, audio attenuators, analog meter drivers, programmable power supplies,
high-speed modems and other applications where low cost,
high speed and complete input/output versatility are
required.

t All

commerCial and industrial temperature range parts are available with
burn-In For ordering information see 1986 Data Book, Section 2.

PIN CONNECTIONS
GENERAL DESCRIPTION

VLC

The DAC-20 series of 2-digit BCD monolithic multiplying
digital to analog converters provide very high-speed performance coupled with low cost and outstanding applications
flexibility.

16-PIN
DUAl-IN-LiNE
PACKAGE

Advanced circuit design achieves 85ns settling times with
very low "glitch" energy and at low power consumption.
Monotonic multiplying performance is attained over a wide 20
to 1 reference current range. Matching to within 1 lSB

EQUIVALENT CIRCUIT

.-

r-~-'---r~--~~~~--~~--~~~~----lr-oIOUT

Manufactured under one or more of the following patents: 4,055,773; 4,056,740; 4,002,639

11-42

1/86, Rev. A

---------f~ DAC-20 2-DIGIT BCD HIGH-SPEED MULTIPLYING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS

(TA = 25° C, unless otherwise noted)

Operating Temperature Range
DAC-20
CP ........................ 0° C to + 70° C
DICE Junction Temperature (TJ) ••••••• -65°C to +150°C
Storage Temperature Range

V+ Supply to V- Supply.. • • • • •• • • • • • • • • • • • • • • • ••••• 36V
Logic Inputs ......................... V- to V- plus 36V
VLC ••••••••••••••••••••••••••••••••••••••••••• V-to V+
Reference Inputs (V14, V15) ..................... V- to V+
Reference Input Differential Voltage (V14 to V15) •••• ±18V
Reference Input Current (114) ••••••••••••••••••••• 5.0mA

ca,

a Package

......................... -65°C to +150°C
P Package ......................... -65° C to + 125° C
Power Dissipation •••••••••••••••••••••••••••••• 500mW
Derate above 100° C •••••.•••••••••••••••••• 10mW/o C

NOTE: Absolute ratings apply to both DICE and packaged parts unless
otherwise noted.

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300° C

ELECTRICAL CHARACTERISTICS
characteristics refer to both lOUT and

at Vs= ±15V, IREF= 2.0mA, O°C:5 TA:5 70°C, unless otherwise noted. Output

~

!'OuT.
DAC-20C

PARAMETER

SYMBOL

CONDITIONS

MIN

TVP

MAX

UNITS

Resolution

BCD 0 to 99 steps

2

Digits

Monotonicity

BCD 99 steps

2

Digits

Nonlinearity

NL

0000 0000 to 1001 1001

Settling Time

ts

To ±1I2 LSB (±0.5% FS)
all bits sWitched ON or
OFF, TA = 25·C (Note 1)

85

150

ns

Propagation Delay
Each Bit
All bits sWitched

tpLH
tpHL

TA =25·C (Note 1)

35

60

ns

Full Tempco

TCI FS

(Note 1)

±10

±60

ppml"C

Output Voltage Compliance
(True Compliance)

Voe

Full-scale current change
< 1/2 LSB « 0.5% FS)
ROUT> 2OMO typical
IREF= 1mA

-10

+18

V

Full Range Output
(Digital Input 1001 1001)

IFR4

TA = 25·C, IREF= 2mA

1.92

Zero-Scale Current

Izs

Output Current Range

lOR

Logic Input Levels
LogiC ''0''
Logic "1"

V,L
V,H

LogiC I nput Current
Logic ''0''
Logic "1"

I,L
I'H

±1I2

V-=-10V
V- = -12V to -18V

2.2
4.2

1.98

2.04

mA

0.2

5

I'A

2
2

VLe=OV
Y,N = -10V to +0 8V
Y,N = 2V to 18V

-2
0.002

V'S

V-=-15V

VTHR

Vs =±15V (Note 1)

-10
-1

Reference Bias Current

1'5

Reference Input Slew
Rate

dlldt

(Note 1)

Power Supply Sensitivity

PSSI FS+
PSSI FS-

V+ = 4.5V to 18V
V- = -4.5V to -18V
IREF= 1mA

Power Supply Current

1+
11+
1-

mA

0.8

VLe=OV

LogiC Input SWing

4

V

±10
±10

I'A

+18

V

+135

V

-3

I'A

8

mNl's

±Q.0003

±0.03

%4.I FS

±0.002

±0.03

%4.V

3.8
-6.5
3.8
-9.1

mA

Vs =±15V,I REF =2mA

2.3
-5.0
2.5
-7.8

Vs =±5V,I REF =1mA
Vs = ±15V, IREF= 2mA

37
152

52
194

mW

Vs =±5V,I REF =1mA

NOTE:
1. Guaranteed by deSign.

1/86, Rev. A

11-43

-'~~----

U

0

~
0

LogiC Threshold Range

Pd

~

0

U

-10

Power Dissipation

LSB

~

.., - - - - - - - - - -

,-',

-

-'-,

----

~

~

U
......
0

II

---------I~ DAC·20 2·DIGIT BCD HIGH·SPEED MULTIPLYING D/A CONVERTER

DICE CHARACTERISTICS
1. VLC
2. TOUT
3. V4.

lOUT

5.
8.
7.
8.

BIT 1 (MSB)
BIT2
BIT3
BIT4

9.
10.
11.
12.
13.
14.
15.
18.

BITS
BIT8
BIT 7
BIT 8 (LSB)

V+
VRE'(+)
VRIF (-)
COMP

For additional DICE Information refer to
1986 Data Book, Section 2.

DIE SIZE 0.085 X 0.085 Inch, 5,525 sq. mila
(2.159 X 1.651 mm, 3.58 sq. mm)

WAFER TEST LIMITS at Vs = ±

15V, IREF = 2.0mA,

TA = 25°C, unless otherwise noted. Output characteristics

refer to both

lOUT and lOUT'

DAC·20G
PARAMETER

SYMBOL

LIMIT

BCD 0 to 99 steps

Resolution
Monotonlclty
Nonlinearity

CONDITIONS

BCD 99 steps
NL

FS = 1001 1001

Output Voltage Compliance

UNITS
Digits MIN
Digits MIN

±112

LSB MAX

Voe

Full-Scale Current Change
<1/2 LSB

+18
-10

VMAX
VMIN

Full-Scale Current

I FS4

VREF = 10V
R'4, R,5 = 5kO

2.04
1.92

mAMAX
mAMIN

Zero-Scale Current

Izs

Output Current Range

lOR

Logic "0" Input Level

V'L

Logic "1" tnput Level

V'H

Logic Input Current
Logic "0"
Logic "1"

I'L
I'H

p.AMAX
V- = -10V
V- = -12V to -IBV

2.1
4.2

mAMIN

0.8

V MAX
VMIN

V'N = -10V to +O.BV
V'N = 2V to lBV

±10
±10

p.AMAX

+IB
-10

V MAX
VMIN

±0.03
±0.03

%Il.I FS MAX
%Il.V

Logic Input Swing

V,s

V-=-15V

Power Supply Sensitivity

PSSIFs+
PSSI F5-

V- = -4.5V to -IBV
V- = -4.5V to -18V
IREF= lmA

Power Supply Current

1+
1-

VS=±IBV
IREF ,,2mA

3.B
-7.B

mAMAX

Power Dissipation

Pd

VS=±IBV
IREF " 2mA

194

mWMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield afler packaging is not
guaranteed for standard product dice. Consult factory to negotiate speCifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS

at Vs = ±15V, IREF = 2.0mA, unless otherwise noted specified. Output

characteristics refer to both lOUT and lOUT.

DAC·20G
PARAMETER

SYMBOL

Reference Input Slew Rate

dlldt

Propagation Delay

tPLH' tpHL
ts

Settling Time

CONDITIONS

TYPICAL

UNITS

B

mAlp.s

TA = 25°C, Any Bit

35

ns

To ± 1/2 LSB, All Bits Switched
ON or OFF, TA = 25°C

85

ns

11·44

1/86, Rev. A

--------I~ DAC-20

2-DIGIT BCD HIGH·SPEED MULTIPLYING D/A CONVERTER

TYPICAL REFERENCE PERFORMANCE CHARACTERISTICS
REFERENCE AMP
COMMON·MODE RANGE
(DIGITAL INPUT 10011001)

FULL·SCALE CURRENT VI
REFERENCE CURRENT
(DIGITAL INPUT 1001 1001)

2.8

5.0

2.4

1
~

0:
0:

""
~

5

2.0

V~. -,~V

V- .1_5V

I

I

'.1

2mA

·1

ili0:

L.I,mA

!:;

'iEF

4.0

1
....

IV•• 1,6v

V

3.0

0:

2.0

j!:

I

-1'1

[

0.0
~18

~12

~4

-8

[
12

1/

'.0

IAt.o12mA

POSITIVE COMMON MODE IS ALWAYS (V+) -1.5V;
NEGATIVE COMMON MODE RANGE IS V- PLUS
(tREF

x SOOn)

-y

-

-

9 2.0

3.0

2.0

4.0

l

-'"

6.0

'2

REFERENCE INPUT FREQUENCY
RESPONSE (DIGITAL INPUT
10011001)

OUTPUT CURRENT
OUTPUT VOLTAGE
(OUTPUT VOLTAGE COMPLIANCE)
(DIGITAL INPUT 10011001)
VB

~

iO

:s

........

'.2

i'

~

-4

5
0

~

~

0:

0.4

-8
-'0
-'2

0.0
-15 -60 -25

0

25

50

75

100 125 150 175

-'4 0.1

2.4

'.1

r-..
R'4 .. R,S " 1kn
Rt. "50on

0.5

I
r
r

0.4

2.0

5.0

-12

FREQUENCY (MHz)

I I
! I

~

[

II

[

I J

-8

-4

12

16

OUTPUT VOLTAGE (VOLTS)

CURVE 1: Cc " 15pF, V1N .. 2.0Vp_p CENTERED AT
+1.0V, LARGE SIGNAL.
CURVE 2' Cc = 15pF, V IN = 50mVp_p CENTERED AT
+200mV, SMALL SIGNAL.

POWER SUPPLY
CURRENT VI V+

POWER SUPPLY
CURRENT VI V-

8.0

1.0
BITS "HIGH" OR "LOW"

1....

BITS MAY BE "HIGH" OR "LOW"

I-

....

1i0:i

1i0:i

4.0

il:

iil

1+

0:

~

""~

4.0

M

iil

I I I Ii_
!- WI~H I~EF J0.2JA _

0:

~

'.0

=2mA

I- WITH IREF '" lmA

0:

0:

"">
t

I - WITH IREF

=<

.§ 6.0

6.0

~

'.0

M

MUUUUW~N~~W

V+, POSITIVE POWER SUPPLY (Vdc)

1+

I-

M~~_~~~~~~~

V-, NEGATIVE POWER SUPPLY (Vdc)

11-45

<:

0'

~

iAEF .I'mA

IREF" O.2mA

-16

10

I

r

0.0
1.0

I

IREF'" 1mA

0.8

II I

TEMPERATURE (oC)

v-, ~5V

'.2

\

vRtS " ov
I I I I II
0.2

L-,L
II

2.0

-I

~

0.8

'0\

-2

!:;

>

:I:

q

./

~

Z

~

2.8

i'

~

U

o
o

TA "'TMIN TOTMAX

~

'6

LOGIC INPUT VOLTAGE (VOLTS)

THE RECOMMENDED RANGE FOR OPERATION WITH
WITH A OC REFERENCE CURRENT IS +O.2mA
TO +4.0mA.

NOTE:

PLUS 2.6V,

2.0

'.1

~

I

00

'.0

IREF' REFERENCE CURRENT (mA)

VTH - VLC VI TEMPERATURE

~

'[

!.!

L
0.0

V,S. REFERENCE COMMON MODE VOLTAGE (VOLTS)

NOTE:

LIMIT FOR

V~I·

V

0.0

16

'"

V

"

0.4

6.0

/

0

0.8

~

V

!/

il

'.2

8.0

FOR
~MIN ITO iMA~- I - i-,L1MIT
V-; -'6V

T1·

I I

T A· TMIN TO TMAX

LOGIC INPUT CURRENT
VI INPUT VOLTAGE

1/86, Rev. A

---------t~ DAC-20 2-DIGIT BCD HIGH-SPEED MULTIPLYING D/A CONVERTER

TYPICAL REFERENCE PERFORMANCE CHARACTERISTICS
OUTPUT VOLTAGE COMPLIANCE
'I TEMPERATURE

POWER SUPPLY CURRENT
VI TEMPERATURE
80

,.
~

ALL BITS "HIGH" OR "LOW'

~

1-

6.0
'REF'" 2.DmA

iiia:

w

a:

~

I

v- .. -15V

1.

'2

il

40

~

!

v+ = +16V

a:

1+

~ 20

-4

~

-8
ooL-J-~

-26

0

25

50

75

100 125

150

-75 -50 -25

TEMPERATURE (CCI

__L-J-~__L-J-~~
0

25

50

75

laO 125

150

TEMPERATURE {CCI

The unused output must be connected to ground or some
voltage source capable of sourcing 1.65 times IREF. A detailed
discussion of reference input operation begins on the next
page.

BASIC OUTPUT CONNECTIONS
With complementary current outputs, the DAC-20 may be
used with either positive true or negative true (complementary) logic. Current appears at the "true" output (10) when a
"1" is applied to a logic input. As the BCD-coded input
increases, the sink current at Pin 4 increases proportionately,
in the fashion of a "positive logic" D/A converter. When a "0"
is applied to a logic input, that current is turned OFF at Pin 4
and ON at Pin 2 (lQj which is used for negative true or
"negative logic" D/A converters.

Both outputs have an extremely wide voltage compliance
enabling fast direct current-to-voltage conversion through a
resistor tied to ground or other voltage source. Positive
compliance is 36V above V- and is independent of the
positive supply. Negative compliance is given by V- plus
(I REF X 8000) plus 2.5V.

POSITIVE VOLTAGE OUTPUT
NEGATIVE TRUE LOGIC INPUTS

POSITIVE TRUE LOGIC INPUTS

+,ov
V REF

5000len 14

-=DECIMAL
INPUT

+15V -16V

-=-

-15V

BCD INPUT
MID
LID

10

0

0000

0000

0

10

0001

0000

020mA

+1.0V

20

0010

0000

0.40mA

30

0011

0000

O.SOmA

40

0100

0000

80

1000

99

1001

DECIMAL
INPUT

+15V -15V

BCD INPUT
MSD
LSD

iO

0

1111

1111

0

10

1110

1111

020mA

+1.0V

+2.0V

20

1101

1111

0.4OmA

+2.0V

+3.0V

30

1100

1111

o SOmA

+3.0V

0.80mA

+4.0V

40

1011

1111

O.80mA

+40V

0000

1.SOmA

+8.0V

80

0111

1111

1.60mA

+8.0V

1001

198mA

+9.9V

99

0110

0110

1.98mA

+9.9V

Eo
0

11-46

Eo
0

1/86, Rev. A

---------I~ DAC-20 2-DIGIT BCD HIGH-SPEED MULTIPLYING D/A CONVERTER

NEGATIVE VOLTAGE OUTPUT
POSITIVE TRUE lOGIC INPUTS

NEGATIVE TRUE LOGIC INPUTS

MSO

LSD

~,.---"'---.

+10V
V REF

5000krl. 14

500kil

+1SV -15V

DECIMAL
INPUT

-15V

BCD INPUT
MSD
LSD
0000

0000

10

0001

0000

20

0010

30

0011

10

-15V

+15V -15V

DECIMAL
INPUT

Eo
0

BCD INPUT
MSD
LSD

~

0

1111

1111

0

10

1110

1111

o 20m A

Eo
0

020mA

-1.0V

-1.0V

0000

0.40mA

-20V

20

1101

1111

0.40mA

-2.0V

0000

o BOmA

-30V

30

1100

1111

O.BOmA

-30V

40

0100

0000

O.BOmA

-40V

40

1011

1111

O.BOmA

-4.0V

BO

1000

0000

1 BOmA

-B.OV

BO

0111

1111

1.BOmA

-BOV

99

1001

1001

1.98mA

-9.9V

99

0110

0110

1.9BmA

-9.9V

REFERENCE OPERATION
POSITIVE

NEGATIVE
MSD

MSD

LSD

~~

B1 82 63 B4 65 86 B7 BB

~R~~\ 14 r'",-"'-"....I"-...o;....-..;=",",",
o---.,......NV''-'I

VREF{+)

IREF

DAC-20

VREF{-)

DAC-20

R1S 15

FOR FIXED REFERENCE,
TTL OPERATION,
TYPICAL VALUES ARE
VREF = +10 COOV
RREF = 5 COOk
R'5"",RREF

Cc = 00lpF
VLC '" OV (GROUND)

LSB

MSB

RREF
(R14)'4

VREF{+)

R15,S

r
o lJ.1F

-=-

V REF {+)

LSD

~~

B1 82 83 B4 65 B6 B7 B8
MSB

16
Cc

v-

FOR FIXED REFERENCE,
TTL OPERATION,
TYPICAL VALUES ARE

13

CQMP

coov

v+

VREF = +10
RREF = 5 COOk
R,S "'" RREF
Cc =' 001IJF
VLC = OV (GROUND)

VLC

16

f'F CC~l::
v-

13

V+

VLC

VREFH
I FS = RREF

I FS '" RREF

TO'"

10 +
IREF x 165
FOR ALL LOGIC INPUT STATES

iQ ""

10 +
IREF x 165
FOR ALL LOGIC INPUT STATES

REFERENCE AMPLIFIER SETUP

The DAC-20 is a multiplying converter in which the output
current is the product of a digital number and the input reference current. The reference current may be fixed or may vary
from nearly zero to +4.0mA. The full range output current is a
linear function of the reference current and is given by:

In positive reference applications an external positive reference voltage forces current through R14 into the VREF(+)
terminal (Pin 14) of the reference amplifier. Alternatively. a
negative reference may be applied to VREF(-) at Pin 15; reference current flows from ground through R14 into VREF(+). as
in the positive reference case. This negative reference con-

11-47

1/86, Rev. A

II

----------I1fMD

DAC-20 2-DIGIT BCD HIGH-SPEED MULTIPLYING D/A CONVERTER

nection has the advantage of a very high impedance presented at Pin 15. The voltage at Pin 14 is equal to and tracks the
voltage at Pin 15 due to the high gain of the internal reference
amplifier. R1s(nominally equal to R 14 ) is used to cancel bias
current errors and may be eliminated with only a minor
increase in error.
When a DC reference in used, a reference bypass capacitor is
recommended. A 5V TTL logic supply is not recommended
as reference. If a regulated power supply is used as a reference, R14 should be split into two resistors with the junction
bypassed to ground with a O.ltLF capacitor.
For most applications the tight relationship between IREFand
IFR will eliminate the need for trimming IREF. If required,
full-scale trimming may be accomplished by adjusting the
value of R14.
The reference amplifier must be compensated by using a
capacitor form Pin 16 to V-. For fixed reference operation, a
O.OltLF capacitor is recommended. For variable reference
applications, see section entitled "Multiplying Operation."

VTH e:: VLC +1.4V
+15V CMOS
VTH = +7,6V

VTH '" +1 4V

+1SV
9.1kn
VLe

Fastest settling times are obtained when Pin 1 sees a low
impedance. If Pin 1 is connected to a 1kO divider, for example, it should be bypassed to ground by a 0.01 tLF capacitor.
MULTIPLYING OPERATION

Bipolar references may be accommodated by offsetti ng V REF
or Pin 15. The negative common-mode range of the reference
amplifier is given by: VCM-= V- plus (IREFX 8000) plus 2.5V.
The positive common mode range is V+ less 1.5V.
AC reference applications will require the reference amplifier
to be compensated using a capacitor from Pin 16 to V-. The
value of this capacitor depends on the impedance presented
to Pin 14: for R14 values of 1.0, 2.5 and 5.0kO, minimum value
of Cc are 15, 37, and 75pF. larger values of R14 require

62k.\1

ECl

The logic input threshold is l.4V above VLC. For TTL and DTl
interface, simply ground Pin 1. When interfacing ECl, an
IREF = 1mA is recommended. For interfacing other logic
families, see the figure. Pin 1 will source lOOtLA typically, so
the external circuitry must be designed to accommodate this
current. Note that the threshold voltage has the temperature
dependence of two forward biased diodes. The two VLC
setting circuits shown, include temperature compensation.

The DAC-20 provides excellent multiplying performance
with an extremely linear relationship between IFS and IREF
over a range of 2mA to 4tLA. Monotonic operation is maintained over a typical range of I REF from lOOtLA to 2mA.

LOGIC INPUT OPERATION AND INTERFACING

TTL

For V-= -15V, the logic inputs may swing between -10V and
+ 18V. This enables direct interface with a + 15V CMOS logic,
even when the DAC-20 is powered from a +5V supply. Minimum logic threshold voltage are given by: V- plus (IREF x
8000) plus 2.5V. The logic threshold may be adjusted over a
wide range by placing an appropriate voltage at the logic
threshold control pin (Pin 1, VLc).

CMOS, NMOS, PMOS

ACCOMMODATING BIPOLAR REFERENCES
V'

20kn

13kn

14

VIN~

DAC-20

3kn

3kn
TO PIN 1

39k.\1

20kn

15
TO PIN 1
VLe

VLe
R3

6.2kn
'REF> PEAK NEGATIVE SWING OF 'iN
-S.2V
14

LOGIC THRESHOLD CONTROL

VIN~

The DAC-20 design incorporates a unique logic input circuit
which enables direct interface to all popular logic families and
provides maximum noise immunity. This feature is made
possible by the large input swing capability, 2JJ.A logic input
current and completely adjustable logic threshold voltage.

R 15 (OPTIONAL) 15

DAC-20

RREF'" R16
V REF (+) MUST BE ABOVE. PEAK POSITIVE

SWING OF Y,N

11-48

1/86, Rev. A

---------I~ DAC-20 2-DIGIT BCD HIGH-SPEED MULTIPLYING D/A CONVERTER

proportionately increased values of Cc for proper phase
margin.

however, an artificial ground may be useful to insure logic
swings, etc., remain between acceptable limits.

For fastest response to a pulse, low values of R'4 enabling
small Cc values should be used. If Pin 14 is driven by a high
impedance such as a transistor current source, none of the
above values will suffice and the amplifier must be heavily
compensated, which will decrease overall bandwidth and
slew rate. For R'4 = 1kO and Cc = 15pF, the reference amplifier slews at 4mAl JJ.s enabling a transition from I REF=O to I REF
= 2mA in 500ns.

Power Consumption may be calculated as follows:
Pd = (1+) x (V+) + (1-) x (V-). A useful feature of the DAC-20
design is that supply current is constant and independent of
input logic states; this reduces the size of the power supply
bypass capacitors.
BURN-IN CIRCUIT

Operation with pulse Inputs to the reference amplifier may be
accommodated by the alternate compensation scheme shown
above. This technique provides lowest full-scale transition
times. An internal clamp allows quick recovery of the reference amplifier for a cutoff (IREF = 0) condition. Full-scale
transition (0 to 2mA) occurs in 120ns when the equivalent
impedance at Pin 14 is 2000 and C c = O. This yields a reference slew rate of 16mVI JJ.s, which is relatively independent of
RIN and V1N values.
NOTES

PULSED REFERENCE OPERATION
-18V

, RESISTORS ARE RS lA TYPE
2 CAPACITORS ARE CKR05BX102KL TYPE
3 CONNECT 0 O1 .. F SOV CAPACITORS FROM

V+ AND V- TO GROUND FOR EACH 6

VREF{+j

9

)

DEVICES
OPTIONAL RESISTOR
FOR OFFSET 1NPUTS

:: RREF

r

I

TEMPERATURE PERFORMANCE

'4

The nonlinearity and monotonicity specification of the
DAC-20 are guaranteed to apply over the entire rated operating temperature range. Full-scale output current drift is tight,
typically ± 10ppm/o C, with zero-scale output current and
drift essentially negligible compared to 1/2 LSB.

ovJL
TYPICAL VALUES
R 1N " 5k

V1N (+) '" lOV

NO CAP

The temperature coefficient of the reference resistor R'4
should match and track that of the output resistor for minimum overall full-scale drift.

POWER SUPPLY CONSIDERATIONS
SETTLING TIME OPTIMIZATION

The DAC-20 operates over a wide range of power supply
voltages from a total supply of 9V to 36V. When operating at
supplies of ±5V or less, IREF S; 1rnA is recommended. Low
reference current operation decreases power consumption
and increases negative compliance, reference amplifier
negative common-mode range, negative logic Input range,
and negative logic threshold range; consult the various
figures for guidance. For example, operation at -4.5V with
IREF = 2mA is not recommended because negative output
compliance would be reduced to near zero. Operation from
lower supplies is possible. However, at least 8V total must be
applied to insure turn-on of the internal bias network.

The DAC-20 is capable of extremely fast settling times, typically 85ns at IREF = 2.0mA. Judicious circuit design and
careful board layout must be employed to obtain full performance potential during testing and application. The output
capacitance of the DAC-20, including the package, isapproximately 15pF; therefore the output RC time constant dominates settling time if RL > 5000.
Fastest operation can be obtained by using short leads, minimizing output capacitance and load resistor values, and by
adequate bypassing at the supply, reference and VLC terminals. Supplies do not require large electrolytic bypass
capaCitors as the supply current drain is independent of input
logic states; 0.1JJ.F capaCitors at the supply pins provide full
transient protection.

Symmetrical supplies are not required, as the DAC-20 is
quite insensitive to variations in supply voltage. Battery
operation is feasible as no ground connection is required:

11-49

1/86, Rev. A

I

DAC-86
COMDAC® COMPANDING
D/A CONVERTER (w255 LAW)
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•

.iiiiiiiiiiiiiliiiiii

+V FS -

Conforms With Bell System 1'"255 Companding Law
Meets 03 Compandor Tracking Specifications
Both Encode and Decode Capability
Tight Full-Scale Tolerance Eliminates Calibration
Low Full-Scale Drift Over Temperature
Extremely Low Noise Contribution
Multiplying Reference Inputs
Simplifies PCM System Design
High Reliability
Low Power Consumption and Low Cost
Two Grades Available

ANALOG
OUTPUT

0COMDAC®
TRANSFER
CHARACTERISTIC

-V FS -

o

x 000 0000
DIGITAL INPUT

111 1111

GENERAL DESCRIPTION
The DAC-86 monolithic COMDAC® D/A Converter provides
a 15 segment linear approximation to the Bell System ,.,-255
companding law. The law is implemented by using three bits
to select one of eight binarily-related chords (or segments)
and four bits to select one of sixteen linearly-related steps
within each chord. A sign bit determines signal polarity, and
an encode/decode input determines the mode of operation.
Accuracy is assured by specifying chord end point values,
step nonlinearity, and monotonicity over the full operating
temperature range. Typical applications include PCM carrier
systems, digital PBX's intercom systems, and PCM recording.
For CCITT "A" Law models, refer to the DAC-89 data sheet.

PIN CONNECTIONS & ORDERING INFORMATION

ENCODE/DECODE SELECT
1 = ENCODE
SIGN BIT INPUT

1"" POSITIVE
MOST SIGN I FreANT CHORD
BIT INPUT
SECOND CHORD BIT INPUT
LEAST SIGNIFICANT
CHORD BIT INPUT
MOST SIGNIFICANT STEP
BIT INPUT

1

18

In(1+l'lxl)
< <
_
Y(x) - sgn(x)
In(1 + 1') -1 - x - 1
for a normalized coding range of ± 1
where: x = input signal level
Y = output compressed signal level
I' = 255
This law is implemented with a eight chord (or segment)
piecewise linear approximation with 16 linear steps in each
chord. Dynamic range of 72dB in both polarities is achieved
with eight-bit coding.

EQUIVALENT CIRCUIT

00

=

16

DECODe OUT E/D Sa = 01

4

15

ENCODE OUT E/O S8 = 10

5

14 ENCODe OUT E/O S8
13

BELL 1'"255 LAW TRANSFER CHARACTERISTIC
The DAC-86 transfer characteristic is a piecewise linear
approximation to the Bell System ,.,-255 law expressed by:

POSITIVE POWER SUPPL Y

DECODE OUT E/D S8

=

I

, ", ""

STEP
INPUTS

CHORD
INPUTS

SB

87 86 65 84

E/D

11

NEGATIVE POWER SUPPLY
NEGATIVE REFERENCE
INPUT

SECOND STep BIT INPUT

7

THIRD STEP BIT INPUT

8

11

m~G~VE REFERENCE

LEAST SIGNIFICANT STep
BIT INPUT

9

10

lOGIC THRESHOLD
CONTROL

l8-PIN HERMETIC DUAL-IN-LINE
(X-Suffix)
REFERENCE
AMPLIFIER

GRADE
DAC-86EX
DAC-86CX

TEMP. RANGE
-25° C!+85° C
-25° C!+85° C

ACCURACY
±1/2 Step
±1 Step

13
V~

11-50

V+

VLe

1/86, Rev. A

------------I~ DAC-86 COMDAC@COMPANDINGD/ACONVERTER

ABSOLUTE MAXIMUM RATINGS
V+ Supply to V- Supply

•••••••••••••••••••••••••••

36V

Logic Inputs

.................. V- plus SV to V- plus 36V

VLC Swing ••.•••••••••••••••••••••••••• V- plus SV to V+

Operating Temperature

Analog Current Outputs

Storage Temperature

••••••• V- plus SV to V- plus 36V

••••.•••••••••••

•••••••••••••••••

-25°C to +S5°C

-65° C

Reference Inputs •••••••••••••••••••••.•.•••••• V- to V+

Power Dissipation

Reference Input Differential Voltage

Derate Above 100°C ..........................

Reference Input Current

•.••.•••.••••.• ± 1SV

•••••••••••••••••••••••• 1.25mA

ELECTRICAL CHARACTERISTICS

at Vs

to +

150° C

•••••••••••.••.•.••..•....•••• 500mW

10mW/oC

Lead Temperature (Soldering, 60 sec) •••.••••••••• 300°C

= ±15V, IREF = 52S.uA. -25° C:S TA:S +S5° C, for all 4 outputs, unless otherwise

noted.

DAC-86E
PARAMETER

SYMBOL

CONDITIONS

Resolution

8 chords with 16 steps each

Dynamic Range

20 log (17.15''1 0 ."

Monotoniclty

Sign-Bit + or-

Chord End-Point Accuracy
All Chords

Error relative to Ideal values
at IFS = 2007.75pA

Encode DeCision Level Current

Additional output
encode/decode = 1

MAX

MIN

TYP

MAX

UNITS

±128

±128

±128

±128

±128

±128

Steps

72

72

72

72

72

72

dB

128

To within ±1/2 step

Full-Scale Drift (C7 ) (Note 2)

alFS

Full temperature range

Output Voltage Comphance

Voe

Full-Scale Symmetry Error

10 (+) - 10 (-)

Decode or encode pair
Input Code 111 1111

Zero-Scale Current (Co)

I zs

Measured at selected output
with 0000000 Input

Disable Current (All bits high)

'DIS

Leakage of output disabled
by ElD and SB

1/2

5/8

±1/16

±1/10

-5

" 1/2 step

+18

Logic Input Levels. LogiC "0"

V,L

VLe-OV

LogiC Input Levels. LogiC "1"

V ,H

VLe = OV

liN

Y,N = -5V to + 18V

LogiC Input Swing

V's

V-=-15V

Step

±1/10

±1/4

Step

+18

Volts

0

.....:l

-5

±1/40

±1/4

Step

1/40

1/8

1/40

1/4

Step

75

nA

±1

Step

0.8

0.8

Volts

120

120

pA

+18

Volts

75

±1/2
20

42

20

mA

Volts

-5

+18

-3

1'2

Reference Input Slew Rate

dl/dt

Power Supply SenSitiVity Over
Supply Range (Refer to
Characteristic Curves)

PSSI FS+
PSSI F5-

V+ = 4 5V to 18V. V- = -15V
V- = -10 8V to -18V. V+ = 15V

Power Supply Current

1+
11+
1-

Power Dissipation

Pd

-12

025

-5

-3

-12

025

pA
mAlJls

±1/20
±1/1O

±1/2
±1/2

±1/20
±1/10

±1/2
±1/2

Step

Vs =+5V.-15V.I FS =20mA
Vs = +5V. -15V. I FS = 2.0mA
Vs = ±15V. I FS = 2.0mA
Vs = ±15V, I FS = 2.0mA

27
-6.7
2.7
-6.7

45
-93
45
-9.3

2.7
-67
27
-67

4.5
-93
45
-93

mA

Vs = +5V, -15V, I FS = 2.0mA
Vs = ±15V, I FS = 2.0mA

114
141

167
207

114
141

167
207

mW

NOTE:
1. Inacompanding DACtheterm LSB IS not used because the step size within
each chord is different For example, In the first chord around zero (Col
step size is O.5~A, while In the last chord near full-scale (C 7) step size is
64~A Settling time vanes for each of the chord bits and step bits and a
maximum specification IS misleadmg. In decode operatIOn, the DAC-86

and OP-16 combination will decode eight channels In the encode mode,
the DAC-86 and CMP-01 combinatIOn Will encode eight channels. Both
encode and decode statements assume a 5.2~sec channel time.
Guaranteed by design

1/86, Rev. A

11-51

-----

U

C)

0

~
~

Reference Bias Current

-----=----~.--------

3/4

±1/8

4.2

Logic Input Current

~

0

1/2

±1/40

Error relative to Ideal values
at IFS = 2007.75pA
I FSR

1/4

~

Step

J,tsec

Full-scale current change

Output Current Range

Steps

~
i:.J.:I

±1

±1/2
3/8

ts

Step Accuracy
All Chords

TYP

128

Setthng Time (Note 1)

U)

DAC-86C

MIN

--~--.---.~------

--

--

"...,---~-------.-----

~
,.....

C)
,.....

Q

II

------------1!EMD DAC-88 COMDACIS COMPANDING D/A CONVERTER
OUTPUT CURRENT DC TEST CIRCUIT

"'

LINE SELECTION TABLE

+VREF

""

ENCODEI
DECODE

SIGN
BIT
1

10E(+)

2

1

0

10E(-)

(E01 /R2 )

3

0

1

100(+)

(E02/R3 )

4

0

0

100H

(E02/R4 )

TEST
GROUP
1

18.94kll

~RAEFJ

11

"'2

20k1l

-15V

+16V

(E01 /R1 )

NOTE: Accuracy Is specified In the test circuit using the tables below to be
within the specified proportion of a step at the maximum value In each chord.
Monotonlclty Is guaranteed for all Input codes.

R1 .. A2 "' R3" A4 .. 2.5kll
·VREF IS ADJUSTED BEFORE TESTING EACH DEVICE

TO PROVIDE IDEAL FULL SCALE OUTPUT CURRENT.

OUTPUT
MEASUREMENT

':"

CONDENSED CURRENT OUTPUT TABLES (lREF = 528j.1A)
IDEAL DECODE OUTPUT CURRENT IN MICROAMPS AT CHORD ENDPOINTS

~
STEP

0000
15

1111
STEP SIZE

4

6

000

001

010

011

100

101

110

111

0.0

8.25

24.75

57.75

123.75

255.75

519.75

1047.75

7.5

23.25

54.75

117.75

243.75

495.75

999.75

2007.75

2

4

8

16

32

64

0.5

IDEAL ENCODE OUTPUT CURRENT IN MICROAMPS AT CHORD ENDPOINTS

~
STEP
0

15

0

4

000

001

010

011

100

101

110

111

0000

0.25

8.75

25.75

59.75

127.75

263.75

535.75

1079.75

1111

7.75

23.25

55.75

119.75

247.75

503.75

1015.75

2039.75

16

32

64

STEP SIZE

0.50

4

NOTE:

These tables may be extended to Include all of the encode/decode currents
(Ideal with SI REF =528I'A) by multiplying anyof the numbers in the normalized
tables by 0.51'A.

PARAMETER DEFINITIONS

CHORD ENDPOINTS

FULL-SCALE DRIFT

The maximum code in each chord; used to specify accuracy.

The change in output current over the full operating temperature
with VREF= 10.000V, Rll = lB.94kO, and R12= 20kO.

STEPS

FULL-SCALE SYMMETRY ERROR

Increments in each chord which divides the chord into
16 equal levels.

The difference between 100(-) and 100(+) or the difference
between 10E(-) and 10E(+) at full-scale output.

OUTPUT LEVEL NOTATION

OUTPUT VOLTAGE COMPLIANCE

Each output current level may be deSignated by the code
Ie,s where C = chord number and S = step number. For
example, 10,0 = zero-scale current; 10,1 = first step from zero;
10,15 = endpoint of first chord (Co); 17,15 = full-scale current.

The maximum output voltage swing at any current level
which causes < 1/2 step change in output current.

CHORDS

DYNAMIC RANGE

Groups of linearly-related steps in the transfer function. Also
known as segments.

Ratio of full-scale current to step size in chord zero, expressed
in dB. This can be measured peak or peak-to-peak with the
same result.

11·52

1/86, Rev. A

-------------im DAC-88 COMDAce COMPANDING D/A CONVERTER
magnitude AID converter. However, a compressing AID has
one significant difference. In a conventional (linear converter), the step size Is a constant percentage of full-scale. In a
compressing AID converter, the step size Increases as the
output changes from zero-scale to full-scale.

BASIC ENCODE OPERATION
(COMPRESSING AID CONVERSION)
BASIC ENCODE CONNECTIONS
±&V ANALOG INPUT

When the DAC Is used In the feedback loop of a successive
approximation analog to digital converter (ADC) the DAC
outputs are used as decision levels to determine the edges of
the quantizing bands. When the DAC Is used In the decode
mode the outputs correspond to the center of the quantizing
bands. The encode mode output exceeds the decode mode
output by one-half step. See AN-39 for detailed explanation.

r-I

'*

GROUND

FOA
SINGLE
ENDED
INPUTS

SUCCESSIVE

APPROXIMATION
REGISTER

DIGITAL
OUTPUTS

ISAAI

SIGN
BIT

tt1*S=~
JCHOAD
I1
BITS
BITS
JSTEP

123458789

VFlEF
+fOV

A"

2Ok"
-16V

+1BV

ENCODE DECISION LEVELS
Compressing A/D conversion with the DAC-86 requires a
comparator, an EXCLUSIVE-OR gate, and a successive
approximation register - the usual elements in any sign-

ENCODING SEQUENCE
An encoding sequence begins with the sign-bit decision.
During this time the comparator functions as a polarity detector. The encode/decode (E/D) input is held at a logic "0". In
this mode, current flows into the decode outputs, and the
comparator is effectively disconnected from the DAC. Once
the Input polarity has been determined, the ElD input toggles
to a logic "1" allowing current to flow into IOE(+) or IOE(-)'
For positive Inputs, currentflows Into IOE(+) through R1, and
the comparator's output Is entered as the answer for each
successive decision. For negative inputs, current flows into
IOE(-) through R2 developing a negative voltage which Is
compared with the analog Input. An EXCLUSIVE-OR gate
Inverts the comparator's output during negative trials to
maintain the proper logic coding, all ones for full-scale and
all zeros for zero-scale.
The bits are converted with a successive removal technique,
starting with a decision at the code 011 1111 and turning off
bits sequentially until all decisions have been made.
C = chord no. (0 through 7)
S = step no (0 through 15)

NORMALIZED ENCODE LEVEL (SIGN BIT EXCLUDED) (Ic,s= 2[2 c (S + 17) -16.5]

~
STEP

0

1

2

3

4

5

6

7

000

001

010

011

100

101

110

111
4319

0

0000

1

35

103

239

511

1055

2143

1

0001

3

39

111

255

543

1119

2271

4575

2

0010

5

43

119

271

575

1183

2399

4831

3

0011

7

47

127

287

607

1247

2527

5087

4

0100

9

51

135

303

639

1311

2655

5343

5

0101

11

55

143

319

671

1375

2783

5599

6

0110

13

59

151

335

703

1439

2911

5855

7

0111

15

63

159

351

735

1503

3039

6111

8

1000

17

67

167

387

767

1567

3167

6367

9

1001

19

71

175

383

799

1631

3295

6623

10

1010

21

75

183

399

831

1695

3423

6879

11

1011

23

79

191

415

863

1759

3551

7135

12

1100

25

83

199

431

895

1823

3679

7391

13

1101

27

87

207

447

927

1887

3807

7647

14

1110

29

91

215

463

959

1951

3935

7903

1111

31

95

223

479

991

2015

4063

8159

4

8

16

32

64

128

256

15

STEP SIZE

11-53

1/86, Rev. A

•
•

------------IIEMD DAC-86 COMDACII!> COMPANDING D/A CONVERTER
100(+) forcing a positive voltage atthe operational amplifier's
output. When the sign-bit input is low, logic "0", all of the
output current flows into 100(-) through R2 forcing a negative
voltage output. The sign-bit steers current into 100(+) or
100(-), therefore the output will always be symmetrical,
limited only by the matching of R1 and R2.

ENCODE TRANSFER CHARACTERISTICS
(AID CONVERSION)
DIGITAL
OUTPUT 1+)

DECODE TRANSFER CHARACTERISTIC
(D/A CONVERSION)

ANALOG
OUTPUT (+)

DIGITAL
OUTPUT (-J

BASIC DECODE OPERATION
(EXPANDING Df A CONVERSION)
0/A conversion with the OAC-86 is implemented by using an
operational amplifier connected to the decode outputs. The
decode mode of operation is selected by applying a logic "0"
to the encode/decode input. This enables the 100(+) or
100(-) to be selected by the sign-bit input. When the sign-bit
input is high, a logic "1 ", all of the output current flows into

ANALOG
OUTPUT 1-)

NORMALIZED DECODE OUTPUT (SIGN BIT EXCLUDED) (lc,s=2[2 C (5+ 16.5) -16.5J

~
STEP

C =chord no. (Othrough 7)

5=step no (Othrough 15)

0

1

2

3

4

5

6

7

000

001

010

011

100

101

110

111

0

0000

0

33

99

231

495

1023

2079

4191

1

0001

2

37

107

247

527

1087

2207

4447

2

0010

4

41

115

263

559

1151

2335

4703

3

0011

6

45

123

279

591

1215

2463

4959

4

0100

8

49

131

295

623

1279

2591

5215

5

0101

10

53

139

311

655

1343

2719

5471

6

0110

12

57

147

327

687

1407

2847

5727

7

0111

14

61

155

343

719

1471

2975

5983

8

1000

16

65

163

359

751

1535

3103

6239

9

1001

18

69

171

375

783

1599

3231

6495

10

1010

20

73

179

391

815

1663

3359

6751

11

1011

22

77

187

407

847

1727

3487

7007

12

1100

24

81

195

423

879

1791

3615

7263

13

1101

26

85

203

439

911

1855

3743

7519

14

1110

28

89

211

455

943

1919

3871

7775

15

1111

30

93

219

471

975

1983

3999

8031

16

32

64

128

256

STEP SIZE

4

11-54

1/86, Rev. A

-----------I~ DAC-86 COMDAC8 COMPANDING D/A CONVERTER

NORMALIZED TABLES

BASIC DECODE CONNECTIONS

The encode and decode tables are used to calculate the ideal
output current at any point. For example, in decode mode at
13.7 (011 0111) find 343. 343/8031 x IFS = 85.75/lA (IFS =
2007.75/lA). Alternatively. use the condensed current tables
and add up the number of steps.

R1

25kO
R11
18.94kO
(RREf)

IREF

BASIC REFERENCE CONSIDERATIONS

R12

20kn

Full-scale output current is ideally 2007 .75/lA when the reference current is 528/lA in the decode mode. In the encode
mode IFS= 2039.75/lA due to the additional 112 step (32/lA).
A percentage change in IREF will produce the same percentage change in output current.

IREF=~
IDEAL VALUES
'REF = 528pA
IFS = 2016j.1A

The large step size at full-scale allows the use of inexpensive
references in many applications. In some situations VREF may
even be the positive power supply. For example, with V+ =
15V, RREF= 15V/528/lA or 28.4kO. When using a power supply
as a reference. R11 becomes two resistors, R11 A and R11 B,
and the junction bypassed to ground with a 0.1/lf monolithic
capacitor.

NOTE: THIS CONFIGURATION WILL DECODE 24 CHANNELS

REFERENCE AMPLIFIER SETUP
The CAe-86 is a multiplying CIA converter. The output current is the product of the normalized digital input and the
input reference current. The reference current may be fixed
or may vary from nearly zero to + 1.0mA. The full-scale output
current is a linear function of the reference current.

DECODE OUTPUT VOLTAGE
E/D SB B1 B2 B3 B4 B5 B6 B7 VOLT

o

1

(+) ZERO-SCALE +1 STEP 0

1

0

0

0

0

0

0

1

o

1

0

0

0

0

0

0

0

OV

000000000

OV

POS FULL-SCALE
(+)

ZERD-SCALE

(-) ZERO-SCALE

(-) ZERO-SCALE +1 STEP 0

0

0

0

o

0

1

1

NEG FULL-SCALE

In external reference applications a positive reference voltage forces currentlhrough R11 in the the VR(+) terminal (pin
11) of the reference amplifier. Alternatively, a negative reference may be applied to VR(-) at pin 12; reference current
flows from ground through R11 into VR(+). This negative
reference connection has the advantage of presenting a very
high impedance at pin 12. The voltage at pin 11 isequal toand
tracks the voltage at pin 12 due to the high gain olthe internal

5019V

0

0

0

0.0012

-00012

0

-5.019V

I

SYSTEM TEST CIRCUIT
MULTIPLEXER

MULTIPLEXER
D/A

AiD

.

OAeSAMPLE/HOLD
ANALOG
CHANNELS

CMNn

IN

1-_ _-', r-:;;;:;:--"---~

j

ANALOG
CHANNELS

OUT

REF01/02

CROSSTALK

-75dB

IDLE CHANNEL NOISE

21dBnc

(AVERAGE CHANNEL VALUE)

NOTES
118kHz SAMPLING CONDITIONS AID CONVERSION TIME
1554j.1s
2 AUDIO TEST ANALYZER CONTAINS AC MESSAGE FILTER

AND 3kHz FLAT FILTER

11-55

1/86, Rev. A

-----------l1£MD

DAC-88 COMDAC® COMPANDING D/A CONVERTER

accomplished by selecting Rll or by using a potentiometer
for Rll.

reference amplifier. R12 (nominally equal to Rll) is used to
cancel bias current errors and may be eliminated with a
minor increase in error.

Using lower values of reference current reduces negative
power supply current and increases reference amplifier negative common-mode range. While the recommended operating range of DC reference current is 0.1 rnA to 1.0mA, monotonic operation is maintained over an even wider range.

POSITIVE REFERENCE OPERATION
VRH

+10V

RnA

LOGIC INPUT AND POWER SUPPLY
CONSIDERATIONS

DIGITAL INPUTS

INTERFACING CIRCUIT FOR
ECl, CMOS, HTl, AND NMOS lOGIC INPUTS

ENCODE
OUTPUTS

DECODE
OUTPUTS

Eel

CMOS, HTL, NMOS

V'

VREF
IREF = RREF

DECODE OUTPUTS
IDEAL VALUES

ENCODe OUTPUTS
IDEAL VALUES

IREF"" 528.uA
IFS '" 201611A

IREF "" 528iJA

IFS

=

13k.Q

2048J,1A

NOTES

,
2

RECOMMENDED WHEN VREF IS V+ OR THE LOGIC POWER SUPPLY
PINS'1 AND 12 ARE eQUAL IN VOLTAGE VREF IS IMPRESSED
ACROSS Rll (RREF)

20k.l1

"A"

"A"

39kS1

20k.l1

52V
NOTES.
1
SET THE VOLTAGE "A" TO BE AT THE DESIRED LOGIC INPUT
SWITCHING THRESHOLD
2
ALLOWABLE RANGE OF LOGIC THRESHOLD IS TYPICALLY
-5V TO +13V WHEN OPERATING THE DAC-S6 ON ±15V SUPPLIES

NEGATIVE REFERENCE OPERATION

ENCODE
OUTPUTS

lOGIC INPUTS
The DAC-86 interfaces with various logic families by referencing VLC (pin 10) at a potential which is 1.4V below the
desired logic input switching threshold. However, this
voltage source must be capable of sourcing and sinking a
changing current at pin 10.

DECODE
OUTPUTS

ENCODE OUTPUTS

DECODE OUTPUTS
IDEAL VALUES

IDEAL VALUES

IREF = 528iJA
lfS = 2016iJA

IREF '" 528}JA
IFS = 2048/.1A

The negative voltage at the logic inputs must be limited to
+ 10V with respect to V- (pin 13).

NOTE
PINS 11 AND 12 ARE EQUAL IN VOLTAGE VREF IS IMPRESSED

POWER SUPPLIES

ACROSS All (RREFl

Power supply current drain is relatively independent of
voltage and temperature and completely independent of the
logic input states.
When operating with V- between -15V and -l1V, output
negative voltage compliance, Voc(-), reference input
amplifier common-mode voltage range, and logic input
negative voltage range are reduced by an amount equivalent
to the difference between -15V and the V- supply in use.
Operation with V+ between +5V and +15V affects VLC and
the reference amplifier common-mode positive voltage range
in the same manner.

REFERENCE AMPLIFIER OPERATION
For most applications a + 10.0V reference, such as the PMI
REF-Ol, is recommended for optimum full-scale temperature
performance. (This also minimizes the contributions of reference amplifier Vos and TCVos). For most applications the
tight relationship between IREF and IFS eliminates the need
for trimming IREF; but if desired full-scale trimming is

11-56

1/86, Rev. A

-----------l1EMD DAC-88 COMDAC· COMPANDING D/A CONVERTER
SYSTEM PERFORMANCE CHARACTERISTICS

STANDARD OUTPUT CONNECTIONS

SIGNAL TO QUANTIZING DISTORTION vs INPUT LEVEL
VREF

~

"OY

5.

."on

z

1••1M

...9
II:

0

I;; 40
C

.'2

~

;!

20kn

0

~
~

3.

"in
"
Z

20

--5.

STANDARD ENCODE/DECODE
CONNECTIONS REQUIRE
1VocHI <: IVAmul
-40

-30

-2.

-1.

+1.

INPUT lEVEL IdBmO_1_ _ C MSG WTG
I.ILAW _ _ _ _ 3kHz FLAT

OUTPUT COMPLIANCE EXTENSION CONNECTIONS
GAIN TRACKING

Rl

COMPARATOR

VREF
+10V

...

,."on

+15
_ +10

~

:3 -+05
z

~

00

.'2

~~~~~~~~~~~~~~~~~

20kn

~
~

~

-05

-1.

-3.

-20

-1.

EXTENDED RANGE CONNECTION ALLOWS

-0

V Anth <: Vocl-l/n
WHERE n .. R2JR1 + R2

INPUT LEVEL (dBmOI

-16

VA MUST DRIVE Rl

+ R2

" L A W - - - C MSGWTG
____ 3kHz FLAT

NEGATIVE OUTPUT VOLTAGE COMPLIANCE Voc(-)
OUTPUT VOLTAGE COMPLIANCE
The DAC-86 has true current outputs with wide voltage compliance that enables single ended and balanced load drive
capability. Positive voltage compliance is +18V and negative
voltage compliance is -5.0V with IREF= 528~A and V = -15V.
Negative voltage compliance Voc(-) for other values of IREF
and V- may be obtained from the table, or calculated as
follows:
VOC(-) min = (V-)

V-

1.0mA

IFS
2.0mA

-12V

-2.SV

-2.0V

-O.4V

-1SV

-S.SV

-S.OV

-3.4V

-1SV

-S.SV

-S.OV

-6.4V

4.0mA

MINIMUM NEGATIVE COMPLIANCE
VOC(-) MIN = (V-) + (2 IREF 1.6kO) + S.4V

+ (2 IREF X 1.6kO) + 8.4V

Output voltage compliance can be extended in both encode
and decode modes using the connections shown above.

DICE
For applicable DICE information, see DAC-88 Data Sheet.

11-57

1/86, Rev. A

II

DAC-88
COMDAOs> COMPANDING
D/A CONVERTER (~-25S LAW)
Precision M0l101ithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•
•
•
•

GAIN TRACKING

IMPROVED ACCURACY over DAC-86
IMPROVED SPEED over DAC-86
Conforms With Bell System 1'"255 Companding Law
Meets 03 Compandor Tracking SpeCifications
Both Encode and Decode Capability
Tight Full-Scale Tolerance Eliminates Calibration
Low Full-Scale Drift Over Temperatura
Extremely Low Noise Contribution
Multiplying Reference Inputs
Simplifies PCM System DeSign
High Reliability
Low Power Consumption and Low Cost
Fully Specified Dice Available

-15
fA) GAIN TRACKING PERFORMANCE OF DAC-88IN 8-CHANNEL SYSTEM

GENERAL DESCRIPTION
BELL 1'"255 LAW TRANSFER CHARACTERISTIC

The DAC-88 monolithic COMDACI8> D/A Converter provides
a 15 segment linear approximation to the Bell System 1'"255
companding law. The law is implemented by using three bits
to select one of eight binarily-related chords (or segments)
and four bits to select one of sixteen linearly-related steps
within each chord. A sign bit determines signal polarity, and
an encode/decode input determines the mode of operation.

The DAC-88 transfer characteristic is a piecewise linear
approximation to the Bell System 1'255 law expressed by:
Y(x) = sgn(x)

In(1+l'lxl)
< <
In(1 + 1') -1 - x - 1

for a normalized coding range of ± 1
where: X = input signal level
Y = output compressed signal level
I' = 255

Accuracy is assured by specifying chord end point values,
step nonlinearity, and monotonicity over the full operating
temperature range. Typical applications include PCM carrier
systems, digital PBX's, intercom systems, and PCM recording. For CCITT "A" Law models, refer to the DAC-89 data
sheet.

This law is implemented with an eight chord (or segment)
piecewise linear approximation with 16 linear steps in each
chord. DynamiC range of 72dB in both polarities is achieved
with eight-bit coding.

PIN CONNECTIONS & ORDERING INFORMATION
EQUIVALENT CIRCUIT
ENCODE/DECODE SELECT
1 = ENCODE
SIGN BIT INPUT
1 = POSITIVE

MOST

1

POSITIVE POWER SUPPl Y

2

DECODE OUT E/D S8 '" 00

SIGNIFICA~JT~~~~~

SECOND CHORD BIT INPUT
LEAST SIGNIFICANT
CHORD BIT INPUT
MOST

STEP
INPUTS
87 86 B5 B4

DECODE OUT E/D SB = 01
ENCODE OUT E/D

sa = 10

14

ENCODE OUT E/D

sa = 11

13

NEGATIVE POWER SUPPLY

4
5

SIGNIFIC~~~~;~~

SECOND STEP BIT INPUT

7

THIRD STEP BIT INPUT

8

LEAST SIGNIFICANT STEP
BIT INPUT

9

CHORD
INPUTS
8382 B1

sa

'/0

~J~'tTIVE REFERENCE

L.._ _- '

11

r~~~~'VE REFERENCE

10

LOGIC THRESHOLD
CONTROL

VRf'l <>-',,+-,......--.,

18-PIN HERMETIC DUAL-IN-LiNE
(X-Suffix)

REFERENCE
AMPLIFIER

,.

DAC-8S

Grade
DAC-88EX

Temp. Range
-25°C/+85°C

13

Accuracy
±1/4 Step

V-

11-58

18

V·

VLe

1/86, Rev. A

-------------I1fMD

DAC·88 COMDAC~ COMPANDING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
V+ Supply to V- Supply. • • • • • • • •• • •• • • • • • . • • • • • • • •• 36V
VLC Swing •••.••••••••••••••••••.•••••• V- plus SV to V+
Analog Current Outputs ••••••• V- plus SV to V- plus 36V
Reference Inputs ••.•••.•••••••••••••••••.•...• V- to V+
Reference Input Differential Voltage ••••.•••.•••••• ± 1SV
Reference Input Current ••••••••.••••••••.•.•••• 1.2SmA
Logic Inputs •••••••••••••.•••• V- plus SV to V- plus 36V

Operating Temperature ••••••••.•••••••

-2S·C to +SS·C

Storage Temperature •••••••••.••••••• -6S·C to +1S0·C
Power Dissipation •••••••••.•••.•.•••••••••••••• SOOmW
Derate Above 100·C •...•........•••..••••.••• 10mW/·C
Lead Temperature (Soldering, 60 sec) •••.••••.•.•• 300·C
NOTE: Absolute ratings apply to both DICE and packaged parts. unless
otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs=± 1SV, IREF= S2SI'A, -2S· C:STA:S+SS· C, all 4 outputs, unless otherwise noted.
DAC·88E
PARAMETER

SYMBOL

Resolution

CONOITIONS
8 chords with 16 steps each

MIN

TYP

MAX

UNITS

±128

±128

±128

Steps

72

72

72

V')

~

~

~

Dynamic Range

20 log (i7,1s110,11

Monotonicity

Sign-Bit + or -

Chord End-POint Accuracy
Chord Zero

Error relative to Ideal values
at I FS = 2007 75"A

±1/4

Step

0

Chord End-Point Accuracy
All Chords Other Than Zero

Error relative to Ideal values
at I FS = 2007 75"A

±1/2

Step

C)

Encode Decision Level Current

Additional output
encode/decode = I

5/8

Step

Settling Time (Note I)
Settling Time In Chord Zero

128

dB
Steps

U

0
3/8

1/2

ts

To within ± 1/2 step

500

ns

Tseo

To within ±1/2 step

500

ns

Full-scale Drift (C 7 ) (Note 3)

Full temperature range

Output Voltage Compliance

Voe

Full-Scale current change
S 1/2 step

Full-Scale Symmetry Error
(Note 2)

10 (+)-1 0 (-)

Decode or encode pair
Input Code 1111111

Zero-Scale Current (Co)
(Note 2)

Izs

Measured at selected output
with 0000000 Input

±1/16

±I/IO

Step

+18

Volts

±1/4O

±1/8

Step

1/40

1/8

Step

5

-5

Leakage of output disabled
by ElDand SB

100

nA

Step Accuracy
Chord Zero

Error relative to ideal values
at I FS = 2007 75"A

±1/4

Step

Step Accuracy
All Chords Other Than Zero

Error relative to ideal values
at I FS = 2016"A

±1/2

Step

Output Current Range

42

Vle = OV

Logic Input Current

Y,N = -5V to + 18V

Logic Input Swing

V'S

V-=-15V

-3

V+ = 4.5V to 18V, V- = -15V
V- = -10.8V to -18V, V+ = 15V

__

._----

"A

+18

Volts

-12

±1/20
±I/IO

"A
mAl"s

±1/2
±1/2

Step

DAC-88 and CMP-DI combination will encode eight channels. Both
encode and decode statements assume a 5.2"s channel time.
2. Current specifications relate to differential currents between (+) and H
output leads. At the selected outputs, equal idle currents are present
simultaneously on both current output leads.
3. Guaranteed by design.

1/86, Rev. A

11-59

.

120

025

NOTES:
I. In a companding DAC the term LSB is not used because the step size
within each chord is different. For example, in the first chord around zero
(Co) step size IsO.5"A, while in the last chord near full-scale (C 7 ) step size
Is 84"A. Settling time varies for each of the chord bits and step bits and a
maximum specification is misleading. In decode operation, the DAC-88
and OP-16 combination Will decode 24 channels In the encode mode, the

-~

rnA
Volts
Volts

-5

dl/dt

Power Supply Sensitivity Over
Supply Range (Refer to
Characteristic Curves)

0
08

2

Reference Bias Current
Reference Input Slew Rate

2.0

Vle = OV

Logic Input Levels. Logic "I"

-~

.~.------

~

~
0
~

~

C)

......
Q

Disable Current (All bits high)
(Note 2)

Logic Input Levels, Logic "0"

~

II

------------1M

DAC-88 COMDAC- COMPANDING DlA CONVERTER

ELECTRICAL CHARACTERISTICS at Vs= ±15V, IREF= 528~A, -25"C S TAS +85"C, all 4 outputs, unless
otherwise noted. (Continued)
DAC-88E
PARAMETER

SYMBOL

CONDITIONS

MIN

TYP

MAX

UNITS

Power Supply Current

1+
11+
1-

Vs =+5V.-15V.I FS =2.0mA
Vs = +5V, -15V, IFS = 2.0mA
Vs =±15V,I FS =2.0mA
Vs =±lSV.I FS =2.0mA

2.7
-6.7
2.7
-8.7

5.5
-12
5.5
-12

mA

Power Dissipation

Pd

Vs = +SV. -lSV. IFS = 2.0mA
Vs = ±15V. I FS = 2.0mA

114
141

207
262

mW

Full-Scale Current Deviation
From Ideal Deviation
(See Tables) (Note 2)

IFS(D)
IFS(E)

VREF 10.000V, TA = 25' C
Rll = 19.53kO, R12 = 20kO

±1/2
±1/2

Step

Idle Current (Note 2)

II

10

!SA

NOTE:
2. Current specifications relate to differential currents between (+) and (-)
output leads. At the selected outputs. equal Idle currents are present
simultaneously on both current output leads.

DICE CHARACTERISTICS

1. ElD
2.
3.
4.
5.
6.
7.
8.
9.

SIGN-BIT
BIT 1 (MSB)
BIT2
BIT3
BIT4
BITS
BIT6
BIT 7 (LSB)

10. VLC
11. VR (+)
12. VR (-)
13. V14. IOE (+)
15. 10E (-)
16. 100 (+)
17.100(-)
18. V+

For additional DICE information refer to
1986 Data Book, Section 2.
DIE SIZE 0.123 X 0.085 Inch, 10,455 sq. mils
(3.124 X 2.159 mm, 6.745 sq. mm)

WAFER TEST LIMITS at Vs= ±15V, IREF= 5281'A, TA=25°C, all 4 outputs, unless otherwise noted.
DAC-88N
PARAMETER

SYMBOL

Resolution

CONDITIONS
8 chords with 16 steps each

Dynamic Range

20 log (1 7, ,si'o, ,)

Monotonicity

Sign-Bit+ or-

Chord End-Point Accuracy
Chord Zero

(NOTE 3)
LIMIT

UNITS

±128

Steps MIN

72

dB MIN

128

Steps MIN

Error relative to Ideal values
atl Fs = 2007.7SI'A

±1I4

Step MAX

Chord End-Point Accuracy
All Chords Other Than Zero

Error relative to Ideal values
at I FS = 2007. 751'A

±1/2

Step MAX

Encode Decision Level
Current

Additional output
encode/decode = 1

3/8
5/8

Step MIN
Step MAX

-5
+18

Volts MIN
Volts MAX

Output Voltage Compliance

Voe

Full-scale current change
S 1/2 step

11-60

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _-I~DAC.88 COMDAce COMPANDING D/A CONVERTER

WAFER TEST LIMITS

at Vs= ±15V, IREF= 5281'A, TA = 25°C,

8114 outputs,

unless otherwise noted. (Continued)

DAC·88N
(NOTE 3)
PARAMETER

SYMBOL

CONDITIONS

Full-Scale Symmetry Error
(Note 2)

10 +-1 0 -

Decode or encode pair
Input Code 1111111

Zero-Scale Current (Note 2)

Izs

Disable Current (All bits high)
(Note 2)

lOIS

LIMIT

UNITS

±1/8

Step MAX

Measured at selected output
0000000 input

1/8

Step Max

Leakage of output disabled
by ElD and SB

100

nAMAX

Step Accuracy
Chord Zero

Error relative to Ideal values
at IFS = 2oo7.75"A

±1/4

Step MAX

Step Accuracy
All Chords Other Than Zero

Error relative to Ideal values
at IFS = 2016"A

±1/2

Step MAX

4.2

mAMIN

Output Current Range

I FSR

Logic Input Levels. Logic "0"

VIL

VLC=OV

0.8

Volts MAX

Logic Input Levels. Logic "I"

VIH

VLC=OV

2

Volts MIN

Logic Input Current

liN

VIN= -5V to +18V

120

"A MAX

VIS

V-=-15V

-5
+18

Volts MIN
Volts MAX

Logic Input Swing
Reference Bias Current

112

-12

"A MAX

Power Supply Sensitivity Over
Supply Range

PSSI FSPSSI FS-

V+ = 4.5V to 18V
V- = -10.8V to -18V

±112
±1/2

Step MAX
Step MAX

1+
1-

Vs =±15V.IFS=2.0mA

5.75
-12.0

mAMAX

1+
1-

Vs = ±15V. IFS= 20mA

5.75
-12.0

mAMAX

IFSD

VREF 10.000V. TA = 25'C

±1/2

Step MAX

IFSE

Rll = 19.53kO
R12 =20kO

±1/2

Step MAX

Power Supply Current

Full-Scale Current Deviation
From Ideal Deviation
(See Tables) (Note 2)

NOTE:
Electrical tests performed at wafer probe to the limits shown. Due to variations in assembly methods and normal Yield loss. yield aiter packaging is not guaranteed
for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

TYPICAL ELECTRICAL CHARACTERISTICS

at Vs = ± 15V, and TA = 25° C, unless otherwise noted.

DAC·88N
PARAMETER

SYMBOL

CONDITIONS

TYPICAL

UNITS

Settling Time (Note 1)

ts

To within ± 1/2 step

500

ns

Settling Time in Chord Zero

Tsco

To within ±1/2 step

500

ns

Full-Scale Drift (C 7)

 COMPANDING D/A CONVERTER

When the DAC is used in the feedback loop of a successive
approximation analog to digital converter (ADC) the DAC
outputs are used as decision levels to determine the edges of
the quantizing bands. When the DAC is used in the decode
mode the outputs correspond to the center of the quantizing
bands. The encode mode output exceeds the decode mode
output by one-half step. See AN 39 for detailed explanation.

BASIC ENCODE OPERATION
(COMPRESSING AID CONVERSION)
ENCODE DECISION LEVELS
Compressing AID conversion with the DAC-88 requires a
comparator, an EXCLUSIVE-OR gate, and a successive
approximation register - the usual elements in any sign
magnitude AID converter. However, a compressing AID has
one significant difference. In a conventional (linear converter), the step size is a constant percentage of full-scale. I n a
compressing A/D converter, the step size increases as the
output changes from zero-scale to full-scale.

ENCODING SEQUENCE
An encoding sequence begins with the sign-bit decision.
During this time the comparator functions as polarity detector
only. The Encode/Decode (ElD) input is held ata logic "0". In
this mode current flows into the decode outputs, and the
comparator is effectively disconnected from the DAC. Once
the input polarity has been determined, the ElD input toggles
to a logic "1" allowing current to flow into 10E(+) or 10E(-)'

BASIC ENCODE CONNECTIONS

For positive inputs, currentflows into 10E(+) through R1, and
the comparator's output is entered as the answer for each
successive decision. For negative inputs, current flows into
10E(-) through R2 developing a negative voltage which is
compared with the analog input. An EXCLUSIVE-OR gate
inverts the comparator's output during negative trials to
maintain the proper logic coding, all ones for full-scale and
all zeros for zero-scale.

±5VANALOG
INPUT
~

+
r-I

-:;:-

FOR

GROUND

Rl

2."

SINGLE

ENDED
INPUTS

SUCCESSIVE
APPROXIMATION

L.

-15V

__

-t~~REiG~'S:TEiRj'S~AR~'q~--<> OUTPUTS
DIGITAL

The bits are converted with a successive removal technique,
starting with a decision at the code 011 1111 and turning off
bits sequentially until all decisions have been made.

SIGN BIT

m-mSICHORO
I rl=
BITS

I

STEP

BASIC DECODE OPERATION
(EXPANDING DJ A CONVERSION)

BITS

V REF

1 2 3 4 5 6 7 8

9

+10V

0/ A conversion with the DAC-88 is implemented by using an
operational amplifier connected to the decode outputs. The
decode mode of operation is selected by applying a logic "0"
to the Encode/Decode input. This mode enables the 100(+)
or 100(-) to be selected by the sign-bit input. When the signbit input is high, a logic "1", all of the output currentflows into

Rll
v R'
OAe-S8

11+-

l~:~r

12 'REF
V R-

R12
20kn

-15V

+15V

DECODE TRANSFER CHARACTERISTIC
(O/A CONVERSION)

ENCODE TRANSFER CHARACTERISTICS
(AID CONVERSION)

ANALOG
OUTPUT 1+)

ANALOG
INPUT (-)

ANALOG
INPUT (+)

ANALOG
OUTPUT I-J

11·63

1/86, Rev. A

------------l1fHD

DAC-88 COMDAC@COMPANDINGD/ACONVERTER

I OO( +) forcing a positive voltage at the operational amplifier's
output. When the sign-bit input is low, logic "0", all of the
output current flows into 100H through R2 forcing a negative
voltage output. The sign-bit steers current into 100(+) or
100H, the output will therefore always be symmetrical,
limited only by the matching of Rl and R2.

NORMALIZED TABLES
The encode and decode tables are used to calculate ideal
output current at any pOint. For example, in decode mode at
13 ,7 (011 0111) find 343. 343/8031 x IFs = 85.75,..A (lFS =
2007.75,..A). Alternatively, use the condensed current tables
and add up the number of steps.
C = chord no. (0 through 7)
8 = step no. (0 through 15)

NORMALIZED ENCODE LEVEL (SIGN BIT EXCLUDED) (Ic,s = 2[2 c (8 + 17) -16.5J
CHORO

o
2

o

4

5

001

010

011

100

101

110

111

0000

35

103

239

511

1055

2143

4319

0001

39

111

255

543

1119

2271

4575

0010

43

119

271

575

1183

2399

4831

0011

47

127

287

607

1247

2527

5087

STEP

000

0100

9

51

135

303

639

1311

2655

5343

0101

11

55

143

319

671

1375

2783

5599

0110

13

59

151

335

703

1439

2911

5855

0111

15

63

159

351

735

1503

3039

6111

1000

17

67

167

367

767

1567

3167

6367

9

1001

19

71

175

383

799

1631

3295

6623

10

1010

21

75

183

399

831

1695

3423

6879

11

1011

23

79

191

415

863

1759

3551

7135

12

1100

25

83

199

431

895

1823

3679

7391

13

1101

27

87

207

447

927

1887

3807

7647

4

6

14

1110

29

91

215

463

959

1951

3935

7903

15

1111

31

95

223

479

991

2015

4063

8159

16

32

64

128

256

STEP SIZE

4

NORMALIZED DECODE OUTPUT(SIGN BIT EXCLUDED) (lc,s=2[2 c (8+ 16.5)-16.5J

o

C = chord no. (0 through 7)
8 = step no. (0 through 15)

4

5

6

000

001

010

011

100

101

110

111

o

33

99

231

495

1023

2079

4291

37

107

247

527

1087

2207

4447

41

115

263

559

1151

2335

4703

45

123

279

591

1215

2463

4959

131

295

623

1279

2591

5215

o

0000

2

0010

3

0011

4

0100

8

49

5

0101

10

53

139

311

655

1343

2719

5471

6

0110

12

57

147

327

687

1407

2847

5727

0001

4

0111

14

61

155

343

719

1471

2975

5983

8

1000

16

65

163

359

751

1535

3103

6239

9

1001

18

69

171

375

783

1599

3231

6495

10

1010

20

73

179

391

815

1663

3359

6751

11

1011

22

77

187

407

847

1727

3487

7007

12

1100

24

81

195

423

879

1791

3615

7263

13

1101

26

85

203

439

911

1855

3743

7519

14

1110

28

89

212

455

943

1919

3871

7775

1111

30

93

219

471

975

1983

3999

8031

4

8

16

32

64

128

256

15

STEP SIZE

11-64

1/86, Rev_ A

-------------t~ DAC-88 COMDACI!J COMPANDING D/A CONVERTER

SYSTEM TEST CIRCUIT
MULTIPLEXER

MULTIPLEXER

DIA

AID
DAC·

••

SAMPLE/HOLD
ANALOG
CHANNELS

CMp·01

1-_ _-''- r-:=::-I~--~

DAC·

••

IN

ANALOG
CHANNELS
OUT

REF·
01/02
SAR

CROSSTALK
~ 75dB
IDLE CHANNEL NOISE
21dBnc
(AVERAGE CHANNEL VALUE)

NOTES
118kHz SAMPLING CONDITIONS AJD CONVERSION TIME
1554jJs
2 AUDIO TEST ANALYZER CONTAINS AC MESSAGE FILTER
AND 3kHz FLAT FILTER

BASIC REFERENCE CONSIDERATIONS
Full-scale output current is ideally 2007 .75/JA when the reference current is 528/JA in the decode mode. In the encode
mode IFS= 2039.75/JA due to the additional 1/2 step (32/JA). A
percentage change in I REF will produce the same percentage
change in output current.

The large step size at full-scale allows the use of inexpensive
references in many applications. In some applications VREF
may even be the positive power supply. Forexample, with V+
= 15V, RREF = 15v/528/JA or 28.4kO. When using a power
supply as a reference, R11 becomes two resistors, R11 A and
R11B, and the junction bypassed to ground with a 0.1/Jf
monolithic capacitor.
REFERENCE AMPLIFIER OPERATION

DECODE OUTPUT VOLTAGE
E/D 58 81 82 83 84 85 86 87

POS FULL-SCALE
(+1 ZERO-SCALE +1 STEP
(+1 ZERO-SCALE
(-I ZERO-SCALE
(-I ZERO-SCALE +1 STEP
NEG FULL-SCALE

The DAC-88 is a multiplying D/A converter. The output current is the product of the normalized digital input and the
input reference current. The reference current may be fixed
or may vary from nearly zero to + 1.0mA. The full-scale output
current is a linear function of the reference current and is
given for all four outputs in the figures above.

Eo

5.019V

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

o 0012V
o
o

OV
OV
-o.0012V

REFERENCE RECOMMENDATIONS

1 -5.019V

For most applications a + 10.0V reference, such as the PMI
REF-01, is recommended for optimum full-scale temperature
performance.

BASIC DECODE CONNECTIONS

POWER SUPPLIES

DIGITAL INPUTS

Power supply current drain is relatively independent of voltage and temperature and completely independent of the
logic input states.

R11
18.94
00
(RREFI
ANALOG
OUTPUT
±5V

IDEAL VALUES
-12V

+12V

>5V

NOTE:
THIS CONFIGURATION WILL DECODE 24-CHANNELS

IREF '" 528jJA
IFS'" 2007.7SjJA

When operating with V- between -15V and -11V, output
negative voltage compliance, VOC(-) , reference input amplifier common-mode voltage range, and logic input negative
voltage range are reduced by an amount equivalent to the
difference between -15V and the V- supply. Operation with
V+ between +5V and +15V affects VLC and the reference
amplifier common-mode positive voltage range in the same
manner.

11-65

1/86, Rev. A

II

_ _ _ _ _ _ _ _ _ _-I~DAC.88 COMDAC. COMPANDING D/A CONVERTER

capability. Positive voltage compliance is + 18V and negative
voltage compliance is -5.0V with IREF= 528"A and V =-15V.
Negative voltage compliance Voc(-) for other values of IREF
and V- may be obtained from the table. or calculated as
follows:

STANDARD DUTPUT CONNECTIONS

Voc(-) min = (V-)

""

'894
kn

+ (2 IREF X 1.6kO) + 8.4V

Output voltage compliance can be extended in both encode
and decode modes using the connections shown in the
compliance extension diagram.

RREf

",.

ZOkn

NEGATIVE OUTPUT VOLTAGE COMPLIANCE VOC(-)
STANDARD ENCODE/DECODE
CONNECTIONS REQUIRE

IVOC!_II':; IVAmllxl

COMPLIANCE EXTENSION CONNECTIONS

IFS

V-

1.0mA

2.0mA

-12V

-2.SV

-2.DV

-D.4V

-lSV

-S.SV

-S.DV

-3.4V

-lSV

-S.SV

-S.DV

-S.4V

4.0mA

MINIMUM NEGATIVE COMPLIANCE
Rl

VocH MIN = (V-)

COMPARATOR

""

+ (2 IREF 1.SkO) + S.4V

SIGNAL TO QUANTIZING DISTORTION vs INPUT LEVEL

1894
kn

RREF

jJLAW - - - C MSG WTG
_ _ _ _ 3kHz FLAT

"12

20kU

VAm8X ~ Vael-I/n
WHERE n '" R2/R, + R2
VA MUST DRIVE Rl + R2

zo

OUTPUT VOLTAGE COMPLIANCE

-so
INPUT LEVEL (dBmO)

The DAC-88 has true cu rrent outputs with wide voltage compliance that enables single ended and balanced load driving

APPLICATIONS
PHOTODIODE LINEARIZING CIRCUIT

ce

elK
SAR

0

ee
DATA
OUT

elK
IOE_

CMP+
DAC-88

CMP-

SIB

11·66

1/86, Rev. A

DAC-89
COMDAC® COMPANDING
D/A CONVERTER ("1\' LAWI
Precision Monolithics Inc.

+v FS -

FEATURES
•
•
•
•
•
•
•
•
•

11-Blt Accuracy and Resolution Around Zero
Sign Plus 66dB Dynamic Range
True Current Outputs: -5V to +18V Compliance
Tight Full-Scale Tolerance Eliminates Calibration
Low Full-Scale Drift Over Temperature
Low Power Consumption and Low Cost
Ideal for PCM and 8-Blt ",p Applications
Outputs Multiplexed for Time Shared Applications
Fully Specified Dice Available

. . . . . . . ..

ANALOG
OUTPUT

0-

COMOAC®
TRANSFER
CHARACTERISTIC

GENERAL DESCRIPTION
The DAC-89 monolithic COMDAC® converter provides the
complete decode function for "A" Law PCM CODECs. The
DAC-89 may be configured in an encoder, decoder, or timeshared between encoding and decoding.

AX
Y = 1 + In A

A= 87.6
The DAC-89 implements this law with an eight chord (or
segment) piecewise linear approximation for each polarity
with sixteen linear steps in each chord. The first two chords
are co-linear and of equal step size, and may be considered
as one chord of 32 steps. Step sizes of the remaining six
chords are binarily related to the first chord.

y= 1+lnAX 1/A~X~1
1 + In A

PIN CONNECTIONS & ORDERING INFORMATION

CHORD BIT INPUT

EQUIVALENT CIRCUIT

POSITIVE POWER SUPPLY
DECODE OUT E/D S8 = 00

4

5

CHORD
INPUTS

6382 61

S6

E/D

ENCODE OUT EJO S8 = 10
14

ENCODE OUT E/O S8 = 11

13

NEGATIVE POWER SUPPLY
NEGATIVE REFERENCE
INPUT
POSITIVE REFERENCE
INPUT
LOGIC THRESHOLD
CONTROL

7

THIRD STEP BIT INPUT

8

11

LEAST SIGNIFICANT STEP
BIT INPUT

9

10

TOP VIEW
18-PIN HERMETIC DUAL-IN-LINE
(X-Suffix)
GRADE
DAC-89EX

STEP
INPUTS

87 86 8584

DECODE OUT E/D S8 = 01

MOST SIGNIFICANT STEP

BIT INPUT
SECONe STEP BIT INPUT

0 ~ X ~ 1/A where:

Y = Output signal level of the compressor (encoder).

The DAC-89 output is an approximation to the CCITT "A"
law which can be expressed as:

2

, ", ""

X = Normalized input signal level of the compressor
(encoder), V1NNFS.

CCITT "A" LAW CHARACTERISTIC

1

0000

DIGITAL INPUT

Specifying chord end-pOint values assures accuracy chord
nonlinearity, and monotonicity over the full operating
temperature range. For companding D/A converters with
Bell ,.,.255 law conformance, refer to the DAC-88 data sheet.
For industrial, process control, and audio applications, see
the DAC-86 data sheet.

ENCODE/DECODE SELECT
1"" ENCODE
SIGN BIT INPUT
1 '" POSITIVE
MOST SIGNIFICANT CHORD
BIT INPUT
SECOND CHORD BIT INPUT
LEAST SIGNIFICANT

x 000

o ", ""

TEMP RANGE
-25°C/+85°C

REFERENCE
AMPLIFIER

13

ACCURACY
±1/4 step

v-

11-67

VLe

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _--1~DAC-88 COMDAC- COMPANDING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
V+ Supply to V- Supply. . . . . . .. . . . . .. .. .. .. . . . . .... 36V
VLC Swing. . • . . . . • • • • . • • . • • • . • • • . • • • . •• V- plus 8V to V+
Analog Current Outputs ••.•••• V- plus 8V to V- plus 36V
Reference Inputs ......••••••.•••.••.•••...•••. V- to V+
Reference Input Differential Voltage .•..••.•....... ± 18V
Reference Input Current .••••.•••.•.. '••.•.•••.•. 1.25mA
Logic Inputs .................. V- plus 8V to V- plus 36V

Operating Temperature Range .•...•.... -25°C to +85°C
Storage Temperature Range. . . . . • . • . •. -65° C to + 150° C
Power Dissipation .............................. 500mW
Derate Above 100° C .......................... 10mW/o C
Lead Temperature (Soldering, 60 sec) ..••.•..••.•• 300°C
NOTE: Absolute ratings apply to both DICE and packaged perts unless
otherwise noted,

ELECTRICAL CHARACTERISTICS at Vs=± 15V, IREF= 512",A, -25° C:5 TA:5+85°C, all 4 outputs, unless otherwise noted.
In a companding DAC the term LSB is not used because the step size within each chord is different. For example, in the first
chord around zero (Co) step size is 1.0",A, while in the last chord near full-scale (C7) step size is 64",A.
DAC-89E
PARAMETER

SYMBOL

CONDITIONS

MIN

TYP

MAX

UNITS

±128

±128

±128

Steps

Resolution

8 chords with 16 steps each

Dynamic Range

20 log (1 7, ,sllo,o)

Monotonicity

Sign Bit + or-

Chord End-Point Accuracy
Chord Zero

Error relative to Ideal values
at I FS = 20161'A

±1/4

Step

Chord End-Point Accuracy
All Chords Other Than Zero

Error relative to ideal values
atlFS = 20161'A

±1/2

Step

Step Accuracy
Chord Zero

at 'FS = 20161'A

±1/4

Step

Step Accuracy
All Chords Other Than Zero

Error relative to ideal values
at IFS = 20161'A

±1/2

Step

Encode Decision Level Current

Additional output
Encode/Decode = 1

3/4

Step

Settling Time (Note 1)

ts

To within ± 1/2 step

Full-Scale Drift (Note 3)

AI FS

Full temperature range

±1/4

Step

Output Voltage Compliance

VOC

Full-scale current change
S 1/2 step

+18

Volts

Full-SCale Current Deviation
from Ideal (See Tables) (Note 2)

'FS(D)
IF.(E)

VREF 10.oo0V TA = 2SoC
Rll = 19.53kO, R12 = 20kO

±1/2
±1/2

Step
Step

Full-Scale Symmetry Error
(Note 2)

' 0 (+) -10(-)

Decode or Encode pal r

±1/8

Step

Zero-Scale Current
(Note 2)

Izs

Measured at selected output
with 000 0000 Input

1/2

3/4

Step

Dlseble Current (Note 2)

lOIS

Disabled by EtD and SB

S.O

100

Idle Current (Note 2)

I,

as

66
128

Error relative to ideal values

1/4

Output Current Range

I FSR

VREF= 2S,oooV TA = 2SoC

V,L

VLC=OV

Logic Input Levels, Logic "1"

V,H

VLC=OV

Logic Input Current

'iN

V,N =-SVto+18V

Logic Input SWing

V's

V-=-lSV

"2

Reference Input Slew Rate

dlldt

Power Supply Sensitivity Over
Supply Range (Refer to
Characteristic Curves)

PSSI FS+
PSSI F5-

±1/20
-S

±1/40

ns

10

Logic Input Levels, Logic ''0''

Reference Bias Current

112
SOO

1/4

4,2

2.0

0

mA

0.8

Volts
Volts

-5
-3

120

I'A

+18

Volts

-12

0.25

NOTES:
1. Settling time varies for each of the chord bits and step bits and a maximum
specification may be misleading. In decode operation, the DAC-89 and
OP-16 combination will decode 8 channels. In the encode mode, the
DAC-89 and CMP-Dl combination will encode 8 channels. Both encode
and decode statements assume a 3,91's channel time,

nA
I'A

2,0

V+ = 4.5V to 18V. V- = -15V
V- = -10.8V to -18V, V+ = lSV

dB
Steps

±1/20
±1/10

I'A
mAIl's

±1/2

±1I2

Step

Current specifications relate to differential currents between (+) and (-)
output leads. At the selected outputs, equal idle currents are present
simultaneously on both current output leads,
3. Guaranteed by design.
2

11·68

1/86, Rev. A

-----------l1fMD DAC-89 COMDACII COMPANDING D/A CONVERTER
ELECTRICAL CHARACTERISTICS at VS= ±15V, IREF= 512j.1A, -25° CS TAS +85° C, all 4 outputs, unless otherwise noted.
In a companding DAC the term LSB is not used because the step size within each chord is different. For example, In the first
chord around zero (Co) step size is 1.0j.lA, while in the last chord near full-scale (C7) step size Is 64j.1A. (Continued)
DAC-89E
PARAMETER

SYMBOL

CONDITIONS

TVP

MAX

UNITS

1+

Vs = +5V. -15V. I FS = 2 OmA
Vs =+5V,-15V,I Fs=2.0mA
Vs = ±15V, IFS = 2.0mA
Vs = ±15V, I FS = 2.0mA

2.7
-6.7
2.7
-6.7

5.5
-12
5.5
-12

mA

Vs = +5V, -15V. I FS = 2.0mA
Vs =±15V,I FS =2.0mA

114
141

207
262

mW

1-

Power Supply Current

1+
1-

Power Dlssipatton

MIN

"->
r:(

~

DICE CHARACTERISTICS

1.
2.
3.
4.
5.
8.
7.
8.
9.

ElD
SIGN-BIT
BIT 1 (MSB)
BIT2
BIT3
BIT4
BIT5
BIT8
BIT 7 (LSB)

10.
11.
12.
13.
14.
15.
18.
17.
18.

~

0

VLC

VR (+)

U

VR (-)

C)

V-

0

10E (+)
10E (-)
100(+)
100(-)

~

V+

0

~

~

For additional DICE Information refer to
1988 Data Book, Section 2.

C)

DIE SIZE 0.123 X 0.085 Inch, 10.455 sq. mils
(3.124 X 2.159 mm, 8.745 sq. mm)

6

II

WAFER TEST LIMITS at Vs= ±15V, IREF= 528"A, TA=25°C, all 4 outputs, unless otherwise noted.
DAC-89N
(NOTE3)
PARAMETER

SYMBOL

Resolution

CONDITIONS
8 chords with 16 steps each

Dynamic Range

LIMIT

UNITS

±128

Steps MIN

66

dBMIN

128

Steps MIN

Error relative to ideal values
at IFS = 2007.75I'A

±1/4

Step MAX

Chord End-point Accuracy
All Chords Other Than Zero

Error relative to ideal values
at IFS= 2007.751'A

±1I2

Step MAX

Encode Decision Level
Current

Additional output
encode/decode = 1

114
3/4

Step MIN
Step MAX

-5
+18

Volts MIN
Volts MAX

±1/8

Step MAX

Monotonlcity

Sign-Bit + or-

Chord End-point Accuracy
Chord Zero

Output Voltage Compliance

Voe

Full-Scale Symmetry Error
(Note 2)
Zero-Scale Current (Note 2)
Disable Current (All bits high)
(Note 2)

Full-scale currenl change
S 1/2 step
Decode or encode pal r
Input Code 1111111

Izs

Measured at selected output
0000000 input

1/4

Step Max

lOIS

Leakage 01 output disabled
by ElDand 58

100

nAMAX

11-89

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _-l~DAC-88 COMDAce COMPANDING D/A CONVERTER

WAFER TEST LIMITS at VS= ±15V. IREF= 5~1I~A. TA= 25°C. all 4 outputs. unless otherwise noted. (Continued)
DAC-89N
(NOTE 3)
PARAMETER

SYMBOL

LIMIT

CONDITIONS

UNITS

Step Accuracy
Chord Zero

Error relat,ve to ,deal values
at I FS = 2007.7Sp.A

±1I4

Step MAX

Step Accuracy
All Chords Other Than Zero

Error relative to ,deal values
at I FS = 2016p.A

±1/2

Step MAX

Output Current Range
Logic Input Levals, Logic "0"
Logic Input Levels, Logic "I"

VLe=OV

Logic Input Current

V'N=-SVtO+18V

VREF= 2S.000V, TA = 2SoC

4.2

mAMIN

VLe = OV

0.8

Volts MAX

2

Volts MIN

V-=-ISV

Logic Input Swing

p.AMAX

-S
+18

Volts MIN
Volts MAX

-12

p.AMAX

V+ = 4 SV to 18V
V- = 10 8V to -18V

±1I2
±1I2

Step MAX
Step MAX

1+
1-

Vs = ± ISV, I FS = 2.0mA

S.S
-12.0

mAMAX

1+
1-

Vs=±ISV,I FS =2.0mA

S.7S
-12.0

mAMAX

VAEF 10.000V. TA = 2SoC

±1I2

Step MAX

Rll = 19.53kO
R12=20kO

±1/2

Step MAX

Reference Bias Current
Power Supply Sensitivity Over
Supply Range (Refer to
Characteristic Curvas)

Power Supply Current

120

Full-Scale Current Deviation
From Ideal Deviation
(See Tables) (Note 2)

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging Is not
guarenteed for standard product dice. Consult factory to negotiate spaciflcations based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V. and TA = 25° C. unless otherwise noted.
DAC-89N
PARAMETER

SYMBOL

CONDITIONS

TYPICAL

UNITS

Settling Time (Note 1)

ts

To within ± 1/2 step

SOO

ns

Settling Time In Chord Zero

Tseo

To within ±1/2 step

SOO

ns

Full-Scale Drift (C7)

~IFS

Full temperature range

±1/2O

Step

Reference Input Slew Rate

dl/dt

Power Dissipation
Idle Current (Note 2)

02S

mNp.s

Po

Vs +SV,-ISV

114

mW

Po

Vs =±ISV

141

mW

10

p.A

I,

NOTES:
1. In a companding DAC the term LSB Is not used because the step s,ze
within each chord is different. Forexample, in the first chord around zero
(Co) step size is O.Sp.A. While in the last chord near full-scale (C 7) step size
is 64p.A. Settling time varies for each of the chord bits and step bits and a
maximum speCification is misleading
2. Current specifications relate to differential currents between (+) and (-)
output leads. At the selected outputs, equal Idle currents are present
simultaneously on both current output leads
3. See DAC-89E for typical values.

11-70

1/86, Rev_ A

------------l~ DAC-89 COMDAC® COMPANDING D/A CONVERTER

OUTPUT CURRENT DC TEST CIRCUIT
R1

LINE SELECTION TABLE
DIGITAL INPUTS

+VREF

TEST
GROUP

ENCODE/
DECODE

SIGN
BIT

OUTPUT
MEASUREMENT

R11

1953kn
(RREF)

10E(+)

11

2

12
R12

>OkG

0

3

0

4

0

0

(E01/R1 )

10E(-)

(E01/R2 )

100(+)

(E02/R3)

100(-)

(E02/R4)

NOTE: Accuracy is specified in the test circuit using the tables below to be

Al '" R2 = R3 = R4 = 2 Skll
·VAEF IS ADJUSTED BEFORE TESTING EACH DeVICE
TO PROVIDE IDEAL FULL-SCALE OUTPUT CURRENT

within the specified proportion of a step at the maximum value in each chord.
Monotonlcity is guaranteed for all Input codes.
-:"

CONDENSED CURRENT OUTPUT TABLES
IDEAL DECODE OUTPUT CURRENT IN MICROAMPS AT CHORD ENDPOINTS
4
CHORO
o
000

001

010

011

100

101

110

111

0000

0.5

165

33

66

132

264

528

1056

1111

15.5

31.5

63

126

252

504

1008

2016

8

16

32

64

STEP

o
15

STEP SIZE

IDEAL ENCODE OUTPUT CURRENT IN MICROAMPS AT CHORD ENDPOINTS
4
CHORO
o
2
STEP

000
0000

15

1111

16

001

010

011

100

101

110

111

17

34

68

136

272

544

1088

32

64

128

256

512

1024

2048

16

32

64

STEP SIZE

4

NOTE:
These tables may be extended to include aU of the encode/decode currents
(ideal with IREF= 512"A) by multiplying any of the numbers in the normalized
tables by 0.5"A.

IDEAL OUTPUT CURRENT
The difference between the (+) and (-) currents (encode or
decode) at any code.

PARAMETER DEFINITIONS
STEP NONLINEARITY
Step size deviation from ideal within a chord.

OUTPUT VOLTAGE COMPLIANCE

ENCODE CURRENT

The maximum output voltage swing at any current level
which causes <1/2 step change in output current.

The difference between 10E(+) and 100(+) or the difference
between 10E(-) and 100(-) at any code.

CHORDS

FULL-SCALE DRIFT

Groups of linearly-related steps in the transfer function. Also
known as segments.

The change in output current over the full operating
temperature with VREF = 10.000V. R11 = 19.53kO. and
R12 = 20kO.

The maximum code in each chord; used to specify accuracy.

FULL-SCALE SYMMETRY ERROR

STEPS

The difference between 100(-) and 100(+) or the difference
between 10E(-) and 10E(+) at full-scale output.

Increments in each chord which divides the chord into 16
equal levels.

CHORD ENDPOINTS

11-71

1/86, Rev. A

II

_ _ _ _ _ _ _ _ _ _--I~DAC-89 COMDAC@COMPANDINGD/ACONVERTER

OUTPUT LEVEL NOTATION
Each output current level may be designated by the code Ie s
where C = chord number and S = step number. For example,
10 .0 = zero-scale current: 10 .1 = first step from zero; 10,15 =
endpoint of first chord (Co); 17.15 = full-scale current.
DYNAMIC RANGE
Ratio of full-scale current to step size in chord zero expressed
in dB.

BASIC ENCODE OPERATION
(COMPRESSING AID CONVERSION)

For positive inputs, current flows into 10E(+) through R1, and
the comparator's output is entered as the answer for each
successive decision. For negative inputs, current flows into
10E(-) through R2 developing a negative voltage which is
compared with the analog input. An EXCLUSIVE-OR gate
inverts the comparator's output during negative trials to
maintain the proper logic coding, all ones for full-scale and
all zeros for zero-scale.
The bits are converted with a successive removal technique
starting with a decision at the code 0111111 and sequentially
turning off bits until all decisions have been made.
ENCODE DECISION LEVELS

ENCODING SEQUENCE
An encoding sequence begins with the sign-bit decision.
During this time the comparator functions as a polarity detector only. The Encode/Decode (E/D) input is held at logic "0".
In this mode current flows into the decode outputs, and the
comparator is effectively disconnected from the DAC. Once
the input polarity has been determined, the ElD inputtoggles
"1" allowing current to flow into 10E(+) or 10E(-).

Compressing A/D conversion with the DAC-89 requires a
comparator, an EXCLUSIVE-OR gate, and a successive
approximation register - the usual elements in any signmagnitude AID converter. However, a compressing AID has
one significant difference. In a conventional (linear) converter, the step size is a constant percentage of full scale, but in a
compressing A/D converter, the step size increases as the
output changes from zero-scale to full-scale.

BASIC DECODE CONNECTIONS

When the DAC is used in the feedback loop of a successive
approximation analog to digital converter (ADC), the DAC
outputs are used as decision levels to determine the edges of
the quantizing bands. When the DAC is used in the decode
mode the outputs correspond to the center of the quantizing
bands. The encode mode output exceeds the decode mode
output by one-half step. See AN 39 for detailed explanation.

:t5V ANALOG INPUT

r ----

..!...
-=-

R2

2.5k

(GROUND FDA
SINGLE-ENDED
INPUTS)
SUCCESSIVE

DIGITAL

APPROXIMATION
REGISTER {SARI

ENCODE TRANSFER CHARACTERISTIC
(AID CONVERSION)

OUTPUTS

DIGITAL
SIGN

rH$a=t=~}

OUTPUT (+)

BIT
CHORD
BIT

~}STEP
BITS
1 2 3 4 5 6 7 8 9
81 83 85 B7
S8 82 84 B6
11

VREF
+10V

E/D

V R (+)
DAC-89

.......

Rll
19.53kn
(AREFJ

'REF

VAH 12
R12

2OkCOMPANDINGD/ACONVERTER

NORMALIZED ENCODE DECISION LEVELS (SIGN-BIT EXCLUDED)
NORMALIZED ENCODE DECISION
CHORO
STEP

3

4
,00

,0,

110

, ,1

68

0"
,36

272

544

,088

2176

0
000

0000

00,

0,0

34

6

000,

4

36

72

,44

288

576

,,52

2304

00,0

6

38

76

,52

304

608

,2,6

2432

00"
0,00

8
,0

40

80

,60

320

640

,280

2560

42

84

,68

336

672

,344

0,0,

,2

44

88

176

352

704

1408

2688
28,6

0,,0

,4

46

92

184

368

736

,472

2944

0",
,000

16

48

96

,92

384

768

,536

3072

,8

50

,00

200

400

800

,600

3200

9
,0

,00,

20

52

,04

4,6

832

,664

3328

,0,0

22

54

,08

208
2,6

432

864

1728

3456

"

,0"
,,00

24

56

,,2

224

448

896

,792

3584

26

58

,,6

232

464

928

,856

37,2

13
,4

,,0,

28

60

,20

240

480

960

,920

3840

",0

30

62

,24

248

496

992

,984

3968

""

32

64

,28

256

5,2

,024

2048

"4096

,6

32

64

,28

,2

,5

STEP SIZE

·Vlrtual Decision Level

BASIC DECODE OPERATION
(EXPANDING D/A CONVERSION)

NORMALIZED TABLES
The encode and decode tables are used to calculate ideal
output current at any code point. For example, in decode
mode at 13,7 (0110111) find 188. 188/4032 times IFSof 20161'A
equals 941'A.

D/ A conversion with the DAC-89 is implemented by using an
operational amplifier connected to the decode outputs. The
decode mode of operation is selected by applying a logic "0"
to the Encode/Decode input. This mode enables the 100
outputs, disables the 10Eoutputs, and allows 100(+) or 100(-)
to be selected by the sign-bit input. When the sign-bit input is
high, logic "1", the output current flows into 100(+) forcing a
positive voltage at the operational amplifier's output. When
the sign-bit input is low, logic "0", the output current flows
into 100(-) through R2 forcing a negative voltage output. The
sign-bit steers current into 100 (+) or 100H, the output will
therefore always be symmetrical, limited only by the matching of R1 and R2.

E/D SB B1 B2 B3 B4 B5 B6 B7

,,,

Eo

5.040V

POS FULL-SCALE

0

(+1 ZERO-SCALE+' STEP

0

,

0 0

0

0

0

0

1

0.OO12V

(+1 ZERO-SCALE

0

1 0

0

0

0

0

0

0

0.004V

(-I ZERO-SCALE

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

(-I

ZERO-SCALE +1 STEP

NEG FULL-SCALE

0

1

,

0.004V
-o.0012V
-5.040V

DECODE TRANSFER CHARACTERISTIC
(D/A CONVERSION)

BASIC DECODE CONNECTIONS

ANALOG

OUTPUT (+)
R1

2.5kn

R12

.ok"

VREF
IREF = RREF
IDEAL VALUES
'REF = 512j.1A
IFS = 2016,uA
ANALOG
OUTPUT (-)

11-73

1/86, Rev. A

I

~ DAC-89

COMDAC® COMPANDING D/A CONVERTER

NORMALIZED DECODE OUTPUT (SIGN-BIT EXCLUDED)
NORMALIZED DECODE OUTPUT
CHORD
STEP
000
0

4

6

001

010

011

100

101

110

111

0000

33

66

132

264

528

1056

2112

0001

35

70

140

280

560

1120

2240

0010

37

74

148

296

592

1184

2368

3

0011

39

78

156

312

624

1248

2496

4

0100

41

82

164

328

656

1312

2624

43

86

172

344

688

1376

2752

0101

11

0110

13

45

90

180

360

720

1440

2880

0111

15

47

94

188

376

752

1504

3008

8

1000

17

49

98

196

392

784

1568

3136

9

1001

19

51

102

204

408

816

1632

3264

10

1010

21

53

106

212

424

848

1696

3392

6

11

1011

23

55

110

220

440

880

1760

3520

12

1100

25

57

114

228

456

912

1824

3648

13

1101

27

59

118

236

472

944

1888

3776

14

1110

29

61

122

244

488

976

1952

3904

1111

31

63

126

252

504

1008

2016

4032

8

16

32

64

128

15

STEP SIZE

2

BASIC REFERENCE CONSIDERATIONS
Full-scale output current is ideally 20l61'A when the reference
current is 5121'A in the decode mode. In the encode mode I FS
= 20481'A due to the additional one-half step (32I'A). A percentage change in IREF caused by changes in VREF or RREF
will produce the same percentage change in output current.

Output voltage compliance can be extended in both encode
and decode modes using the output compliance extension
connections. (Figures 1 and 2).
NEGATIVE OUTPUT VOLTAGE COMPLIANCE VoC<-)

The large step size at full scale allows the use of inexpensive
references in many applications. In some applications VREFmay
even be the positive power supply. For example, with V+ =
l5V, RREF= l5V/5l21'A or 29.3kD. When using a powersupply
as a reference, Rll should be two resistors, RllA and RllB,
and the junction bypassed to ground to provide decoupling.
OUTPUT VOLTAGE COMPLIANCE

V-

1.0mA

IFS
2.0mA

4.0mA

-12V

-2.BV

-2.DV

-DAV

-lSV

-S.BV

-S.DV

-3.4V

-lBV

-B.BV

-B.DV

-S.4V

MINIMUM NEGATIVE COMPLIANCE

VocH MIN = (V-) + (2 IREF 1.6k!l) + BAV

The DAC-89 has true current outputs with wide voltage compliance that enables single ended and balanced load driving
capability. Positive voltage compliance is + l8V and negative
voltage compliance is-5.0V with IREF= 5l21'A and V= -15V.
Negative voltage compliance Voc(-) for other values of IREF
and V- may be obtained from the table, or calculated as
follows:

IDLE OUTPUT CURRENT
In the selected output state (encode or decode), equivalent
idle currents are present on the (+) and H output leads. The
output will be symmetrical with the external resistor matching
determining the overall system accuracy.

VocH min = (V-) + (2 IREF X 1.6kD) + 8.4V

11-74

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _---l~DAC-89 COMDAC® COMPANDING D/A CONVERTER

OUTPUT COMPLIANCE EXTENSION CONNECTIONS
STANDARD ENCODE/DECODE CONNECTIONS

VREF
+10V

DIGITAL INPUTS

R12
20kn

STANDARD ENCODE/DECODE
CONNECTIONS ReQUIRE

IVOC{-)I';;; IVAmaxl

Figure 1

EXTENDED RANGE CONNECTIONS

R1

COMPARATOR

VREF

+10V
R11
1953kn
(RREF)

R12
20kn

EXTENDED RANGE CONNECTION ALLOWS

II

eo A2/R, + R2
VA MUST DRIVE Al + R2

WHERE n

Figure 2

IDEAL DECODE OUTPUT CURRENT IN MICROAMPS (SIGN-BIT EXCLUDED)
IDEAL DECODE OUTPUT
CHORD
0
001

010

011

100

101

110

111

05

165

33

66

132

264

528

1056

0001

1.5

175

35

70

140

280

560

1120

0010

2.5

18.5

37

74

148

286

592

1184

0000

5

3

000

STEP

0011

3.5

19.5

39

78

156

312

624

1248

0100

45

205

41

82

164

328

656

1312

0101

5.5

215

43

86

172

344

688

1376

0110

65

22.5

45

90

180

360

720

1440

0111

7.5

23.5

47

94

188

376

752

1504

1000

8.5

245

49

98

196

392

784

1568

1001

9.5

25.5

51

102

204

408

816

1632

10

1010

10.5

265

53

106

212

424

848

1696

11

1011

11.5

275

55

100

220

440

880

1780

12

1100

12.5

28.5

57

114

228

456

912

1824

13

1101

13.5

29.5

59

118

236

472

944

1888

14

1110

14.5

305

61

122

244

488

976

1952

15

1111

155

31.5

63

126

252

504

1008

2016

4

8

16

32

STEP SIZE

2

----

----------

-----~---------

64

1/86, Rev. A

11-75

---

-----

-~,--

DAC-IOO
IO-BIT CURRENT-OUTPtn:
D/A CONVERTER
l'"n:CISIOII

MOlloltthlt.:s IIIC

FEATURES
•
•
•
•
•

Faat Settling ...•.•.• 225naec (8 Blta), 375naec (10 Blta)
Stable ....••...•....•... Tempcol to ±15ppm/oC Max
Commercial, Indultrlal and Military Modell Available
TTL Compatible Logic Inputl
Wide Supply Range ..................... ±6V to ±18V

The small size, wide operating temperature range, and high
reliability construction make the DAC-100 Ideal for aerospace
applications. Other applications include use in servopositioning systems, X-V plotters, CRT displays, programmable power supplies, analog meter movement drivers,
waveform generators and high speed analog-to-digltal
converters.

GENERAL DESCRIPTION
The DAC-100 Isa complete 10-bit resolution digital-to-analog
converter constructed on two monolithic chips in a single
16-pin DIP. Featuring excellent linearity vs. temperature
performance, the DAC-100 includes a low tempco voltage
reference, ten current source/switches and a high stability
thin-film R-2R ladder network. Maximum application flexibility is provided by the fast current output, matched bipolar
offset and feedback resistors. Resistors are included for use
with an external op amp for voltage output applications.

PIN CONNECTIONS

16-PIN HERMETIC
DUAL-IN-LiNE
PACKAGE
(Q-SuHlx)

Although all units have 10-bit resolution, a wide choice of
linearity and temperature coefficient options are provided to
allow price/performance optimization.

ORDERING INFORMATIONt
N.L.·
%FS
MAX

TEMPCO·
ppm/oC
MAX

INDUSTRIAL TEMPERATURE
Vo = ±2.5V15V
Vo = ±5V110V

MILITARY TEMPERATURE
Vo = ±2.5V15V
Vo=±5V110V

±O.OS

±1S

DAC100AAC7

±O.OS

±30

DAC100ABC7

DAC100ABC8

±O.OS

±60

DAC100ACCS/883

DAC100ACC6/883

DAC100ACC7

DAC100ACC8

±O.10

±30

DAC100BBCS/883

DAC100BBC6/883

DAC100BBC7

DAC100BBC8

DAC100BCCS/883

DAC100BCC6/883

DAC100BCC7

DAC100CCCS/883

DAC100CCC6/883

±O.10

±60

±O.10

±120

±O.20

±60

±O.20

±120

±O.30

±120

t All

commercial and industrial temperature range parts are available with
burn-In For ordering Information see 1986 Data Book, Section 2.
• Part number construction: The 1st letter following DAC-l00 (A-D) refers to

COMMERCIAL TEMPERATURE
Vo=±5V110V
Vo = ±2.5V15V

DAC100AAC8

DAC100ACC3

DAC100ACC4

DAC100BCC8

DAC100BCC3

DAC100BCC4

DAC100CCC7

DAC100CCC8

DAC100CCC3

DAC100CCC4

DAC100DDC7

DAC100DDC8

DAC100DDC3

DAC100DDC4

the nonlinearity specification; the 2nd letter (A-D) refers to the full-scale
tampeo; the letter Q refers to the package, and the end numeral indicates the

output voltage and temperature.

SIMPLIFIED SCHEMATIC

I
v+~~~~~tt:it~:1-c
3

AS

612kH

I.
RS

BIPOLAR REF

11-76

As

ANALOG
OUTPUT

-FOR 10V OR t6V OPERATION,
RS .. 4.BBkfl (PACKAGE 03,
06, Q71
FOR 6V OR t26V OPERATION,
RS = 2.44k0 (PACKAGE 04,

as,

OS)

1/86, Rev. A

------------1\fMD DAC·100 10·BIT CURRENT·OUTPUT D/A CONVERTER
ABSOLUTE MAXIMUM RATINGS

(Note 2)
DICE Junction Temperature. •• • • • • • • •• -2S· C to + 1S0· C
Storage Temperature Range ••••••••••• -6S·C to +1S0·C
L.ead Temperature (Soldering, 60 sec) ••••••••••• +300·C

V + Supply to V- Supply .................... 0 to +36V
V+ Supply to Output ••••••••••••••••••••••• 0 to + 1SV
V- Supply to Output ••••••••••••••••••••••• 0 to -1SV
L.oglc Inputs to Output ................... -1V to +6V
Power Dissipation (Note 1) .................... SOOmW
Operating Temperature Range 03, 04 •••• O· C to + 70· C
OS, 06, 07, OS ..................... -SS·Cto+12S·C

NOTES:
1. Rating applies to ambient temperature of 100' C. Above 100' C. derate at
10mW/'C.
2. Ratings apply to DICE and packaged parts. unless otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs = ± 1SV, -25· C S TAS + SS· C for 07 and OS devices; O· C S TAS +70· C for 03 and
04; -SS· C S TA S + 12S· C for OS and 06 devices, unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

DAC·100

NL

(±1/2LSB -10 bits)
(±1/2LSB - 9 bits)
(±1/2LSB - 8 bits)
(±3/4LSB - 8 bits)

Resolution
Nonlinearity
(For nonllnearlty/tempco
combinations, see Ordering
Information)
Full·Scale Tempco
(See Full'Scale Test Circuit)

Settling Time TA = 25'C

Full·Range Output VOltage
(Limits guarantee adjustabillty
to exact 10.0 (5.0)V with a
2000 Trimpot. between
Adjust and V-I

MIN

TVP

MAX

10

Tc

ts

V. R

to ±0.05% FS
to±O.l%FS
to ±0.2% FS
to ±0.4% FS
to ±0.8% FS

UNITS
Bits

ABCD-

±0.05
±0.1
±0.2
±0.3

-A
-B
-C
-D

±15
±30
±60
±120

ppm/'C

ALL
ALL
ALL
ALL
ALL

375
300
225
150
100

ns

Connect FS Adjust to V10V Models (03. 05, 07)
(See Full·Scale Test Circuit)
5V Models (04, 06, 08)
V,N =0.7V

10

%FS

Vzs

U

Measured with respect to

ALL

Logic Inputs: High

V,NH

Logic Inpuls: Low

V,NL

Measured with respect to
output pin

ALL

Logic Input Current, Each Input

liN

V,N =Oto+6V

ALL

Logic Input Resistance

R'N

V,N = Oto +6V

ALL

Logic Input Capacitance

output pin

C 'N

ALL

Output Resistance

Ro

ALL

Output Capacitance

Co

ALL

2.1

%FS
V

0.7
5

V
/loA

3

mO

500

kO

pF

13

pF

Applied Power Supplies: V+

ALL

+6

+18

Applied Power Supplies: V-

ALL

-6

-18

V

±0.10

% per Volt

Power Supply Sensitivity

Pss

Vs= ±6V to ±18V

Power Consumption

Po

Vs= ±15V
Vs= ±6V
Vs= ±15V

03,04
03,04
05.06, 07. 08

ALL
200
80
200

V

300
mW
250

Positive Supply Current

1+

Vs =+15V
Vs= +15V

03,04
05.06,07,08

10
8.33

mA

Negative Supply Current

1-

Vs = -15V
Vs =-15V

03,04
05,06,07,08

-10
-8.33

mA

11·n

~

~

~

5.55

0.013

0

I

V

ALL

V,N = 2.1V

~

C)

(See Basic Unipolar Voltage Output Circuit)
Zero·Scale Output Voltage

~

0

11.1

5

~

1/86, Rev. A

C)

.....
0

I

_ _ _ _ _ _ _ _ _ _--IIfMDDAC-100 10-BIT CURRENT-OUTPUT D/A CONVERTER

DICE CHARACTERISTICS
DAR-01

DAI-G1

DIE SIZE 0.089 X 0.063 Inch, 5607 sq. mils
(2.26 x 1.6 mm, 3.616 sq. mm)

DIE SIZE 0.060 x 0.067 Inch, 5360 sq. mils
(2.032 x 1.70 mm, 3.45 sq. mm)

1. Ra
2.
3.
15.
16.

2.
3.
4.
5.
6.
7.
8.
9.

VOUTPUT
FULL-SCALE ADJ

RS

R - Pads are connected to similarly
marked pads on DAI-01

VOUTPUT
BIT 10 (LSB)
BIT9
BIT8
BIT7
BIT6
BIT5

10.
11.
12.
13.
14.

BIT4
BIT 3
BIT2
BIT 1 (MSB)
V+

R - Pads are connected to similarly
marked pads on DAR-01

Note: Pads 4 - 14, See DAI-G1

Note: Pads 1, 2, 15, 16, See DAR-01

These die versions are available on special order; contact your PMI sales office.

WAFER TEST LIMITS at TA = 25°C for the R-2R Ladder Network comprised of R1-R8, R12, R23, R34, R45 and R56 when
connected to an ideal DAI-01, unless otherwise noted.
DAR-01-N
PARAMETER

CONDITIONS

Nonlinearity

VRI =3.2V

TYP

MIN

-

DAR-01-G
MAX

MIN

±0035

TYP

MAX

UNITS

±0.05

%

WAFER TEST LIMITS at TA = 25°C, VR1 = 3.2V, unless otherwise noted.
DAR-01
PARAMETER

CONDITIONS

Resistance Rl

Absolule Measurement

Ratio RCI to Rl

,

MIN

TYP

MAX

UNITS

2.56

3.84

kO

Ideal = 1.00503 to 1

-1

+1

%

Ratio Rl to RSI

Ideal = 1.29959 to 1

-1

+1

%

Ratio Rl to RS2

Ideal = 1.29959 to 1

-1

+1

%

Ratio RB to Rl

Ideal = 192211 to 1

-1

+1

%

.

\

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal Yield loss. yield after packaging IS not
guaranteed for standard product dice. Consult factory to negotiate speCifications based on dice lot qualification through sample lot assembly and testing.

11-78

1/86, Rev. A

------------I1rHD

DAC-100 10-BIT CURRENT-OUTPUT D/A CONVERTER

TYPICAL ELECTRICAL CHARACTERISTICS in common to all grades.
DAR-01
PARAMETER

CONDITIONS

Absolute Temperature CoeffiCient

All Resistors

Tracking Temperature CoeffiCient

All Resistors with Respect to Rl

TYP

MIN

MAX

UNITS

±180

ppm/'C

3

ppm/'C

WAFER TEST LIMITS at TA = 25° C when connected to an ideal DAR-01, unless otherwise noted.
DAI-01-N
PARAMETER

SYMBOL

CONDITIONS

Nonllneanty

NL

Vs =±15V

VMCR

Vs =±15V

Internal Reference

Voltage

MIN

DAI-01-G

TYP

MAX

MIN

TYP

±0.05
6900

66

6.6

MAX

UNITS

±0.1

%

6.900

V

WAFER TEST LIMITS at Vs = ± 15V, TA = 25° C when connected to an ideal DAR-01, unless otherwise noted.
DAI-01
PARAMETER

MIN

CONDITIONS

Resolullon
Analog Output Current

All Bits Low, V- Connected to FS Adjust

Zero-Scale Output Current

All Bits High, V- Connected to FS Adjust

Logic Input "0"

Measured with Respect to Output

Logic Input "I"

Measured with Respect to Output

Supply Current

All Bits High, V- Connected to FS Adjust

Power Supply Rejection

Vs = ±6V to ±18V

MAX

UNITS

10

TYP

10

Bits

1840

2274

IJ.A

±0011

%I F5

07

V

8.33

mA

0.1

%IFsfV

21

V

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to vanations In assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ±15V, and when connected to an ideal DAR-01, unless otherwise
noted.
DAI-01-N
PARAMETER

MIN

CONDITIONS

Full-Scale Temperature CoeffiCient (Note)

TYP
±60

DAI-01-G
MAX

MIN

TYP

MAX

I

UNITS

±60

ppm/'C

NOTE:
Full-Scale Temperature CoeffiCient is defined as the change in output voltage
measured," the basIc unipolar voltage output test CirCUit shown on the
DAC-l00 data sheet and IS expressed In ppm between 25'C and either
temperature extreme diVided by the corresponding temperature change.

1/86, Rev. A

11-79

----

-_.-

.

__

._--

-----------I~ DAC-100 10-BIT CURRENT-OUTPUT D/A CONVERTER

BASIC CONNECTIONS

REDUCED RESOLUTION APPLICATION

BASIC UNIPOLAR VOLTAGE OUTPUT CIRCUIT

v+

TYPICAL SETTLING TIME
FOR 1/2 SCALE CHANGE
"'" 1 OIlS

lOUT
6

VOUT

LOGIC CODING - The DAC-100 uses complementary or
inverted binary logic coding, i.e., an all "zeroes" Input produces a full range output, while an all "ones" Input produces a
zero-scale output. Each lesser significant bit's weight is onehalf the previous more significant bit's value. High logic input
turns the bit "OFF," low logic Input level turns the bit "ON".

BASIC BIPOLAR VOLTAGE OUTPUT CIRCUIT

LOGIC COMPATIBILITY - The input logic levels are directly
compatible with TTL logic and may also be used with CMOS
logic powered from a single +5 volt supply.
NONLINEARITY (NL) - The maximum deviation from an
ideal straight line drawn between the end pOints, expressed
as a percent of full-scale range (FSR) or given In terms of
LSB value. The end pOints are zero-scale output to full-scale
output for unipolar operation and minus full-scale to positive
full-scale for bipolar operation.
VOUT

V+

BIPOLAR OPERATION - The DAC-100 may be converted to
bipolar operation by injecting a half-scale current into the
output; this is accomplished by connecting the internal bipolar resistor to a +6.4 volt reference. Trimming of the zero
output may be facilitated by placing a 5000 adjustable resistance in series with the +6.4 volts.

APPLICATIONS INFORMATION
FULL RANGE OUTPUT ADJUSTMENT - The output current of the DAC-100 may be reduced to produce an exact
10.000 (5.000) volt output by connecting a 2000 adjustable
resistance between the full-scale adjust pin and V-. Adjustment should be made with an input of all "zeroes."

VOLTAGE AT OUTPUT PIN - The DAC-100 is designed to
be operated with the voltage at the output pin held very close
to zero volts. Input logic threshold levels are directly affected
by output pin voltage changes; voltage swings at the output
may cause loss of linearity due to improper switching of bits.
Large voltage swings may cause permanent damage and
should be avoided. Proper operation can be obtained with
output voltages held within ±0.7 volts; a pair of back-to-back
silicon diodes tied from the output to ground is a convenient
way of clamping the output to this limit.

LOWER RESOLUTION APPLICATIONS - The DAC-100
may be used in applications requiring less than 10 bits of
resolution. All unused logic inputs must be tied to logic high
for proper operation. "Floating" logic inputs can cause
improper operation.

11-80

1/86, Rev. A

-------------l~ DAC-100 10-BIT CURRENT-OUTPUT D/A CONVERTER

TYPICAL APPLICATIONS

BINARY-COOED-DECIMAL D/A CONVERSION
'5V

EXTERNAL REFERENCE CONNECTION
+15V

+1SV

ANALOG SUM OF TWO DIGITAL NUMBERS

v*CAN BE EXPANDED TO 3 DIGITS BY ADDITION OF A THIRD OAe AND
99 TO 1 CURRENT DIVIDER

INTERFACING WITH CMOS LOGIC

The DAC-100 requires only about 1/LA of input current into
each logic stage. This enables use with CMOS inputs as long
as one rule is observed; logic input voltages should not
exceed 6.5 volts or V+, whichever is smaller. To provide an
understanding of this rule, it is necessary to discuss the logic
input stage design.
LOGIC INPUT STAGE DESIGN

For simplicity, only one of the ten identical input circuits is
shown below. The DAC-100 uses a fast current-steering
technique that switches a bit-weighted current between the
positive supply (V+) and the analog output, which is usually
constrained to be at zero volts (virtual ground) by an external
summing amplifier.

v-

DAC-100 - LOGIC INPUT STAGE
DIGITALLY PROGRAMMED LEVEL DETECTOR
ANALOG
OUTPUT
ZERO
VOLTS

VREF

1 = VIN

> VREF

O=VtN-Oo-o---:-+---t---f'
REFERENCE
INPUT

CAPACITIVE LOADING
The output operational amplifier provides stable operation
with capacitive loads up to 100pF.

ANALOG

REFERENCE OUTPUT
Reference output current, lref' should not exceed

GROUND

100~A.

INTERFACING WITH CMOS LOGIC
The DAC-20810gic input stage requires approximately 1~A,
lin, and is capable of operation with inputs between -5V and
(V+ -0.7 Volts). The wide input voltage range allows direct
CMOS interface with no additional components required.

Full-Scale output voltage is trimmed using.the circuit configuration shown above. Low tempco metal-film resistors are
recommended. External components should be mounted
near the package to ensure good temperature tracking.
REFERENCE INPUT BYPASS
Low noise and fast settling operation can be obtained by
bypassing the Reference Input to Analog Ground with a
O.01~F monolithic capacitor.

EXTERNAL REFERENCES
Positive external reference voltages may be applied to the
reference input terminal to improve Full-Scale temperature
coefficient. External references are used when cascading
several converters or when tracking is required between system elements.

GROUNDlflfG
Separate digital and analog grounds have been provided for
optimum noise rejection. Best results will be obtained when
analog and digital ground are connected together at one
point only. This configuration ensures neglible digital currents flowing in analog ground.

MULTIPLYING OPERATION
Two-quadrant multiplying operation is achieved by applying
an analog input (0 to +10V) to the Reference input terminal.
The DAC output is the scaled product ofthe input voltage and
the digital code.

APPLICATIONS INFORMATION
SIGN - MAGNITUDE CODING TABLE
SIGN BIT
MSB

LOWER RESOLUTION APPLICATIONS
For applications requiring less than 8-Bit resolution, connect
unused logic inputs to ground.

LSB

+ FULL-SCALE
-1 LSB

UNIPOLAR OPERATION
Operation as an 8-Bit binary converter may be implemented
by connecting the Sign-Bit to +5V for positive Full-Scale
output, and OV for negative Full-Scale.

+ HALF-SCALE

0 0 0 0 0 0 0

ZERO-SCALE (+)

+5 VOLT OPTION
The output voltage range can be modified by connecting the
5V option pin (pin 13) to the analog output (pin 14). The 5V

11-86

0 0 0 0 0 0 0 0

ZERO-SCALE (-)
-HALF-SCALE

0
0

-FULL-SCALE
+1 LSB

0

0 0 0 0 0 0 0 0
0 0 0 0 0 0 0

1/86, Rev. A

------------I~ DAC-208 9-BIT VOLTAGE-OUTPUT D/A CONVERTER

UNIPOLAR OPERATION
5VOPTION

+15V

MSB 82 83 84

85 B6 87

88 58

REF-Ol

15

REF

DAC-208

IN
ANALOG DIGITAL

vANALOG
GROUND

V+

GROUND GROUND

10

12

OUTPUT

'6

9

RANGE
o -+5V

10Mf

SHOWN WITH
EXTERNAL REFERENCE

v+

APPLICATIONS
AUDIO ATTENUATOR
V+

2QkH

V-

+5V

3V :$;V REF :,,>10V

AUDIO
INPUT

10kn

€9

> __.....___

AUDIO
OUTPUT

---1~NEF

}-----<>8

CAG-20B

+5V
ANALOG
GROUND

DIGITAL
GROUND

INPUTS

Ql

Q2

03 04

05 06

Q7

as

74LS116

DIGITAL
INPUT

11-87

RESET

ENABLE

1/86, Rev. A

II

DAC-210
ll-BIT VOLTAGE-OUTPUT D/A CONVERTER
110 BITS PLUS SIGN)
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

•
•
•
•
•
•
•
•
•
•

The DAC-210 is a complete, monolithic 10-bit plus sign DAC
with a ± 10V output. A precision voltage reference, a logic
controlled polarity switch and output amplifier are included.
Linearity, monotonicity, and full-scale temperature coefficient are guaranteed over the full operating temperature
range. Ease of application is achieved by the total D/A system
specs given for nonlinearity and zero-scale offset. System
specs eliminate the complex error budget analysis required
by less "complete" DACs. Sign-magnitude coding minimizes
the "major-carry" zero-code errors inherent in offset coding
schemes. Reliability is enhanced by a monolithic design,
100% burn-in, and a hermetic DIP package. MIL-STD-883
Class B processing is available on -55 0 C to +1250 C grades.

Complete ........... Includes Reference and Op Amp
Bipolar Output. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ± 10V
Sign-Magnitude Coding
No Bipolar Offset Adjustment Required
10-Blt Linearity Maintained over Full Temperature
Multiplying Operation
Fast.............................. 1.5/Ls Settling Time
Monotonlclty Guaranteed
Reliable ............................. 100% Burned-In
Models with MIL-STD-883 Class B Processing Available

ORDERING INFORMATIONt
18·PIN HERMETIC DUAL IN-LINE PACKAGE
TEMPCO
(ppmrc)
±40
±60
±30 Typ

NL%FS

MILITARY

COMMERCIAL

±O.O5
±O.O5
±O.10

DAC210AX'
DAC210BX'

DAC210EX
DAC210FX
DAC210GX

PIN CONNECTION

• For devices processed in total compliance to MIL-STD-883. add /883 after
part number. Consult factory for 883 data sheet.
t All commercial and industrial temperature range parts are available with
burn-in. For ordenng information see 1986 Data Book. Section 2.

18-PIN DIP
(X-Suffix)

SIMPLIFIED SCHEMATIC
DIGITAL lOGIC INPUTS
(

BIT 1
1

W2

W3

W4

WS

W.

W.

WS

W9

23456789

BIT 10 '\
10

SIGN BIT

olGI
TAL
GND

R/2

v-

11·88

1/86, Rev. A

~ DAC-210 11-BIT VOLTAGE-OUTPUT D/A CONVERTER

ELECTRICAL CHARACTERISTICS- MILITARY AND COMMERCIAL GRADES atVs=±15V,-55°e::; TA::;+ 125°e,
for A and B grades. ooe::; TA::; +70oe for E, F and G grades, unless otherwise noted.
DAC-210AlE
PARAMETER

SYMBOL

Resolution

Including Sign

II

II

II

10

10

9

Nonlinearity
Zero-Scale
Offset Voltage
Bipolar Full Range
Voltage Symmetry

NL

TA = 2SoC
TA = Full Range
TA = Full Range (A onlyl

-

MAX

MIN

TVP

DAC-210G

MIN

Monotonlclty

TYP

DAC-210B/F

CONOITIONS

MAX

±O.OS
±O.OS
±0.07S

±O.OS
±OlO

Vzs

TA=2SoC
TA = Full Range

±O.OS
±0.06

±O.I
±O.I

V FRS

TA =2SoC
TA = Full Range

40
SO

60
70

Vzss

TA = Full Range

Tc

I ntemal Reference
External Reference

MIN

TYP

MAX

UNITS
Bits
Bits

±O.IO
%FS
%FS

60

mV

SO

(VFR+-I VFR-I I
Zero-Scale
Voltage Symmetry

2

mV

(Vzs+ - Vzs-I
Gain Tempco
Output Voltage
Range

VOR +

RL =2kO

VCR-

Differential
Nonlinearity

DNL

TA =25°C

Selliing Time

Ts

(Note 41

Reference Input
Slew Rate

SRREF

Reference Input
Impedance

Z,N

Reference Input
MultiplYing Range

IVR m

Reference Amplifier
Bandwidth

BW

Reference Output
Voltage

VREF

DACOutput
Current

10

Reference Output
Current

IREF

Output Slew Rate

SRo

Logic Input Current

liN

LogiC "0" Input
Voltage

V,NL

Logic "I" Input
Voltage

V,NH

Power Supply
Sensitivity
(Note 21

Pss

Positive Supply
Current
Negetlve Supply
Current

ForO 1% TYPical
Nonlinearity
(Note II

±40

±60

±IS
+IIS
-10.0

+10.0
-II.S

+10.0
-I1.S

+II.S
-10.0

±I

+10.0
-II.S

±I

ppm/oC
+11 S
-10.0

V

±I

LSB

I.S

I.S

I.S

I'S

I.S

I.S

I.S

VII's

200

200

200

MO

10

3

3

10

3

0

7.6

S

0

100

10

V

±2

7.6

S

0

100

±IO

±2

0.&
2.0

V

S

mA

100

10

10
-SVSV,SV+

I'A

10
±IO

±2

0.&
2.0

VII's
±IO

I'A

0.&

V

2.0

V

TA =2SoC

O.OlS

O.OS

O.OlS

O.OS

O.otS

0.1

TA = Full Range

o.olS

0.1

O.OlS

0.1

O.otS

0.1

1+

7

9

7

9

1-

-10

-12

-10

-12

%VFsN

NOTES:
1. Guaranteed by design.
2. Power Supplies - The DAC-210 Will operate within specifications for
power supplies ranging from ±I2V to ±I&V. Power supplies should be
bypassed near the package with a 0.1 I'F disk capacitor.

- ---------

-10

9

mA

-12

mA

3. Guaranteed by VOR test, RL = 2kO.
4. To within ±SmV of final sellied value, (±IO volt output step, RL = 2kO. I

1/86, Rev. A

11-89

-~

U

0

~

~

~......

C)

76

-

~

C)

MHz

(Note 31

~
0

±30
±30

±30

~

-------

------- ------

CS

II

-------------l~ DAC-210 11-BIT VOLTAGE-OUTPUT D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS

Operating Temperature Range
DAC-210A, B ....................... -55° C to + 125° C
DAC-210E, F, G ........................ O°C to +70°C
DICE Junction Temperature (Tj) ....... -65° C to + 150° C
Storage Temperature Range ........... -65°C to +150°C
V+ Supply to Analog Ground .................. 0 to + 18V
Analog Ground to Digital Ground. . . . . . . .. .. .. 0 to ±0.5V
Logic Inputs to Digital Ground ........ -5V to (V+ -0.7V)

V+ Supply to V- Supply ............................ 36V
Internal Reference Output Current .. . . . . . . . . . . . . .. 300ltA
Reference Input Voltage ....................... 0 to + 10V
Internal Power DisSipation ...................... 500mW
Derate Above 100°C .......................... 10mW/oC
Lead Temperature (Soldering, 60 sec) ............. 300°C
Output Short-Circuit Duration ................. Indefinite
(Short-circuit may be to ground or either supply.)
NOTE: Absolute rallngs apply to both DICE and packaged parts unless
otherwise noted.

DICE CHARACTERISTICS
1. 81 (MS8)
2.82
3.83
4.84
5. 85
6. 86
7. 87
8. 88

10.
11.
12.
13.
14.
15.

810 (LSB)
DIGITAL GROUND
yANALOG GROUND
ANALOG OUTPUT
REFERENCE INPUT

16. y+

17. REFERENCE OUTPUT
18. SIGN BIT
NOTE: For5voltoutputoptlon (+5Vonly) * Isconnectedtoanalogoutput **IS
connected to analog ground
9. 89

DIE SIZE 0.117 X 0.086 Inch, 10,062 sq. mils
(2.972 X 2.18 mm, 5.942 sq. mm)
For additional DICE information refer to
1986 Data Book, Section 2.

WAFER TEST LIMITS at Vs = ± 15V, + 10V full-scale output, TA =25°C, unless otherwise noted.
PARAMETER

CONDITIONS

Resolution

Bipolar Output
Unipolar Output

DAC-210N
LIMIT

DAC-210G
LIMIT

DAC-210GR
LIMIT

11
10

11
10

11
10

Bits MAX

±0,05

±01

±0,2

%FSMAX

±5

±10

±10

mVMAX

±1

±2

±2

mVMAX

UNITS

10

Monotonicity
Nonlinearity

Sign-Bit High, All Other

Zero-Scale Offset

Inputs Low

Zero-Scale Symmetry

Vzs + - Vzs-

Full-Scale Bipolar

Bits MIN

± 10V Full-Scale

±40

±80

±80

mVMAX

Power Supply
Rejection

Vs~±12Vto±18V

005

0,05

0,1

%VFslV MAX

Power Consumption

IOUT~O

300

300

300

mWMAX

0,8

0,8

0,8

V MAX

2

2

VMIN

11,5
10

115
10

11,5
10

V MAX
VMIN

-10
-115

-10
-11,5

-10
-11.5

V MAX
VMIN

±1

±1

±1

LSBMAX

Symmetry

Logic Input "0"
Logic Input "1"
V+ (Sign-Bit High)

Analog Output Voltage
(All Bits High)

V- (Sign-Bit Low)

Differential Nonlinearity

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to variations In assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate speCificatIOns based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V and + 10V full-scale output, unless otherwise noted.
DAC-210N
TYPICAL

DAC-210G
TYPICAL

DAC-210GR
TYPICAL

UNITS

Internal Reference

15

30

30

ppm/'C

ts

To ±112 LSB
10 Volt Step

1,5

1,5

1.5

I'S

liN

TA~25'C

PARAMETER

SYMBOL

CONDITIONS

Full-Scale Tempeo

TCVFS

Settling Time
(TA ~ 25'C)
Logic I nput Current

I'A
11-90

1/86, Rev. A

--------------l~ DAC-210 11-81T VOLTAGE-OUTPUT D/A CONVERTER

CONNECTION INFORMATION

TYPICAL APPLICATIONS

FULL-SCALE ADJUSTMENT - Full-scale output voltage
may be trimmed by use of a potentiometer and series resistor
as shown; however, best results will be obtained if a low
tempco resistor is used or if pot and resistor tempcos match.
Alternatively, a single pot of 2: 75kO may be used.

10-BIT SIGN-MAGNITUDE ADC
ANALOG
INPUT
±10V
+15V

CLOCK START +5V

SERIAL
DATA

FULL SCALE ADJUSTMENT CIRCUIT
DIGITAL INPUTS

,r--MSB

LSB SIGN BIT V+

REFERENCE
OUTPUT
FULL

REFERENCE

SCALE ~......::'N::,P;:.UT'--I---+_---<
ADJUST

POT
LOWTC

10Kr.!

62K"
ANALOG
GROUND

v-

OIGITACGROUND

REFERENCE INPUT BYPASS - Lowest noise and fastest
settling operation will be obtained by bypassing the reference input to analog ground with a O.01I'F disk capacitor.

62kn
NOTES

1 CONNECT END OF CONV TO START
FOR CONTINUOUS OPERATION
2 FOR NON-CONTINUOUS OPERATION,
HOLD START LOW FOR ONe CLOCK
CYCLE CONVERSIONS BEGIN ON THE
NeXT LOW TO HIGH TRANSITION
3 CONVERSION IS COMPLETED IN 12
CLOCK CYCLES

VARIABLE REFERENCES - Operation as a two-quadrant
multiplying DAC is achieved by applying an analog input
varying between 0 and + 10V to the reference input terminal.
The DAC output is then the scaled product of this voltage and
the digital input.

APPLICATIONS INFORMATION

,,,-

LOWER RESOLUTION APPLICATION - For applications
not requiring full 10-bit resolution, unused logic inputs
should be tied to ground.

GROUNDING - For optimum noise rejection', separate digital and analog grounds have been brought out. Best results
will be obtained if these grounds are connected together at
one point only, preferably at the power supply, so that the
large digital currents do not flow through the analog ground
path.

CAPACITIVE LOADING - The output operational amplifier
provides stable operation with capacitive loads up to 100pF.
REFERENCE OUTPUT - For best results, reference output
current should not exceed 100I'A.

SIGN - MAGNITUDE CODING TABLE
SIGN-BIT
+ FULL-SCALE
-1 LSB

MSB

INTERFACING WITH CMOS LOGIC - The DAC-210's logic
input stages require about 11'A and are capable of operation
with inputs between -5 volts and V+. This wide input voltage
range allows direct CMOS Interface with no additional
components.

LSB

1 1 1 1 1 1 1 1 1 1

+ HALF-SCALE

1 0 0 0 0 0 0 0 0 0

ZERO-SCALE (+)

o0 0
o0 0
o0

ZERO-SCALE (-)

0

-HALF-SCALE

0

-FULL-SCALE
+1 LSB

0

0 0 0 0 0 0 0

USE WITH EXTERNAL REFERENCES - Positive polarity
external reference voltages referred to analog ground may be
applied to the reference input terminal to improve full-scale
tempco, to provide tracking to other system elements, or to
slave a number of DAC-210's to the reference output of any
one of them.

0 0 0 0 0 0 0
0 0 0 0 000

111111111

11-91

--------

----

1/86, Rev_ A

II

DAC-312
12-BIT HIGH-SPEED MULTIPLYING
D/A CONVERTER
Pn.."clsion MOllollthlcs

Jill:.

FEATURES
•
•
•
•
•
•
•
•
•
•

excellent power supply rejection ratio of ±.001% Fs/%.:lV.
Operating over a power supply range of +5/-11 V to ±18Vthe
device consumes 225mW at the lower supply voltages with an
absolute maximum dissipation of 375mW at the higher
supply levels.

Guaranteed Differential Nonlinearity ........•.. 0.012%
Nonlinearity ..•..•.••.•.•.•.•.•.....••.....•. 0.025%
FaIt Settling Time ............................. 250nl
High Compliance ....................... -5V to +10V
Differential Outputs ......................... 0 to 4mA
Guaranteed Monotonlclty .. .. .. .. .. .. .. .. . .. ... 12 Bltl
low Full-Scale Tempco •..••. • • . • . . . • • . . • •• 10ppm/· C
Circuit Interface to TTL, CMOS, ECl, PMOS/NMOS
low Power Conlumptlon ...•.•.•..•.•.....••. 225mW
Indultry Standard AM8012 Pinout

With their guaranteed specifications, Single chip reliability
and low cost, the DAC-312 device makes excellent building
blocks for AID converters, data acquisition systems, video
display drivers, programmable test equipment and other
applications where low power consumption and complete
input/output versatility are required.

GENERAL DESCRIPTION

PIN CONNECTIONS & ORDERING INFORMATIONt

The DAC-312 series of 12-bit multiplying digital-to-analog
converters provide high speed with guaranteed performance
to 0.012% differential nonlinearity over the full commercial
operating temperature range.
Based on the segmented design approach pioneered by PMI
with the COMDAC@ line of data converters, the DAC-312
combines a 9-bit master DlA converter with a 3-bIt (MSB's)
segment generator to form an accurate 12-bit DlA converter
at low cost. This technique guarantees a very uniform step
size (up to ±1I2 LSB from the ideal), monotonicity to 12 bits
and integral nonlinearity to 0.05% at its differential current
outputs. In order to provide the same performance with a
12-blt R-2R ladder deSign, an integral nonlinearity over
temperature of 1/2 LSB (0.012%) would be required.

20-PIN HERMETIC
DUAl-IN-LiNE
PACKAGE
(R-Sufflx)

MODEL

TEMP RANGE

DNL

DAC312BR
DAC312FR
DAC312ER

-SS·CI +12S·C
O· CI+ 70· C
0·CI+70·C

±1 LSB
± 1 LSB
±1/2 LSB

Military Temperature Range Devices
With MIL-STD-883 Class B Processing

The 250nssettling time with low glitch energy and low power
consumption are achieved by careful attention to the circuit
design and stringent process controls. Direct interface with
all popular logiC families is achieved through the logic
threshold terminal.

MODEL

TEMP

DNL

DAC312BRl883

-SS·C/+12S·C

±1 LSB

• For devices processed In total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet.
tAli commercial and Industrial temperature range parts are available with
burn-In. For ordering Information see 1986 Data Book, Secllan 2.

High compliance and low drift characteristics (as low as
10ppm/·C) are also features of the DAC-312 along with an

FUNCTIONAL DIAGRAM
82

V(+I

83

84

B6

Be

87

B8

89

a10

811

LSB
812

-

~----~~~-+~~~~-Hr-~-+~~--~~Io

~--~1-~~+T~~++~~HH~Hr~~-r~~

16
COMP

17
V{-)

Manufactured under one or more of the following patents: 4,055,773; 4,058,740; 4,092,839

11-92

1/86, Rev. A

-----------t~ DAC-312 12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
Operating Temperature
DAC-312B .......................... -55°C to +125°C
DAC-312E, DAC-312F ................... 0° C to + 70° C
DICE Junction Temperature. • • • •• . • . •. -65° C to + 150° C
Storage Temperature (Tj ) •••••••••• • • ••

-65° C to + 125° C

Lead Temperature (Soldering, 60 sec) ............. 300° C
Power Supply Voltage ............................. ± 18V

ELECTRICAL CHARACTERISTICS

Logic Inputs

.............................. -5V to +18V

Analog Current Outputs

................... -8V to + 12V

Reference Inputs V 14, V15 •••••••••••••••.••••••. V- to V+
Reference Input Differential Voltage (V14, to VIS) ..•• ±18V
Reference Input Current (114) ..•.••.••••.••••..•• 1.25mA
NOTE: Absolute ratings apply to both DICE and packaged parts. unless
otherwise noted.

at Vs = ±15V, IREF= 1.0mA, -55°C:5 TA:5125°C for DAC-312B, O°C:5 TA:5 70°C for

DAC-312E, DAC-312F, unless otherwise noted. Output characteristics refer to both lOUT and lOUT.

DAC-312E
PARAMETER

SYMBOL

CONDITIONS

MIN

TVP

DAC-312B/F
MAX

MIN

Resolution

12

12

Monotonicity

12

12

Differential
Nonlinearity

D.N.L.

Nonlinearity

N.L.

Deviation from ideal step size

Deviation from ideal straight line
VREF = 10.OOOV
R'4 = R'5 = 10.000kO

Full-Scale Current

3.967

Full-Scale Tempco
Output Voltage
Compliance

VOC

D.N.L. Specification guaranteed
over compliance range

Full-Scale
Symmetry

I FSS

IIFSI-IIFSI

Zero-Scale Current

IZS

Settling Time

ts

All bits switched
50% point logic swing to
50% point output (See Note)

Propagation
Delay - all bits
Output Resistance

Output Capacitance
Logic Input Levels "0"

V,L

Logic Input Levels "I"

V,H

Sensitivity

dl/dt

R'41eQI = 8000
Cc = OpF (See Note)

PSSI FS+
PSSI FS-

V+=+13.5Vto+16.5V, V-=-15V
V-= -13.5V to -16.5V, V+ = +15V

Power Supply
Range

V-

Power Supply
Current

1+
11+

%FS

3.999

4.063

mA

±IO
±O.OOI

±40
±0.OO4

ppm/OC
%FS/OC

+10

Volts

+10

3.935

-5

±I

±0.4

0.10

500

250

500

ns

25

50

25

50

ns

>10

>10

MO

20

20

pF

0.8

0.8

40

-5

+18

-5

+18

o

-2

o

-0.5

-2

±0.0005
±O.00025

±O.OOI
±O.OOI

%FS/%aV

18
-10.8

Volts

-0.5

-

V+ = +5V. V- = -15V
V+ = +15V. V- = -15V

Volts

4

4

1-

Volts
Volts

2
40

V+=+5V. V-=-15V
V+=+15V, V-=-15V

±2

250

±0.OOO5
±0.00025

4.5
-18

V+

Power
Dissipation

±0.05

2

Logic Input Swing

Power Supply

±0.05

±30
±0.003

VLC = GND

Reference Bias
Current
Reference Input
Slew Rate

Bits
%FS
LSB

±IO
±O.OOI

V,N =-5 to +18V

Logic Input Current

Bits

010
To ±1/2 LSB, all bits
switched ON or OFF (See Note)

UNITS

±0.0250
±I

4.031

±0.4

MAX

±0.0125
±0.5

3.999

-5

TVP

±O.OOI
±O.OOI
18
-10.8

4.5
-18

3.3
-13.9
3.9
-13.9

7
-18
7
-18

3.3
-13.9
3.9
-13.9

7
-18

225
267

305
375

225
267

305
375

7

mA

-18
mW

NOTE: Guaranteed by design.

11-93

1/86, Rev. A

II

----------I~ DAC-312 12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

DICE CHARACTERISTICS
1. B1 (MSB)
2. B2
3. B3

11. B11
12. B12 (LSB)

5. B5
6. B6

13.
14.
15.
16.

7.87

17. V-

8. B8

18. 10
19.
20. V+

4.84

VLC/AGND
VREF (+)
VREF (-)
COMP

ro

9.89
10. B10

For additional DICE information refer to
1986 Data Book, Section 2.
DIE SIZE 0.140 X 0.095 inch, 13,300 sq. mils (3.56 X 2.41 mm, 8.58 sq. mm)

WAFER TEST LIMITS

at Vs

= ± 15V, IREF =1.0mA, TA =25 ·C, unless otherwise noted. Output characteristics refer to

both lOUT and lOUT'

PARAMETER

SYMBOL

CONDITIONS

DAC-312N
LIMIT

DAC-312G
LIMIT

UNITS

12

12

Bits MIN

Resolution
Monotonlcity
Nonlinearity
Output Voltage
Compliance

VOC

Full-Scale
Current

Full-Scale Current
Change <1/2 LSB

VREF = 10.0OQV
R,., R,s = 10 OOOkfl

12

12

Bits MIN

±005

±0.05

%FSMAX

+10
-5

+10
-5

V MAX
VMIN

4.031
3.967

4.063
3.935

mAMAX
mAMIN
~AMAX

Full-Scale Symmetry

I FSS

±1

±2

Zero-Scale Current

I zs

0.1

0.1

~AMAX

±0.012
±1/2

±0025
±1

%FSMAX
Bits (LSB) MAX

0.8

0.8

V MAX

+18
-5

VMAX
VMIN

Differential
Nonlineanty

DNL

Deviation from
ideal step size

Logic Input Levels "0"

V,L

VLC = GND

Logic Input Levels "1"

V,H

VLC = GND

VMIN

Logic Input SWing

V's

+18
-5

Reference Bias
Current

115

-2

-2

~AMAX

Power Supply
SensitivIty

PSSI FS+
PSSI FS _

V+ = +13.5V to +16.5V. V-= -15V
V- = -13.5V to -16.5V, V+ = +15V

±0.001
±0.001

±0.001
±0.001

%/%MAX

Power Supply
Current

1+
1-

Vs =±15V
'REFS 1.0mA

-18

-18

Power
Dissipation

Po

Vs =+15V
I REF ",0mA

375

375

mAMAX
mWMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations In assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS

at 25°C; Vs = ±15V, and IREF

= 1.0mA, unless otherwise noted. Output

characteristics refer to both lOUT and lOUT'

PARAMETER

SYMBOL

Reference Input
Slew Rate

dl/dt

Propagation Delay

tpLH • tpHL

Any Bit

Settling Time

ts

To ±1/2 LSB. All
Bits SWitched ON
orOFF.

Full-Scale

TC ,FS

DAC-312N
TYPICAL

CONDITIONS

DAC-312G
TYPICAL

8

11-94

UNITS
mA/~s

25

25

ns

250

250

ns

±10

±10

ppm/·C

1/86, Rev_ A

----------l~ DAC-31212-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

TYPICAL PERFORMANCE CHARACTERISTICS

OUTPUT CURRENT vs
OUTPUT VOLTAGE
(OUTPUT VOLTAGE COMPLIANCE)
50

36 VH"!-15V

1
t-

~

24

"

20

t::>

I!:::>
0

TA =TMIN TOTMAX

I

r

I

08

00
-14

I
I
-10

-6

-

~::>

-1 1

I

I
I

I
I

I

I

I

-

N~TESI

f--

r rI I
1

I

I

r

10

-14

14

r 1
-10

-6

1

ALL BITS ON

1

I

I

1

1

POWER SUPPLY CURRENT
vs TEMPERATURE

/"
V

-138
-136

~~

-4

1

POWER SUPPLY CURRENT vs
POWER SUPPLY VOLTAGE
-142

-2

I
10

-2

..;.,

J

V 1S REFERENCE COMMON MODE (VOLTS)

-140

FOR OTHER V!-! OR !REF'

~~;~~;~~TL.i~~~~~~~E:
~~,~
"
::-

:

1

-!OTMA~_

TA=+MIN

'REF = 0 2mA

04

:

IS ALWAYS iV-)+1 B+(lREFX3kSH

DB

-2

f' ! :~:~~S~I~~~~~N~ci~'6TFES ~ I
'I OUTPUT VOLTAGE WITH
•
VH" -16V, I REF"':; 1 OmA
I

10

IS ALWAYS (V+) -1 25V
NEGATIVE COMMON MODE VOL rAGE

'REF=05mA

16

16~~nl~

:: ~::::;i~~:"(\~}f~..:i+'-+_-l

POSITIVE COMMON MODE VOLTAGE

OUTPUT VOLTAGE (VOLTS)

~L-~~~~~~~~~
14

-100

-50

0

50

100

150

200

TEMPERATURE fC)

TRUE AND COMPLEMENTARY
OUTPUT OPERATION

Vs" ±15V
- I REF = 1 DmA-

"

I

-134
-132

-13

"

-14
-15

r--

12

I
I

a:: -12

~

rl

0

-"

-10

-

"
-

:~~11 WHffii
~

28

32

r

a:
a: 24
::>
t- 20

I

IRE~=02lA

I

12

1l~

t-

~

04

a:

.,

36

E

ALL, BITS ON

IREF =05mA

16

I I

'REF = 1 OmA

40

.I

I

I

I

28

a:
a:

::>

H~I-I.J08V I

32

42

I

IREF - 1 OmA

40

OUTPUT COMPLIANCE
vs TEMPERATURE

REFERENCE AMPLIFIER
COMMON-MODE RANGE

o

8

'~EF ~ 1 O~A

J(-l

10

12

14

16

18

20

] EtH·jj J
-100

REFERENCE AMPLIFIER
SMALL-SIGNAL
FREQUENCY RESPONSE
11111111

10

JJillllL

10

I

-8

-10
-12
001

150

200

250

GAIN ACCURACY vs
REFERENCE CURRENT
10

I 11111111
I 11111111

I I
I I

~

R14 = R15 - lkn
RL = loon, Cc = 5pF
V15 = oV
VIN =1Vp-p
CENTERED AT +500mV

"'''
~

--<

100

12

I

RL = 200n.
VR1S= OV
VIN = 5OmVp·p
CENTERED AT +l00mV

-4

50

REFERENCE AMPLIFIER
LARGE-SIGNAL
FREQUENCY RESPONSE

R14 - R15 = lkn

-2

0

~

>

~ 04
a:
03

-8

.,

-10

"

01

--<

1 CC OpF
2 CC=5pF
3 CC = lOpF

FREQUENCY (MHz)

"

~ 07
~ 06
0
~ 05

a

-4

I

10

09

~ 08

"

-2

Ii'll
01

(111111111111)

IREF= 1 OmA

TEMPERATURE (OC)

POWER SUPPLY VOLTAGE (VOLTS)

12

-50

(0000 0000 0000)

-12
001

~
z 02

01

10

FREQUENCY (MHz:)

11-95

0

0

01

02

03

04

05

06

07

08 09

10

REFERENCE INPUT CURRENT (mAl

1/86, Rev. A

----------I~ DAC-312 12-81T HIGH-SPEED MULTIPLYING D/A CONVERTER

BASIC CONNECTIONS
NEGATIVE LOW IMPEDANCE OUTPUT OPERATION

POSITIVE LOW IMPEDANCE OUTPUT OPERATION

OAe-3l2
OAC-312

409.
'FR" 4096

FOR COMPLEMENTARY OUTPUT (OPERATION AS A NEGATIVE LOGIC DAC),

CONNECT INVERTING INPUT OF OP-AMP TO
TO GROUND

FOR COMPLEMENTARY OUTPUT (OPERATION AS A NEGATIVE LOGIC oAe),
CONNECT NON-INVERTING INPUT OF OP-AMP TO
(PIN 19), CONNECT

ro (PIN 191. CONNECT '0 (PIN 18)

ro

'0 (PIN 18) TO GROUND

ACCOMMODATING BIPOLAR REFERENCES

BASIC NEGATIVE REFERENCE OPERATION
RREF

r,...------..

VREFH

REF

14~--------------'

-=

DAC-312

o-___"R,,1·r--'5..
L..-_ _ _ _J

14

,.

R'N

"::"

x 4 X 'REF

NOTE'

RREF SETS IFS: R16 IS fOR A
DAC-312

BIAS CURRENT CANCELLATION

'REF;;" PEAK NEGATIVE SWING OF 'IN

RECOMMENDED FULL-SCALE
ADJUSTMENT CIRCUIT
LOWTC

14

-

Rl.

(OPTIONAL)

,.

VREF o-~__4~5k~"~~14r--------------,
+5 ooov

DAC -312

IREF{+) .... 'mA

HIGH INPUT
IMPEDANCE

POT

BASIC POSITIVE REFERENCE OPERATION
LSB
81 B2 B3 B4 B5 S6 87 BB 89 810 S11 812

IREFR15

APPROX.
Skll

PULSED REFERENCE OPERATION

MS.

V REF (+)

«

J

RREF
(R16) 14

t

lOkll

VREF (+) MUST BE ABOVE PEAK POSITIVE SWING OF VIN

V REF (+)

OAC-312

39k~15

:! R

'0

(OPTIONAL RESISTOR
REF

FOR OFFSET INPUTS

O~~R~'N~~'~___~4!:Jr-------------~ 18

V
L

REQ .... 800n

15

Rp

DAC-312

16

19

20 13
FOR FIXED REFERENCE,
TTL OPERATION,
TYPICAL VALUES ARE.

ov.JL

VREF· +10.000V

VH
V REF (+)
4096
IFR = RREF
X 4096 X 4

TYPICAL VALUES:
RIN "1kSl

V{+) VLC RREF = 10 OOOkO
R15 = RREF

Rp = 4kO
VIN(+) = 1V

Cc' 001.F
V LC " OV (GROUND)

--L-.

ro

1

10 +
= IFR FOR ALL
LOGIC STATES

REO· RIN

11-96

1

1

+ R; + RREF

1/86, Rev. A

----------IIEMD

DAC-312 12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

BASIC CONNECTIONS
INTERFACING WITH VARIOUS LOGIC FAMILIES
TTL

ECl

HTl

V TH '" +1 4V

V(+)

OAe-3l2
2Dkn

13kn

V lC
13

TO PIN 13

Vle

V le

R

i

20kn

+15V

TO PIN 13

V TH '" V LC + 14V

39kn

62kn

400pA

FOR +15V CMOS
V TH =+76V

lkH

-S2V

Vle

V

1

z = 6 2V

NOTES
1 seT THE VOLTAGE "Au TO THE DESIRED LOGIC
INPUT SWITCHING THRESHOLD

01M

-=-

'

2

ALLOWABLE RANGE OF LOGIC THRESHOLD IS
TYPICALLY -5V TO +135V WHEN OPERATING
THE OAe ON :t15V SUPPLIES

BIPOLAR OFFSET (TRUE ZERO)
ROFF

R1

2 aOOmA

5000kn

5k"

II

REF-ol
+10V
OAC-3l2

VREF
R14"'-- =R15
1.0mA

812

R15

R2

OPTIONAL FOR
2's COMPLEMENT

10kn

5k"

OPERATION

MS8

lS8

NOTE.
CODE MAY BE COMPLEMENTED BY REVERSING '0 &

CODE FORMAT

OUTPUT SCALE

Offset binary;
true zero output

Positive full-scale
Positive full-scale -LSB
+LSB
Zero-scale
-LSB
Negative full-scale +LSB
Negative full-scale

2's complement;
true zero output
MSB complemented
(Need inverter at 81 )

iQ

MSB
Bl
B2 B3 B4 B5 B6 B7 B8

Positive full-scale
Positive full-scale -LSB
+1 LSB
Zero-scale
-1 LSB
Negative full-scale + LSB
Negative full-scale

1
0
0
1
0
0

1
0
0

1
0
0

1
0
0

1
0
0

0
0

0
0

0
0

0
0

0
1
0
0

1
0
0

1
0
0

1
0

0
1
0
0

B9 Bl0 Bll

1
0
0

1
0
0

0
0

0
1
0
0

LSB
10
B12 (mAl

(mAl

VOUT

3.999
3998

0000
0.001

99951

-9.9951

0

2.001
2.000
1.999

99902
1998
0.0049
1999
0.000
2.000 -0.0049

0
0

0001
0.000

3998
3999

3999
3.998

0000

99951

0001
1998
1999

99902
00049
0000
-00049

1
0
0

0

0

0
1
0

2001
2000
1999

2000

0.001
0000

3998
3999

----

~---------

-10.000

-9.9951

-10000

1/86, Rev. A

11-97

--~-~--'"-~------

10

--~

- _ . _ - - .-. . - - - - -

----

-----------I1fMD

DAC-312 12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

BASIC CONNECTIONS
BASIC UNIPOLAR OPERATION
A1
25k
A1'

10k"
REF-01
+10V
OAC-312

B12

V REF

R14"'-- =R15
1.0mA

A15

A2

10kO

2.51<

MSB

LSB

NOTE
CODE MAY BE COMPLEMENTED BY REVERSING '0 &

CODE FORMAT

OUTPUT SCALE

Straight Binary,

Positive full-scale
Positive full-Scale - LSB
LSB
Zero-scale

unipolar with true

mput code, true
zero output

Complementary binary;
unipolar with
complementary Input
code, true zero output

Positive full-scale
Positive full-scale -LSB
LSB
Zero-scale

iQ

MSB
B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11
1

o
o
o
o

1
1
1
1 1
1 1
1
1
000000000
0000000000

o o o o o o
o o o o o o

o o
o o

1

1

1

o
o

1

LSB 10
B12 (mA)

1

o
1

o
o
1

o
1

10
(mA)

VOUT

3999
3998
0.001
0000

0000
0.001
3998
3.999

9.9976
9.9951
00024
00000

0000
0001
3998
3999

3.999
3998
0001
0.000

9.9976
99951
0.0024
00000

SYMMETRICAL OFFSET OPERATION
A1
25k

?»-<_ _

A1'
10kn

~\
IO~

REF-Ol
+10V
DAC-312

V_OOUT

.'2
V REF

R14=-- =R15
10mA

A15
10kn

M~B
CODE MAY BE COMPLEMENTED BY REVERSING

OUTPUT SCALE

Straight offset binary,

Positive full-scale

symmetrical about zero,
no true zero output

Positive full-scale -LSB
(+ I Zero-scale
I-I Zero-scale
Negative full-scale -LSB
Negative full-scale

o
o
o

Positive full-scale

o

Positive full-scale -LSB
I +I Zero-scale
I-I Zero-scale
NegatIVe full-scale -LSB
Negative full-scale

'a & iQ

MSB
B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 B11

CODE FORMAT

no true zero output
MSB complemented
(need Inverter at B1)

125k

LSB

NOTE

1 's complement,
symmetrical about zero,

R2

R3
25k

"
OPTIONAL
~.

1
1
1

o
o
1

1
1

1

1

1
1

1
1
0
1

1

1

1

1

1

LSB 10
B12 (mA)

1

1

1

o
o

1

1
1

1

1
0
1

1
1

o

o

o

1

1

1

1
0
1

o

1

o

0
0

o
o

0
0

o o

o o

0
0

o o
o o

o
o

1
1

1
1

1
1

1
1

1
1

1
1

1
1

1

1

1

o

1

o
o
1

o

o

1

o o
1

1

1
1

1

1

o o o o o o o o o
1

1

1

o

o

1

1

1

o o o o o o o o
o o o o o o o o

1

11-98

1

1

1

o

o

o

1

o

10
(mA)

VOUT

3999 000
99976
3998 0001
99927
2000 1999 00024
1999 2000 -00024
0.001 3998 -99927
0000 3.999 -9.9976
3999 0000 9.9976
3998 0001
99927
2000 1999 00024
1999 2000 -00024
0001 3998 -99927
0000 3999 -99976

1/86, Rev. A

----------I~ DAC-312 12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

APPLICATIONS INFORMATION
REFERENCE AMPLIFIER SETUP
The DAC-312 is a multiplying D/A converter in which the
output current is the product of a digital number and the
input reference current. The reference current may be fixed
or may vary from nearly zero to +1.OmA. The full range output
current is a linear function of the reference current and is
given by:

4095
IFR = 4096 X 4 X (IREF) = 3.999IREF'
where I REF = 114
I n positive reference applications, an external positive reference voltage forces current through R14 into the V REF (+)
terminal (pin 14) of the reference amplifier. Alternatively, a
negative reference may be applied to V REF (-) at pin 15 Reference current flows from ground through R14 Into VREF(+) as
in the positive reference case. This negative reference connection has the advantage of a very high impedance presented at pin 15. The voltage at pin 14 is equal to and tracks
the voltage at Pin 15 due to the high gain of the internal
reference amplifier. R15 (nominally equal to R14) is used to
cancel bias current errors.
Bipolar references may be accomodated by offsetting VREF
or pin 15. The negative common-mode range of the reference
amplifier is given by: VCM-= V- plus (I REF X 3kO) plus 1.23V.
The positive common-mode range is V+ less 1.BV.
When a DC reference is used, a reference bypass capacitor is
recommended. A 5.0V TTL logic su pply is not recommended
as a reference. If a regulated power supply is used as a
reference, R14 should be split into two resistors with the
junction bypassed to ground with a 0.1/LF capacitor.
For most applications the tight relationship between IREFand
IFS will eliminate the need for trimming IREF. If required, full
scale trimming may be accomplished by adjusting the value
of R14, or by using a potentiometer for R14. An improved
method of full-scale trimming which eliminates potentiometer T.C. effects is shown in the Recommended Full-Scale
Adjustment circuit.
The reference amplifier must be compensated by uSing a
capacitor from pin 16 to V-. For fixed reference operation, a
0.01/LF capacitor is recommended. For variable reference
applications, see section entitled "Reference Amplifier Compensation for Multiplying Applications."
MULTIPLYING OPERATION

The DAC-312 provides excellent multiplying performance
with an extremely linear relationship between IFS and IREF
over a range of 1mA to 1/LA Monotonic operation is maintained over a typical range of IREF from 100/LA to 1.OmA.
Although some degradation of gain accuracy will be realized

at reduced values of IREF. (See Gain Accuracy vs Reference
Current).
REFERENCE AMPLIFIER COMPENSATION FOR
MULTIPLYING APPLICATIONS

AC reference applications will require the reference amplifier
to be compensated using a capacitor from pin 16 to V-. The
value of this capacitor depends on the impedance presented
to pin 14 for R14 values of 1.0, 2.5 and 5.0kO, minimum values
of Cc are 5,10, and 25pF. Largervalues of R14 require proportionately increased values of Cc for proper phase margin.
For fastest response to a pulse, low values of R14 enabling
small Cc values should be used. If pin 14 is driven by a high
impedance such as a transistor current source, none of the
above values will suffice and the amplifier must be heavily
compensated which will decrease overall bandwidth and
slew rate For R14 = 1kO and Cc=5pF, the reference amplifier
slews at 4mA//Ls enabling a transition from IREF= 0 to IREF=
1 mA In 250ns.
Operation with pulse inputs to the reference amplifier may
be accommodated by an alternate compensation scheme.
This technique provides lowest full-scale transition times.
An internal clamp allows quick recovery of the reference
amplifier from a cutoff (IREF = 0) condition. Full-scale
transition (0 to 1mAl occurs in 62.5ns when the equivalent
impedance at pin 14 is BOOO and Cc = O. This yields a
reference slew rate of BmA//Ls which is relatively independent of RIN and VIN values.
LOGIC INPUTS

The DAC-312 design incorporates a unique logic input circuit
which enables direct interface to all popular logic families
and provides maximum noise immunity. This feature is made
possible by the large input swing capability, 40/LA logic input
current, and completely adjustable logic threshold voltage.
For V- = -15V, the logic inputs may swing between -5 and
+10V. This enables direct interface with +15V CMOS logic,
even when the DAC-312 is powered from a +5V supply. Minimum input logic swing and minimum logic threshold voltage
are given by : V- plus (IREF X 3kO) plus 1.BV. The logic
threshold may be adjusted over a wide range by placing an
appropriate voltage at the logic threshold control pin (pin 13,
VLc). The appropriate graph shows the relationship between
VLC and VTH over the temperature range, with VTH nominally
1.4 above VLC. For TTL interface, simply ground pin 13. When
interfacing ECL, an IREF:S 1mA is recommended. For interfacing other logic families, see block titled "Interfacing With
Various Logic Families". For general setup of the logic control circuit, it should be noted that pin 13 will sink 7mA typical;
external circuitry should be designed to accommodate this
current.

11-99

1/86, Rev_ A

81

-----------IIEHD

DAC-312 12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

ANALOG OUTPUT CURRENTS

Both true and complemented output sink currents are provided where '0+ iQ= I FR. Current appears atthe "true" output
when a "1" is applied to each logic input. As the binary count
increases, the sink current at pin 18 increases proportionally,
in the fashion of a "positive logic" 01 A converter. When a "0"
is applied to any input bit, that current is turned off at pin 18
and turned on at pin 19. A decreasing logic count increases iQ
as in a negative or inverted logic 01 A converter. Both outputs
may be used simultaneously. If one of the outputs is not
required it must still be connected to ground or to a point
capable of sourcing I FR; do not leave an unused output pin
open.
Both outputs have an extremely wide voltage compliance
enabling fast direct current-to-voltage conversion through a
resistor tied to ground or other voltage source. Positive compliance is 25V above V- and is independent of the positive
supply. Negative compliance is + 10V above V-.
The dual outputs enable double the usual peak-to-peak load
swing when driving loads in quasi-differential fashion. This
feature is especially useful in cable driving, CRT deflection
and in other balanced applications such as driving centertapped coils and transformers.
POWER SUPPLIES

The OAC-312 operates over a wide range of power supply
voltages from a total supply of 20V to 36V. When operating
with V-supplies of-10Vor less, IREF:51 mA is recommended.
Low reference current operation decreases power consumption and increases negative compliance, reference amplifier
negative common-mode range, negative logic input range,
and negative logic threshold range; consult the various figures for guidance. For example, operation at -9V with 'REF =
1mA is not recommended because negative output compliance would be reduced to near zero. Operation from lower
supplies is possible, however at least 8V total must be applied
to insure turn-on of the internal bias network.
Symmetrical supplies are not required, as the OAC-312 is
quite insensitive to variations in supply voltage. Battery
operation is feasible as no ground connection is required;
however, an artificial ground may be used to insure logic
swings, etc. remain between acceptable limits.

TEMPERATURE PERFORMANCE
The nonlinearity and monotonicity specifications of the
OAC-312 are guaranteed to apply over the entire rated operating temperature range. Full-Scale output current drift is
tight, typically ± 10ppmfO C, with zero-scale output current
and drift essentially negligible compared to 1/2 LSB.
The temperature coefficient of the reference resistor R14
should match and track that of the output resistor for min-

imum overall full-scale drift. Settling times of the OAC-312
decrease approximately 10% at-55° C; at+ 125 0 C an increase
of about 1S% is typical.
SETTLING TIME

The OAC-312 is capable of extremely fast settling times,
typically 2S0ns at I REF= 1.0mA. Judicious circuit design and
careful board layout must be employed to obtain full performance potential during testing and application. The logic
switch design enables propagation delays of only 2Sns for
each of the 12 bits. Settling time to within 1/2 LSB of the LSB
is therefore 2Sns, with each progressively larger bit taking
successively longer. The MSB settles in 2S0ns, thus determining the overall settling time of 2S0ns. Settling to 10-bit
accuracy requires about90 to 130ns. The output capacitance
of the OAC-312 including the package is approximately 20pF;
therefore, the output RC time constant dominates settling
time if RL > soon
Settling time and propagation delay are relatively insensitive
to logic input amplitude and rise and fall times, due to the
high gain of the logic switches. Settling time also remains
essentially constant for IREF values down to O.SmA, with
gradual increases for lower IREF values lies in the ability to
attain a given output level with lower load resistors, thus
reducing the output RC time constant.
Measurement of the settling time requires the ability to accurately resolve ± 1/2 LSB of current, which is ±SOOnA for4mA
FSR. In order to assure the measurement is of the actual
settling time and not the R.C. time of the output network, the
resistive termination on the output of the OAC must be
SOO ohms or less. This does, however, place certain limitations on the testing apparatus. At IREF values of less than
O.SmA, it is difficult to prevent RC damping of the output and
maintain adequate sensitivity. Because the OAC-312 has 8
equal current sources for the 3 most significant bits, the
major carry occurs at the code change of 000111111111 to
111000000000. The worst case settling time occurs at the zero
to full-scale transition and it requires 9.2 time constants for
the OAC output to settle to within ± 1/2 LSB (0.012S%) of its
final value.
The OAC-312 switching transients or "glitches" are on the
order of SOOmV-ns This is most evident when switching
through the major carry and may be further reduced by
adding small capacitive loads at the output with a minor
sacrifice in transition speeds.
Fastest operation can be obtained by using short leads, minimizing output capacitance and load resistor values, and by
adequate bypassing at the supply, reference, and VLC terminals. Supplies do not require large electrolytic bypass capacitors as the supply current drain is independent of input logic
states; 0.1 /LF capacitors at the supply pins provide full transient protection.

11-100

1/86, Rev. A

-----------i~ DAC-312

12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

DIFFERENTIAL vs INTEGRAL NONLINEARITY
Integral nonlinearity. for the purposes of the discussion. refers to the "straightness" of the line drawn through the individual response pOints of a data converter. Differential nonlinearity. on the other hand. refers to the deviation of the
spacing of the adjacent points from a 1 LSB ideal spacing.
Both may be expressed as either a percentge of full-scale
output or as fractional LSBs or both. The following figures
define the manner in which these parameters are specified.
The left figure shows a portion of the transfer curve of a DAC
with 1/2 LSB INL and the (implied) DNL spec of 1 LSB. Below
this is a graphic representation of the way this would appear
on a CRT. for example. if the DI A Converter output were to be
applied to the Y input of a CRT as shown in the application
schematic titled "CRT Display Driver". On the right is a portion of the transfer curve of a DAC specified for 2 LSB INL
with 1/2 LSB DNL specified and the graphic display below it.

One of the characteristics of an R-2R DAC in standard form is
that any transition which causes a zero LSB change (i.e. the
same output for two different codes) will exhibit the same
output each time that transition occurs. The same holds true
for transitions causing a 2 LSB change. These two problem
transitions are allowable for the standard definition of monotonicity and also allow the device to be specified very tightly
for INL. The major problem arising from this error type is in
AID converter implementations. Inputs producing the same
output are now represented by ambiguous output codes for
an identical input. Also. 2 LSB gaps can cause large errors at
those input levels (assuming 1/2 LSB quantizing levels). It
can be seen from the two figures that the DNL specified DI A
converter will yield much finer grained data than the INL
specified part. thus improving the ability of the AID to resolve
changes in the analog input.

DIFFERENTIAL LINEARITY COMPARISON
D/A CONVERTER WITH

DIA CONVERTER WITH
±2 LSB INL, ±1/2 LSB DNL

±1/2 LSB INL, ±1 LSB DNL

c IDEAL OUTPUTS
• ACTUAL OUTPUTS

c IDEAL OUTPUTS
• ACTUAL OUTPUTS

SEGMENT OF l2-BIT DAe
TRANSFER CURVE FOR

I

INL = ±2lse
DNL = ±1I2 LSB

0000 0010 0100 0110 1000

0001

0011 0101

0111

1010

1001

1100 "10

1011

1101

0010 0100 0110 1000 1010

1111

0011

0101 0111 1001

1'00 1110

1011 1101

0000

1111

0001

DIGITAL INPUTS

DIGITAL INPUTS

VIDEO DEFLECTION BY DAC's

VIDEO DEFLECTION BY DAC's

-ENLARGED "POSITIONAL" OUTPUTS

ENLARGED "POSITIONAL" OUTPUTS

11-101

1/86, Rev. A

----------I~ DAC-312 12-BIT HIGH-SPEED MULTIPLYING D/A CONVERTER

DESCRIPTION OF OPERATION
The OAC-312 is divided into two major sections, an 8segment generator and a 9-bit master/slave O/A Converter.
In operation the device performs as follows (See Simplified
Schematic):

EXPANDED TRANSFER CHARACTERISTIC
SEGMENT (001 010011)

The three most significant bits (MSB's) are inputs to a 3-to-8
line decoder. The selected resistor (R5 in the figure) is connected to the master/slave 9-bit O/A Converter. All lower
order resistors (R1 through R4) are summed into the 10 line,
while all higher order resistors (RS through R8) are summed
into the TO line. The R5 current supplies 512 steps of current
(0 to 0.499mA for a 1mA reference current) which are also
summed into the 10 or TOlines depending on the bits selected.
In the figure, the code selected is: 100 110000000. Therefore,
2mA (4 X 0.5mA/segment) +0.375mA (from master/slave 0/A
Converter) are summed into 10 giving an 10 of 2.375mA. TO
has a current of 1.S25mA with this code. As the three MSB's
are incremented, each successively higher code adds 0.5mA
to 10 and subtracts 0.5mA from iQ, with the selected resistor
feeding its current to the master/slave 01 A Converter; thus
each increment of the 3 MSB's allows the current In the 9-bit
O/A Converter to be added to a pedestal consisting of the
sum of all lower order currents from the segment generator.
This configuration guarantees monotonicity.

1501
1600
1499

1001
0.99

END SEGMENT (001)

t(01000000~)-;010 111111111,J

tlo"OOOOooooo)

DIGITAL INPUT

SIMPLIFIED SCHEMATIC

V(+)

B2

.3

84

B5

86

87

98

B9

Bl0

Bl1

19

18

17

COMP

V(-)

11-102

'0

1/86, Rev. A

----------l~ DAC·312 12·BIT HIGH· SPEED MULTIPLYING DIA CONVERTER

12-BIT FAST AID CONVERTER

12-81T FAST AID CONVERTER

SERIAL
DATA OUT
CONVERSION TIME vs ACCURACY

125

CLOCK

\

\
LSB

100

(WORST CASE)\
OAe-3l2
CMP-05

!>-

075

:>

050

~

u

f---

~

REF-Ol

025

OAe-3l2

0.00
100

COMP

1\

DAC-~

CMP-05
{TVP}

200

300

\

'"

400

1\

500

600

700

800

CONVERSION TIME PER TRIAL (nS)

10000kS1

CONVERSION
TIME (ns)

NOTE
DEVICE(S) CONNECTED TO ANALOG INPUT MUST BE CAPABLE OF SOURCING 40mA
A BUFFER (eg aUF·03) MAY BE REQUIRED

TYP

WORST
CASE

SAR

33

55

CMP-05

92

125

TOTAL

375ns

680ns

X 13

4

9~s

8

8~s

II

11·103

1/86, Rev. A

DAC-888
BYTEDAC® 8-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE"
MULTIPLYING D/A CONVERTER
PreCISIOIl MUIlolithics Inc

FEATURES
•
•
•
•
•
•
•
•
•

latches may also be operated in a transparent mode by
holding both control pins low. Additionally, the DAC-888 has
a data hold time requirement of zero nanoseconds.

8-Blt Level TrIggered Latch
8-Blt p.P Compatible
Easily Interfaced to All 8-Blt Processors
nL Logic Compatible
~ and WR Inputs
High Output Impedance and Compliance
Proven DAC-08 Analog Flexibility and Reliability
Nonlinearity to ±O.1% Max
Low Power Dissipation ••.••.••••.•••••••••• 134mW

The Analog section consists of a "Field-Proven" DAC-08 01 A
Converter. Monotonic multiplying performance is attained
over a wide 20 to 1 reference current range. Matching to
within 1 LSB between reference and full-scale currents
eliminates full-scale adjustment in most applications.
DAC-888 applications include graphic display drivers, highspeed modems, AID converters, programmable waveform
generators and power supplies, analog meter drivers, audio
encoders and programmable attenuators; and other applications where low cost, high speed and buffered flexibility are
required.

ORDERING INFORMATIONt
l8-PIN HERMETIC DUAL-IN-LiNE PACKAGE
NL
'laFS

MILITARY
TEMP.

INDUSTRIAL
TEMP.

0.1
0.19

DAC888AX'
DAC888BX'

DAC888EX
DAC888FX

PIN CONNECTIONS

• For devices processed In total compliance to M I L-STO-883, add 1883 after
part number. Consult factory for 883 data sheet.
t All commercial and industrial temperature range parts are available with
burn-In For ordering Information see 1986 Data Book, Section 2

18-PIN HERMETIC
DUAL-IN-LiNE
PACKAGE
(X-Suffix)

GENERAL DESCRIPTION
The BYTEDAC@> DAC-888 is a buffered 8-bit digital-toanalog converter designed specifically for 8-bit bus oriented
systems. The data inputs are connected to level-triggered
latches. Two active-low control pins are provided for ease of
interface to virtually all available microprocessors. The

DB,

7

LSB DBO

8

FUNCTIONAL DIAGRAM
D~ DBs DBs 084 DB3

DB2 DB, DBO

MSS

LSB

....._ _ _ _ _+'4'---0 'OUT
8-BIT MULTIPLYING OAC

COMP

v-

GND

v+

Manufactured under one or more of the following patents 4,055,773,4,056,740,4,092,639

11-104

1/86, Rev. A

~ DAC-88a BYTEDACIII a-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
Operating Temperature
DAC-888 AlB ........................ -Sso C to + 12So C
DAC-888 ElF .......................... -2SO C to +8So C
DICE Junction Temperature (T)l .....•• -6So C to +150°C
Storage Temperature •....•••....•.•.. -65° C to + lS00 C
Power Dissipation .....•..•..•.•..•.....•.••.•.. 300mW
Derate above 100° C •.•..•.••••••..•.••.••.... 10mW/o C
Lead Temperature (Soldering, 60 sec) ..•..••.••... 300° C
V+ Supply to V- Supply .......................... , 18.1V

Logic Inputs ................................ OV to S.SV
Analog Current Outputs .•...•.•.••.•..•.•.•••••• -SmA
Reference Inputs (V'O to V,,) ...•.•...•.••.•.••• V- to V+
Reference Input Differential Voltage
(V1QtoV,,) •..•.•••....•••..•.•....•..•.......•.. ±lSV
Reference Input Current ...•..••.•..•.•••....•..... SmA
NOTE: Abso)ute ratings apply to both DICE and packaged parts, unless
otherwise noted

ELECTRICAL CHARACTERISTICS at V+ = +SV, V-= -12V, IREF= 2mA, TA = -SsoC to +12SoC for DAC-888A/B,
otherwise noted. TA = 25° C to +8So C apply for DAC-888E/F. Output characteristics refer to both lOUT and tc:mT.
DAC-888A1E
PARAMETER

SYMBOL

CONDITIONS

MIN

TVP

unless

MIN

TYP

MAX

UNITS

Resolution

8

Bits

Monotonicity

8

Bits
±O.,

Nonlinearity
Full-Scale Tempco
Output Voltage
Compliance

Output Impedance

TCI FS

(See note)

Voe

Full-Scale Current
Change < , /2 LSB

-5

Full Range Current

IFR

Full Range
Symmetry

I FRS

IFA14 -I FR13

Zero-Scale Current

Izs

Output Current
Range

I FSR

Reference- Bias
Current

18

Power Supply

PSSI FR +
PSSI FR-

±50
+5

±'O
-5

>20

ROUT
VREF = 5.00V
Rl1 , R,o = 2.500kO
TA=25°C

Sensitivity

±'O

'.94

2,

I REF =3mA

V+ = 4.5V to 5.5V
V-= -'O.BV to -,3 2V
IAEF= 'mA

±0.'9

%FS

±80

ppm/oC

+5

V

>20

,.99

2.04

±,

±8

,94

2.04

mA

±,

±8

~A

0.2

2.9

2.9

2.'

MO

'.99

0.2

2

~A

mA

-,

-3

-,

-3

~A

±0.0003
±0.0002

±O.O,
±O.O,

±0.OOO3
±0.OO02

±O.O,
±O.O,

%ol.IFsI%ol.V+
%ol.IFsI%ol.V-

Power Supply
Current

1+
1-

IREF= 2mA

,2
6

'6
9

,2
6

,6
9

mA

Power Dissipation

Pd

IREF= 2mA

,34

,90

,34

,90

mW

Logie Input Levels
Logic Input "0"
Logie Input "'"

V,L
V,H

Logic Input Current
Logic Input "0"
Logic Input "'"

I,L
I'H
NOTE: Guaranteed by design.

0.8

O.B
2

V,N = 0.8V
V,N =50V

-2
0.'

11-105

-,0

-2
0.'

,

-,0

~
>-4

~

DAC-8888/F

MAX

rJ)

V

~A

1/86, Rev. A

~

0

U
C)

0

~

0
~

~......

C)

......

0

II

~

DAC-aaa BYTEDAC@ a-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING D/A CONVERTER

ELECTRICAL CHARACTERISTICS -

A.C. PARAMETERS Vs = +5V, -12V, IREF= 2mA, TA = 25°C.

DAC-888A1E
PARAMETER

MIN

TYP

MAX

300

400

DAC-888B/F

SYMBOL

CONDITIONS

MIN

Settling Time

ts

From CE & WR Negative Level
to ±1/2LSB, All Bits Switched
ON or OFF, (See note)

Reference Input
Slew Rate

dlldt

(See Note)

4

4

Data Input Setup Time

tDs

(See note)

150

150

ns

Data Input Hold Time

tDH

(See note)

10

10

ns

Chip Enable/Write Pulse Width

tENW

(See note)

250

250

ns

TYP

MAX

300

400

8

UNITS
ns

mA/p.s

NOTE: Guaranteed by design.

DAC-888 PIN DESCRIPTION
SYMBOL

DESCRIPTION

DBo - DB7

DATA BIT - Bits 0-7 are digital. active-high Inputs DB7 IS assigned as the MSB.

PINS 1-8

TI

CHIP ENABLE - An active low input control which IS the deVice enable Input terminal.

WR

WRITE CONTROL - An active low control which enables the microprocessor to write data to the DAC.

PIN 17

lOUT. lOUT

CURRENT OUTPUT - Complementary current outputs, which when added. equal IFs

PINS 13-14

VAEF +. VREF-

VOLTAGE REFERENCE -

PINS 10-11

COMP

COMPENSATION - The reference amplifier frequency compensating terminal

PIN 16

Differential inputs that accept a negative, positive, or bipolar Input and are used to set IFs.

PIN 12

FUNCTIONAL DIAGRAM AND TIMING DIAGRAM FOR 8-BIT OPERATION

.

TO 8-BIT DATA BUS

DATA
INPUT

'~
-----------~
VALID
~tDH

I

WR.CE---------~

O-:~~~--_f.;;;;;:_--=-_;;;;;:;\---~c.:IOUT

lOUT ----------n"7.n-.
'DUT--------~-~~~
seE

NOTE~

NOTE: If Input data changes after WR • TI low, lOUT/lOUT will change.
The last data Input before WR +CE high will be latched. It is suggested. but
not mandatory, that data be valid from WR • CE low to WR + CE high.

OPERATION TABLE

CE
I

o
o

WR OUTPUT
X

NO CHANGE

I

NO CHANGE

0

UPDATE LATCHES (TRANSPARENT)

11-106

1/86, Rev. A

~ DAC-888 BYTEDAC@ 8-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING D/A CONVERTER

DICE CHARACTERISTICS
1. DB7 (MSB)
2.DB6
3.DB5
4.DB4
5.DB3
6.DB2
7.DB1
8. DBO (LSB)
9. GROUND

10.
11.
12.
13.
14.
15.
16.

VREF (+)
VREF (-)
COMP
lOUT
lOUT

V-

WR
17. CE
18. V+

For additional DICE information refer to
1986 Data Book, Section 2.

DIE SIZE 0.139 x 0.126 Inch; 17, 514 sq. mils
(3.53 x 3.2mm.; 11.296 sq. mm.)

WAFER TEST LIMITS

at

Vs = +5, -12V,

IREF=

2mA,

TA =

25°C,

unless otherwise noted. Output characteristics refer to both

lOUT and lOUT.

PARAMETER

SYMBOL

DAC-888N

DAC-888G

LIMIT

LIMIT

UNITS

8

Bits MIN

±01

±019

%FSMAX

CONDITIONS

Resolution

Monotomclty

Bits MIN

Nonlinearity

Output Voltage
Compliance

Voe

Full-Scale Current
Change < 1/2 LSB
ROUT> 20MO Typ.

+5
-5

+5
-5

VMAX
VMIN

Full Range Current

IFA14

VREF = 5.00V
Rl1 , R10 = 2.500kO
TA = 25°C

204
194

204
1.94

mAMAX
mAMIN

IFR14-IFR13

±8

±8

I'AMAX

2

I'AMAX

Full Range Symmetry

I FRS

Zero-Scale Current

I zs

Output Current Range

I FSR

Reference 818S Current

18

Power Supply

PSSI FR +
PSSI FR -

V+ = 4.5V to 5 5V
V-=-4 5V to -12V, IREF= 1mA

Power Supply
Current

1+
1-

Power Dissipation

Pd

Logic Input Levels
Logic Input "0"
LogiC Input "1"

V,L
V,H

Sensitivity

LogiC Input Current

21

2.1

mAMIN

-3

-3

I'AMAX

±0.01
±001

±0.01
±0.01

%MFS/%Ll.V+ MAX
%MFS/%Ll.V- MAX

IREF = 2mA

16
9

16
9

mAMAX

IREF= 2mA

190

190

mWMAX

08

0.8

V MAX
VMIN

-10

-10
1

IREF= 3mA

V,N=O 8V
V,N =50V

I,L
I'H

I'AMAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to vanatlons In assembly methods and normal yield loss, Yield after packaging is not
guaranteed for standard product dice Consult factory to negotiate speCificatIOns based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS V+ = +5V, -12V,

IREF =

2mA,

TA =

25° C,

unless otherwise noted.

DAC-888
PARAMETER

SYMBOL

Reference Input Slew Rate

dl/dt

CONDITIONS

From CE Negative Edge to ± 1/2 LSB,
All bits Switched ON or OFF

TYPICAL

UNITS

8.0

mAil's

Settling Time

ts

300

ns

Data Input Setup Time

t DS

100

ns

Data Input Hold Time

tDH

0

ns

Chip Enable
Write Pulse Width

tENW

200

ns

11-107

1/86, Rev. A

I

~

DAC-8aa BYTEDAC* a-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING D/A CONVERTER

DIGITAL INFORMATION
The BYTEDAC® DAC-888 is a monolithic microprocessor
compatible DI A converter consisting of an 8-bit level
triggered latch, control circuitry and one 8-bit multiplying
D/A converter housed in an 18-pin dual in line package (DIP).
The DAC-888 accepts 8-bit binary bytes at the data inputs.
Data access is accomplished when WR and CE are low.
During the low state of CE and WR, the latches are transparent, therefore, data should be valid from 100ns prior to WR
and"CE low until"CE or WR high. When"CE or WR goes high,
the data stored in the latches will hold the selected output
indefinitely.
ANALOG INFORMATION
BASIC POSITIVE REFERENCE OPERATION
MSB
LSB
DB7 DBe DBS 084 DB3 DBZ DBl 080

17

Rla 10

VREF+

lOUT

lOUT

IREF_

BASIC NEGATIVE REFERENCE OPERATION

,,-

RREF
REF

In positive reference applications, an external positive reference voltage current flows through R'0 into the VREF(+) terminal of the reference amplifier. Alternatively, a negative
reference may be applied to VREF(-); reference current flows
from ground through R,o into VREF(+) as in the positive reference case. This negative reference connection has the
advantage of a very high impedance presented at pin 11. The
voltage at pin 10 is equal to and tracks the voltage at pin 11
due to the high gain of the internal reference amplifier. Rll
(nominally equal to R,o) is used to cancel bias current errors;
R" may be eliminated with only a minor increase in error.
For most applications the tight relationship between IREFand
IFR will eliminate the need for trimming IREF. If required,
full-scale trimming may be accomplished by adjusting the
value of R,oor by using a potentiometer for R,o. An improved
method of full-scale trimming which eliminates potentio-

,.
+6.000V

or may vary from nearly 0 to +4.0mA. The full range output
currrent is a linear function of the reference current and is
given by:

MSB
LSB
DB7 DBS DB5 DB4 DBa Daz DB1 DBa

OAe

~'~'~--~~~~~~~[-~V_~~G~N~D~'~~~~T~-----r'~3 lOUT
R11

12

15

•.00.F

.,'~

-=

18

.'Yi

FOR FixeD REFERENCE, TTL OPERATION TYPICAL
VALUES ARE'

'FR "" +:RRee: x ~

RREF = 2.5000k
R1' = RREF

FOR ALL LOGIC STATES

14

lOUT + loUr ... IFR

REFERENCE AMPLIFIER SET-UP

lOUT

IFR=~)(.2.§§.

RREF
256
NOTE, RREF seTS IFS. R11 IS FOR BIAS CURRENT CANCELLATION
AND IS EQUAL TO RREF

The DAC-888 is a multiplying D/A converter in which the
output current is the product of a digital number and the
input reference current. The reference current may be fixed
BASIC UNIPOLAR NEGATIVE OPERATION
MSB
LSB
DB7 DBS OBS DB4 DBa Daz DB1 DBa

DB7 DBB DB5 DB4 DB3 DB2 DB, DBO lomA lemA

Eour
'REF =
2.000mA
11

HALF-SCALE + 1 LSB
HALF-SCALE
HALF-SCALE -1 LSB

Eo

'.008 0.984 -2520 -2.460
1000 0.992 -2.500 -2460
0.992 1.000 -2460 -2 SOO

1

ZERO-SCALE +1 LSB 0
ZERO-SCALE

11-108

Eo

1992 0.000 -4.980 0.000
,984 0008 -4.960 -0.020

FULL·SCALE
FULL-SCALE -, LSB

1
0

0008 , 984 -0.020 -4960
0000 1992 -0.000 -4.980

1/86, Rev. A

-m

DAC-881 BYTEDACII I-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING DlA CONVERTER

RECOMMENDED FULL-SCALE ADJUSTMENT CIRCUIT

TABLE 1. REFERENCE AMPLIFIER COMPENSATION
REF. INPUT RESISTANCE

SUGGESTED Cc

1kn
2.Skn
Skn

1SpF
37pF
7SpF

Msa
LS8
DB7 DBa DBs DB4 DB3 DBZ OS, DBO

NOTE: A 0.Q1 p.F capacitor Is suggested for fixed references.

17

For fastest response to a pulse, low values of R10, enabling
small Cc values, should be used. If pin 10 Is driven by a high
current source, none of the above values will suffice and the
amplifier must be heavily compensated which will decrease
overall bandwidth and slew rate. For R10 = 1kn and Cc =
1SpF, the reference amplifier slews at 4mA/"s, enabling a
transition from IREF = 0 to IREF = 2mA in SOOns (see Figure,
pulsed reference operation).

18
RIO
LOW T,C.

IREF+ - z.nm:!A_~4.5~lcl!!!0~_....f,;;;;,:;--.....,I;=---;;;~.-_~~14!....o
+6.000v 0
"EF

Bipolar references may be accommodated by offsetting VREF
or pin 11, as shown in Figure below. The negative commonmode range of the reference amplifier is given by VCM = Vplus (lREF X 1kn) plus 2.SV. The positive common-mode
range is V+ less 1.SV.
meter TC effects is shown in the Recommended Full Scale
Adjustment Circuit.
Using lower values of reference current reduces negative
powersupply current and increases reference amplifier negative common mode range. The recommended range for
operation with a DC reference current is +0.2mA to +4.0mA.
The reference amplifier must be compensated by using a
capacitor from pin 12 to V-. For fixed reference operation a
O.01"F capacitor is recommended. For variable reference
applications, see "Reference Amplifier Compensation for
Multiplying Applications" section.

REFERENCE AMPLIFIER COMPENSATION FOR
MULTIPLYING APPLICATIONS
AC reference applications will require the reference amplifier to be compensated using a capacitor from pin 12 to V-.
The value of this capacitor depends on the impedance presented to pin 10 (see Table 1).

When a DC reference is used, a reference bypass capacitor is
recommended. A S.OV TTL Logic supply is not recommended as a reference. If a regulated power supply is used as
a reference, R10 should be split into two resistors with the
junction bypassed to ground with a 0.1"F capacitor.

ANALOG OUTPUT CURRENTS
Both true and complemented output sink currents are provided, where 10 + iQ = IFR. Current appears at the "true"
output when a "1" is applied to each logic input. As the binary
count increases, the sink current at pin 14 increases proportionally in the fashion of a "positive logic" D/A converter.
When a "0" is applied to any input bit, that current is turned
off at pin 14 and turned on at pin 13. A decreasing logic count
increases iO as in a negative or inverted logic D/A converter.
Both outputs may be used simultaneously. If one of the
outputs is not required it must still be connected to ground or
to a pOint capable of sourcing IFS; do not leave an unused
output pin open.

ACCOMMODATING BIPOLAR REFERENCES
MSa

LSB

Msa

DB7 DBa DII6 DB4 DB3 DB. DBI DOo

LSB

DB7 DBe DBS D84 DB3 DB2 DBl DBa

DAC

VIN

-

HIGH INPUT
IMPEDANCE
RREF - R11
+VREF MUST BE ABOVE PEAK POSITIVE SWING OF VIN

IREF ;;.. NEGATIVE SWING OF liN

11-109

---- ----------

LV- ,,;;;~R~'~'
~'~'I_~~O~R~~=::,-4V:;;;_......!G::N!2D_rn;n:~~+::IUT~_-+'::..3()
(OPTIONAL)

1/86, Rev. A

I

~

DAC-888 BYTEDAC~ 8-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING DlA CONVERTER

8080 INTERFACE

B1

B2 _ _ _ _ _"
ADDRE~

8228

><::::

___________________________________

8080
.p

><::::

DATA ___________________________________
~"'~

v- v+
ADDRESS
8US

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___J~

::~: ::::::::::,~,:::::::::::::::::::::::::::

ADDRESS
DECODER

6800,6801,6809 INTERFACE

B2
ADDRESS

RJW
"XX
.p

VMA

RIW

DATA
ADDRESS
8US

e!!

B2

lOUT

VMA

lOUT

/

'/"

'"

~

'--../
X

8085 INTERFACE

elK

"'~____________....J/
DATA _ _ _ _ _ _ _ _...JX~

808•
•p

_______________>C

"DtMEM .../
MAPP~g {

ADDRESS
BUS
I/O/~

ADDRESS

CE

'-

....Jr

~,,____________________________________

'DUT ______________________~ ~-----------------'OUT
)c~___________________

DECODER

11-110

1/86, Rev. A

~

DAC·881 BYTEDACfI> I·BIT HIGH·SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING DlA CONVERTER

PULSED REFERENCE OPERATION
Msa
LSB
DB7 DB6 DBS 084 DB3 DB2 DBl DBo

+VREF

<;>
I

16

OPTIONAL RESISTOR ~
FOR OFFSET INPUTS
~

1"

ov.rL

RIN

I

*

TYPICAL VALUES
RIN =5k

10

Rp

+VIN·1OV

NO CAP
·UNBALANCED INPUTS WILL CAUSE SLIGHT DEGRADATION OF
'FS ACCURACY DUE TO lOS OF REFERENCE AMPLIFIER

POSITIVE LOW IMPEDANCE OUTPUT OPERATION
MS8

LSB

DB7 CBs CBs D84 DB3 DB2 DB1 DBO

17

,.
10

RL

VREF+

lOUT

>--'--<> Eour

OAe

11

VREFCOMP

v-

lOUT
GNO

v+

o TO

+IFS x RL

II

'FR":: 'REF

FOR COMPLEMENTARY OUTPUT (OPERATION AS NEGATIVE LOGIC DAel.

CONNECT INVERTING INPUT OF OP AMP TO lOUT CONNECT lOUT TO
GROUND

NEGATIVE LOW IMPEDANCE OUTPUT OPERATION
Msa
LSB
DB7 CBs DB5 084 DB3 CB2 DB1 DBo

17
16

>-""--0 EOUT

10

o TO
RL

11

-IFS

x RL

IFR"~IREF

FOR COMPLEMENTARY OUTPUT (OPERATION AS NEGATIVE
LOGIC OAC,) CONNECT NQNINVERTING INPUT OF OP AMP
TO lOUT CONNECT lOUT TO GROUND

11·111

1/86, Rev. A

-WD

DAC-888 BYTEDAce 8-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING DlA CONVERTER

BASIC BIPOLAR OUTPUT OPERATION
Mse

LSB

DB7 DBs DB& 084 DB3 082 OB1 DBo

+6,ODOV

17

,.
6.DOOIen

IREF+=
2.00DmA

5.000kn

o-..2.!!10t-_-1rv;;;;;:---=-;;;;;~__-I~14~E~O:!!:UT!..J
13

11

DB7
POSITIVE FULL-SCALE
POSITIVE FULL-SCALE -1 Lse
ZERO-SCALE+1Lse
ZERO-SCALE
ZERO-SCALE -1 Lse

1
1

0

NEGATIVE FULL-SCALE +1 Lse
NEGATIVE FULL-SCALE

0
0

EOUT

DB8

DB5

DB4

DB3

1
1

1
1

1
1

1
1

0
0

0
0

0
0

0
0

0
0

0
0

0
0

0
0

0
0

0
0

0
0

0
0

DB2

DB1

DBa

Eo

Eo

1

-4.960
-4.920

5.000
4.960

-0.040
0.000
0.040

0.080
0.040
0.000

4.900
5.000

-4.920
-4.980

0
1

0

1
1

0

OFFSET BINARY OPERATION
MSS
LSB
DB7 DBS DBS DB4 DBa DB2 DBl DBo

+600av

+16V

+5V

+5V~6L-~~~_ _-r-~:--~~~~\-

VOUT

__~~-l_-t

>--4--() EOUT

REF-02

25k1l

12
15
o.01pF

-12V

DB7

-16V

18

+5V

DBI

DBS

DB4

DB3

DB2

DB1

POSITIVE FULL-SCALE
POSITIVE FULL-SCALE -1 LSe
ZERO-SCALE
NEGATIVE ZERO-SCALE +1LSe
NEGATIVE FULL-SCALE

0
0

DBD

Eo

1
0

4.980
4.920

0

0

0

0

0

0

0

0.000

0
0

0
0

0
0

0
0

0
0

0
0

1
0

-4.960
-5.000

11-112

1/86, Rev. A

~

DAC-8aa BYTEDAce a-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING DlA CONVERTER

BASIC BIPOLAR OUTPUT OPERATION
Both outputs have an extremely wide voltage compliance,
enabling fast direct current-to-voltage conversion through a
resistor tied to ground or other voltage source. Positive
compliance is 18V above V- and is independent of the
positive supply. Negative compliance is given by V- plus
(IREF X 1kO) plus 2.5V.

supplies is possible. However, at least 8V must be applied to
insure turn-on of the internal bias network.
Symmetrical supplies are not required, as the DAC-888 is
quite insensitive to variations in supply voltage.
Power consumption may be calculated as follows:
P d = (1+) (V+) + (1-) (V-).

POWER SUPPLIES

TEMPERATURE PERFORMANCE

The DAC-888 operates over a wide range of power supply
voltages from a total supply of 9V to 15V. When operating at
supplies of ±5V or less, IREF::; 1mA is recommended. Low
reference current operation decreases power consumption
and increases negative compliance, reference amplifier negative common-mode range, negative logic input range, and
negative logic threshold range; consult the various figures
for guidance. For example, operation at -4.5V with IREF =
2mA is not recommended because negative output compliance would be reduced to near zero. Operation from lower

The nonlinearity and monotonicity specifications of the
DAC-888 are guaranteed to apply over the entire rated
operating temperature range. Full-scale output current drift
is tight, typically ± 10ppm/o C, with zero scale output current
and drift essentially negligible compared to 1/2 LSB.
The temperature coefficient of the reference resistor R10
should match and track that of the output resistor for
minimum overall full-scale drift. Settling times of the DAC888 decrease approximately 10% at -55 0 C; at +1250 C an
increase of about 15% is typical.

Z-80 INTERFACE

IREF

;,;;m:;~

zso

••

/
/

\

WI;

iVii

X

DATA==::><

CE---.....

ADDRESS
BUS

,I

lOUT

;,;;m:;

I

X

lOUT

6502 INTERFACE

02

ADDRESS

IREF

RiW

6602

••

RM

DATA

CE

ADDRESS
BUS

02

lOUT
ADDRESS

rouT

DECODER

11-113

/

\

x::::::::x
'--./

><

1/86, Rev. A

~

DAC-88a BYTEDACII> a-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING DlA CONVERTER

8048 INTERFACE

DATA BUS

.p

ALE~

BUS

FLOATING

,~-----------------X

ADDRESS

X

FLOATING

X

DATA

>(E:NG

'OUT
'OUT

\

CE

\

\VA

''OUT--------------~x:x~O~O~------------------OUT _______ I I
NOTE

OAe OUTPUT MAY CHANGE

UNTIL DATA VALID

'SOFTWARE SAR' AID CONVERTER (WITH 8048 MICROPROCESSOR)
07 06 05 04 03 02 01 DO

ViR

CEW

+15V

WRITE
STROBE

+15V

1

2

3

4

5

6

7

8 16

>=--+--OVOUT

DAC-888

REF-02

001pF
-16V

'9k"

-12V

+5V

10k"

-L.

ANALOG
-:.- GROUND

DIGITAL
VI GROUND
+15V

11-114

1/86, Rev. A

--ffHD

DAC-8aa BYTEDACI!> a-BIT HIGH-SPEED "MICROPROCESSOR COMPATIBLE" MULTIPLYING D/A CONVERTER

SUCCESSIVE APPROXIMATION ANALOG-TO-DIGITAL CONVERSION PROGRAM
LISTING USING DAC-888 AND SYM 1 PCB WITH 6502~P WITH FLOW CHART

LOCATION

DATA

MNEMONIC

COMMENTS

500
502
504
506
508

A900
A208
9500
A980
A8

LOA #00
LOX #08
STA ,X
LOA #80
TAY

Clear
Set Index Register
Clear Memory at 08 H
Tnal Bit
TOY

LOAD INDEX WITH 8

CLEAR LOCATION 8 IN RAM

LOAD A WITH 8016

TRANSFER (Al TO Y

509
50C
50F
511
513

800010
AD 00 1C
2901
FO 01
98

STA 1000 (Cont.)
LOA 1COO
AND A, #01
BEQ'+1
TYA

Output
Read Compo
Mask It
Branch if CMP ~ 0
Get Tnal Bit

514
515

18
7500

CLC
ADC,X

517
519
51A

9500
98
4A

STA ,X
TYA
LSR

Clear Carry
Result Summed With
PrevIous Test
Save it
Get Tnal Value
Next Bit

51B
51C
51E
520

A8
1500
90 E9
4C 00 05

TAY
ORA,X
BCC '-23
JMP 500

Save it
Next Data
Continue For 8 Trials
Do Over

OUTPUT (Al TO DAG 888

READ COMPARATOR OUTPUT

MASKITWITH0116

ADD (Al TO TRIAL VALUE (LOC 6)
SAVE IT IN LOC 8

SHIFT TRIAL BIT RIGHT TO CARRY

NOTE: 32 Bytes 2601'S

CARRY
~

1

BURN-IN CIRCUIT

II

,.
17
16
15
+5V

DAC-888

"
13

-12V

o ' .. F, 50V

12

11
10

10kn

11-115

1/86, Rev. A

])JlC:-1508Jl/1408Jl
8-BIT MULTIPLYING
D/A CONVERTERS
Precision MOIlolithics Inc.

FEATURES
•
•
•
•
•
•
•
•

The R-2R ladder divides the reference current into eight
binarily-related components which are fed to the switches. A
remainder current equal to the least significant bit is always
shunted to ground, therefore the maximum output current is
255/256 of the reference amplifier input current. For example,
a full-scale output current of 1.992mA would result from a
reference input current of 2.0mA,

Improved Direct Replacement for MC1S0S/MC140S
0.19% Nonlinearity Maximum Over Temperature Range
Improved Settling Time ................... 2S0ns, Typ
Improved Power Consumption ........... 1S7mW, Typ
Compatible with TTL, CMOS Logic
Standard Supply Voltages +S.OV and -S.OV to -1SV
Output Voltage Swing ................. +O.SV to -S.OV
High-Speed Multiplying Input ............... 4.0mA/l's

The DAC-150SA/140SA is useful in a wide variety of applications, including waveform synthesizers, digitally programmable gain and attenuation blocks, CRT character generation, audio digitizing and decoding, stepping motor drives,
programmable power supplies and in building tracking and
successive approximation analog-to-digital converters.

ORDERING INFORMATIONt
IS· PIN DUAL·IN-LINE PACKAGE
RELATIVE
ACCURACY
'10 FS

±O,19'1o
±O,39'1o
±O,78'1o

HERMETIC
MILITARY
DAC1508A-80'

For significantly improved speed and applications flexibility
your attention is directed to the DAC-OS S-bit high-speed
multiplying D/A converter data sheet. For D/A converters,
which include precision voltage references on the chip,
please refer to the DAC-210 or the DAC-100 data sheet.

PLASTIC
COMMERCIAL
COMMERCIAL
DAC1408A-80
DAC1408A·70
DAC1408A-60

DAC1408A-8P
DAC1408A-7P
DAC1408A-6P

• For devices processed in total compliance to MIL-STD-883, add 1883 alter
part number Consult factory for 883 data sheet.
tAil commerCial and industrial temperature range parts are available with
burn-in For ordering Information see 1986 Data Book, Section 2.

PIN CONNECTIONS

GENERAL DESCRIPTION
The DAC-150SA/140SA are S-bit monolithic multiplying
digital-to-analog converters consisting of a reference current amplifier, R-2R ladder, and eight high-speed current
switches. For many applications, only a reference resistor
and reference voltage need be added. Improvements in
design and processing techniques provide faster settling
times combined with lower power consumption while retaining direct interchangeability with MC150S/140S devices.

16-PIN DUAL-IN-LINE
PACKAGE
HERMETIC (0)
EPOXY (P)

SIMPLIFIED SCHEMATIC
Msa
A1

A2

A3

A4

AS

A6

A7

lSB
AS

GND

VRE ,(+1

,.

0-+-_------,

VAEF(-lo-'5+---I
COMPEN

11-116

1/86, Rev. A

_ _ _ _ _ _ _ _ _---IIfMDDAC-150lAl140lA I-BIT MULTIPLYING DlA CONVERTERS

ABSOLUTE MAXIMUM RATINGS
Power Supply Voltage
VCC ••••••••••••••••••••••••••••••••••••••••• +5.5Vdc
VEE •••••••••••••••••••••••••••••••••••••••• -16.5Vdc
Digital Input Voltage, V5 through V I 2 •••••••••• +5.5,OVdc
Applied Output Voltage ••••••••••••••••••• +0.5, -5.2Vdc
Reference Current, 114 ••••••••••••••••••••••••••••• 5mA
Power Dissipation (Package limitation), Pd
Ceramic Package (or Epoxy B Package) ••••••• l00mW
Derate above TA = +25°C •••••••••••••••••• 6.7mW/oC

ELECTRICAL CHARACTERISTICS

=

Derate above T A = + 100° C for
Epoxy B Package .......................... 5.3mW/oC
Operating Temperature Range, TA
DAC-150SA •••••••••••••••••••••••• -55°C to +125°C
DAC-140SA •••••••••••••••••••••••••••• O°C to +75°C
DICE Junction Temperature (TJ) ••••••••• -65° C to 150° C
Storage Temperature Range, T stg •••••• -65°C to + 150° C
Plastic Package Only ••••••••••••••• -65° C to + 125° C
NOTE: Ratings apply to both DICE and packaged parts. unless otherwlsa
noted.

=

=

at VCC
+5Vdc, VEE
-15Vdc, VREFlR14
2mA, -55°C S TA S +125°C for
DAC-150SA-S, 0° C S TA S + 75° C for DAC-140SA, unless otherwise noted. All digital inputs at logic high level.

PARAMETER
Relative Accuracy (error relative to
Full·Scale 10 )
DAC·1508A·B, DAC·I408A-8
DAC·140BA·7
DAC·I40BA·6

SYMBOL

DAC-1508A11408A
MIN
TVP
MAX

CONDITIONS

UNITS

~

~

~

0

±0.19
±0.39
±0.7B

Er

Settling Time to within 112 LSB
(includes tPLH)

ts

TA = +25°C, (Note I)
TA = +25°C, (Note I)

Propagation Delay Time

tpLHI tPHL

Output Full·Scale Current Drill

TCl o

Digital Input Logic Levels (MSB)
High Level, Lagle "I"
Low Level, Logic "I"

V,H
V,L

Digital Input Current (MSB)

I'H
I'L

Relerence Input Bias Current (Pin 15)

1'5

250
30

U
%IFS

100

ns
ppm/oC

±20

0

0

ns

~

0

~

0.8
High Level, V,H = 5.0V
Low Level, V,L = 0.8V

0
-0.4

Vdc

0.04
-0.8

mA

-I

-3

pA

0
0

2.0
2.0

2.1
4.2

mA

1.9

1.99

2.1

mA

0

4

pA

+0.5
+0.5

Vdc

Output Current Range

lOR

VEE =-5V
VEE =-15V

Output Current

10

VREF = 2.000V, RI4 = 10000

Output Current

101m,n)

All bits low

Output Voltage Compliance
(E,S 0.19% at TA = +25°C)

Vo

IREF = ImA
VEE =-5V
VEE =-10V

Reference Current Slew Rate

SRI REF

4

Output Current Power Supply
Sensitivity

PSSl o_

0.5

2.7

"AN

Power Supply Current

Icc
lEE

All bits low

+9
-7.5

+14
-13

mA

Power Supply Voltage

VCCR
VEER

TA=+25°C

+5
-15

+5.5
-16.5

Vdc

Pd

All bits low
VEE =-5Vdc
VEE =-15VdC
All bits high
VEE =-5Vdc
VEE =-15Vdc

82
157

135
265

Power Di88ipation

V)

-0.6
-5

+4.5
-4.5

mAIl's

mW

70
132

NOTE:
I. Guaranteed by design.

11·117

1/86, Rev. A

~

0
......

0

II

------------11EM!> DAC-1508A11408A 8-BIT MULTIPLYING DlA CONVERTERS
DICE CHARACTERISTICS
1. N.C.
2. GROUND
3. VEE

4.10
5. A1 (MSB)

6. A2

7. A3
8. A4

9. AS
10. A6
11. A7
12. A8(LSB)
13. Vce
14. VREF(+)
15. VREF(-)
16. COMP

For additional DICE information refer to
1986 Data Book, Seclion 2.
DIE SIZE 0.085 X 0.082 Inch, 5270 sq. mils
(2.16 X 1.58 mm, 3.39 sq. mm)

WAFER TEST LIMITS at V+ = 5V, V- = 15V, IREF= 2mA,

TA =

25°C, unless otherwise noted.
DAC-1408A-G

PARAMETER

SYMBOL

CONDITIONS

Resolution
Monotoniclty

LIMIT

UNITS

8

Bits MIN

8

Bits MIN

±O 19

%FSMAX

+05
-06
-5

VMAX
VMIN
VMIN

2,±0.1

mAMAX

4

IJAMAX

2.1
4.2

mAMAX

Logic "0" Input Level

0.8

V MAX

Logic "I" Input Level

2

VMIN

Nonlinearity
Full-Scale Current Change, 'REF = lmA
<1/2 LSB V-=-5V
V-=-10V

Output Voltage Compliance
Full-Scale Current
Zero-Scale Current
Output Current Range

IzS

(All Bits Low)

lOR

V-=-5V
V-=-15V

Logic Input Current
Logic "0"
Logic "1"

Low Level, Vll = -0.8V
High Level, VIH = 5V

Reference Bias Current
Output Current Power Supply
SensitiVIty
Power Supply Current
(All Bits Low)

PSSlo1+
1-

Power Supply Voltage Range
Power DiSSipation
(All Bits Low)

V-=5V
V-=-15V

±10
±10

IJAMAX

-3

IJAMAX

2.7

IJAIVMAX

+14
-13

mAMAX

+5, ±0.5
-16.5,-4.5

V MAXIMIN

135
265

mWMAX

NOTE:
Electrical tests are performed at wafer probe to the lImIts shown. Due to varlallons In assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dIce Consult factory to negotIate specIfications based on dice lot qualificallon through sample lot assembly and testing

TYPICAL ELECTRICAL CHARACTERISTICS at V+ = +5V, V-= -15V, TA = 25° C, VLcand lOUT connected to ground, and
I REF = 2mA, unless otherwise noted. Output characteristics refer to lOUT only.
DAC-1408G
PARAMETER

SYMBOL

Reference Input Slew Rate

dlldt

CONDITIONS

Propagation Delay

Any BIt

Settling Time

To ± 1/2 LSB, All Bits Switched
ONorOFF

ts

11-118

TYPICAL

UNITS

4

mNIJS

30

ns

250

ns

1/86, Rev. A

------------1~ DAC-150aAl140aA a-BIT MULTIPLYING DlA CONVERTERS

APPLICATIONS
RELATIVE ACCURACY TEST CIRCUIT
MSB
A'
A2
12 BIT
D-TO A

A3

o TO +10V

CONVERTER

A'

(to 01%

OUTPUT

--0--

ERROR MAX)

AS

A6
A7

~
5k!l

LSB A"I'

Al1A12
50kI!

VREF

<;;.

2V

~

I~*
Vee

R"
950n

-

,.

MSB
S

1

1

ERR OR
(1V = 1%)

,r;?

'3

6
7

B

8·BIT

COUNTER

OAC 150SAI
OAC 1408A

,."

~

"

12

1kn

:!:'t
--

USE WITH NEGATIVE VREF

R14

R16

A'

Ol;:

R15

A'

A2

A4

II

-

Vee-

USE WITH POSITIVE V REF

R14~

A3

3 2

A2

15

R15

.....-o--JW-.

A3

VnEFH

"Lr

1-''''4-Q----"\MR_'4_____ VREF(+)

's

...r-L

R'S

A4

AS

AS

AS

AS

A7

A7

AS

AB

RL

SEE
TEXT
FOR

SEE
TEXT
FOR

VALUES

VALUES

OFe

OF e

1/86, Rev. A

11-119

~-------

~--~-

------

-----------IIEMD DAC-150aAl140aA a-BIT MULTIPLYING DlA CONVERTERS
TRANSIENT RESPONSE AND SETTLING
TIME TEST CIRCUIT
Vee

+20Vdc

13

The compensation capacitor value must be increased with
increases in R14 to maintain proper phase margin; for R14
values of 1.0, 2.5 and 5.0kO, minimum capacitor values are 15,
37, and 75pF. The capacitor may be tied to either VEE or
ground, but using VEE increases negative supply rejection.

,.
EIN

I Io.'"F

51

... 26pF

Cc

VEE
EO FOR SETTLING TIME MEASUREMENT
(ALL BITS SWITCHED LOW TO HIGH)

USE WITH CURRENT-TO-VOLTAGE CONVERTING OP AMP
Vee

VREF" 2.OVdc

=~4 =- 6~~~ ~

, Okn

r----,

'3

MSB

r.:':::.4<>---_ _ _-'\IVv_ _~-__ VREF

A'

,.

A2
A3

R'.

AS

A8
A7

Vo

VEE
THEORETICAL Va

VO'

Yffl

(Rol

[¥+¥+'V+~+~HI+~+~J
so

ADJUST VREF R15 OR RO
THAT Va WITH ALL DIGITAL
INPUTS AT HIGH LEVEL IS EQUAL TO 9 961 VOLTS
Va = 1k
2V (5k)
=

A negative reference voltage may be used if R14 is grounded
and the reference voltage is applied to R15 as shown. A high
input impedance is the main advantage of this method.
Compensation involves a capacitor to VEE on Pin 16, using
the values of the previous paragraph. The negative reference
voltage must be at least 4.0V above the VEE supply. Bipolar
input signals may be handled by connecting R14 to a positive
reference voltage equal to the peak positive input level at
Pin 15.
When a DC reference voltage is used, capacitive bypass to
ground is recommended as a reference voltage. If a well
regulated 5.0V supply, which drives logic is to be used as the
reference, R14 should be decoupled by connecting it to
+5.0V through another resistor and bypassing the junction of
the two resistors with 0.1 p.F to ground. For reference voltages
greater than 5.0V, a clamp diode is recommended between
Pin 14 and ground.
If Pin 14 is driven by a high impedance such as a transistor
current source, none of the above compensation methods
apply and the amplifier must be heavily compensated,
decreasing the overall bandwidth.

A4

A8

Pin 14 regardless of the setup method or reference voltage
polarity. Connections for a positive voltage are shown on the
preceding page. The reference voltage source supplies the
full current 114' For bipolar reference signals, as In the
multiplying mode, R15 can be tied to a negative voltage
corresponding to the minimum input level. It is possible to
eliminate R15 with only a small sacrifice in accuracy and
temperature drift.

'2 "4 i ii 32 i4 Us +256
['+'+'+'+'+'+'
'J

1OV[~J= 9961V

OUTPUT VOLTAGE RANGE
The voltage on Pi n 4 is restricted to a range of -0.6V to +0.5V
when VEE = -5V due to the current switching methods
employed in the DAC-150SA-S.

The negative output voltage compliance of the DAC-150SA-S
is extended to -5.0V where the negative supply voltage is
more negative than -10V. Using a full-scale current of
1.992mA and load resistor of 2.5kO between Pin 4 and ground
will yield a voltage output of 256 levels between 0 and
-4.9S0V. The value of the load resistor determines the
switching time due to increased voltage swing. Values of RL
up to 5000 do not significantly affect performance but a
2.5kO load increases "worst case" settling time to 1.2p.s
(when all bits are switched on). Refer to the subsequenttext
section of Settling Time for more details on output loading.

GENERAL INFORMATION AND APPLICATION NOTES
OUTPUT CURRENT RANGE
The output current maximum rating of 4.2mA may be used
only for negative supply voltages more negative than -7.0V,
due to the increased voltage drop across the resistors in the
reference current amplifier.

REFERENCE AMPLIFIER DRIVE AND COMPENSATION
The reference amplifier provides a voltage at Pin 14 for
converti ng the reference voltage to a cu rrent, and a turnaround circuit or current mirror for feeding the ladder. The
reference amplifier input current, 114, must always flow into

11-120

1/86, Rev. A

------------t~ DAC-150aAl140aA a·BIT MULTIPLYING D/A CONVERTERS

ACCURACY
Absolute accuracy is the measure of each output current
level with respect to its intended value, and is dependent
upon relative accuracy and full-scale current drift. Relative
accuracy is the measure of each output current level as a
fraction of the full-scale current. The relative accuracy of the
DAC-150SA-S is essentially constant with temperature due to
the excellent temperature tracking of the monolithic resistor
ladder. The reference current may drift with temperature,
causing a change in the absolute accuracy of output current.
However, the DAC-150SA-S has a very low full-scale current
drift with temperature.

The DAC-150SA-S/DAC-140SA series is guaranteed accurate
to within ± 1/2 LSB at a full-scale output current of 1.992mA.
This corresponds to a reference amplifier output current
drive to the ladder network of 2.0mA, with the loss of one LSB
(S.OJLA), which is the ladder remainder shunted to ground.
The input current to Pin 14 has a guaranteed value of between
1.9 and 2.1 mA, allowing some mismatch in the NPN current
source pair. Testing relative accuracy is accomplished by the
circuit labelled "Relative Accuracy Test Circuit". The 12-bit
converter is calibrated for a full-scale output current of
1.992mA. This is an optional step since the DAC-150SA-S
accuracy is essentially the same between 1.5 and 2.5mA.
Then the DAC-150SA-S circuit's full-scale current is trimmed
to the same value with R14 so that a zero value appears at the
error amplifier output. The counter is activated and the error
band may be displayed on an oscilloscope, detected by
comparators, or stored in a peak detector.
Two S-bit D/A converters may not be used to construct a
16-bit accuracy D/A converter. 16-bit accuracy implies a
total error of±1/2 of one part in 65, 536, or±0.00076% which
is much more accurate than the ±0.19% specification provided by the DAC-150SA-S.

MULTIPLYING ACCURACY
The DAC-150SA-S may be used in the multiplying mode with
eight-bit accuracy when the reference current is varied over a
range of 256:1. If the reference current in the multiplying
mode ranges from 16JLA to 4.0mA, the additional error
contributions are less than 1.6JLA. This is well within eight-bit
accuracy when referred to full scale.

A monotonic converter is one which supplies an increase in
current for each increment in the binary word. Typically, the
DAC-150SA-S is monotonic for all values of reference current
above 0.5mA. The recommended range for operation with a
DC reference current is 0.5 to 4.0mA.
SETTLING TIME
The "worst case" switching condition occurs when all bits are
switched "ON", which corresponds to a low-to-high transition for all bits. This time is typically 250ns for settling to
within ± 1/2 LSB, for S-bit accuracy, and 200ns to 1/2 LSB for
7 and 6-bit accuracy. The turn off is typically under 100ns.
These times apply when RL::; 500n and Co::; 25pF.

The slowest single switch is the least significant bit. In
applications where the D/A converter functions in a positivegoing ramp mode, the "worst case" switching condition does
not occur, and a settling time of less than 250ns may be
realized.
Extra care must be taken in board layout since this is usually
the dominant factor in satisfactory test results when measuring settling time. Short leads, 100ilF supply bypassing for low
frequencies, and a minimum scope lead length are all
mandatory.

11·121

II

1/86, Rev. A

DAC-8012
CMOS 12-BIT MULTIPLYING
D/A CONVERTER "WITH MEMORY"
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

feature "memory" read-write operation. Data is loaded into the
latches by a single 12-bit wide word, and can be read back on
the same data lines. The DAC-8012 can be directly connected to
12-bit and 16-bit busses.

Data Readback Capability
12-81t Resolution and Linearity
Low Gain Tempco: 2ppm/o C
Fast TTL/CMOS Compatible Data Latches
3-State Outputs TTL/CMOS Compatible
Single +5V to +15V Supply
Small 2D-Pin 0.3" DIP
LowCost
Latch-Up Resistant
Ideal for Battery-Operated Equipment

The read back function makes the DAC-8012 particularly wellsuited for applications in automatic test equipment, industrial
automation, and other multi-channel microprocessor-controlled
systems that require keeping track of the current DAC output
data without using an extra memory location for each channel.
Low power dissipation and single-supply operation from +5Vto
+15V makes the DAC-8012 an excellent choice in low-power
and remote sytems and digital systems with a large number of
analog outputs. Four-quadrant multiplying capability and 12bit linearity allows the DAC-8012 to be used in low-noise, widebandwidth, low-distortion, digitally-controlled preCision attenuator and filter applications.

ORDERING INFORMATIONt
PACKAGE: 2o-PIN'
MAXIMUM
GAIN ERROR
RELATIVE
TA = +2S'C
Vee = +SV
ACCURACY
±1/2 LSB
±1 LSB

±1 LSB
±3 LSB

MILITARY"
TEMR
-SS'C to
+12S'C

INDUSTRIAL COMMERCIAL
TEMR
TEMR

-25'C 10
+8S'C

O'Cto

DAC-8012AR
DAC-8012BR

DAC-8012ER
DAC-8012FR

DAC-8012GP
DAC-8012HP

PIN CONNECTIONS

+70'C

• Package Designation'
Suffix R Hermetic DIP
Suffix P: Plastic DIP
Consult Factory for Dice Informatton
•• For deVices processed In total compliance to MIL-STD-883. add 1883 after

2D-PIN EPOXY DIP
(P-Sufflx)
2D-PIN HERMETIC DIP
(R-Sufflx)

part number Consult factory for 883 data sheet

t All

commercial and industrial temperature range parts are ava"able with
burn-In For ordering information see 1986 Data Book. Section 2.

GENERAL DESCRIPTION
The DAC-8012 is a monolithic 12-bit CMOS Multiplying DAC
with on-board data latches and three-state output buffers that

FUNCTIONAL DIAGRAM

?

J
I

RO'WA

"FB

12-81T
MULTIPLYING
DAC

...

r-

INPUT DATA

J

os

1

LATCHES

,..

'"

I I

~J

THREE-STATE
OUTPUT BUFFERS

~

I

~

.......

OkND

AGND

J

A

'i7

J

OUTl
)

/

D
11-122

DBll ....

DBQ

1/86, Rev. A1

----------I~ DAC-8012 CMOS 12-BIT MULTIPLYING D/A CONVERTER "WITH MEMORY"

ABSOLUTE MAXIMUM RATINGS

Storage Temperature .................... -65°C to +150°C
Lead Temperature (Soldering, 60 sec) .....•....... +300° C

(TA = 25°C, unless otherwise noted.)
VDD to DGND ................................ -0.3\1, +17V
Digital Input Voltage to DGND .................. -0.3V, VDD
AGND to DGND .............................. -0.3V, VDD
VRFB , VREF to DGND .................•............. ±25V
VPIN 1 to DGND ............................... -0.3V, VDD
Power Dissipation (Any Package) to +75° C ........ 450mW
Derates Above +75°C by ...................... 6mWfOC
Operating Temperature Range
Military (AR, BR) Grades ............... -55°C to +125°C
Industrial (ER, FR) Grades .............. -25°C to +85°C
Commercial (GP, HP) Grades .............. O°C to +70°C
Dice Junction Temperature. . . . . . . . . . . . . . . . . . . . . .. +150°C

CAUTION:
Stresses above those listed under"Absolute Maximum Ratings" may cause
permanent damage to the device This Isa stress rating only and functional
operation at or above this specification IS not Implied Exposure to above
maximum ratmg conditions for extended periods may affect deVice
reliability
Do not apply voltages higher than VDO or less than GND potential on any
terminal except VAEF .
The dlQltallnputs are zener protected, however, permanent damage may
occur on unprotected units from high-energy electrostatic fields Keep
units In conductive foam at all times until ready to use Use proper antistatic handling procedures
Remove power before inserting or removing units from their sockets

=+5Vor +15V, V REF =+10V, VOUT 1 =Ov, AGND =DGND =OV; TA =-55°C to +125°C
=-25°C to +85°C apply for DAC-8012ER/FR, TA =O°C to +70°C apply for DAC-8012GP/HP, unless

ELECTRICAL CHARACTERISTICS at VDD
apply for DAC-8012AR/BR, TA
otherwise noted.

DAC-8012A/E/G
PARAMETER

SYMBOL

CONDITIONS

MIN

TYP

DAC-8012B/F/H

MAX

MIN

TYP

MAX

UNITS

STATIC ACCURACY

12

Resolution
Relative
Accuracy

12

Bits

INL

TA = Full Temp Range

±1/2

±1

LSB

DNL

TA = Full Temp Range

±1

±1

LSB

G FSE

TA = +25°C
TA = Full Temp Range

±1
±2

±3
±4

LSB

±5

±5

ppmloC

0002
0004

0002
0004

%1%

10

10

200
25

200
25

300

300

Differential
Nonlinearity

(Note 1)
Gain Error

(Notes 2, 3)
Gain Temperature
CoeffiCient
.6.Galn/.6.Temperature

TCG FS

(Notes 4. 5)
DC Supply Relectlon
aGalnl avoo (Note 4)

Output Leakage
Current at OUT 1

PSR

T = +25°C
T: = Full Temp Range(IlNoo = ±5%)
TA = +25°C, RD/WR = DS = OV,
All digital inputs = OV

ILKG

nA
TA = Full Temp Range
AlB VerSions
ElF IG/H VerSions

DYNAMIC PERFORMANCE
Propagation Delay

(Notes 4, 6, & 7)
Current Settling Time

(Notes 4. 7)
Glitch Energy
(Note 4)

tpo

TA = +25°C
(OUT 1 Load = 1000. CEXT = 13pF)

.

TA = Full Temp Range
(To 1/2 LSB)
10UT1 Load = 1000

Q

TA = Full Temp Range
VREF =AGND

TA

AC Feedthrough
at IOUT1

FT

(Notes 4. 11)

ns

/lS

=+25°C

400
500

400
500

TA = Full Temp Range
VREF = ±10V. f = 10kHz

nVs

mvp_p

REFERENCE INPUT
Input ReSistance

(Pin 19 to GND)
(Note 12)

TA = Full Temp Range
RREF

11

Input ReSistance

11-123

- -

.--~~---

15

11

15

kO

1/86, Rev. A1

III

~ DAC-8012 CMOS 12-BIT MULTIPLYING D/A CONVERTER "WITH MEMORY"

ELECTRICAL CHARACTERISTICS at VDD=+5V, VREF= +10V, VOUTI = OV, AGND = DGND = OV; TA =-55·C to +125·C apply
for DAC-8012AR/BR, TA = -25·C to +85·C apply for DAC-8012ER/FR, TA = O·C to +70·C apply for DAC-8012GP/HP, unless
otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

DAC-8012A/E/G
MIN
TYP
MAX

DAC-8012B/F/H
MIN
TYP
MAX

70
ISO

70
ISO

UNITS

ANALOG OUTPUTS
Oulput CapaCItance
(Note 4)
C OUT1

COUT

Voo =+SV or +1SV
TA =Full Temp. Range
DBO-DBI I =OV, RDIWR =OS =OV
DBO-DBI I =Voo, RD/WR =OS =OV

pF

DIGITAL INPUTS
Input High Voltage
Input Low Voltage

VINH
VINL

TA

=Full Temp. Range

Input Current
(NoteS)

liN

TA
TA

=+2S'C

Input Capacitance
DBO-DBll
RD/WR, OS
(Note 4)

CIN

TA

=Full Temp

Output High Voltage

VOH

10

=400p.A

Output Low Voltage

VOL

10

=-1

2.4

2.4

=Full Temp. Range
Range

O.S

O.S

10

10

12

12

6

6

V

p.A

pF

DIGITAL OUTPUTS
4.0

4.0

6mA

Three-State Output
Leakage Current
SWITCHING CHARACTERISTICS
(Note 9)

See Timing Diagram

Write to Data Strobe
Setup Time

twsu

TA
TA

=+2S'C

Data Strobe to
Write Hold Time

tWH

TA

Read to Data Strobe
Setup Time

t Asu

Data Strobe to
Read Hold Time

tAH

Write Mode Data Strobe
Width

tWAS

Read Mode Data Strobe
Width

tAOS

Data Setup Time

tosu

Data Hold Time

=Full Temp. Range
TA =+2S'C

V

0.4

0.4

V

10

10

p.A

ns

0

0

=Full Temp. Range

0
0

0
0

ns

TA
TA

=+2S'C
=Full Temp. Range

0
0

0
0

ns

TA
TA

=+2S'C

Range

0
0

0
0

ns

TA

=Full Temp. Range

180
2S0

ISO
2S0

ns

220
290

220
290

ns

210
2S0

210
2S0

ns

0
0

0
0

ns

=Full Temp
TA =+2S'C

=+2S'C
=Full Temp. Range
TA =+2S'C
TA
TA

TA

=Full Temp. Range

tOH

TA
TA

=+2S'C

=Full Temp

Range

Data Strobe to
Output Valid Time
(Note 13)

tco

TA
TA

=+2S'C
=Full Temp

Range

300
400

300
400

ns

Output Active Time
from Deselection
(Note 13)

toTO

TA
TA

=+2S'C
Range

21S
37S

21S
37S

ns

100

TA =Full Temp Range
(All Digital Input. VINL or VINH )

2

mA

100

TA =Full Temp Range
(All Digital Inputs OV or Voo )

100

p.A

=Full Temp

POWER SUPPLY

Supply Current

10

11-124

100

10

1/86, Rev. A1

IfMD

DAC-S012 CMOS 12-BIT MULTIPLYING D/A CONVERTER "WITH MEMORY"

ELECTRICAL CHARACTERISTICS at V DD =+15V, V REF =+10V, VOUT1 = OV, AGND =DGND =OV; TA =-55°C to +125°C apply for
DAC-S012AR/BR, TA =-25°C to +S5°C apply for DAC-S012ER/FR, TA =O°C to +70°C apply for DAC-S012GP/HP, unless otherwise
noted.
DAC-8012A/E/G
PARAMETER

SYMBOL

CONDITIONS

MIN

Inpul High Voltage
Input Low Voltage

V,NH
V,NL

TA

Input Current
(Note B)

liN

TA =+25°C
TA =Full Temp Range

Input Capacitance
DBO-DBII
RD/WR, OS
(Note 4)

C 'N

TA =Full Temp. Range

Output High Voltage

VOH

10

Output Low Voltage

VOL

10

TYP

DAC-8012B/F/H

MAX

MIN

TYP

MAX

UNITS

DIGITAL INPUTS
13.5

=Full Temp. Range

13.5

V

1.5

1.5

10

1
10

JJ.A

12
10

12
10

pF

rJ)

~

;;.Li

~

~

DIGITAL OUTPUTS

=3mA
=-3mA

135

13.5

Three-State Output
Leakage Current
SWITCHING CHARACTERISTICS
(Note 9)

See Timing Diagram

Write to Data Strobe
Setup Time

t wsu

TA
TA

Data Strobe to
Write Hold Time

tWH

Read to Data Strobe
Setup Time

tRSU

Data Strobe to
Read Hold Time

tRH

Write Mode Data Strobe
Width

tWAS

TA = +25°C
TA =Full Temp Range

Read Mode Data Strobe
Width

tADS

Data Setup Time

15

V

10

10

JJ.A

a

TA = +25°C
TA =Full Temp. Range

a
a
a
a

TA = +25 Q C
TA =Full Temp Range

a

0

=+25°C
=Full Temp

Range

V

1.5

~
0

f:"'

~....

ns

ns

100
120

100
120

ns

TA = +25°C
TA =Full Temp Range

110
150

110
150

ns

tosu

TA = +25°C
TA =Full Temp. Range

90
120

90
120

ns

Data Hold Time

tOH

TA = +25°C
TA =Full Temp Range

a
a

a
a

ns

Data Strobe to
Output Valid Time
(Note 13)

tco

TA =+25°C
TA =Full Temp Range

lBO
220

180
220

ns

Output Active Time
for Deselectlon
(Note 13)

tOTD

TA =+25°C
TA =Full Temp Range

180
250

180
250

ns

100

TA =Full Temp Range
(All Digital Inputs V,NL or V,NH )

100

TA =Full Temp Range
(All Digital Inputs OV or Voo )

=Full Temp. Range

0

ns

a
a

TA

U

C)

ns

a
a
a

TA = +25°C

0

....C)
0

POWER SUPPLY

Supply Current

mA

10

NOTES:
12~bit monotonic over full temperature range.
Includes the effects of 5ppm max gam T.C.
Using internal RFB DAC register loaded With 1111 1111 1111 Gam error
adjustable uSing the circUIts of Figures 4 and 5
4. GUARANTEED but NOT TESTED
5 Typical value IS 2ppm/o C for Voo = +5V.
6. From digital input change to 90% of final analog output

2.
3.

8
IS

9
10
11
12

11-125

-~.~-~-~--.

100

10

100

JJ.A

All dlgltalmputs =OV to Voo , or Voo to OV
Logic mputs are MOS gates, tYPical Input current (at +25°C) IS less than
InA
Sample tested at +25 0 C to ensure com pliance
Feedthrough can further be reduced by connecting the metal lid on the
sldebraze package (Suffix R) to DGND
Resistor T C =+100ppmfOC max
CL =100pf

1/86, Rev. A 1

--

-

.-

---------l~ DAC-B012 CMOS 12-BIT MULTIPLYING D/A CONVERTER "WITH MEMORY"

DICE CHARACTERISTICS

1. OUT 1
2. AGND
3.DGND
4. DB11 (MSB)
5. DB10
6.DB9
7.DBB
B.DB7
9.DB6
10. DB5

11.
12.
13.
14.
15.
16.
17.
1B.
19.
20.

DB4
DB3
DB2
DB1
DBO (lSB)
OS
RD/WR
Voo
VREF
RFB

For additional DICE information refer 10
19B6 Dala Book. Section 2.
DIE SIZE 0.121 X 0.112 inch. 13.552 sq. mils
(3.07 X 2.B5 mm. B.75 sq. mm)

WAFER TEST LIMITS at VD D = +5V or +15V. VREF = +1aV. VOUT 1 = av. AGND = DGND = av, TA = 25°C. unless otherwise noted.
DAC-8012G
PARAMETER

SYMBOL

CONDITIONS

Relative Accuracy

INL

Endpoint Linearity Error

LIMIT

UNITS

±1/2

LSB MAX

±1

LSB MAX

±3

LSB MAX

DAC Latches Loaded with 0000 0000 0000
Pad 1

±10

nAMAX

RREF

Pad 19

6/15

kil MINI
kilMAX

Output High Voltage

VOH

VDD

40

VMIN

Output Low Voltage

VOL

VDD

04

VMAX

24
135

VMIN

0.8
15

VMAX

±1

/lA MAX

Differential Nonlinearity

DNL

Gain Error

GFSE

DAC Latches Loaded with 1111 1111 1111

Output Leakage

'LKG

Input Resistance

Digital Input

High

V1NH

Digital Input
Low
Input Current
Supply Current
DC Supply Rejection
(~Galnl ~VDD)

V 1NL

'iN

=5V. 10 =400/lA
=5V, 10 =-1 6mA
VDD =5V
VDD =15V
VDD
VDD
Y'N

=5V
=15V
=OV or VDD

All Digital Inputs V1NL or V1NH
'DD

All Digital Inputs OV or VDD

PSRR

VDD

=±5%

01
0004

mAMAX

%1% MAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to vanatlons in assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

TYPICAL ELECTRICAL CHARACTERISTICS at VDD =+5Vor+15V. VREF =+1aV. VOUT1 =av; TA =25°C. unless otherwise noted.
DAC-8012G
PARAMETER

SYMBOL

Digital I nput Capacitance

CIN

CONDITIONS

COUT1

DAC Latches Loaded with 0000 0000 0000

COUT1

DAC Latches Loaded with 1111 1111 1111

TYPICAL

UNITS

12

pF

70

pF

150

pF

300

ns

Output Capacitance

Propagation Delay

VDD = 15V

VDD = 5V

11-126

1/86, Rev. A 1

---------IIEM!>

DAC-B012 CMOS 12-BIT MULTIPLYING D/A CONVERTER "WITH MEMORY"

LOAD CIRCUITS FOR SWITCHING TESTS

LOGIC INFORMATION

-u

D/A CONVERTER SECTION

FOR DATA GOING HIGH
TO DATA BIT

Figure 1 shows a simplified circuit of the D/A Gonvertersection
of the DAG-8012, and Figure 2 gives an approximate equivalent
switch circuit. R is typically 11kO.

~'00PF~3kn

The binary-weighted currents are switched between OUT 1 and
AGND by N-channel switches, thus maintaining a constant
current in each ladder leg independent of the switch state.

FOR DATA GOING LOW

The capacitance at the OUT 1 terminal, GOUT " is code
dependent and varies from 70pF (all switches to AGND) to
150pF (all switches to OUT 1). One of the current switches is
shown in Figure 2.

T3kn

TO DATA BIT

--1.

The input resistance at VREF (Figure 1) is always equal to RLDR
(RLDR is the R/2R ladder characteristics resistance and is equal
to value "R"). Since the input resistance at the VREF pin is
constant, the reference terminal can be driven by a reference
voltage or a reference current, ac or dc, of positive or negative
polarity. (If a current source is used, a low-temperaturecoefficient external RFB is recommended to define scale factor.)

~100PF

PARAMETER DEFINITIONS
RELATIVE ACCURACY

Sometimes referred to as endpoint nonlinearity, and is a
measure of the maximum deviation from a straight line passing
through the endpoints of the DAG transfer function. Relative
Accuracy is measured after the zero and full-scale points have
been adjusted, and is normally expressed in LSB or as a
percentage of full scale.

The internal feedback resistor (R FS ) has a normally closed
switch in series as shown in Figure 1. This switch improves
performance over temperature and power supply rejection;
however, when the circuit is not powered up the switch assumes
an open state.

DIFFERENTIAL NONLINEARITY

This is the difference between the measured change and the
ideal change between any two adjacent codes. A differential
nonlinearity of ±1 LSB maximum over the full operating
temperature range will ensure that a device is monotonic (the
output will not decrease for an increase in digital code applied).

BURN-IN CIRCUIT
"0

C:~,

VREF Voo

C2

~---.... +

GAIN ERROR

Gain or full scale error is the amount of output error between
the ideal output and the actual output. The ideal output is VREF
minus 1 LSB. The gain error is adjustable to zero using external
resistance.

OUT1

R"

AGND

VREF

DGND

Voo

C4

20

19

R1
1kO

18

011 (MSB)

D10

OUTPUT CAPACITANCE

The capacitance from OUT1 to AGND.

Os

D.

!lSB) DO

D8

D1
D2

PROPAGATION DELAY

07

This is measured from the digital input change to the analog
output current reaching 90% of its final value.

06

03

os

D4

10

"

13

R2
5kO

12
11

FEEDTHROUGH GLITCH ENERGY

This is a measure of the amount of charge injected to the analog
output from the digital inputs, when the digital inputs change
states. It is the area of the glitch and is specified in nVsec; it is
measured with VREF = AGND.

11-127

1/86, Rev. A1

II

---------I1fMD

DAC-8012 CMOS 12-BIT MULTIPLYING D/A CONVERTER "WITH MEMORY"

TIMING DIAGRAM

NOTES.

Yoo
Voo

=+5Y; tr =IF =20ns
=+15V, t, ::: If '" 40ns

ALL INPUT SIGNAL RISE AND
FALL TIMES MEASURED FROM
10% TO 90% OF Voo'
TIMING MEASUREMENT REFERENCE LEVEL

IS

FIGURE 1: Simplified DIA Circuit of DAC-8012

VIM 2"'"

V1L •

FIGURE 3: Digital InputlOutput Structure
,-------Ril

DB9

DB11

DBl

DBO
(lSB)

(MSB)

WR

FIGURE 2: N-Channel Current Steering Switch
TO LADDER

FROM~
0----1

INTERFACE
LOGIC

AGND

OUT 1

DIGITAL SECTION
Figure 3 shows the digital 1/0 structure for one bit. When the
data strobe (DS) and the RD/WR lines are held low, data at the
digital input is fed through the input buffers and the data latches
which control the DAC current output switches are transparent.
Data is latched when either DS or RDIWR go high. When the data
strobe DS is held low and the RDIWR line is held high, the
three-state buffer becomes active and the data from the latches is

fed through the three-state buffers to the digital input/output
lines. This is known as the Read Cycle, or data read back.
The input buffers are simple CMOS inverters designed such
that when the DAC-B012 is operated with Voo = +SV, the buffers
convert TTL input levels (2.4V and O.BV) into CMOS logic levels.
When the digital input is in the region of 1.0V to 3.0V. the input
buffers operate in their linear region and draw current from
the power supply. To minimize power supply currents, it is
recommended that the digital input voltages be as close to the
supply rails (Voo and DGNO) as is practically possible. The
DAC-B012 may be operated with any supply voltage in the range
SV SVoo SlSV. With Voo =+lSV, the input logic levels are CMOS
compatible only, i.e., 1.SV and l3.SV.
The three-state output buffers, in the active mode, provide TTLcompatible digital outputs with a fan-out of one TTL load when
the DAC-B012 is operated with +SV power supply. When
powered from +lSV, the output buffers provide output logic
levels of 1.SV and 13.SV. Three-state output leakage is typically
10nA.

11-128

1/86, Rev. Al

---------I1fMD

DAC-8012 CMOS 12-BIT MULTIPLYING D/A CONVERTER "WITH MEMORY"

FIGURE 4: Unipolar Binary Operation

TABLE I: Recommended Trim Resistor Value vs. Grades
TRIM
RESISTOR
R1
R2

HP/FR/BR
1000
330

GP/ER/AR
200
6.80

VOUT

TABLE II: Unipolar Binary Code Table for Circuit of Figure 4
BINARY NUMBER IN
DAC REGISTER
DBll-0BO

ANALOG OUTPUT

"SEE TABLE 1

BASIC APPLICATIONS
Figures 4and 5 show simple unipolar and bipolar circuits using
the DAC-8012. Resistor R1 is used to trim for full scale. The
following versions: DAC-8012AR, DAC-8012ER, DAC-8012GP,
have a guaranteed maximum gain error of ±1 LSB at +25·C and
Vee = +5V, and in many applications the gain trim resistors are
not required. Capacitor C1 provides phase compensation and
helps prevent overshoot and ringing when using high speed op
amps. The circuits of Figures 4 and 5 have constant input
impedance at the VREF terminal.

1111

1111

1111

-V1N '

1000

0000

0000

-V1N '

0000

0000

0001

-V1N '

0000

0000

0000

e
e

095
4096 }
048
4096
} = -1/2 VIN

{40~6 }
o Volts

TABLE III: 2's Complement Code Table for Circuit of Figure 5
ANALOG OUTPUT

DATA INPUT

The circuit of Figure 4 can either be used as a fixed reference
D/A converter so that it provides an analog output voltage in the
range 0 to -VIN (the inversion is introduced by the op amp); or
V1N can be an ac signal in which case the circuit behaves as an
attenuator (2-Quadrant Multiplier). V1N can be any voltage in the
range -20V :5VIN :5+20V (provided the op amp can handle such
voltages) since VREF is permitted to exceed Vee. Table II shows
the code relationship for the circuit of Figure 4.
Figure 5 and Table III illustrate the recommended circuit and
code relationship for bipolar operation. The D/A function itself
uses offset binary code, and inverter U, on the MSB line,
converts 2's-complement input code to offset binary code. The
inverter U, may be omitted if the inversion is done in software,
using an exclusive OR instruction.

01 1 1

1111

1111

+V1N ' {2047 }
2048

0000

0000

0001

+V1N '

0000

0000

0000

1111

1111

1111

-V1N '

1000

0000

0000

-VIN'

{20:8 }

o Volts
{20:8 }

e

048
2048 }

R3, R4 and R5 must match within 0.01 % and should be the same
type of resistors (preferably wire-wound or metal foil), so that
their temperature coefficients match. Mismatch of R3 value to
R4 causes both offset and full scale error. Mismatch of R5 to R4
and R3 causes full scale error.
FIGURE 5: Bipolar Operation (2's Complement Code)

DATA INPUT

11-129

1/86, Rev. A 1

~
~

~

0

U

l'i
0

~
0

~

~

l'i
......

0

II

-----------l~ OAC·BOil! CMuS i2·BiT MULTIPLYING CIA CONVERTER "WITH MEMORY"

APPLICATIONS HINTS

INTERFACING THE DAC-8012 TO MICROPROCESSORS

Output Ollset: CMOS D/A converters exhibit a code-dependent
output resistance that causes a code-dependent error voltage
at the output of the amplifier. The maximum amplitude of this
offset, which adds to the D/A converter nonlinearity, is 0.67 Vas
where Vas is the amplifier input-offset voltage. To maintain
monotonic operation, it is recommended that Vas be no greater
than 10% of 1 LSB over the temperature range of operation.

Figure 6 shows the interface configuration for the 68000 16-bit
microprocessor. No external logic is required to write data into
the DAC orto read back data from the DAC-8012 latches. Analog
circuitry has been removed for clarity.
FIGURE 6: 68000 l6-Bit Microprocessor to DAC-8012 Interface

General Ground Management: AC or transient voltages between AGND and DGND can cause noise injection into the
analog output. The simplest method of ensuring that voltages at
AGND and DGND are equal is to tie AGND and DGND together
at the DAC-8012. It is recommended that two diodes (1 N914 or
equivalent) be connected in inverse parallel between AGND
and DGND pins in complex systems where AGND and DGND
tie on the backplane.

68000

ADDRESS

CPU

BUS~l

I

ADDRESS

l

os

I DECODER I

CTACK

R/W

RD/Vim
~

DB11

DATA BUS

Digital Glitches: When RD/WR and DS are both law, the latches
are transparent and the D/A converter inputs follow the data
inputs. Some bus systems do not always have data valid for the
whole period during which RD/WR is low. This will allow invalid
data to briefly appear at the DAC inputs during the write cycle.
This can cause unwanted glitches at the DAC output. Retiming
the write pulse RD/WR, so that it only occurs when data is valid,
will eliminate the problem.

~

'1

r

DBO

FIGURE 7: 8-Bit Processor to DAC-8012 Interface
"Qo = DECODED ADDRESS FOR LATCH TO DAC OPERATIONS
01 = DECODED ADDRESS FOR DATA BUS TO INPUT LATCH OPERATION
02
DECODED ADDRESS FOR OUTPUT LATCH TO DATA sus OPERATION

=

A15

t-----------------------------'
ADDRESS BUS

Aot----,

00 '0,*

PROCESSOR
SYSTEM

~P---i=~=t=====

8-BIT DATA BUS

11·130

1/86, Rev. A1

DAC-8212

JPMI}

DUAL 12-BIT BUFFERED
MULTIPLYING CMOS D/A CONVERTER
PreCISIon Monolithics lnc.

PRELIMINARY
FEATURES

GENERAL DESCRIPTION

• lWo Matched 12-Blt DACs on One Chip
• Direct Parallel Load of All 12 Bits for High
Data Throughput
• On Chip Latches for Both DACs
• 12-Blt Endpoint Linearity (:t:1/2 LSB)
• + 5V to + 15V Single Supply Operation
• DACs Matched to 1%
• Four-Quadrant Multiplication
• Low Power Consumption

The DAC-8212 contains two 12-bit multiplying digital-to-analog
converters. Excellent DAC-to-DAC matching and tracking results from monolithic construction. The DAC-8212 consists of two
thin-film R-2R resistor-ladder networks, two 12-bit data latches,
one 12-bit input buffer, and control logic. Operation from a +5 or
+ 15 volt single power supply disSipates only 20mW of power.
Digital input data is directed into one of the DAC data latches
determined by the DAC selection control line DAC A/DAC B.
The data load cycle is similar to the write cycle of a random
acce§!..mem2!X, activated by the WR and CS control inputs.
With WR and CS lines logic low, the input latches are transparent,
allowing input digital data to flow directly to the DAC selected by
the DAC A/DAC B select input.

APPLICATIONS
•
•
•
•
•

Automatic Test Equipment
Robotics
Programmable Instrumentation Equipment
Digilal Gain/AHenuation Control
Ideal for BaHery-Operated Equipment

PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE: 24-PIN**
RELATIVE
ACCURACY

E~:~R

±1/2LSB
±1/2LSB

±1 LSB
±6 LSB

MILITARY*
INDUSTRIAL
COMMERCIAL
TEMPERATURE TEMPERATURE TEMPERATURE
-55'C TO + 125'C -25'C TO + 85'C O'CTO +70'C
DAC8212AV
DAC8212BV

DAC8212EV
DAC8212FV

24-PIN
HERMETIC DIP

DAC8212GP
DAC8212HP

• For deVices processed In total compliance to MIL-STD-883, add 1883 after part
number. ConsuH factory for 883 data sheet
"Package deslgnabon SuffiX V Hermebc DIP, SuffiX p. Plastic DIp
t All commeraal and Industrial temperature range parts are available With burn-In.
For ordenng Information see 1986 Data Book, Section 2

FUNCTIONAL DIAGRAM

Voo

DACAl
DACB

co

Wi<

This preliminary product Information Is based on testing of a limited number of devices. Final specifications may vary. Please contact local sales
office or distributor for final data sheet.
11-131
1/86, Rev. A

III

-----t1fMi} DAC-S212 DUAL 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER -

PRELIMINARY

ABSOLUTE MAXIMUM RATINGS
(TA = +25°C, unless otherwise noted.)
VDDtoAGND ....•....•......•............. OV, +17V
V DD to DGND .•...•.•...•..•....•.......... OV, +17V

Dice Junction Temperature •................... + 150°C
Storage Temperature .•...•...•.•...... -65·C to + 150·C

AGND to DGND ....•.......•.•...........•..•.. VDD
DGND to AGND ........••....•.......•......... VDD
Digital Input Voltage to DGND ...•....... -0.3V, VDD +0.3

Lead Temperature (Soldering, 60 sec) ...•..•..... +300·C
CAUTION:
I. Do not apply voltages h'gher than voo or less than GND potential on any
terminal except VREF .
2. The digital control Inputs are zener-protected; however, permanent damage
may occur on unprotected units from high-energy electroslaticfields. Keep units
in conductive foam at all times until ready to use.
3. Do not insert this device Into powered sockets; remove power before insert,on or
removal.
4. Use proper anti-static handling procedures.
5. Stresses above those listed under "Absolute Maximum Ratings' may cause
permanent damage to the device.

VPIN 2, VPIN 24 to AGND .•..•......•.... -0.3V, VDD +0.3
VREF A, VREF B to AGND ...•.•......•........... ± 25V
VRFB A, VRFB B to AGND .••......•••............ ±25V
Power Dissipation (Any Package) to + 75°C ....... 450mW
Derate Above + 75°C by .•................•.• 6mW/oC
Operating Temperature Range
AV, BV Versions .•.............••..• -55·C to + 125·C
EV, FV Versions •......•........•..•• -25·C to +85°C
GP, HP Versions ....•....••.....•..... O·C to + 70·C

at VDD = +5Vor +15V, VREF A = VREF B = +10V, OUT A = OUT B = OV; TA = -55·C
= -25·C to +85°C apply for DAC-8212EV/FV; TA = O·C to + 70·C apply for DAC-8212GP/HP,

ELECTRICAL CHARACTERISTICS
to + 125°C apply for DAC-8212AVlBV; TA
unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

MIN

TVP

MAX

UNITS

±1/2

LSB

±I

LSB

Specifications apply to both DAC A and DAC B

STATIC ACCURACY

12

Resolution

N

Relative Aocuracy

NL

Endpoint Linearity Error

Differential Nonlinearity

DNL

All Grades are Monotonic

Full Scale Gain Error
(Note I)

GFSE

Output Leakage Current
Out A (Pin 2)/Out B (Pin 24)

'LKG

Input Resistance
(VRE~' VREFB)

RREF

VREFAIVRE.e
(Input Resistance Mstch)

~VREFA,B

Bits

TA = +25'C

DAC-8212A/E/G
DAC-8212B/F/H

±I
±6

TA = Full Temp. Range

DAC-8212NE/G
DAC-8212B/F/H

±2
±7

Data In 0000 0000 0000

TA = +25'C
TA = Full Temp. Range

5

8
0.1

LSB

±50
±400

nA

15

kll

±I

%

DIGITAL INPUTS
Digital Input High

DlgHallnput Low

Input Current
Input capacrtance
(Note 2)

V'NH

V'NL

Voo= +5V

2.4

VDO = +15V

13.0

V

Voo= +5V

0.8

Voo= +15V

1.5

V
TA =+25'C
TA = Full Temp. Range

I'N

Voex = OV or Voo

C'N

DBO-DBII
CS,DACA/DAC B

WR,

11-132

.001

±I
±IO

t-""
WRITE SELECT

1---..

"ow

""'----

}

S~~~~I8I ~or ~~AI AlB, A/W

tw

M
0
U~ LOW PULSES
MINIMUM
TIME BETWEEN
DATA HOLD nME (An'ER
... HIGH)

=

WRITE CYCLE (tDSW

+ iNw)

2Onst.

180 nstDsw MIN
80nstNw
20 ne tH
280nstw

MODE SELECTION TABLE
DS1

DS2

Alii

RIW

DACA

DACB

DACC

DACD

H

H

X

X

HOLD

HOLD

HOLD

HOLD

L

H

H

L

WRITE

HOLD

HOLD

HOLD

L

L

H

L

WRITE

HOLD

WRITE

HOLD

L

H

H

H

READ

HOLD

HOLD

HOLD

L

L

H

H

HOLD

HOLD

HOLD

HOLD

L

H

L

L

HOLD

WRITE

HOLD

HOLD

L

L

L

L

L

HOLD

WRITE

HOLD

WRITE

L

H

L

H

HOLD

READ

HOLD

HOLD

L

L

L

H

HOLD

HOLD

HOLD

HOLD

H

L

H

L

HOLD

HOLD

WRITE

HOLD

H

L

H

H

HOLD

HOLD

READ

HOLD

H

L

L

L

HOLD

HOLD

HOLD

WRITE

H

L

L

H

HOLD

HOLD

HOLD

READ

= Low State

H

= High State

X

= Irrelevant

11·139

1/86, Rev. A

III

PM-S62
12-BIT MULTIPLYING CURRENT-OUTPUT
D/A CONVERTER
Pn:'ClSIOil MOllolithlcs Inc.

FEATURES
•
•
•
•
•
•
•
•

and Increased output resistance. The PM-562 Is pin compatible with the AD562.

Nonlinearity ......................... ± 1/4 LSB (Max)
Settling Time ............................ 1.5 ~s (Typ)
No Laser TrImming Used In Fabrication
Intarnal Range and Offset Scaling Resistors
Guaranteed Monotonlclty Over Temperature
TTL or CMOS Logic Input Compatibility, Pin Selectable
Low Power Consumption •. . . . . . . . . • . • .. 130mW (Typ)
Dlractly Pin Compatible with AD582

A highly stable trim method; selective shorting of zener
diodes, provides 13-bit accuracy without the need for laser
trimming. Reliability of this trimming method has been
proven in several other PMI products with many years of
reliability history. Internal scaling resistors plus an external
op amp simplifies construction In voltage output applications,
while maintaining accuracy over wide operating temperature
ranges. The PM-562Is recommended for 12-blt accuracy DlA
applications where single-chip reliability, small size and low
cost are primary considerations.

ORDERING INFORMATIONt
NONLINEARITY

@2S·C(LSB)

PMI MODEL NO.

TEMP. RANGE

±1/4
±1/2
±1/2
±1/4
±1/2

PM562AV'
PM562BV'
PM562FV
PM562GV
PM562HV

-55· C/+ 125· C
-55·C/+125·C
-25· C/+65· C
0·C/+70·C
0·C/+70·C

PIN CONNECTIONS
V+

• For devices processed in total compliance to MIL-STD-~83. add /883 after
part number. Consult factory for 883 data sheet.
fAil commercial and Industrial temperature range parts are available with
burn-In. For ordering Information see 1984 Data Book. Section 2.

GENERAL DESCRIPTION
The PM-562 Is a 12-blt monolithic multiplying dlgltal-toanalog converter consisting of a reference current amplifier,
an R-2R ladder network, range and offset scaling resistors,
and 12 high-speed current switches. Improvements provided
by the PM-562 include greater negative power supply range

TOP VIEW
24-PIN HERMETIC DIP
'Connect VLCto V+ for CMOS logic Inputs. Leave VLCopsn circuit or
grounded for TTL inputs.

EQUIVALENT CIRCUIT
V+

i,

VLC

MSB
81

2

24

B2

B3

23

B4

86

21

22

B6

20

,.

87

B8

,.

B9

17

,.

LSS
B10 811 812

16

14

,.
11

PM-582

SUM

VREF(+I

VREf(-I

IREF~

4

5

•

19'~fll
20kfI

IFS=4X IREF

~ V.....

/"

I

LOGIC INPUT SWITCH DRIVERS

L
I
I
I

I
I
I

I

I
I

'J

I
I
I

I
I
I

I

I
I

'1

"''' ,.
"'''

l-'j j l~ j l~ j l~ j~~~

•
12

20V SPAN

tOV SPAN

10
GND

(113 SHOWN)

.........

"'"

......

... ......

•

BIPOLAR OFFSET
"R" OUT

7

BIPOLAR OFFSET

9.951dl

--

"R" IN

•
V-

Manufactured under tfilo. following patent: 4.055.773.

11·140

1/86, Rev. A

_ _ _ _ _ _ _ _-IIfMi)PM-58212-BIT MULTIPLYING CURRENT-OUTPUT D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
Supply Voltage (V+) •••••••••••••••••••••••••••••• + 18V
Supply Voltage (V-) •••••••••••••••••••••••••••••• -18V
V+ to V- •••••••••••••••••••••••••••••••••••••••••• 36V
Logic Inputs •••••••••••••••••••••••• V- to (V- plus 36V)
Summing Junction (Pin 4) •••••.••••••••••••••• V- to V+
CMOS/TTL Threshold (Pin 2) •••••••••••••••••• V- to V+
lOUT (Pin 9) ••••••••••••••••••••••••••••••• -5V to +18V
Span Resistors •••••••••••••••••••••••••••••••••••• 36V

Operating Temperature Range
PM-562A1B ......................... -55· C to + 125· C
PM-562F ••••••••••••••••••••••••••••• -25·Cto+85·C
PM-562G/H ••••••••••••••••••••••••••••• O· C to + 70· C
Storage Temperature Range ••••••••••• -65· C to + 150· C
Power Dissipation at TA = 125· C •••••••••••••••• 1000mW
Lead Temperature (Soldering, 60 sec) ••••••••••••• 300·C

ELECTRICAL CHARACTERISTICS atV+

=5V, V- =-15V, VREF (+) =+10.0000V, VREF(-) =OV, TA=+25·C,

~

unless otherwise noted.
PM-562A1G
MIN
TVP
MAX

PARAMETER

SYMBOL

CONDITIONS

Resolution

N

TA = Full Range

12

TA = Full Range

12

Monotonlclty

PM-562B/F/H
MIN
TVP
MAX
12

UNITS
Bits

12

Bits

~

~

0

Nonlinearity

NL

±1/4

±1/2

LSB

U

Differential Nonlinearity

DNL

±1/2

±1/2

LSB

Settling Time

Is

0
0

Output Voltage Compliance

Voc

Full-Scale Output
Current Range

IFR

Zero-Scale Current

Izs

Output Resistance

Ro

Output Capscitance

Co

Reference Input Impedance

ZIN

To ± 112 LSB, all bits
ON or OFF

1.5

1.5

- +10/-1.5
VREF (+) = +10.0000V Unipolar
Bipolar
R2=500

-1.6
±0.8

All bits OFF

I'S

V

- +10/-1.5

-2.0
±1.0

-2.4
±1.2

.005

0.05

-1.6
±0.8

-2.0
±1.0

-2.4
±1.2

mA

.005

0.05

%FS

2

Pln5

MO

30

30

pF

20

20

kO

Gain Error

I FSE

R2 = 500. (Note I)

±0.2

±0.2

%FS

Bipolar Zero-Scale Error

IBZSE

R1 = 500, (Note I)

±0.1

±0.1

%FS

Full-Scale Gain
Adjustment Range

AI FSR

R2= 1000 Trimpot. (Note I)

±0.25

±0.25

%FS

Bipolar Zero-Scale
Adjustment Range

±0.25

AlsZSR

R1 = 1000 Trimpot. (Note I)

Power Supply
Gain Sensitivity

+Pss
-P SS

V+=5VorV+=15V
V-=-15V

Supply Current

1+
l-

V+ = 4.75V to 15.8V
V- = -15V ± 10%

TTL Logic
Input Voltage

VIH
VIL

Pln2
Opsn Circuit

IIH= 100nA
IlL =-I00I'A

2.0

CMOS Voltage
Logic Input

VIH
VIL

Pln2
tied to Pin 1

IIH= l00nA
IlL =-I00I'A

70

±0.25
2
6

5
-7

11-141

%FS
2
6

18
-25

5
-7

18

-25

2.0
0.8

0.8
70

30

30

ppmFSI%
mA
V
%V+

1/86, Rev. A

~

,
0

~

~

0

0

---------I1fHD

PM-56212-BIT MULTIPLYING CURRENT-OUTPUT CIA CONVERTER

ELECTRICAL CHARACTERISTICS at V+ = 5V, V-=-15V, VREF (+) = +10.0000V, VREF (-) = OV,-55'C :STA:S+125'C for
PM-562A!B, -25' C:S TA:S +85'C for PM-562F, O'C:S TA:S +70'C for PM-562G/H, unless otherwise noted.
PM-562A/G
PARAMETER

SYMBOL

CONDITIONS

Zero-Scale Temperature
Coefficient

TCI ZS

Unipolar Leakage
Current Change (Note 2)

Bipolar Zero-Scale
Temperature Coefficient

TCl sZS

Bipolar (Note 2)

Full-Scale Gain
Temperature Coefficient
Differential Nonlinearity
Temperature Coefficient

TC ONL

MIN

TYP

PM-5628/F/H

MAX

MIN

TYP

MAX

UNITS
ppmFS/'C

4

ppmFS/'C

Excludes VREF (Note 2)

ppmFS/'C

(Note 2)

ppmFS/'C

NOTES:
1. See connection diagram.

2. Guaranteed by design.

CONNECTION DIAGRAM
+15V

VLe

MSB
81
82

83

84

85

B6

87

B8

89

810

B'1

LSB
812

11

---------jI

10

I

5k!)

PM-562

SPAN
RESISTOR

{

I
I
I

I

5k!)

I
I
I

REF AMP
SUMMING

JUNCTION

0---'-1----....- - - - - - ,
+15V

r - - - .....I

W\r--....::.f---'lM,-.....-+e-...

.....

I
I

I
I
I

I

UNIPOLAR
OFFSET

I
I

ADJUST

I

I
I

L________________________________________ Q'
_
_________________ I

I
I

I

I

BIPOLAR

-lSV

loon OFFSET ADJUST

I

~

NOTE:
Set VLC = V+ for CMOS logic inputs. Leave V LC open circuit or grounded for
TTL Inputs.

BURN-IN CONFIGURATION

PM·S62

+ 1 0 V o - - - - - - - - - -....-+----+--0 ~O~6 +10V)
v-15V

TTL DIGITAL INPUTS

NOTES:
1. Calibration Procedure First set digital inputs B, thru B'2 to digital
zero and adjust R, until VOUT = 0 OOOOV Next set digital Inputs B,
thru B'2 to digital one's and adjust R2 until VOUT = 9.9976V.

2.

Bypass V+ and V- supplies with O.01IlF in parallel with 1.uF

capacitors

Figure 2 +10V Unipolar Voltage Output

11-145

1/86, Rev. A

I

----------l1EM!) PM-56212-BIT MULTIPLYING CURRENT-OUTPUT D/A CONVERTER
-10 DODV

OUTPUT ADJ

O.OOOV
OUTPUT ADJ

VREF
"'10000V

3,12

-

> ....--oVOUT
(-10V TO +10VI

-15V

v-

TTL DIGITAL INPUTS

NOTES:
1. Calibration Procedure· Set VREF = 10.0000V. Next set dlgllallnputs
8, thru 8'2 to digital zero and adjust A, until VouT =-10.0000V. Set
digital inputs 8, thru 8'2to 100000000000 (I.e. only turn ON MSB)

Figure 3

and adjust R2 until VOUT = O.OOOOV.
Bypass V+ and V- supplies with 0.01"F
capacitors.

In

parallel with 1"F

±10V Bipolar Voltage Output

PRECISION X-Y POSITIONING SYSTEM

4-20mATRANSMITTER

lOUT

II

-15V

GND +15V

4-20mA

'-_ _-t-....V·OUT

11-146

1/86, Rev. A

---------I~ PM-5S2 12·BIT MULnPLYING CURRENT·OUTPUT D/A CONVERTER

DIGITALLY CONTROLLED R.F. OSCILLATOR
+V

I

11·147

1/86, Rev. A

PM-S6SA
COMPLETE HIGH-SPEED 12-BIT
MONOLITHIC D/A CONVERTER
Preci"'IOll Monolithics Inc.

PRELIMINARY
FEATURES

oto 10V, 0 to 5V, -5 to 5V, -10 to 10V output voltages using an

•
•
•
•
•
•
•
•

external op-amp and the internal resistors. The digital inputs are
TTL-compatible and may be driven by 5V or 15V CMOS.

Very Fast-Settling (Sample Teated) •••••••• 250ns Max
High-Stability Burled·Zener Reference on Chip
Linearity Guaranteed Over Temperature •• 1/2 LSB Max
Monotonlclty Guarenteed Over Temperature
Low Power Conaumptlon •••.•••..••.••• 345mW Max
Operation Guaranteed et :t 12V Supplies
Inputs Compatible with 5V and 15V Logic
AD565A Second·Source Pin Compatibility with Improved
Tempco, Reference Load Capability, and Trimming

Logic delay to analog output is typically only 40ns, and settling to
0.01% of full-scale is factory tested. Great attention has been
given to minimizing glitch energy and amplitude.
The subsurface zener reference Is very stable, both over temperature and time. Typical gain tempco is only 10ppm/oC. The zener
reference has less wideband noise than a typical bandgap type
and is also free from excessive low-frequency noise. The reference is trimmed to 2 LSB Typ.

ORDERING INFORMATIONt
PACKAGE
LINEARITY
OVER
FULL· SCALE
TEMPERATURE TEMPCO, MAX
RANGE
±1/2LSB
±3/4LSB
±1/2LSB
±3/4LSB

EPOXY
DIP
24-PIN

15ppm/'C
30 ppm/'C

PIN CONNECTIONS

HERMETIC
OPERATING
DIP
TEMPERATURE
24-PIN
RANGE
MIL
MIL
COM
COM

PM565AAV'
PM565ABV'
PM565AGP
PM565AHP

15ppm/'C

30 ppm/'C

• For devices processed in total compliance to MIL-STD-663, add 1883 after part
number. Consult factory for 663 data sheet.
t All commercial and Industrial temperature range parts are available with burn-In.
For ordering information see 1986 Data Book. Section 2.

GENERAL DESCRIPTION

AEFOUT

4

AEFGNO

6

AEFIN

6

BIPOLAR OFFSET

8

10VSPAN

10

20YSPAN

11

POWEAGNO

12

24·PIN
HERMETIC DIP
(V·Sufflx)

The PM-565A is a high-speed 12-bit bipolar digital-to-analog
converter which provides a trimmed reference and application resistors. The output is a 0-2mA current, which can be converted to

EQUIVALENT CIRCUIT
REF OUT

v+

BIPOLAR OFFSET

20VSPAN

10VSPAN
SkU

r--+--+-oIoUT

REFINlo-=t-"",,/\r-4-.;

BkO
FIEF GND

C>-''I--.-JVV..........

v-

81 82 83 B4 85 88 87 B8 B9 810 811 812
(MSBI-4-- DIGITAL LOGIC INPUTS ________ (LSB)

This preliminary product Information Is based on testing of a limited number of devices. Final specifications may vary. Please contact local.ales
office or distributor for final data aheet.

11-148

1/86, Rev. A

PM-7226
QUAD 8-BIT CMOS D/A CONVERTER
WITH VOLTAGE OUTPUT
Pn:CISIOIl MOJloiJthu.:s Inc.

ADVANCEPRODUCTINFO~TION
FEATURES
•
•
•
•
•
•
•
•
•

CROSS REFERENCE

Four Voltage Output DACs on a Single Chip
Microprocessor Compatible
Internal Output Buffer Amplifiers
Common 8- Bit Data Bus
AdJustment-Free .......••.•..... ±2LSB Total Error
Guaranteed Monotonlclty
Single or Dual Supply Operation
Latch-Up Resistant
Space Saving 20-Pln 0.3 Inch Wide DIP

APPLICATIONS
•
•
•
•
•
•
•

Automatic Test Equipment
Industrial Automation
Process Controls
Instrumentation Equipment
Medical Equipment
Multi-Channel Microprocessor-Controlled Systems
X-Y Graphics

PMI

ADI

TEMPERATURE
RANGE

PM7226AR
PM7226ER
PM7226GP

AD7226TD
AD7226BQ
AD7226KN

MIL
IND
COM

GENERAL DESCRIPTION
The PM-7226 places four a-bit voltage output CMOS digital-toanalog converters on a single chip. By using a single a-bit data
bus port, the device fits into a space saving 20-pin 300 mil DIP
package. The internal output buffer amplifiers can each drive up
to 5mA from either single or dual supply rails. Each DAC has
individually addressable data latches for easy microprocessor
interface.
The compact size, low power, and economical cost-per-channel,
makes the PM-7226 attractive for applications requiring multiple
D/A converters without sacrificing circuit-board space.
The PM-7226 is a direct replacement for AD7226.

ORDERING INFORMATIONt
PACKAGE: 2o-PIN""

PIN CONNECTIONS

TOTAL
COMMERCIAL
INDUSTRIAL
MILITARY"
UNADJUSTED TEMPERATURE TEMPERATURE TEMPERATURE
ERROR
±2LSB

PM7226GP

PM7226ER

PM7226AR

20-PIN HERMETIC DIP
(R-Suffix)

• For devices processed in total compliance to MIL·STD·883, add 1883 after part
number. Consult factory for 883 data sheet.
"Package Designation: SuffiX R: Hermetic DIP; Suffix P: Plasbc DIp
t All commercial and Industrial temperature range parts are available with burn·in.
For ordering Information see 1986 Data Book, Sactlon 2.

20-PIN EPOXY DIP
(P-Sufflx)

FUNCTIONAL DIAGRAM
v,""

> - _ - r2-oVOUTA

>---+-i.!.'OVQUTB

MSB
7
8-BIT

I

DATA BUS \r14!!jI_ _"'_'_U_US_--,
LSB

>-_-I"20'OVoU'<:

~ ~:
AD

17

This advance product Information describes a product In development at the time of this printing. Final specifications may vary, Please contact
local sales office or distributor for IInal data sheet.
1/86, Rev. A
11-149

II

PM-7524
CMOS 8-BIT BUFFERED MULTIPLYING
D/A CONVERTER
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

CROSS REFERENCE

±1/8 LSB Endpoint Linearity
Microprocessor Compatible
On-Board Data Latches
Guaranteed Monotonlclty Over Full Temperature Range
Low Power Consumption " , ' , ' , " , ' , , ' , ' SmW @ +SV
+SV to +1SV Operation
Full Four Quadrant Multiplication
TTL/CMOS Compatible
Latch-Up Resistant
No Schottky Diodes Required

APPLICATIONS
•
•
•
•
•

Microprocessor Controlled Circuits
Precision AGC Circuits
Bus Structured Instruments
Function Generators
Digitally Controlled Attenuators and Power Supplies

TEMPERATURE
RANGE

PMI

ADI

PM7524AO
PM7524BO
PM7524BO

AD7524UD
AD7524TD
AD7524SD

MILITARY

PM7524EO
PM7524FO
PM7524FO

AD7524CD
AD7524BD
AD7524AD

INDUSTRIAL

PM7524GP
PM7524HP
PM7524HP

AD7524LN
AD7524KN
AD7524JN

COMMERCIAL

PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE: 16-PIW'

NONLINEARITY

MILITARY'
TEMPERATURE
-55'C to +125'C

INDUSTRIAL
TEMPERATURE
-25'C to +85'C

COMMERCIAL
TEMPERATURE
Q'C to +7Q'C

±1/8(LSB)
±1/4(LSB)

PM7524AO
PM7524BO

PM7524EO
PM7524FO

PM7524GP
PM7524HP

16-PIN EPOXY DIP
(P-Suffix)

• For devices processed," total compliance to MIL-STD-883, add 1883 after
part number. Consult factory for 883 data sheet
"Package Designation Suffix 0 Hermetic DIP. Suffix P Epoxy DIP
tAil commercial and Industnal temperature range parts are available with
burn~jn. For ordenng information see 1986 Data Book, Section 2

GENERAL DESCRIPTION

16-PIN HERMETIC DIP
(Q-Suffix)

FUNCTIONAL DIAGRAM

The PM-7524 is an 8-bit monolithic multiplying digital-toanalog converter with input latches. It is compatible with all
popular 8-bit microprocessors including the 6800, 8080, 8085,
and Z80. It's load cycle is similar to that of a RAM's write cycle.

10kn

10kn

10kn

PMI's tightly controlled thin-film resistor processing provides
1/8 LSB linearity without laser trimming. The design incorporates a matching MOS transistor in series with the feedback
resistor to achieve a gain and linearity error tempco of
2ppm/oC.

20kn

L--++-4--I-+-+-~I--H-+---+-oOUT2

The PM-7524 exhibits excellent performance on a single +5V to
+15V power supply. It is TTL compatible at +5V and dissipates
less than 10 mW; at +15V it is CMOS compatible and dissipates
less than 30 mW.

' - i - -......+--......+---.!I--.....-:--.---~ aun
10kn

'-",fI/'r-+-QRFB

PMI's improved latch-up resistant design eliminates the need
for external protective Schottky diodes.

WR

The PM-7524 is manufactured using thin-film resistors on an
advanced oxide-isolated silicon-gate CMOS process.
OB7 (MSB)

DB.

DB'

DBO /LSBj

GND

DATA INPUTS

11·150

1/86, Rev. A

----------I~ PM-7524 CMOS 8-BIT BUFFERED MULTIPLYING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS

(TA = +25°C, unless

otherwise noted)
V DD (to GND) ................................ -0.3V, +17V
V REF (to GND) ..•...•..................•.•........ ±25V
RFB (to GND) ......•......................•......• ±25V
Digital Input Voltage To GND .....•........... -0.3V to V DD
Output Voltage (Pin 1, Pin 2) .................. -0.3V to V DD
Power Dissipation (Package)
Ceramic (Suffix 0)
To +75°C ...••...•..........•..........•....•. 450mW
Derates Above +75° C By .................••... 6mW/o C
Plastic (Suffix P)
To +70°C ..................................... 670mW
Derates Above +70° C By .................... 8.3mW/o C
Operating Temperature Range
Military (AO, BO Versions) .............. -55°C to +125°C
Industrial (EO, FO Versions) ............. -25° C to +85° C
Commercial (GP, HP Versions) .•..•........ O°C to +70°C

Dice Junction Temperature .•..........•..•........ +150° C
Storage Temperature ..................... -65°C to +150°C
Lead Temperature (Soldering, 60 sec) ............... +300° C
CAUTION:
Do not apply voltages higher than VDO or less than GND potential on any

terminal except VAEF (Pin 15) and RFB (Pin 16)

4

The digital control Inputs are zener protected, however, permanent
damage may occur on unconnected units from high energy electrostatic
fields Keep units In conductive foam at all times until ready to use
Use proper anti-static handling procedures
Absolute MaXimum Ratings apply to both packaged deVices and DICE
Stresses above those listed under Absolute MaXimum Ratings may cause
permanent damage to the deVice

ELECTRICAL CHARACTERISTICS at V REF = +10V; V OUT1 = V OUT2 = OV; TA = -55°C to +125°C apply for PM-7524AO/BO;
TA = -25°C to +85°C apply for PM-7524EO/FO; TA = O°C to +70°C apply for PM-7524GP/HP, unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

MIN

TYP

MAX

MIN

TYP

MAX

UNITS

STATIC ACCURACY
PM-7S24A/E/G
Resolution

Relative Accuracy

(Notes 1, 2)

N

INL

PM-7S24B/F/H

Voo::: +5 or +15V

TA = Full Temp Range

Bits

Voo ::: +5V
TA ::: Full Temp Range

±02

±02

Voo=+15V
TA = Full Temp Range

±005

±01

±10
±14

±10
±14

±05
±O 6

±O 5
±O 6

%FSR

Voo ::: +5V
Gain Error

(Note 3)

G FSE

TA = +25°C
TA = Full Temp Range

%FSR

Voo ::: +15V

TA = +25°C
TA ::: Full Temp Range
Voo ::: +5V

Gain T C
(Notes 4. 5)

TCG FS

TA = Full Temp Range

±O 004

±0004

Voo::: +15V
TA ::: Full Temp Range

±0001

±OO01

%FSRrC

Voo = +5V
DC Power Supply
Rejection

AGain/A Voo
(Notes 3. 6)

TA ::: +25°C

PSR

TA = Full Temp Range

008
016

0002
001

008
016

0001
0005

002
004

0001
0005

002
004

%FSR/%

Voo = +15V
TA ::: +25°C

TA = Full Temp Range
Voo ::: +5V
TA ::: +25°C

Output Leakage

TA = Full Temp Range
(Notes 7. 8)

Current
IOUT1' IOUT2

0002
001

±50
±400

±50
±400
nA

ILKG

Voo ::: +15V
TA :::+25°C

TA = Full Temp Range
(Notes 7. 8)

11-151

±50
±200

±50
±200

1/86, Rev_ A

I

-----------1\fMD

PM-7524 CMOS 8-BIT BUFFERED MULTIPLYING D/A CONVERTER

= +1DV; VOUT1 = VOUT2 = DV; TA = -55°C to +125°C apply for PM-7524AQ/BQ;
=-25°C to +85°C apply for PM-7524EQ/FQ; TA =DOC to +7DoC apply for PM-7524GP/HP, unless otherwise noted. (Continued)

ELECTRICAL CHARACTERISTICS at VREF
TA

PARAMETER

SYMBOL

CONDITIONS

MIN

TYP

MAX

MIN

TYP

MAX

UNITS

DYNAMIC PERFORMANCE

Input to 90% of
Final Analog
Output Current)
(Note 4)

PM-7524A1B

VDD = +5V
TA :::: +25°C

Propagation Delay
(From Dlgllal

tpD

PM-7524E/F/G/H

150
200

TA = Full Temp Range
(Note 9)

150
175
ns

VDD = +15V
TA = +25°C
TA = Full Temp Range
(Note 9)

65
90

65
80

PM-7524A/E/G

Output Current
Settling Time

(To 1/2 LSB)
(Note 4, 15)

ts

VDD = +5V
TA :::: +25°C
TA = Full Temp Range
(Note 9)

300
350

300
350
ns

Voo :::: +15V

TA :::: +25°C
TA = Full Temp Range
(Note 9)
Voo

AC Feedthrough
IOUT"I OUT2

PM-75248/F/H

FT

(Note 4)

200
250

200
250

025
050

0,25
050

20

20

120
30

120
30

30
120

30
120

= +5V or +15V

TA = +25°C
TA = Full Temp Range
(Note 10)

%FSR

REFERENCE INPUT
Input Resistance

(Pin 15 to GND)
(Note 11)

R'N

Von = +5V or +15V
TA = Full Temp Range

k!l

ANALOG OUTPUTS
Output Capacitance
COUT1 (Pin 1)

Voo

COUT2 (Pin 2)
Co

Voo

COUT1 (Pin t)
COUT2 (Pin 2)

= +5V or +15V

TA = Full Temp Range
(Note 4,12)

pF

= +5V or +15V

TA = Full Temp Range
(Note 4,13)

DIGITAL INPUTS

Digital Input
High

Digital Input
Low

V,H

V,L

=OV or VDO

+24

+135

+135

V

VDD = +5V
TA = Full Temp Range

+08

+08

VDD = +15V
TA = Full Temp Range

+1 5

+15

±1
±10

±1
±10

20

20

TA = +25°C
TA = Full Temp Range

C'N

TA = Full Temp Range
V ,N = OV

Voo

V

= +5V or +15V

liN

Input Capacitance
DBO-DB?

WR,CS
(Note 4)

+24

VDD = +15V
TA = Full Temp Range

Voo

Input Current
VIN

VDD = +5V
TA = Full Temp Range

JJA

= +5V or +15V

11-152

pF

1/86, Rev. A

_ _ _ _ _ _ _ _ _;~ PM-7524 CMOS a-BIT BUFFERED MULTIPLYING D/A CONVERTER

ELECTRICAL CHARACTERISTICS
TA

at VREF

= +10V;

VO UT1

= VO UT2 = OV;

TA

= -55°C to

+125°C apply for PM-7524AO/BO;

= -25°C to +85°C apply for PM-7524EO/FO; TA = OOC to +70o C apply for PM-7524GP/Hp, unless otherwise noted. (Continued)

PARAMETER

SYMBOL

CONDITIONS

MIN

TVP

MAX

MIN

TVP

MAX

UNITS

SWITCHING
CHARACTERISTICS
(Note 4)
PM-7524A/B

Vee = +5V
Chip Select to
Write Setup Time
(tWA = tes )
(Note 14)

TA = +25°C

tes

Ch I P Select to
Write Hold Time

PM-7524E/F/G/H

TA = Full Temp Range

170
240

170
220

Vee = +15V
TA = +25'C
TA = Full Temp Range

100
150

130

ns

100

PM-7524A/E/G

PM-7524B/F/H

PM-7524A1B

PM-7524E1F/G/H

Vee = +5V or +15V
TA = Full Temp Range

o

ns

Vee = +5V
TA = Full Temp Range

170
240

170
220

Vee = +15V
TA = +25'C
TA = Full Temp Range

100
150

100
130

Vee = +5V
TA = +25'C
TA = Full Temp Range

135
170

135
170

Vee = +15V
TA = +25'C
TA = Full Temp Range

60
100

60
BO

TA = +25°C

Write Pulse Width

(les :2: tWA' leH :2: 0)

Data Setup Time

tes

PM-7524A1E/G
Data Hold Time

Vee = +5V or +15V
TA = Full Temp. Range

ns

ns

PM-7524B/F/H
10

10

ns

II

POWER SUPPLY
Vee = +5V
Supply Current
(All Digital

TA = +25'C
TA = Full Temp. Range

Inputs = V'L or V,H )

Vee = +15V
TA = +25'C
TA = Full Temp Range

Supply Current
(All Digital
Inputs = OV or Vee)

Veo = .5V or ·15V
TA = +25'C
TA = Full Temp Range

NOTES:
1. Guaranteed monotonic over full temperature range and at Vee = +5V and
+15V
2. FSR (Full Scale Range) = VAEF - 1 LSB
3 USing Internal feedback resistor
4 Guaranteed by design and not production tested
5 Gain TC measured from +25' C to TMIN or from +25' C to T MAX
6. t!.Vee = ±10%
7. DBO-DB7 = OV; WR = CS = OV, VAEF = ±10V, for IOUT1'
B. DBO-DB7 = Vee, WR = CS = OV, VAEF = ±10V, for IOUT2

1
2

rnA
2
2

100
500

100
500

/lA

9 OUT110ad = 1000, C EXT = 13pF, WR = CS= OV, DBO-DB7=OVtoVee or
Vee to OV
10 VAEF = ±10V, f = 100kHz, DBO-DB7 = OV, WR = CS = OV
11. Temperature coeffiCient approximately equals +300ppm/'C
12 DBO-DB7 = Veo' WR = CS = OV
13 DBO-DB7 = OV, WR = CS = OV
14 See Timing Diagram
15. Extrapolated' ts (1/2 LSB) = tpe + 6.2r, where r = the measured first time
constant of the final RC decay

11-153

1/86, Rev. A

----------t1fMD

PM-7524 CMOS 8-BIT BUFFERED MULTIPLYING DIA CONVERTER

DICE CHARACTERISTICS
1.0UT1
2.0UT2
3. GND
4. DB7(MSB)
5. DB6
6. DB5
7. DB4
8. DB3

9.
10.
11.
12.
13.
14.
15.
16.

DB2
DB1
DBO (LSB)
CS
WR
Voo
VREF
RFB

For additional DICE information refer to
1986 Data Book, Section 2.
DIE SIZE 0.067 x 0.070 inch; 4690 sq. mils
(1.7 x 1.78 mm, 3.03 sq. mm)

WAFER TEST LIMITS at VREF = +10V, Voun = VOUT2 = OV; TA = +25°C.
PM-7524G
PARAMETER

SYMBOL

CONDITIONS

LIMIT

UNITS

STATIC ACCURACY
N

Vee = +5V or +15V

8

Bits MIN

Relative Accuracy
(Notes 1, 2)

INL

Vee
Vee

=+5V
=+15V

±002
±005

%of FSR MAX

Gam Error
(Note 3)

GFSE

Voe

=+5V or +15V

±005

%of FSR MAX

DC Power Supply
Rejection
aGain/aVoe
(Notes 3, 4)

Vee

=+5V

008

PSR

Voo

=+15V

002

Resolution

Output Leakage
Current

% of FSR/% MAX

ILKG

Vee =+5V or +15V
(Notes 5, 6)

±50

nAMAX

R'N

Vee

=+5V or +15V

5/20

kO MINIMAX

Digital Input
High

V,H

Voe

=+5V

V DD = +15V

+24
+135

VMIN

Digital Input
Low

V,L

Voe =+5V
Voe = +15V

+08
+15

V MAX

liN

Voo

=+5V or +15V

±1

/lAMAX

lee

Vee
Vee

=+5V
=+15V

lee

Vee

=+5V or +15V

'OUT1, 'OUT2

REFERENCE INPUT
Input Resistance
(Note 7)
DIGITAL INPUTS

Input Current

(V'N

=OV or Vee)

POWER SUPPLY
Supply Current
(V ,N =V,L or V,H)
Supply Current
(V'N

= OV or Voo)

mAMAX

100

/lAMAX

NOTES:
Guaranteed monotonic over full temperature range and at VDD = +5V and
4 ~VDD = ±10%
+15V
5 DBO-DB7 =OV, WR =CS =OV, VAEF =±10V, for IOUT1
2 FSR (Full Scale Range) =VAEF - 1 LSB
6
DBO-DB7 =Voe, WR =CS =OV, VREF =±10V, for IOUT2
3 USing Internal feedback resistor
Temperature coefficient approximately equals +300ppm/oC
Electncal tests are performed at wafer probe to the limits shown Due to vanatlons In assembly methods and normal Yield loss, Yield after packagmg IS not guaranteed
for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

11-154

1/86, Rev. A

------------l\fMD

PM-7524 CMOS a-BIT BUFFERED MULTIPLYING D/A CONVERTER

DEFINITIONS

FIGURE 1: PM-7524 Functional Diagram

RESOLUTION
The resolution of a DAC is the number of states (2") that the
full-scale range (FSR) is divided (or resolved) into, where n is
equal to the number of bits. Resolution in no way implies
linearity.

VREF

20kO

RELATIVE ACCURACY
Relative accuracy or end-point nonlinearity is a measure of
the maximum deviation from a straight line passing through
the end-points of the DAC transfer function. It is measured
after adjusting for ideal zero and fUll-scale and is expressed
in % or ppm of full-scale range or (sub) multiples of 1 LSB

GAIN
Ratio of the DAC's external operational amplifier output voltage to the V REF input voltage when using the DAC's internal
feedback resistor.

2Ok"

10k"

20kn

.

2Ok"

2Ok"

QUT2

10k{l

oun
A.B

cs
iNA

PROPAGATION DELAY
The time for the output current to reach 90% of its final value
from a given digital input signal.
SETTLING TIME
Time required for the output function of the DAC to settle to
within 1/2 LSB for a given digital input stimulus, i.e., zero to
full scale.

10ks}

10kll

DB7 (MSa)

DB.

DB.

DBO (LSB)

logic are shown The switches are binarily weighted and
switch the ladder current between IOUT1 and IOUT2 bus lines;
this switching allows a constant current to be maintained in
each resistor leg regardless of the switch state.

EQUIVALENT CIRCUIT ANALYSIS

FEEDTHROUGH ERROR
Error caused by capacitive coupling from V REF to output with
all switches off.

Figure 2 shows an equivalent circuit for the PM-7524 with all
digital inputs Law. The OUT1 and OUT2 leakage current
source is the combination of surface and junction leakages to
the substrate. The 1/256 current source represents the constant 1-bit current drain through the ladder termination resistor. The situation is reversed with all digital inputs HIGH,
i.e., the current output is now switched to the OUT1 terminal.
The output capacitance is dependent upon the digital input
code, and is therefore modulated between the low and high
values.

OUTPUT CAPACITANCE
Capacitance from IOUT1 and IOUT2 terminals to ground.

FIGURE 2: PM-7524 Equivalent Circuit
(All Digital Inputs LOW)

GAIN ERROR
Gain error or full-scale error is a measure of the output error
between an ideal DAC and the actual device output. Ideal
output is equal to VREF - 1 LSB.

OUTPUT LEAKAGE CURRENT
Current which appears on IOUT1 terminal with all digital
inputs low or on IOUT2 terminal when all inputs are high.

A.B
R .... 'Okn

CIRCUIT DESCRIPTION
. - - - - -.....-~---oOUT'

CIRCUIT INFORMATION
The PM-7524 is an 8-bit multiplying CMOS digital-to-analog
converter with on-board data latches. It is fabricated using a
highly stable thin-film R-2R resistor ladder network and eight
N-channel current switches. A voltage or current reference
and an operational amplifier are all that is required in the
majority of applications.

V AE •

e>--J\II/V-_----_---......- - - - - o O U T 2

Figure 1 shows a simplified circuit of the PM-7524 converter.
The R-2R ladder, current steering switches, and interface

11-155

-~~~~~~~~~~~------~--

- - ------------

1/86, Rev. A

-----------I1fMD

PM-7524 CMOS a·BIT BUFFERED MULTIPLYING D/A CONVERTER

INTERFACE LOGIC

FIGURE 3: Supply Current vs Logic Level

MODE SELECTION
The mode selection is controlled by the CS and WR inputs.

14

1400
TA " +25°C

ALL DIGITAL INPUTS

12

WRITE MODE
The PM-7524 is in the WRITE mode when both the CS and WR
are both LOW; the input latches are transparent and the output
immediately follows the data input logic. See the MODE
SELECTION TABLE.

~
}"

TIED TOO ETHER

ooo~

1

10

Voo
-'+5V

BOO"

}

~
C

F

HOLD MODE
The MODE SELECTION TABLE shows the output results when
either CS or WR is HIGH. The output holds the value corre·
sponding to the last digital inputs prior to CS or WR assuming
the HIGH state.

1200

BOO

'i

400

~

,OO=+15V
4

o

V

o

'"

"-

........

200

i'-.

10

12

o

14

V IN (VOLTS)

MODE SELECTION TABLE
~

WR

MODE

L

L

WRITE

DAC responds to data bus
(DBO - DB7) inputs (trans·
parent)

H

X

HOLD

Data bus (DBO locked out

X

H

HOLD

DAC holds last data present
when Wl'i or~ assumes a
HIGH state

DAC RESPONSE

APPLICATIONS
FIGURE 4: Unipolar Binary Operation
(2-Quadrant Multiplication)

DB7) is

DATA

INPUTS

L = Low State, H = High State, X = Don't Care.

VOUT

cs
w-

WRITE CYCLE TIMING DIAGRAM
NOTES:
1. At AND R2 USED ONLY IF GAIN

GNO

ADJUSTMENT IS REQUIRED.
2 C1 PHASE COMPENSATION (10pF - 15pF)
MAY BE REQUIRED WHEN USING HIGH
SPEED AMPLIFIERS TO PREVENT
RINGING OR OSCILLATION.

TABLE I: Unipolar Binary Code Table
DIGITAL INPUT
MSB
LSB

DATA IN

(DBO - DB7)

-----'

ANALOG OUTPUT
-VREF

(~~ )

0 0 0 0 0 0

-VREF

( 129 )
256

0 0 0 0 0 0 0

-VREF

( 128 )
256

0

-VREF

(~~ )

0 0 0 0 0 0 0

-VREF

(2k)

0 0 0 0 0 0 0 0

-VREF

(

NOTES:
1. AU INPUT SIGNAL RISE AND FALL TIMES

MEASUREDFROM10IAlT080IIJoQFYDD Voo
=+5Y,t,"'t,= 2One; VDD"'+1~
.on..

1,=,,=

2 TlMlNGMEASUREMENTFlEFERENCELEVEL
I8VIH+VIL

2

VREF
--2-

3. tDS+tDHISAPPROXIMATELYCONSTANTAT
145n1 MIN AT +25OC, Voo '" +5Y AND

twr;;

170111 MIN THE PMa7524IB SPECIFIED FOR
A MINIMUM IIJH OF 1Dnl, HOWEVER, IN
APPlICAnONS WHERE IbH> 1Onl,

to. MAY

BE REDUCED UP TO THE LIMIT fDa " 8Sns.
tDH = 8On..

Supply current (100) versus Logic input voltage (V,N) is shown in
Figure 3. This plot shows the supply current for both Voo = +5V
and Voo =+15V.

2~6 )

=0

NOTE:
_

1LSB

11-156

= (2

8)

(VREF)

1

= 2s6

(VREF)

1/86, Rev. A

-----------l~ PM-7524 CMOS 8-BIT BUFFERED MULTIPLYING D/A CONVERTER

FIGURE 5: Bipolar (4-Quadrant) Operation
±10V
(AC OR DC)

VREF

Voo
A5

Your

DATA

INPUTS

I

LSB

11

CS

12

VA

13
OND
NOTES.
1. ADJUST Rl FOR Vo ur " OV AT CODE 10000000.
2. Cl PHASE COMPENSATION (10 - 15pF) MAY BE
REQUIRED IF A11S A HIGH SPEED AMPLIFIER.

OND

FIGURE 7: PM-7524/MC6800 Interface

TABLE II: Bipolar (Offset Binary) Code Table
DIGITAL INPUT
MSB
LSB

ANALOG OUTPUT
AO-A15

(.J£L)
128
o

0 000 0

o

000 000

VMA

o

o

-VREF

I-......----I~

6800

(1~8)

o

000 000

( 127 )
128

o

000 0 0 0 0

( ~)
128

.2t--.......,

II

DO-07

NOTE:
1LSB

1
= (2- ') (VREF) = 128

(VREF)

FIGURE 6: PM-7524/8085A Interface

11-157

1/86, Rev. A

----------I~ PM-7524 CMOS 8-BIT BUFFERED MULTIPLYING DIA CONVERTER

FIGURE 8: Power Generation Connection

FIGURE 9: Divider
(Digitally Controlled Gain)

-::::=+';;,:;;....1

OAT A 9-.....

INPUT "0" 6--+.....--+-1

cso-++-.....+-t

v,

WRo-++.....++-1

>-----<~----o VOUT

v,

EQUATIONS
-YIN

VOUT

Av::

=

----0-

-VOUT

"""'YiN

1

=:

-

0

WHERE. Av

= VOLTAGE GAIN

AND WHERE'

D=D2~7+D;6+

OBO
,.-

DBN = lorD

EXAMPLES

o
o

= 00000000, Av = - AOL (OP AMP)
= 00000001, Av = - 256

0= 10000000,Av =
o

-Ws

=:-2

= 11111111, Av = - ~

CIRCUIT EQUATIONS
V, =: - (VREF)
V2 =
(VAEF )
Vn = - (YREF)
Vn = + (VREF)

+

(D)
(02)
(Dn). n AN ODD INTEGER
(On). n AN EVEN INTEGER

WHERE

D=D:'7+D~6+

OBO
,.-

AND
DBn

= lorO

11-158

1/86, Rev. A

PM-7528
DUAL 8-BIT BUFFERED
MULTIPLYING CMOS D/A CONVERTER
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•

CROSS REFERENCE

On-Chip Latches For Both DACs
+5V To +15V Single Supply Operation
DACs Matched To 1%
Four-Quadrant Multiplication
TTL/CMOS Compatible
8-Bit Endpoint Linearity (±1/2 LSB)
Full Temperature Operation
Low Power Consumption
Microprocessor Compatible

PMI

TEMPERATURE
RANGE

ADI

PM7528AR
PM7528BR
PM7528BR

AD7528UD
AD7528TD
AD7528SD

MILITARY

PM7528ER
PM7528FR
PM7528FR

AD7528CO
AD7528BO
AD7528AO

INDUSTRIAL

PM7528GP
PM7528HP
PM7528HP

AD7528LN
AD7528KN
AD7528JN

COMMERCIAL

APPLICATIONS
•
•
•
•
•
•
•

Digital Gain/Attenuation Control
Digital Control Of Filter Parameters
Digitally-Controlled Audio Circuits
X-Y Graphics
Digital/Synchro Conversion
Robotics
Ideal For Battery-Operated Equipment

GENERAL DESCRIPTION
The PM-7528 contains two 8-bit multiplying digital-to-analog
converters. Excellent DAC-to-DAC matching and tracking results from monolithic construction. The PM-7528 consists of
two thin-film R-2R resistor-ladder networks, two data latches,
one input buffer, and control logic. Operation from a 5 to 15 volt
single power supply dissipates only 20mW of power.

ORDERING INFORMATIONt
PACKAGE: 20-PIN"
MILITARY'
INDUSTRIAL
COMMERCIAL
RELATIVE
GAIN TEMPERATURE TEMPERATURE TEMPERATURE
ACCURACY ERROR -55°C TO +125°C -25°C TO +85°C DOC TO +7DoC

±1/2 LSB
±1/2 LSB

±2 LSB
±1 LSB

* For devices processed

PM-7528BR
PM-7528AR

PM-7528FR
PM-7528ER

PM-7528HP
PM-7528GP

total compliance to MIL-STO-883, add /883 after
part number Consult factory for 883 data sheet
** Package designation Suffix A Hermetic DIP, Suffix P PlastiC DIp
t All commercial and industrial temperature range parts are available with
burn-In For ordering InformatIOn see 1986 Data Book, Section 2
In

Digital input data is directed into one of the DAC data latches
determined by the DAC selection control line DAC A/DAC B.
The 8-bit wide input data path provides TTL/CMOS compatibility. The data load cycle is similar to the write cycle of a
random access memory. The PM-7528 is bus compatible with
most 8-bit microprocessors, including the 6800, 8080, 8085, and
Z80.
The PM-7528 saves PC board space with the narrow 20-pin 0.3"
wide DIP. A set of internal tracking span resistors is included,
minimizing external parts.

PIN CONNECTIONS

FUNCTIONAL DIAGRAM
20-PIN EPOXY DIP
(P-Suffix)
20-PIN HERMETIC DIP
(R-Suffix)

11-159

1/86, Rev_ A

II

_ _ _ _ _ _ _ _-I~PM-7528 DUAL 8-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS
(TA = +25°C, unless otherwise noted.)
VDD to AGND .................................. OV, +17V
VDD to DGND .................................. OV, +17V
AGND to DGND .............•....•................ VDD
DGND to AGND ................................... VDD
Digital Input Voltage to DGND ...••.....••....•. -0.3\/, VDD
VplN 2, VPIN 20 to AGND .............•.......... -0.3V, VDD
VREF A, VREF S to AGND ........................... ±25V
VRFB A, VRFB S to AGND ........................... ±25V
Power Dissipation (Any Package) to +75°C ........ 450mW
Derate Above +75°C by ......•.•...........•.• 6mWfOC
Operating Temperature Range
AR, SR Versions ............•••....... -55°C to +125°C
ER, FR Versions ...•.................... -25°C to +85°C
Gp, HP Versions .............•••........•• O°C to +70°C

Dice Junction Temperature ...•.•.•....•......•... +150°C
Storage Temperature .•.................. -65°C to +150°C
Lead Temperature (Soldering, 60 sec) .•........... +300°C
CAUTION:
1.

Do not apply voltages higher than VDD or less than GND potential on any
terminal except VREF
The digital control inputs are zener-protected, however, permanent damage

may occur on unprotected units from high-energy electrostatic fields
Keep Units In conductive foam at all times until ready to use
Do not insert this device into powered sockets, remove power before
insertion or removal
4

Use proper anti-static handling procedures,

5

Stresses above those listed under"Absolute Maximum Ratings" may cause
permanent damage to the device

ELECTRICAL CHARACTERISTICS at VDD =+5V, VREF A =VREF S =+10V, OUT A =OUT S =OV; TA =-55°C to +125°C apply for
PM-7528ARISR; TA =-25 0 Cto +85°C apply for PM-7528ER/FR; TA =O°C to +70°C apply for PM-7528GP/HP, unless otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

PM-7528A/E/G
MIN
TYP
MAX

MIN

PM-7528B/F/H
TYP
MAX

UNITS

STATIC ACCURACY
(Note 1)
Resolution
Relative Accuracy

(Note 2)
Differential Nonlinearity

(Note 3)

N
NL
DNL

Full Scale
Gain Error

Bits

GFSE

(Note 4)

TA = +25°C

TA

=

Full Temp. Range

±1/2

±1/2

LSB

±1

±1

LSB

±1
±3

±2
±4

LSB

±0.007

±0.007

%IOC

±50
±400

±50
±400

nA

15

15

kO

±1

%

Gain Temperature

Coefficient
(ll.Gain/ll.Temperature)
(Notes 4, 10)

TCG FS

Output Leakage Current
Out A (Pin 2)/Out B (P," 20)
(NoteS)

'LKG

Input Resistance
(VREF A, VREF B)
(Note 6)

RREF

VREF AIVREF B
(Input Resistance Match)

ll.VREFA, B

TA = +25°C
TA = Full Temp. Range

0.1

±1

01

DIGITAL INPUTS
(Note g)
Digital Input High

V 1NH

Digital Input Low

V1Nl

Input Current
(Note 7)

lIN

Input Capacitance
(Note 10)

C,N

24

2.4

V

08
001

TA = +25°C

TA

=

Full Temp. Range

±1
±10

.001

0.8

V

±1
±10

I'A

DBO-DB7

10

10

WR, CS, DAC A/DAC B

15

15

11-160

pF

1/86, Rev. A

----

~--~-

IfMD

----

---

PM-7528 DUAL B-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER

ELECTRICAL CHARACTERISTICS at V DD = +5V, VREF A = VREF B= +10V, OUT A

= OUT B = OV; TA = -55°C to +125°C apply for

PM-7528AR/BR; TA =-25°Cto +85°C apply for PM-7528ERlFR; TA = O°Cto +70°C apply for PM-7528GP/HP, unless otherwise noted.
(Continued)

PM-7528A/E/G
PARAMETER

SYMBOL

CONDITIONS

MIN

TYP

PM-7528B/F/H

MAX

MIN

TYP

MAX

UNITS

SWITCHING CHARACTERISTICS
(Notes 10, 11)
ChIp Select to
Write Set-Up Time

les

Chip Select to
Wrote Hold TIme

tCH

DAC Select to
Wrote Set-Up TIme

t AS

DAC Select to
Write Hold TIme

tAH

Data ValId to
Write Set-Up Time

tos

Data Valid to
Wrote Hold TIme

tOH

Write Pu Ise Width

tWR

TA
TA

=+25°C
=Full Temp. Range

TA = +25°C

=Full Temp Range
TA =+25°C
TA =Full Temp Range
TA =+25°C
TA =Full Temp. Range
TA =+25°C
TA =Full Temp. Range
TA

TA
TA

=+25°C
=Full Temp

Range

200
230

200
230

ns

20
30

20
30

ns

200
230

200
230

ns

20
30

20
30

ns

110
130

110
130

ns

0

0

ns

180
200

180
200

ns

E2

All DIgital tnputs V,NL or V,NH

mA

100

100

All DIgital Inputs OV or Voo

100

f.lA

AC PERFORMANCE
CHARACTERISTICS
(Note 13)

0
0

~

0

~

~

-

0

Q

DC Supply RejectIon
(;';;-;528 DUAL ii-BiT BUFFERED MULTIPLYING CMOS D/A CONVERTER

TYPICAL PERFORMANCE CHARACTERISTICS
TOTAL UNADJUSTED ERROR
vs DIGITAL INPUT
1/4

TA ~ 25"C

FULL-SCALE GAIN ERROR
vs TEMPERATURE

RELATIVE ACCURACY

vs REFERENCE VOLTAGE
1/2 ,..-...,--,-,--,--r-""""""T-""'-'-'"

-OJA

015
Voo'" +lSV

Voo '" 1SV

VREF '" +10V

TA = 2S"C
014

1/4

-1/4

1--t--i-t---+-r---t--t-t--t--1
MAXIMUM POSITIVE ERROR

FOR THIS TYPICAL UNIT LINEARITY
TRACKS WITHIN 0 1% BETWEEN
OAe A AND OAe B

013

012

1/4

DAca

-1/4

f--t--i-t--t-I--t--t-t--t--1

64

12.

192

256

-6-4-20

DIGITAL CODE

OUTPUT
LEAKAGE CURRENT
vs TEMPERATURE

•

VAEF

-50

-25

25

50

FEEDTHROUGH vs
FREQUENCY

-10
100

R~ .I,b~~

...

'It
OP-Ol

-40

PHASE RELATIVE
TO PHASE AT 1KHz

VDD=~y

10

ffi

ew

-~
1
-75

:1

....." /~
-50

-25

25
50
75
TEMPERATURE (OC)

100

it

III~

-20
VREF(lN) '" 6
Voo'" 15V

-40

VR~~ 11111

TA '" 2SOC

OIGIT~~IIIINPUT '" 1~?llll111
-60

125

1k

10k

lOOk

~

OP-Ol

III

FREQUENCY (Hz)

11-166

1/

-80

~~~i' 'II
1M

125

VREF = 6 YAMS
RL = lOon

_60

It:on

100

Voo =' ~~U
-20

-20

":=

75

GAIN AND PHASE SHIFT
vs FREQUENCY
WITH RESISTIVE LOAD
AND OP-01 AMPLIFIER
GAIN RELATIVE TO
GAIN AT 1KHz

"~

010
-75

TEMPERATURE (oC)

=-lOV

w

10

REFERENCE VOLTAGE (VOLTS)

10

1000

"0,

~

011

AT EACH REFERENCe VOLTAGE, ALL
DIGITAL CODES WERE TESTED, THEN
THE MAXIMUM ERROR IS PLOTTED HERE

-1/4

1
"«

i""""'- r-..

"

-100

-120
10M

1k

10k

lOOk

1M

10M

FREQUENCY (Hz)

1/86, Rev. A

_ _ _ _ _ _ _ _----1~PM-7528 DUAL a-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER

TYPICAL PERFORMANCE CHARACTERISTICS
TOTAL HARMONIC DISTORTION
vs FREQUENCY
Voo

-20

21~~

III

ANALOG CROSSTALK
vs FREQUENCY
Vo~ ~ljM~

I

VREF = 6 VRMS
TA = 2SoC

001

I I

VREFB = 6 VRMS

-20

VOUTA
RATIO = - -

OP-Ol OUTPUT OP AMP

POWER SUPPLY REJECTION
vs TEMPERATURE

+ttHtt-+t++H+\I--+-++ftIIlI

11111~REfB

-40

-40~HH+W~++~~+~H+\I--+~ftIIlI

-60

-60

r-

-80

H-t+I-tttIt--+H:II~/++-ttttlI\t-t+t-ttllll

-

-80

-100

-120

01

10

100

0001

JJ~JlLlL AMPLIFlER~~1-ttHtIII

Voo

FREQUENCY (kHz)

lOOk

1M

-

00001

::::1:k:::::::::I:II:ilL:11111:11'0:01:::::~:::::
10k

4 75V TO 525V= ----'

r--- ~
000001
-75

10M

-50

Vao = 14V TO l6V

-I

-25

I I
25

50

75

lOa

125

TEMPERATURE (OC)

FREQUENCY (Hz)

SUPPLY CURRENT
vs TEMPERATURE

-

Voo

15V

V 1NL ""

5V

=

01

i,.....-'"

001
-75

V
-50

~
-25

....

~INH·r5V f--

25

50

75

100

II

125

TEMPERATURE (OC)

VOLTAGE SWITCHING MODE CHARACTERISTICS
RELATIVE ACCURACY

GAIN AND PHASE
vs FREQUENCY

vs REFERENCE VOLTAGE

......
-10

-20

....
-1 ~--'~;;--I--45

-90

vREF(IN) = 100mV RMS + 2Voc
voo = 15V

DIGIT~fIlIINPUT
10

10

REFERENCE VOLTAGE (VOL TS)

=

'~I~~I1111
100

1000

10000

FREQUENCY (kHz)

11-167

1/86, Rev. A

_ _ _ _ _ _ _ _--1!EMDPM-7528 DUAL 8-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER

PARAMETER DEFINITIONS

AC FEEDTHROUGH
AC signal due to capacitive coupling from VREF to output with
all switches "off."

RELATIVE ACCURACY
Relative accuracy, or endpoint nonlinearity, is a measure of the
maximum deviation from a straight line passing through the
endpoints of the DAC transfer function. It is measured after
adjusting for zero and full scale, and is normally expressed in
LSB's or as a percentage of full scale reading.

INTERFACE LOGIC INFORMATION
DAC SELECTION
Both DAC latches share a common 8-bit input port. The control
input DAC AlDAC B selects which DAC can accept data from
the input port.

DIFFERENTIAL NONLINEARITY
Differential nonlinearity is the difference between the measured
change and the ideal 1 LSB change between any two adjacent
codes. A specified differential nonlinearity of ±1 LSB maximum
over the operating temperature range ensures monotonicity.

MODE SELECTION
The inputs CS and WR control the operating mode of the
selected DAC. See Mode Selection Table below.

GAIN ERROR
Gain error, or full-scale error, is a measure of the output error
between an ideal DAC and the actual device output. The ideal
full-scale output is VREF minus 1 LSB. Gain error of both DAC's
in the PM-7528 Is adjustable to zero with external resistance.

WRITE MODE
When CS and WR are both low, the selected DAC is in the write
mode. The input data latches of the selected DAC are transparent and its analog output responds to the data on the data bit
lines DBO-DB7.

OUTPUT CAPACITANCE
CapaCitance from OUT A or OUT B to AGND.

HOLD MODE
The selected DAC latch retains the data which was present on
the data lines just priorto CS or WR assuming a high state. Both
analog outputs remain at the values corresponding to the data
in their respective latches.

DIGITAL CHARGE INJECTION
The amount of charge injected from the digital inputs to the
analog output when the inputs change states. This is normally
specified as the area of the glitch in either pAsecs or nVsecs,
depending upon whether the glitch is measured as a current or
voltage signal. Digital charge injection is measured with VREF A,
VREF B = AGND.

MODE SELECTION TABLE
DACAlDACB

CS

WR

DACA

PROPAGATION DELAY
This is a measure of the internal delays of the circuit. It is
defined as the time from a digital input change to the analog
output current reaching 90% of its final value.
CHANNEL-TO-CHANNEL ISOLATION
The portion of input signal from one DAC's reference input
which appears at the output of the other DAC, expressed as
a ratio in dB.

DACB

L

L

L

WRITE

HOLD

H

L

L

HOLD

WRITE

X

H

X

HOLD

HOLD

X

X

H

HOLD

HOLD

L = Low State

x =Don't Care

H = High State

CIRCUIT INFORMATION-D/A SECTION

DIGITAL CROSSTALK
The glitch energy transferred to the output of one converter,
due to a change in digital input code to the other converter,
specified in nVsec.

The PM-7528 contains two identical8-bit multiplying digital-toanalog converters, DAC A and DAC B. Each DAC includes a
stable thin-film R-2R resistor ladder and eight NMOS current
steering switches. Figure 1 shows a simplified equivalent circuit

WRITE CYCLE TIMING DIAGRAM

~S-'I'-'CH)_
CHIP SELECT - - - -....{

'AS-l-'AH
IlACA/0AC B - - - -....{

_

WRiTE

~'WR~
"--

_ _ _ _ _......
DATA IN 10BO-OB7I

VOO

_ _ VOO

~

t='os

y_-----

VOO

-I-

~TA IN STABLE

NOTES
, ALL INPUT SIGNAL RISE AND FALL

t':_H____

X

11-168

TIMES MEASURED FROM 10% TO 90%
OFV oo
Voo· +5V, tr = tf" 2On5,

voo

Voo· +15V, tr = tf = 40ns
2 TIMING MEASUREMENT REFERENCE
LEVEL IS V 1H + V1l

2

1/86, Rev. A

----------1~ PM-752a DUAL a-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER

of either DAC. The inverted R-2R ladder takes a voltage or
current reference and divides it in a binary manner among the
eight current steering switches. The number of switches
selected to the output (OUT) add their currents together
forming an analog output current representation of the switch
selection. The DAC OUT and analog ground (AGND) should be
maintained at the same voltage for proper operation.

1 LSB of the reference current (lREF)' COUT is the parallel
combination of the NMOS current steering switches. The value
of COUT depends on the number of switches connected to the
output. The range of COUT is 50pF to 120pF maximum. The
equivalent output resistance Ro varies with input code from
O.8R to 3R, where R is the nominal ladder resistor of the R-2R
ladder.

CIRCUIT INFORMATION-DIGITAL SECTION
The digital inputs provide TTL input compatibility (VINH = 2.4,
VINL = O.8V) when the PM-7528 operates with Voo of +5V. The
digital inputs effect the amount of quiescent supply current as
shown in Figure 3. Peak supply current occurs as the digital
input (VIN) passes through the transition voltage. Maintaining
the digital input voltages as close as possible to the supplies
(Voo and DGND) minimizes supply current consumption.
When operating the PM-7528 from CMOS logic the digital
inputs are driven very close to the supply ralls, minimizing
power consumption.

2"

L+~44-r~~~t-~~-~~--oOUT

~-++-~:~~~--+~~~--~AGNO
OAC DA4A LATCHES
AND DRIVERS

Digital input protection from electrostatic discharge and electrostatic buildup occurs in the input network shown in Figure 4.

FIGURE 1: Simplified functional circuit for DAC A or DAC B.
EQUIVALENT CIRCUIT ANALYSIS

30

The equivalent circuit of DAC A shown in Figure 2 is similar to
DAC B. DAC A and DAC B both share the analog ground pin 1
(AGND). With all digital inputs high, the reference current flows
to OUT A. A small leakage current (lLEAKAGE) flows across
internal junctions, doubling every 10°C. The R-2R ladder
termination resistor generates a constant 11256 current which is

TA '" 25 Q C
ALL DIGITAL INPUTS
TIED TOGETHER

,L:r- Voo

OD
V

-----,
VREF C>--'I,/Vv-_--.......p----__ VOUT

ANALOG OUTPUT
(DAC A or DAC B)

~~~~(7)o-------+-------.

127 )
+VIN ( 128
+15V

10000001
10000000

o

01111111
00000001

127 )
-VIN ( 128

00000000

128 )
-VIN ( 128

FIGURE 7: PM-7528 in Single Supply, Voltage
Switching Mode

11-172

1/86, Rev. A

_ _ _ _ _ _ _--tIf.Mi)PM-752a DUAL a-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER

PROGRAMMABLE WINDOW COMPARATOR

input signal range depends on the reference and polarity. that Is
the test input range is 0 to minus VREF' The A and B data latches
are programmed with the upper and lower test limits. A signal
within the programmed limits will drive the output to logic high.

A programmable window-comparator In Figure 8 will determine
if voltage inputs applied to the DAC feedback resistors are
within limits programmed Into the PM-7528 data latches. The

~~~~~:~;

<>-.,....---------,

VDD
RFBA

3

'7
1kn

DATA INPUTS

PASS/FJ!JC
OUTPUT

1 AGNO

C1-+--+---'~

WI'!_+_I-.....:.:'8+J
cro:A/DAC B-+-I---'+J

19 RFe B

FIGURE 8: Digitally Programmable Window Comparator (Upper and Lower Limit Detector).
MICROPROCESSOR INTERFACE

I

DACA}

CPU
6800

ADDRESS BUS

A8-A1S

ADDRESS BUS

AO-A15

IDACA}

CPU
8085

PM-752S·

PM-762S"

IOAe B)

8

ALE

"A

;;r

ADDR/DATA BUS

ADO-AD7

DATA BUS

.. ANALOG CIRCUITRY HAS BEEN OMITTED FOR CLARITY.

·ANALOG CIRCUITRY HAS BEEN OMITTED FOR CLARITY

.. A .. DECODED 7528 ADQR OAC A
A + 1 " DECODED 7528 ADDR OAC B

DECODED 7628 ADDR OAC A

A + , .. DECODED 7528 ADDR OAC B

NOTE

8085 INSTRUCTION SHLD (STORE H 8t L DIRECT) CAN UPDATE
BOTH DACSWITH DATA FROM HAND L REGISTERS.

FIGURE 9: PM-7528 Dual DAC to 6800 CPU Interface.

FIGURE 10: PM-7528 Dual DAC to 8085 CPU Interlace.

11-173

_.- - - - -

--------

1/86, Rev. A

_ _ _ _ _ _ _ _--1~PM-7528 DUAL B-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTER

DIGITALLY CONTROLLED SIGNAL ATTENUATOR

generate logarithmic attenuation, Table 4 was generated based
on the equation:
Digital Input = 256 X exp (-Attenuat ion (dB))
2o

Figure 11 shows the PM-7528 configured as a two-channel
programmable attenuator. Applications include stereo, audio,
and telephone signal-level control applications. In order to

Voo <>-_--"7'"

DBa

PM-7628

DB7

/ ' - - - -____

\ r - - - -.......

1+''''5--------oCll
1+'",6- - - - - - - - O W R
I+:...-..-------og~~

:1

1---r:'S:...-..--------oV rN B

VourB

AGNO

1

DGND

5

FIGURE 11: Digitally-Controlled Dual Telephone Attenuator.
TABLE 4: Attenuation vs. DAC A, DAC B Code for the Circuit of Figure 11
ATTN. dB

DAC INPUT CODE

CODE IN
DECIMAL

ATTN. dB

DAC INPUT CODE

CODE IN
DECIMAL

0

11111111

255

8.0

01100110

102

0.5

11110010

242

8.5

01100000

96

1.0

11100100

228

9.0

01011011

91

1.5

11010111

215

9.5

01010110

86

2.0

11001011

203

10.0

01010001

81

2.5

11000000

192

10.5

01001100

76

3.0

10110101

181

11.0

01001000

72

3.5

10101011

171

11.5

01000100

68

4.0

10100010

162

12.0

01000000

64

4.5

10011000

152

12.5

00111101

61

5.0

10010000

144

13.0

00111001

57

5.5

10001000

136

13.5

00110110

54

6.0

10000000

128

14.0

00110011

51

6.5

01111001

121

14.5

00110000

48

7.0

01110010

114

15.0

00101110

46

7.5

01101100

108

15.5

00101011

43

11-174

1/86, Rev. A

PM-7S33
CMOS LOW COST lO-BIT MULTIPLYING
D/A CONVERTER
PreciSIon MOl1oJ-ithics Inc.

FEATURES

GENERAL DESCRIPTION

•
•
•
•
•
•
•
•
•
•

The PM-7533 IS a 10-bit 4-quadrant multiplying DAC. It is
manufactured using thin film on an oxide-isolated, silicongate, monolithic CMOS wafer fabrication process. PMI's
advanced thin-film resistor processing provides true 10-bit
linearity and excellent long-term stability achieved without
laser trimming.

10-Bit Resolution
Full Four-Quadrant Multiplication
Nonlinearity: 1/2 or 1 LSB
TTL/CMOS Compatible
Improved Gain Error and Linearity Error from +5V to +15V
Low Power Consumption
Low Feedthrough Error
Low Cost
AD7520 and AD7533 Replacement
Full Temperature Operation

APPLICATIONS
•
•
•
•
•
•
•
•
•

Digital/Synchro Conversion
Programmable Gain Amplifiers
Ratiometric A/D Conversion
Function Generator
CRT Graphics Generator
Digitally-Controlled Attenuator
Digitally-Controlled Power Supplies
Digital Filters
Linear Automatic Gain Control

The PM-7533 is pin and function equivalent to the AD7520
and AD7533.
The PMI PM-7533 applications flexibility allows direct interface to TTL or CMOS circuitry and operation from +5V to
+15V power supplies. Output scaling is provided by the internal feedback resistor and an external op amp; both positive
and negative reference voltages can be accommodated.

PIN CONNECTIONS

16-PIN EPOXY DIP
(P-Suffix)
16-PIN HERMETIC DIP
(Q-Suffix)

ORDERING INFORMATIONt
PACKAGE: 16-PIN"

NONLINEARITY
±O 1% (±1 LSB)
±O 05% (±1/2 LSB)

INDUSTRIAL
COMMERCIAL
MILITARY'
TEMPERATURE TEMPERATURE TEMPERATURE
-55'C 10 +125'C -25'C 10 +85'C
D'C 10 +7D'C
PM7533BQ
PM7533AQ

PM7533FQ
PM7533EQ

PM7533HP
PM7533GP

'For deVices processed In total compliance to MIL-STD-883, add 1883 after
part number Consult factory for 883 data sheet
t All commerCial and industrial temperature range parts are available with
burn-In. For ordering information see 1986 Data Book, Section 2
"Package Designation' Suffix Q Hermetic DIP, Suffix P Epoxy DIP

FUNCTIONAL DIAGRAM
10kn

20HZ

10kn

10kn

20kn

2Ok"

20kn

20kn

CROSS REFERENCE
TEMPERATURE
RANGE

PMI

ADI

PM7533AQ
PM7533BQ
PM7533BQ

AD7533UD
AD7533TD
AD7533SD

PM7533EQ
PM7533FQ
PM7533FQ

AD7533CD
AD7533BD
AD7533AD

INDUSTRIAL

PM7533GP
PM7533HP
PM7533HP

AD7533LN
AD7533KN
AD7533JN

COMMERCIAL

L--+-+.......-hi----4.....",>--++-*----Q 'oun
'--'c--.......-+----ct-;--~~-+-;_......-:-:~-o 'OUT1

MILITARY

AFEEDBACK
BIT 1 (MSa)

BIT 2

BIT 3

BIT 10 (LSB)

DIGITAL INPUTS (OTL/TTL/CMOS COMPATIBLE)
(SWITCHES SHOWN FOR DIGITAL INPUTS "HIGH")

11-175

1/86, Rev. A

---------t1EMD

PM·7533 CMOS LOW COST 1C1·BIT MULTIPLYING D/A CONVERTER

ABSOLUTE MAXIMUM RATINGS (TA = +25°C, unless
otherwise noted)
Voo (to GND) . " ........................... -0.3V, +17V
VREF (to GND) .................................... ±25V
RFB (to GND) . " . " .•• " " . " " . " " " " " " " " . ±25V
Digital Input Voltage Range ••••••••••.••.•• -0.3V to Voo
Output Voltage (Pin 1, Pin 2) ••.••••••.••••• -0.3V to Voo
Power Dissipation (Package)
Ceramic (Suffix Q)
To +70°C ••••..••••.•.•.•••••.•..••••.•••.• 450mW
Derates Above +75°C By ••••••••..•••••••• 6mW/oC
Plastic (Suffix P)
To +70°C •...••.•••••••••••••••••••.••••••• 670mW
Derates Above +70°C By ••.•••••.••••••.. 8.3mW/oC
Operating Temperature Range
Military (AQ, BQ Versions) •••••••.••• -55°C to +125°C
Industrial (EQ, FQ Versions) ••.••..•.•. -25°C to +85°C
Commercial (GP, HP Versions) ..••..•..• O°C to +70°C

Dice Junction Temperature .•••••.•••••••••••••• +150°C
Storage Temperature •••.••••••••••••.. -65°C to +150°C
Lead Temperature (Soldering, 60 sec) •..•.•.••.•. +300°C
CAUTION
1. Do not apply voltages higher than Vee or less than GND potential on any
terminal except VAEF (Pin 1S) and RFB (Pin 16)
2. The dIgital control Inputs are zener protected. however, permanent
damage may occur on unconnected Units from high energy electrostatic

fields Keep units in conductive foam at all tImes untIl ready to use
Use proper anti-stallc handling procedures
4. Absolute MaXImum Ratings apply to both packaged deVIces and DICE
Stresses above those hsted under Absolute Maximum Ratings may cause
permanent damage to the deVIce

ELECTRICAL CHARACTERISTICS at VDD = +15V, VREF = +10V, AGND = DGND = OV, VOUT1 = VOUT2 = OV, TA = -55°C
to +125°C apply for PM·7533AQ/BQ, TA = -25°C to +85°C apply for PM-7533EQ/FQ, TA = O°C to +70°C apply for
PM-7533GP/HP, unless otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

PM-7533A/E/G

PM-7533B/F/H

MIN

MIN

TYP

MAX

TVP

MAX

UNITS

STATIC ACCURACY
ResolutIon

10

Relative Accuracy
(Note 1)

10

BIts

±O.OS

±01

%FSR

TA =+2S'C
TA = Full Temp. Range

±1.4
±1.S

±1.4
±1S

%FS

PSRR

TA =+2S'C
TA = Full Temp. Range

O.OOS
0008

O.OOS
0.008

%/%

Output Leakage Current
loun (Pin 1) (Note 6)

ILKG1

TA =+2S'C
TA= Full Temp Range

±SO
±200

±SO
±200

nA

Output Leakage Current
IOUT2 (Pin 2) (Note 7)

I LKG2

TA = +2S'C
TA = Full Temp. Range

±SO
±200

±SO
±200

nA

ts

TA = +2S'C (Note 10)
TA = Full Temp. Range

600
800

600
800

ns

TA =+2S'C
TA = Full Temp. Range

±O.OS
±0.1

±O.OS
±01

%FSR

20

kO

Gain Error

(Notes 2, 3)
Power Supply Rejection

~

1.5
0.5

--

~
~

§ -0.'
~

.......
1.0

0.'

"

-1.0
_10~_~_~_~_-L_-L_~

o

±2

±a

±4

±S

±10

-1.6

258

0

±12

0

1024

768

10

0

VREP (VOLTS)

DIGITAL CODe (DECIMAL)

Voo (VOL TS)

GAIN ERROR vs VREF

RELATIVE ACCURACY
(NONLINEARITY) VI
DIGITAL CODE

LINEARITY ERROR va
SUPPLY VOLTAGE

1.0...--..,...--,--..,...-.....---,-.....,
TA .. +2SoC

16

2.0

1.5

TA " +25°C

Veo" +15V

Veo • +6V TO +16V

1.0

DIGITAL INPUT-,,11,,1111

~>
:i

V REF • +lav

-- --

0.6

=
u
~

:iw

>
~

-0.5

512

~

\--t--+--+--+--t---i

-0.5

=

1.5

1.0

\

0.5

-1.0

-100~-±~2-~±4-~±6---±~8--±~10~-±~'2

-1.5

o

512

256

VAEF (VOLTS)

00

1024

768

10

DIGITAl. CODE (DECIMAL)

LOGIC INPUT THRESHOLD
VOLTAGE va
SUPPLY VOLTAGE (VDD)

VREF FREQUENCY
RESPONSE
12

TA ' +~'Jl~
Veo" +15V

15

Voo (VOLTS)

16

""'

TA = +25~C
V REF =+lOV

VREF" 10V PEAK AC
DIGITAL INPUT = 1111",,11
AI. = 100n

12

-3

-6

o

-12
0001

001

0.1

10

10

o

V,NH '"

2.4Y"

V'NL •

0.8~"""""'... ...
10

15

20

Veo (VOLTS)

FREQUENCY (MHz)

11-180

1/86, Rev. A

_ _ _ _ _ _ _ _--fIfMDPM-7533 CMOS LOW COST 10-BIT MULTIPLYING D/A CONVERTER

DEFINITIONS
RESOLUTION
The resolution of a DAC is the number of states (2") that the
full-scale range (FSR) is divided (or resolved) into, where n is
equal to the number of bits. Resolution in no way implies
linearity.
RELATIVE ACCURACY
Relative accuracy or end-point (nonlinearity) is a measure of
the maximum deviation from a straight line passing through
the end-points of the DAC transfer function. It is measured
after adjusting for ideal zero and full-scale and is expressed
in % or ppm of full-scale range or (sub) multiples of 1 LSB.

controlled by CMOS inverters. Most applications require the
addition of only an operational amplifier and a current or
voltage reference.
An inverted R-2R ladder network in a simplified D/A converter circuit is shown in Figure 1. The current through each
ladder leg is switched between loun and IOUT2 under the
control of the digital inputs. This allows a constant current to
be maintained in each ladder leg regardless of the digitalinput switch states.

10kO

ZOkn

SETTLING TIME
Time required for the output function of the DAC to settle to
within 1/2 LSB for a given digital input stimulus, i.e., zero to
full scale.

OUTPUT LEAKAGE CURRENT
Current which appears on IOUT1 terminal with all digital
inputs low or on IOUT2 terminal when all inputs are high.

CIRCUIT DESCRIPTION
The PM-7533 isa 10-bit multiplying D/A converter. It consists
of a silicon-chrome thin-film R-2R resistor ladder network
and ten pairs of NMOS current steering switches, all on a
monolithic chip. The NMOS current steering switches are

20kO

20kO

20kO

'--++-......-++-.......'<'-++-......- - 0 0 'OUT2
L.....f--.......;--~:-"'*-~-...-:::-::-<> 'OUT1
RFEEDBACK

BIT 1 (MSS)

GAIN ERROR
Gain error or full-scale error is a measure of the output error
between an ideal DAC and the actual device output.

OUTPUT CAPACITANCE
Capacitance from loun and IOUT2 terminals to ground.

20kn

810

GAIN
Ratio of the DAC's external operational amplifier output
voltage to the VREF input voltage when using the DAC's
internal feedback resistor.

FEEDTHROUGH ERROR
Error caused by capacitive coupling from VREFtO output with
all switches off.

10kO

10kSl

BIT 2

BIT 3

BIT 10 (LSB)

DIGITAL. INPUTS (OTl/TTL/CMOS COMPATIBLE)
(SWITCHES SHOWN FOR DIGITAL INPUTS "HIGH'"

FIGURE 1: SIMPLIFIED DAC CIRCUIT

Figure 2 shows one of ten digital input CMOS inverters
driving an NMOS switch. The size of devices 1, 2, and 3 are
optimized to make the digital inputs DTL/TTL/CMOS compatible over the full military temperature range. The input
stage drives the two inverters (4, 5) and (6, 7), which drives
the two NMOS switches (8 and 9). The switch "ON" resistances are binarily-scaled so that the voltage drop across
each switch is the same; that is, switch S1 in Figure 1 (8 and 9
of Figure 2) was designed foran "ON" resistance of 20 ohms,
switch S2 for40 ohms, etc. With a 10V reference input, switch
S1 current is 0.5mA, switch S2 is 0.25mA, etc. This will
maintain a constant 10mV drop across each switch. It is
essential that each switch voltage drop be equal so that the
D/A converter accuracy is maintained.

v+------...----~--...--

t TO R-2R RESISTOR LADOER

IOUT2

IOUT1

FIGURE 2: CMOS SWITCH

11-181

1/86, Rev_ A

I

----------l~

PM-7533 CMOS LOW COST 10-BIT MULTIPLYING D/A CONVERTER

EQUIVALENT CIRCUIT ANALYSIS
Figures 3 and 4 show equivalent circuits of the DAC with all
digital inputs high and low respectively. With all digital inputs
in the high state as shown in Figure 3, the reference current is
switched to the IOUT1 terminal, and the IOUT2 terminal is
open-circuited. Only the output capacitance, surface,
leakages, and junction leakages appear at the I OUT2 terminal.
The 1/1024 current source is a constant 1-bit current drain
through the termination resistor of the R-2R ladder network.
The ILEAKAGE current source represents a combination of
surface and junction leakages to the substrate. The "ON"
capacitance of the output NMOS switch is higher on the
IOUT1 terminal when all digital inputs are high (MOS transistor gate capacitance increases with applied gate voltage).

RFEEDBACK
""'10kn

o-_N'-_---.......p_---_-.....---<>'oun

V REF

r----_-----IY-_---.......
V REF

DB9-DBO
NOTES
1 R1 AND R2 USED ONLY IF GAIN ADJUSTMENT IS REQUIRED
2 C1 MAY BE REQUIRED WHEN USING HIGH SPEED OP AMPS

FIGURE 4: EQUIVALENT DAC CIRCUIT
(All digital Inputs LOW).

FIGURE 5: UNIPOLAR BINARY OPERATION
(2-QUADRANT MULTIPLICATION)

11-182

1/86, Rev. A

-----------l1fHD

PM-7533 CMOS LOW COST 10-BIT MULTIPLYING D/A CONVERTER

are set to 1000000000 and adjusting R1 for a zero output
voltage (less than 10% of 1 LS8). Resistors R3, R4 and R5
must be selected for matching and tracking in order to keep
offset and full scale errors to a minimum. Resistors R1 and R2
temperature coefficients must be taken into account if they
are used C1 phase compensation capacitor may not be
needed and should be selected empirically. The digital Input
code versus analog output voltage is shown in Table II.

TABLE I: UNIPOLAR BINARY CODE TABLE
DIGITAL INPUT
MSB
LSB

NOMINAL ANALOG OUTPUT
(VOUT as shown in Figure 5)
( 1023)
1024

o

0 0 0 0 0 0 0

( 513 )
1024

o

0 0 0 0 0 0 0 0

( 512 )
1024

TABLE II: BIPOLAR (OFFSET BINARY) CODE TABLE
DIGITAL INPUT
MSB
LSB

( 511 )
1024

o

NOMINAL ANALOG OUTPUT
(VOUT as shown in Figure 6)

(~)
512

1 1 1

o

0 0 0 0 0 0 0 0

-VREF

(_1_)

o

0 0 0 0 0 0 0 0 0

-VREF

(~)
1024

1024

o

(5~2

0 0 0 0 0 0 0

o
o

00000 0 0 0 0
NOTES:
1 Nominal full scale for the CirCUit of Figure 5 IS given by
1023 )
FS ~ ~VAEF ( 1024
2 Nominal LSB magnitude for the Circuit of Figure 5

LSB

IS

o

Figure 6 shows a simple bipolar output circuit using the
PM-7533 and a PMI OP-215 dual op amp. The circuit uses
offset binary coding and a fixed DC voltage for VREF.
Digitally-controlled attenuation of an AC signal occurs when
the signal is used as the signal source at V REF . Negative
output full-scale is adj usted by setting the digital inputs to all
zeros and adjusting the value of the VREF voltage or R5. The
zero-scale output voltage is adjusted while the digital inputs

15

5~2 )

000 0 0 000

(~)
512

000 0 0 0 000 0

(~)
512

o

VIN

(

given by

~ VAEF (1~24) or VAEF (2-°)

)

NOTES:
1 Nominal full scale for the Circuit of Figure 6 IS given by

FSR

~ VAEF (~~;)

2 Nominal LSB magnitude for the

LSB

CircUit

of Figure 6

IS

given by

~ VAEF (5~2)

III

V REF

~~n

V OUT

DATA INPUT

FIGURE 6: BIPOLAR OPERATION (4-QUADRANT MULTIPLICATION)

11-183

1/86, Rev. A

---------I~ PM-7533 CMOS LOW COST 10-BIT MULTIPLYING D/A CONVERTER

The PM-7533 may be used in the voltage output operation as
shown in Figure 7. This circuit configuration will lend itself to
single-supply operation because signal inversion does not
occur. The output should be buffered due to its high output
resistance (10kfl) to prevent loading errors. The reference
voltage should be kept to +1.5 volts maximum to keep
nonlinearity errors to less than 1 LSB as shown in Figure 8.

TA = +2SoC
Vao

By connecting the DAC in the feedback of an op amp as
shown in Figure 9, the circuit behaves as a programmable
gain amplifier (analog/digital divider). The transfer function is:
-1

Vo= (~+~V: . . ~ )
2'

22

2'0

-2

=+5V

~~
o

where A, ... AlO assumes a value of 1 or O.

256

512

768

1024

DIGITAL CODe (DECIMAL)

FIGURE 8: VOLTAGE MODE
+5V

V,N

0-------.,

DIGITAL INPUT {OJ

(±10mV TO ±lOV

RANGE
DEPENDING ON
GAIN SETTING)

PM-7533

NOTES

>----.....-oVOUT =-VIN X 0210

1 V REF ";;;+1 5V TO KEEP NONLINEARITY ERRORS < 1 LSB
2 ROUT"" 10kn THIS HIGH OUTPUT RESISTANCE SHOULD
BE BUFFERED TO PREVENT LOADING ERRORS
3 THIS CIRCUIT WILL LEND ITSELF TO SINGLE SUPPLY
OPERATION

(±lOV RANGE)

o

FIGURE 7: VOLTAGE OUTPUT OPERATION

1023
512

256

GAIN TABLE
AV (VourIVIN)

-100097
-2

-4

-512
-1024
OPEN LOOP

FIGURE 9: PROGRAMMABLE GAIN AMPLIFIER

11-184

1/86, Rev. A

PM-7541
CMOS 12-BIT MONOLITHIC MULTIPLYING
D/A CONVERTER
Precision MOl1olithics Inc.

FEATURES
•
•
•
•
•
•
•
•
•
•

CROSS REFERENCE

Full Four-Quadrant Multiplication
12-Bit Endpoint Linearity (±1/2 LSB)
Pretrimmed Gain
TTL/CMOS Compatible
Low Power Consumption
Low Feedthrough Error
Direct Replacement for AD7521 and AD7541
Superior Power Supply Rejection from +5V to +15V
Low Gain and Linearity Tempcos (TYP 2ppm of FSR/°C)
Latch-Up Resistant

PMI

TEMPERATURE
RANGE

ADI

PM7541AX
PM7541BX

AD7541TD
AD7541SD

MILITARY

PM7541EX
PM7541FX

AD7541BD
AD7541AD

INDUSTRIAL

PM7541GP
PM7541HP

AD7541KN
AD7541JN

COMMERCIAL

GENERAL DESCRIPTION
The PMI PM-7541 is a 12-bit, 4-quadrant multiplying digital-toanalog converter. It is manufactured using an advanced oxideisolated, silicon-gate, monolithic CMOS technology.

APPLICATIONS
•
•
•
•
•
•
•
•

Digltal/Synchro Conversion
Programmable Amplifiers
Ratiometric AID Conversion
Function Generator
CRT Graphics Generator
Digitally-Controlled Attenuator
Digitally-Controlled Power Supplies
Digital Filters

Laser-trimmed thin-film resistors on CMOS circuitry provide
true 12-bit linearity and excellent absolute accuracy. The low
power dissipation, together with NMOS temperature-compensating switches, assures the performance over the full temperature range. It is a pin-compatible replacement for Analog
Devices AD7521 and AD7541 with equal or better performance.

PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE: 18-PIW·

NONLINEARITY
1LSB
112 LSB

MILITARY·
INDUSTRIAL
COMMERCIAL
TEMPERATURE TEMPERATURE TEMPERATURE
-SS·C TO +12S·C -2S·C TO +8S·C O·C TO +70·C
PM7541BX
PM7541AX

PM7541FX
PM7541EX

18-PIN EPOXY DIP
(P-Suffix)

PM7541HP
PM7541GP

• For devices processed In total compliance to MIL-STD-883, add 1883 after
part number. Consult factory for 883 data sheet.
(also available in Side Braze-XB)
t All commerCial and industrial temperature range parts are available with
burn-in For ordering information see 1986 Data Book, Section 2.
** Package Designation"
Suffix X Hermetic DIP (XB - Side Braze)
Suffix P: Epoxy DIP

18-PIN HERMETIC DIP
(X-Suffix)

FUNCTIONAL DIAGRAM

20kn

20kn

20kS"!

20kn

20kH

20kn

SPOT

NMOS
SWITCHES

'--+-+.......--f-f-4---r,'--++......-H------ 'OUT!
RFEEDBACK
BIT 1 (MSB)

BIT 2

81T3

BIT 12 (LSB)

DIGITAL INPUTS (OTL/TTL/CMOS COMPATIBLE)
(SWITCHES SHOWN FOR DIGITAL INPUTS "HIGH")

11-190

1/86, Rev. A

----------I!EHD

PM-7541 CMOS 12-BIT MONOLITHIC MULTIPLYING D/A CONVERTER

FIGURE 2: CMOS SWITCH
v+------~----~--~--

+TO R-2R RESISTOR LADDER

IOUT2

One of the twelve CMOS switches is shown in Figure 2. The
digital input stage, devices 1, 2, and 3, drives the two inverters,
devices 4,5,6, and 7; these inverters in turn drive the two output
current steering switches, devices 8 and 9. Devices 1, 2, and 3
are designed such that the digital control inputs are DTL, TTL,
and CMOS compatible overthe full military temperature range.

loun

FIGURE 4: PM-7541 EQUIVALENT CIRCUIT
(ALL DIGITAL INPUTS HIGH)
RFEEDBACK
R = 10kn

The twelve output current-steering switches are in series with
the R-2R resistor ladder, and therefore, can introduce bit errors.
It is essential then, that the switch "ON" resistance be binarily
scaled so that the voltage drop across each switch remains
constant. If, for example, switch 1 of Figure 1 was designed with
an "ON" resistance of 10 ohms, switch 2 for 20 ohms, etc., then
with a 10 volt reference Input, the current through sWitch 1 IS
0.5mA, switch 2 is 0.25mA, etc., a constant 5mV drop will then
be maintained across each switch.

O--¥W~----.......- - - -.....-

.....---OIOUT1

V REF

, - - - - -. . . .- - - - _ Q 10UT2

EQUIVALENT CIRCUIT ANALYSIS
Figures 3 and 4 show the equivalent circUits for all digital Inputs
LOW and HIGH respectively. The reference current IS switched
to IOUT2 when all inputs are LOW and IOUT1 when Inputs are
HIGH. The ILEAKAGE current source IS the combination of
surface and Junction leakages to the substrate, the 1/4096
current source represents the constant 1-bit current drain
through the ladder terminating resistor. The output capacitance
is dependent upon the digital Input code, and IS therefore
modulated between the low and high values.

DYNAMIC PERFORMANCE
OUTPUT IMPEDANCE
The output resistance, as In the case of the output capacitance,
IS also modulated by the digital input code. The resistance
looking back Into the IOUT1 terminal, may be anywhere
between 10kfl (the feedback resistor alone when all digital
Inputs are low) and 7.5kfl (the feedback resistor In parallel with
approximately 30kfl of the R-2R ladder network resistance
when any single bit logiC IS high). The static accuracy and
dynamic performance will be affected by thiS modulation. The
gain and phase stability of the output amplifier, board layout,
and power supply decoupllng will all affect the dynamiC
performance of the PM-7541. The use of a compensation
capacitor may be reqUired when high-speed operational
amplifiers are used It may be connected across the amplifiers
feedback resistor to provide the necessary phase compensation
to Critically damp the output.

FIGURE 3: PM-7541 EQUIVALENT CIRCUIT
(ALL INPUTS LOW)
RFEEDBACK

R = 10kn

.....- - Q 'oun

r----~--

II

The considerations when using high-speed amplifiers are:
1. Phase Compensation (See Figures 5 and 6).
2. Power supply decoupling at the device socket and use of
proper grounding techniques.

c>--A/'o/I.--..----....- - - -.....- - - - - o I O U T 2

1/86, Rev. A

11-191

~~~~

-----

-----~-

----------

-----------l1fH!) PM-7541 CMOS 12-BIT MONOLITHIC MULTIPLYING D/A CONVERTER
APPLICATIONS INFORMATION

Unused digital inputs must always be grounded or taken to
Voo; this will prevent noise from triggering the high impedance
digital input resulting in output errors. It is also recommended
that the used digital inputs be taken to ground or Voo via a high
value (1 MO) resistor; this will prevent the accumulation of static
charge whenever the PC card is disconnected from the system.

APPLICATION TIPS
Linearity depends upon the potential of loun and IOUT2 (pins 1
and 2) being exactly equal to GND (pin 3). In most applications,
the DAC is connected to an external op amp with its non inverting input tied to ground, see Figures 5 and 6. The amplifier
selected should have a low input bias current and low drift over
temperature. The amplifier's Input offset voltage should be
nulled to less than ±200/lV (less than 10% of 1 LS8).

OUTPUT AMPLIFIER CONSIDERATIONS
For low speed or static applications, AC specifications of the
amplifier are not very critical. In high-speed applications, slew
rate, settling time, open-loop gain, and gain/phase margin
specifications of the amplifier should be selected forthe desired
performance. It has already been pOinted out that an offset can
be caused by including the usual bias current compensation
resistor in the amplifier's non inverting input-terminal. This
resistor should not be used. Instead, the amplifier should have a
bias current which is low overthe temperature range of interest.

The operational amplifiers usual bias current compensation
resistor in the non inverting input should not be used, the input
should be connected directly to ground with a low-resistance
wire. This resistor can cause a variable offset voltage contributing an error. All pins going to ground should be taken to a
common point to avoid ground loops. The Voo power supply
should have a low noise level and not have transients greater
than +17V.

FIGURE 5: UNIPOLAR BINARY OPERATION (2-QUADRANT)

-15V

FIGURE 6: UNIPOLAR BINARY OPERATION (2-QUADRANT)
R1
VREF

-lOY
BIT 1
(MSS)

r-*".....,I::-.,A FEEDBACK

1--"""'...---,-------,
R2
l1d1

'oun

BIT 12
(LSBj
-15V

11-192

1/86, Rev. A

-----------l1fMD

PM-7541 CMOS 12-BIT MONOLITHIC MULTIPLYING D/A CONVERTER

The static accuracy is affected by the variation in the DAC's
output resistance. This variation is best illustrated by using the
circuit of Figure 8 and the equation:
Error Voltage = Vos (1

+ :~B)

where Ro = function of digital code.
Ro "" 1OkO for more than 4-bits of logic 1.
Ro "" 30kO for any single bit logic 1.
Therefore, the offset gain varies as follows:
At code 001111111111: VERROR

1

=Vos

( 1 + 10kO)
10kO

UNIPOLAR BINARY OPERATION (2-QUADRANT)
The circuits of Figures 5 and 6 can either be used as a fixed
reference D/A converter, or as an attenuator with an AC input
voltage. I n the fixed reference mode, the DAC provides an
analog output voltage in the range of zero to plus or minus VREF ,
depending on VREF polarity. The reference input voltage can
range between -20V to +20V; this is due to the ability of VREF
being able to exceed VDD , the limiting factor being the op amp
voltage range. Table 1 shows the code relationship for the
circuit of Figure 6. R1 can be omitted with a resulting maximum
gain error of 0.3% of full scale.

=2 Vos
Table 1: Code Table for Circuit of Figure 6

.
_
(
10kO) _ 4 \,(
At code 010000000000. VERROR 2 - Vos 1 + 30kO -:3 os

DIGITAL INPUT
111111111111
100000000000
011111111111
000000000000

The error difference is 2/3 Vos.
Since one LSB has a weight (for VREF = +10V) of 2.5mV for the
PM-7541 DAC, it is clearly important that Vos be nulled, either
using the amplifier's nulling pins or an external network.
APPLICATIONS
Figures 5, 6, and 7 show simple unipolar and bipolar circuits
with their associated waveforms using the PM-7541 and two
PMI types of output amplifiers. A small feedback capacitor
should be used across the amplifier to help prevent overshoot
and ringing when using high-speed op amps. Resistor R1 is
used totrimforfull scale, lowtempco (approximately 50ppm/°C)
resistors or trim pots should be selected when gain adjustments
are required.

o

BIPOLAR BINARY OPERATION (FOUR-QUADRANT)
The recommended circuit and code relationship is shown in
Figure 7 and Table 2. The digital Input is offset binary coded and
multiplies VREF per Table 2. Resistors R3 and R4 should beequal
within 0.1% at all temperatures, but need not track the resistors
within the PM-7541. The network comprised of R5, R6, and R7
sums 1/2 LSB of current into IOUT2 to ensure correct coding at
zero. R1 can be adjusted to produce the outputs shown in Table
2. However, when the application permits it, R1 and R2 should
be omitted. The maximum gain error in this condition is 0.3% of
full scale. R5 may be replaced by a 1000 fixed resistor; the
maximum zero error is then 0.015% of full scale. The input offset
voltage of both amplifiers should be adjusted to less than 0.1 mV
and be better than 0.5mV over the temperature range of interest. With VREF set to 10V, R5 is adjusted so that with code
100000000000, VO UT = OV ±0.2mV. R1 is adjusted so that code
000000000000 causes VOUT to equal VREF .

FIGURE 7: BIPOLAR OPERATION (4-QUADRANT)
R1

VREF o--'W~~------------

R2

BIT 1

NOMINAL ANALOG OUTPUT
-0.99975 VREF
-0.50000 VREF
-0.49975 VREF

(MSB)

Table 2: Code Table for Circuit of Figure 7
DIGITAL INPUT
111111111111
100000000001
100000000000
010000000000
000000000000

BIT 12
(LSB)

11-193

..

_ ..._ - - - -

._-- . _ - - -

NOMINAL ANALOG OUTPUT
-0.99951 VREF
-0.00049 VREF

o
+0.50000 VREF
+1.00000 VREF

1/86, Rev. A

--_

... _ - - - -

a

----------l~ PM-7541 CMOS 12-BIT MONOLITHIC MULTIPLYING D/A CONVERTER

OFFSET ADJUSTMENT
1. Adjust VREF to approximately +10V.
2. Set R5 to zero.
3. Connect all digital inputs to "Logic 1".
4. Adjust IOUT2 amplifieroffsettrimpot forOV ±0.1 mV at IOUT2
amplifier output.
5. Connect a short circuit across R4.
6. Connect all digital inputs to "Logic 0".
7. Adjust IOUT2 amplifier offset trimpot for OV ±0.1 mV at IOUT1
amplifier output.
S. Remove short circuit across R4.
9. Connect MSB (Bit-1) to "Logic 1" and all other bits to
"Logic 0".
10. Adjust R5 for OV ±0.2mV at VOUT'
GAIN ADJUSTMENT
1. Connect all digital inputs to Voo.
2. Monitor VOUT for -VREF ( 1 - ;,,) volts reading while
adjusting Rt
FIGURE 8: SIMPLIFIED CIRCUIT

ANALOG/DIGITAL DIVISION
The transfer function for the PM-7541 connected in the multiplying mode as shown in Figure 6 is:
A1 A2 A3
A12)
Vo = -VIN ( ~ + 2> + 2> + ........ ~
where Ax assume a value of 1 for an "ON" bit and 0 for an
"OFF" bit.
The transfer function is modified when the DAC is connected in
the feedback of an operational amplifier as shown in Figure 9,
it now is:

The above transfer function is the division of an analog
voltage (VREF ) by a digital word. The amplifier goes to the rails
with all bits "OFF" since division by zero is infinity. With all bits
"ON", the gain is 1 (±1 LSB). The gain becomes 4096 with the
LSB, bit 12, "ON".

FIGURE 9: ANALOG/DIGITAL DIVIDER
BIT 12
(LSB)

o-_-",'VY_-"M""""'_'VVv---

ETC

+15V

"FB
PM-7541

"
Your

11-194

1/86, Rev. A

PM-7542
12-BIT (4-BIT BYTE INPUT)
MULTIPLYING CMOS D/A CONVERTER
Precision MOl1olithics Inc.

ADVANCEPRODUCTINFO~TION
FEATURES
± 1/2 LSB Nonlinearity Over Temperature

•
•
•
•
•
•
•
•
•

GENERAL DESCRIPTION
The PM-7542 is a 12-bit monolithic, CMOS digital-to-analog
converter. It consists of three 4-bit data registers, a 12-bit DAC
register, a 12-bit multiplying DAC, and address decoding.

Low Gain Tempco ....•............... 5ppm/oC Max
Microprocessor Compatible
Four Quadrant Multiplication
Low AC Feedthrough
Low Power Dissipation ...................... 40mW
LowCost
Small 16-Pin 0.3" DIP Package
Latch-Up Resistant

Data is accepted in three 4-bit bytes that facilitate direct interfacing to 4 or a-bit microprocessors. Data loading is similar to a
static RAM's write cycle. A CLR pin is provided to reset the DAC's
internal registers to zeros.

CROSS REFERENCE
APPLICATIONS
•
•
•

Industrial Automation
Process Controls
Instrumentation Equipment

ORDERING INFORMATIONt
PACKAGE: 16-PIN DIP"
GAIN
ERROR

MILITARY'
TEMPERATURE

INDUSTRIAL
TEMPERATURE

COMMERCIAL
TEMPERATURE

±1 LSB
±6LSB

PM7542AO
PM7542BO

PM7542EO
PM7542FO

PM7542GP
PM7542HP

• For devices processed in total compliance to MIL-STD-883, add /883 after part
number. Consult factory for 883 data sheet.
"Package Designation: Suffix Q: Hermetic DIP; SuffiX P: Epoxy DIP.
t All commercial and Industrial temperature range parts are available with burn-In.
For ordering Information see 1986 Data Book, Section 2.

PMI

ADI

TEMPERATURE
RANGE

PM7542AO
PM7542BO
PM7542BO

AD7542GTD
AD7542TD
AD7542SD

MILITARY

PM7542EO
PM7542FO
PM7542FO

AD7542GBD
AD7542BD
AD7542AD

INDUSTRIAL

PM7542GP
PM7542HP
PM7542HP

AD7542GKN
AD7542KN
AD7542JN

COMMERCIAL

FUNCTIONAL DIAGRAM

R"
OUT1
OUT2

PIN CONNECTIONS

AGND

V,,
DGND

16-PIN HERMETIC DIP
(Q-Suffix)

.0

t==~~~~DO(LSBI

16-PIN EPOXY DIP
(P-Suffix)

D1

D'

D3(MSB)

This advance producllnlormation describes a product in development al the lime 01 this prinling. Final specifications may vary. Please contact
local sales office or distributor lor final data sheet
11-195

1/86, Rev. A

III

PM-7543
12-BIT SERIAL-INPUT
MULTIPLYING CMOS D/A CONVERTER
Precision MOllolithics Inc.

ADVANCEPRODUCTINFO~TION
FEATURES
•
•
•
•
•
•
•
•
•
•
•
•

CROSS REFERENCE

± 1/2 LSB Nonlinearity Over Temp
Low Gain Tempco .•....•...•....••... 5ppm/oC Max
Serial Load on Positive or Negative Strobe
Asynchronous CLEAR Input for Initialization
Low AC Feedthrough
No Schottky Diode Output Protection Required
Four Quadrant Multiplication
Low Power Dissipation
Single Supply Operation
LowCost
Sma Ii 16·Pln 0.3" DIP
Latch·Up Resistant

PACKAGE: 16-PIN DIP"

±1/2LSB
±1/2LSB

±1 LSB
±6LSB

MILITARY'
TEMPERATURE
-SS·C to + 12S·C
PM7543AQ
PM7543BQ

INDUSTRIAL
COMMERCIAL
TEMPERATURE TEMPERATURE
-2S·C to + 8S·C
O·Clo +70·C
PM7543EQ
PM7543FQ

PM7543GP
PM7543HP

• For devices processed in total compliance to MIL-STD-883, add 1883 after part
number. Consult factory for 883 data sheet.
"Package Designation: Suffix Q: Hermetic DIP; Suffix P: Epoxy DIP.

t

ADI

TEMPERATURE
RANGE

PM7543AO
PM7543BO
PM7543BO

AD7543GTD
AD7543TD
AD7543SD

MILITARY

PM7543EO
PM7543FO
PM7543FO

AD7543GBD
AD7543BD
AD7543AD

INDUSTRIAL

PM7543GP
PM7543HP
PM7543HP

AD7543GKN
AD7543KN
AD7543JN

COMMERCIAL

GENERAL DESCRIPTION

ORDERING INFORMATIONt

RELATIVE
GAIN
ACCURACY ERROR

PMI

All commercial and industrial temperature range parts are available with burn-in.

For ordering information see 1986 Data Book, Section 2.

PIN CONNECTIONS

The PM-7543 is a monolithic, serial input, 12-bit CMOS digitalto-analog converter. It is intended for applications where serial
input data is used. Serial input reduces pin count and is, therefore, well-suited where PC board space is at a premium.
The PM-7543 contains a 12-bit serial-in parallel-out input register, a 12-bit shift register, and a 12-bit DAC. A signal at the
strobe pin clocks the serial data input on the leading or trailing
edge of the strobe signal; selection is at the user's discretion.
Data at the serial-in input register is then loaded onto the shift register with the load input controls. A CLR input pin is provided to
asynchronously reset the shift register.
The PM-7543 operates from a single +5V supply and is packaged in a small 16-pin 0.3" wide DIP.

FUNCTIONAL DIAGRAM

VREF

,.

r------......U..!..nO

0-"'+-------1

UT1
OUT 2

t---+"'-OAGND

16·PIN EPOXY DIP
(P·Suffix)
16·PIN HERMETIC DIP
(Q·Suffix)

CLR

13

Li52

9

STB1<>-"+--~r,

SRI

STB4
VDD
STB2

DGND

This advance product information describes a product in development at the time of this printing. Final specifications may vary. Please contact
local sales office or distributor for final data sheet.
1/86, Rev. A
11·196

P~-7545/P~-7645
12-BIT BUFFERED
MULTIPLYING CMOS D/A CONVERTERS
Precision Monolithics Inc

GENERAL DESCRIPTION

FEATURES
•
•
•
•
•

Preadlusted Full Scale •.• ± 1 LSB Maximum Gain Error
Low Gain Temperature Coefficient •••••••••• 2ppm/°C
Small 20-Pln 0.3" Wide DIP
PM-7545 TTL Compatible for VDD = 5V
PM·7645 TTL and 5V CMOS Compatible for VDD = 15V

The PM·7545/PM·7645 are 12·bit CMOS multiplying DACs with
internal data latches. Digital data is input in a 12·bit wide data
format, while CS and WR control inputs are active low. During
this time the latches are transparent allowing digital inputs direct
connection to the DAC. When WR is returned to logic high, the
current data word in the latch is saved.
The PM·7545 operates from 5 to 15 volt power supplies, offering
TTL logic compatibility at VDD of 5V and CMOS logic compat·
ibility at VDD of 15V. The PM·7645 is specified for operation at
VDD of 15V, offering TTL logic input compatibility.

ORDERING INFORMATIONt
PACKAGE: 2D-PIN"
MILITARY""
INDUSTRIAL COMMERCIAL
MAXIMUM TEMPERATURE TEMPERATURE TEMPERATURE
GAIN ERROR
·55°C to
·25°C to
DOC to
+7DOC
+125°C
+85°C
±1 LSB
±3LSB
±1 LSB
±3LSB

PM7545AR
PM7545BR
PM7645AR
PM7645BR

PM7545ER
PM7545FR
PM7645ER
PM7645FR

PIN CONNECTIONS

PM7545GP
PM7545HP
PM7645GP
PM7645HP

2o-PIN EPOXY DIP
(P·Sufflx)

'Package designation:
Suffix R: Hermetic DIP
Suffix P Plastic DIP
"For devices processed in total compliance to MIL-STD-883. add /883 after
part number. Consult factory for 883 data sheet.
tAli commercial and industrial temperature range parts are available with
burn-In. For ordering Information seo 1986 Data Book. Section 2.

2o-PIN HERMETIC DIP
(R·Sufflx)

CROSS REFERENCE
TEMPERATURE
RANGE

PMI

ADI

PM7545AR
PM7545BR
PM7545BR
PM7545BR

AD7545GUD
AD7545UD
AD7545TD
AD7545SD

MILITARY

PM7545ER
PM7545FR
PM7545FR
PM7545FR

AD7545GCQ
AD7545CQ
AD7545BQ
AD7545AQ

INDUSTRIAL

PM7545GP
PM7545HP
PM7545HP
PM7545HP

AD7545GLN
AD7545LN
AD7545KN
AD7545JN

COMMERCIAL

FUNCTIONAL DIAGRAM

PM-7545/PM-7645

r-------------~jl--~-oOUT
VREF 0-+------1

12-BIT
MULTIPLYING OAe

r----+-OAGNO

WRo-+----'r"-....

080-+----1..",

11-197

INPUT DATA LATCHES

DGNO

1/86, Rev. B

--------i~ PM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

ABSOLUTE MAXIMUM RATINGS
(TA = 25°C unless otherwise noted.)
VDD to DGND .•...•..•.•..•....•........••... -0.3V, +17V
Digital Input Voltage to DGND .....•.....•...•.. -0.3V, VDD
AGND to DGND ......•.•••••..•••.•.•....•... -0.3V, VDD
VRFB , VREF to DGND .•..••...•...•. . . . • . . . . . . • . . . .. ±25V
VPIN 1 to DGND ............................... -0.3V, VD D
Power Dissipation (Any Package) to +75°C ...•••.• 450mW
Derates Above +75°C by ..••.•••........•..... 6mW/oC
Operating Temperature Range
Military (AR, BR) Grades .......•....... -55°C to + 125°C
Industrial (ER, FR) Grades ••...•...•..•. -25°C to +85°C
Commercial (Gp, HP) Grades .........•.... O°C to +70°C
Dice Junction Temperature. . . • . . . •• • . . . . . . . . . . . .. +150°C
Storage Temperature ...•...•....•....•.• -65°C to +150°C
Lead Temperature (Soldering, 60 sec) .••.••....... +300°C

CAUTION:
1.

Stresses above those listed under "Absolute Maximum Ratings" may cause
permanent damage to the device This IS a stress rating only and functional
operation at or above this specification is not Implied. Exposure to above
maximum rating conditions for extended periods may affect device reliability.

Do not apply voltages higher than VDD or less than GND potential on any
terminal except VREF

The digital inputs are zener protected, however, permanent damage may
occur on unprotected units from high-energy electrostatic fields. Keep units
in conductive foam at all times until ready to use, Use proper anti-static
handling procedures
Remove power before inserting or removing units from their sockets

ELECTRICAL CHARACTERISTICS atVDD = +5V, VREF = +10V, VOUT= OV, AGND = DGND =OV; TA =-55°C to +125°C apply for
PM-7545AR/BR, TA = -25° C to +85° C apply for PM-7545ER/FR, TA = 0° C to +70° C apply for PM-7545GP/HP, unless otherwise noted.
For 15V operation, see pages 4 and 5.

PARAMETER

SYMBOL

PM-7S4SA/E/G
TYP
MAX
MIN

CONDITIONS

PM-7S4SB/F/H
MIN
TYP
MAX

UNITS

STATIC ACCURACY
12

ResolutIOn

N

Relative
Accuracy

INL

TA = Full Temp. Range

±1/2

±112

LSB

Differential
Nonlinearity

DNL

TA = Full Temp Range
(Note 1)

±1

±1

LSB

G'SE

TA=+25°C
TA = Full Temp Range

±1
±2

±3
±4

LSB

TCG,s

(Note 4)

±S

ppm/'C

PSS

T = +2S'C
A
(:.V = ±S%)
TA = Full Temp Range
DD

0002
0004

0002
0004

%1%

10

10

AlB Versions
E/F/G/H VerSions

200
SO

200
SO

300

300

Gain Error

(Notes 2,3)

12

Bits

Gain Temperature

Coefficient

±2

±S

±2

.aGain/.aTemperature

DC Supply Rejection
:'Gainl:'VDD
Output Leakage
Current at OUT

TA = +2S'C. WR = CS = OV.
All Digital Inputs = OV
IlKG

nA
TA = Full Temp. Range

DYNAMIC PERFORMANCE
Propagation Delay
(Notes 4. S. 6. 7)

tpD

TA = +2S'C
(OUT Load = 100n. CEXT = 13pF)

Current Settling Time

t,

TA = Full Temp Range
(To 1/2 LSB) (Note 4)
lOUT Load = 100n

Digital Charge
Injection

AC Feedthrough
at lOUT

Q

fJS

TA=+25°C'
TA = Full Temp. Range

300
400

VRE ' = AGND (Note 4)
FT

ns

TA = Full Temp Range
VRE ' = ±10V. f = 10kHz
All Digital Inputs = OV

300
400

S

11-198

nVs

mVp~p

1/86, Rev. B

--

--------

---

\fMDPM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

ELECTRICAL CHARACTERISTICS atVDD = +5V, V REF = +10V, VOUT= OV, AGND =

DGND = OV; TA = -55°C to +125°C apply for

PM-7545AR/BR, TA = -25° C to +85° C apply for PM-7545ERlFR, TA = 0° C to +70° C apply for PM-7545GP/HP, unless otherwise noted.
For 15V operation, see pages 4 and 5. (Continued)

PARAMETER

PM-7S4SA/E/G
MAX
MIN
TYP

SYMBOL

CONDITIONS

RREF

TA = Full Temp Range
Input Resistance

COUT

TA = Full Temp Range
DBO-DB11 =OV. WR =CS =OV
DBO-DB11 = Voo ' WR = CS =OV

PM-7S4SB/F/H
MIN
TYP
MAX

UNITS

REFERENCE INPUT
Input Resistance
(Pin 19 to GND)

11

15

11

15

kn

70
150

pF

ANALOG OUTPUTS
Output Capacitance
(Note 4)
COUT

70
150

~

DIGITAL INPUTS
Input High Voltage
Input Low Voltage
Input Current
Input Capacitance
DBO-DB11, WR, CS

V,NH
V 1NL

liN

C'N

TA

= Full Temp

2.4

24

Range

08

08

10

10

TA = +25°C

= Full Temp Range
TA = Full Temp Range
V,N =0 (Note 4)
TA

8

V

pA

pF

SWITCHING CHARACTERISTICS
(Notes 4, 8, 9)

See Timing Diagram

Chip Select to
Write Setup Time

tcs

TA
TA

=+25"C
= Full Temp. Range

Chip Select to
Write Hold Time

tCH

TA

= Full Temp

Range

0

Write Pulse
Width

tWR

TA
TA

=+25"C
=Full Temp

Range

250
380

175
270

250
380

175
270

ns

Data Setup
Time

tos

TA
TA

=+25"C
= Full Temp

Range

140
210

100
150

140
210

100
150

ns

Data Hold Time

tOH

TA

= Full Temp

Range

10

100

TA =Full Temp Range
(All Digital Inputs V,NL or V,NH )

100

TA =+25"C
TA = Full Temp Range
(All Digital Inputs OV or VOO )

280
380

200
270

280
380

200
270

ns

0

ns

ns

10

mA

100
100

5

100
100

pA

NOTES:
2
3
4
5

12~blt monotoniC over full temperature range
Includes the effects of 5ppm max gain TC
USing Internal RFB DAC register loaded with 1111 1111 1111 Gam error IS
adjustable uSing the Circuits of Figures 4 and 5
GUARANTEED and NOT TESTED
From digital Input change to 90% of final analog output

11-199

6
7
8

~

0

U

C)

0

~
0

~

~

C)

......

POWER SUPPLY

Supply Current

tn
~

All digital Inputs =OV to Voo , or Voo to OV
Logic Inputs are MOS gates, tYPical Input current (at +25" C) IS less than
1nA
Sample tested at +250 C to ensure compliance
Chip select CS must be COinCident or present before and/or afterwnte WR,
that IS, les 2: tWR' tCH 2: 0

1/86, Rev. B

0

II

~ PM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

ELECTRICAL CHARACTERISTICS at V DD =+15V, VREF =+10V, V OUT =OV, AGND =DGND =OV; TA =-550 C to +1250 C apply for
PM-7545/PM-7645ARlBR, TA =-250 C to +850 C apply for PM-7545/PM-7645ERlFR, TA =00 C to +700 C apply for PM-7545/PM-7645GPlHp,
unless otherwise noted.
PM-7545A/E/G
PM-7645A/E/G
PARAMETER

SYMBOL

CONOITIONS

MIN

TYP

PM-7545B/F/H
PM-7645B/F/H

MAX

MIN

TYP

MAX

UNITS

STATIC ACCURACY
12

12

Bits

Resolution

N

Relat,ve
Accuracy

INL

TA

Range

±112

±1/2

LSB

Differenllal
Nonhneanty

DNL

TA = Full Temp. Range
(Note 1)

±1

±1

LSB

Gam Error
(Notes 2, 3)

G FSE

TA
TA

±1
±2

±3
±4

LSB

Gain Temperature
Coefficient
.6.Galni.6. Temperature

TCG FS

(Note 4)

±5

ppm/oC

DC Supply ReJeclion
AGa,n/AVoo

PSS

TA
TA

0002
0004

0.002
0.004

%/%

10

10

Output Leakage
Current at OUT

ILKG

=Full Temp

=+25°C
=Full Temp

Range

±2

=+25° C
(AV =±5%)
=Full Temp Range 00
TA =+25°C, WR =CS =OV,
All Dig,tal Inputs =OV
TA = Full Temp Range

±5

±2

nA

AlB Versions
ElF/G/H Vers'ons

200
50

200
50

TA =+25°C
(OUT Load

300

300

DYNAMJC PERFORMANCE
Propagat,on Delay
(Notes 4, 5.6, 7)

tpo

Current Setting Time

ts

TA =Full Temp Range
(To 1/2 LSB) (Note 4)
lOUT Load

D'g,tal Charge
Injection

AC Feedthrough
at lOUT

=1000. C EXT =13pF)

ns

ps

=1000

Q

TA =+25°C
TA = Full Temp Range
VREF =AGND (Note 4)

FT

TA =Full Temp Range
VREF =±10V, f =10kHz
All D'g,tallnputs =OV

RREF

TA =Full Temp Range
Input Res'stance

300
400

300
400

nVs

mVp_p

5

REFERENCE INPUT
Input Resistance
(Pin 19 to GND)

11

11

15

15

kO

70
150

pF

ANALOG OUTPUTS
Output Capacitance
(Note 4)

C OUT

TA = Full Temp Range
DBO-DBll = OV, WR =CS =OV
DBO-DBll = VOD' WR

COUT

70
150

=CS =OV

DIGITAL INPUTS
Input High Voltage
Input Low Voltage

V'NH
V'NL

TA

Input High Voltage
Input Low Voltage

V'NH
V'NL

TA = Full Temp Range. PM-7645

Input Current

I'N

TA = +25°C
TA = Full Temp. Range

Input Capacitance
DBO-DB11, WR. CS

C'N

TA =Full Temp Range
V'N =0 (Note 4)

=Full Temp

Range. PM-7545

135

13.5
15

2.4

24

11-200

1.5

V

V

08

08

10

1
10

IlA

8

8

pF

1/86, Rev. B

----------t1fMD

PM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

ELECTRICAL CHARACTERISTICS at V DD =+15\1, V REF =+10V, VOUT =OV, AGND =DGND =OV; TA =-55 0 C to +1250 C apply for
PM-7545/PM-7645AR/BR, TA =-250 C to +85 0 C apply for PM-7545/PM-7645ER/FR, TA =00 C to +700 C apply for PM-7545/PM-7645G P/Hp,
unless otherwise noted. (Continued)

PARAMETER

PM-7545A/E/G
PM-7645A/E/G
MIN
TYP
MAX

SYMBOL

CONDITIONS

IDD

TA = Full Temp Range
(All Digital Inputs V,NL or V,NH )

'DD

TA = +25°C
TA = Full Temp Range
(All Digital Inputs OV or VDD )

PM-7545B/F/H
PM-7645B/F/H
MIN
TYP
MAX

UNITS

POWER SUPPLY

Supply Current

SWITCHING CHARACTERISTICS
(Notes 4. 8. 9)

See Timing Diagram

Chip Select to

TA=+25°C
TA ;:;; Full Temp Range

Write Setup Time

tcs

mA
100
100

100
100

/lA

PM-7545 A/E/G
180
200

120
150

g

PM-7545 B/F/H
180
200

120
150

ns

teH

Write Pulse
Width

tWR

TA=+25°C
TA =Full Temp Range

160
240

100
170

160
240

100
170

ns

Data Setup
Time

t DS

TA=+25°C
TA = Full Temp Range

90
120

60
80

90
120

60
80

ns

Data Hold Time

tDH

TA

= Full Temp

~

0

Chip Select to
Wnte Hold Time

TA

~

>Ll

Range

ns

U

(j

=Full Temp

Range

SWITCHING CHARACTERISTICS
(Notes 4. 8, 9)

See Timing Diagram

Chip Select to
Wnte Setup Time

tes

TA =+25°C
TA = Full Temp Range

Chip Select to
Write Hold Time

teH

Write Pulse
Width

tWR

Data Setup
Time
Data Hold Time

TA

= Full Temp

10

10
PM-7645 AlE/G

150
210

ns
PM-7645 B/F/H

150
210

Range

ns

Range

150
210

150
210

ns

t DS

TA=+25°C
TA =Full Temp Range

225
300

225
300

ns

tDH

TA = Full Temp Range

10

10

ns

TA

= Full Temp

NOTES:
12-blt monotonic over full temperature range
Includes the effects of 5ppm max gain TC.
USing Internal RFB . DAC register loaded with 1111 11111111 Gain error
adjustable using the cirCUits of Figures 4 and 5.
GUARANTEED and NOT TESTED.
From digital Input change to 90% of final analog output.

IS

11-201

~

0
E:-'

~....

(j
ns

TA =+25°C

0

All digital Inputs =OV to VDD • or VDD to OV.
LogiC Inputs are MOS gates, tYPical mput current (at +25 0 C) IS less than
InA
Sample tested at +25 0 C to ensure compliance
Chip select CS must be cOincident or present before and/or afterwnte WR,
that IS, tcs :;::: tWA' tCH :;::: 0

1/86, Rev_ B

....
0

II

_ _ _ _ _ _ _-I!fMDPM-7545/PM-764512-BIT BUFFERED MULTIPLYING CMOS DIA CONVERTERS

DICE CHARACTERISTICS

1. OUT
2. AGND
3. DGND
4. DB11 (MSB)
5. DB10
6.DB9
7. DB8
8. DB7
9.DB6
10. DB5

11. DB4
12. DB3
13. DB2
14. DB1
15. DBO (LSB)
16. CS
17.WR
18. Voo
19. VREF
20. RFB

For additional DICE inlormation refer to
1986 Data Book, Section 2.

DIE SIZE 0.093 X 0.101 inch, 9.393 sq. mils
(2.36 X 2.57mm, 6.07 sq. mm)

WAFER TEST LIMITS at VDD =+5 or +15\/, VREF =+10V, VOUT =OV, AGND

=DGND =Ov.
PM-7545G/PM-7645G

PARAMETER

SYMBOL

CONDITIONS

Relative Accuracy

INL

Endpoint Linearity Error

Differential Nonlinearity

DNL

Gain Error

GFSE

DAC Latches Loaded with 1111 1111 1111

Output Leakage

ILKG

Input Resistance

LIMIT

UNITS

±1/2

LSB MAX

±1

LSB MAX

±5

LSB MAX

DAC Latches Loaded with 000000000000
Pad 1

±10

nA MAX

RREF

Pad 19

7/15

kO MIN/kO MAX

Digital Input
High

V1NH

V = 5V
OD
PM-7545 only
VOD = 15V

24
135

VMIN

Digital Input
Low

V1Nl

V = 5V
DD
PM-7545 only
VOD = 15V

08
15

V MAX

Digital Input
High

V1NH

VOD = 15V PM-7645 only

24

VMIN

Digital Input
Low

V1NL

VOD = 15V PM-7645 only

08

V MAX

Input Current

'iN

VIN

±1

fJ.AMAX

Supply Current
'

DC Supply Rejection
(aGaln/avDD )

=OV or Voo

All Digital Inputs V1NL or V1NH
DO

PSS

All Digital Inputs OV or VOD
8VOD

=±5%

01
002

mAMAX

%/% MAX

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to vanatlons In assembly methods and normal Yield loss, Yield after packaging IS not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing

11-202

1/86, Rev. B

----------1~ PM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS DIA CONVERTERS

TYPICAL ELECTRICAL CHARACTERISTICS at Voo
unless otherwise noted. (Note 1)

=+5V or +15V, AGND =DGND =OV, VREF =+10V, OUT =OV; TA =25°C,
PM-7545G/PM-7645G

PARAMETER

SYMBOL

Digital Input Capacitance

C'N

CONDITIONS

pF
50

OAC Latches Loaded with 0000 0000 0000

Output Capacitance

pF

COUT

Propagation Delay
(Notes 2, 3, 4)

UNITS

TYPICAL

110

OAC Latches Loaded with 1111 1111 1111

140

Voo = 15V

tpo

Voo = 5V

PM-7545 only

ns

230

NOTES:
1 These charactenstlcs are for design gUidance only and are not subject to test
2 From digital mput change to 90% of fmal analog output.
3. OUT load =100(1, C EXT = 13pF
4 CS = WR = O. OBO to OB11 = OV to Voo or Voo to OV

TYPICAL PERFORMANCE CHARACTERISTICS
FULL-SCALE GAIN ERROR
vs TEMPERATURE

LOGIC THRESHOLD VOLTAGE
vs SUPPLY VOLTAGE

05
04

OUTPUT LEAKAGE CURRENT
vs TEMPERATURE

15

1000
::::VREF=-lOV

Voo'" 15V

03
02
1

./
./

-0 1

J

~

-02

~

-04
-05
~

-

p~ ~

-03

~
_

~

0

•

TEMPERATURE

~

@

~

ill

PM-7645

10

/
1

15

-75

-50

-25

Voo SUPPLY VOLTAGE

(~C)

25

50

75

100

125

TEMPERATURE (OC)

SUPPLY CURRENT
vs TEMPERATURE
PM-7545

SUPPLY CURRENT
vs TEMPERATURE
PM-7645
10

Voo

= 5V

F
V,N

VDD

15V
VIN=24V -

24V

T"'I...
01

v"

/

01
VIN =

001

75

-50

-25

oav

0
25
50
75
TEMPERATURE (0C)

"
100

001
-75 -50

125

OBV..-

V
-25

25

50

75

100

125

TEMPERATURE (OC)

11-203

1/86, Rev. B

II

--------l~ PM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

PARAMETER DEFINITIONS

LOGIC INFORMATION

RELATIVE ACCURACY
Sometimes referred to as endpoint nonlinearity, and is a
measure of the maximum deviation from a straight line passing
through the endpoints of the DAC transfer function. Relative
Accuracy is measured after the zero and full-scale points have
been adjusted, and is normally expressed in LSB or as a
percentage of full scale.

WRITE CYCLE TIMING DIAGRAM

----VDD

DIFFERENTIAL NONLINEARITY
This is the difference between the measured change and the
ideal change between any two adjacent codes. A differential
nonlinearity of ±1 LSB maximum over the full operating temperature range will ensure that a device is monotonic (the
output will increase for an increase in digital code applied).

DATA IN
(DBO-DBll)

------'
MODE SELECTION

WRITE MODE'
CS AND WR LOW, OAe RESPONDS
TO DATA BUS (DBO-DB11) INPUTS.

GAIN ERROR
Gain or full scale error is the amount of output error between the
ideal output and the actual output. The ideal output is VREF
minus 1 LSB. The gain error is adjustable to zero using external
resistance.

HOLD MODE:

EITHER B OR WR HIGH, DATA BUS
(DBO-OB11) IS LOCKED OUT; OAe
HOLDS LAST DATA PRESENT WHEN
WR OR ES ASSUMED HIGH STATE

NOTES:

Voo'" +5V, tr '" tf = 20ns

voo =+15V; tr = tf = 40ns
ALL INPUT SIGNAL RISE AND FALL TIMES MEASURED FROM

~~~I~~ ~~A~~~rBENT REFERENCE LEVEL IS VIH; VIL.

OUTPUT CAPACITANCE
The capacitance from OUT to AGND.

THE PM·7645 USES AN INPUT LOGIC HIGH OF 5 VOLTS

PROPAGATION DELAY
This is measured from the digital input change to the analog
output current reaching 90% of its final value.

D/A CONVERTER SECTION
FIGURE 1: Simplified D/A Circuit of PM-7545

DIGITAL CHARGE INJECTION
This is a measure of the amount of charge injected to the analog
output from the digital inputs, when the digital inputs change
states. It is the area of the glitch and is specified in nVsec; it is
measured with VREF = AGND.
I
I

BURN-IN CIRCUIT

OB11
(MBB)

+10
VREF

PM-7545/PM-7645

~

G
R3

L...!

RFB

~

AGND

VREF

flL

DGND

Voo

~

WR

H

OUT

100kO~ 011 (MSB)
5

~
~
~
~
~

010

cs ~

09

(LSB) DO

08

01

07

02

06

03

05

D4

~"~

Voo

Rl
tkO

5kO

14

E..!3......
J2.......

~

R2
5kO

Voo::: 15V± 5%
VREF = 10V ± 5%

Cl

080
(LBB)

Figure 1 shows a simplified circuit of the D/A Converter section
and Figure2 gives an approximate equivalent switch circuit. R is
typically 11 kQ.
The binary-weighted currents are switched between OUT
and AGND by N-channel switches, thus maintaining a constant
current in each ladder leg independent of the switch state.

R4

1L-

DB9

=C2 =47J1f

C3 = C4 = 001,uf

The capacitance at the OUT terminal, COUT, is code dependent
and varies from 70pF (all switches to AGND) to 150pF (all
switches to OUT). One of the current switches is shown in
Figure 2.
The input resistance at VREF (Figure 1) is always equal to RLDR
(RLDR is the R/2R ladder characteristics resistance and is equal
to value URn). Since the input resistance at the VREF pin is
constant, the reference terminal can be driven by a reference
voltage or a reference current, ac or dc, of positive or negative
polarity. (If a current source is used, a low-temperaturecoefficient external RFB is recommended to define scale factor.)

11-204

1/86, Rev. B

--------Im

PM·7545/PM·7645 12·BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

The internal feedback resistor (R FS ) has a normally closed
switch in series as shown in Figure 1. This switch improves
performance over temperature and power supply rejection;
however when the circuit is not powered up the switch assumes
an open state.

FIGURE 2: N·Channel Current Steering Switch
TO LADDER

FROM~

INTERFACE ~
LOGIC~

not required. Capacitor C1 provides phase compensation and
helps prevent overshoot and ringing when using high speed op
amps. The circuits of Figures 4 and 5 have constant input
impedance at the VREF terminal.
The circuit of Figure 4 can either be used as a fixed reference
D/A converter so that it provides an analog output vOltage in the
rangeOto -VIN (the inversion is introduced by the op amp); orVIN
can be an ac signal in which case the circuit behaves as an
attenuator (2·Quadrant Multiplier). VIN can be any vOltage in the
range -20 :s VIN :s +20 volts (provided the op amp can handle
such voltages) since VREF is permitted to exceed Voo. Table 2
shows the code relationship for the circuit of Figure 4.

....

FIGURE 4: Unipolar Binary Operation
our

AGND

Voo

R2*

DIGITAL SECTION
V OUT

R1'

Figure 3 shows the digital structure for one bit. The digital
signals CONTROL and CONTROL are generated from CS
and WR.

0611-080

*SEE TABLE 1.

FIGURE 3: Digital Input Structure
. . . - - - _ T O AGNO SWITCH

TABLE I: Recommended Trim Resistor Value vs. Grades
DIGITAL

TO OUT SWITCH

INPUT

CONTROL

CONTROL

The input buffers are simple CMOS inverters designed such
that when the PM· 7545 is operated with Voo = 5V, the buffers
convert TTL input levels (2.4V and 0.8V) into CMOS logic levels.
When the digital input is in the region of 1.0volts to 6.0volts, the
input buffers operate in their linear region and draw current
from the power supply. To minimize power supply currents, it is
recommended that the digital input voltages be as close to the
supply rails (Voo and DGND) as is practically possible. The
PM·7545 may be operated with any supply voltage in the range
5:S Voo:S 15 volts. With Voo = +15V, the input logic levels are
CMOS compatible only, i.e., 1.5V and 13.5V. The PM·7645
operates with Voo = 15V only; the buffers convert TTL input
levels (2.4V and 0.8V) into CMOS logic levels.

TRIM
RESISTOR

CR

HP/FR/BR

R1

2000

1000

200

R2

680

330

6.80

GP/ER/AR

TABLE II: Unipolar Binary Code Table for Circuit of Figure 4
BINARY NUMBER IN
DAC REGISTER

ANALOG OUTPUT

1111

1111

1111

}
-VIN . { 4095
4096

1000

0000

0000

2048}
-VIN . { 4096

0000

0000

0001

-VIN • {

0000

0000

0000

-1/2VIN

40~6 }
a Volts

BASIC APPLICATIONS
Figures 4 and 5 show simple unipolar and bipolar circuits using
the PM·7545/PM·7645. Resistor R1 is used to trim for full scale.
The following versions (pM·7545AR, PM·7545ER, PM·7545GP)
have a guaranteed maximum gain error of ±1 LSB at +25°C and
Voo = +5V, and in many applications the gain trim resistors are

11·205

1/86, Rev. B

I

---------I~ PM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

Figure 5 and Table 3 illustrate the recommended circuit and
code relationship for bipolar operation. The D/A function itself
uses offset binary code. The inverter U, on the MSB line,
converts 2's-complement input code to offset binary code. The
inverter U, may be omitted if the inversion is done in software.
R3, R4 and R5 must match within 0.01 % and should be the same
type of resistors (preferably wire-wound or metal foil), so that
their temperature coefficients match. Mismatch of R3 value to
R4 causes both offset and full scale error. Mismatch of R5 to R4
and R3 causes full scale error.
TABLE III: 2'8 Complement Code Table for Circuit of Figure 5
DATA INPUT

ANALOG OUTPUT
{ 2047 }
2048

0111

1111

1111

+V1N '

0000

0000

0001

+VIN . {

0000

0000

0000

1111

1111

1111

-VIN • {

1000

0000

0000

-V1N '

20~8}

APPLICATION HINTS
Output Offset: CMOS D/A converters exhibit a code-dependent
output resistance that causes a code-dependent error voltage
at the output of the amplifier. The maximum amplitude of this
offset, which adds to the D/A converter nonlinearity, is 0.67 Vas
where Vas is the amplifier input-offset voltage. To maintain
monotonic operation, it is recommended that Vas be no greater
than 10% of 1 LSB over the temperature range of operation.
General Ground Management: AC or transient voltages between AGND and DGND can cause noise injection into the
analog output. The simplest method of ensuring that voltages at
AGND and DGND are equal is to tie AGND and DGND
together at the PM-7545/PM-7645. It is recommended that two
diodes (1 N914 or equivalent) be connected in inverse parallel
between AGND and DGND pins in complex systems where
AGND and DGND tie on the backplane.
Digital Glitches: When WR and CS are both low, the latches are
transparent and the D/A converter inputs follow the data inputs.
Some bus systems do not always have data valid for the whole
period during which WR is low. This will allow invalid data to
briefly appear althe DAC inputs during the write cycle. This can
cause unwanted glitches at the DAC output. Retiming the write
pulse WR, so that it only occurs when data is valid, will eliminate
the problem.

a Volts

20~8 }

{ 2048 }
2048

FIGURE 5: Bipolar Operation (2's Complement Code)

11

ANALOG
COMMON

U,
(SEE TEXT)

'2

·SEE TABLE 1

DATA INPUT

11-206

1/86, Rev. B

_ _ _ _ _ _ _---1~PM-7545/PM-7645 12-BIT BUFFERED MULTIPLYING CMOS D/A CONVERTERS

INTERFACING THE PM·7545/PM·7645 TO
MICROPROCESSORS
The PM-7545 can be directly interfaced to either an 8 or 16-bit
microprocessor via its 12-bit wide data latch using the CS and
WR control signals.
An 8-bit processor interface configuration is shown in Figure 6.
It uses two memory addresses, one forthe lower 8-bits and one
for the upper 4-bits of data into the DAC via the latch.
FIGURE 6: 8-Bit Processor to PM-7545/7645 Interface

Connection to an 8-bit processor with a full 16-bit wide address
bus (such as the 6800, 8080, Z80) is shown in Figure 7. The 12
lower address lines are fed directly to the PM-7545; this allows
the PM-7545 to use 4k bytes for its address location. The
address field of the instruction is organized so that the lower
12-bits contain the DAC data. Data is written into the DAC using
a single write instruction.
FIGURE 7: Connecting the PM-7545/7645 to an 8-Bit Microprocessor via the Address Bus
Vl

A15I------------------'\
ADDRESS BUS
AO

ex:
~
~

A15
16-81T ADDRESS BUS

AO

~

DBll

CPU

DBO

CPU

PM-7545!
PM-7645

ViR t:>------l
WI!

C)

WI!

0

~

DATA BUS

8-BIT DATA BUS

DO

DO
~50

U

"CS

07

071----_-'

0

'" DECODED ADDRESS FOR DAC

0

*0, = DECODED ADDRESS FOR LATCH

f-;-<

~......

C)

......
Cl

II

11-207

1/86, Rev. B

PM-7548
12-BIT 18-BIT BYTE INPUT)
MULTIPLYING CMOS D/A CONVERTER
Prl"cisiol1 Monolithics Inc.

ADVANCEPRODUCTINFO~TION
CROSS REFERENCE

FEATURES
•
•
•
•
•
•
•
•
•
•

12-BIt Resolution with an 8-Blt Data Bus
Direct Interface to Most 8-Blt Microprocessors
Selectable Input Data Format (Left or Right-Justified)
TTL/CMOS Compatible
Gain Drift •••.•..•.......••....•...•. 5ppm/oC Max
Guaranteed Monotonic Over Full Temperature Range
Single Supply Operation
Four Quadrant Multiplication
Latch-Up Resistant
Small 20-Pin 0.3" DIP Package

PMI

ADI

TEMPERATURE
RANGE

PM7548AR
PM7548BR

AD7548TD
AD7548SD

MILITARY

PM7548ER
PM7548FR

AD7548BO
AD7548AO

INDUSTRIAL

PM7548GP
PM7548HP

AD7548KN
AD7548JN

COMMERCIAL

PIN CONNECTIONS
APPLICATIONS
•
•
•
•
•

8·Bit Mlcroprocessor·Controlied Systems
Industrial Automation
Process Controls
Servo Control Systems
Programmable Filters and Amplifiers

20-PIN HERMETIC DIP
(R-Sufflx)
20-PIN EPOXY DIP
(P·Sufflx)

ORDERING INFORMATIONt
PACKAGE: 2O·PIN DIP"
RELATIVE
GAIN
ACCURACY ERROR
±1/2LSB
±1 LSB

±3LSB
±6LSB

MILITARY'
TEMPERATURE

INDUSTRIAL
COMMERCIAL
TEMPERATURE TEMPERATURE

-55"<:10 +125"<:

-25"<: 10 + 85"<:

O"C 10 + 7O"C

PM7548AR
PM7548BR

PM7548ER
PM7548FR

PM7548GP
PM7548HP

FUNCTIONAL DIAGRAM

• For deVIces processed in total compliance to MIL-STD-883, add 1883 after part
number. Consult factory lor 883 data sheet.
"Package Daslgnation: Suffix A: Hermetic DIP; Suffix P: Epoxy DIP.
t All commercial and Industrial temperature range perts are available with burn-In.
For ordering inlormation see 1986 Data Book, Section 2

VDD

18

Rn
o"",191--I(~R~-2~R~LA~D~DE:R~F=~~===~ AGND
[~~~E:=F=~Fir==f,5~ CTRL
20

GENERAL DESCRIPTION

""OF

The PM-7548 is a 12-bit monolithic, CMOS digital-to-analog
converter with an 8-bit data input bus. Simple interfacing to 8-bit
microprocessors results when using a two-step data-load operation from the 8-bit data bus. The PM-7548 allows either left or
right-justified data. This simplifies data-formatting when used
with various types of microprocessors.

OUT

DF/I5i5ii

15.() Lilli:
t+--Ir
t+--If!!17.() Wii

In multiple DAC applications, it sometimes becomes necessary to
simultaneously update the outputs. The PM-7548 is ideally
suited for this application, providing a separate LDAC control pin
that immediately updates an analog output from the input
registers.

1-_+1ll!.6o am

DGND

This advance productlnlormatlon describes a product In development at the time 01 thll printing. Final specifications may vary. Please contact
local lalel office or distributor lor final data sheet.

11·208

1/86, Rev. A

J~38510/11301/11302
JAN 8-BIT MULTIPLYING
D/A CONVERTERS
Precision Monolithics Inc.

POWER AND THERMAL CHARACTERISTICS

GENERAL DESCRIPTION
This data sheet covers the electrical requirements of the monolithic 8-bit digital-to-analog converters found in MIL-M-38510/113.
Devices supplied to this data sheet are manufactured and tested
at PMl's MIL-M-38510 certified facility and are listed in
QPL-38510.
Complete device requirements will be found in MIL-M-38510
and MIL-M-38510/113 for Class B processed devices.

Package

case oufllne

Maximum allowable
power dlsalpallon

Mexlmum
BJ - C

Maximum
BJ - A

Dual-in-hne

E

400mW at TA = 125'C

35'C/W

120'C/W

PIN CONNECTIONS & ORDERING INFORMATION

Device Types shall be as follows:
01 DI A Converter, 8 bit, 0.19% linearity
02 D/A Converter, 8 bit, 0.10% linearity

GENERIC CROSS-REFERENCE INFORMATION
This cross-reference information is presented for the convenience of the user. The Generic-Industry types listed may not
have identical operational performance characteristics across
the military temperature range or reliability factors equivalent to
the MIL-M-38510/113 devices.
Military Device Type

Generic-Industry Type

01

DAC-08

02

OAC-08A

Jan Device Type
JM38510/11301BEC
JM38510/11302BEC
JM38510/11301 BEB
JM38510/11302BEB

PMI Device Type
OAC08Q 113851 0
OAC08AQ1/38510
OAC08Q2/38510
OAC08AQ2I38510

Linearity
0.19%
0.10%
0.19%
0.10%

NOTES: Lead finish as follows
BEC" Gold Plate. side braze package
BEB: Tin Plate. CERDIP Package

CASE OUTLINE
Per MIL-M-38510, Appendix C, Case Outline 0-2 (16-Lead 1/4"x
718", dual-in-line). Package type deSignator "E".

II

SIMPLIFIED SCHEMATIC

r--t-~-+~--+~-+~-h--h_-h~_+4:"2-0- lOUT

-TOUT
VREF

(+)

0--'-"4+--.--+_ _,

VREF (_) 0--"'54-_-1"",

16
CQMP

3
-Vee

11-209

1/86, Rev. A

----------l~ JM3a510111301111302 JAN a-BIT MULTIPLYING DlA CONVERTER

ABSOLUTE MAXIMUM RATINGS
Junction Temperature ........................... 175·C
Storage Temperature ••••••••••••••••• -65· C to + 150· C

Supply Voltage [+VCC- (-VCC)] ••••••••••••••••• 36Vdc
Voltage, Digital Input to Negative Supply
[Vlogic- (-Vcc)] ........................... 0 to 36Vdc
Voltage, Logic Control (VLc) .............. -Vccto+Vcc
Reference Voltage Input [(VI4, VIS)] ........ -Vcc to +Vcc
Reference Input Current (114) ....................... 5mA
Reference Input Differential Voltage
[(VlrVls)] ................................... ±18Vdc
Lead Temperature (Soldering, 60 sec) ••••••••••••• 300·C

RECOMMENDED OPERATING CONDITIONS
Supply Voltage Range ............... ±5Vdc to ± 15Vdc"
Ambient Temperature Range .......... -55·C to +125·C
"NOTE:

A slight degradation In linearity can occur when the supply voltage Is near the
±SV end of the recommended operating range.

at ±Vcc = ±15Vdc; Source resistance = 50 ohms; IREF = 2.mA; Figure 1; Ambient
=-55· C to +125· C, unless otherwise noted.

ELECTRICAL CHARACTERISTICS
temperature range

01 LIMITS
PARAMETER

SYMBOL

CONDITIONS

~(I)

Measure 10 , (ION-l oN-,) 2: 0 at

02 LIMITS

MIN

MAX

MIN

MAX

0

16

0

16

0

16

0

16

each major carry point

Monotontclty
~iiT

Measure

UNITS

/loA

iQ. (ION-l oN-,) 2:0at

each major carry pOint
Output Symmetry

~IFS

I FS - I FS

-8

8

-4

4

/loA

Full-Scale Current
Temperature Coefficient

To (lFS)
To (IFS)

All input bits high. Measure 10
All input bits low. Measure

-50

50

-50

50

ppm/DC

I FS

All input bits high.
194

2.04

1984

2

mA

IFS

All Input bits low,

Izs

All Input bits low
Measure 10
All Input bits high,
Measure iO

Full-Scale Current

Zero-Scale Current

~
INL+
Positive Bit Errors
INI.+

Measure iO

Measure

i;;

Measure iO
(I Negative bit errors)/iFS

~INL

Measure 10
IINL+ I-IINL-I

~Ijij[

M~ureiQ
IINL + I-IINL-I

NL+

Measure 10
IINL+ I + I ~INL I
M~ureiQ
IINL+ I + I ~I'jijL I

liil+

-1

/loA

0

019

0

0.10

-019

0

-0.10

0

-005

005

-0.03

003

0

019

0

010

%

0

019

0

010

%

%

(I Positive bit errors)/iFS

INL-

Positive Relative Accuracy

-2

Measure 10
(IPosltlve bit errors)IIFS

Measure 10
(INegatlve bit errors)IIFS

Positive and Negative

Negative Relative Accuracy

Measure Ie

INLNegative Bit Errors

Bit Error Difference

ro

NL-

Measure 10

NI.-

IINL-I + I ~INL I
Measure iQ
IIiiII.-1 + I ~IiilI.l

Bit Error
Bit error is the deviation of the analog output from its Ideal value (after
zero-scale and full-scale errors have been calibrated out) when turning on an
IndiVidual bit. This IS measured for all n bits.
Bit error (analog value) = Vn - (FSRl2 n)
Where Vn = analog output with bit n on only
FSR = full-scale range
n = number of bits

Summation Nonlinearity (INL)

Summation nonlinearity IS the sum of all positive bit errors or all negative bit
errors, whichever Is larger By summing up all the bit errors in one direction,
you obtain the worst possible nonlinearity (I e. if bit 2 lsI LSB high and bit4 is
1/2 LSB high. then bits 2 and 4 together Will be 1 1/2 LSBs high. This IS
essentially the same as Integral nonlinearity since the bit errors are superImposed on each other to give the worst case nonlinearity.

11-210

1/86, Rev. A

~

JM31510111301I11302 JAN I-BIT MULTIPLYING DlA CONVERTER

ELECTRICAL CHARACTERISTICS
temperature range = -55 0

C to + 1250 C.

at ±VCC= ±15Vdc; Source resistance = 50 ohms; IREF= 2.0mA; Figure 1; Ambient
unless otherwise noted.

01 LIMITS
PARAMETER

SYMBOL
IFSR,
IFSR,

Output Current Range
IFSR2
IFSR2
Reference BIas Current

CONDITIONS
All input bits high. Measure 10 •
-Vee =-10V. VREF =lSV
All Input bIts low. Meesure i;;:
-Vee = -10V, VREF = lSV
All input bIts hIgh, Measure 10
-Vee =-12V, VREF = 2SV
All input bits low, Measure i;;
-Vee =-12V, VREF = 2SV

UNITS

2.1
mA

4.2

4.2

0

pA

I'H

All input bits Y,N = 18V, each
Input meesured separately

-0 OS

10

-0 OS

10

,.A

I,L

All input bits Y,N = 10V, each
Input measured separately

-10

All input bits high, Measure la,
V ,O = 18V
All Input bits low, Measure
V,O = 18V

i;;:

1.90

All input bits high, Measure IOo
V,o =-10V
All Input bits low, Meesure iQ.
V,o =-10V

1.90

Low Level Input Current

IFS+
IF8+
IFs IF!>

~IFse

Change In Full Scale Current
Due to Voltage Compliance
~IFSC

PSSIF8+ 1
PSSIFS+l
PSSIFS+2
PssIFS+2
Pss I FS-l

Power Supply SensitivIty
From-Vee

MAX

-3

High Level Input Current

" Power Supply Sensitivity
From+Vee

2.1

02 LIMITS
MIN

0

All Input bits low

Full-Scale Current
At -10V Compliance

MAX

-3

I REF -

Full-Scale Current
At + 18V Compliance

MIN

Pss IFS -l

PSS IFS -2

All input bits high, Measure 10 •
2SoC S TAS 12SoC
TA =-55°C
V'a = 18V to -10V
All Input bits low, Measure i;;'
2SoC S TAS 12SoC
TA=-SSoC
V'a = 18V to -10V
All input bits high. Measure la,
+Vee= 4.SV to +5.SV, -Vee =-18V
All Input bits low, Measure i;;:
+Vee = 4.SV to +S.SV. -Vee = -18V
All input bits high, Measure la,
+Vce = 12V to 18V, -Vee = :l8V
All input bits low, Measure la,
+ Vee = 12Vto 18V, -Vee =-18V
All Input bits high, Measure la,
+Vee = 18V. -Vce = -12V to -18V
All input bits low. Meesure
+ Vee = 18V, -Vee=-12V to -18V
All Input bits high. Measure IOo
+ Vee = 18V, -Vee = -4.5V to -S.5V
IREF= lmA

208

1.94

pA

2.04

mA

2.08

1.94

2.04

mA

~

aU~
C)

~

a
-4
-8

8

-4
-8

4
8

-4
-8

4
8

-4
-8

4
8

-4

4

-4

4

-8

8

-8

8

-8

8

-8

8

-2

2

-2

2

4

~

pA

~

C)

.....
Q

pA

All Input bits low, Measure
+Vee = 18V, -Vee = -4.SV to -S.5V
IREF= lmA

11-211

~

a

ro:

10

Pss IFS-2

-10

en

1/86, Rev. A

---------I~ JM3a510/11301/11302 JAN a·BIT MULTIPLYING D/A CONVERTER

ELECTRICAL CHARACTERISTICS at ±Vcc = ± 15Vdc; Source resistance
temperature range = -55 0 C to +1250 C. unless otherwise noted.

= 50 ohms; IREF = 2.0mA; Figure 1; Ambient

>-__-+________

~K4

NOTES.
1. THE REFERENCE D/A CONVER·
TER SELECTED SHOULD HAVE
A RESOLUTION OF 8 BITS OR
MORE, AND A LINEARITY OF
0.015% OR BETTER.
2, THE VOLTAGE REFERENCE
SHOULD HAVE AN ACCURACY
OF ±O.02% OR BETTER.
3. ALL RELAYS ARE SHOWN IN
THEIR NON·ACTUATED STATE.

Figure 1. Test Circuit For Static Tests
+5V FOR VOUT
OV FOR Vi5UT

VOUT

-["J-

VIN

10V~--- - 90%
tr< 10ns

Y,N
-

-

-

10%

OV

*\.__

OV - - - - - .
\---O.5V
VOUT OR

-2V

~O~T ___il~

1.5V

-.jilt~

Figure 3. Test Circuit For Slew Rate, Device Types 01, 02
11·212

1/86, Rev. A

----------I~ JM3a510/11301111302 JAN a-BIT MULTIPLYING D/A CONVERTER

rVI. (SEE NOTE 11

I

---------.,I
OPTIONAL WITH fET PROBE

I
I
I
I
I
I
I
______ ..JI

I
I

1kll

I
O.7V
1k1l

'----

NOTES:

100k.Q

1

2kn

FOR TURN·ON VI. = 2.7V; FOR TURN-OFF VI. = D.7V

2.

MINIMIZE CAPACITANCE AT THIS NOOE BY USING
SHORT RUNS AND ADEQUATE SPACING BETWEEN

RUNS.
DIODES MUST BE SCHOTIKV TYPE (MBD 501 OR
EQUIVALENT).
BANDWIDTH OF OSCILLOSCOPE. USED FOR WAVEFORM MEASUREMENT SHOULD BE SOMHz MINIMUM; SATURATION OF PREAMP MUST BE AVOIDED.

3.
4.
-16V

_I I _

r-

----J"

:3

tr = tf

_ I
~,f

< 10n5

I

4-

---:~90%~

ov VIN

VaUT

: I

::

::

~

,
I

I
I

I

1

!-'SHL-j

:

±1/2LSB

1f{ff

'I
5°~l
~

~
tPHL

-------+---,'

~-------~

I

" 'PLH

1""":

~----~--I

T,'/2LSB '" 4mV

f.-tSLH--..t

TURN-ON

TURN-OFF

Figure 2_ Test Circuit For Propagation Delay and Settling Time, Device Types 01 and 02

III

BURN-IN
+18V

Devices supplied by PMI have been subjected to burn-in
per method 1015 of MIL-STD-883 using test condition C or
test condition F with the circuit shown in Figure 4.

C2

F
C1

16

R1

15

"

13

12

11

10

OUT.

R1=9kfl±5%

-18V

C1 "O.OOl~F. 50V
C2 = 0 01~F, 50V

Figure 4. Test Circuit, Burn-In and Operating Life Test

11-213

1/86, Rev_ A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

AID

CONVERTERS

· .
PreCISIon

12-3

Introduction

12-3

Definitions

12-7

ADC-910
High-Speed "Microprocessor
Compatible" A/D Converter

12-19

ADC-8208
Microprocessor-Compatible 8-Bit
CMOS A/D Converter

12-2

AID

CONVERTERS
Precision Monolithics Inc.

INTRODUCTION

supply sensitivity, and noise. Proper circuit layout
and grounding are also important factors in
achieving rated performance.

Analog-to-digital converters (A DC) translate
analog input voltages into an equivalent digital
value. PMI's ADCs use the general purpose
successive approximation conversion (SAC)
technique. The basic SAC architecture consists of
a DAC, comparator, input scaling resistors, and a
successive approximation register. A clock and a
voltage reference make a complete ADC system.
PMI uses both bipolar and CMOS technologies to
optimize ADC designs. The bipolar technology
lends itself to high speed ADC conversion and
provides the complete ADC function including an
internal reference. The CMOS technology offers
ADC designs with very low power consumption
and good interface versatility at lower prices. PMl's
improved bipolar and oxide-isolated CMOS
processes offer the best conversion speeds in the
SAC architecture.

Aperture- The period of time during which an
analog input to an ADC should not change more
than 1/2 LSB. Determines maximum sampling
frequency of the ADC.
ADC-An acronym for Analog-to-Digital
Conversion, sometimes written as AID.
Binary Coding-The digital output of an ADC is
considered binary-coded when outputs start at all
zeros for a zero-scale analog input counting up to
all-ones' output code for a full-scale analog input
signal.
Bipolar Mode-An ADC configured to convert
both positive and negative input Voltages.
Bipolar Offset-The analog displacement of one
half of full scale range when the ADC operates in
the bipolar mode. The offset is generally derived
from the converter reference circuit.

A SAC ADC has inherent ratioing capability
between the input voltage and the input reference
level. The digital output is proportional to VINIVREF.
Important criteria for selection of AID converters
include accuracy, conversion speed and resolution.
Other requirements include temperature stability,
output coding, output data format, reference
stability, clock and power needs.
At the digital output many interface data formats
are available to match the wide variety of
application needs. The ADC-910 provides a very
complete microprocessor-controlled interface
which contains data registers, control registers and
hand-shaking lines. This architecture provides a
complete microprocessor compatible system that
allows total software control of all ADC functions.
For stand-alone applications the ADC-8208 offers
direct data output, start conversion and very low
power dissipation.

Bus-A parallel path of binary information signalsusually 4, 8, or 16 bits wide.
Byte-A binary digital word often 8 bits long.
Chip Select (CS)-A logic input signal activating
data or control information transfer between the
ADC and the digital circuitry. This signal is usually
active low designated by a bar over the top of (CS).
CS is used in conjunction with additional qualifying logic signals to determine the exact activities
to take place.
Clock (CK)- The operation of the successive
approximation conversion process requires a
clock. The clock sets the basic conversion rate.
Some ADCs have an internal clock, using an
external capacitor to set the frequency. Most ADCs
have an external clock input.
Code Width-The amount of input voltage change
which occurs between output code transitions
expressed in LSBs of full scale. The ideal code
width is one LSB.

DEFINITIONSANALOG-TO-DIGITAL CONVERTERS

Code Width Uncertainty-The dynamic variation
or jitter in the code width due to noise.

Accuracy-ADC total accuracy consists of the
sum of integral nonlinearity (INL) error, zero error,
gain error and reference error specifications. The
INL is the most critical specification since no
additional user calibration can be easily performed.
Additional specifications affecting ADC device
accuracy includes zero drift, gain drift, power

Coding- The format of the ADC digital output
data. Common formats include binary, offset
binary, complementary binary, two's complement,
low byte and high byte.
12-3

AID
CONVERTERS
Precision Monolithics Inc.

Command Register-An internal register of the
ADC that can be programmed by the user to select
various modes of operation. For example, unipolar
or bipolar conversion selection, range selection,
data output format, etc.

Half-Step Offset-The half-step offset is necessary
in ADCs to center the analog input voltages half
way between the output code transition pOints.
The SAC converter architecture inherently has a 1
LSB transition voltage for the first code change if a
half-step offset is not inserted.
High Byte (HB)-In ADCs with resolutions greater
than 8 bits, some products are offered in a high
byte, low byte format to simplify interface to 8-bit
microprocessor systems. The high byte contains
the most significant bit and some or all of the
upper 8 bits of the ADC output. Generally bits are
loaded in a right or left justified format. Individual
product data sheets should be consulted.

Control Lines-Digital input/output pins that
activate/monitor and control ADC operation. For
example, chip select, write, low byte, high byte,
start convert, end of conversion, conversion
complete, busy, read, etc.
Conversion Complete (CC)- This digital output
signal indicates the end of conversion. When this
signal is in the opposite state the ADC is considered
to be in the "busy" state.

Input Impedance (RIN)-Analog input resistance
of the ADC. In SAC converters input impedance is
primarily resistive, but should be buffered by an
amplifier with a low output impedance at the clock
frequency of the SAC ADC to maintain a stiff input
voltage to the ADC.

Conversion Time (tc)- The amount of time
elapsed from the start-conversion control signal
until the conversion complete signal occurs.
Differential Nonlinearity (DNL)-The worst-case
deviation of distance between transition voltages
from the ideal 1 LSB code width. The parameter
uses units of LSBs. DNL is one of an ADC's errors
which cannot be removed by user adjustments.

Input Range-The analog voltage input range
which results in a zero code to full-scale code
digital output in a natural binary-coded ADC. This
parameter is identical to full-scale range. Common
preset ranges include 0 to 10V, -10 to +10V, -5 to
+5V, etc.

Drift-See Gain Drift, Zero Drift, and Offset Drift
specifications.
End Of Conversion (EOC)-This digital output
signal flags the conversion complete and data
ready condition of an ADC.

Integral Nonlinearity (INL)- This error
specification is a measure of the straightness of
the ADC transfer function. It is measured as the
worst-case deviation in LSBs from a straight line
drawn through the center of the first and last code
widths. INL is one of the key error specifications of
an ADC which cannot be calibrated by the user.
Left Justified Data-In the byte-oriented dataoutput format, data bit sets shorter than 8 bits are
placed starting in the left side of the data output
transfer register. This could apply to the upper or
lower byte. For example, a 12-bit ADC will have 4
extra bits which could be left justified.

Full Scale (FS)-Maximum input voltage of an
ADC. A full-scale input voltage to a natural binarycoded ADC outputs the all-ones code.
Full-Scale Error- This parameter applied to ADCs
is the same as the gain error specification. See
Gain Error.
Full-Scale Range (FSR)-The difference between
maximum and minimum analog values for an ADC
input.
Gain Drift- The change in the full-scale transition
voltage measured over the operating temperature
range This parameter has units of % of full scale,
ppm of full scale, or LSB. It may also be expressed
as % of FS/oC, ppm of FS/oC, etc.

Least Significant Bit (LSB)-The binary digit with
the smallest numerical weighting, normally one
LSB equals full-scale range divided by (2 n), where
n is the number of bits of the ADC. This parameter
may be expressed in millivolts.

Gain Error- The difference between the actual
full-scale transition voltage and the ideal full-scale
transition Voltage. The units of this parameter are
in LSBs or percent of full scale.

Low Byte (LB)-In ADCs with resolutions greater
than 8 bits, some products are offered in a high

12-4

AID
CONVERTERS
Precision Monolithics Inc.

byte, low byte format to simplify interface to 8-bit
microprocessor systems. The low byte contains
the least significant bit and some or all of the low 8
bits of the ADC output. Generally bits are loaded
in a right or left justified format. Individual product
data sheets should be consulted.

the LSB of the ADC which are assigned the same
output code. The ±112 LSB limit to resolution is
known as the fundamental quantization error of an
ADC.

Major Transition-The digital output code change
between one-and-all-zeros code and zero-and-allones code. For example an 8-bit ADC major
transition occurs between the codes 1000 0000 and
01111111 (half-scale).

Reference (VREF)-Reference voltages determine
ADC system's absolute conversion accuracy at full
scale. At other voltages the INL and reference
determine absolute accuracy. References may be
internally or externally supplied.

Missing Code-As an analog input voltage is
swept from zero input to full scale, the output
digital codes should change by 1 LSB as each
transition voltage is reached. If an output code is
skipped, we have a missing code.
Most Significant Bit (MSB)-The binary digit of
the resulting ADC conversion with the largest
numerical weighting. Normally the MSB has a
weighting of one-half full-scale range.

Relative Accuracy-Relative accuracy error,
expressed in LSBs or % of full scale, is the
deviation of the analog value at any code from its
theoretical value after gain and offset is calibrated.
PMI's usage of this specification is identical to the
integral nonlinearity specification.

R-2R Ladder-A resistor network providing the
basic binary-current-division used in SAC ADCs.

Resolution-The resolution of an ADC is the
number of states (2n) that the analog input voltage
is divided (or resolved) into, where n is equal to the
number of bits.

Offset Binary Code-A digital binary code used to
represent plus or minus input polarity analog
voltages. Negative full-scale voltage is assigned allzeros code. Zero input voltage has the MSB high
(logic one) and the remaining bits at zero. Positive
full-scale voltage is assigned all-ones code.

Right Justified Data-In the byte-oriented dataoutput format, data bit sets shorter than 8 bits are
placed starting in the right side of the data output
transfer register. This could apply to the upper or
lower byte. For example, a 12-bit ADC will have 4
extra bits which could be right justified.
Sampling Frequency- The rate at which the ADC
can continuously convert analog inputs into digital
outputs. In SAC ADCs clock frequency and data
transfer overhead determines the sampling
frequency.
Short Cycling-Termination of the conversion
sequence of an ADC to less than the total number
of clock periods required for a full resolution
conversion.
Span-Sometimes used to describe the full-scale
analog input voltage. Some ADCs have resistorselectable analog input ranges.
Status Register-A register indicating current
status of the analog-to-digital conversion with a
busy signal or a conversion complete Signal.
Successive Approximation Conversion (SAC)Successive approximation converters compare an
analog input against an accurately-known binary

Offset Drift-The change with temperature of
analog zero for an ADC operating in the bipolar
mode. It is generally expressed in ppm of FSR/o C
or LSBs.
Offset Error- The error at analog zero for an ADC
operating in the bipolar mode. It has units of % of
FSR or LSBs.
Overrange-When analog input voltages exceed
the input range, an overrange condition is in effect.
Normally the digital output code stays at all ones
for positive overrange and all zeros for negative
overrange with standard binary coding.
Power Supply Sensitivity (PSS)- The change in
the full-scale transition voltage due to a change in
power supply voltage. The units are in LSB, % of
FS per % of supply voltage, or ppm of FS per % of
supply voltage.
Quantization Error-An ADC of n bits can only
identify 2n output codes; however, there exists an
infinite number of analog input values adjacent to

12·5

AID
CONVERTERS
PrCCISJ()lI MOIlOlttlllCS

Inc.

fraction of a reference input. Starting with half the
reference value (the MSB) the converter
determines whether the input is greater than or
less than this value. Next the converter divides the
remaining value in half again determining whether
the input value is in the upper or lower quarter.
This process is continued until all bits have been
tried. The digital result of each trial is a numerical
representation of the analog input.
Successive Approximation Register (SAR)-A
digital circuit that controls the operation of a
successive approximation ADC and accumulates
the output digital word in its register.
Three-State Output Buffer-A digital output circuit
that can be programmed to output a logic low,
logic high or a high output impedance state. These
devices are generally connected to digital buses.
Transition Noise-Width of the transition voltage
region.

Transition Voltage-The transition voltage is the
center of a finite band of analog input voltages
where the digital output code of the ADC toggles
between two adjacent codes.
Unipolar Mode-Operation of an ADC with zero to
full-scale voltage inputs of one polarity only.
Word-A set of binary digits that represent the
fundamental register size of the digital circuits
being used. For example, Z80, 8085, and 6800
microprocessor systems use 8-bit words, while
68000 and 80186 microprocessor systems have
16-bit words.
Zero Drift-The change with temperature of
analog zero for an ADC operating in the unipolar
mode. It is generally expressed in ppm of FS/o C,
or LSBs.
Zero Error-The error at analog zero for an ADC
operating in the unipolar mode. This is a user
adjustable parameter.

12-6

ADC-910
HIGH-SPEED "MICROPROCESSOR COMPATIBLE"
A/D CONVERTER
Precision Monolithics Inc.

PRELIMINARY
read/write inputs and 3 Chip Select inputs to control the 10
data lines is included. Interrupt enable, start conversion and
bipolar/unipolar mode selection are controlled by the data
bus. The use of high-speed Linear Differential Logic results
in fast (6/Ls) conversion time and low power dissipation.

FEATURES
•
•
•
•
•
•
•
•

Includes Clock, Reference, 3-State Buffered Outputs
Fast Conversion Time ............................ 6/Ls
Four Input Ranges .. + /-2.5V, + /-5.0V, +5.0V and + 10.0V
112 LSB INL
No Missing Codes Over Temperature
Low ESD Sensitivity Due to Rugged Bipolar Processing
Software Programmable Unipolar/Bipolar
Easily Interfaced to 8 and 16·Blt /LP Bus

PIN CONNECTIONS

ORDERING INFORMATIONt
PMI MODEL NO.

TEMP RANGE

ADC910AT"
ADC910ET
ADC910GT

-55' C/+ 125'C
-25'C/+85'C
O'C/+70'C

DIGITAL
GNO
(LSB)BO

6

• For devices processed in total compliance to MIL-STD-883, add /883 after
part number. Consult factory lor 883 data sheet.
tAil commercial and industrial temperature range parts are available with
burn-in. For ordering information see 1986 Data Book, Section 2.

21

~~~LOG

18

g~~:ET

17 V REF OUT
16 VAEF ADJUST

GENERAL DESCRIPTION
The ADC-910 is a 10-bit A/D converter designed specifically
for interfacing with microprocessors. 3-state data outputs
allow direct connection to an a-bit data bus in an MSB byte of
a bits and an LSB byte of 2 bits. A command register with

28-PIN HERMETIC DIP
(T-Sufflx)

SIMPLIFIED SCHEMATIC
MSB
LSB
89 BS 87 86 B5 B4 83 B2 B1 80

Y YY

CS3

CS2

?-

START
UNI!SIP

INE
COMMAND
ENABLE
HIGH BYTe

t

!

CLOCK

CClK

V REF IN

V AEF OUT

-

~.
~~~
~e~

--

~

I

'MSB

...........

f-<> V+
ANALOG

.~"
~yFFERS II~~"l
I~"

~- f-<>

GNO

v-

J

VINB

"~

"

R

VINA

COMP

SAR

R

-

3-STATE BUFFERS (81

2R

caMP
OFFSET

2R

REF

V REF ADJ

-

REGISTER

CSi

CLOCK IN

DIGITAL
GNO

.J,-

'\

~

10-BIT OAC

R

~

~7

U

This preliminary product Information Is based on testing of a limited number of devices. Final specifications may vary. Please contact local sales
office or distributor for final data sheet.

12-7

1/86, Rev. A

------1~ ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER -

PRELIMINARY

ABSOLUTE MAXIMUM RATINGS
Operating Temperature Range
ADC-910AT/BT •.••..•.•..•.•..•..••• -SS·C to +12S·C
ADC-910ET/FT ...••....•..........••. -2S·C to +8S·C
ADC-910GT/HT •.•.••••...•......•.•... O·C to +70·C
Storage Temperature Range •..•.•....• -6S·C to +1S0·C
Power Dissipation at TA = 12S·C ••...•••..•..•.. 1000mW
Lead Temperature (Soldering, 60 sec) ••...•••••••• 300·C
Supply Voltage (V+) ••••.•.••••••.•..••...•.••••.••.• 6V

ELECTRICAL CHARACTERISTICS at V+

= SV,

V-

Supply Voltage ( V-) •••••••••••••••.•.••.•••..••••..• 6V
V+ to V- ••••.•••••••••.••.•••....•••••..•.•.••.••.• 12V
Logic Inputs ..•.•.•.••••.•••••••.....•••.••• +6V, -0.3V
Logic Outputs (in 3-state) ••.••.••...•.••.•••• +6V, -0.3V
VINA ..•••••...•••...•••..•.•.•.•••.•.••.••••.•.•.• 1SV
VINB •••••..•••••••••••.••..•....•••••••.•.•.••.••. 7.SV
Reference Inputs .•...•••••.•..•••.••.•••.•••.••.•• 3.0V
Digital Ground to Analog Ground Voltage ••.••.•••.• O.SV

= -SV,

VREF

= 2.SV, fCLK = O.SMHz;

= -SS·C to +12S·C apply for

TA

ADC-910AT/BT, unless otherwise noted.
PARAMETER
Integral Nonlinearity

(Note 3)
Gain Drift
(Note 1)

ADC-910BT
MIN
TYP MAX

ADC-910AT
MIN
TYP MAX

SYMBOL

CONDITIONS

INL

TA = 25°C
TA = Full Temp. Range

TCG FS

External Reference
Internal Reference

1/2
3/4

LSB

25
40

4.75V < V+ < 5 25V

UNITS

500

30
50

ppm FS;oC

600

MVIV

Positive Supply Current

1+

30

40

30

40

mA

Negative Supply Current

I-

SO

60

50

60

mA

Reference Line RegulatIon

ELECTRICAL CHARACTERISTICS at V+

= SV,

V-

= -5V,

VREF

= 2.SV,

fCLK

= O.SMHz;

TA

= -2S·C to +8S·C

apply for

ADC-910ET/FT, unless otherwise noted.
PARAMETER
Integral Nonllneanty

(Note 3)
Gain Drift
(Note 1)

ADC-910ET
MIN
TYP MAX

SYMBOL

CONDITIONS

INL

TA =25°C
TA = Full Temp. Range

1/2
1/2

TCG FS

External Reference
Internal Reference

20
35

4 75V < V+ < 5 25V

ADC-910FT
MIN
TYP MAX

UNITS
LSB

500

25
45

ppm FS/oC

600

MVIV

Positive Supply Current

1+

30

40

30

40

mA

Negative Supply Current

I-

SO

60

50

60

mA

Reference Line RegulatIon

ELECTRICAL CHARACTERISTICS at V+

= 5V,

V-

= -SV,

VREF

= 2.SV,

fCLK

= 0.5MHz;

TA

= O·C

to +70·C apply for

ADC-910GT/HT, unless otherwise noted.
PARAMETER
Integral Nonlinearity

(Note 3)
Gain Dnft

(Note 1)

ADC-910GT
MIN
TYP MAX

ADC-910HT
MIN
TYP MAX

SYMBOL

CONDITIONS

INL

TA = 25°C
TA = Full Temp. Range

TCG FS

External Reference
Internal Reference

10
25

10
25

ppm FS/oC

4.75V < V+ < 5 25V

1/2
3/4

UNITS
LSB

300

300

MVIV

PositIVe Supply Current

1+

30

30

mA

Negative Supply Current

I-

SO

50

mA

Reference Line Regulation

12-8

1/86, Rev. A

!EM!) ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER ELECTRICAL CHARACTERISTICS at V+ =

5V, V-

PRELIMINARY

= -5V, VREF = 2.5V, fCLK = O.5MHz; TA = 25°C, unless otherwise noted.
ADC-910AT/ET/GT

CONDITIONS

Resolution

N

TA ~ Full Temp Range

10

10

Bits

TA ~ Full Temp Range (Notes 2, 3)

10

10

Bits

No MIssing Codes Guaranteed
Gain Error
Unipolar Mode

Offset Error
Bipolar Mode

Offset Error

G FSE

VRE ,

V ZSE

TA

~

~

MAX

ADC-910BT/FT/HT

SYMBOL

Resolution for which

MIN

TYP

PARAMETER

MIN

TYP

MAX

2 500V (Notes 2, 3)

UNITS

LSB
1/2

Full Temp Range

LSB

VasE

1.5

LSB

15

LSB

Bipolar Mode ZeroScale Offset Drift

TCVzs

TA ~ Full Temp Range (Note 1)

Analog Input Impedance

RINA

Pin 20

35

Analog Input Impedance

RIN B

Pin 19

175

25

35

175

25

35

kfl

Reference Input Resistance

RREF

Pin 22

175

2.5

3.5

175

25

35

kfl

Reference Voltage Output

VRE,OUT

Pin 17, Untnmmed

245

250

2.55

245

2.50

255

V

RT~

±50

Reference Voltage
Tnm Range
Reference Output

Supply Sensitivity

Negative Power

TA

+P ss

= Full

Temp Range

4.5V to 5.5V, VRE , External

mV

15

15

mV/mA

1/2

1/2

LSB

C/J

~
~

E2

~

0

U

.....l

~
,......

C)
,......
-P ss

-4.5V to -5 5V, VRE , External

Conversion Time

Tc

fCLK

~

1 MHz (Note 5)

ConverSion Time

Tc

fCLK

~

0 5MHz (Note 6)

Supply Sensitivity

kfl

±50

1mA< I < SmA,

Load Reg ulation

Positive Power

10kfl

35

Digital Input High

V 1NH

TA ~ Full Temp Range

Digital Input Low

V1NL

TA ~ Full Temp Range

1/2

LSB
~s

12
20

12

Digital Input Current

IINH

TA ~ Full Temp Range

04

IJNL

TA ~ Full Temp Range

to

Digital Output High

VOH

I OH ~ -400~A,
TA ~ Full Temp Range

Digital Output Low

VOL

10L ~ 16mA,
TA ~ Full Temp Range

Digital Output Current

10H

VOH~24V

Digital Output Current

10L

VOL ~ 04V

Three-State
Output Leakage

102

TA

24

V
08

V

20

~A

04
20

37

01

10
24

04

-400

Full Temp Range

~s

2.0
08

Digital Input Current

~

t/2

~A

37

01

V

04

-400

V
~A

16

16

mA

10

10

~A

NOTES:
Change In 25°C value from 25°C to TM,n or TMax
Tested In the 5V unipolar mode at 6~s conversion time
Tested In the ±5V bipolar mode at 12J.ts converSion time
Measured mdlvldually on V+ and V- supplies
Applies to 5V Input unipolar operation, see Figure 1 for connections
6 Applies to 10V mput unipolar operation, and ±5V/±lOV Input bipolar
operation, see Figure 1 for connections

12-9

1/86, Rev. A

Q,

0

~

C)

0

~

-<

II

------1m

ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER - PRELIMINARY

APPLICATIONS INFORMATION
CIRCUIT OPERATION (refer to the simplified schematic)
The ADC-910 uses a successive approximation type A/D
conversion routine. When a start command is received by the
command register, the SAR, DAC and comparator begin a
bit-by-bit trial against the analog input voltage. When all ten
bits have been tried, the ten data outputs of the SAR will
contain a 10-bit digital representation of the analog input
voltage.
When the conversion is complete, a read command and a
chip selection will output the data through the 3-state output
buffers. Selecting CS1 will output the eight MSBs (the high
byte) and selecting CS2 will output the two LSBs (the low
byte). Selecting both CS1 and the CS2 will cause all ten data
bits to be output through the 3-state output buffers.
When the conversion is complete, the SAR sends an end of
conversion (EOC) signal to the command register, which
turns on the interrupt output open-collector NPN transistor
(INT), providing the interrupt disable bit (lNE) is set to "0".
The EOC Signal is also multiplexed into the input of the
3-state buffer for bit 9 (B9). Also, at this time, the overrange
signal appears at the SAR output and IS multiplexed into the
input of the 3-state buffer for bit 8 (B8). These two bits of
information compnse the status register, which is multiplexed to the data bus with a read command and a selection
of CS3.
Unipolar/bipolar mode selection and the enabling/disabling
of the interrupt output is done when the start of conversion

command is entered. In the unipolar mode, the IMSBcurrent
source is turned off. For bipolar mode operation, the I MSB
current source is applied to the summing mode of the
comparator. This provides the proper offset of IMSB to do a
bipolar conversion.
BASIC CONNECTIONS (refer to Fig. 1)
Power Supply Connections: The ADC-910'is operated on ±S
volt power supplies. +S volts is applied to pin 1S and -S volts
is applied to pin 23. These lines should be bypassed near the
device with a 0.1/LF capacitor in parallel with a large value
capacitor such as 10/LF.
Analog and Digital Ground: Separate analog and digital
grounds are provided to maintain optimum noise rejection.
Care should be maintained to insure that digital switching
noise is not introduced into the analog ground line. This can
be accomplished by making the final ground point as close
(physically and electrically) as possible to the analog ground
pin of the ADC-910.
Analog Inpuls: There are two analog voltage inputs to the
ADC-910. VINA (pin 20) accepts input Signals between 0 volts
and + 10volts in the unipolar mode and between-S volts and
+S volts in the bipolar mode. V INB (pin 19) accepts input
signal levels between 0 volts and +S volts in the unipolar
mode and between -2.S volts and +2.S volts in the bipolar
mode. The input resistance is nominally Skfl forV INA and 2.Skfl
for VINB. The comparator offset pin (pin 18) is left open when
using VINA, and is tied to analog ground when using VINB.

12-10

1/86, Rev. A

-----t~ ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER -

FIGURE 1: BASIC CONNECTIONS

PRELIMINARY

*
+

Ace-91O

...!i
...E.
..E.

BS

V+

B8

V 1NA

~

B7

VINB

p!-----o VIN (5V RANGE)

....!!

B6

....!!'.
2
2
-2.

DATA
LINES

+5V

6k(l

B4

VREF OUT

B3

V REF IN

B2

B,

-2.

BO

2.

,

TO MICROPROCESSOR
2

WR OUTPUT

('

1

~
~

~

ANALOG
GND

f2---.

SWITCH OPEN FOR
O_IOV INPUT RANGE.
SWITCH CLOSED FOR

0-5V INPUT RANGE.

10kn

'i7
iNT

DIGITAL
GND

elK IN

4

t:-----a

+5V

25
CCLKn

RD

J.
_
-

~OOPF - FOR INTERNAL CLOCK

WR

CSi CS2 CS3

v-

23

~g~~:Ji~t~t2L5 ~~O~~'

-5V

~

,2Sf 3
ADDRESS BUS' \

VIN (10V RANGE)

~

VREF ADJ

AD OUTPUT

TO MICROPROCESSOR

'5

CaMP 18
OFFSET

B6

~

TO MICROPROCESSOR

rna

CBVPASS

+5V

CBYPASS

ADDRESS

DECODER

FIGURE 2: START CONVERSION AND OPERATING MODE SELECTION
(WRITE MODE WR = "LOW", CS3 = "LOW")
MSB
B9 B8

LSB
B7

B6

BS

B4

B3

B2

Interrupt) (UniPOlar/)
( Disable/
BI polar
Enable
Select

B1

BO

Start)
( Bit

,----,-I
X 1x--,---I
X --,-----,-I
X I--,---x
1 _ 0-,---0---,---------,I
1,----,X 1x--,--I
X --'----1.1
X I---,--x
1 ~--,-----,I

[EJ
[EJ

Start Conversion,
Bipolar Mode,
Interrupt Enabled.
Start Conversion,
Unipolar Mode,
Interrupt Disabled.

x = "Don't Care"
FIGURE 3: READING DATA AND STATUS
(READ MODE RD = "LOW")
LSB
B1

BO
CS1 = "Low"
Read High Byte

CS2="Low"
Read Low Byte

CS3 = "Low"
Read Status

12-11

1/86, Rev. A

------l1fMD

ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER - PRELIMINARY

FIGURE 4: ADC-910 TIMING DIAGRAMS
(a) Write Cycle

SAR/R~SEBT DATA
LATCHED

I

2ND
3RD

4TH

5TH

6TH 7TH

STATUS DATA
10TH READY

8TH 9TH

\

CLOCK

eS3

-'*, :'

---:-"'1:.-__

, '

,I[

, twc -

I
,

r-

I

, ,

,

I

tWR - :

200ns MIN
tel - 150ns TVP
tWR -lOOns MIN

tOH -OnsMAX
tos - lOOns TYP

(b) Read Cycle

"\-I-_ _ _f
I

'~--------------

~
,I
,,

:+

~--------~,-'-,
I

I

: i

I

D A T A - - - - - - - - - + : - + - ! ~~~3=-S~T~AT~E~-----------

tRI_:

tRA---.1

tAl - 1500s TYP

I
I

tRA - 200ns MAX
tRW -150ns MIN

ij

\

I

I
-I

jtRW-:

tOF _lOOns TYP

:

*--: :

I
r---tOF

i

(c) Stand-Alone

STATUS DATA

READY

\

CLOCK

,

,

,

,

,

I

I

I_I
I I

~tsc
I

I
I

,:--f:.+'------------------+'----:

CS1~

I I41-----------------~1-----WR~I
I

: -: 0-

TNT

I

I'

I

I I

I
I

I
:

I
:

I
~ lOOns....)
I MIN
I

I
I
I

i

t cw

,
I
ct::1

twr-l r-I tww

I, :\.f\...

tsc -

:
I

lOOns MIN

tew - 50ns MIN
tWI - 50ns TYP
tel - 150ns TYP

I
tCI_1
I

:
I
I
'-I

reference adjust pin (pin 16).

Voltage Reference: The voltage reference for the ADC-910 is
nominally +2.5 volts. To use this internal reference, the reference output pin (pin 17) should be tied to the reference input
pin (pin 22). Adjustment of the reference voltage may be
done by applying a 10kO trimpot between the reference output and analog ground with the center tap wiper tied to the

To use an external reference with the ADC-910, simply apply
it to the VREF input pin (pin 22). This voltage should be
bypassed to analog ground with a O.1/-1F capacitor.

12-12

1/86, Rev. A

_ _ _ _~~ ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER - PRELIMINARY

Clock: For internal clock operation, the external capacitor
(CCLK) sets the conversion rate. The conversion rate graph
provides the relationship of CCLK and temperature to conversion rate. The CCLK capacitor is connected between CCLK
(pin 25) and the V- supply (pin 23), see Figure 1. The clock
input (pin 26) is connected to the V+ supply (pin 15). Internal
clock operation exhibits a conversion time variation from
device to device for a given CCLK, due to capacitor and
internal resistor tolerances of the basic R-C oscillator. For
operation at the upper frequencies of 0.5 and 1MHz, an
external clock input is recommended.

16-bit bus, single 10-bit word reading is possible. When using
an 8-bit data bus, the "high byte" and "low byte" can be
multiplexed onto a single 8-bit bus as indicated in Fig. 5.
To read all 10 bits at once, the RD (pin 1). CSl (pin 28) and CS2
(pin 27) are all held "low". This turns on 3-state output buffers
and all data bits can be read.
To read the 8-bit "high byte", the RD (pin 1) and CSl (pin 28)
lines are held "low".
To read the 2-bit "low byte", the RD and CS2 lines are held
"low".

For external clock operation, no clock capacitor is required.
The CCLK pin (pin 25) should be tied to the -5 volt supply and
the external clock is applied to the clock input (pin 26).
1.0MHz clock maximum may be used. This will result in a 6j.Ls
conversion time. Slower clock rates will result in slower
conversion speeds.
Conversion time

Included on the ADC-910 is a 2-bit status register which is
multiplexed onto the data bus on lines B9 and B8.
To read the status register, RD (pin 1) and CS3are held "low".
End of conversion (EOC) is indicated by a "lOW" bit9 (pin 14)
and overrange (OR) is indicated by a "high" in bit 8 (pin 13).

~ 6 x -1f

CALIBRATION (10 Volt unipolar mode only)
To adjust out gain error, a trimpot may be inserted in series
with the analog input voltage input. Assuming a 2.500 volt
reference is applied at the reference input, gain error trimming is accomplished by adjusting the input trimmer so that
the final digital output code transition occurs for an input
voltage of VA = 9.985 volts (this is the transition from
1111111110 to 1111111111). When using the internal reference
or an adjustable external reference, gain error trimming may
be accomplished by adjusting the reference voltage until the
final digital output code transition occurs at VA = 9.985 volts.

CLK

40
35

~
w

:0;

30
25

;::
z
o

20

>

'5

~

8 '0

.
r----.. t-.
l""- t- t--

l"- I"- r--

C EXT - 3000pF

CEXT

=

DRIVING THE ANALOG INPUT
To insure 10-bit accuracy the input to the ADC-910 must be
driven by a source which has an output impedance of less
than 0.5 ohms at 1MHz.

1900pF

CEXT = 800pF

o

-75

-50

-25

25

50

75

100

125

INTERFACING THE ADC-910 TO THE 6502j.LP
(refer to Fig. 5)
An example of direct connection to an 8-bit data bus is shown
in Fig. 5. Notice that the two least significant bits are connected to data bits B3 and B4. This allows a 10-bit data
transfer over an 8-bit bus. In this example, a Synertek Systems SYM-l Educational Computer Board supports the
6502j.LP. The flow charts and op codes for a variety of conversion exercises are shown below.

CHIP SELECT, READ AND WRITE INPUTS (refer to Fig. 2)
Start Commands: To start a conversion the WR input (pin 2)
must be held "low" while CS3 (pin 3) is held "low" and a logic
"high" is applied to bit 2 (pin 7). Another way to start a
conversion is to hold CSl (pin 28) and WR (pin 2) "low" for a
complete clock cycle.
Operating mode selection is done when the start command is
applied. As with the start command, WR and CS3 are held
"low". A logic "high" applied to bit 4 (pin 9) disables the
interrupt and a logic "low" enables the interrupt. A logic
"high" applied to bit 3 (pin 8) selects unipolar mode and a
logic "low" selects bipolar-mode operation.

INTERFACING THE ADC-910 TO THE MC68000
(refer to Fig. 6)
An example of a direct connection to a 16-bit data bus is
shown in Fig. 6. The 10-bit output of the ADC-910 is connected directly to the 10 least significant bits of the MC68000
data bus. In this example a Motorola MC68000 Computer
Board supports the 68000j.LP. A flow chart and assembly
language program is shown below for a simplified 10-bit wide
conversion.

READING DATA AND CONVERSION STATUS
(refer to Fig. 3)
Data can be read in two ways: a Single 10-bit word or in a 8-bit
"high byte" with a 2-bit "low byte". When interfacing to a

12-13

1/86, Rev. A

-----t1£'MD

ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER - PRELIMINARY

FIGURE 5: ADC-910 INTERFACE TO 6502~P ON SYM-1 BOARD
+

~CBYPASS

ADC-910

MSB

D7

SYM-l
(6502,uPJ

+5V
DB9

D6

DB8

DS

DB7

D4

DBS

D3

DBS

D2

DB4

DI

I

DO

F
V
Skn

IRQ
~2

R/W

iii

:R>

~

V+

V'NA ~VIN(IOV RANGE)
V ,NB ~VIN(5V RANGE)
COMP
OFFSET
V REF OUT

V REF IN ~

DB3
DB2

V REF AOJ

DBl

r---:

GND ~

~7

DIGITAL
GND

CCLK~
WR

1:

V

CBVPASS~

74LS138

L

I

Y4 ~C5A
Y5 $1801

8

VS

G2

V7

*

CLOCK

AD

¢

Gl

10kn

ANALOG

DBO

CS1CS2CS3

II

SWITCH OPEN FOR
O....10V INPUT RANGe.
SWITCH CLOSED FOR
O.... 5V INPUT RANGE.

INT

Al AD

A

'1

~

r---

$1802
$1803

ADDRESS
DECODER

FIGURE 6: ADC-910 INTERFACE TO MC68000 COMPUTER BOARD
+

I':" CBYPASS

ADC-910

MSB

DO

Me6S00a

+5V

DB9

V+

08

DB8

07

D87

V'NA
V ,NB

DS

DB6
DBS

OFFSET

0,

DB'

VREF OUT

03

DB3
DB2

01

DBl

00

DBO

V REF IN ~
VREF ADJ

r---:

ANALOG
GND

r---

SWITCH OPEN FOR

o.... ,ov INPUT RANGE.
SWITCH CLOSED FOR
0 __ 5V INPUT RANGE.

10kn

~7

DIGITAL
GND

......

.....

R/W

L

_ _ ADDRESS
BUS

I

'1

~

r---

CLOCK

lMHzCLOCK

CTACK

RANGE)

Y,N (5V RANGEl

COMP

DS

D2

~VIN(10V
r-----o

V

CCLK

WR

es; CS2 CS3

1

El

f\

\
/

AD

74LS138
E3

V-

1

*

*

-5V

BYPASS

C

ADDRESS

DECODER

OR GATE
OPEN COLLECTOR

r

12-14

1/86, Rev. A

----l~ ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER -

PRELIMINARY

ADC-910 INTERFACE SOFTWARE AND FLOW CHART FOR 8502I'P (SYM-1)
Unipolar Mode, Interrupt Disabled

PC

MNEMONIC

OPCODE

COMMENT

0302
0304
0307
0308
0309
030C
030F
0312
0314
0317

LDA#$07
STA$1803
NOP
NOP
LDA$1801
STA$03AO
LDA$1802
AND#$03
STA$03A1
JMP$0302

A907
8D0318
EA
EA
AD 0118
8DA003
AD 0218
2903
8D A1 03
4C 02 03

Select Mode
Start Conversion
Wait 21'S
Wait 21'S
Read High Byte
Store High Byte
Read Low Byle
Mask Bits B7-B2
Store Low Byte
Goto 0302

1/86, Rev. A

12-15

----

-

-

-

--- -

~

~

--

-----~-

------1~ ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER -

PRELIMINARY

ADC-910 INTERFACE SOFTWARE AND FLOW CHART FOR 6502}tP (SYM-1)

Polled Mode, Overrange Test

PC

MNEMONIC OPCODE

COMMENT

0200

LDA#$OX

A90X

0202
0205
0208
020A
020C
020F
0211
0213
0216
0219
021C
021F

STA$1803
LOA$1803
ANO#$80
BNE-7
LOA$1803
AND#$40
BEQ+3
JMP$0227
LDA$1801
STA$024E
LDA$1802
AN 0#$03

800318
AD 0318
2980
DO F9
AD 0318
2940
FO 03
4C 27 02
AD 0118
80 4E 02
AD 0218
2903

0221
0224

STA$024F
JMP$0200

80 4F 02
4C 00 02

0227
022A

LDA$1800
JMP$0200

AD 00 18
4C 00 02

Select Mode: X =
1,3,5 or 7 (see below)
Start Conversion
Read Status Register
Mask for EOC Bit
Loop Until EOC
Read Status Reg ister
Mask for OR Bit
If Not OR, go to 0216
Else Go to 0227
Read High Byte
Store High Byte at 024E
Read Low Byte
Mask for 2 LSBs
(Low Byte)
Store Low Byte at 024F
Loop Back to Start
(0200)
Light LED for OR
Loop Back to Start
(0200)
High Byte
Low Byte

024E
024F
Mode
X=1:
X=3:
X=5.
X=7:

12-16

Selection:
Bipolar Mode, Interrupt Enabled
Unipolar Mode, Interrupt Enabled
Bipolar Mode, Interrupt Disabled
Unipolar Mode, Interrupt Disabled

1/86, Rev. A

------t~ ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE" AID CONVERTER -

PRELIMINARY

ADC-910 INTERFACE SOFTWARE AND FLOW CHART FOR 6502",P (SYM-1)
Interrupt-Driven Conversion
INTERRUPT SERVICE ROUTINE:

PC

MNEMONIC OPCODE

0200
0202
0205
0207
020A

LOA#$02
STA$A679
LOA#$10
STA$A678
LOA#$03

A902
8079 A6
A912
80 78 A6
A903

020C
020F

STA$1803
JMP$20A

800318
4COA02

COMMENT
Set Interrupt Vector

Select Mode (Unipolar,
Interrupt Enabled)
Start Conversion
Jump to 20A
(Loop Until Interrupt)

Interrupt Service Routine
0212
0215
0218
021B
0210
0220
0222
0225

LOA$1801
STA$024E
LOA$1802
ANO#$03
STA$024F
LOA#$03
STA$1803
RTI

AD 0118
804E02
AD 0218
2903
80 4F 02
A903
800318
40

Read High Byte
Store High Byte at 024E
Read Low Byte
Mask Out Bits 9-4
Store Low Byte at 024F
Select Mode
Start Conversion
Return from Interrupt

~

~

~

0

U

~

......
0
......
Q,

0

~

0
0

~

12-17

1/86, Rev. A

-----l~ ADC-910 HIGH-SPEED "MICROPROCESSOR COMPATIBLE"

AID CONVERTER - PRELIMINARY

ADC-910 INTERFACE SOFTWARE AND FLOW CHART FOR 16-BIT/LP
MC68000 Computer Board Start Sequence

PC

MNEMONIC

COMMENT

1000
1002
1008
100A
100C
100E
1014

MOVEQ #12, DO
MOVE DO, $50000
NOP
NOP
NOP
MOVE $20000, 01
ANDI #1023, 01

Select Mode*
Start Conversion

1008

JMP $1000

Delay
Read Data
Mask out B15-Bl0
leaving B9-BO
Jump to 1000

* Loading a decimal 12 Into DO will apply the following binary word to the
command register at the start of the converSion'

89888786858483828180
0000001100
This results

In

unipolar mode selection with the Interrupt disabled.

ADC-910 INTERFACE SOFTWARE AND FLOW CHART FOR 16-BIT/LP
MC68000 Computer Board Polling Status Register

PC

MNEMONIC

COMMENT

1000
1002
1008
100E

MOVEQ#X,DO
MOVE DO, $50000
MOVE $50000,01
AND #512,01

1012
1016
101C

BNE.L $1008
MOVE $50000,02
AND #256,02

1020
1024
102A
102E
1032
1036

BEQ.L $102E
MOVE 03,$40000
JMP $1000
MOVE $20000,04
AND #1023,04
JMP $1000

Select Mode*
Start Conversion
Read Status Register into 01
Mask for EOC Bit (1000000000 =
512 Decimal)
Loop Until EOC
Read Status Register
Mask for OR Bit (0100000000 =
256 Decimal)
Branch to $102E Unless OR
Light OR Indicator
Start Over
Read and Store 10-Bit Data
Mask Unwanted 6 LSBs
Start Over

'For 81polar Mode with Interrupt Enabled· X = 4 DeCimal
For Unipolar Mode with Interrupt Enabled. X

= 12 DeCimal

For Bipolar Mode. Interrupt Disabled. X = 20 DeCimal
For Unipolar Mode, Interrupt Disabled· X = 28 Decimal

12-18

1/86, Rev. A

ADC-8208
MICROPROCESSOR- COMPATIBLE
8 -BIT CMOS AID CONVERTER
Precisioll MOllolithics Inc.

ADVANCEPRODUCTINFO~TION
FEATURES

The ADC-8208 provides an improved second-source to the
AD7574, providing a three-times faster conversion time.

• Fast Conversion Time ........................ 5f1S
• 8·Blt Accuracy ..•.............. ± 1/2 LSB Max INL
• Memory·Mapped Interface
• Low Power Dissipation ...................... 30mW
• Operates from Single + 5V
• Fits AD7574 Sockets, with Improved Conversion Time
• Space Saving 0.3 Inch Wide 18·Pin DIP

This low power device is ideal for process control, instrumentation, navigation, and general data-acquisition systems.

PIN CONNECTIONS

GENERAL DESCRIPTION

18·PIN CERDIP
(Q·Sufflx)

The ADC-8208 is an 8-bit microprocessor compatible AID converter which uses the successive-approximation conversion
technique to provide a 5f1s maximum conversion time. Control
logic and three-state data output buffers constitute the memorymapped microprocessor interface. The CS and RD control lines
reset the converter, start conversion, and read output data. The
BUSY output indicates conversion in progress.

18·PIN EPOXY DIP
(P·Sufflx)

FUNCTIONAL DIAGRAM
eLK

Vee

?

V,.
60,s

r8>

SA"

rl~1

INTERFACE
& CONTROL

LOGIC

THREESTATE
BUFFERS
DATA OUT
DBO-OB7

r
AGNO

~
b

DGNO

This advance product Information describes a product In development at the time of this printing. Final specifications may vary. Please contact
local sales olllce or distributor for IInal data sheet.

12·19

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3 ':
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13

Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

ANALOG SWITCHES
MULTIPLEXERS
Precision M0l101ithics Inc.

13-3

Introduction

13-3

Definitions

13-7

SW-Ol/SW-02
Quad SPST JFET Analog'switches

13-14

SW-05
Dual SPST JFET Analog Switch

13-22

13-48 MUX-08/MUX-14
8-Channel/Dual 4-Channel JFET
Analog Multiplexers
13-59 MUX-161MUX-28
16-Channel/Dual 8-Channel JFET
Analog Multiplexers
13-69 MUX-88
8-Channel Analog Multiplexer For
PCMCODECs

SW-06
Quad SPST JFET Analog Switch

13-74

13-33 SW-201/SW-202
Quad SPST JFET Analog Switches
13-40

SW-7510/SW-7511
Quad SPST JFET Analog Switches

13-2

DMX-88
8-Channel Analog De-Multiplexer

ANALOG SWITCHES
MULTIPLEXERS
Precision Mouolithics Inc.

INTRODUCTION
Analog multiplexers and switches find applications in data acquisition, metrology, telemetry,
process control and telephony systems. Multiplexers are multiple analog switches which
share a common output. An on-chip address
decoder selects the appropriate input by means
of a binary code. All channels may be deactivated by an enable/disable control pin.
In the past multiplexers/switches have been
manufactured with hybrid, monolithic CMOS or
dielectrically isolated CMOS technologies. The
merging of ion implant techniques with the
standard bipolar process creates a fourth technological alternative - the bipolar-JFET
process. High-quality ion implanted p-channel
FET's can now be compatibly processed with
bipolar devices.

with "RON" placing fundamental limits on signal
acquisition time. Low "RON" and "COUT" insures
minimum elapsed time between the channel
select command and the acquisition of data to
within a specified error band. High cross talk
and off isolation specifications prevent unselected input signals from affecting the signal
path.
PMI offers a wide selection of single-ended and
differential multiplexers and switches. Sixteen
and eight-channel multiplexers as well as differential eight and four-channel devices are available. Dual and Quad SPST switches in normally
closed and open configurations are also available. All devices are pin-for-pin replacements for
many industry standard CMOS devices.

DEFINITIONS

The cost of hybrid devices limits their use to
applications which require the extremely low
"RON" resistance made possible by discrete
FET's. MOS technologies are inherently plagued
by SCR "latch up" problems and analog signal
overvoltage destruction. The use of buried layers and expensive dielectric isolation processing
can eliminate the SCR failure mode, but the
overvoltage blowout problems can be solved
only by adding large series input resistance with
each switch. This increases system errors since
the equivalent "RON" may typically be over
1000 ohms.
JFET switches have no SCR "latch up" tendency
and can withstand analog input overvoltages
while maintaining low "RON" resistance. In
addition, the special handling required with
CMOS devices is not necessary with JFET
switches.
In selecting analog multiplexers, attention must
be paid to several key specs. Break-before-make
switching insures no two-channel inputs are
simultaneously connected. This prevents input
sensor damage and misoperation. Acquiring
analog input signals within a specified time and
error band are primary concerns affected by
"RON" resistance and "COUT" capacitance specifications. A low "RON" insures minimum signal
attenuation and maximum accuracy. The
"COUT" capacitance forms on R-C time constant

Analog Current Range (lA, Is) - The minimum
range of currents the switch is capable of
conducting in the ON state without degrading
ON resistance. It is measured as the value of
conduction current that does not cause more
than a doubling of the RON value for the
product grade.
Analog Input Leakage Current (IS(OFF» - The
algebraic sum of diode current losses from an
OFF-channel source input to the power supplies, ground and through the channel. Specified

IO(OFF). IS(OFF) Test Condition Definitions

10 (OFF)

NOTES
,
SWITCHES TURNED "OFF" VIA ENABLE PIN OF DECODER
LOGIC FOR ID (OFF), IS (OFF) MEASUREMENTS

2

13-3

IS (OFF) IS TESTED AT EACH INPUT (SOURCE) TERMINAL

JPMI)

ANALOG SWITCHES
MULTIPLEXERS

Precision MOllolithics Inc.

Charge Transfer (a) - Charge transfer appears
as a voltage step (pedestal) on the output capacitor after switch turn OFF. The undesirable
charge AC couples directly from the logic-control
driver to the switch contact.

as an absolute value, as the direction of current
flow is not predictable.
The
algebraic sum of diode current losses from an
OFF-channel "0" output to the power supplies,
ground and through the channel. Specified as an
absolute value, as the direction of current flow is
not predictable.
Analog Output Leakage Current (IO(OFF» -

Crosstalk (CT) - The proportionate amount of

cross-coupling from an analog input channel to
another output channel, expressed in dB.
Digital Input Capacitance (COlO) - The capacitance between a digital input and ground.

Analog Output-To-Input Capacitance (COS(OFF»

- The equivalent capacitance which shunts an
open switch effectively between "S" and "0"
output.

Insertion Loss - Insertion loss measures the
amount of signal power absorbed by the switch
ON resistance at a given measurement frequency. Insertion loss is defined in decibels as a
ratio of the output-voltage amplitude (Vo) versus the input-voltage amplitude (Vs) with a
specified load impedance.

Analog Input Capacitance (CS(ON» - The capac-

itance between an analog "S" input and ground
with the channel ON.
The capacitance between an analog "S" input and ground
with the channel OFF.

Analog Input Capacitance (CS(OFF» -

Insertion Loss (dB)

Analog Output Capacitance (CO(OFF» - The
capacitance between the analog (DRAIN) output
and ground with the channel OFF. Highfrequency transmission and output settling time
characteristics are highly influenced by this
parameter in conjunction with RON.

IVol

= 20 log I vsl

At low frequencies this equation simplifies to:
Insertion Loss (dB) = 20

10g~RL:L RON)

Logic "0" Input Current (liNd - The current
flowing into a digital input when a specified lowlevel voltage is applied to that input.

Analog Output Capacitance (CO(ON» - The
capacitance between the analog "0" output and

ground with the channel ON.

Logic "0" Input Voltage Level (VINd -

The maximum (or most-positive) digital low-level input
voltage for which proper operation of the device
is guaranteed.

The range of
analog-voltage amplitudes, with-respect-to
ground, over which the analog switch operates
(ON/OFF) within the RON and leakage specifications - IS(OFF), IO(OFF) and IO(ON) + IS(ON).
Analog Voltage Range (VA) -

Logic "1" Input Voltage Level (VINH) -

The minimum (or least-positive) digital high-level input
voltage for which proper operation of the device
is guaranteed.

Break-Before-Make Delay (tOPEN) - The
elapsed time between the turn-off of one analog
input and the subsequent turn-on of another
input as determined by the appropriate instantaneous change in the digital input code for
both inputs measured between the outputs' 50%
transition points.

Negative Voltage Supply (V-) - The most negative voltage supply with respect to ground.
Positive Voltage Supply (V+) - The most positive voltage supply with respect to ground.

Channel Capacitance (CSS(OFF). COO(OFF» -

The capacitance between the O(S) terminals of
any two channels.

13·4

ANALOG SWITCHES
MULTIPLEXERS
Precision Monolithics Inc.

ON Resistance Variation (~RON) - The variation of ON resistance produced by the specified analog input voltage change with a constant
load current.

OFF Isolation (ISO(OFF}) - The proportionate
amount of a high-frequency analog input signal
which is coupled through the channel of an OFF
device. This feedthrough is transmitted through
COS(OFF) to a load comprised of CO(OFF) in
parallel with an external load. Isolation generally
decreases by 6dB/octave with increasing
frequency.

~RON(%)=

RON @ VA = -10V - RON @ VA = +10V
RON@ VA=OV
X 100%

ON Resistance (RON) - The series ON channel resistance measured between "S" input
and "D" output terminals under specified
conditions.

ON Channel Analog Leakage Current (IO(ON) + IS(ON})
Current loss (or gain) through an ON-channel
resistance creating a voltage offset across the
device. As the direction of current flow is not
predictable, only the magnitude is specified at
various temperature ranges.

ON Resistance Match (RON Match) - The
channel-to-channel matching of ON resistance when channels are operated under identical conditions.
RON Match =

Ri - RAVG
R
AVG

x

Output Enable Delay Time OFF (tOFF(EN}) Multiplexers - The time required to disconnect
the analog output from the analog input determined by the digital address input code. it is
measured from the 50% pOint of ENABLE input
logic change to the time the output reaches
10% of the initial value.

100%

where
N = # of channels in package (i.e., for MUX-08
N = 8, for MUX-16 N = 16, etc.)
Ri = Each channel's ON resistance

Output Enable Delay Time ON (tON(EN}) Mutiplexers - The time required to connect the
analog output to the analog input determined by
the digital address input code. It is measured
from the 50% pOint of the ENABLE input logic
change to the time the output is within 90%
of final value.

1 N

RAVG=

N~

Ri
i=1

Output ON Switching Time (tON) - The time
required to connect the analog output to the
analog input. The time is measured from the
50% point of the logic input change to the
time the output reaches 90% of the final
value.

I(ON} Test Condition Definitions

Output OFF Switching Time (tOFF) - The time
required to disconnect the analog output from
the analog input. The time is measured from the
50% pOint of the logic input change to the time
the output reaches 10% of the initial value.

S2 0---+----<>"
S30---+----<>"

Outpul Settling Time (Is) - The elapsed time for
the analog output to reach its final value within a
specified error band after the corresponding
digital input code has been changed. It is
measured from the 50% point of the logic input
change to the time the output reaches final
value within specified error band.

SEQUENCE EACH

SWITCH ON
NOTES
1
2

TEST IS REPEATED FOR EACH SWITCH WITH SOURCE CONNECTED TO DRAIN
VD '" +10V AND ~10V

13-5

ANALOG SWITCHES
MULTIPLEXERS

JPMI)
Precision Monolithics Inc.

Power Supply Rejection (PSRR) - The ratio of
the change in switch contact voltage (Vo) to the
change in voltage supply (V+ or V-) that causes
it.

+PSRR (dB)

= 20 log

(tNo)
AV+

-PSRR (dB)

= 20 log

( AVO)
AV-

Switching Time (tTRAN) - Multiplexers - The
time required to switch and slew from one
analog input channel to another analog input
with a full-scale differential between inputs with
a high impedance output load. The time is
measured from the 50% pOint of the logic input
change to the time the output reaches 80% of
the final value.
Total Harmonic Distortion (THO) - The ratio of
the signal power at the fundamental frequency
to the signal power of all harmonics observed at
the switch output (Vo) with a pure sinusoid
applied to the switch input (Vs).

13·6

SW-Ol/SW-02

®IPMI)

QUAD SPST JFET ANALOG SWITCHES
(TEMPERATURE COMPENSATED RONI
Precision Monolithics Inc.

FEATURES

vs. analog input signals. The junction FET construction also
reduces static discharge destruction prevalent in CMOS
devices.

•
•
•
•
•
•
•

Low RON vs Temperature ................... 0.03%I O C
Low Absolute RON .............................. 850
Low RON Variation vs Analog Signal ............... 7%
High Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 300ns
Low Leakage Current .......................... 0.2nA
Overvoltage and Supply Loss Protected
SW-01 Is Improved Pin Compatible Device for DG201,
ADG201, LF11201
• SW-02 Is Improved Pin Compatible Device for DG202,
LF11202, IH202

Low RON sensitivity to temperature and voltage is complemented by guaranteed high-speed operation and low-leakage
currents. Logic inputs may operate directly from either
CMOS or TTL logic levels and are supply voltage independent. The SW-01/02 are protected during supply voltage
power loss and against input signal overvoltages.

PIN CONNECTIONS

ORDERING INFORMATIONt
16-PIN HERMETIC DUAL-IN-LINE PACKAGE
FUNCTION

MILITARY'

INDUSTRIAL

N.C.
N.O.

SW01BO
SW02BO

SW01FO
SW02FO

* For devices processed In total compliance to M I L-STD-a83, add /883 after

part number. Consult factory for 883 data sheet
tAil commercial and Industnal temperature range parts are available with
burn-in For ordering Information see 1986 Data Book, Section 2

16-PIN DUAL-IN-LINE PACKAGE
(Q-Package)

GENERAL DESCRIPTION
The SW-01/02 are four-channel single-pole, single-throw
analog switches which offer operating characteristics unavailable in other JFET or CMOS devices. A unique circuit
design provides a nearly constant RON over the full operating
temperature span. RON drift typically runs under 300ppm/o C.

CONTROL LOGIC
SWITCH STATE
SW-02
SW-01

LOGIC
INy

a

The SW-01/02 are pin compatible with the DG201/202. An Ion
Implanted FET switch inherently exhibits low RON variations

NOTE:

ON
OFF
INy

OFF
ON

= INPUT 1-4

SIMPLIFIED SCHEMATIC DIAGRAM (TYPICAL SWITCH)

II
DIGITAL

01

INPUT
INX

13-7

1/86, Rev. A

------------I~ SW-01/SW-02 QUAD SPST JFET ANALOG SWITCHES

ABSOLUTE MAXIMUM RATINGS (TA = 25° C, unless otherwise noted).
Operating Temperature Range
SW-01/02BQ ....................... -55°C to +125°C
SW-01/02FQ ........................ -25° C to +85° C
DICE Junction Temperature (Ti ) ....... -65°C to +150°C
Storage Temperature Range . . . . . . . . . .. -65° C to + 150° C
Power Dissipation (Q-Package) ................. 900mW
Lead Temperature (Soldering, 60 sec) ............. 300°C
Maximum Junction Temperature ................. , 150° C
V+ Supply to V- Supply ........................... , 36V
V+ Supply to Ground .............................. 36V

Logic Input Voltage .........•.. (V- or -4V) to V+ Supply
Analog Input Voltage
Continuous ....... V- Supply -25V to V+ Supply +25V
ForV+=V-=O .......................•....•... ±15V
Maximum Current Through Any Pin ............... 30mA
Peak Current,
(Pulsed at 1ms, 10% Duty Cycle) .........••...... 70mA
NOTE: Absolute ratings apply to both DICE and packaged parts, unless
otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs = ±15V, and TA = 25°C, unless otherwise noted.
SW-01/02B
PARAMETER

SYMBOL

CONDITIONS

MIN

"ON" Resistance

RON

-10V"VA ,,10V, 10 " 1rnA
(Note 1)

RON Match
Analog Voltage Range

VA

Full Temperature Range (Note 8)

ARoN vs VA

ARON

VA" 10V, 10 " lmA

+10
-10

SW-01/02F

TYP

MAX

TYP

MAX

85

100

85

120

n

4

10

4

10

%

10

%

+11
-15
7

MIN

+10
-10

+11
-15

10

UNITS

V

Analog CUrrent Range

IA

VA" 10V

5

5

rnA

Source CUrrent In "OFF" Condition

IS(OFf)

Vs

~

10V, Va

-10V

02

02

nA

Dram Current In "OFF" Condition

10(OFF)

VS

~

10V, Vo ~ -10V

0.2

0.2

nA

VS

~

±10V, (Note 2)

58

58

dB

Leakage Current In "ON" Condition

10(ON)+

~

nA

'S{ON)

"OFF" Isolation

ISO OFF

Test Figure 2

70

70

Crosstalk

CT

Test Figure 3

Turn-On-Tlme

TON

Test Figure, (Note 3)

300

400

300

400

ns

Turn-Off-Time

TOFF

Test Figure 1, (Note 3)

200

300

200

300

ns

Break-Betore-Make Time

TON-ToFF

Test Figure 1, (Note 7)

100

Source Capacitance

CS{OFF)

VA" 10V

dB

100

ns
pF

Drain Capacitance

CO(OFF)

VA" 10V

LogiC "1" Input Voltage

V,NH

Full Temperature Range (Note 8)

Logic "0" Input Voltage

VtNL

Full Temperature Range (Note 8)

55

55

pF
V

08

Logic "1" Input Current

',NH

2 :5V'N :515V

Logic "0" Input Current

I'Nl

0" V,N " 0.8V

Positive Supply Current

1+

(Note 5)

6.3

Negative Supply Current

1-

(NoteS)

32

Ground Current

IG

(NoteS)

3.0

08

V
~A

3
8.0

~A

6.3

9.0

mA

4.5

3.2

5.5

rnA

4.0

3.0

4.5

rnA

NOTES:
VA == OV, 10 ::::; 100.uA SpeCifIed as a percentage of RAVERAGE where
R

3
4

AVERAGE -

RON1 + RON2 + RON" RON4
4

The conditions listed specify the worst case leakage current. The leakage
currents apply equally to source or dram.
Sample tested
Parameter tested at TA = 125°C for military temperature range device
Power supply and ground currents specified for sWitch "ON" or "OFF"
The "OFF" state consumes highest power

TCR~

RON@TwRON@25'C Xl00,
R ON @25'CX(TH 25'C)

where

TH~125'CforBgrade
TH~85'CforFgrade

SWitching IS guaranteed to be break-before-make
Guaranteed by leakage currents and RON tests For normal operation
analog signal voltages should be restricted to less than (V+) - 4V

13-8

1/86, Rev. A

------------l~ SW-01/SW-02 QUAD SPST JFET ANALOG SWITCHES

ELECTRICAL CHARACTERISTICS at Vs
SW-01/02F, unless otherwise noted.

= ±15V, and -55°C:S TA:S + 125° C for SW-01/02B and -25°C:S TA:S 85°C for
SW-01/02B
MIN

PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

-10V" VA" 10V, ID " 1mA
(Note 1)

RON Match
RON Temperature

TYP

SW-01/02F

MAX

MIN

TYP

120

MAX

UNITS

140

n

10

15

10

15

%

003

020

003

0.15

%/oC

TC R

VA = OV, ID = 100"A, (Note 6)

Source Current in "OFF" Condition

IS(OFF)

Vs = 10V, VD = -10V, (Note 4)

10

10

nA

Drain Current in "OFF" Condition

ID(OFF)

Vs = 10V, VD = -10V, (Note 4)

10

10

nA

Leakage Current in "ON" Condition

ID(ON)+
IS(ON)

Vs = ±10V, (Notes 2, 4)

10

10

nA

Turn-On-Time

TON

Test Figure 1, (Note 3)

500

600

500

600

ns

Turn-Oil-Time

TOFF

Test Figure 1, (Note 3)

400

500

400

500

ns

Break-Before-Make Time

TON-ToFF

Test Figure 1, (Notes 3, 7)

100

LogiC "1" Input Current

IINH

2" V,N " 15V

Logic "0" Input Current

IINL

0" V,N " O.BV

Positive Supply Current

1+

(NoteS)

11

Negative Supply Current

1-

(NoteS)

6

Ground Current

IG

(NoteS)

Coefficient - Average

100

ns
"A
"A
12

mA
mA

6

mA

AVERAGE-

::J

~
Vl

RON1 + RON2 + RON3 RONO

4

~

2. The conditions listed specify the worst case leakage current. The leakage
currents apply equally to source or drain.
3. Guaranteed by design.
4. Parameter tested at TA = 125° C for military temperature range device.
5. Power supply and ground currents specified for switch "ON" or "OFF".
The "OFF" state consumes highest power.

6. TC R=

RON@TH-RON@25°C X100;
R ON @ 25°C X (TH - 25°C)

where

~

p..
......

b

NOTES:
1. VA = OV, 10= 100!,A. Specified as a percentage of RAVERAGEwhere:
R

~

::r::

~
......
~
Vl
0
0

TH=125°CforBgrade
T H = 85°C for F grade

7. Switching is guaranteed to be break-before-make.

~

III

13-9

1/86, Re'!_ A

------------1!EMD SW-01/SW-02 QUAD SPST JFET ANALOG SWITCHES
DICE CHARACTERISTICS

SW-01

SW-02

DIE SIZE 0.100 X 0.088 Inch, 8800 Iq. mill
(2.540 X 2.438 mm, 8.183 Iq. mm)

DIE SIZE 0.100 X 0.088 Inch, 8800 Iq. mill
(2.540 X 2.438 mm, 8.183
mm)

1.
2.
3.
4.
5.
8.
7.
8.

1.
2.
3.
4.
5.
8.
7.
8.

SWITCH (1) ADDRESS (IN1)
SWITCH (1) DRAIN (Dl)
SWITCH (1) SOURCE (Sl)
NEGATIVE SUPPLY
GROUND
SWITCH (4) SOURCE (S4)
SWITCH (4) DRAIN (D4)
SWITCH (4) ADDRESS (IN4)

9.
10.
11.
12.
13.
14.
15.
16.

.q.

SWITCH (3) ADDRESS (IN3)
SWITCH (3) DRAIN (D3)
SWITCH (3) SOURCE (S3)
NO CONNECTION
POSITIVE SUPPLY
SWITCH (2) SOURCE (S2)
SWITCH (2) DRAIN (D2)
SWITCH (2) ADDRESS (IN2)

SWITCH (1) ADDRESS (IN1)
SWITCH (1) DRAIN (Dl)
SWITCH (1) SOURCE (Sl)
NEGATIVE SUPPLY
GROUND
SWITCH (4) SOURCE (S4)
SWITCH (4) DRAIN (D4)
SWITCH (4) ADDRESS (IN4)

9.
10.
11.
12.
13.
14.
15.
18.

SWITCH (3) ADDRESS (IN3)
SWITCH (3) DRAIN (D3)
SWITCH (3) SOURCE (S3)
NO CONNECTION
POSITIVE SUPPLY
SWITCH (2) SOURCE (S2)
SWITCH (2) DRAIN (D2)
SWITCH (2) ADDRESS (IN2)

For additional DICE Information refer to 1986 Data Book, Section 2.

WAFER TEST LIMITS at Vs = ±15V, TA = 25° C, unless otherwise noted.
SW-01/02N

SW-01/02G

LIMIT

LIMIT

UNITS

100

120

o MAX

VA= OV.loS 100!,A

10

10

% MAX

VAS 10V, 'DS lmA

10

10

% MAX

9

mAMAX

4.5

5.5

mAMAX

4.0

4.5

mAMAX

±10

±10

VMIN

PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

-10V S VAS 10V, 'DS lmA

ARON

RON Match
ARoNvs VA
Positive Supply Current

1+

(Note 1)

Negative Supply Current

1-

(Note 1)

Ground Current

IG

Analog Voltage Range

VA

(Note2)

Logic "1" Input Voltage

V,NH

(Note2)

Lagle "0" Input Voltage

V,NL

(Note2)

Logic "0" I nput Current

ItNL

Logic "1" Input Current

I'NH

2

VMIN

0.8

VMAX

OS V,N S 0.8V

3

!'AMAX

2 S V,NS 15V

3

!'AMAX

0.8

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ±15V and TA = 25° C, unless otherwise noted.
SW-01/02N

SW-01/02G

TYPICAL

TYPICAL

90

90

0

VA = 0,1 0 = 100!,A

0,03

0.03

%I'C

PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

-55'C S TAS 125'C

RON Temperature Coefficient

TC R

UNITS

Turn-On-Tlme

TON

RL = lk, CL = 13pF

300

300

ns

Turn-Off-Time

TOFF

R L = lk. C L = 13pF

200

200

ns

Drain Current in "OFF"
Condition

10(OFFI

Vs = 10V. Vo

0.2

0.2

nA

"OFF" Isolation

ISO OFF

f =500kHz. RL =6800

58

58

dB

Crosstalk

CT

f

=500kHz. RL = 6800

70

70

dB

=-10V

NOTES:
1. Power supply and ground current specified for switch "ON" or "OFF".

2. Guaranteed by RON and leakage current measurements.

13-10

1/86, Rev. A

------------t!EHD SW-01/SW-02 QUAD SPST JFET ANALOG SWITCHES
TEST CIRCUITS

APPLICATIONS INFORMATION
This analog switch employs ion-implanted JFETs in a switch
configuration designed to assure break-before-make action.
The turn-off time is much faster than the turn-on time to
guarantee this feature over the full operating temperature
and input voltage range. Fabricated with Bipolar-JFET processing rather than CMOS, special handling is not necessary
to prevent damage to these switches. Because the digital
Inputs only require a 2V logic "1" Input level, powerconsuming pull-up resistors are not required for TTL compatibility to insure break-before-make switching as is most
often the case with CMOS switches. The digital inputs utilize
PNP input transistors where input current is maximum at the
logic "0" level and drops to that of a reverse-biased diode
(about 10nA) as the Input voltage Is raised above -1.4V.

LOGIC 0 • SN ON

LOGIC
INPUT
tr <.200'

3.0V

'0%

tf'" 20nl

SWITCH
INPUT

SWITCH
OUTPUT

+15V

Vo • Vs RL +

SWITCH

~~s(on)

SWITCH

INPUT

8,

Vs • -6V

O-I-------......---1~ Vo

OUTPUT

GNO

The "ON" resistance, RON, of the analog switches Is constant
over the wide input voltage range of -15V to +11V with
VSUPPLY = ± 15V. For normal operation, however, positive
input voltages should be restricted to 11V (or 4V less than the
positive supply). This assures that the Vas of an OFF switch
remains greater than its Vp, and prevents that channel from
being falsely turned ON. Individual switches are "ON" without power applied.

-15V
(REPEAT TEST FOR IN2. INa AND IN4)

INVERT LOGIC INPUT FOR SW-02

Proper switching requires the "Source" terminal to be connected to the input driving signal.

TEST FIGURE 1

PROGRAMMABLE ATTENUATOR (1 to 0.0001)
VA'" 5VRMS
AT f = 500kHz

Your

1."MSl

RL·8800

101kil
14

"OFF" ISOLATION· 20 LOG

[."~~T]

10kSl

910n

11

13
2.7,10,15

SW-01

V+

V-

TEST FIGURE 2

VA'" 5V RMS
AT f = 600kHz

VD'
(- 1 ATTENUATION WHEN ALL SWITCHES OFF)

CL'" 7pF
VD2

TEST FIGURE 3

13-11

1/86, Rev. A

--------------i~ SW-01/SW-02 QUAD SPST JFET ANALOG SWITCHES

TYPICAL PERFORMANCE CHARACTERISTICS

(SW-01/02)
OVERVOLTAGE
CHARACTERISTIC

POWER SUPPLY
LOSS CHARACTERISTIC
150

4.0

90

V+ '" +15.0V

Vi -'i.DV
=

3.0

100

-'n
I

:5

I

1'01102

2.0

1

"ON" RESISTANCE VI
ANALOG VOLTAGE (VA)

10 _

_VA

1/

Rl=oo

r

I

S

-20 -10

10

: VA

30

40

75
V+ '" +15V

v-

RL =

-5.0

-10.0

00

SWITCHING TIMES VI
TEMPERATURE

--

200

-

180

+10

RON VI TEMPERATURE

..-

~~V

250

150

+5

ANALOG INPUT VOLTAGE - VA(VOL TS)

130

300

200

-5

-10

350

220

1

65

100

SWITCHING TIME VI
ANALOG VOLTAGE

TON

= -15V
TA = 25"C _
IS = 1rnA

70

5.0

ANALOG INPUT VOLTAGE (VOLTS)

~

80

~

Fl

-50

-100

20

g

SW-Ot/02

./

.......

240

V

~

-

85

/

50

ANALOG INPUT VOLTAGE - VA (VOLTS)

- ---

260

=.l

-150

-10
-40 -30

280

V+ = V-

/"

V

...--V

120
110

1--

,./

,./v"OFF

g

100

z 90
a:

,.........

0

80

"

RON = 85n @25°C

70
100

60

TOFF

160

-10

-5

10

50
-75

40
-50

-25

0

+25

+50

+75 +100 +125

-55 -35 -15

TEMPERATURE (QC)

ANALOG INPUT VOL lAGE - VA (VOLTS)

CROSSTALK AND "OFF"
ISOLATION VI FREQUENCY
100

5
25
45
65
TEMPERATURE fOC)

85

105 125

SWITCH CAPACITANCE VI
ANALOG VOLTAGE

LEAKAGE CURRENT VI
TEMPERATURE
16

10,000

Cs (ON)

14

90
12
80

i"-...

10 (ON)

1'-,
70

60

"'"f'..

I

40
tOOk

CROSSTALK ' -

"N

1M

10

1,000

I

Nil

FREQUENCY (Hz)

(ON)

ISI~lh

I

Cs (OFF)

10 (OFF)

e- -

,/11'"

OFF1ISJAJION
50

+ IS

II
10M

100
-60

V

L rL /
./

~ 1/

-20

20

60

TEMPERATURE 1°C)

13-12

100

140

-10

-

riD (iFF)
-6

V

,.......
/

V

l-

-2

'0

ANALOG INPUT VOLTAGE -VA (VOLTS)

1/86, Rev. A

----------~!EHD SW-01/SW-02 QUAD SPST JFET ANALOG SWITCHES

The SW-01/02 designs have been optimized for low "ON"
resistance variation with temperature, signal voltage, and
supply voltage changes. Fast switching response and low
leakage currents at high temperature are also key performance improvements over older circuit designs.

and analog current near ideal signal transfer accuracy is
obtainable.
Applications needing very high analog current capability
(>SmA) or where the switch is driven from high source
impedances (>100.0.) should use the SW-201 (Pin Compatible
to SW-01) or the SW-202 (Pin Compatible to SW-02) highcurrent quad switches.

The static-electricity-resistant JFET switches and additional
overvoltage-protection circuitry make the precision switches
extremely durable in most application environments.

Although the SW-201/202 do not offer the same "ON"
resistance temperature coefficient, many other premium
characteristics are similar. In addition, the SW-201/202 offer
exceptionally low signal distortion over a wide signal voltage
and~equency range.

The SW-01/02 are well suited to applications requiring
analog currents --i---,--t---

"(;g

45

I--

ANALOG VOLTAGE (VOLTS)

V+ '" 16V
V-" -16V

"...
~

25

ilw-40

~',

-0.6

V+

a:

\\

\

---

'.

'-

~

I

IS(OFF)

-0.4

/

20

1

I~\,I

25"C

5

LEAKAGE CURRENT AT 125°C
VB ANALOG VOLTAGE

IO(OFF)

-0.2

SLOPE" 1/ RON

....... I---'"

TEMPERATURE

0.2

LOGIC INPUT CURRENT
VI VOLTAGE

15a:

-55 -35 -15

LEAKAGE CURRENT

V SD VOLTAGE ACROSS SWITCH (VOLTS)

!...

o

VI ANALOG VOLTAGE

V

-20
-2

I---'" V

ANALOG VOLTAGE (VOLTS)

L

/
II

V

20

±SUPPLY VOLTAGE (VOLTS)

SWITCH CURRENT
vsVOLTAGE

40

0

20

°0~--~--~--~--1~2-~1.~~,8

..... ......

SO

;

TA=2SOC

0

ii

VB

II

;;,:
,g

...

~

6

a:

::>

-200
-300

"~ 4

)

-100

a:

./

1+

il

./

r-

-400
-16

-10
-5
10
lOGIC INPurVOLTAGE (VOLTS)

0

"

-

..

-3'

-1.

.-

•

2.
45
TEMPERATURE

13-18

6.

fe)

8. 10.

12.

.+

.°0~--~--~--~---'~2---+'15~~'8
±SUPPLY VOLTAGE (VOL TSj

1/86, Rev. A

--------------1~ SW·OS DUAL SPST JFET ANALOG SWITCH

TYPICAL PERFORMANCE CHARACTERISTICS

SWITCHING RESPONSE

VB

SWITCHING TIME
TEMPERATURE

SWITCHING TIME
ANALOG VOLTAGE

VB
600

600

V+ = 15V

V+"' 15V

v-

V-" -15V
TA=2IfC
RL" lkn

500

60 0

CL.·30pF
VA --5V

.
w

I200

~

IA

r- V

t!l 300

z

1i!:

t'OFF

-I'

(SEE SWITCHING TIME TEST CIRCUIT)

r- I- r-

t'OFF MEASURED TO 90%
OF Vo (SEE TEST CKT)

~11 -9 -7

tON

5 -3 -1

~ 30 0

i

........

100

INSERTION LOSS
VB FREQUENCY

....- --::::

0

~~

>=

= -15V
VA = -5V
RL = lk.l1
CL = 30pF

1

--

-

tOFF

3

5

r=

~ 200

-

......
7

:;...-:::::: :::::::

~

.:;::; ~ r-

tOFF

100

9

11

0
-55 -35 -15

5

25

45

65

85

ANALOG VOLTAGE (VOLTS)

TEMPERATURE (oC)

OFF ISOLATION AND
CROSSTALK VB FREQUENCY

DISTORTION VB
FREQUENCY

105 125

100

~ .....

a;

"::l

80

i

-2
-4

.......

UllWL

.......

~~

u 60

o

~

mrSOLATION

1l

-6

z

-8

RL = lkf2
v+ 15V
v- = -15V

=

~
tt

20

o

,

~A,', ~~~~

'Ok

40

5

VA=lVRMS

-'0

2

'M

TOOk

FREQUENCY

'OM

RL' 680nlill
VIN = lVRMS
V+ = 15V
v- = -1SV
TA=25°C

o

100M

TOOk

'Ok

(Hz)

POWER SUPPLY REJECTION
VB FREQUENCY
'00
90

"

I'-

80
70

'\

60

30

V+ = 15V

IIII I

28

~~SRR

24

I\.

..s-

20
16

RL = lkn
VIN = lVRMS
SWITCH CLOSED

12

LOG!~~

-PSRA '" 20

LOG:~~

o
'00

lk

10k
FREQUENCY (Hz)

I- v-· -,5V
=

TA 2SOC
f lMHz

=

v+ = 15V
v- = -15V

+PSRR = 20

20

----

CS(ON)

CS{OFF)

-

TOOk

OVERVOLTAGE
CHARACTERISTIC

32

llU

TA = 25°C

40

'0

CAPACITANCE VB
ANALOG VOLTAGE

-PSRR

50

10M

1M

FREQUENCY (Hz)

'M

o-11

-

:,.....V
>-f- ~OIOFFI

-9 -7 -5 -3 -1
1
3
5
7
ANALOG VOLTAGE (VOLTS)

13·19

9

11

-10

0

10

20

30

40

ANALOG VOL TAGE (VOLTS)

1/86, Rev. A

-------------Im

SW·05 DUAL SPST JFET ANALOG SWITCH

TYPICAL PERFORMANCE CHARACTERISTICS (Single Supply Operation)
"ON" RESISTANCE

SWITCHING TIME vs
ANALOG VOLTAGE

vs SUPPLY VOLTAGE

'00

,-'--r--r-....,-....,-....,---,
v- ""

80

g
w

"z

500

RL = 4.7kn

~L='3PF
v+ = 2av

400

w

">=

~ 300

;:

::

40

~ 200

z

v- '" ov
ov

)

GND '"

.5.
60

~

~

0.4

600

OV
GND = OV
TA = 25°C
IS = lmA +-+-+_+_+~

'oN

\
\

'DO

12

16

20

24

00

28

V+ SUPPLY VOLTAGE (VOLTS)

8

12

16

20

24

~ i;:".'S(OFF)

-'"

-

t'OFF
20

0

IOIOFFI

V
lolON)

0

0

LEAKAGE CURRENT
vs ANALOG VOLTAGE

V+" 2av

v-

-0.4

-0.6 0

28

=OV
GND =

ov

TA'" 25°C
FOR IO(OFF). Vs = lOV
FOR IS(OFF). Vo = lOV

8

12

16

20

24

28

ANALOG VOL rAGE (VOL 15)

ANALOG VOLTAGE (VOLTS)

TEST CIRCUITS
OFF ISOLATION TEST CIRCUIT

CROSSTALK TEST CIRCUIT
VA = 5VRMS
AT f = 500kHz

NCo-~S'+-o~A-~rD~'~~~___V~D'
D
A--~--~--1r---oVOUT

50n

RL

"OFF" ISOLATION = 20 LOG

.

[~]
Your

50n
CROSSTALK = 20 LOG

['!.A.]
VOl

SWITCHING TIME TEST CIRCUIT

LOGIC
+15V

Vo = Vs Rl +

SWITCH
INPUT

INPUT

~ (ON)
SWITCH
OUTPUT

S,

~=~VO-+-----ai A-~-o-,--~>-oVo

tr

<

20n5

tf

<

20n5

SWITCH "ON" FOR LOGIC 0 , -_ __

3.0V
50%

50%

SWITCH

INPUT

RL

'kn

SWITCH
OUTPUT

SWITCH OUTPUT WAVEFORM SHOWN FOR Vs '" CONSTANT
WITH LOGIC INPUT WAVEFORM AS SHOWN.
Va IS THE STEADY STATE OUTPUT WITH SWITCH ON.

REPEAT TEST FOR IN2

13-20

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _ _ _-;~ SW-05 DUAL SPST JFET ANALOG SWITCH

APPLICATIONS INFORMATION
The SW-05 provides a rugged J FET alternative to the industry
standard CMOS generic DG200 device. The basic differences in process (CMOS vs Bipolar-JFET) effecting switch
characteristics result from the parallel connected enhancement mode FETs used in CMOS versus the single depletion
mode JFET used by the Bipolar-JFET process. The junction
technology is far less susceptible to electrostatic damage
(ESD), and offers a higher resistance to radiation exposure.
No extensive threshold shifts take place as commonly found
in CMOS.

CMOS logic. Since the SW-05 incorporates a standard two
forward diode drop logic voltage reference, pin 7 in the metal
can and pin 12 in the DIP package were left unconnected.
This allows direct plug-in compatibility with DG200's requirmg external logic threshold adjustment to their VREF pin
when operating from supplies other than ± 15 volts. No logic
threshold adjustment is necessary with the SW-05 operating,
for example, at ± 12 volts.
The addition of a 7474 latch in front of the SW-05 (Figure 1)
results in a pulse input latching SPDT analog switch. A
positive edge of an input pulse to the TTL D type flip flop
causes the Q andQ outputs to change state. Taking advantage of the complementary outputs turns the SW-05 into a
break-before-make SPDT analog switch. The short dead time
between switch closures prevents damaging current flowing
between the two ~ow impedance sources. The photograph
illustrates the dead time when VOUT IS pulled to ground by the
1k ohm termination resistor.

The basic JFET switch design inherently results in a more
linear "ON" resistance overthe designed analog signal range
of -15 to +11 volts. The "ON" resistance is independent of
analog voltage and supply voltage, but does have a positive
temperature coefficient of O.4%/ o C. Leakage currents stay in
the low picoamps at room temperature providing very high
"OFF" ~esistance characteristics.
The logic control inputs are TTL input compatible with full
400mV noise immunity over the full operating temperature
range. The PNP input structure requires very little logic drive
current resulting in minimum output loading to both TTL and

The initialize input, connected to the D flip flop reset (clear)
input, resets Q, sending an active low to the IN, terminal of
the SW-05 closmg switch 1.

FIGURE 1: Pulse Input Latching SPOT Analog Switch

+1SV

"'V
PULSE

'4

INPUT

n

I.
V+
CHANNEL 1

Vee

,.---=.;S'¥4'--_ _ _OO+5V
SOURCE 1

f '-----4----'t.:!!!.,

01 5

3 elK1

0' 5
D FLIP FLOP
INITIALIZE 1

1/27474

1---.-oOVOUT

SW.o5
CHANNEL 2

''''

ll'lf6-......- - 4 - -.......~,
2 01

GNO

'---=1"-----0 +2V

II

SOURCE 2

PIN OUT FOR
METAL CAN PACKAGE
SHOWN

-15V

13-21

1/86, Rev. A

SW-06
QUAD SPST JFET
ANALOG SWITCH
Precision MOIlOlithics Inc.

FEATURES

GENERAL DESCRIPTION

• Two Normally Open and Two Normally Closed SPST
Switches with Disable
• Switches can be Easily Configured as a Dual SPDT or
a DPDT
• Highly Resistant to Static Discharge Destruction
• Higher Resistance to Radiation Than Analog Switches
Designed with MOS Devices
• Guaranteed RON Matching ...•....••...•...•. 10% Max
• Guaranteed Switching Speeds ....••. TON = 500ns Max
TOFF = 400ns Max
• Guaranteed Break-Before-Make Switching
• Low "ON" Resistance ...•.......•.......•.•.. 80n Max
• Low RON Variation from Analog Input Voltage ....... 5%
• Low Total Harmonic Distortion .....•......•...•. 0.01%
• Low Leakage Currents at High Temperature:
TA = 125° C ................. 100nA Max
TA = 85°C ................... 30nA Max
• Digital Inputs TT"l/CMOS Compatible and Independent
ofV+
• Improved Specifications and Pin Compatible to LF-11333/
13333
• Dual or Single Power Supply Operation

The SW-06 is a four channel single-pole, single-throw analog
switch that employs both bipolar and ion-implanted FET
devices. The SW-06 FET switches use bipolar digital logic
inputs which are more resistant to static electricity than
CMOS devices. Ruggedness and reliability are inherent in
the SW-06 design and construction technology.
Increased reliability is complemented by excellent electrical
specifications. Potential error sources are reduced by minimizing "ON" resistance and controlling leakage currents at
high temperatures. The switching FET exhibits minimal RON
variation over a 20V analog signal range and with power
supply voltage changes. Operation from a single positive
power supply voltage is possible. With V+ = 36V, V- = OV, the
analog signal range will extend from ground to +32V.
PNP logic inputs are TTL and CMOS compatible to allow the
SW-06 to upgrade existing designs. The logic "0" and logic
"1" input currents are at micro-ampere levels reducing loading on CMOS and TTL logic.

FUNCTIONAL DIAGRAM
V'

PIN CONNECTIONS
SWitches are shown

12

3

the logic "0"
mput state and DIS = "1"
In

IN 1

~Sl

O-'----jr-,

t.2...o01

TOP VIEW
LEVEL

SHIFT

IN 2

6

_~S2

0--4---<..-....

t.2....a

16-PIN DUAL-IN-LINE
PACKAGE
(Q OR P Package)

SW-06BRC/883
LCCPACKAGE
(RC-Sufflx)

IN 3

O-:'---i-ar--....

IN 4

0-:.;"'-1-<11..-....

-[~

t

15

ORDERING INFORMATIONt
PACKAGE IS
16-PIN DIP
HERMETIC
EPOXY

ORDER
PART
NUMBER
SW06BO'
SW06FO
SW06GO
SW06GP

DIS

LCC
SW06BRC/BB3

OPERATING
TEMPERATURE
RANGE

13

GND

02

S4

D4

V-

TRUTH TABLE

MIL
INO
COM
COM

SWITCH STATE

'For devices processed In total compliance to MIL-STD-BB3, add IBB3 after
part number Consult factory for BB3 data sheet
t All commercial and Industrial temperature range parts are available with
burn-In For ordering information see 19B6 Data Book, Section 2

13-22

DISABLE
INPUT

LOGIC
INPUT

CHANNELS
1&2

CHANNELS
3&4

0
10rNC
10rNC

X
0
1

OFF
OFF
ON

OFF
ON
OFF

1/86, Rev. A

---------------1~ SW-06 QUAD SPST JFET ANALOG SWITCH

ABSOLUTE MAXIMUM RATINGS (Note 1)
Operating Temperature Range
SW-06BQ, BRC ...................... -55°C to +125°C
SW-06FQ ............................ -25°C to +85°C
SW-06GP .............................. O°C to +70°C
Storage Temperature Range ........... -65°C to +150°C
Power Dissipation (Note 2)
Q Package ................................... 900mW
P Pac kage ................................... 500mW
Lead Temperature (Soldering 60 sec) ............. 300° C
Maximum Junction Temperature .................. 150°C
ELECTRICAL CHARACTERISTICS at V+

V+ Supply to V- Supply ............................ 36V
V+ Supply to Ground .............................. 36V
Logic Input Voltage ............ (-4V or V-I to V+ Supply
Analog Input Voltage Range
Continuous ............. V- Supply to V+ Supply +20V
Maximum Current Through
Any Pin Including Switch ....................... 30mA
NOTES:
Absolute ratings apply to both DICE and packaged parts. unless otherwise
noted.
2.

a Package derated 12mW/' C above 75' C, P Package derated 10mW/' C
above 25'C.

= 15V, V- = -15V and T A = 25° C,

unless otherwise noted.

SW-OS8
PARAMETER

SYMBOL

"ON" Resistance
RON Match Between
SWitches

RON Match

MAX

Vs = OV, Is= lmA

60

Vs = ±10V, Is = 1mA

65

IS = lmA (Note 8)
Is= lmA

Analog Current Range

VS =±10V

Applied

Source Current In
"OFF" Condltton

"OFF" Condition

Source Current

In

"ON" Condition

IO(OFF)

(Note 5)

Logical "1" Input

Full Temperature Range

Voltage

(Notes 6, 8)

Voltage

VIN

V1N

Logical "0" Input

Time

65

100

100

150

20

20
+10

+11

-10

-15

10

15

12

See SWitching Time
Test Circuit (Notes 6,9)
(Note 9)

mA

20

10

20

%

03

20

03

20

03

10

nA

03

20

03

20

03

10

nA

03

20

03

20

03

10

nA

20

v

20

08

08

v

10

0 8V

tON-tOFF

%

10

08

Test Circuit (Notes 6, 9)
tOFF

n

v

10

= 2 OV to 15 OV
=

UNITS

15

20

See SWitching Time

Turn-On-Time

8reak-8efore-Make

80

(Note 4)

Current

Turn-Oft-Time

150

-15

(Notes 6, 8)

Logical "1" Input

100

+11

Full Temperature Range

Logical "0" Input

100

-10

= Vo = ±lQV

Vs

MAX

60

+10

(Note 5)

IS{ON)+

TYP

80

-15

(Note 5)

IO(ON)

MIN

+11

Vs = 10V, Vo = -10V

Drain Current In

MAX

-10

Vs = 10V, Vo = -10V
IS(OFF)

SW-OSG

TYP

+10

-lOV:S Vs 'S::. lOV, Is = 1 OmA

Voltage

MIN

10

Vs = OV, Is = 100"A (Note 1)

Analog Voltage Range

~RON vs

MIN

SW-OSF

TYP

CONDITIONS

50

15

50

15

50

15

100

340

500

340

600

340

700

ns

200

400

200

400

200

500

ns

140

50

140

50

140

ns

Source Capacitance

CSIO FF)

Vs = OV (Note 5)

70

70

70

pF

Drain Capacitance

CD (OFF)

Vs = OV (Note 5)

55

55

55

pF

Channel "ON"

CD(ON)+

Vs = Vo = OV (Note 5)

15

15

15

pF

Vs = 5V RMS, RL = 6aOn,
C L = 7pF, f = 500kHz (Note 5)

58

58

58

dB

70

70

70

dB

Capacitance
"OFF" Isolation

CS{ON)
ISO (OFF)

Vs = 5V RMS, RL = 680n,

Crosstalk
PoSItive Supply
Current
Negative Supply
Current
Ground Current

C L = 7pF, f = 500kHz (Note 5)
1+

All Channels "OFF",
DIS = I l f ) (Note 5)

50

60

50

90

60

90

mA

1-

All Channels "OFF",
DIS = «0)) (Note 5)

30

50

40

70

40

70

mA

30

40

30

40

30

50

mA

All Channels "ON" or
"OFF" (Note 5)

13-23

1/86, Rev. A

-------------I~ SW-06 QUAD SPST JFET ANALOG SWITCH

ELECTRICAL CHARACTERISTICS at V+= 15V, V-=-15V, -55°C:O; TA:O;+125°C forSW-06BQ, -25°C:o; TA:O; +85°C for
SW-OSFQ and 0° C :0; TA

:0;

70° C for SW-OSGP, unless otherwise noted.
SW-06F

SW-068
PARAMETER

SYMBOL

CONDITIONS

Temperature Range

TA

Operatrng
~

Vs

"ON" Resistance

RON

RON Match Between

OV, IS ~ 10mA

Vs = ±10V, IS
~

OV, IS ~

= 10mA
100~A

Vs

Analog Voltage Range

VA

Is ~ 1 OmA (Note 8)
Is~ 1 OmA

Analog Current Range

IA

aA ON With Applied

Is~

Source Current In
IS(OFF)

"OFF" Condition

+10
-10

1 OmA
10V, Vo

~

~

MIN

125

-25

TYP

SW-06G
MAX

MIN

TYP

85

MAX

UNITS

70

'C

110

75

125

75

175

80

110

80

125

80

175

25

10

%

+11
-15

V

20

-10V:::;Vs .5+10V,

Vs

MAX

75

(Note 1)

Vs~±100V

.:1RON

Voltage

TYP

-55

RON Match

SWitches

MIN

+11
-15

+10
-10

+11
-15

+10
-10

n

12

11

11

mA

10

12

15

%

-10V,
60

30

60

nA

60

30

60

nA

100

30

60

nA

V,N ~ 2 OV to 15.0V
(Note 4)

10

10

15

~A

V,N

10

10

15

~A

TA = Max Operating Temp

(Notes 5, 7)
Vs

Drain Current In

10(OFF)

"OFF" Condition

10V, Vo

~

~

-lOV,

TA = Max Operating Temp

(Notes 5, 7)
Vs

Leakage Current In

IS(ON)+

"ON" Condition

ID(ON)

Logical "1" Input
IrNH

Current
Logical "0" Input

'INL

Current

~

Vo

~

±10V,

TA = Max Operating Temp

(Notes 5, 7)

~

0 8V

See SWitching Time

Turn-On-Time

tON

Turn-Off-Time

tOFF

Test CirCUit (Notes 2,6)
See SWltchrng Time

Break-Before-Make
tON-tOFF

TIme

Test CircUIt (Notes 2, 6)

(Note 3)

440

900

500

900

1000

ns

300

500

330

500

500

ns

70

70

50

ns

PositIve Supply Current

1+

All Channels "OFF" (Note 5)

90

13.5

135

mA

Negative Supply Current

1-

All Channels "OFF" (Note 5)

75

10.5

105

mA

6.0

75

75

mA

Ground Current

All Channels "ON" or "OFF"

IG

(Note 5)

NOTES:
1. Vs = OV, Is = 100",A Specified as a percentage of RAVERAGE where

A
2.

3.

- RONl + RON2 + RON3 + RON4

AVERAGE -

4

Guaranteed by design.
SWitch IS guaranteed to provide break-before-make operation
Current tested at V = 2.0V. ThiS IS worst case condition
SWitch being tested ON or OFF as indicated, V ,NH ~ 2 av or V ,NL ~ a.BV,
per logic truth table
Also applies to disable Pin.
Parameter tested only at TA = +125°C for military grade device.
Guaranteed by RON and leakage tests For normal operation maximum
analog Signal voltages should be restricted to less than (V+) -4V.
Sample tested

,N

5.
6.
7.

B.
9

13-24

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _ _ _-j~ SW-06 QUAD SPST JFET ANALOG SWITCH

DICE CHARACTERISTICS
1. IN (1)
2. 0 (1)

3. S (1)
4. GND
5. V- (SUBSTRATE)
6. S (2)
7.0(2)
8. IN (2)
9. IN (3)
10. 0 (3)
11. S (3)
12. V+
13. DISABLE
14. S (4)
15. 0 (4)
16. IN (4)

DIE SIZE 0.100 X 0.096 inch, 9600 sq. mils
(2.540 X 2.438 mm, 6.193 sq. mm)

WAFER TEST LIMITS at V+

For additional DICE information refer to
1986 Data Book, Section 2.

=15V, V- = -15V, TA =25°C, unless otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

-IOV"VA " 10V, Is" ImA

SW-06N

SW-06G

LIMIT

LIMIT

UNITS

80

100

nMAX
%MAX

VA = OV, Is" 100~A

IS

20

aRONvsvA

aR oN

-IOV" VA" 10V, Is" ImA

10

20

% MAX

Positive Supply Current

1+

(Note I)

60

9.0

mAMAX

Negative Supply Current

1-

(Note I)

50

70

mAMAX

Ground Current

IG

(Note I)

40

4.0

mAMAX

Analog Voltage Range

VA

Is=lmA

±IOO

±IO.O

VMIN

LogiC "1" Input Voltage

V 1NH

(Note 3)

20

2.0

VMIN

Log'c "0" Input Voltage

V 1Nl

(Note 3)

08

0.8

V MAX

LogiC "0" Input Current

IINL

OV" V'N" 0 8V

50

50

~AMAX

Logic "1" Input CUrrent

IINH

2 OV" V'N" 15V (Note 2)

Analog Current Range

IA

Vs= ±IOV

RON Match Between Switches RON Match

~AMAX

10

mAMIN

NOTE:

Electrical tests are performed at wafer probe to the limits shown. Due to vanatlons In assembly methods and normal Yield loss, Yield after packaging is not
guaranteed for standard product dice. Consult factory to negotl8te specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at V+ = 15V, V- = -15V and

TA

= 25° C, unless otherwise noted.

SW-OSN

SW-06G

TYPICAL

TYPICAL

60

60

!l

tON

340

340

ns

tOFF

200

200

ns

0.3

nA

PARAMETER

SYMBOL

CONDITIONS

"ON" ReSistance

RON

-IOV" VA" 10V, Is" ImA

Turn-On-Time
Turn-Off-Time

Dram Current In

UNITS

ID(OFF)

Vs = 10V, V D = -IOV

03

"OFF" Isolation

ISO(OFF)

f = 500kHz, RL = 680n

58

58

dB

Crosstalk

CT

f = 500kHz, RL = 680n

70

70

dB

"OFF" Condition

NOTES:
1 Power supply and ground current specified for sWitch "ON" or "OFF"
2 Current tested at VIN = 2 OV This IS worst case condition

Guaranteed by RON and leakage tests

13-25

1/86, Rev. A

--------------1~ SW-06 QUAD SPST JFET ANALOG SWITCH

TYPICAL PERFORMANCE CHARACTERISTICS
"ON" RESISTANCE VI
POWER SUPPLY VOLTAGE
120

"ON" RESISTANCE vs
ANALOG VOLTAGE

! 25'e +---+--+--+--+---l

110

TA

100

~~==_~~~

+-+-+--+--+---1

90

z 60

ti

50
40

v+ ~ 15V

v- = -15V

70

IS= lmA

80

§

75

V+ '" +lSV
v- '" -15V
70 rTA'25'C - - - r - - - + - - - j
65~-_~

70 r-~~-+---+--~--~RON --

RON vs TEMPERATURE

75r-----~-----r----_r-----,

_ _~~_ _+-_ _- ;

I--

~ 60f=::~=t~~-lr-~--~~:::j

65

~ 55 ~--~--~~--+-----;

g 60
is
0: 55

50~--~--~~--+----;

50

45~--~--~---+----;

45

30

20

Z

... V
/~

~

V
V

±13

±14

±15

±16

±17

±18

±19

-10

___

_ _-"

~

__

~

40L-_ _
±12

~

10
-5

40
-55

10

0

-25

ANALOG INPUT VOLTAGE-VA (VOL TS)

POWER SUPPLY VOL rAGE (VOL TSI

LEAKAGE CURRENT
vs ANALOG VOLTAGE

SWITCH CURRENT
vs VOLTAGE
20

25

50

75

100

125

TEMPERATURE (OC)

LEAKAGE CURRENT
vs TEMPERATURE

-100

lOOnA
c_SUPPLY"=±15V

TA

<1

.§
>-

r--

10

V+ = 15VJ
V-=-lSV
TA=2SoC

--

V

iii

~

1/

~

"
~

/

~

~

>-

0

.......-

1

/

8:;' :r-25'~t-

IL

~

~

"w
""
"~

~~~~~~~~;~~~~~-rJDE

/

I'

t----LESS THAN 10 PI CO AMPS

-20
-2

TA =

TA

,

_Vl

!Z=

Y

II

-10

125::2

-00 1
-15

-1

V SD VOLTAGE ACROSS SWITCH

.X

-10

10pA

10

-5

-55

15

-25

SUPPLY CURRENT vs
TEMPERATURE

SUPPLY CURRENT vs
SUPPLY VOLTAGE

10

0

50

25

75

100

125

TEMPERATURE (OC)

ANALOG VOLTAGE {VOLTS}

(VOLTS)

SWITCH CAPACITANCE
vs ANALOG VOLTAGE
2

10
V+ '" 15V

TA

v- '" -15V

J25"C

v+
2.

DIS= '0'

olS= '0'

t-

f=lMHz

= 15V

V-=-15V

TA" 25°C

24

~

r- t- +-

1+

r- t--

1

~

f--+--

-55 -35 -15

20

w

1+

t-,...

~

J""..
25

45

65

TEMPERATURE (OC)

85

...

105

--

,...-

l-:'""

.-

±SUPPLY VOLTAGE (VOLTS)

13-26

16

~

12

~

s

I"- r- ~

C (ON)

s

C (OFF)

•

~

~

7

9

Co (OFF)

12

125

"~

15

,.

i i
-11 -9 -7 -5

-3 -1

1

3

5

11

ANALOG VOLTAGE (VOLTS)

1/86, Rev. A

-------------IrfHD

SW-06 QUAD SPST JFET ANALOG SWITCH

TYPICAL PERFORMANCE CHARACTERISTICS
TON/ToFF
SWITCHING RESPONSE

V5

SWITCHING TIME
vs TEMPERATURE

SWITCHING TIME
ANALOG VOLTAGE

400

350

600
Vs

Vs = ±15V
RL = lkS2
CL = 10pF
TA=2S0C -

'-...

-0.5

300

300

-2
VO(V)

-4

tI..

250

-6

200

'50
-'0

'" --

Jy

400

I'-- ~

VO(V)

=±lSV

Rl = lkn
CL = 10pF

200

'00

TOFF

V
........V

........

--

o

-5

-55

-25

25

-2

..

80

"~

70

iij:

60

:)

75

100

125

TOTAL HARMONIC
DISTORTION

'0

'00
~

50

TEMPERATURE (OC)

CROSSTALK AND "OFF"
ISOLATION V5 FREQUENCY

"
~
g

~

...-V

~

ANALOG INPUT VOLTAGE (VOLTS)

INSERTION LOSS
V5 FREQUENCY

TL

~
~

TA = 25°C

II

v+

~ V+

= 15V

f--

+15V
l5V

VA = lV rms
TA = 25°C

%~4

'0

(;t-

0
~4~1
10,\<
18

V

:==

V-=-lSV

I~~ -lOon

r--

""-

0'

iil

~SO

........

o

40
'Ok

lOOk

'M

'M

lOOk

100M

'OM

AL = lkD.

r-....

UII

00'

'OM

lk

'00

FREQUENCY (Hz)

10k

100k

FREQUENCY (Hz)

FREQUENCY (Hz)

POWER SUPPLY REJECTION
V5 FREQUENCY

OVERVOLTAGE
CHARACTERISTICS

90~~-r~mr-rnT~-rnT"~""~~~

~H"Idl:ttIt--t-H-t1

+PSRR = 20 log

~~~

701--f-+H-fl'IfI':--"rt..H-H

-PSRR '" 20 log

~~~

80

\

1111111

6°I--f-+Hfflrr-~+*mk-rHt~-+++~

I" +~~~~

Hr-H-I-ftlI+---I-t1-fl1lIt--H'>litltJ--l+t-I-ftlII

50

TA'" 25°C

40

V+=15V

30

V-= -lSV

i'-PSRR

20
_,~

'k

'Ok

lOOk

'M

-20

'OM

__

~

-10

__

~

__- L__

'0

~

__

20

~

__

30

~

40

ANALOG VOLTAGE-VA (VOLTS)

FREQUENCY (Hz)

13-27

1/86, Rev. A

-------------I~ SW-06 QUAD SPST JFET ANALOG SWITCH

TYPICAL PERFORMANCE CHARACTERISTICS (OPERATING SINGLE SUPPLY)
"ON" RESISTANCE

LEAKAGE CURRENT

VI ANALOG VOLTAGE

vs VANALOG

SUPPLY CURRENT
vs SUPPLY VOLTAGE
10

01

rr-

1>-

1201--l-+-+-+---l-+--I

~

l00t-~--r-1--r-~--r-~

13
t!
;;2

60~~---+--~--t-~---+--I

~

401--+-+-+--+-+-+---1
__

~

__

~

__

12

~~

16

__

~

m

__

-02

-03

FO~ IS(OFF), V D"'12V

~

1+

-

-OS

J

-OS

V+ 28V
V-=OV

-07

TA=25°C

-08

I
12

ANALOG VOLTAGE (VOLTS)

TA=25°C

--

'OION}.../

-04

~

M

V

r- FOR ID(OFF), VS= 12V

16

20

ANALOG VOLTAGE (VOL TSj

24

t-----

GND '" 0

'-

GND~V

20~-l--+--+--+---l--+--1

o

IS~
I

-01

140 1--l--+--+--+-----,--+--1

oL-~

Jo

JO(OFF)

28

12

I--

1-

16

20

24

28

SUPPLY VOL rAGE (VOLTS)

SWITCHING TIME
vs SUPPLY VOLTAGE
1000

NOTE: These single-supply-operation characteristic curves
are valid when the negative power supply V- is tied to
the logic ground reference pin "GND". TTL input
compatibility is still maintained when "GND" is the
same potential as the TTL ground. t'OFF is measured
from 50% of logic input waveform to 0.9 Vo. The
analog voltage range extends from 0 to V+ -4V, the
switch will no longer respond to logic control when
VA is within 4 volts of V+.

900
800

1

r-

RL = 4 7K
C L '" 13pF

V+ '" 2aV

-

V- =OV
:-GND'" OV
TA = 25°C ' - -

700

w

:IE 600
~

~

500
~
~ 400

~

300

I -i--

200 1\

tON

.....

tOFF

100
12

16

20

24

28

ANALOG VOLTAGE (VOLTS)

SIMPLIFIED SCHEMATIC DIAGRAM (TYPICAL SWITCH)

DIGITAL
INPUT
IN x

13-28

1/86, Rev. A

-------------t~ SW-06 QUAD SPST JFET ANALOG SWITCH

OFF ISOLATION TEST CIRCUIT

CROSSTALK TEST CIRCUIT

VA'" 5VRMS
AT f '" 500kHz

VA = 5VRM S

AT f '" 500kHz

NC

D

81

VOl

01

Vour
CL

50n

RL

RL

52
"OFF" ISOLATION

=

20 LOG

[~J
Vaur

D2

NC

son
CROSSTALK = 20 LOG

SWITCHING TIME TEST CIRCUIT

+1SV

LOGIC
INPUT

+15V

SWITCH

INPUT

Vs=-sv

tr
tf

SWITCH

30V

< 20nS
< 20nS

OUTPUT

5,

O-+------if: ....-i--c.....-----1r----

IN 1,IN4

t

L

04

1
01

S2'S3

r

S4

~

I~ _ _~-

I~ _ _

L

;

[>-- -

OIN2,IN3
-

03

1
02

D4

DUAL DPST

o IN ',IN 2
IN3,IN4

01-IN_,:...,'N_2_ _

t
r' r'

[> _ _

t1S1'~

S2

_

01

03

~_

'S3

0~IN3,'N4[> _ J~ _r~

1

04

r r
L L

[> __~ _ _

DPDT

02

fD3
APPLICATIONS INFORMATION

t

switches, The PNP transistor inputs require such low input
current that the SW-06 approaches fan-ins of CMOS input
devices, These bipolar logic inputs exceed any CMOS input
circuit in resistance to static voltage and radiation susceptability, No damage will occur to the SW-06 if logic high voltages are present when the SW-06 power supplies are OFF,
When the V+ and V- supplies are OFF, the logic inputs
present a reverse bias diode loading to active logic inputs,
Input logic thresholds are independent of V+ and V-supplies
making single V+ supply operation possible by simply connecting GND and V- together to the logic ground supply.

This single analog switch product configures, by appropriate
pin connections, into four switch applications, As shown in
Figure 1, the SW-06 connects as a QUAD SPST, a DUAL
SPOT, a DUAL oPST, ora DPDTanalog switch, This versatility increases further when taking advantage of the disable
input (DIS) which turns all switches OFF when taken active
low,
lon-implantation of the JFETanalog switch achieves low ON
resistance and tight channel to channel matching, Combining the low ON resistance and low leakage currents results in
a worst case voltage error figure VERROR@ 125° C= IOION) X
RSOION)= 100nAX 1000= 11 microvolts, Thisamountof error
is negligible considering dissimilar-metal thermally-induced
offsets will be in the 5 to 15 microvolt range,

ANALOG VOLTAGE AND CURRENT
ANALOG VOLTAGE
These switches have constant ON resistance for analog voltages from the negative power supply (V-) to within 4 volts of
the positive power supply, This characteristic shown in the
plots results in good total harmonic distortion, especially
when compared to CMOS analog switches that have a 20 to
30 percent variation in ON resistance versus analog voltage,
Positive analog input voltages should be restricted to 4 volts
less than V+ assuring the switch remains open circuit in the
OFF state, No increase in switch ON resistance occurs when
operating at supply voltages less than ±15 volts (see plot),
Small signals have a 3dB down frequency of 70MHz (see
insertion loss versus frequency plot),

LOGIC INPUTS
The logic inputs (INx) and disable input (DIS) are referenced
to a TTL logic threshold value of two forward diode drops
(1.4V at 25°C) above the GND terminal. These inputs use
PNP transistors which draw maximum current at a logic "0"
level and drops to a leakage current of a reverse biased diode
as the logic input voltage raises above 1.4 volts, Any logic
input voltage greater than 2,0 volts becomes logic "1", less
than 0,8 volts becomes logic "0" resulting in full TTL noise
immunity not available from similar CMOS input analog
13-30

1/86, Rev. A

-------------l1fMD

SW-06 QUAD SPST JFET ANALOG SWITCH

ANALOG CURRENT
The analog switches in the ON state are JFETs biased in their
triode region and act as switches for analog current up to the
IA specification (see plot of los vs Vos). Some applications
require pulsed currents exceeding the IA spec. For example.
an integrator reset switch discharging a shunt capacitor will
produce a peak current of IA(PEAK) = VCAp/Ros)ONI' In this
application, it is best to connect the source to the most
positive end of the capacitor, thereby achieving the lowest
switch resistance and fastest reset times. The switch can
easily handle any amount of capacitor discharge current
subject only to the maximum heat dissipation of the package
and the maximum operating junction temperature from
which repetition rates can be established.

Switching transients occurring at the source and drain contacts results from AC coupling of the switching FETs gate-tosource and gate-to-drain coupling capacitance. The switch
turn ON will cause a negative going spike to occur and the
turn OFF will cause a positive spike to occur. These spikes
can be reduced by additional capacitance loading, lower
values of R L , or switching an additional switch (with its extra
contact floating) to the opposite state connected to the spike
sensitive node.

DISABLE NODE
This TTL compatible node is similar to the logic inputs IN x
but has an internal 2/JA current source pUll-Up. If disable is
left unconnected. it will assume the logic "1" state, then the
state of the switches is controlled only by the logiC inputs IN x.

SWITCHING
POWER SUPPLIES

Switching time tON and tOFF characteristics are plotted versus VANALOG and temperature. In all cases, tOFF is designed
faster than tON to insure a break-before-make interval for
SPOT and OPOT applications. The disable input (OIS) has
the same switching times (tON and tOFF) as the logic inputs
(INx)·

This product operates with power supply voltages ranging
from ± 12 to ±18 volts; however, the specifications only guarantee device parameters witl:1 ± 15 volt ±5% power supplies.
The power supply sensitive parameters have plots to indicate
effects of supply voltages other than ± 15 volts.

TYPICAL APPLICATIONS
4-CHANNEL SAMPLE HOLD AMPLIFIER

OPERATION FROM SINGLE POSITIVE POWER SUPPLY

-15V
V+ -<+36V

13

SW06

VI

V+

+15V

1/4 SW06

VOUT

V2

14

15

V3

11

10

V4

VOUT

1"

0

1000pF

DISABLE ALL
SWITCHES

13-31

1/86, Rev. A

-------------I~ SW-06 QUAD SPST JFET ANALOG SWITCH

HIGH OFF ISOLATION SELECTOR SWITCH (Shunt-Series SWitch)

+1SV

+15V

SW-06

2'

v+

2.

+15V

-15V

NC
-15V

CONTROL
INPUT

HI-CHANNEL 1 ON
LO-CHANNEL 2 ON

THIS SWITCH ARRANGEMENT IMPROVES OFF ISOLATION BY 30dS

SINGLE POLE DOUBLE THROW SELECTOR SWITCH WITH BREAK-BEFORE-MAKE INTERVAL

+15V

lOGIC IN

12

(VOLTS)

v+
~~______~11~____~~~~r7~,'_0__-1__--oVO
+4V

"

Vo
(VOLTS)

During the 88M interval the 1kfl resistor pulls the output to ground assuring
that no shorting between V, and V2 occurs,

13-32

1/86, Rev. A

SW-201/SW-202
QUAD SPST JFET
ANALOG SWITCHES
Precision MOllolithics Inc.

may be independently digitally controlled. Each SW-201
switch is normally closed (NC), whereas each SW-202 is
normally open (NO) when the corresponding digital control
input is a zero. The SW-201 and SW-202 are otherwise
identical.

FEATURES
SW-201
• Normally "ON" for Logic 0 Input
• Improved Performance and Pin Compatible With DG-201,
LF11201/13201, H1201, and IH201
SW-202

The judicious combination of bipolar and FET devices in a
single monolithic IC results in a product with performance
characteristics and ruggedness that are superior to those of a
similar circuit fabricated using CMOS technology.

• Normally "OFF" For Logic 0 Input
• Improved Performance and Pin Compatible With
LF11202/12202/13202 and IH202
Both SW-201 and SW-202

Increased reliability is complemented by excellent electrical
specifications. Potential error sources are reduced by minimizing "ON" resistance and controlling leakage currents at
high temperatures. The switching FET exhibits minimal RON
variation over a 20V analog signal range and with power
supply voltage changes. Operation from a single positive
power supply voltage is possible. With V+ = 36V, V-= OV, the
analog signal range will extend from ground to +32V.

•
•
•
•
•
•
•

Highly Resistant to Static Discharge Destruction
Guaranteed Break-Before-Make Switching (tOFF < tON)
Low "ON" Resistance ....................... 80n Max
Guaranteed RON Matching .................. 15% Max
Low RON Variation from Analog Input Voltage ..... 5%
High Analog Current Operation ............ 10mA Min
Low Leakage Currents at High Temperatures:
TA = 125°C .................. 60nA Max
TA = 85°C ................... 30nA Max
• Guaranteed Switching Speeds:
tON = 500ns Max tOFF = 400ns Max
• Digital Inputs are TTL and CMOS Compatible
• Dual or Single Supply Operation

The PNP logic inputs are TTL and CMOS compatible. Logic
input currents are at micro-ampere levels which improvescircuit
fan in.
PIN CONNECTIONS
16-PIN DUAL-IN-LINE PACKAGE
(0 or P Package)

ORDERING INFORMATIONt
DIP

PACKAGE
16-PIN HERMETIC
16-PIN HERMETIC
16-PIN EPOXY

SWITCH CONFIGURATION
NC
NO
SW201BO'
SW201FQ
SW201GP

SW202BO'
SW202FO
SW202GP

OPERATING
TEMPERATURE
RANGE
MIL
INO
COM

For devices processed In total compliance to MIL-STO-B83, add /883 after
part number Consult factory for 883 data sheet
tAil commercial and Industrial temperature range parts are available with
burn-In For ordenng Information see 1986 Data Book, Section 2
*

SW-201 CONTROL LOGIC
SWITCH
LOGIC
o
ON

GENERAL DESCRIPTION
The SW-201 and SW-202 each consist of four independent,
single-pole, single-throw (SPST) analog switches, which

SW-202 CONTROL LOGIC
LOGIC
SWITCH

o

OFF

OFF

ON

SIMPLIFIED SCHEMATIC DIAGRAM (ONE SWITCH)

010

~---...._---...._-+

__+ __+J""'"f1

ANAL.OG

,.

:IN(S)

OUT 101

DIGITAL
INPUT

Manufactured under the following patent' 4,228,367

13-33

1/86, Rev. A

-----------l~ SW·201/SW·202 QUAD SPST JFET ANALOG SWITCHES

ABSOLUTE MAXIMUM RATINGS

(Note 1)

V+ Supply to Ground •••••••••••.••••••••••••••••••

SW·201BQ, SW·202BQ ••••••••••••••

-55'C to +125'C

SW-201FQ, SW-202FQ •••••••••••••••

Continuous ••••.•••.•••

-25'C to +85'C

500llsec pulse ..... V- Supply -15V to V+ Supply +20V
Maximum Current Through Any Pin ••••••••••••••• 30mA

-65'C to +150'C

P-Suffix • • . • .. .. .. .. .. • • • • • .. .. .. ... -65' C to + 125' C
Power Dissipation (Note 2) ...................... 900mW

NOTES:
1. Absolute ratings apply to both DICE and packaged parts, unless otherwise
noted.
2. Derated 12mW/' C above 75' C.

Lead Temperature (Soldering, 60 sec) •••••••.•.••• 300'C
Maximum Junction Temperature ••••••••••••.••••• 150'C
V+ Supply to V- Supply............................ 36V

ELECTRICAL CHARACTERISTICS at

V±

= ±15V and

TA

=

25'C, unless otherwise noted.

SW-201B
SW-202B
PARAMETER
"ON" Resistance
RON Match Between

Switches
Analog Voltage Range

SYMBOL

CONDITIONS

RON

VA = OV, Is=1mA
VA = ±10V, Is = lmA

RON Match

VA = OV, 10= 100~A;
(Note 1)

VA

Is = 1.0mA (Note 6)
Is= 1.0mA

Analog Current Range

'A

Vs = ±10V

J:

90

"e-

80
RON-

60

1100

f_~)25O~

1000

r-vs= :!:.15V

900

~

700

ill

600

~

1I

~O

"'c;"

400

w

30

0

>

20
10
±15

:!:.16

:t17

:!:.18

POWER SUPPLY VOLTAGE (VOLTS)

±19

V+

I
II

28

I--

o ./
o 1

75

100

125

";0z

~

,/

~ 15~

f = 1MHz

V-=-15V
TA = 25°C

24
20

~

w

V

V

300
200

50

SWITCH CAPACITANCE vs
ANALOG VOLTAGE
32

I I I
I I I

/

25

-25

TEMPERATURE (OC)

I-- CURRENT INTO SOURCE

100
:!:.14

-55

~

40

:!:.13

10pA ' - -__l - - - '__--'-__-'---''---'-__-'

10M

800

50

±12

"

SWITCH CURRENT vs
VOLTAGE

120

70

1M

FREQUENCY (Hz)

"ON" RESISTANCE vs
POWER SUPPLY VOLTAGE
110

10

-5

ANALOG INPUT VOLTAGE (VOLTS)

~"'~~

50

-25

"" ---

....TOFF

0.\8

46

~

200

l00nAr---r-~---r--~--r--.--,

1&0

'/

40
-55

1\

150

25

-25

~~

TA '" 25"C

/'f'

z

0

250

100

65

z

V

CROSSTALK AND "OFF"
ISOLATION vs FREQUENCY

75
70

'-...

300

TEMPERATURE (OC)

RONvsTEMPERATURE

9

TOFF

o
-55

50

350

Vs = ±15V
Rl = lkn
CL '" lOpF
TA=2SoC -

I

./

ANALOG INPUT VOLTAGE-VA (VOLTS)

§

-

V

100

45~--1_--_t--_+--~

./

V V

400

~

TA - 2SCC

v+ '" +15V

SWITCHING TIME vs
ANALOG VOLTAGE

16

--

CS(ON)

12
Cs (OFF)

~

~

/'

/'
2

Co (OFF)

I I

3

4

5

6

7

8

SWITCH CURRENT (rnA)

13-37

9

10

11 12

-11 -9 -7 -5

-3

-1

1

3

5

7

9

11

ANALOG VOLTAGE (VOLTS)

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _----t!lliDSW-201tSW-202 QUAD SPST JFET ANALOG SWITCHES

TYPICAL PERFORMANCE CHARACTERISTICS
SW-201

SW·202

tOWtOFF SWITCHING RESPONSE

tOWtOFF SWITCHING RESPONSE

-ov
-ov

TOP TRACE' LOGIC INPUT (5V!DIVJ
BOTTOM TRACE SWITCH OUTPUT (1V/DIV)

TOP TRACe LOGIC INPUT (5V!DIV)
BOTTOM TRACE' SWITCH OUTPUT (1V/DIV)

OFF ISOLATION TEST CIRCUIT

SW-201 WAVEFORMS

VA'"' 5V RMS
@f=500kHz

"-+--t--......--<>

VOUT

* 3.5V------_....

.on

LOGIC INPUT

50%

tr,tf<20ns

ov==========~~~=====q--~~~==
O.lVa
SW-201 OUTPUT

[~J

"OFF" ISOLATION = 20 LOG

~e============f===~~==~==~~---

VOUT

SWITCHING TIME TEST CIRCUIT
+15V

*SWITCH OUTPUT WAVEFORM SHOWN FOR Vs = CONSTANT WITH LOGIC INPUT
WAVEFORM AS SHOWN Vo IS THE STEADY STATE OUTPUT WITH SWITCH ON

R,

Vo=:Vs - - RL + ros(ON)

SWITCH
INPUT

5,

Vs =~5V

o--t------O"il

SWITCH
OUTPUT

0,

"--t-<)-~-__.-o Vo

I

____ .JI

R,
1Kn

SW-202 WAVEFORMS

LOGIC
INPUT

~15V

REPEAT TEST FOR 1N2. 1N3, AND IN4

* 35V LOGIC INPUT

ov

CROSSTALK TEST CIRCUIT

tr,tf<20ns

50%
O.lVO

SW-201 OUTPUT

.,

VA = 5V RMS
@f"500kHz

50n

09VO

Vo

j

-5V

VD'

I--'ON-

I--'OFF~

.0
*SWITCH OUTPUT WAVEFORM SHOWN FOR Vs = CONSTANT WITH LOGIC INPUT
WAVEFORM AS SHOWN Vo IS THE STEADY STATE OUTPUT WITH SWITCH ON

CAOSSTALK=20 LOG V02
V"

13-38

1/86, Rev. A

-----------l1fHD

SW-201/SW-202 QUAD SPST JFET ANALOG SWITCHES

APPLICATIONS INFORMATION

OPERATION FROM SINGLE POSITIVE POWER SUPPLY

This analog switch employs ion-implanted JFETs in a switch
configuration designed to assure break-before-make action.
The turn-off time is much faster than the turn-on time to
guarantee this feature over the full operating temperature
and input voltage range. Fabricated with Bipolar-JFET processing rather than CMOS, special handling is not necessary
to prevent damage to these switches. Because the digital
inputs only require a 2V logic "1" input level, powerconsuming pull up resistors are not required for TTL compatibility to insure break-before-make switching as is most
often the case with CMOS switches. The digital inputs utilize
PNP input transistors where input current is maximum at the
logic "0" level and drops to that of a reverse-biased diode as
the input voltage is raised above ~ 1.4V.

V+

.:s: +38V

13

v.
1/4SW201
O:-:;VANALOO:'SY+-4V

YO UT

o-----"if---if I
I
I

I

r------,

I

IL...TTL
g:T~MOS ~:-0-'1"""--'
______ ...

The "ON" resistance, RON, of the analog switches is constant
over the wide input voltage range of -15V to +11V with
VSUPPLY = ±15V. For normal operation, however, positive
input voltages should be restricted to 11V (or 4V less than the
positive supply). This assures that the VGsof an OFF switch
remains greater than its Vp, and prevents that channel from
being falsely turned ON. Individual switches are "ON" without power applied.

TYPICAL APPLICATIONS
PROGRAMMABLE GAIN NONINVERTING AMPLIFIER WITH SELECTABLE INPUTS
+15V -15V

~~-------~-~--~--~-OVOUT
+15V

+lSV

1X
18kn

99kn

999kn

~--=J--------A ~"C>-+''----+10X

~--4-------A ~"C>-+"-----f----"00X

"0-+---+---+---+ 1000X
2k"

13-39

100kn

100kn

1/86, Rev. A

S~-7510/S~-7511

®IPMI)

QUAD SPST JFET
ANALOG SWITCHES
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

• Pin Complltlble with AD7510 01, AD7511 01
• JFET Switches Rllther thlln CMOS
• Highly Reslstllnt to Static Dlschllrge Dllmage
• Radiation Resistant
• No SCR Llltch-up Problems
• Low "ON" Resistance - 750 MIIX
• Superior "OFF" Isolation and Crosstalk
• Digital Inputs Compatible with TTL and CMOS
• No Pull-Up Resistors Required to Insure
Break-Before-Make Action with TTL Inputs

The SW-7510/7511 are monolithic linear devices, each containing four independently selectable SPSTanalog switches.
The SW-7510 operates normally-open with logic-low inputs.
The SW-7511 operates normally-closed with logic-low inputs.
All logic inputs are fully TTL input compatible.
Performance advantages include exceptionally high "OFF"
isolation, low leakage current and low crosstalk. Data conversion, position controllers, choppers, demodulators and
programmable-gain amplifiers are popular SW-7510/7511
circuit applications.
The PMI Bipolar-JFET process reduces susceptibility to
electrostatic destruction and offers a high resistance to
radiation exposure. Plus, total freedom from the intrinsic
SCR latch-up problems encountered in equivalently manufactured CMOS products.

ORDERING INFORMATIONt
TYPICAL
25°C
RESISTANCE

PACKAGE
HERMETIC
DIP

TEMPERATURE
RANGE

60n

SW7510AO"
SW7510EO

MIL
INO

80n

SW7510BO"
SW7510FO

MIL
INO

60n

SW7511AO"
SW7511EO

MIL
INO

80n

SW7511BO"
SW7511FO

MIL
INO

PIN CONNECTIONS
16-PIN HERMETIC DUAL-IN-LINE
(O-Sufflx)
CONTROL LOGIC:
SW-7510: Switch "ON"
for Address "HIGH"
SW-7511: Switch "ON"
for Address "LOW"

"For devices processed In total compliance to MIL-STO-883, add 18S3 after
part number. Consult factory for 883 data sheet
t All commerCial and industrial temperature range parts are available with
burn-In For ordering information see 1986 Data Book, Section 2.

SCHEMATIC DIAGRAM (Typical SW-7510 Switch)

LOGIC
INPUT

G A O U N O o - - - + - - - - - - + - - - -.....- - . . J

13-40

1/86, Rev. A

-------------l~ SW-7S10/SW-7S11 QUAD SPST JFET ANALOG SWITCH

ABSOLUTE MAXIMUM RATINGS

(TA = 25°C, unless otherwise noted).
V+ Supply to Ground .............................. 36V
Logic Input Voltage ...•.....•.. (-2V or V-) to V+ Supply
Analog Input Voltage
Continuous ............. V- Supply to V+ Supply +20V
1% Duty Cycle and Driving
all 4 Inputs with

Operating Temperature Range,
SW-7510/7511AO, SO •...•..•.•..•.. -55°C to +125°C
SW-7510/7511 EO, FO ..•.••..••.••.•• -25°Cto+85°C
DICE Junction Temperature (Ti) ...•... -65°C to +150°C
Storage Temperature Range .•...•••..• -65°C to +150°C
Power Dissipation •..•...•••••...•••.•..•....•.. 500mW
Derate above 100°C ••.••••.•..•.•..••.•..•• 10mW/oC

500l"s pulse •..•.•. V- Supply -15V to V+ Supply +20V
Maximum Current Through Any Pin •.•...•••.•.... 25m A

Lead Temperature (Soldering, 60 sec) •.•••........ 300°c
Maximum Junction Temperature ......•.•.••.••... 150°C
V+ Supply to V- Supply ............................ 36V

ELECTRICAL CHARACTERISTICS

at Vs

NOTE: Absolute ratings apply to both DICE and packaged parts. unless
otherwise noted

= ± 15V and TA = +25° C, unless otherwise noted.
SW-7S10AlE
SW-7S11A/E

SW-7S10B/F
SW-7S11B/F

PARAMETER

SYMBOL

CONDITIONS

TYP

MAX

"ON" Resistance

RON

Vo=OV, 108= 1mA

60

75

"RON vs Vo (Vs)

"RON

VD'" 10V, ID= 1mA

15

RON Match 01 Switches

RON Match

Vo = OV, los = 1mA

1.5

Analog Voltage Range

VA

Is = 1mA (Note 5)

"OFF" Leakage Current

IS{OFF)' IO(OFF)

Vs =+10V, VD =-10V, (Note 1)

+ iOION)

"ON" Leakage Current

IS(ON)

Logic "1" Voltage

V 1NH

(NoteS)

Logic "0" Voltage

V 1NL

(NoteS)

I'NL

V,N = +O.4V

Logic "0" CUrrent

MIN

+10
-10

MIN

TYP

MAX

80

100

15
10

+11
-15

15
+10
-10

10

Vs = Va = +10V, (Note 1)
20

10

30

nA

30

nA
V

O.B
35

15

OB

V

3.5

JJ.A

Logic Input Capacitance

C DiG

V,N = +O.4V

"ON" Switching Time

tON

Vs = -5V, RL = 1kn, C L = 7pF, (Note 4)

350

450

450

550

"OFF" Switching Time

tOFF

Vs= -5V. RL = 1kn. C L = 7pF, (Note 4)

260

300

350

450

"OFF" Isolallon

ISO OFF

(Note 2)

Crosstalk

CT

(Note 3)

Analog "OFF" Capacitance

CS(OFF1' CO(OFF)

Vs= IV, Vo=O

Analog "ON" Capacitance

eS(ON)' COlON)

Vs=OV,Vo=O

Feedthrough Capacitance

COS(OFF)

Vs=OV

COOIOFF)

Vs=OV
Vs=OV

04
04

Positive Supply Current

1+

Logic Inputs at "0" or "1"

50

90

30

90

mA

Negative Supply Current

1-

LogiC Inputs at "0" or "1"

2B

50

17

5.0

mA

Channel Capacitance

CSSIOFFJ'

1.5

%

V

20

15

n
%

+11
-15

10

UNITS

1.5

pF
ns

~

p..
.......

b

::J

~
VJ
~

::r::

U

f-<
.......

ns

~

66

dB

C)

70

70

dB

0

6.5

65

pF

14

14

pF

OB

OB

pF

04
04

pF

VJ

66

NOTES:
1 The conditions listed specify the worst case leakage currents The leakage
currents apply equally to source (8) or drain(D)
OFF Isolation IS measured by drivmg the source of any OFF SWitch and
observing the voltage which appears on the drain The conditions are.

RL = 6BOn, C L = 7pF, Vs = 5V RMS ' 1= 100kHz
Crosstalk IS measured by driVing source of any OFF sWitch and obserVing
voltage which appears on any other "ON" output drain The conditions

are RL = 6BOn. C L = 7pF, Vs = 5V RMS • 1 = 100kHz
4 Sample tested
5. Guaranteed by RON and leakage tests For normal operation maximum
analog signal voltages should be restricted to less than (V+) -4V

13-41

~

1/86, Rev. A

~

II

-------------l~ SW·7510/SW·7511 QUAD SPST JFET ANALOG SWITCH

ELECTRICAL CHARACTERISTICS at Vs = ±15V, -55°C $; TA$; +125°C for SW·7510AO, SO and SW·7511AO, SO; and
-25°C $; TA $; +85° C for SW·7510EO, FO and SW·7511 EO, FO, unless otherwise noted.
SW·7510A/E
SW·7511 AlE
MIN

TYP

SW·7510B/F
SW·7511B/F

MAX

MIN

TYP

MAX

UNITS

PARAMETER

SYMBOL

CONDITIONS

"ON" Aesistance

AON

Vo= OV.IOS= 1mA

~RON VB.

"RON Drift

Vo=OV.los= 1mA

Analog Voltage Range

VA

Is= 1mA (Note 4)

"OFF" Leakage Current

IS{OFFI,ID(OFF)

"ON" Leakage Current

'SION)

Logic "1" Voltage

V,NH

(Note 4)

Logic "0" Voltage

V,NL

(Note 4)

0.8

0.8

V

Logic "0" Current

I'NL

V ,N = +0.4V

5.0

7.0

J.lA

"ON" Switching Time

tON

Vs = -5V, RL = 1kO, C L = 7pF (Note 2)

600

1000

ns

"OFF" SWitching Time

tOFF

Vs= -5V, AL = 1kO, CL = 7pF (Note 2)

500

750

ns

Positive Supply Current

1+

Logic Inputs at "0" or "1"

13

13

mA

Negative Supply Current

1-

Logic Inputs at "0" or "1"

7.5

7.5

mA

Temperature

+ 'D(ON)

100

150

0.4
+10
-10

+11
-15

+10
-10

°

0.5

%/'C

+11
-15

V

Vs = +10V. Vo = -10V. (Notes 1. 3)

90

100

nA

VS = Vo = +10V. (Notes 1. 3)

90

100

nA

2.0

2.0

V

NOTES:
1. The conditions listed specify the worst case leakage currents. The leakage
currents apply equally to source (S) ordrain(D).
2. Guaranteed by design.
3. Tested at 125'C only for "A" and "8" grades.
4. Guaranteed by RON and leakage tests.

13·42

1/86, Rev. A

-----------I1fMD

SW-7510/SW-7511 QUAD SPST JFET ANALOG SWITCH

DICE CHARACTERISTICS
SW-7510 (SWITCH ON FOR ADDRESS HIGH)

SW-7511 (SWITCH ON FOR ADDRESS LOW)

DIE SIZE 0.091 X 0.083 Inch, 7553 sq. mils
(2.311 X 2.108 mm, 1.918 sq. mm)
1.
2.
3.
4.
5.
6.
7.
8.

NEGATIVE SUPPLY (SUBSTRATE)
GROUND
ADDRESS (A1)
ADDRESS (A2)
ADDRESS (A3)
ADDRESS (A4)
DISABLE (NO CONNECT)
POSITIVE SUPPLY

9.
10.
11.
12.
13.
14.
15.
16.

DRAIN (04)
SOURCE (S4)
DRAIN (03)
SOURCE (S3)
DRAIN (02)
SOURCE (S2)
DRAIN (01)
SOURCE (S1)

1.
2.
3.
4.
5.
6.
7.
8.

NEGATIVE SUPPLY (SUBSTRATE)
GROUND
ADDRESS (A1)
ADDRESS (A2)
ADDRESS (A3)
ADDRESS (A4)
DISABLE (NO CONNECT)
POSITIVE SUPPLY

9.
10.
11.
12.
13.
14.
15.
16.

DRAIN (04)
SOURCE (S4)
DRAIN (03)
SOURCE (S3)
DRAIN (02)
SOURCE (S2)
DRAIN (01)
SOURCE (S1)

For additional DICE Information refer to 1986 Data Book, Section 2.

WAFER TEST LIMITS at V+

PARAMETER

= + 15V, V- = -15V, TA =25°C, unless otherwise noted.
SW-7510NI
SW-7511N

SW-7510GI
SW-7511G

LIMIT

LIMIT

UNITS

75

100

OMAX

SYMBOL

CONDITIONS

RON

Vo=OV.los=1mA

Logic "1" Voltage

V 1NH

(Note 1)

2.0

20

VMIN

Logic "0" Voltage

V 1NL

(Note 1)

0.8

0.8

V MAX

Logic "0" Current

I'NL

V'N= +O.4V

3.5

3.5

IJ.AMAX

Positive Supply Current

1+

Logie Inputs at "0"

9

9

mAMAX

Negative Supply Current

1-

Logie Inputs at "0"

5

mAMAX

"ON" Resistance

NOTES:
1. Guaranteed by RON and leakage tests.
Electrical tests are performed at wafer probe to the limits shown. Due to vanatlons in assembly methods and normal yield loss, Yield after packaging IS not
guaranteed for standard product dice. Consult factory to negotiate speCifications based on dice lot qualification through sample lot assembly and testing

TYPICAL ELECTRICAL CHARACTERISTICS at V+ = + 15V, V- = -15V and TA = 25°C, unless otherwise noted.

PARAMETER

SYMBOL

SW-7510GI
SW-7511G

TYPICAL

TYPICAL

Vo=OV, 10S= 1mA

60

80

0

Vo=OV, los= 1mA

0.4

0.5

%/oC

"ON" Switching Time

Vs = -5V, RL = 1kO, C L = 7pF

350

450

ns

"OFF" Switching Time

Vs = -5V, RL = 1kO, C L = 7pF

260

350

"ON" Resistance
RON vs. Temperature

RON Drift

CONDITIONS

SW-7510NI
SW-7511N

13-43

UNITS

ns

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _---1!fMDSW-7510/SW-7511 QUAD SPST JFET ANALOG SWITCH

TYPICAL PERFORMANCE CHARACTERISTICS (Apply to all models. unless otherwise noted)
LARGE-SIGNAL SWITCHING

LARGE-SIGNAL SWITCHING

-

C I:~

Ii
~- :~

,:I

I-

~

_.-

.1

rj
~
I= ~

"

[~

,.
I"

I

M

1

:VI
I

SMALL-SIGNAL SWITCHING

,

ILl

•

100m
" VI
I

-

-

[H"E~
I

-

SMALL-SIGNAL SWITCHING
WITH FILTERING

..
- --~

.,,-+-•

I,I
-~ .

[~
" ,~

1';::'1
II.

i1
"'.,::1
.

1-

.~

~-

Ir:ii
ilil 1-

1Ii:~
.

' li:J

-II
'H-+

I
II.

-

,.

II
"

1::;:.1

,= ',,~

II

Il

-

SMALL-SIGNAL SWITCHING
WITH FILTERING

t~
:
, I

~
!~

111

-

SMALL-SIGNAL SWITCHING

II

I. '
-~".

~

.1

1 ,-11:

-

II.

SMALL-SIGNAL SWITCHING
WITH FILTERING

SMALL-SIGNAL SWITCHING

NOTE:
Upper Photo Traces: Logic Control Signal Ax (5v/DIV)
Lower Photo Traces: Switch Outputs Vo

13-44

1/86, Rev. A

-----------I~ SW-7510/SW-7511 QUAD SPST JFET ANALOG SWITCH

TYPICAL PERFORMANCE CHARACTERISTICS (Apply to all models, unless otherwise noted)
NORMALIZED RON vs
SWITCH CURRENT (Is)

"ON" RESISTANCE (RON) vs
ANALOG VOLTAGE (VA)

II

120
110
100

v+ "

r
f-------

+15V

3

V-=-lSV
TA=25°C
lOS = ±lmA

g

80

~

70

t

2 -

V+ = +15V
V-=-1SV
1 _ TA =25°C

90

t---

SW7S10!11B, F

t---

E

60

8

50

---

-

9

1,ov/

;/

/

0

Sl751L'1~

RONvsTEMPERATURE
200

180

/

=

140
12 a

/'

V

t-~'O
~'5)
v- -15V

160 t-- .loS = 1 rnA
VA = OV

§

SW7510Jl 1, B. F

100

~

V A =-10V

..... n .H-

80
60

40

40

V

.J.-1"

z

1-:::

::::: I--"

f-""SW7510/11, A, E

7
30

20

20
-10 -8

-6

-4

4

-2

6

8

o.6

-5 -4

10

-3

-2

-1

SWITCH LEAKAGE CURRENTS
VB ANALOG INPUT VOLTAGE
300

v~ -

270

v-

a
,1,5V

180

10 (ON) + IS (ION)

IJI

90
60

Ih

30

lIL

-6

-4 -2

0

2

o-

60a

.........

Cs

-s

(OFF)

I---"
3

-6 -4

-2

a

-20

I

4

B

10

-15V
TA = 25°C
SEE AC TEST
f-CIRCUIT

0:

o

r- r--;;

IINL -

30a

-

.....

10a
-60 -40 -20

55o

0

20

40

60

80 100 120 140

TEMPERATURE (OC)

40 0

~

~ -70

-90

10k

lOOk

0

,~ 0

-'SW7~1O/1;, A, I~ __

IIII~

1M

13-45

V
0<
~

V

:-'- , ""'"""'"

30

FREQUENCY (Hz)

100 120 140

SW7510!11, B, F
TON

,/'

Clilllill

'"
~I

v:

TON

350

3
~
~

50o

t/

0 20
40 60 80
TEMPERATURE (OC)

V- = -150
Vs = -5V

45 0

OFF ISOLATIO;;'

~ -60

o .... 0

20a

~±,oj'A;E

SWITCHING TIMES
vs TEMPERATURE
60 a

11111111

40 a

t-

-60 -40 -20

65 a

J, ~ l,~UIII
r-v- '"

~~~~~G~~~LL~~7~~'o"

500

...... I----.

a

2

CROSSTALK AND "OFF"
ISOLATION vs FREQUENCY

r-- -.......

1+ SW7510!11, A, E

[-t---{..J L

1

a

-10 -8

DIGITAL INPUT
CURRENT IINL
VB TEMPERATURE

r---.

-----

5

VA (VOLTS)

I

100 120 140

lON) t--"-

8

10

80

~ +,Jv

2

B

60

V+
V-=-15V

-

VA (VOLTS)

_tol"5J

40

8

C

80a
70a

0

0

~

20

SUPPLY CURRENTS
VB TEMPERATURE

2

4

0

TEMPERATURE (OC)

2

V-=-15V

90

a
-60 -40 -20

10 (OFF)

a

-10 -8

1000

4

/

120

1100

6'

,I IO~FI
IS

150

5

v+ =+1SV
v-= -15V
TA = 25°C

8_

210

4

SWITCH CAPACITANCES
vs ANALOG VOLTAGE (VA)

= -15V
TA = 25"C

240

3
IS (rnA)

VA-ANALOG INPUT VOL TAGE (VOLTS!

25

J~

,/'

V
I--- I--"

SW7510/11, A, E
TOFF

20o _ ~~~~'0/1', B, F

10M

15a
-60 -40

20

0

20

40

60

80

laO 120 140

TEMPERATURE (eC)

1/86, Rev_ A

_ _ _ _ _ _ _ _ _ _-1~SW-7510/SW-7511 QUAD SPST JFET ANALOG SWITCH

AC TEST CIRCUITS
CROSSTALK MEASUREMENT CIRCUIT
+15V

-15V
CROSSTALK'" 20 LOG

v-

0~~

1----...--,

0,

'pF

VSUPPLY = ± 15V. Higher input voltage is tolerable provided
that some form of current limiting is employed (such as that
of an op-amp output stage) to avoid exceeding junction
temperature and power dissipation requirements. For normal
operation, however, positive input voltages should be
restricted to 11V (or 4V less than the positive supply). This
assures thatthe Vas of an OFF switch remains greater than its
Vp, and prevents that channel from being falsely turned ON.
Individual switches are "ON" without power applied.
Proper switching requires the "Source" terminal be connected to the input driving signal.

'pF

LATCHING DPDT SWITCH
FOR SW-7511 REVERSE VOL rAGE TO A, & A2

+15V

AD

-a

.

a

5

PR

ISOLATION MEASUREMENT CIRCUIT
t15V

INPUT 1

'/2

-15V

v,

,.

V'

SN7474

VOFF ISOLATION

S,

~

20 LOG

~~,1

OUTPUT 1

INPUT 2

CLR

INPUT 3
SW-7510

son

0,

A,

Al
OUTPUT 2

'pF

INPUT 4
FOR SW-7S11 CONNECT A, TO +5V

SWITCHING TIME TEST CIRCUIT
+15V

Truth Table
State of Switches

-15V

Command

+~~ J"1..I"L

"v+,--,lv-'"

Ao
1

r-~--JA

o
1

SW-751'

o

After Command
82 and 83
81 and 83
same
same
on
off
off
on
INDETERMINATE

INTEGRATOR WITH ANALOG RESET AND
START/STOP CAPABILITY

APPLICATIONS INFORMATION

1/.

This analog switch employs ion-implanted JFETs in a switch
configuration designed to assure break-before-make action.
The turn-off time is much faster than the turn-on time to
guarantee this feature over the full operating temperature
and input voltage range. Because the digital inputs only
require a 2.0V logic "1" input level, power-consuming pull-up
resistors are not required for TTL compatibility to insure
break-before-make switching as is most often the case with
CMOS switches. The digital inputs utilize PNP input transistors where input current is maximum at the logic "0" level and
drops to that of a reverse-biased diode (about 10nA) as the
input voltage is raised above ~ 1.4V.

SW7510Fa
RL

r- 14

r-"W'v_T
1/.
SW7510FQ

I

~RESET

13

-,sv
+15V

Vo = llRC
.r---<~-{)

The "ON" resistance, RON, of the analog switches is constant
over the wide input voltage range of -15V to +11V with

J VINdt

NOTE: Applicallons show SW-7510. For SW-7511 applicallons the logic is
inverted.

13-46

1/86, Rev. A

------------I~ SW-7510/SW-7511 QUAD SPST JFET ANALOG SWITCH

TYPICAL APPLICATIONS
ACTIVE LOW-PASS FILTER WITH DIGITALLY SELECTED BREAK FREQUENCY

,.0
+15V

fCO

r"
fC3

,.

120

CONTROL

fC2

z

,.

"""

.0

.....

w

":;a 40

12

>

SELECT

fCl

OPEN
..... ...... OP-02
LOOP GAIN

[

SELECT

TTL

---.,

'"

~

iC4, l-fL

fCl

fC3

'C2

""'-

............... ............. ...............

............. ~ f~

"'-.....

10

SELECT

.............

.....

~ fl.,

"'-..... ................

............. ............... ...............
fO

1K

fOO

10K

100K

fM

FREQUENCY - Hz

AL (VOLTAGE GAIN BELOW BREAK FREQUENCY)
~------------------~---oVOUT

R3

Rf

~ 100 (40dB)

Ie (BREAK FREQUENCY)

~ 211"~3CX

IL (UNITY GAIN FREQUENCY)

NOTE: Applications show SW-7510. For SW-7511 applicallons the logic
inverted.

~

= 211"~'CX

IS

13-47

1/86, Rev_ A

MUX-08/MUX-24
8-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS
10VERVOLTAGE AND POWER SUPPLY LOSS PROTECTEDj
Precision Mono1ithics Inc.

break-before-make action without the need for external pullup resistors over the full operating temperature range,

FEATURES
•
•
•
•
•
•
•
•

JFET Switches Rather Than CMOS
Low "ON" Resistance ...................... 2200 Typ
Highly Resistant to Static Discharge Damage
No SCR Latch-Up Problems
Digital Inputs Compatible With TTL and CMOS
125 0 C Temperature Tested Dice Available
MUX-OS Pin Compatible With DG50S, HI-508A, IH510S,
IH6108, LF1150S/1250S/1350S, AD7506
MUX-24 Pin Compatible With DG509, HI-509A, IH520S,
IH6208, LF11509112509/13509, AD7507

For single sixteen-channel and dual eight-channel models,
refer to the MUX-16/MUX-28 data sheet.

PIN CONNECTIONS

ORDERING INFORMATIONt
PACKAGE
25'C ON
HERMETIC PLASTIC
RESISTANCE
DIP
DIP
220n

MUXOBAO'
MUXOBEO

300n

MUXOBBO'
MUXOBFO

220n

MUX24AO'
MUX24EO

LCC

16-PIN DUAL-IN-LINE PACKAGE
(Q or P Package)

MIL
INO
COM

MUXOBEP

300n

TEMPERATURE
RANGE

MUXOBBRCIBB3
MUXOBFP

MIL
INO
COM

MUX24EP

MIL
INO
COM

MUX24FP

MIL
INO
COM

MUX24BO'
MUX24FO

MUX-OSBRC/SS3
LCC
(RC-Suffix)

Mll-STO-883, add /883 after
part number Consult factory for 883 data sheet.
tAil commerCial and Industrial temperature range parts are available WIth
burn-In. For ordenng information see 1986 Data Book, Section 2

* For devices processed In total compliance to

FUNCTIONAL DIAGRAMS
GENERAL DESCRIPTION
The MUX-08 is a monolithic eight-channel analog mUltiplexer which connects a single output to one of the eight
analog inputs depending upon the state of a 3-bit binary
address.

GND

The MUX-24 is a monolithic four-channel differential analog
multiplexer configured in a double pole, four-position (plus
OFF) electronic switch array. A two-bit binary input address
connects a pair of independent analog inputs from each
four-channel input section to the corresponding pair of
independent analog outputs.
All switches in the MUX-08/MUX-24 are turned OFF by
applying logic "0" to the ENABLE pin, thereby providing a
package select function.

ENABLE

Al

AO

v.
GNO

Fabricated with Precision Monolithics' high performance
Bipolar-JFET technology, these devices offer low, constant
"ON" resistance, low leakage currents and fast settling time
with low crosstalk to satisfy a wide variety of applications.
These multiplexers do not suffer from latch-up or static
charge blow-out problems associated with similar CMOS
parts, The digital inputs are designed to operate from both
TTL and CMOS levels while always providing a definite

v-

DRAIN

A

13-48

SlA

S2A

S3A

S4A

SlB

S2B

S3B

S4B DRAIN

B

1/86, Rev. A

-------l~ Mux-oa/MUX-24 a-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

ABSOLUTE MAXIMUM RATINGS

(Note)

Operating Temperature Range

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300°C
Maximum Junction Temperature •.•••••••••••••••• 150'C

MUX-OS/24-AO, BO, BRC •.••••••.•• -55°C to +125°C
MUX-oS/24-EO, FO •••••••..••.•••••• -25° C to +S5° C
MUX-oS/24-EP, FP •••••••..••••.•••••••• 0° C to + 70° C
DICE Junction Temperature (TI ) •.•••••
Storage Temperature Range. • • • • • • • . ••
P-Suffix ••••••••••••••• • • • • • • • • • • •••

V+ Supply to V- Supply. • • • • •• • • • • • • • • • • • • • • • • • • • •. 36V
Logic Input Voltage. • • • • • • • • • •• (-4V or V-) to V+ Supply

-65°C to +150°C

Analog Input Voltage .... V-Supply-20Vto V+ Supply+20V

-65° C
-65° C

Maximum Current Through Any Pin

to + 150° C
to +

Power Dissipation ••••••••••••••.••••••••••••••• 500mW
Derate above 100° C

NOTE: Absolute rallngs apply to both DICE and packaged parts, unless
otherwise noted

•••••••••••••••••.•••••• 10mW/o C

ELECTRICAL CHARACTERISTICS

at V+ = + 15V, V- = -15V and TA = 25' C, unless otherwise noted.

MUX-08A1E
MUX-24A1E
PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

Vs ';10V, Is'; 200~A

ARoN With Applied Voltage

ARoN

-10V'; Vs'; 10V, Is =

MIN

RON Match Between SWitches

VA

(Note6)

Source Current (Switch "OFF")

Is 10FF,

Vs = 10V. Vo = -10V (Note 1)

Digital Input Current

300

IO{OFF)

Vs= 10V. Vo =-10V (Note 1)

MUX-OB
MUX-24

IO(ONl

Vo = 10V (Note I)

MUX-OB
MUX-24

liN

Y,N = O.4V to 15V

'INL(EN)

VEN = 0 4V

Digital Input Capacitance

COIG

TYP

MAX

300

400

UNITS

n
%

+10 +104
-10
-15

Digital "0" Enable Currant

MIN

15

Vs = OV, Is =

+ISION,

MAX

220

200~A

RON Match

Leakage Current (Switch "ON")

TYP

MUX-OBB/F
MUX-24B/F

200~A

Analog Voltage Range

Drain Current (Switch "OFF")

•••••••.••••••• 25mA

125° C

20
+10 +10.4
-10
-15

%
V

001

10

001

20

nA

01
005

10
10

01
005

20
20

nA

01
005

10
10

0.1
0.05

20
20

nA

10

~A

10

~A

10
10
3

4
3

pF

tTRAN

1B

Output Settling Time

ts

10V Step to 0.10%
10V Step to 0 05%
10V Step to 0.02%

13
15
23

17
19
25

~s

Braak-Belore-Make Delay

tOPEN

Figure 3 (Test CorCUlt)

OB

10

~s

tON (EN)

(Note 5) Figure 2
(Test CorCUlt)

tOFF (EN)

(Note 5) Figure 2
(Test CorCUlt)

MUX-OB
MUX-24

01
02

ISO OFF

(Note 4) Figure 5
(Test CircUit)

MUX-08
MUX-24

60
66

60
66

dB

Crosstalk

CT

(Note 3) Figure 4
(Test CorCUlt)

MUX-08
MUX-24

70
76

70
76

dB

Source Capacitance

es (OFF)

SWitch "OFF",
Vs= OV, Vo= OV

MUX-08
MUX-24

25

25

Drsln Capacitance

CD(OFF)

SWitch "OFF",
Vs=OV, Vo=OV

MUX-08
MUX-24

7
4

7
4

pF

Input to Output Capacitance

COSIOFF'

(Note 4)

MUX-OB
MUX-24

03
015

03
015

pF

Positive Supply Currant
(All Digital Inputs
Logic "0" or "1 ")

1+

V+= 15V
V+=5V

10
8

12

6
5

12

Negative Supply Current
(All Digital Inputs
Logic "0" or "1 ")

1-

V+=-15V
V+ =-5V

30
25

38

20
18

38

Enable Delay "ON"
Enable Delay "OFF"
"OFF" Isolation

21

04
05

1B

02
03

21

3
p.,

.......

(Notes 2,5) Figure 1
(Test CorCUlt)

Switching Time

~
I-I-1

~s

~

~
I-I-1
::r:
U
tI)

r-<
.......

~

C)

2

~s

04
06

~s

0

~

II

pF

mA

mA

NOTES: See next page

1/86, Rev. A

13-49

~---------

._----.

-----------~-

_ _ _ _ _ _-I~MUX.oa/MUX.24 a·CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

ELECTRICAL CHARACTERISTICS at V+ = 15V, V- = -15V and -55° C:5 TA:5 125° C, unless otherwise noted,
MUX"()BAI
MUX-24A
PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

Vs S 10V, Is S 200/AA

.:IoR ON With Applied Voltage

.:IoR ON

-10V S Vs S 10V, Is = 200"A

RON Match Between Switches

RON Match

Vs =OV,l s =200"A

MIN

TVP

MUX·OBBI
MUX-24B

MAX

MIN

TVP

400

MAX
500

15

UNITS

0

4,5

%

10

15

%

+10 +10.4
-10
-15

+10 +10.4
-10
-15

V

Analog Voltage Range

VA

(NoteS)

Source Current (Switch "OFF")

'SIOFFI

Vs = 10V. Vo = -10V (Notes 1, 7)

25

50

nA

'OIOFFI

Vs= 10V, Vo =-10V
(Notes 1, 7)

MUX-oa
MUX-24

100
50

500
500

nA

MUX-08
MUX-24

100
50

500
500

nA

Drein Current (Switch "OFF")
Leakage Current (Switch "ON")

1010NI
+ISIONI

Vo= 10V (Noles 1, 7)

Digital "1" Input Voltage

V 1NH

(NoteS)

Digital "0" Input Voltage

VINL

(NotaS)

0.7

0.7

V

Digital Input Current

liN

VIN = 0.4V to 15V

20

20

"A

Digital "0" Enable Current

IINLIENI

V EN = 0.4V

20

20

"A

Positive Supply Current

1+

All Digital Inputs Logic
"0"or"1"

15

15

rnA

Negative Supply Current

1-

All Digital Inputs
Logic "0" or "1"

5

5

rnA

2

V

ELECTRICAL CHARACTERISTICS at V+ = 15V, V-= -15V and -25°C:5 TA:5 +B5°C for MUX-OBEQ, FQ and
MUX-24EQ, FQ; 0° C:5 TA:5 +70°C for MUX-DBEP, FP and MUX-24EP, FP, unless otherwise noted.
MUX"()BEI
MUX-24E
PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

VsS 10V,I s S 2oo"A

.:IoR ON With ApplIed Voltage

.:loR ON

-10V S Vs S 10V, Is = 200"A

RON Match Between SWItches

RON Match

Vs = OV, Is = 200"A

MIN

TVP

MUX-OBFI
MUX-24F

MAX

MIN

TVP

400

MAX
500

15

UNITS

0

4.5

%

10

15

%

+10 +10.4
-10
-15

+10 +10.4
-10
-15

V

Analog Voltage Range

VA

(Note 6)

Source Current (Switch "OFF")

ISIOFFI

Vs= 10V, V o =-10V (Notes 1, 7)

10

10

nA

Drain Current (SWitch "OFF")

'OIOFFI

Vs= 10V, Vo =-10V
(Notes 1, 7)

MUX-oS
MUX-24

100
50

100
50

nA

Leakage Current (Switch "ON")

'OIONI
+ISIONI

Vo= 10V (Notes 1, 7)

MUX-oS
MUX-24

100
50

100
50

nA

Digital "1" Input Voltage

VINH

(Note 6)

DIgital "0" Input Voltage

VINL

(NoteS)

O.S

- O.S

V

Digital Input Current

'iN

VIN = 0.4V to 15V

20

20

"A

Digital "0" Enable Current

IINLIENI

VEN =0.4V

20

20

"A

Positive Supply Current

1+

All Digital Inputs Logic
"0"or"1"

15

15

rnA

Negative Supply Current

1-

All DigItal Inputs
Logic "0" or "1"

5

5

rnA

NOTES:
1 Conditions applied to leakage tests insure worst case leakages. Exceeding 11 V on the analog input may cause an "OFF" channel to turn "ON"
2. RL = 10MO, C L = 10pF.
3. Crosstalk Is measured by driVing channel S with channel 4 "ON".
RL = 1MO, C L = 10pF, Vs = 5V RMS, I = 500kHz.

2

V

4.

"OFF" isolation is measured by driving channelS WIth ALL channels "OFF"
RL = 1kO, C L = 10pF, Vs=5V RMS, I = 500kHz. COS is computed Irom the
OFF isolation measurement.
5 Sample tested.
S Guaranteed by leakage current and RON tests.
7. Leakage tests are performed only on military temperature grades at 125'C.

13-50

1/86, Rev. A

--------1~ Mux-oa/MUX-24 a-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

DICE CHARACTERISTICS (125°C TESTED DICE AVAILABLE)
~!!!!!!IIIIIII!!!!!!11 MUX-OB

MUX-24

DIE SIZE 0.090 x 0.061 Inch, 5490 sq. mils
(2.286 X 1.549 mm, 3542 sq. mm)

9.
10.
11.
12.
13.
14.
15.
16.

1. AO
2. ENABLE
3. Y- (SUBSTRATE)
4. S1

5. S2
6. S3
7. S4
8. DRAIN

S8
S7
S6
S5

9.
10.
11.
12.
13.
14.
15.
16.

1. AO
2. ENABLE
3. Y- (SUBSTRATE)
4. S1 A
5. S2A
6. S3A

Y+
GND
A2
A1

7. S4A
8. DRAIN A

DRAIN B

S4 B
S3 B
S2 B
S1 B

Y+
GND
A1

For additional DICE informalion refer 10
1986 Data Book, Section 2.

WAFER TEST LIMITS at V +

PARAMETER

= 15V, V - = -15V, TA = 25 ·C, unless otherwise noted. (Note 1)

SYMBOL

CONDITIONS

RON

Vs = OV,
Is = 200~A

Digital "1" Input Voltage

V,NH

(Note 2)

Digital "0" Input Voltage

V,NL

(Note 2)

Digital "0" Input Current

I'NL

V,N =04V

"ON" Resistance

Digital "0" Enable Current

ItNL(ENI

Positive Supply Current
(All Digital Inputs Logic "0")

1+

Negative Supply Current
(All Digital Inputs Logic "0")

1-

Analog Input Range

VA

V ,N =04V

MUX-OB/
MUX-24NT

MUX-OB/
MUX-24N

MUX-OB/
MUX-24G

LIMIT

LIMIT

LIMIT

300
400

300

400

O.B

O.B

O.B

10
20

10

10

TA = 125"C

10

10

TA =125"C

10
20
12
15

12

12

TA =125"C

3.B
5

38

3.B

TA = 125"C

±10

±10

±10

TA = 125"C

UNITS
nMAX
VMIN

(Note2)

V MAX
~AMAX

~AMAX

mAMAX

mAMAX
V MIN

NOTE:
Electrical tests are performed at wafer probe to the limits shown Due to variations in assembly mehtods and normal Yield loss, yield after packaging is not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testmg.

TYPICAL ELECTRICAL CHARACTERISTICS at V+ = 15V, V-=-15V and TA = 25°Cfor MUX-08/24N & G, TA= 125°C for
MUX-08/24NT, unless otherwise noted.
MUX-OB/
MUX-24NT

MUX-OS/
MUX-24N

MUX-OS/
MUX-24G

TYPICAL

TYPICAL

TYPICAL

(Note 1)

17

13

2.1

~s

ts

10V Step to 0.1% (Note 1)

21

1.5

1.9

~s

PARAMETER

SYMBOL

CONOITIONS

Switching Time

tTRAN

Output Settling Time

UNITS

Break-Belore-Make Delay

t OPEN

(Note 1)

O.B

OB

1.0

~s

Crosstalk

CT

(Note 1)

70

70

70

dB

~RON

20

05

0.5

nA

+10.4/-15

+104/-15

+10.4/-15

V

With Applied Voltage

~RoN

-10V ,; Vs'; 10V, Is =

Leakage Current (Switch "ON")

IO{ON)

V D= 10V (Note 1)

Analog Input Range

VA

200~A

NOTES:
1. The data shown is extrapolated from measurements made on the
packaged deVices

%

2

13-51

Guaranteed by leakage current and RON tests.

1/86, Rev. A

IEMD

MUX-08/MUX-24 8-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

MUX-08
LOGIC STATE

MUX-24
LOGIC STATE
EN

"ON"
CHANNEL

X

L

NONE

L

H

EN

"ON"
CHANNEL

Al

Ao

X

L

NONE

X

L

H

L

A2

Al

AO

X

X

L

L

L

L

H

H

2

L

H

H

2

L

H

L

H

3

H

L

H

3

L

H

H

H

4

H

H

H

4

H

L

L

H

5

H

L

H

H

6

H

H

L

H

7

H

H

H

H

8

TYPICAL PERFORMANCE CHARACTERISTICS (Applies to all grades, unless otherwise noted.)
MUX-08
BREAK-BE FORE-MAKE
SWITCHING

MUX-08
LARGE-SIGNAL SWITCHING

RL = lkn, CL = lDpF, V'f, B '" 10V

RL = lMn, C L = 10pF,

VOLTAGE = 2V/DIV
TIME = 200ns/DIV

VOLTAGE = 5V/DIV
TIME = l/-1s/0lV

MUX-08
SMALL-SIGNAL SWITCHING
WITH FILTERING

RL = lMn, CL '" 500pF, Vl '" SOOmV, Va = +500mV
VOLTAGE = SOOmV/DIV
TIME = lMS!DIV

V, = -10V, Vs '" +10V

MUX-08
SMALL-SIGNAL SWITCHING

V, = -50amV, V8 = +500rnV

Rl = lMn, Cl = 10pF,
VOLTAGE SOOmV/DIV

=

TIME'" ')"s/DIV

MUX-08
SMALL-SIGNAL SWITCHING
WITH 21's SAMPLE TIME

V, = -SOamV, Va = +500mV

MUX-08
SMALL-SIGNAL SWITCHING
WITH FILTERING AND
2.51's SAMPLE TIME

RL = lMn, CL = 10pF,
VOLTAGE'" SOOmV/DIV

VOLTAGE = SOOmV/DIV

TIME = SOOns/DIV

TIME = SOOns/Drv

AL = lMn, Cl = 500pF. V, = -SOamV, Va = +500mV

NOTE:
Top waveforms: Digital Input 5v/DIV
Bottom waveforms: Multiplexer Output

13-52

1/86, Rev_ A

-------I~ Mux-oa/MUX-24 a-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

TYPICAL PERFORMANCE CHARACTERISTICS (Applies to all grades, unless otherwise noted.)
MUX-08 CROSSTALK AND
OFF ISOLATION PERFORMANCE
OF CHANNEL 8

MUX-08 CROSSTALK AND
OFF ISOLATION PERFORMANCE
OF CHANNEL 8

~

140

CERAMIC

(0) PACKAGE

i

100

::::

r--.

~

o
z
o

80

~

60

~

40

,iii
.,.
OL-~~~LLUill~LUUill~LU~

lk

10k

lOOk

1M

~

>

IIII~I I I

I

V(+)

=

ISOL~~ION

11111

r-l::-~t-+
~v+.: 15V

1-

~

10

11111

05

RL = 1Mn, CL = 10pF, Vs = +5V RMS
OFF-ISOLATION
RL = lkn, C l = 10pF, Vs = +5V RMS

-

a
1k

10M

1M

lOOk

10k

-50

10M

-25

25
50
75
TEMPERATURE (DC)

FREQUENCY 1Hz)

FREQUENCY (Hz)

"ON" RESISTANCE (RON)
vs ANALOG VOLTAGE (VA)

ENABLE DELAY TIMES
vs TEMPERATURE

~~~-~:~:~~

-1SV
RL = 10Mn
CL _ 10pF

_ 15

CROSSTALK

20

---=t: - -

v -

j

11:::

OFF

+1SV

V(-) = -15V
TA = 2SOC

a

2

~ W~~STH

o
>0

25

f1:""

~ 120
~

TRANSITION TIMES
VB TEMPERATURE

lOa

125

RONVS SWITCH VOLTAGE (VSD)
600

V+ = +15V
v- = -15V
VS2 - VSB = OV
RL = 1 k.l1
-,L--~---"f-7"9---/

20

MUX-OBA, E

I

VSl

='

s
w

VSl = lOV

u

MUX-OSB, F

15

f--+-+-+-+-f-+---1IS = 200j./A

2V

~~m~
10~

05~
~A.E.

f--+-+-+-+-f-+---1~~~ ~~~J­
3001--+-+-+-+-f-+-+:;';:';=r---1
f----+-+-+--+-t-+-+IS '" 200iJA

~

"
0
rr."

200

S
~

f-~-+-+~-f-+--+-t-+---1

300

~o

200

TA

0

25'le

MUX-24A.E -tOFF

-50

-25

25

50

75

100

-10

125

TEMPERATURE (Oe)

,
4
-8 -6 -4 -2
VA - ANALOG INPUT VOLTAGE (VOL TS\

500

r-'
r-

v+ -

15V

v- = -15V
500 ~ TA =25°C

S
w
u

"i;;
~

400

iii

-

~

"0
"
~

"-~ ~1i's
0

300

MUX-OB8,F

MUX.24B,y

A:

,.,

./

L'

MJX-08~.E

MUX-24A,E

I

v- '" -15V
IS'" 100/J-A
VA = OV

§
w

U

"i;;
~

300

z

200

V

"

~

LJ

~1;-~

MUX-'4. ,:::;:

/~

1>;::

v

'" -15V
T A == 2SoC

1.0

13
~

01

;2

;

:/'

'"~ 001
~

100

25
50
75
TEMPERATURE (OC)

13-53

100

125

--

I=--

f--

~-

r--- r-1=
r----

~
r----

0001
-25

800

ID(OFF)

=

I-S b-.

r---50

C-

r--

~
~

100
-2000
- 1200
-400 0 400
1200
2000
-1600
-800
SOD
1600
IS - SWITCH CURRENT (,uA)

MUX-24A,E_

10

MUp!-F--

~

0
I

I

;t1~

SWITCH LEAKAGE
CURRENTSvs
ANALOG INPUT VOLTAGE

V
, / / ~ V""'-

iii
0

100-

200

I

V+ OI'5V
400

/'

100
-800 -600 -400 -200
200 400 600
Vs - SWITCH VOLTAGE (mV)

10

RON vs TEMPERATURE

RON vs SWITCH CURRENT (Is)
600

--

--

V+ '" +15V +-+~-f-+--+-t--j
V- '" -15V +-+~-+-+--+-t--I

~~~:~~~:~-

D

-

I
Z

100 f-~-+-+~-f-+--+-t-+---1

~f'

~

z

0

I

V+ '" +15V
v- = -15V
TA '" 25°C

400

i;;

o

r-r--

r-

w
u

"

f----+-+-+--+-t-+-+~~~:~~!:~

~

500

-15

--

IS(OFF)

I'.
IOION)

\

-10
-5
10
VA - ANALOG INPUT VOLTAGE (VOLTS)

15

1/86, Rev. A

_ _ _ _ _ _-t~MUX-oa/MUX-24 a-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

TYPICAL PERFORMANCE CHARACTERISTICS (Applies to all grades, unless otherwise noted.)

SWITCH LEAKAGE CURRENTS
vs TEMPERATURE
100

125
12

12

~

11

1>-

f-10

V+ _ +15V

1

10V

r,
f---~

"u
w

10

§ 0;: ro~v
~ Vo·

~a:

f---

~

'S~NI

'?IO,,/

+

01

fT

'-!-

7

~

1+
MUX-OSB,F
MUX·24B,F

,

I--- I -

-

Vs+ - +15V
Vs- = -1SV

10

/'

001

1

M1UX.24?,F

1

TEMPERATURE (OC)

MUX-24
SMALL-SIGNAL SWITCHING

MUX-24
SMALL-SIGNAL SWITCHING
WITH FILTERING

-25

25

RL'" lMn, Cl

=

lOpF,

50

75

100

125

v, = -sqOmV,

V4 = +500mV

VOL TAGE

=

SOOmV/DIV, TIME = lj.!s/DtV

MUX-24
SMALL-SIGNAL SWITCHING
WITH FILTERING
AND 2.5JLS SAMPLE TIME

RL = lMn, Cl = 500pF,

v,

= -SOOmV,
V4 = +500mV
VOLTAGE = SOOmV/DIV, TIME = SOOns/DIV

-55

-25

c~BI

,..... . . .

100

125

v.

T

COlOr'

I

--

1

0
25
50
75
TEMPERATURE fOCi

-55

i--i--

r--

r--

MUX-OBB,F

'stOFF)

I

v-· -15V

r--

MUX·24A,E

2

I),LI
Q

MUX-OBA,E

4

I--

TA=25 C

r--

9

B I'-f':::
~

!!}

f--- r--

MUX 08A,E
MUX-24A,E

2

': 5

~

1:'
~

-;/

1+

>6

'""
"
x

- - 'OIO~

MUX-08
SWITCH CAPACITANCES
vs ANALOG INPUT VOLTAGE

SUPPLY CURRENTS
vs TEMPERATURE

1

es{OFF)

......

I
-12

-8

-4

12

VA - ANALOG INPUT VOLTAGE (VOLTS)

RL'" lMfl, CL '" 500pF, Vl '" -500mV,
V4 '" +500mV
VOLTAGE'" 500mV/DIV, TIME'" l"s!DIV

MUX-24
BREAK-BEFORE-MAKE
SWITCHING
2Y

IpS

2L

20(N

RL'" lkfl, CL '" 10pF, Vl, 4 '" lOV
VOLTAGE'" 2V!DIV, TIME'" 200ns/DIV

MUX-24
SMALL-SIGNAL SWITCHING
WITH 2JLs SAMPLE TIME

v,

RL = lMn, CL = 10pF,
= -SOOmV,
V4'" +500mV
VOLTAGE'" 500mV!DIV, TIME'" 500ns!DIV

MUX-24
LARGE-SIGNAL SWITCHING

RL eo lMfl, CL '" 10pF, Vl '" -10V, V4 '" +10V
VOLTAGE'" 5V!DIV, TIME'" l/-1s!DIV

NOTE:
Top waveforms: Digital Input 5V1DIV
Bottom waveforms: MUltiplexer Output

13-54

1/86, Rev. A

--------1~ Mux-oa/MUX-24 a-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

TYPICAL PERFORMANCE CHARACTERISTICS (Applies to all grades, unless otherwise noted.)

DIGITAL INPUT CURRENTS
VB TEMPERATURE

l+v- ,~v

=
= -15V

MUX-24
CROSSTALK AND OFF
ISOLATION PERFORMANCE
OF CHANNEL 3A

MUX-24
SWITCH CAPACITANCES vs
ANALOG INPUT VOLTAGE
12

I
10

ALL DIGITAL
INPUTS ARE
LOGIC "0"

~ 140

~ l+ .1'5J
v-

=

"~

r---- ........... r-......

IJ/:

-M=-=

,

IINL

-55

--::

_

CS(OFF)

'tf

80

0

II

>=

~)

:l
~
~
~

J-

0

60

V+ = 15V
v- = -15V

"

20

!t
-25

0
25
50
75
TEMPERATURE (OC)

100

125

-12

-8

-4

12

III
I

f!f:VD
Cos

0

~lil~:F' "'"

N'1i~S~ATlO~

I I: I ill r--

TA=25°C
CROSSTALK
RL = lMn, CL = 10pF, Vs = 5V RMS
RON (SWITCH NO 4) = 300n
OFF ISOLATION
RL = lkrl., CL ~ 10pF, Vs = 5V RMS

40

I

>,

hlll

No

~ 100

""0
z

Co (ON)

o

i" [J~~~~TAU i

120

~

-15V

lk

lOOk

10k

VA - ANALOG INPUT VOLTAGE (VOLTS)

1M

m

10M

FREQUENCY (Hz)

A.C. TEST CIRCUITS
BREAK-BEFORE-MAKE TEST CIRCUIT

TRANSITION TIME TEST CIRCUIT

* S1S

+15V

" SlB
S2A, S2B

+1SV

S2A, S2B
S3A, S3B

S3A, S3B

S4B

S4.

v+
+5V

EN

V+

S1(SlA)

NC(DB) S1(SlA)

+10V

Ne(DB)

+5V

A2(Ne}

r----1>---+----I A1

LOGIC
INPUT

LOGIC
INPUT

son

MUX-08
(MUX 24)

I----+-..--ovo
50"

( ) DENOTES MUX-24

Figure 1

S~:~7 t--------,

A2(NC)

.----1>---+----1 A 1

-lV

S8(S4A)

EN

( ) DENOTES MUX-24

Figure 3

ENABLE DELAY TIME TEST CIRCUIT

+lSV

+15V

" S18
S2A, S28
S3A, S3B

II

CROSSTALK MEASUREMENT CIRCUIT
'" SlA, S2A

S4A

S18-S4B

CHANNEL 4(4A) IS ON

S4B, S4A

V+

V+

Ne(DB) 81 (SlA)

+5V

-lV

S1-57

EN
NC(DB)

,----1 EN

1---------,

I-I

A2(NC)

Al
Al

LOGIC
INPUT

AO

MUX-08
(MUX-24l

son

t-.....-.--oVo
son

Figure 2

( ) DENOTES MUX-24

( ) DENOTES MUX-24

Figure 4

13-55

1/86, Rev. A

--------l1fM!} Mux-oa/MUX-24 a-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS
A.C. TEST CIRCUITS

APPLICATIONS INFORMATION
These analog multiplexers employ ion-implanted JFETs in a
switch configuration designed to assure break-before-make
action. The turn-off time is much faster than the turn-on time
to guarantee this feature over the full operating temperature
and input voltage range. Fabricated with Bipolar-JFET processing, special handling as required with CMOS devices, is
not necessary to prevent damage to this multiplexer. Because
the digital inputs only require a 2.0V logiC "1" input level,
power-consuming pull-up resistors are not required for TTL
compatibility to insure break-make switching as is most often
the case with CMOS multiplexers. The digital inputs utilize
PNP input transistors where input current is maximum at the
logic "0" level and drops to that of a reverse-biased diode
(about 10nA) as the input voltage is raised above'" 1.4V.

OFF-ISOLATION MEASUREMENT CIRCUIT
+15V

• SlA. S2A

""

SlB-S4B

ALL CHANNELS
ARE OFF

+04V

v+
EN

51-87 r - - - - - - - ,
(.)

NC(DB}

A2(Ne}

III

AD

50n

(':~~:~:l

r -......-----1SB(53A)

D(DII}r-_......--<>VD

The "ON" resistance, RON, of the analog switches is constant
over the wide input voltage range of -15V to +11V with
VSUPPLY = ±15V. Higher input voltage is tolerable provided
that some form of current limiting is employed (such as that
of an op-amp output stage) to avoid exceeding junction
temperature and power dissipation requirements. For normal
operation, however, positive input voltages should be restricted to 11V (or4V less than the positive supply). This assures
that the VGS of an "OFF" switch remains greater than its Vp ,
and prevents that channel from being falsely turned "ON".
When operating with negative Input voltages, the gate-tochannel diode will be turned on if the voltage drop across
an "ON" switch exceeds -0.6V. While this condition will
cause an error in the output, it will not damage the switch. In
lab tests, the multiplexer output has been loaded with a
O.Q1I'F capacitor in the circuit of Figure 1. With V1 = -10V and
Va = +10V, the logic input was driven at a 1kHz rate. The
positive-going slew rate was 0.3V1I'S which is equivalent to a
normal loss of 3mA. The negative-going slew rate was
0.7V1l's which is equivalent to a "reverse" loss of 7mA. Note
that when switch 1 is first turned "ON" it has a drop of -20V
across its terminals. In spite of that fact, the current is limited
to approximately twice its normal loss.

( I DENOTES MUX-24

Figure 5

SWITCHING TIME WAVEFORMS

SWITCH

OUTPUT
VD
(SEE FIG 1)

SWITCH
OUTPUT
VD
(SEE FIG 2)

CROSSTALK AND OFF-ISOLATION
Crosstalk and off-isolation performance is influenced by
the type of package selected. Epoxy (P) packaged devices
typically exhibit a 12dB improvement in off-isolation
(f = 500kHz) performance when compared to ceramic (0)
packaged devices. Epoxy packaged devices typically exhibit
a 15dB improvement in crosstalk (f = 500kHz) performance
when compared to ceramic (0) packaged devices.

VD

SWITCH
OUTPUT

60%

VD
(SEE FIG 3)

3.5V

SINGLE SUPPLY OPERATION OF JFET MULTIPLEXERS
PMI's JFET multiplexers will operate from a single positive
supply voltage with the negative supply pin at ground
potential. The analog signal range will include ground.

LOGIC INPUT

..%

For complete single supply operation information, refer to
application note, AN-32.

13-56

1/86, Rev. A

-------l~ Mux·oa/MUX·24 a·CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

SIMPLIFIED MUX-08 SCHEMATIC

v'o---__~~~~~~--_+--~--------~------------_.

DRAIN
--i--4>-OOUTPUT

IVol

I
I
I

DECODING
MATRIX

r----+--+------~-----+-----

J

Be.

FOR MUX - 24 SWITCH PAIRS S, - Ss. 82 84 - Sa ARE TURNED ON BY
A REPROGRAMMED DECODING MATRIX AND ~ IS NO LONGER USED

THE COMMON ANALOG BUS IS SPLIT IN HALF TO PROVIDE lORAIN B) OUTPUT

The simplified MUX-oS/MUX-24 schematic shows that logic
trip points are determined by two forward diode drops. An
Internal clamping diode between V- and ground prevents
excessive current flow between V+ and ground in the event
that V- becomes open circuit. The decoding matrix is
accomplished by a programmed diode array. The switch cell
consists of P channel JFET's with appropriate blocking
diodes which ruggedizes the circuit's overvoltage and supply
loss characteristics.

CMRR TEST CIRCUIT
Vsup =±1SV

AO= A,=OV
EN" +5V

PIFFERENTIAL MULTIPLEXERS

Xl000
GAIN

One characteristic unique to differential multiplexers
(MUX-24) is the ability to reject common-mode signals from
becoming differential error signals. Common-mode rejection
is a parameter which defines the amount of rejection in terms
of dB. The MUX-24 exhibits a 106dB at 60Hz and 101dB at
400Hz of CMRR using the test circuit of Figure 6.

CONFIGURATION

Figure 6

13-57

CMRR = 20 log

(VO
)
1000XVj

1/86, Rev. A

-------t1fMD

Mux-oa/MUX-24 a-CHANNEL/DUAL 4-CHANNEL JFET ANALOG MULTIPLEXERS

TYPICAL PERFORMANCE CHARACTERISTICS
OVERVOLTAGE V-I
CHARACTERISTIC

OVERVOLTAGE V-I
CHARACTERISTIC

30

30

TA-25°C

DUAL·8UPPL Y

VN "+5V
VE .. -15V FOR VA>O
Ve -t1aV FOR VA <0

I

I
J
-10
-40

-20

SINGLE-SUPPL Y

fA'" 26°C

OPERATION

Vs'" ±1SV
20

POWER-LOSS V-I
CHARACTERISTIC
30

-

20

TA" 2S G C
vs(+) .. OV (GROUND)
VN'"OV
Ve--1OV

OPERATION

Vs(+) <= 1SV
VsI-J '" OV (GROUND)

VN"'t5V
VE" -10V

20

~ r--

f.-

/
20

-10
-40

40

-20

I
20

VAIVOLTSJ

40

-10
-40

-20

0

--

20

40

VAIVOlTSI

VAIVOLTSI

OVERVOLTAGEIPOWER-LOSS MEASUREMENT TEST CIRCUIT
VI+I .. +16V

-'F

VI+I
.. (S4AI

0----1 NC(OBI
MUX·08
A2INe) (MUX-24)

+60V

o----1~--;Al

SI(~~71-

_ _ _--,

AO

lMn

V(-J· -15V
( I DENOTES MUX·24

13-58

·S1A-S3A
S1B-S48

1/86, Rev. A

MUX-16/MUX-28
16-CHANNEL/DUAL 8-CHANNEL
JFET ANALOG MULTIPLEXERS (OVERVOLTAGE PROTECTED)
Precision MOllolithics Inc.

FEATURES

GENERAL DESCRIPTION

• JFET Switches Rather Than CMOS

The MUX-16 is a monolithic 16-channel analog multiplexer
which connects a single output to 1 of the 16 analog inputs
depending upon the state of a 4-bit binary address. Disconnection of the output is provided by a logical "0" at the
ENABLE input, thereby providing a package selection
function.

•
•
•
•
•
•
•
•
•
•
•

Highly Resistant To Static Discharge Damage
No SCR Latch-up Problems
Low "ON" Resistance - 2900Typlcal
Low Leakage Current
Digital Inputs Compatible With TTL and CMOS
Break-Before-Make Action
12So C Temperature-Tested Dice Available
Overvoltage Protected
Supply Loss ProteclJon
MUX-1B Pin Compatible With DGS06, HI-SOBA, AD7S0B
MUX-28 Pin Compatible With DG507, HI-507A, AD7507

ORDERING INFORMATIONt
PACKAGE

25°C
RESISTANCE

HERMETIC DIP

2900"
2900
40004000
2900"
2900
40004000

MUX16AT
MUX16ET
MUX16BT
MUX16FT
MUX28AT
MUX28ET
MUX28BT
MUX28FT

LCC

TEMPERATURE
RANGE

MUX16BTC/883

MIL
INO
MIL
INO
MIL
INO
MIL
INO

"For devices processed in total compliance to MIL-STD-883, add /883 after
part number. Consult factory for 883 data sheet
commercial and industrial temperature range parts are available With
burn-in. For ordering information see 1986 Data Book, Section 2.

t All

The MUX-28 is a monolithic 8-channel differential analog
multiplexer configured in a double pole, 8-position (piUS
OFF) electronic switch array. A 3-bit binary input address
connects a pair of independent analog inputs from each
8-channel input section to the corresponding pair of independent analog outputs. Disconnection of both inputs is
provided by a logical "0" at the ENABLE input, thereby offering a package select function.
Fabricated with Precision Monolithics' high performance
Bipolar-JFET technology, these devices offer low, constant
"ON" resistance. Performance advantages include low leakage currents and fast settling time with low crosstalk to
satisfy a wide variety of applications. These multiplexers do
not suffer from latch-up or static discharge blow-out problems associated with similar CMOS parts. The digital inputs
are deSigned to operate from both TTL and CMOS levels
while always providing a definite break-before-make action
without the need for external pull-up resistors. For single
8-channel and dual 4-channel models, refer to the
MUX-08/MUX-24 data sheet.

FUNCTIONAL DIAGRAMS

MUX-16

EN
A3
A2

~o--t~~--~~~-r~--r-~

MUX-28

V+

EN

GND

A,

v-

~~II~~~~-r~~~--~

A2

13-59

v+
GND

v-

1/86, Rev. A

-------I~

MUX-16/MUX-28 16-CHANNEL/DUAL 8-CHANNEL JFET ANALOG MULTIPLEXERS

ABSOLUTE MAXIMUM RATINGS(Ratings apply to both DICE and packaged parts, unless otherwise noted.}
Operating Temperature Range,

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300°C

MUX-16/28-AT, BT, BTC •••••••••••••
MUX-16/28-ET, FT •••••••••••••••••••
Dice Junction Temperature (Tj ) ••••••
Storage Temperature Range •••••••••••

-55°C to +125°C

Maximum Junction Temperature •••••••••••••••••• 150"C

-25° C to +85° C

V+ Supply to V- Supply ••••••••••••••••••••••••••••

-65°C to +150°C
-65"C to +150"C

36V

Logic Input Voltage •••••••••••• (V- or -4V) to V+ Supply
Analog Input Voltage •.•. V-Supply-20Vto V+Supply+20V

Power Dissipation ••••••••••••••••••••••••••••• 1000mW

Maximum Current Through Any Pin

••••••••••••••• 25mA

ELECTRICAL CHARACTERISTICS at Vs= ±15V and TA = 25°C, unless otherwise noted.

MUX-16A/E
MUX-28A/E
PARAMETER

SYMBOL

CONDITIONS

"ON" Resistance

RON

Vs c5 10V. Is c5 200~A

-----------

I

J

I

FOR MUX-28 SWITCH PAIRS 5, - 59_ 52 - 5'0S8 - 5'6 ARE TURNED ON BY
A REPROGRAMMED DECODING MATRIX AND A31S NO LONGER USED
THE COMMON ANALOG BUS IS SPLIT IN HALF TO PROVIDE (DRAIN B) OUTPUT

v-

13·68

1/86, Rev. A

MUX-88
8-CHANNEL ANALOG MULTIPLEXER FOR PCM CODECS
(OVERVOLTAGE PROTECTED)
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

•

The MUX-88 is a monolithic eight-channel analog multiplexer
ideally suited to shared-channel PCM CODEC systems.
One-of-eight channels is selected upon the decoding of a3 bit
binary address. An enable input (En) disables all switches
when logic low providing package select. All logic control
inputs have true TTL input compatibility eliminating the need
for pull-up resistors necessary for some CMOS equivalent
products.

Compatible with Standards for Noise and Crosstalk In
Telephony Systems
• Pin Compatible with DGS08, HI-S08A, LF11S08
• JFET Switches Rather Than CMOS
• Low "ON" Resistance - 220.0 Typical
• Low Output Leakage Current - 100nA Max
• Digital Inputs Compatible with TTL and CMOS
• Input Overvoltage and Supply Loss Protected

Fabricated with Precision Monolithics' high performance
Sipolar-JFET technology, this device offers low "ON" resistance, low leakage, fast settling time and excellent crosstalk
isolation (98dS @ 20kHz). These characteristics make this
device suitable for meeting system level communication
requirements in shared-channel PCM CODECs.

ORDERING INFORMATIONt
RON

MODEL

TEMP RANGE

400.0
520.0

MUX-88EQ
MUX-88FQ

INO
INO

tAil commercial and mdustnal temperature range parts are available with
burn-in. For ordenng information see 1986 Data Book, Section 2

Additional ruggedization results from built-in overvoltage,
supply loss, and latch-up free circuit characteristics.

PIN CONNECTIONS
FUNCTIONAL DIAGRAM

ENABLE

A2

AO

A1

TOP VIEW
16-PIN HERMETIC DUAL-IN-LiNE
(Q-Sufflx)

v+
GND

TRUTH TABLE
v-

"ON"

DRAIN

58

57

56

S5

54

53

52

51

13-69

EN

CHANNEL

X

L

NONE

L

H

L

H

H

2

H

L

H

3

A2

A1

AO

X

X

L

L

L
L
L

H

H

H

4

H

L

L

H

5

H

L

H

H

6

H

H

L

H

7

H

H

H

H

8

1/86, Rev. A

II

------------;/fMD

Mux-aa a-CHANNEL ANALOG MULTIPLEXER FOR PCM CODECS

ABSOLUTE MAXIMUM RATINGS (TA = 2SoC, unless otherwise noted)
Operating Temperature Range,
MUX-88EQ, FQ ....................... -2So C to +8So C
Storage Temperature Range. . . . . . . . . .. -6So C to + 1S0° C
Power Dissipation .............................. SOOmW
Derate above 100°C ........................ 10mW/oC
Lead Temperature (Soldering, 60 sec) ............. 300° C

V+ Supply to V- Supply. . . . . . . . . . . . . . . . . . . . . .. . . . .. 36V
V+ Supply to Ground ............................... 18V
Logic Input Voltage (Note S) ........... (V- or -4V) to V+
Analog Input Voltage .... V- Supply-20V to V+ Supply +20V
Maximum Current Through Any Pin ............... 2SmA

ELECTRICAL CHARACTERISTICS for V+ = -1SV and -2soC:5 TA:5 8SoC, unless otherwise noted.
MUX-88E
PARAMETER

SYMBOL

CONDITIONS

"ON" Aesistance

Vs = OV. Is = 200~A

t.AON With Applied Voltage

-10V:5 Vs:5 10V, Is =

MIN

TYP

MUX-88F

MAX

200~A

1.5

RON Match

Vs = OV, IS = 200~A

'SIOFF)

Vs= 10V, Vo= -10V, (Note 1)

10

Vs= 10V, Vo = -10V, (Note 1)

1010FF)

1010N)+ I SION ) Vo= 10V, (Note 1)

Digital "1" Input Voltage

V,NH

Digital "0" Input Voltage

MAX

520
4.5

RON Match Between SWitches

Drain Current (SWitch "OFF")

TYP

400

Source Current (Switch "OFF")

Leakage Current (Switch "ON")

MIN

25

UNITS
il

%

30

il
10

nA

100

100

nA

100

100

nA

0.8

0.8

V

20

20

~A

20

20

~A

15

15

mA

(Note 5)

V

(Note 5)

Digital Input Current
Digital "0" Enable Current

I)NLIEN)

Positive Supply Current

1+

All Digital Inputs Logic "0"

Negative Supply Current

1-

All Digital Inputs Logic "0"

Switching Time

trAAN

Figure 1, (Note 2)

1.8

Output Settling Time

ts

10V Step 0.10%
10V Step 0.05%
10V Step 0.02%

1.3
1.5
2.3

mA
21

2.2

2.5

~s

1.7
1.9
2.5

~s

Break-Belore-Make Delay

t OPEN

0.8

1.0

~s

Enable Delay "ON"

tONIEN)

1.0

1.2

~s

Enable Delay "OFF"

tOFFIEN)

0.2

0.2

~s

"OFF" Isolation

ISO OFF

88

88

dB

Crosstalk

dB

(Note 4)

CT

(Note 3)

98

98

Source Capacitance

CSIOFF)

Switch "OFF", Vs = OV, Vo = OV

2.5

2.5

pF

Drain Capacitance

COIOFF)

Switch "OFF", Vs= OV, Vo= OV

7

7

pF

Input to Output Capacitance

COSIOFF)

(Note 4)

0.3

0.3

pF

NOTES:
1. Condltions applied to leakage tests insure worst case leakages. Exceeding l1V on the analog input may cause an "OFF" channel to turn "ON"
2. Sample tested. The measurement conditions of Figure 1 Insure worst case
tranSition time.
3. Crosstalk is measured by driving channel 8 with channel 4 ON.
A L = lMil, C L = 1OpF, Vs = 5V AMS, I = 20kHz. (See Figure 2)
4. OFF Isolation is measured by driving channel 8 With ALL channels OFF
AL = 1kil, C L = 10pF, Vs = 5V AMS, I = 20kHz Cos is computed Irom the
OFF isolation measurement.
Guaranteed by RON and leakage current testing. For normal operatIon
maximum analog signal voltages should be restricted to less than
(V+) -4V.
DICE
For applicable DICE ,"Iormatlon see MUX-08/MUX-24 data sheet

13-70

1/86, Rev. A

----------I~ Mux-aa a-CHANNEL ANALOG MULTIPLEXER FOR PCM COOECS

TYPICAL PERFORMANCE CHARACTERISTICS
SWITCH LEAKAGE CURRENTS
VB ANALOG INPUT VOLTAGE

SWITCH LEAKAGE CURRENTS
vs TEMPERATURE

'0

'00
V+ - +15V
-15V

v+

+15V

"

1

TA

vs

+10V

'0

v- = -15V

v- - -1SV

v-

1

+2SOC

'0

i!!
a:

"'

'S(oFF)

""'"

,I

:t

-5

'O'ONI~

'0

1

'5

U

>-

~

I-'

i--' .....

.....

cO(OFF)

....

0.'

~

'S(OFFI

V

0.01

-55

-25

25

50

is(Oi F1

'OIONI-

15

100

125

-'2 -10 -8 -6 -4 -2

TEMPERATURE (OC)

VA - ANALOG INPUT VOLTAGE (VOLTS)

LARGE-SIGNAL SWITCHING

....

....

I I I.

~~

TA =+2SoC

-

~

1\
-10

'OIOFFI==

u
w

0001
-15

'0 ~VO"-'OV

::>

~~

0.01

1

V+ = +15V

iiia:

'D(OHI
0'

SWITCH CAPACITANCE vs
ANALOG INPUT VOLTAGE

0

2

4

6

8

10 12

VA - ANALOG INPUT VOLTAGE (VOLTS)

TRANSITION TIMES
VB TEMPERATURE

BREAK-BEFORE-MAKE SWITCHING
v+

= +-'6V

v- = -15V

20

RL

MUX-BB'=

10M!)

CL '" 10pF

J,.
w
i=

MUX-BBE=

'.5
10

05

'R L = 10Mil, C L = 10pF, v, = -10V, VB = +10V
Voltage = SV/Div, Ti me = 1~s/D'v, See
Transition Time CirCUit of Figure 1.

SMALL-SIGNAL SWITCHING

'Voltage= 500mVlDiv, Time= SOOns/DIV, See
Break-Before-Make Circuit of Figure 3.

-50

25

-25

50

75

100

125

TEMPERATURE (OC)

SMALL-SIGNAL SWITCHING
WITH FILTERING

ENABLE DELAY TIME
vs TEMPERATURE
2.5

r=:t===[=+=::;t:::;,:j===F=t=::::=::]
V+ +15V
MUX-88E
=

v- = -15V

20

RZ = lkn

MUX-88F
tON

VSl = +10V

VS2'" Vsa= OV

MUX-88F

VS 1

+2V

MUX-88E
tON

0.5

'R L = 1Mil, C L = 10pF, V, = -SOOmV,
Vss = +SOOmV Voltage = SOOmV/D,v,
Time = 1p.s/Div. See Transition CirCUit of
Figure 1

'RL = 1 Mil, C L = SOOpF, V, = -SOOmV,
VSB = SOOmV Voltage = SOOmV/Div,

Time

=

11ls/Div, See TransitIon Time

Circuit of Figure 1.

-50

-25

25

MUX-88E

MUX-88F

tOFF

tOFF

50

75

100

125

TEMPERATURE (DC)

NOTE:
'Top Waveforms: Digital Input 5V/Div
Bottom Waveforms: Multiplex Output

13-71

1/86, Rev. A

----------I~ Mux-aa a-CHANNEL ANALOG MULTIPLEXER FOR PCM CODECS

TYPICAL PERFORMANCE CHARACTERISTICS (continued)
OFF PERFORMANCE
OF CHANNEL 8

RON VB SWITCH VOLTAGE (Vso)

RON vs SWITCH CURRENT (Is)
600

600

500

S
w

""~

300

I

40

-t

"0

CROSSTALK
RL'" lMn, CL = 10pF, Vs '" +5V RMS,
RON (SWITCH 4) = 300n

20

~~Jo
~
1

r--

0

v- = -15V

w

MUX-88F

~

200

RL = lkrl, CL = 10pF, Vs = +5V RMS

~

- -

~

10k

lOOk

10M

1M

TA =+2SoC

300

r--

'

r--

D

MUX-88~

~~ll'
./

MUX-88E

I

"o

~

200

100

100
1k

400

~

MUX-88E

OFF ISOLATION

>~

v+ = +15V
500

1-'
400

~

"0

rV+- +15V
r-15V
r- Tv-=
A '" +2SoC

-BOO -600 -400 -200

0

200

400

600

-2000-1600-1200-800-400 0

800

400 800 1200 16002000

IS - SWITCH CURRENT blA)

V SD - SWITCH VOLTAGE (mV)

FREQUENCY (Hz)

TYPICAL APPLICATION
EIGHT-CHANNEL SHARED CODEC PCM ENCODER

SUCCESSIVE
APPROXIMATION
REGISTER (SAR)

DIGITAL
OUTPUTS

25k.l1
SIGN

-=

BIT

t 1=H:+:+=::::g 1

CHOAD
BITS

I

CHl

STEP
BITS

CH2
CH3

VREF
+10V

CH4

WAVEFORMS

All

CH5

1'-.- ;::~f

CH6

12 IREF

CHANNEL INPUT ANALOG
A12

CH7

20kn
CH8

MUX OUTPUT SIGNAL

-15V
SAMPLE/HOLD S/H COMMAND

+15V

~

CHANNEL SELECT
LOGIC
SAMPLE/HOLD OUTPUT

CROSSTALK IN PCM SYSTEMS
In PAM or PCM systems crosstalk specifications for components, such as multiplexers, are related to overall system
crosst~lk specifications in a complex manner. Component
specification must, of necessity, refer to the operation of
the multiplexer in a non-sampling mode of operation. When
rapid sequential sampling takes place, such as would be
the case with a typical shared-channel CODEC, crosstalk
will be caused by the off isolation properties of the

multiplexer as well as by storage elements on chip and PC
card stray capacitance. For example, the capacitance has
the effect of conferencing the channels and increasing
crosstalk. Thus, system crosstalk in a shared-channel PCM
CODEC is influenced by multiplexed characteristics as well
as PC card layout and the timing relationship between the
multiplexer and the sample-hold circuit.

13-72

1/86, Rev. A

----------I~ Mux-aa a-CHANNEL ANALOG MULTIPLEXER FOR PCM CODECS

A.C. TEST CIRCUITS
TRANSITION TIME
INPUT
DRIVE

CROSSTALK MEASUREMENT CIRCUIT

5V~

OV

+1SV

RL

eo

15V

lOMn

c L = lOpF
v+
EN

S1

A2

52·57

+10V

MUX·88
A1

S8

AO

D

-lOV

-v

GND

I

RL

-15V

-15V
C

L

-=-

Off ISOLATION MEASUREMENT CIRCUIT

BREAK-BEfORE-MAKE DELAY
INPUT 5V-L""""'I!
DRIVE ov-.T
"--

Figure 2

+15V
+15V
ALL CHANNELS ARE OFF

LOGIC
INPUT .n.
-15V

-15V

Figure 3

Figure 4

APPLICATIONS INFORMATION
These analog multiplexers employ ion-implanted JFETs in a
switch configuration designed to assure break-before-make
action. The turn-off time is much faster than the turn-on
time to guarantee this feature over the full operating
temperature and input voltage range. Because the digital
inputs only require a 2V logic "1" input level. power-consuming
pull-up resistors are not required for TTL compatibility to
insure break-before-make switching as is most often the case
with CMOS multiplexers. The digital inputs utilize PNP input
transistors where input current is maximum at the logic "0"
level and drops to that of a reverse-biased diode (about 10nA)
as the input voltage is raised above ~ 1.4V.
The "ON" resistance. RON. of the analog switches is constant
over the wide input voltage range of -15V to +11V with
VSUPPLY = ±15V. Higher input voltage is tolerable provided
that some form of current limiting is employed (such as that
of an op-amp output stage) to avoid exceeding junction
temperature and power dissipation requirements. For normal

operation. however. positive input voltages should be
restricted to 11V (or 4V less than the positive supply). This
assures that the VGsof an OFF switch remains greater than its
Vp • and prevents that channel from being falsely turned ON.
When operating with negative input voltages. the gate-tochannel diode will be turned on if the voltage drop across an
ON switch exceeds -0.6V. While this condition will cause an
error in the output. it will not damage the switch. In lab tests.
the multiplexer output load capacitor has increased to 0.011'F
in the Transition Time circuit. Figure 1. With VS 1 =-10V and
VS8 = + 10V. the logic input was driven at a 1kHz rate. The
positive-going slew rate was 0.3V1l'sec which is equivalent to
a normal loss of 3mA. The negative-going slew rate was
0.7V/l'sec which is equivalent to a "reverse" loss of 7mA.
Note that when switch 1 is first turned ON it has a drop of
-20V across its terminals. In spite of that fact. the current is
limited to approximately twice its normal loss.

13-73

1/86, Rev. A

DMX-88
8-CHANNEL ANALOG DE-MULTIPLEXER
(LOW CHARGE TRANSFER)
Precision Monolithics Inc.

FEATURES
• Low Charge Transfer - 1SpC Typ
• Compatible with Standards for Noise and Crosstalk In
Telephony Systems
• Pin Compatible with DG50S, HI-50SA, LF1250S/1350S
• JFET Switches Rather Than CMOS
• Low "ON" Resistance - 2200 Typ
• Low Output Leakage Current - 100nA Max
• Digital Inputs Compatible with TTL and CMOS
• No Pull-up Reslltors Required to Ensure
Break-Before-Make Action with TTL Inputs

ORDERING INFORMATIONt
MODEL

TEMP RANGE

4000

DMX88EQ

IND

5200

DMX88FQ

IND

channels. In addition, there is an ENABLE input which
permits turning OFF all channels. Using this function permits
selection of any given circuit in a system employing multiple
devices.
Fabricated with Precision Monolithics' high performance
Bipolar-JFET technology, this device offers low, constant
"ON" resistance. In addition the multiplexer has fast settling
times and low leakage currents necessary to satisfy the
reqUirements of an 8-channel PCM DECODER. This demultiplexer does not suffer from latch-up and is highly
resistant to static charge blow-out problems associated with
similar CMOS parts. The digital inputs are designed to
operate from both TTL and CMOS levels while always
providing a definite break-before-make action without the
need for external pull-up resistors.

PIN CONNECTIONS

tAli commercial and Industrial temperature range parts are available with
burn-In. For ordering Information .ee 1986 Data Book, Section 2.

GENERAL DESCRIPTION
16-PIN HERMETIC
DUAL-IN-LINE
(Q-Sufflx)

The DMX-88 is an 8-channel analog multiplexer optimized
for minimum charge transfer, approximately 4 times lower
than MUX-88 or MUX-OB. This is important when a Multiplexer is terminated in capacitive loads, as in shared-channel
PCM decoder systems. Typical crosstalk at 20kHz is 98dB.
MonOlithic construction makes possible this kind of performance while keeping the price reasonable. The DMX-88
makes use of digital logic to select one-of-eight output

FUNCTIONAL DIAGRAM & TRUTH TABLE

ENABLE

A2

A1

v-

00

00

~

L

NONE

L

H

Ao

X

X

H

H

2

H

L

H

3

L

H

H

H

4

H

L

L

H

H

ro

X

A1

GND

00

"ON"
CHANNEL

Az

v+

s

EN

AD

00

00

m

H

H

H

H

L

H

H

H

H

H

6

8

(INPUT)

"X" = DON'T CARE

13-74

1/86, Rev. A

-----------l~ DMx-aa a-CHANNEL ANALOG DE-MULTIPLEXER

ABSOLUTE MAXIMUM RATINGS

(TA = 25·C, unless otherwise noted.)

Operating Temperature Range,
DMX-88EQ, FQ ••••••••••••••••••••••
Storage Temperature Range •••••••••••

-25·C to +85·C

V+ Supply to V- Supply ............................ 36V

-65·C to +150·C

V+ Supply to Ground ............................... 18V
Logic Input Voltage ............ (-4V or V-) to V+ Supply

Power Dissipation •••••••••••••••••••••••••••••• 500mW
Derate About 100·C •••••••••••••••••••••••• 10mW/·C

Analog Input Voltage

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300·C

ELECTRICAL CHARACTERISTICS for V+

DMX-88E
SYMBOL

CONDITIONS

ON Resistance

RON

Vs = OV, Is =

~RON

~RON

-10V S Vs S 10V, Is =

RON Match

Vs = OV, Is =

RON Match Between Switches

••••••••••••••• 25mA

= 15V, V- =-15V and -25·C:$ TA:$ 85·C, unless otherwise noted.

PARAMETER

With Applied Voltage

•••••• V- Supply -20V to V+ Supply

Maximum Current Through Any Pin

MIN

TYP

200~A

DMX-88F
MAX

MIN

TYP

400
200~A

520

1.5

200~A

25
10
-10

10.4
-15

MAX

+10
-10

UNITS
n

4.5

%

30

n

10.4
-15

V

Analog Voltage Range

VA

(Note 5)

Drain Current (Switch OFF)

1010FF)

Vs = 10V, Vo = -10V (Note 1)

10

10

nA

Source Current (Switch OFF)

ISIOFF)

Vs = 10V, Vo = -10V (Note 1)

100

100

nA

Charge Transfer

at

Rs = 0, C L = 200pF
VIN = 0 (Note 4)

25

pC

Leakage Current (Switch ON)

IDIONI+lsIONI Vo= 10V (Note 1)

100

nA

Digital "1" Input Voltage

V'NH

(Note 5)

Digital "0" Input Voltage

V'NL

(Note 5)

18

25

18

~

p.,

100

V
0.8

0.8

V

Digital "0" Input Current

I'NL

V'N = 0.7V

20

20

~A

Digital "0" Enable Current

IINL(EN)

VEN = 0.7V

20

20

~A

Positive Supply Current

1+

All Digital Inputs Logic "0"

15

15

mA

Negative Supply Current

1-

All Digital Inputs Logic "0"

5

mA

Switching Time

trRAN

Figure 3 CL = 10pF

0.8

1.5

~s

Output Settling Time

ts

10V Step 0.10%
10V Step 0.05%
10V Step 0.02%

1.3
1.5
2.3

1.7
1.9
2.5

~s

Break-Before-Make Delay

t OPEN

Enable Delay ON

tONIEN)

CL = 10pF (Figure 1)

Enable Delay OFF

tOFFIEN,

C L = 10pF (Figure 1)

OFF Isolation

ISO IOFF '

(Note 3)

88

88

dB

Crosstalk

CT

(Note 2)

98

98

dB

2.5

2.5

pF

0.3

0.3

0.8

0.2

Drain Capacitance

CD'OFF)

Switch OFF, Vs = OV, VD = OV

Source Capacitance

CSIOFF,

Switch OFF, Vs = OV, Vo = OV

Input to Output Capacitance

CDS(OFF)

(Note 3)

~s

1.2

~s

0.2

~s

pF
pF

NOTES:
1. Conditions applied to leakage tests insure worst case leakages. Exceeding 11V on the analog input may cause an OFF channel to turn ON.
2. Crosstalk is measured by driving channel 8 with channel 4 ON.
RL = 1Mn, C L = 10pF, Vs = 5V RMS. f = 20kHz. (see figure 2)
3. OFF isolation Is measured by monitoring channel 8 with ALL channels
OFF. RL =1kn, C L = 10pF, Vs=5V RMS, f=20kHz. Cos is computed from
the OFF isolation measurement. (see figure 4)
4. Guaranteed by design.
5. Guaranteed by leakage current and RON tests.

13-75

1/86, Rev. A

......

~

~
~
~

~

0
0

~

------------I1fHD DMx-aa a-CHANNEL ANALOG DE-MULTIPLEXER
TYPICAL PERFORMANCE CHARACTERISTICS
SWITCH LEAKAGE CURRENTS
vs ANALOG INPUT VOLTAGE
10

v+
V
TA

1
~
a:
a:

""
~

100

= +15V

-

15V
25"C

!

~

1.0

-

0.10

~

i

SWITCH LEAKAGE
CURRENTS vs TEMPERATURE

IS (DFF'=

"

0.01

-10

-5

-

-vo

IS (ON)

i

15

TA = 2SOC

10

-lOV

10

I

I

--

I"'""

,- ~ ~

I"'""
IS (OFF)

f.[,

....
.... ""

~

0.10

Co (OFF)
10 (OFF)

10

CSiD~

V+:: +15V
= -15V

v-

Vs - +10V

""

~

\

-15

12

~ ~+ :~f~v

a:
a:

10 (OFF)

0,001

SWITCH CAPACITANCE
vs ANALOG INPUT VOLTAGE

001
-55

/
-25

25

50

75

100

125

-12

TEMPERATURE (DC)

VA -ANALOG INPUT VOLTAGE (VOLTS)

....

IS (DNI'_

LARGE-SIGNAL
TRANSIENT RESPONSE

LARGE-SIGNAL
TRANSIENT RESPONSE

(THROUGH A CLOSED SWITCH)

(THROUGH A CLOSED SWITCH)

-4

-B

12

VA - ANALOG INPUT VOLTAGE (VOLTS)

SWITCHING
TIME vs TEMPERATURE
50

CL=I~ ~

4.0

3.0

/'

V

/

v+ = +15V
v- = -15V
RL'" lOMn

2.0

RL'" lMR, Cl -= 1000pF, VIN = ±10V

10

r--

CL'" 10pF

~

~

-25

+25

+50

+75

+100

TEMPERATURE {OCI

SMALL-SIGNAL
TRANSIENT RESPONSE

SMALL-SIGNAL
TRANSIENT RESPONSE

(THROUGH A CLOSED SWITCH)

(THROUGH A CLOSED SWITCH)

ENABLE DELAY
TIME vs TEMPERATURE
5.0
CL = 1000pF

4.0

3.0

~

TDN~ V
:/

I
V+ = +15V

v-

=>

-15V-

RL = 10Mn

I

20

I

1.0

CL = 10pF-

RL = lMr.!, C L = 1000pF, VIN = ±O.5V

TON (EN)

r--r
-26

I
+25

+50

+75

+100

TEMPERATURE (DC)

NOTE:
Upper Waveform Photos: Input Voltage on SOURCE
Lower Waveform Photos: Output Voltage on DRAIN

13-76

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _ _-;~ DMX-81 I-CHANNEL ANALOG DE-MULTIPLEXER

TYPICAL PERFORMANCE CHARACTERISTICS
CHARGE INJECTION VB
ANALOG VOLTAGE

CROSSTALK AND OFF ISOLATION
PERFORMANCE OF CHANNEL 8
sl~.-"nrr~TTnTIm-","rnr-r,"nm

i:
~
o

r1" "0

40

Hr-H'joj.jjjp..HH++Hl-+t:

~

IIII

~ OOHr-H~~~H++Hl-++~~~H+~
!il
I

40 I-CR'-O'-SUSTllAUlLK-LR-'L-'-="",.UJM'-"--',C--'L-'=-L.l''"OP"-F-,'--'-tttttII

~ 20

Vs'" 5V AMS, RON (SWITCH #4) '" 30an
OFF ISOLATION RL"" lkO, CL '" 10pF,

0[0

VS=5VRMS

r::::: :::::

z

i5
~

600

I

::::::::

030

80

§

r---. ........

!;l
W20

I RS

........

10k

lOOk

1M

W

~

:1:

~
~

J

,fl::::t-

"'Kr---.

"'"
<.>

RS '"

a

r:::::

-6

-4

-2

2

~~'S

z

4

8

200

'DO
-2000

'0

-

z5"c

r./

--..-

300

",0

......

+lSV
-15V

DMX·88F

I

f--...

V+

TA "'

v- '"

400

~

is

I
-10 -8

FREQUENCY (Hz)

500

g

RS=25kn'

'0

'OM

_

~ 'O.~

:::::t--

!:OL....J..J..UJJ.w....w.J.W.iII......w..w.lllL....L.J...U.wJI

,.

RON V8 SWITCH CURRENT (Is)

50

ANALOG VOL lAGE (VOLTS)

~ r--

./'
DMX-88E -

-1200
-400 a 400
1200
IS - SWITCH CURRENT (pAl

2000

A.C. TEST CIRCUITS
ENABLE DELAY TIME

EN

INPUT

SWITCHING TIME

tON(EN). toFF(EN)

tTRAN

5V7L
INPur5V7L

OV

DRIVE

+lSV

OV
EN

AD

A'
+5V
+5V

A2

D81--+-+--o Your

'----'

CL
10pF

Figure 3

Figure 1

CROSSTALK MEASUREMENT CIRCUIT

OFF ISOLATION MEASUREMENT CIRCUIT
+15V

Figure 2

OFF ISOLATION", 20LOG ~

Vs

Figure 4

13-77

1/86, Rev. A

------------I~ DMX-88 8-CHANNEL ANALOG DE-MULTIPLEXER

TYPICAL APPLICATION
FOUR-CHANNEL SHARED CHANNEL PCM CODEC
MULTIPLEXER

DE·MUL TIPLEXER

AID

I

D/A

OAeSAMPLE/HOLD

88/89

ANALOG

ANALOG
CHANNELS

CHANNELS

IN

OUT

ENCODE

DECODE

SECTION

SECTION

CHARGE TRANSFER
TYPICAL CHARGE TRANSFER
OF DMX-88

TEST CIRCUIT

TYPICAL CHARGE TRANSFER
OF CONVENTIONAL JFET
SWITCH

+5,OV
VOUT

AL = 1M.I1
CL '" 1000pF

TOP TRACE. ADDRESS INPUT

TOP TRACE ADDRESS INPUT

BOTTOM TRACE- DRAIN OUTPUT

BOTTOM TRACE. DRAIN OUTPUT

APPLICATIONS INFORMATION
These de-multiplexers employ ion-implanted JFETs in a
switch configuration designed to assure break-before-make
action. The turn-off time is much faster than the turn-on time
to guarantee this feature over the full operating temperature
and input voltage range. Because the digital inputs only
require a 2.0V logic "1" input level, power-consuming pull-up
resistors are not required for TTL compatibility to insure
break-before-make switching as is most often the case with
CMOS devices. The digital inputs utilize PNP input transistors where input current is maximum at the logic "O"level and
drops to that of a reverse-biased diode (about 10nA) as the
input voltage is raised above'" 1.4V.
The ON resistance, RON, of the analog switches is constant
over the wide input voltage range of -15V to +11V with
VSUPPLY = ±15V. Higher input voltage is tolerable provided
that some form of current limiting is employed (such as that
of an op-amp output stage) to avoid exceeding junction
temperature and power dissipation requirements. For normal
operation, however, positive Input voltages should be
restricted to 11V (or 4V less than the positive supply). This
assures that the VGsof an OFF switch remains greater than its
Vp, and prevents that channel from being falsely turned ON.

When operating with negative input voltages, the gate-tochannel diode will be turned on if the voltage drop across an
ON switch exceeds -O.6V. While this condition will cause an
error in the output, it will not damage the switch.

CROSSTALK IN PCM SYSTEMS
In PAM or PCM systems crosstalk specifications for components, such as multiplexers or de-m ultiplexers, are related to
overall system crosstalk specifications in a complex manner.
Component specification must, of necessity, refer to the
operation of the multiplexer in a non-sampling mode of
operation. When rapid sequential sampling takes place, such
as would be the case with a typical shared-channel CO DEC,
crosstalk will be caused by the off isolation properties of the
multiplexer as well as by storage elements on chip and PC
card stray capacitance. For example, the capacitance has the
effect of conferencing the channels and increasing crosstalk.
Thus, system crosstalk in a shared-channel PCM CODEC is
influenced by multiplexer characteristics as well as PC card
layout and the timing relationship between the multiplexer
and the sample-hold circuit.

13-78

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

SAMPLE-AND-HOLD
AMPLIFIERS
Precision Monolithics Inc.

14-3

Introduction

14-3

Definitions

14-6

SMP-I0/SMP-ll
Low-Droop-Rate/Accurate Sampleand-Hold Amplifiers

14-15

SMP-81
Telecommunications Sample-andHold Amplifier

14-2

SAMPLE-AND-HOLD
AMPLIFIERS
Pn:CISHHl MOl1ot.thlCS

Inc

INTRODUCTION
Sample-and-hold amplifiers "sample" an analog
input signal and then "hold" the instantaneous
input value upon the command of a logic control signal. Basically the sample-and-hold is an
"analog memory" where a capacitor serves as
the storage element. Applications in which a
time varying input cannot be tolerated require
sample-and-hold circuits. A fast successiveapproximation analog-to-digital converter is
one application. Data acquisition, data distribution, analog delay and telephony require
sample-hold circuits to "freeze" the analog
signal for further signal processing.

temperature performance. "Super Beta"
transistors provides the high input-impedace
amplifier needed for low droop rate and minimal
signal loading.
The SMP-10 and SMP-11 have different droop
rate and settling hold mode times specifications.
The SMP-81 is characterized for the sampling
requirements found in telecommunications
applications.
In addition to precision sample-and-hold amplifiers, two products with related capabilities are
available. The GAP-01 general purpose analog
processor provides the user with two independently switched transconductance amplifiers, a
unity gain buffer and an uncommitted voltage
comparator. The GAP-Q1 is a non-dedicated
functional block which has a wide variety of
applications. The second device is the PKD-01
monolithic peak detector. This device performs
the peak detector function with accuracies
approaching those obtainable with high cost
hybrid modules at a cost approaching the low
cost, low performance discrete designs. Data
sheets for the GAP-01 and PKD-Q1 are located in
the SPECIAL FUNCTIONS SECTION of this
catalog.

PMI sample-and-hold amplifiers are functionally
identical to track-and-hold circuits. They
continously track input signals during the
sample mode. PMI circuits should not be
confused with AC controlled sample-and-holds.
When an AC controlled sample-and-hold is
commanded to sample, it will take a fast sample
and immediately return to the hold mode. It can
not continously track an input signal.
A sample-and-hold circuit consists of an amplifier, switch, and capacitor. Many specifications
are similar to those of switches and operational
amplifiers - bias currents, voltage gain, and
charge injection are examples. These and other
specifications pertaining uniquely to sampleand-hold circuits are defined below.
The SMP-10 and SMP-11 are precision sampleand-hold amplifiers with high accuracy, low
droop rate, and fast signal acquisition time.
These circuits contain a high impedance input
buffer, a diode bridge switch, a transconductance or "Super-Charger" circuit to enhance
'slewing and a high speed output amplifier. The
"Super-Charger" is capable of supplementing
the capacitor charging current whenever the
difference between input and output levels
exceeds a given threshold. Settling to final value
is under control of currents from the diode
bridge, thus minimizing overshoot and instability.
The inherent low offset voltage errors and low
charge injection allows the residual zero-scale
errors to be actively trimmed using PMI's
zener-zapping technology without degrading

DEFINITIONS
Acquisition Time (taq)- The minimum time forthe
output voltage to begin tracking the input voltage, to within a specified error band, after the
inception of the sample command. By convention, acquisition time is defined for sampling of
a DC level. For instance a circuit which is "holding" a 10V output signal, and operating with
zero input volts, is switched to the sample
mode. The acquisition time is then the time
required for the output to decrease to within a
±10mV (O.01%FS) band about ground potential
(see timing diagram).
Aperture Jitter (~ta) - The maximum amount of
deviation in aperture time from sample to sample. Errors resulting from aperture jitter increase

14-3

SAMPLE-AND-HOLD
AMPLIFIERS
PreciS101l MOJ)olithics Inc.

in proportion to the slew rate of the sampled
analog input signal. Also called aperture
uncertainty time.
Aperture Time (tap) - The time between the
inception of the hold command and the time the
circuit output ceases tracking the input signal
(see timing diagram).

be supplied without significant distortion. Full
power bandwidth Fp is related to slew rate SR
by the following equation:
SR
Fp = 21TEp
Using this equation Fp of 160kHz can be computed. This is applicable only for pulsed conditions. Power dissipation limits Fp to 100kHz for
C.W. operation.

Change In Hold Step (~VHS)- Actual hold
step less the hold step measured after sampling
V = O. A change in hold step has two components:
the first is a function of input voltage, the
second is a function of the rise time of the S/H
voltage. Note that rise time of S/H voltage
dV(SIH)fdt also effects ZERO-SCALE ERROR.

Gain Error - Voltage difference between input
and output voltage measured over a specified
voltage range, assuming the ideal gain is unity.
Hold Capacitor Charging Current (ICH) - The
current ICH which charges, or discharges, the
hold capacitor CH while the circuit is in the
sample mode.

Charge Transfer (Qt) - The amount of charge

transferred to the holding capacitor due to the
action of the switch. Charge is transferred to CH
when the circuit is switched to the hold mode.
Charge transfer causes a change in output
voltage VZS as defined by the equation:

Hold Mode Settling Time (tHm) - The time for
all output transients to settle within a specified
error band. Measured from the inception of the
hold command (see timing diagram).

V (V) = at(pC)
ZS
CH(pF)
Note that for at = 5pC and CH = 5000pF offset
error = 1mY. The SMP-10/11/81 has been factory
nulled for CH = 5000pF. For other values of CH
the zero-scale shift can be calculated from the
equation:
~VZs(V)

=

Hold Step (VHS) - Magnitude of step caused in
the output voltage by switching the circuit from
sample mode to hold mode. Hold step is sometimes called pedestal error, or sample to hold
offset (see timing diagram).
Input Bias Current (IB) - Input terminal current
with input voltage held at zero volts.

at
CH -1mV

Input Resistance (RIN) - AC impedance
measured as a ratio of input voltage VIN to input
current liN.

Droop rate dVCH/dt is
the rate of change of output voltage while the
circuit is in the hold mode. dVCH/dt is a direct
function of droop current lOR:
Droop Rate (dVCH/dt) -

Leakage (Droop) Current (lOR) - The current
which flows out of holding capacitor CH while
the circuit is operating in the hold mode. In general
droop current lOR is defined positive when its
direction is into the CH pin. This parameter is
sometimes called drift current.

dVCH _ lOR
CH

---cit -

where dVcwdt is expressed in p.V/ms, lOR in
nanoamperes and CH in microfarads (see timing
diagram).
Feedthrough Attenuation Ratio (FA) - Feedthrough attenuation is a measurement of the
off-isolation of the analog switch (specified in
dB). The parameter is a direct function of feedthrough capacitance.

Linearity Error - The maximum deviation from
an ideal straight line drawn between the output
voltage when VIN = 0 and the output voltage
when VIN = maximum analog voltage, expressed
as a percentage of the maximum analog voltage.
Output Resistance (Ro) - An AC change in output voltage as a result of an AC change in load
current.

Full Power Bandwidth (Fp) - The maximum

frequency at which rated output voltage Ep can
14-4

SAMPLE-AND-HOLD
AMPLIFIERS
Precision Monolithics Inc.

Power Supply Rejection Ratio (PSRR) - The
change in output voltage for a change in power
supply voltage when the circuit is in the sample
mode. The best power supply rejection ratio
PSRR is obtained with the power supply voltage
changing at a very low rate (DC). For essentially
DC conditions PSRR for the hold mode of operation is essentially the same as the PSRR for the
sample mode. PSRR is degraded as the frequency of the disturbance increases.
Sample Hold Current Ratio (lCWIDR) - The
ratio of the peak charging current available to
the droop current.
Signal Transfer Nonlinearity - The total input
to output, hold mode error caused by gain nonlinearity, feedthrough, thermal transient, charge
transfer and droop rate. These error terms cannot be corrected by offset and gain adjustments.
Slew Rate (SR) - The maximum possible rate
of change of the output voltage when supplying
the rated output. For a sample-and-hold circuit,
slew rate must be defined with a specified value
of holding capacitor CH. Slew rate can either be
measured by operating the circuit in the sample
mode and applying a step function to the input,

or by applying an input voltage which differs
from the output voltage, with the circuit in the
hold mode, then switching to the sample mode
and observing the rate of change of the output
voltage.
Total Error - The algebraic sum of the following
factors:
i. ZERO-SCALE ERROR
ii. Gain Error
iii. Hold Step Change versus

iv. Hold Step Change versus VIN
Voltage Gain (Av) - The ratio of the output voltage
to the input voltage with the circuit operating in the
sample mode.
Zero-Scale Error (Vzs)- The magnitude of the output voltage when the circuit is switched from sample to hold mode while holding the input at zero
volts. ZERO-SCALE ERROR Vzs is the algebraic
sum of the offset voltage and the charge transfer
hold step voltage (see timing diagram). Vzs can
be adjusted to zero (see ZERO-SCALE ERROR
null adjustment).

TIMING DIAGRAM
HOLD MODE

LOGIC"l"
VSlH
LOGIC "0"

SAMPLE MODE
APERTURE TIME

10

VOLTAGE
(VOLTS)

ACQUISITION
TIME

HOLD MODE
SETTLING
TIME

ZERO-SCALE ERROR
DETAIL

~"""'

_ _"":"_ _ _ ANALOG OUTPUT

I
,

Ov___ _ _ _ _
OFFSET ERROR
Vos......;.....;..;.~~-I-J

dV~~/H)

ZERO-SCALE
ERROR (Vzs)

_ _ _ _ _ _ _ _ _ -ANALOG INPUT

--t-

t

14·5

-------------

SMP.I0/SMP.ll
LOW-DROOP-RATE/ACCURATE
SAMPLE-AND-HOLD AMPLIFIERS
PrCC1SIOll l"lon()l~thics

Inc

FEATURES,

GENERAL DESCRIPTION

SMP·10
• Low Droop Rate ............................. 5.0"Vlml
• Low Signal Transfer Nonlinearity ...•.•..•...... 0.005%
• High Sample/Hold Current Ratio ...•......•....• 2x10'
SMp·11
• Low Droop Rate over Tempereture ..... , ...... 120"Vlml
• High Sample/Hold Current Ratio •...•.......... 1.7x108

The SMP-10/11 are precision sample-and-hold amplifiers that
provide the high accuracy, the low droop rate and the fast
acquisition time required in data acquisition and signal processing systems. Both devices are essentially noninverting
unity gain circuits consisting of two very high input impedance buffer amplifiers connected together by a diode bridge
switch.

BOTH SMP·10 AND SMp·11
• FaIt Acquilition Time, 10Y Step to 0.1% ........... 3.5,,1
• High Slew Rate.. .. .. .. .. .. . .. . .. .. .. . .. ... .... 10Vl,,1
• Low Aperture Time .............................. 50nl
• TrImmed for Minimum Zero-Scale Error ......... 0.45mY
• Feedthrough Attenuation Ratio ................... 96dB
• Low Power DI88lpatlon .........•.............. 160mW
• DTL, TTL & CMOS Compatible LogiC Input
• HA-2420, HA-2425, SHM·IC·1, and AD583 Socket
Compatible

HIGH ACCURACY AND LOW DROOP RATE
The high input impedance and the low droop rates of the
SMP-10 and the SMP-11 are achieved by using bipolar Darlington circuits and an ion implant process that creates
"super beta" transistors.
The output buffer's input stage converts to a super beta
Darlington configuration during the hold mode, which results
in a very low droop rate with no penalty in acquisition time.
The use of bipolar transistors achieves a low change in droop
rate over the operating temperature range.

ORDERING INFORMATIONt

FAST ACQUISITION
A unique super charger provides up to SOmA of charging
current to the hold capacitor, which results in smooth, fast
charging with minimum noise. As the hold capacitor voltage
nears its final value, the low current diode bridge controls the
final settling time. This unique combination of linear functions in a monolithic circuit enables the system designer to
achieve superior performance.

TA=+25°C
PACKAGE

VZI
(mV)

DROOP
RATE IN
p.Vlma

14-PIN DIP
HERMETIC

1.5
3.0
1.5
3.0
1.5
3.0
1.5
3.0
7.0

20
50
20
50
200
500
200
500
900

SMP10AY'
SMP10BY'
SMP10EY
SMP10FY
SMP11AY'
SMP11BY'
SMP11EY
SMP11FY
SMP11GY

LCC

SMP11BRC/883

OPERATING
TEMPERATURE
RANGE
MIL
MIL
COM
COM
MIL
MIL
COM
COM
COM

FUNCTIONAL DIAGRAM

• For devices processed In total compliance to MIL-STD-883. add 1883 after
part number. Consult factory for 883 data sheet.
tAli commerCial and industrial temperature range parts are available with
burn-in For ordering Information see 1986 Data Book, Section 2

(VLC)
LOGIC
CONTROL

PIN CONNECTIONS

OUTPUT

INPUT ,,"",--",......,

3

14·PIN DIP (Y·Sufflx)
• PinS 1 and 8 are not Internally
connected, In unity gain applications,
SMP-10 and SMP-11 can replace
HA-2425, HA-2420, SHM-IC-1 and
AD-583 directly.

4

NULL

.. SamplelHold Control

5tH

MODE

1

Sample
Hold

o

SMP-11BRC/883
LCCPACKAGE
(RC-Sufflx)

VLC = OV for TTL Input compatibility.

Manufactured under the following patents: 4.105,215 and 4.142,117.

14-6

1/86, Rev. A

--------llEMD

SMP-10/SMP-11 LOW-DROOP-RATE/ACCURATE SAMPLE-AND-HOLD AMPLIFIERS

ABSOLUTE MAXIMUM RATINGS (Note)
Supply Voltage (V+ minus V-) ...................... 36V
Power Dissipation .............................. 500mW
Derate Above 100· C .......................... 10mW/· C
Input Voltage. . . . . . . . . . . . . . . . . .. Equal to Supply Voltage
Logic and Logic Reference
Voltage ...................... Equal to Supply Voltage
Output Short-Circuit Duration ................. Indefinite
Hold Capacitor Short-Ci rcuit Duration ............ 60 sec
Storage Temperature Range. . . . . . . . . .. -65· C to + 150· C

Lead Temperature (Soldering, 60 sec) ............. 300·C
Operating Temperature Range
SMP-10AY, BY ...................... -55·C to +125·C
SMP-10EY, FY .......................... 0·Cto+70·C
SMP-11AY, BY, BRC .................. -55·C to +125·C
SMP-11 EY, FY, GY ........................ O·C to 70·C
DICE Junction Temperature (TI ) ....•.. -65·C to +150·C
NOTE: Absolute ratmgs apply to both DICE and packaged parts. unless
otherwise noted.

ELECTRICAL CHARACTERISTICS at Vs = ±15V, CH = 0.005/,F, VLC connected to ground, TA = 25·C, unless otherwise
noted.
SMP-10AlE
SMP-11A1E
PARAMETER
Zero~Scale

Error
(Hold Mode)

SYMBOL CONDITIONS
Vzs

MIN

V'N=O
VS/H = 35V. (Note 3)

Input Bias Current

Droop Rate

TYP

MAX

0.45
35

Leakage
(Droop) Current

Dev,ce Warmed Up SMP-l0
(See Note 2)
SMP-ll
dVm/dt

Input Resistance

MIN

MAX

TYP

MAX

UNITS

1.5

0.60

3.0

1.5

70

mV

65

55

20

90

90

160

nA

30

MIN

0.25
250

20
200

5
60

SMP-11G

TYP

0.10
1.00

Device Warmed Up SMP-l0
(See Note 2)
SMP-ll
See Note 1

SMP-10B/F
SMP-11B/F

5
70
14

4.5

50
500

BO

900

nA
~V/ms

25

20

G!1

0.99963 0.99983

0.99953 0.99978

0.99940 099975

V/v

3.5

35

3.5

~s

5.0

5.0

5.0

~s

50

50

50

ns

1.5

1.5

15

5

5

10

10

50

50

Sample Mode
Voltage Gai n

Av

V'N = ±10V, RL = 5k!1
or V'N = ±5V, RL = 2.5k!1
10V step to withm 10mV
of final value (0.1%)
10V step to withm 1.0mV
offinal value (0.01%)

Acquisition Time

Aperture Time
Hold Mode
Settling T,me

Settling to lmV
of final value.

Charge Transfer

Ot

Slew Rate

SR

SMP-l0
SMP-ll

V'N= ±10V
RL = 2.5k!1

10

Hold Capacitor
Charging Current

30

Sample/Hold
Current Ratio

SMP-l0
SMP-ll

Feedthrough
Attenuation Ratio

Input = 20Vp_p 1kHz
RL = 5k!1, (Note 1)

Full Power
Bandwidth

±10Vp-p
(Dissipation Limited)

Input Voltage Range
and/or Output
Voltage Swing

RL = 2.5k!1

Output Resistance
Power Supply
RejectIOn RatiO

Power
Consumption (DC)

3Xl0.
-

20

50
2Xl0.

8Xl0' 8Xl0.
- 1.5Xl0·

17Xl0·

86

80

96
100

±11

±10.5

±115

0.15

RO
PSRR

Sample Mode
Vs= ±9V to±18V

PD

Sample Mode V'N = 0

82

77
180

160

3

14-7

mA
mA/mA

1.5Xl0·

90

90

dB

100

100

kHz

±11.5

V

±115

±10.5

015

92

NOTES:
1 Guaranteed by deSign.
2. These measurements are made with the deVices warmed-up Itean be seen

-

pC

92
170

72

210

0.15

!1

92

dB

160

240

mW

that there is a selection trade off between droop rate and hold mode
settling time.
Measured 500J,ls after hold command.

1/86, Rev. A

-------I~ SMP-10ISMP-11 LOW-DROOP-RATE/ACCURATE SAMPLE-AND-HOLD AMPLIFIERS

ELECTRICAL CHARACTERISTICS -

SMP-10 ONLY at

Vs

= ±15V, C H = O.005,..F, VLC = OV, TA = 25·C, device fUlly

warmed-up, unless otherwise noted.
PARAMETER

SYMBOL

CONDITIONS

SMP-10AlE
MIN
TYP
MAX

SMP-10B/F
MIN
TYP
MAX

Hold Step

VHS

V,N-O

-1.0

-3.0

linearity Error

NL

Y,N - ±10V, Rl - 5kO

Output NOise

EN(RMS)

Wideband NOise 100Hz
to 100kHz Sample Mode

ELECTRICAL CHARACTERISTICS at

Vs

+1.5

+4.0

+1.5

UNITS

+6.0

mV

0005

0.007

%ofl0V

40

50

jlV RMS

= ±15V, CH = O.005,..F, VLC connected to ground, O·C S; TA S; +70·C, unless

otherwise noted.
SMP-10E
SMP-11E
TYP
MAX
MIN

SMP-10F
SMP-11F
MIN
TYP
MAX

075

20

1.0

4.0

2.7

10

mV

50

90

BO

140

120

250

nA

0.7

5

140

1000

PARAMETER

SYMBOL CONDITIONS

Zero-Scale Error

Vzs

Y,N - 0, VSJH - 3.5V. (Note 3)

Input Bias Current

Ie

V,N-OV

lOR

Device Warmed Up SMP-l0
(See Note 2)
SMP-ll

0.05
0.5

0.25
1.B

0.080
0.6

0.65
2.B

Droop Rate

dVCH/dt

Device Warmed Up SMP-l0
(See Note 2)
SMP-"

10
100

50
360

16
120

130

Voltage Gai n

Av

Sample Mode
V,N -±10V, Rl -5kO
or Y,N - ±5V. Rl - 2.5kO

Power Supply
Relectlon Rallo

PSRR

Sample Mode
Vs -±9Vto±1BV

I lc

VlC-OV

ISJH

Sample Mode
VSJH -0.6V
Hold Mode
VSJH -50V

Leakage
(Droop) Current

Logic Control
Input Current

Logic Input

Differential Logic
Threshold

VTH

o 99955 0 99976

BO

75

SMP-11G
TYP
MAX

099930 0.99970

BO

70

UNITS

nA
jlV/ms

VIV

90

dB

-I

-2

-I

-3

-I

-4

jlA

-5

-15

-5

-15

-5

-15

jlA

0.2

0.2
O.B

560

o 99950 0 99972

90

MIN

1.3

2.0

O.B

1.3

02
2.0

O.B

1.3

nA
20

V

NOTES:
I
Guaranteed by design.
2. These measurements are made with the devices warmed-up It can be seen
that there is a selecllon trade off between droop rate and hold mode
settling time
3. Measured 500jls after hold command

14-8

1/86, Rev. A

_ _ _ _ _--j~ SMP-10/SMP-11 LOW-DROOP-RATE/ACCURATE SAMPLE-AND-HOLD AMPLIFIERS

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, CH = O.005/LF, VLC connected to ground, -55°C :'0 TA:'O + 125°C, unless
otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

Zero-Scale Error

Vzs

V,N = O. VS/H = 3.5V. (Note 3)

Input Bias Current

Ie

V,N=OV

lOR

TA =-55'C
TA =+125'C
TA = Full Range
(See Note 2)

Leakage (Droop) CUrrent

Droop Rate

dVCH/dt

TA =-55'C
TA =+125'C
TA = Full Range
(See Note 2)

1.25

3.0

1.60

55

mV

90

180

160

260

nA

0.050
6

0.50
7.5

0.080
8

1.22
10

7.5

8

10

SMP-11

SMP-10

10
1200

100
1500

16
1600

250
2000

SMP-11

1200

1500

1600

2000

Av

Sample Mode
V ,N = ±10V. RL = 5kU
or V,N = ±5V. RL = 2.5kU

Power Supply
Rejection Ratio

PSRR

Sample Mode
Vs =±9Vto±18V

Logic Control Input Current

I LC

VLC = OV

ISIH

Sample Mode
VSlH = 0.6V
Hold Mode
VSlH = 5.0V

Differential Logic
Threshold

SMP-10B
SMP-11B
MIN
TYP
MAX

SMP-10

Voltage Gain

Logic Input

SMP-10A
SMP-11A
MIN
TYP
MAX

0.99950 0.99972

78

0.99940 0.99968

88

72

UNITS

nA

!,Vlms

VIV

90

dB

-1

-3

-1

-5

!,A

-5

-15

-5

-15

!,A

0.2

nA

0.2

~

VTH

0.6

1.3

20

0.6

1.3

2.0

V

3

~
p..,

~

NOTES:
1.
2.

~

~
......,

Guaranteed by design
These measurements are made with the devices warmed-up. It can be seen
that there is a selection trade off between droop rate and hold mode
settling time
Measured 500!,s after hold command.

Cl
.....:I
0

:r:
0

Cl

~

~
p..,

~

II

14-9

1/86, Rev. A

--------I1E!1D

SMP-10/SMP-11 LOW-DROOP-RATE/ACCURATE SAMPLE-AND-HOLD AMPLIFIERS

DICE CHARACTERISTICS
SMP-10

SMP-11
2. INPUT
3. NULL
4. NULL
5. NEGATIVE SUPPLY
(SUBSTRATE)
7. OUTPUT
9. POSITIVE SUPPLY
11. HOLO CAPACITOR (CH)
13. LOGIC THRESHOLO
CONTROL (VLe)
14. SAMPLE/HOLD
COMMAND

2.
3.
4.
5.

INPUT
NULL
NULL
NEGATIVE SUPPLY
(SUBSTRATE)
7. OUTPUT
9. POSITIVE SUPPLY
11. HOLD CAPACITOR (CH)
13. LOGIC THRESHOLD
CONTROL (VLe)
14. SAMPLE/HOLD
COMMAND
For additional DICE
Information refer 10 1986
Dala Book, Seellon 2.

DIE SIZE 0.081 X 0.086 Inch, 6966 Iq. mill
(2.057 X 2.184 mm, 4.772 Iq. mm)

WAFER TEST LIMITS at Vs = ±15V, CH = O.005/LF, VLC connected to ground,

PARAMETER

SYMBOL

CONDITIONS

Zero-Scale Error

Vzs

V,N=O, VslH =3.5V
Hold Mode, (Note 3)

Input Bias Current

Ie

V,N = OV

Leakage (Droop) Current

lOR

SMP-10
Device Warmed-Up SMP-11

Droop Rate

dVeH/dt

.
SMP-10
Device Warmed-Up SMP-11

Voltage Gain

Av

Sample Mode
V,N =±101/
orV ,N =±5V

Hold Capacitor
Charging Current

leH

Y,N - Vour " ±3V

Input Voltage Range and/or
Output Voltage Swing
Power Supply
Rejection Ratio

Sample Mode
Vs =±9Vto±18V

Power Consumption

Po

Sample Mode Y,N = 0

Logic Control Input Current

ILe

VLe = OV

ISJH

Sample Mode
Vs/H =0.6V
Hold Mode
VSIH = 5V

Differential Logic
Threshold

VrH

= 25°C, unless otherwise noted.

SMP-10N
SMP-11N

SMP-10G
SMP-11G

LIMIT

LIMIT

UNITS

1.5

3.0

mVMAX

60

90

nAMAX

0.10

0.25
2.5

nAMAX

20
200

50
500

poV/ms MAX

0.99963

0.99953

V/V MIN

30

20

mAMIN

±11

±10.5

VMIN

82

77

dBMIN

180

210

mWMAX

-2

-3

poAMAX

-15

-15

poAMAX

0

nAMAX

2.0
0.8

V MAX
VMIN

RL = 2.5kil

PSRR

Logic Input

TA

2.0
0.8

VLe=O

NOTE:
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

TYPICAL ELECTRICAL CHARACTERISTICS at Vs

= ±15V,

CH

= O.005/LF,

VLC connected to ground,

TA

= 25°C,

unless

otherwise noted.

PARAMETER

SYMBOL

CONDITIONS

Acquisition Time

taq

10V step to 0.1% of final value

Aperture Time

tap

Charge Transfer

a,

Y,N = 0, VS/H = 3.5V

Slew Rate

SR

Y,N = ± 10V, RL = 2.5kil
14-10

SMP-10N
SMP-11N

SMP-10G
SMP-11G

TYPICAL

TYPICAL

3.5

3.5

pos

50

50

ns

5

pC

10

10

V/p.S

UNITS

1/86, Rev. A

--------I~ SMP-10/SMP-11

LOW-DROOP-RATE/ACCURATE SAMPLE-AND-HOLD AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS
CHANGE IN HOLD STEP

AMPLITUDE CHANGE IN
HOLD STEP va
INPUT VOLTAGE

vaS/H~
dt

-10

I'
-10

~t.VOUT

ill

In

-

30 (mV)
dV!dt (V/jJ.s)

our "'"

;;

E
~

~

w

"
Ii"
w

START WITH

IN

0
40

NEGATIVE SUPPLY

-1

- 0 VOLTS (30jJ.V)

::>

AT l00V!ps

::; -2
~

IIIII

(+lSV +lV SIN wT)

Il~t;iWIUrlr ~

Q

>-

C H - 0OO5MF

1.0

POSITIVE SUPPLY

z

-0 1

-00 1

REJECTION RATIO = 84dB

60

"~

,

111111
111111111 111111111 11111111 III;~
TYPICAL 0 C POSITIVE POWER SUPPLY
11111111 11111

90 "

3.0
dV!dt

\ 1~~lpICAL til~ NEG~~:~E pb~MM SUPPLY
REJECTION RATIO = 86dB

~

80

w

',"

01

'2

In

9
!E<1

100

'3
Ll,V

~

HOLD MODE
POWER SUPPLY REJECTION

20

" -3

II

10

-15

100

VIN'OOVHo~lgl~dDIEIIIIIII III

NORMALIZED TO
OVOLTSV 1N @25 c C

(10 MEG PARALLEL WITH CHI

-5

-10

o

.,5

.,0

.5

0

10

100

FREQUENCY (Hz)

SMP-10
DROOP RATE
va TEMPERATURE

SMP-11
DROOP RATE
va TEMPERATURE

DROOP CURRENT
VB TEMPERATURE

-1500

!

~

~ -1000

"

~

I

~
a:

g
a: -500

-:I
-55

;

V

-15

5

25

45

65

85

105

:sg

-200

Q

II kkl::t[ Ir

-35

-300

125

r-LJ

/'

100

V

V

I I

/

+111
~
;h

-(
~

lOR

ARROW INDICATES POSITIVE
lOR REGAROLESS OF CH
VOLTAGE POLARITY

-6

200
-55

-35 -15

INPUT BIAS CURRENT
vs TEMPERATURE

25

45

65

85

105

-55 -35

125

-15

80

'\

"-

20
10V STEP TO WITHIN ±10mV
OF FINAL VALUE (0 1%)

o

-100

-50

50
TEMPERATURE

100

rei

150

o
18

16

14

12

10

CH (oF. 1nF = 1000pF)

14-11

85

105

125

-2

~

-3

-

-

~u

-4

(;

-5

g

.if)
--~

SAMPLE

e

25 de

1--125o

>-

..........

-r~
1-'--+5 o

1>G

"

65

-1

~

"'- '-.

1\
40

(

RECOMMENDED HOLD
CAPACITOR IS 5000pF

60

45

LOGIC INPUT CURRENT

12

\

25

TEMPERATURE (OC)

ACQUISITION TIME
va HOLD CAPACITOR

10

N

CH

TEMPERATURE (OC)

100

~

~MP-ll

/

-100

-

1"-1-t-

I

-400

w

~

F::

:1::=::=:::11::::::1
1 1=:::11A

1M

lOOk

INPUT VOLTAGE (VOLTS)

-2000

Q

10k

lk

8tH dV!dt LOGIC INPUT SLEW RATE (V/ps)

'I

e;
I
viLe = 0 OV (PIN 13)

-6

-7

I

I

iPIN

141 vbLTAGE

o
LOGIC INPUT VOL TAGE (VOLTS)

1/86, Rev. A

--------I!mD

SMP·10/SMP·11 LOW·DROOP·RATE/ACCURATE SAMPLE·AND·HOLD AMPLIFIERS

TYPICAL PERFORMANCE CHARACTERISTICS
POWER DISSIPATION VB
INPUT FREQUENCY = Vp SIN wT

GAIN ERROR

500

+3r-----~-----r-----r----~

;;!

RL '" 2.5kn

w ~~--~~----4----=~~~~

400

~ +1~~~~----4------t----~

i

§

;t

~

O/-----+----''''''''''!.=---t----~

I

w

~ -1r-----+-----~----~~~~

liN

V+ = 15V
v- = -15V
V LC '" GND
CH = O.OO5f,!F

TA = 2S C C
SAMPLE MODE

='0

[I

V.

5V.

J

.§
2

0
;:: 300

i5 200
a:

15

I!

-

'0
1.0Vp

~
2

:>

5

r-

~

I[

g

'00

-3~----~----~----~----~

o

14

+5

+10

1\
SOUAR~
.mi'

I~SINE

v- = -15V
V LC =GND
CH = 0 005~F

~

-5

OUTPUT SWING

WAVE

V+=15) III

~ ~/_----+_----4_--~~~~~

-10

y: III

INPUT SWING

LIMITED TO ±15V

~ ~251.

~~

iii

MAXIMUM INPUT SIGNAL
AMPLITUDE VI FREQUENCY

10k

lk

1M

lOOk

~

lOOk

10k

500k 1M

2M

INPUT VOLTAGE (VOLTS)

INPUT FREQUENCY (Hz)

INPUT FREQUENCY (Hz)

OUTPUT RESISTANCE
VI FREQUENCY

OUTPUT WIDEBAND NOISE
VB BANDWIDTH (O.1Hz TO
FREQUENCY INDICATED)

HOLD CAPACITOR
CHARGING CURRENT VI
INPUT OUTPUT VOLTAGE

:0;-

10,000

R~ '~g~D Mo8~"~ R6'~l~~L' MO'~~

10M

12

+30

10

I

1,000

+20
+'0

II

HOLD MODE/

«"

100

o

SAMPLE MODE

~
10

100

lk

10k

~f-"

lOOk

1M

10
10k

1M

lOOk

INPUT FREQUENCY (Hz)

10M

BANDWIDTH (Hz)

TOTAL HARMONIC
DISTORTION VB FREQUENCY

SAMPLE MODE
POWER SUPPLY REJECTION

-

'00

.0

POWER SUPPLY CURRENT
VI POWER SUPPLY VOLTAGE

\';N'!'~ ov

r-i'

l-

~

II- l-

(+15V +1V SIN wT)

60

LLJ

T~ = +2;'C

I1mlll'illll
pW!~ls~ml

I- f-

,....slMPL~ MolE

40

I'
NEGATIVE SUPPLY
20

o

10

FREQUENCY (Hz)

(-mIljll TIIIIII!
100

lk

10k

FREQUENCY (Hz)

14·12

lOOk

1M

1

'2

15

,.

V+ OR V- (VOLTS)

1/86, Rev. A

-------I~ SMP-10/SMP-11 LOW-DROOP-RATE/ACCURATE SAMPLE-AND-HOLD AMPLIFIERS

SMp·10/SMp·11 ACQUISITION TIMES
ACQUISITION TIME
-10VtoOV

ACQUISITION TIME
+10Vto OV

S/H{

l=v•• _.

+10V

1-

I-

. .

ACQUISITION TIME
-1.0V to OV

-

:-

· 1-

=1 - -== == · .1
:== -- ==

I

OV

v

- ACQUISITION TIME
+1.0Vto OV

v

8tH {
+1.OV

.111·

lp

ACQUISITION TIME
-100mV toOV

ACQUISITION TIME
+100mV to OV
8tH {

ror

mV

I

a
OV

I

lp

.,

+100

I
I

II

---.1

Giiiiiiiiii
v

==

I

Cv

lp

I

SAMPLE

•

APPLICATIONS INFORMATION
During the null adjustment. the amplifier should be switched
continuously between the "sample" and "hold" mode. The
error should be adjusted to read zero when the unit is in the
"hold" mode. In this way. both offset voltage errors and
charge transfer errors are adjusted to zero.

As shown in the Figure. the sample/hold mode control is
accomplished by steering the current (1 1) through Q1 or Q2.
thus providing high-speed switching and a predictable logic
threshold. For TTL and DTL interface. simply ground VLc
(Pin 13). For CMOS. HTLand HNIL interface. the appropriate

ZERO-SCALE NULL ADJUSTMENT

LOGIC CONTROL

v+

v-

14

2'
StH

VLe

>--'-----0 OUTPUT
CURRENT TO CONTROL
SAMPLE/HOLD MODES

8tH

•

r----~-~_oV+

kr!

INPUT o-----~

•

VS/H TRIP POINT = V Le + 2VSE
FOR TTL COMPATIBILITY SET V LC = OV

VLC

14-13

1/86, Rev. A

--------l~ SMP-10/SMP-11 LOW-DROOP-RATE/ACCURATE SAMPLE-AND-HOLD AMPLIFIERS

threshold voltage, allowing for 2 diode drops for 01 and VeE
of Q3, should be applied to VLC.
For proper operation, the VLC (logic control) must always be
at least 3.5V below the positive supply and 2.0V above the
negative supply.
Sample-and-hold control voltage (S/H) must always be at
least 2.BV above the negative supply.

LOCAL

ANALOG

GUARDING AND GROUNDING LAYOUT
The use of a ground plane is strongly recommended to
minimize ground path resistances. Separate analog and
digital grounds should be used, and it is advisable to keep
these two ground systems isolated until they are tied back to
the common system ground. Digital currents should not flow
back to the system ground through the analog ground path.
HOLD CAPACITOR RECOMMENDATIONS
The hold capacitor (CH) acts as a memory element and also
as a compensating capacitor for the sample-and-hold
amplifier. Forstable operation, a minimum value of 2000pF is
recommended, with no limit set for the maximum value. The
devices have been internally trimmed for C H= 5000pF. Other
values of CH will cause a zero-scale shift, which can be
calculated from the following equation:

GROUND

tN

( V) = 5 (pC) X 10 3 -1
zs m
CH (pF)

The hold capacitor should have very high insulation
resistance and low dielectric absorption. For temperatures
below 85·C, polystyrene capacitors are recommended,
while teflon capacitors are recommended for higher
temperature applications.

14-14

1/86, Rev. A

SMP-81
TELECOMMUNICATIONS
SAMPLE-AND- HOLD AMPLIFIER
Precision Monolithics Inc.

FEATURES

GENERAL DESCRIPTION

• Meets System Performance Requirements In MultiChannel CODECs
• Trimmed for Minimum Zero-Scale Error ........ 0.6mV
• Low Droop Rate Over Temperature .......... 0.1I'V/l's
• Low Aperture Time ............................. SOns
• Fast Acquisition Time 10V Step to 0.1% .......... 3.51's
• High Slew Rate ............................... 10V/l's
• High Sample-Current to Hold-Current Ratio .. 1.7 X 108
• DTL, TTL & CMOS Compatible Logic Input
• HA-2425, DATEL SHM-IC-1, and AD-583 Socket
Compatible"
• Low Power Dissipation
• Low Cost
• Feedthrough Attenuation Ratio ................. 96dB

The SMP-81 precision sample-and-hold amplifier provides
the high accuracy,low droop rate and fast acquisition ideally
required for PCM encoders. The SMP-81 is a non-inverting
unity gain circuit consisting of two buffer amplifiers of very
high input impedance connected by a diode bridge switch.
HIGH ACCURACY AND LOW DROOP RATE
The high input impedance and low droop rate of the SMP-81
are achieved by PMl's ion implant super beta process. The
. high input impedance permits high source impedance applications without degrading accuracy, and low droop rate.
Other features of the SMP-81 include high accuracy, O.6mV
of combined offset voltage and step transfer error, and very
low feedthrough. A diode bridge switch design allows minimum charge transfer step. On-chip zener-zap trimming
eliminates nulling for most applications.

ORDERING INFORMATIONt

Vzs(mV)

1.6
3.5

HERMETIC
14·PIN DIP

OPERATING
TEMPERATURE
RANGE

SMP-81EY
SMP-81FY

INO
INO

FAST ACQUISITION
A unique super charger or transconductance amplifier provides up to 50mA charging current to the hold capacitor. As a
result, smooth charging of the hold capacitor is achieved
with minimum noise. The super charger, in conjunction with
the high slewing rate input and output buffer amplifiers, permits fast acquisition operation. The adjustable logic input
threshold makes the SMP-81 compatible to all logic families.

t All

commercial and industrial temperature range parts are available with
burn-in. For ordenng Information see 1986 Data Book, Section 2.

FUNCTIONAL DIAGRAM
PIN CONNECTIONS

14-PIN HERMETIC DIP
(Y-Sufflx)
OUTPUT

"Pins 1 and 8 are not mternally connected. In unity gain
applications, the SMP-81 can replace HA-2425, SHM-IC-1
and AO·583 directly

NULL

"SAMPLE/HOLO CONTROL
HIGH=HOLO
LOW = TRACK

14-15

1/86, Rev. A

I

~ SMP-81 TELECOMMUNICATIONS SAMPLE-AND-HOLD AMPLIFIER

ABSOLUTE MAXIMUM RATINGS
Supply Voltage (V+ minus V-) •••••••••••.•••••••.•• 36V
Power Dissipation •••••••.••••.••••••.•••••••••• 500mW

Output Short-Circuit Duration ••••••••••••••••• Indefinite
Hold Capacitor Short-Circuit Duration .••••••.••••• 60sec

Derate Above 100· C •••••••••••.•.•••..•••••.• 10mW/·C

Operating Temperature Range ••••••.•••

Input Voltage ••••••••••••••.•••. Equal to Supply Voltage
Logic and Logic Control Voltage .. Equal to Supply Voltage

Storage Temperature Range. • • . . . • . . ..

ELECTRICAL CHARACTERISTICS

-25·Cto+85·C
-65· C to + 150· C

Lead Temperature (Soldering, 60 sec) ••••••••••••• 300·C

at Vs ±15V, C H = O.005/LF, VLc connected to ground, -25·C :5 TA:5 +85·C, unless

otherwise noted.

SMP-81E
PARAMETER

MIN

SMP-81F

TYP

MAX

VZS

Y,N = 0, VSlH = 3.SV (SOOl'sec
after Hold Command)

0.6

SYMBOL

CONOITIONS

Zero-Scale Error (Hold Mode)

MIN

TYP

MAX

UNITS

1.6

0.9

3.S

mV
nA

Input Bias Current

Ie

V,N = 0

10S

225

120

4S0

Leakage (Droop) Current

lOR

Device Warmed·up

0.5

10

0.5

20

nA

Droop Rate

dVCH/dt

01

20

01

40

mV/msec

Input Resistance

R'N

Voltage Gain

Av

(See Note)
Sample Mode
Y,N = ± 10V, Rl = Skn
or Y,N = ±5V, Rl = 2.Skn

0.6

2.0

0.3

14

Gn

0.99960

0.99980

0.999S5

0.99978

VIV

3.5

3.S

I'S

SO

SO

nsec

10

10

VII's

10V step to within 10mV of
final value (0.1%)

Acquisition Time

taq

Aperture Time

ta~

Charge Transfer

at

V,N = 0, VSlH = 3.5V

Slew Rate

SR

V,N =±10V, Rl =2.Skn

Hold Capacitor Charging
Current

ICH

V,N-VDUT,,±3 volts

30

SO

20

50

mA

Feedthrough Attenuation
Ratio

FA

Input -20Vp_p 1kHz, Rl = SKn
(See Note)

86

96

80

90

dB

Full Power Bandwidth

Fp

±10V~_~

100

kHz

Input Voltage Range andlor
Output Voltage Swing

100

(Dissipation Limited)
±10

Rl = 2.5kn
RD

Power Supply Rejection
Ratio

PSRR

Sample Mode Vs = ±9V
to±18V

Power Consumption (DC)

Po

Sample Mode Y,N = 0

Logic Control Input Current

I lC

Logic Input Current

ISIH

Differential Logic Threshold

VTH
tHM

±10

±".S
0.15

Output Resistance

Hold Mode Settling Time

pC

80

160
-6

Sample Mode VS/H = 0.6V
Hold Mode VSlH = 5.0V
0.8
SV step to within 1mVof

180

-1S
0.6

-45

1.3

2.0

V

0.15

n

90

dB

170
-9

-3

1.S

final value

75

90

±".5

0.8

210

-3

mW
I'A

-15
0.6

-4S

I'A
nA

1.3

2.0

V

1.5

I's

NOTE: Guaranteed by design.
DICE
For applicable DICE information, see SMP-11 Data Sheet.

14-16

1/86, Rev. A

---------t~ SMP-81 TELECOMMUNICATIONS SAMPLE-AND-HOLD AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS
SMP-81 ACQUISITION TIMES
ACQUISITION TIME
+10V TO OV

ACQUISITION TIME
-10V TO OV

II

iiiiiii
V '

IVI

-

-~

:=-

I

-III --

--.

,- -II

11
I

==

/

I

I

v

V

=1P

l1

I

==

1p

V

ACQUISITION TIME
-1.0V TO OV

ACQUISITION TIME
+ 1.0V TO OV
iiii

V

II

ffi.....

V

~

:!!

::l
I=l-.

I'

~
9
0
:r:
I

IJJ

-1

0

00.

00.

1pS

1Pot:

ACQUISITION TIME
-100mV TO OV

~

ACQUISITION TIME
+ 100mV TO OV
WHII
II

~

V

II

.,

~1

I

I
~

~~

I

li:Jv

1p
HOLD

SAMPLE

14-17

~

1/86, Rev. A

-----------1~ SiViP-8, TELECOMMUNiCATiONS SAMPLE-AND-HOLD AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

AMPLITUDE CHANGE
IN HOLD STEP
vs INPUT VOLTAGE
+3r---~--~--~----r---~--~

TYPICAL 0 C. NEGATIVE POWER SUPPLY
REJECTION RATIO'" B6dB

i

+2

t--+.--+--+

~

+1

f---+~~..--+--I--+----j

~

IIIIII11

~

~
~

'pOS~TIV~'SUpp~y'"

~
~
~

"

-1

40

~« ~~H+mffi-++H~~~~~~~H+~

H-++Hltlt-++tM+Hfjjjjf---f-tI'llitt-HtH1Il1

g40~H+mffi-++U~~~~++~~H+~

-2

HOLD MODE
10MEG PARALLEL
WITH CHI

o,1:-0..u=I~OO~w..u~...u..lJJlI-!l::-LJ..lJ.l!l!l:::-l.l.lJ.IlI!'M

INPUT VOL rAGE (VOL fS)

-190

0 .•

-180

0.8

-160

0.7

-140

0.6

ffi

0.5

B

POSITIVE
DIRECTION

OA

~

0.3

~

0.2

'"

~
"t'OR
:h

0.1

-<>.1
-0.2
-30 -20

J".
~

I'

0
VS=t15
SAMPLE MODE

-20

+60

TA '"

o

+10
+80 +90

-40

-20

600

25~C

~

> +1 f-'>-2'-.2'~---+---+-----1

of----~--=~~~-+---+

w

f----~---I---=>\.2"""':-=::~

~
~ -21---+---I---=+---~

-.

~

____

~

____

+40

+60

0
-20

+80

OVOLTS:

~

+.

INPUT VOLTAGE (VOLTS)

____

500

~400 I--z
o

m
mr

~

~200
ffi

100

+40

+60

+80

MAXIMUM INPUT SIGNAL
AMPLlTUDEvsFREQUENCY

I II

I/

10

"~

/

i= 300
~

~

V,N

10V Pk
(20V Pk-Pk)
I I
5V Pk
(10V Pk-Pk)

+20

TEMPERATURE ~C)

POWER DISSIPATION
vs FREQUENCY
INPUT = Vp sin wt

o

-10

+20

TEMPERATURE (OC)

~ +2~~_--j-_ _-+_M_M_P_L_E+M_O_OE_~

-3L-____

V,N

3. 0

10

0.0

+20

+40

~ +3

=

SUPPLY CURRENT
vs TEMPERATURE _

:1

1'0...

i I
+20

0

GAIN ERROR

~

.. -

0

TEMPERATURE (Gel

-1

1--

..

l

"

CH

I

7.0

~~!

ARROW INDICATES POSITIVE
lOR REGARDLESS OF CH
VOL rAGE POLARITY

0.0

SAMPLE-MODE SUPPLY
CURRENTvsTEMPERATURE

-120 ~

I'

OF lOR

FREQUENCY (Hz)

INPUT BIAS CURRENT
vs TEMPERATURE

0.10

.s

o,.LO..J..JJ..lJJllL.~llill1LL.LllllJll::-l.LJ.I!lJll.,..LWlW

FREQUENCY (Hz)

LEAKAGE (DROOP)
CURRENTvsTEMPERATURE

'~"

'"

Y,N = O.DV

20

'<

'"z2

;;J

;;J

~1~5---~10~~~~--~---~~--+1~0---+~'5

~

j

I

F··'t+FlI,,*ffi#~;;f#IIIF~.:tiitltlll

0:

;

5

1111111

(+15V +1V SIN wTI

Ot--+-+~~~

w

111111111

TYPICAL D,C. POSITIVE POWER SUPPLY
REJECTION RATIO = 84dB

80

2 60

a:

SAMPLE-MODE
POWER SUPPLY REJECTION

HOLD-MODE
POWER SUPPLY REJECTION

jfJJ.//

2.5V

1 OV Pk
(2V Pk-Pk)

Pk-PkJ

TEST CONDITIONS
SAMPLE MODE
v+ '" 15V
V- '" -15V
CH "'OOO5j.lF
INPUT CENTERED AROUND GROUND

Pk

~0

I'V

~

z
>-

-

~

+10

lk

10k

lOOk

INPUT FREQUENCY (Hz)

14-18

1M
INPUT FREQUENCY (Hz)

1/86, Rev. A

---------I~ SMP-81 TELECOMMUNICATIONS SAMPLE-AND-HOLD AMPLIFIER

TYPICAL PERFORMANCE CHARACTERISTICS

14.0

HOLD CAPACITOR CHARGING
CURRENTvs
INPUT OUTPUT VOLTAGE

OUTPUT WIDEBAND NOISE
VB BANDWIDTH (O.1Hz
TO FREQUENCY INDICATED)

OUTPUT RESISTANCE
VB FREQUENCY

:/

10,000

RO~~'~~lL~ '~g~~ ~ ~~I~~M~Lk'~~~E

12.0

§

10.0

~

~

+2~

'000

w

+,0

ao

II

6.0

HOLD MO OE

~

~ 4.0

5

1i'~~I~IT iii'liutfi I UI'
i.'"
IIIIII[ IJ

2.0

111111 II

o

'0

,k

'00

10k

,0
lOOk

,M

1/

P"

'00

SAMPLE MODE

II,M

.....::r.'Ok

111I11
'OM

'001<
BANDWIDTH (Hz)

INPUT FREQUENCY (Hz)

ACQUISITION TIME TEST CIRCUIT

v+

+10V---,

v-

INPUT

SMP·81

INPUT

ov _____ IL._ _ _ _ __

I

OUTPUT

+lOv

t

OUTPUTov~

2ku

I

TO SCOPE

FET
SfH

PROBe

0------'

REFERENCE OV - - - - - ; . . . . - - - -

v Le
DIGITAL
GND

I

2k!1

REFERENCE

ERROR
VOLTAGE

~'
--l
r--

ov - - - - - - - -

-t -

-

-

FINAL VALUE

ACQUISITION TIME

APPLICATIONS INFORMATION

The hold capacitor should have very high insulation
resistance and low dielectric absorption. For temperatures
below 85° C, polystyrene capacitors are recommended, while
teflon capacitors are recommended for higher temperature
applications.

HOLD CAPACITOR RECOMMENDATIONS
The hold capacitor (C H) acts as a memory element and also
as a compensati ng capacitor for the sample-and-hold amplifier. For stable operation, a minimum value of 2000pF is
recommended, with no limit set for the maximum value. The
SMP-81 is internally trimmed for CH = 5000pF. Other values
of CH will cause a zero-scale shift, which can be calculated
from the following equation:

f>.V

SMP-S1 LOGIC CONTROL
The sample/hold mode control of the SMP-81 incorporates a
unique logic input circuit, which enables direct interface to
all popular logic families and provides maximum noise
immunity. As shown in Figure 1, the mode control is
accomplished by steering the current (I l ) through 01 or 02,
thus providing high speed switching and a predictable logic
threshold. For TTL and OTL interface, simply ground VLC
(pin 13). ForCMOS, HTLand HNIL interface, the appropriate
threshold voltage, allowing for 2 diode drops for 01 and VeE
of 03, should be applied to VLC.

( V) = 5 (pC) X 10 3 -1
zs m
CH (pF)

A C H of 5000pF has been empirically determined to be an
optimum value for 8-channel shared COOEC operation.

14-19

1/86, Rev. A

---------i!EMD iiiiiji-iii TELECOMMUNICATIONS SAMPLE-AND-HOLD AMPLIFIER
SAMPLE/HOLD MODE INTERFACE CIRCUITRY
V+

v-

~------------.---~w

<>----""1

INPUT

;;>-'-------0

OUTPUT

SiH
(PIN '41

0----------'

S/H
V Le
~

0--------'

V Le

(PIN 13)

I

eH' 5OOO0F

CURRENT
TO CONTROL

SAMPLE!

Figure 2

HOLD MODES

GUARDING AND GROUNDING LAYOUT

Figure 1

For proper operation, the VLC (logic control) must always be
at least 3.5V below the positive supply and 2.0V above the
negative supply.

The use of a ground plane is strongly recommended to minimize ground path resistances. Separate analog and digital
grounds should be used, and it is advisable to keep these two
ground systems isolated until they are tied back to the
common system ground. Digital currents should not flow
back to the system ground through the analog ground path.
A guard trace surrounding the hold capacitor node pin 11,
minimizes PC board leakage problems, see Figure 3.

Sample-and-hold control voltage (S/H) must always be at
least 2.8V above the negative supply.
FROM
{s/H CONTROL
LOGIC
CONTROL DIGITAL
GROUND

ANALOG
INPUT

ZERO-SCALE ERROR NULL ADJUSTMENT

During the null adjustment, the amplifier should be switched
continuously between the "sample" and "hold" mode. The
error should be adjusted to read zero when the unit is in the
"hold" mode. In this way, both offset voltage errors and
charge transfer errors are adjusted to zero. Figure 2 shows
the recommended 10kO trim pot connected to V+ if user
needs better Vzs than 1.6mV.

LOCAL
ANALOG
GROUND

V-

Figure 3

14-20

1/86, Rev. A

---------I~ SMP-81 TELECOMMUNICATIONS SAMPLE-AND-HOLD AMPLIFIER

TYPICAL APPLICATION
EIGHT-CHANNEL SHARED CODEC PCM ENCODER

DIGITAL
OUTPUTS

2.5kn

..0.
BIT

tt=a*t=~::glCHOAD
BITS

CHI

BITS
1ST"

CH2
CH3

VREF

.'OV

123456789

CH4

OU

(':=

CH5

It+-

CH6

12 IREF

0"

CH70-+------'

20kn

CH6o-t;=:;:::~-,J
CHANNEL SELECT

-15V

+15V
WAVEFORMS

LOGIC
CHANNEL INPUT ANALOG

MUX OUTPUT SIGNAL

NULL
SAMPLE/HOLD 8/H COMMAND ~

SAMPLE/HOLD OUTPUT

14-21

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

SPECIAL
FUNCTIONS
Precision Monolithics Inc.

15-3

Introduction

15-3

Definitions

15-4

GAP-Ol
General-Purpose Analog Signal
Processing Subsystem

15-19

PKD-Ol
Monolithic Peak Detector

15-2

SPECIAL
FUNCTIONS
Precision Monolithics Inc.

INTRODUCTION

The PKD-01 functions as an analog peak
detector. Peak detection, peak hold and reset
are all digitally-contollable. Circuit gain is set by
external resistor ratios. The PKD-01 configures
into both positive or negative peak detector
circuits.

This section includes analog subsystems made
up of standard integrated circuit building blocks
such as op amps, comparators, analog switches,
sample/hold, and buffers.
The GAP-01 is a general-purpose analog signal
processing subsystem. This precision circuit
contains circuit building blocks used to
construct synchronous demodulators, absolute
value amplifiers, window comparators, a twochannel S/H amplifier or a two-channel
multiplexer with gain.

DEFINITIONS
Refer to the sample-and-hold amplifer section of
this catalog for the GAP-01 and PKD-01 circuit
definitions.

I
15·3

GAP-Ol
GENERAL- PURPOSE ANALOG SIGNAL
PROCESSING SUBSYSTEM
Precision Monolithics Inc.

FEATURES
•
•
•
•
•
•
•

Several applications exploit the ability to select the signal
path through the GAP-Ol. As a two-channel multiplexer or
analog switch, the GAP-Ol high input impedance offers
advantages when switching high impedance signals. Gain
through the "MUX" is also possible. The GAP-Ol operates as
a sample/hold amplifier in the hold mode when both transconductance amplifiers are unselected. With the on-board
comparator, a two-channel successive approximation analogto-digital conversion (ADC) system may be constructed.
Combining a sign-magnitude, digital-to-analog (DAC) converterwith the GAP-Ol results in a four-quadrant multiplying
DAC. The GAP-Ol contains all the functional devices needed
to perform synchronous demodulation or implement the
absolute value function.

Low Offset Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3mV
Low Zero-Scale Error ........................... 4mV
Low Droop Rate ........................... O.1mV/ms
Wide Bandwidth ............................. 400kHz
Digitally Selected Signal Path
Uncommitted Comparator On Chip
Wide Application Versatility
• Synchronous Demodulator
• Absolute Value Amplifier
• Two-Channel S/H Amplifier
• Two-Channel Multiplexer With Gain

ORDERING INFORMATIONt
Vos
(mV)

Vzs
(mV)

MILITARY

3
6

4
7

GAP01AX'
GAP01BX'

HERMETIC
PLASTIC
INDUSTRIAL COMMERCIAL
GAP01EX
GAP01FX

FUNCTIONAL DIAGRAM

GAP01EP
GAP01FP

COMPARATOR

AMPLIFIER A NULL

-INPUT +INPUT

• For devices processed In total compliance to M I L·STD-883. add 1883 after
part number. Consult factory for 883 data sheet.
tAil commercial and Industrial temperature range parts are available with
burn-in. For ordering informatIOn see 1986 Data Book, Section 2.

OUTPUT

CHANNELA''''''I-_-d

-INPUT

GENERAL DESCRIPTION
Designed as a general-purpose analog processing subsystem, the GAP-Ol combines many commonly used system
building blocks within a single integrated circuit.

VOUT

The basic circuit versatility stems from the GAP-Ol's architecture. The circuit features two differential input transconductance amplifiers, two low-glitch current mode switches,
an output voltage buffer amplifier, and a precision comparator.

CHANNEL B

0-'-1---1

GAP-In

SWITCHES SHOWN FOR

Both transconductance amplifier outputs are switched by
current-mode switches into the voltage follower output
buffer, thus providing two digitally selectable signal paths
through the device. Gain through the two channels may be
different in both sign and magnitude depending upon feedback selection. An external capacitor provides loop compensation and doubles as a hold or "memory" capacitor when the
GAP-Ol functions as a dual-channel sample/hold amplifier.
Offset voltage and charge transfer errors are trimmed
by using the "Zener-Zap" trim technique. The output buffer
features a FET input stage to reduce droop rate error in S/H
applications. A bias current cancellation circuit minimizes
droop error at high ambient temperature.

~"'O
CHANNEL B =0

CONTROL LOGIC

The inclusion of a preCision comparator on chip increases
the GAP-Ol's versatility and cost effectiveness in data
conversion applications. The output high voltage level is set
by external resistors. This scheme maximizes noise immunity and permits interface to all standard logic famil ies .

CiiA

Ch B

0
0

0
1
0

OUTPUT
to C
Channel A
Sum
Hold Last Input
Channel B

Manufactured under the following patent: 4.285.051.

15-4

1/86, Rev. A

---------1~ GAP-Ol GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

ABSOLUTE MAXIMUM RATINGS (Note 2)

Operating Temperature Range
GAP01AX, BX ...................... -55'C to + 125'C
GAP01 EX, FX ....................... -25' C to +85' C
GAP01EP, FP .......................... 0'Cto+70'C
DICE Junction Temperature (T J) ••••••• -65' C to + 150' C

Supply Voltage ................................... ± 18V
Power Dissipation .............................. 500mW
Input Voltage. . . .. . . . .. . . . . . . ... Equal to Supply Voltage
Logic and Logic Ground
Voltage ...................... Equal to Supply Voltage
Output Short-Circuit Duration ................. Indefinite
Amplifier A or B Differential Input Voltage .......... ±24V
Comparator Differential Input Voltage .............. ±24V
Comparator Output Voltage ... Equal to Positive SupplyVoltage
Hold Capacitor Short-Circuit Duration ......... Indefinite
Storage Temperature ................. -65'C to +150'C
Lead Temperature (Soldering, 60 sec) ............. 300'C

(NOTE 1)

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

IS-Pin DIP (X)

IOO'C

IOmW/'C

IS-Pm DIP (P)

50'C

IOmW/'C

NOTES:
1. Maximum package power diSSipation vs. ambient temperature.
2. Absolute rallngs apply to both DICE and packaged parts, unless otherwise
noted.

DICE CHARACTERISTICS

PIN CONNECTIONS

CHANNEL B
LOGIC
GROUND
COMPARATOR
OUTPUT

INVERTING
INPUT
NQNINVERTING

INPUT
BNULL
INVERTING
INPUT (A)
NON INVERTING
INPUT (A)

B NULL

DIE SIZE 0.090 X 0.100 Inch, 9,000 sq. mils
(2.286 X 2.54; 5.8 sq.mm)

INVERTING
INPUT (8)
NON INVERTING
INPUT (8)

1. CHANNEL (8)

2.

v+

3. Your
4.
5.
6.
7.
8.
9.

18-PIN HERMETIC DIP
(X-Suffix)
EPOXY DIP
(P-Sufflx)

CH
(A) NULL
(A) NULL
INVERTING INPUT (AI
NONINVERTING INPUT (AI
V-

10.
11.
12.
13.
14.
15.
16.
17.
18.

NON INVERTING INPUT (8)
INVERTING INPUT (8)
(81 NULL
(81 NULL
COMPARATOR NONINVERTING INPUT
COMPARATOR INVERTING INPUT
COMPARATOR OUTPUT
LOGIC GND
CHANNEL (A)

For additional DICE information refer to
1986 Data Book, Section 2.

I
15-5

1/86, Rev. A

~ GAP-O'I GENERAL-PURPOSE ANALOC SICNAL PROCESSlt~G SUBSYSTEM

ELECTRICAL CHARACTERISTICS

at

Vs = ± 15V, CH = 1000pF, TA = 25° C.
GAP01A/E

PARAMETER

SYMIOL

MIN

CONDITIONS

TVP

GAP01B/F

MAX

MIN

TVP

MAX

UNITS

"am" AMPLIFIERS A, I
Zero-Scale Error

Vzs

4

3

7

mV

Input Offset Voltage

Vos

3

3

6

mV

Input Bias Current

16

80

150

80

250

nA

Input Oflset Current

los

20

40

50

100

Voltage Gain

Ay

Open-Loop
Bandwidth

BW

Ay=1

Common-Mode
Rejection Ratio

CMRR

-IOVSVCMS+IOV

Power Supply
Rejection Ratio

18

PSRR

±9VSVsS±18V

Input Voltage Range

VCM

(Note 2)

Slew Rate

SR

Feedthrough Error

25

10

0.4

80

74

90

VlmV

0.4

MHz

80

dB
dB

86

96

76

96

±11.5

±12

±11.5

±12

V

0.5

Vlp.a

0.5
.1VIN = 20V. CHA = I. CHB = 0 (Note 2)

nA

25

66

80

Acquisition Time to
0.1 % Accuracy

taq

20V Step. AYCL = +1 (Note 2)

41

Acquisition Time to
0.01 % Accuracy

taq

20V Step. AYCL = +1 (Note 2)

45

66
70

80
41

dB
70

liS

p.a

45

COMPARATOR
Input Offset Voltage

Vos

0.5

1.5

I

3

mV

Input Bias Current

16

700

1000

700

1000

nA

Input Offset Current

los

75

300

75

300

Voltage Gain

Ay

2kO Pull-up Resistor to 5V (Note 2)

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio

nA

5

7.5

3.5

7

VlmV

-IOVSVCMS+IOV

82

106

82

106

dB

PSRR

±9VSVsS±18V

76

90

76

90

dB

Input Voltage Range

VCM

(Note 2)

±11.5

±12.5

±11.5

±12.5

Low Output Voltage

VOL

ISINK S 5mA. Logic GND = OV

-0.2

0.15

04

-0.2

0.15

0.4

V

"OFF" Output
Leakage Current

IL

VOUT=5V

25

80

25

80

IIA

Output ShortCircuit Current

Isc

VOUT=5V

12

45

12

45

mA

ts

5mV Overdrive. (Note 3)
2kO Pull-up Resistor to 5V

Response TI me

7

7

ISO

V

ISO

ns

DIGITAL INPUTS-CHA, CHI (Note 3)
Logic "I" Input Voltage

VH

Logic "0" Input Voltage

VL

2

2

Logic "I" Input Current

IINH

VH= 3.5V

0.02

Logic "0" Input Current

IINL

VL =0.4V

1.6

10

0.02

0.07

V

0.8

0.8
0.02
2

V

!IA
10

/loA

0.1

mVims

MISCELLANEOUS
Droop Rate

VOR

TI = +25°C (Note I)

Output Voltage Swing.
AmplifierC

VOP

RL = 2 5k

Short-Circuit Current:
AmplifierC

Isc

±11.5
7

SWitch Aperture Time
SWitch SWitching Time
Slew Rate. Amplifier C

SR

RL = 2.5k

Power Supply Current

ISY

No Load

±12.5
IS

±II
40

7

±12
IS

V

40

mA

tap

75

75

ts

50

50

ns

2.5

2.5

VIliS

5

6

NOTES:
I. Due to limited production test times the droop current corresponds to
junction temperature (TI ).

ns

9

mA

2. Guaranteed by design.
3. Channel A = "I". Channel B = "0".

15-6

1/86, Rev. A

-

-

-

---

--

-

----

------_.-

IfMD

GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

ELECTRICAL CHARACTERISTICS at Vs=± 15V. CH= 1000pF. -55'C :5TA:5125'C forGAP01AX & BX; -25'C :5TA:585'C
& FX. and O' C :5 TA :5 70' C for GAP01 EP & FP.

for GAP01 EX

PARAMETER

SYMIOL

GAP01A/E
MIN
TVP
MAX

CONDITIONS

GAP01B/F
MIN
TYP
MAX

UNITS

"8m" AMPLIFIERS A. I
Zero-Scale error

Vzs

4

7

6

12

mV

Input Offset Voltage

Vos

3

6

5

10

mV

Average Input
Offset Drift

TCVos

(Note II

Input Bias Current

Ie

Input Offset Current

los

Voltage Gain

Av

Common-Mode
Rejection Ratio

CMRR

-10V:5 VCM :5 + 10V

Power Supply
Rejection Ratio

PSRR

±9V:5 Vs:5 ±18V

Input Voltage Range

VCM

(Note II

Slew Rate

SR

Acquisition Time to
0.1% Accuracy

taq

-3

-6

-5

-6

p.V/'C

160

250

leo

500

nA

30

lOa

30

150

7.5
74

82

72

nA

9

V/mV

80

dB
dB

80

90

70

90

±11

±12

±10.5

±12

V

0.4

0.4

VII's

60

60

p.S

20V Step, AVCL = + I

COMPARATOR
Input Offset Voltage

Vos

Average Input
Offset Drift

TCVos

Input Bias Current

18

Input Offset Current

los

Voltage Gain

Av

2kll Pull-up Resistor to 5V, (Note I I

Common·Mode
Rejection Ratio

2.5
(Note II

mV

-4

-6

-4

-6

p.V1'C

1000

2000

1100

2000

nA

lao

600

lOa

600

nA

4

6.5

2.5

6.5

V/mV

80

lao

80

92

dB

72

82

72

86

dB

CMRR

-10V:5 VCM :5 +10V

Power Supply
Rejection Ratio

PSRR

±9V:5 Vs:5 ±18V

Input Voltage Range

VCM

(Note II

±11

Low Output Voltage

VOL

ISINK :5 5mA, Logic GND = 5V

-0.2

"OFF" Output
Leakage Current

'L

VOUT=5V

Output ShortCircuit Current

Isc

VOUT= 5V

Response Time

ts

5mV Overdrive, (Note 31
2kll Pull-up ReSistor to 5V

6

±11
0.15

0.4

25
10

-0.2

V
V

lOa

25

leo

p.A

0

45

10

45

mA

U

ns

~

200

DIGITAL INPUTS-CHA, CHI (Note 3)
VH

Logic "0" Input Voltage

VL

Logic "1" Input Current

I'NH

VH = 3.5V

0.02

Logic "0" Input Current

I'NL

VL = 0.4V

2.5

V
0.8

0.8
0.02

15

2.5

V
p.A

15

p.A

10

mV/ms

MISCELLANEOUS
Droop Rate

VOR

Tj = Max. Operating Temp., (Note 2)

Output Voltage Swing:
AmplifierC

VOP

RL =2.5k

Short-Circuit Current:
AmplifierC

Isc

Switch Aperture Time

tap

Slew Rate: Amplifier C

SR

RL = 2.5k

Power Supply Current

ISY

No Load

10
±11

±12
12

40

75

±10.5

±12

6

12

V

40

75

mA
ns

2
5.5

VII's
8

6.5

10

mA

NOTES: See next page.

15-7

Z

0.4

200

Logic "1" Input Voltage

Vl

0.15

1/86, Rev. A

f::

~
~
U

I-LI
p..
en

---------l~ GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

WAFER TEST LIMITS at Vs =

±15V. CH

PARAMETER

SYMBOL

=

1000pF. TA

=

25°C.

GAP-01N
CONDITIONS

LIMIT

UNITS

"gm" AMPLIFIERS A. B
Zero-Scale Error

Vzs

Input Offset Voltage

Vos

Input Bias Current

18

2S0

Input Offset Current

los

100

nAMAX

Voltage Gain

Av

10

V/mV MIN

Common-Mode
Rejection Ratio

CMRR

-IOV"VeM,,+IOV

74

dBMIN

Power Supply
Rejection Ratio

PSRR

±9V" Vs " ±18V

76

dBMIN

Input Voltage Range

VeM

mVMAX
mVMAX

(Note I)
~V'N

Feedthrough Error

= 20V. CHA = I, CHB = 0 (Note I)

nAMAX

±II.S

VMIN

66

dBMIN

COMPARATOR
Input Offset Voltage

Vos

3

mVMAX

Input Bias Current

18

1000

nAMAX

Input Offset Current

los

300

nAMAX

Voltage Gain

Av

2kn Pull-up Resistor to SV (Note I)

3.S

V/mV MIN

Common-Mode
Rejection Ratio

CMRR

-IOV" VeM " +IOV

82

dBMIN

Power Supply
Rejection Ratio

PSRR

±9V" Vs ,,±18V

76

dBMIN

Input Voltage Range

VeM

(Note I)

Low Output Voltage

VOL

ISINK S SmA, Logic GND = SV

"OFF" Output
Leakage Current

IL

Output ShortCircuit Current

Ise

±II.S

VMIN

0.4
-02

VMAX
VMIN

VOUT=SV

80

/,AMAX

VOUT= SV

4S
7

mAMAX
mAMIN

0.8

VMAX

DIGITAL INPUTS-CHA. CHB (Note 3)
Logic "I" Input Voltage

VH

Logic "0" Input Voltage

VL

VMIN

Logic "I" Input Current

IINH

VH = 3.SV

Logic "0" Input Current

IINL

VL = 0.4V

Droop Rate

VOR

Output Voltage Swing.
Amplifier C

VOP

/'AMAX
10

/'AMAX

T = 2SoC
T~ = 2SoC(See Note 2)

0.1
0.20

mV/ms MAX
mV/ms MAX

R L =2.Sk

±II

VMIN

40

mAMAX
mAMIN

9

mAMAX

MISCELLANEOUS

Short-Circuit Current:
Amplifier C

Ise

Power Supply Current

ISY

No Load

NOTES:
I. Guaranteed by design.
2. Due to limited production test times the droop current corresponds to
junction temperature (T,). The droop current vs. time (after power-on)
curve clarifies this pOint. Since most devices (in use) are on for more than

I second, PMI specifies droop rate for ambient temperature (TA) also. The
warmed-up (TAl droop current specification is correlated to the junction
temperature (T,) value. PMI hasa droop current cancellation circuit which
minimizes droop current at high temperatures.
3. Channel A = "I", Channel B = "0".

Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing.

15-8

1/86, Rev. A

--------l1fHD

GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ±15V, CH = 1000pF, TA = 25°C.
GAP-01N
PARAMETER

SYMBOL

CONDITIONS

TYPICAL

UNITS

05

VII'S

"Um" AMPLIFIERS A, B
Slew Rate

SR

Acquisition Time to 0.1 % Accuracy
Acquisition Time to 0.01 % Accuracy

tag

20V step, AVCL = 1

41

I'S

tag

20V step, AVCL = 1

45

p.s

ts

5mV Overdrive
2kn Pull-up Resistor to +5V

150

ns

ns

COMPARATOR

Response Time
MISCELLANEOUS
Switch Aperture Time

tap

75

Switching Time

ts

50

ns

Slew Rate: Amplifier C

SR

2.5

VII's

RL = 2.5kn

TYPICAL PERFORMANCE CHARACTERISTICS
SMALL-SIGNAL OPEN-LOOP
GAIN/PHASE VI FREQUENCY

ACQUISITION TIME VI
INPUT VOLTAGE STEP SIZE
~r-----'------r----~-----,

DROOP CURRENT
VB TEMPERATURE
0

TA = 25"C
CH = 1000pF

,

45

z

""

8

'35
'80

i!:

10k

lOOk

1M

10M

1

1
1
IL

~
'O~~~~-----r-----+----~

1k

8

,

S

90

A~ OP~RATIING

UJ'TS
TEMPERATURE

40~----1------r-----+--~~

0~0----~------'~0----~'5----~20

FREQUENCY (Hz)

2

L

0
-75 -50

,/
-25

INPUT STEP (VOLTS)

AMPLIFIER CHARGE INJECTION
ERROR VI INPUT VOLTAGE
AND TEMPERATURE

HOLD-MODE POWER SUPPLY
REJECTION VI FREQUENCY

+'0

0

50

75

100 125 150

LOGIC INPUT CURRENT VB
LOGIC INPUT VOLTAGE

'00

POLARITY OF ERROR
MAY BE POSITIVE OR

TA =2S0C
VIN = OV
CH '" 1000pF_

C H "" 1000pF

TA "'2S"C

NEGATIVE

80

l-

I-r- -551c

(+15v +lV SIN wT)

-5V

+5V

+12S"C

-0.5

-, 0

-

60

NEGATIVE SUPPLY
(-15V +1V SIN w)

40

I
r- +2S"C

'" ~
......,

~

+2~C

-ss"c

ERROR (mV)

+'0v

20

+'2lC

-

-2

LOGIC 0

CHANNEL A = 1
CHANN)EL B = 0

o
'0

'00

,k

FREQUENCY (Hz)

15-9

100k

'M

-2

LOGIC GRJUND = OV

I

-3
'Ok

I

LOJ,C'

POSITIVE SUPPLY

'OV

25

TEMPERATURE fOCI

I

I

I

-1
LOGIC INPUT VOL TAGE (VOLTS)

1/86, Rev_ A

----------l!EMD

GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

TYPICAL PERFORMANCE CHARACTERISTICS
COMPARATOR INPUT BIAS
CURRENT VB DIFFERENTIAL
INPUT VOLTAGE

COMPARATOR OUTPUT
VOLTAGE VB OUTPUT
CURRENT AND TEMPERATURE

+3

10

;=

Vs .1±15V

i<

lNPJr CURR1ENT Muir

TA "'2SoC

i!:

'":E:w

1l
2:

+2

>-

~

"0w
"~

+1

f5

OTHER

OTHER

'"

f--INPUT

'"
"::1"

f--INPUT

ATOV

AT-1OV

OTHE
_INPUT
AT +1 OV

0

0.8

~

>

a;

02

00

>-

;;;

-02
-1

-10

-5

10

~
~

~

z

"'"

0

3

80

~

TEST CONDITION

~

w

z
z

~

"'/

0
~
~
w

z
z

w

"

"

c:>
>:="
0"

'" \

w

AMPLIFIER A (B) OFF, INPUT = 20V Pk-Pk
AMPLIFIER B (8) ON, INPUT = OV

lk

10k

lOOk

OmV

~~
1M

-5mV

-50

100M

FREQUENCY (Hz)

~

80

100

120

I

-05

0.0

05

"

10

15

20

VB

TEMPERATURE

0

C-

uV
-5mV ' - _......_ ' - - - ' ' - - ' _....._
50

100

150

200

250

300

-50

50

100

150

200

......_
250

.....
300

TIME (nsac)

COMPARATOR OUTPUT
RESPONSE TIME (2kO
PULL-UP RESISTOR, TA = 25° C)

5V
w

w

~

~

3V

OmV

>

>~

":=

2V

-5mV

i<

"0

0

"C;
~

OV

0

>

~

:=
TIME (20;,ts/DIV )

:\
-1.0

+5mV

>

"

\

4V

~

0

TO +5V

INPUT VOL TAGE (mV)

LARGE-SIGNAL
NON INVERTING RESPONSE

"C;

>-

~RL =1.0kn

TO +5V

T A '!25'C

w

0

RL "'2.0kn_

1"'; ,I'"

,

RESISTOR'" 2kO

~
ov

2

-15

140

>

>

c

"C;

60

TIME (nsac)

LARGE-SIGNAL
INVERTING RESPONSE

~w

5

5o

1/ IV
_T)5507~ 1'+125
_

OV
+5mV

o
100

2V

~

~

10

3V

CI

20

1

40

pLLL-U~

4V -

1V

~

""

~

COMPARATOR RESPONSE TIME
5V

;!

60 I--C H '" 1000pF, AMPLIFIER
A AND B CONNECTED
IN +1 GAIN
40

3

10 - OUTPUT SINK CURRENT (rnA)

T~' 25°~

100

h ~V

o
20

CHANNEL TO CHANNEL ISOLATION
VB FREQUENCY

in

w

"

~
~

V

15

=VIN

NON-INVERTING INPUT = OV

4

2:

~

~

INPUT VOLTAGE (VOL IS)

120

c:

J~

0.4

INVERTING INPUT

~

_55°C~

>

i<

-15

+25°~=-",

06

Vs = ±15V
TA =25°C

I

+1 250C

BE LIMITED TO LESS-THAN lmA

>-

e
1
>-

COMPARATOR TRANSFER
CHARACTERISTIC

>~

0

>-

;;;

lV
OV

TIME (20;,ts/DIV )

15-10

TIME (50ns/DIV )

1/86, Rev. A

--

---- - - - - - - -

--------IlfMD

GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

TYPICAL PERFORMANCE CHARACTERISTICS
COMPARATOR
SETTLING TIME FOR
-10V TO OV STEP INPUT

COMPARATOR
SETTLING TIME FOR
+10V TO OV STEP INPUT

COMPARATOR OUTPUT
RESPONSE TIME (2kO
PULL-UP RESISTOR, TA = 25° C)

ov

>

~

5V

w

~
~

"

OUTPUT SWING OF
COMPARATOR VI
SUPPLY VOLTAGE

,.
"

10

0:

0

:;
~

-2

0
w

"

-6

I-

-10

"

-1'

"
~

0

r--- V+ •

-

...

...

~ ... ~

0:

0:

3V

OmV

I-

2V

-6mV

-1.

~

~

0:

§"

V-

+2SoC

+125°C,_

~

"

...
~

~

-2

I!!o.

-10

16

12
SUPPLY VOLTAGE +V AND

"'-

r-

V-

"

-1.

-I

r- V~SUPPLY

-2

~

r-=-..

-10 f-- V-SUPPLY

~

12
SUPPLY VOLTAGE +V AND

.--

...-=""I
+25°C

-56'6
+2SOC

"""'=

,.

-1.

I--i-----

~120'c ~
~

-1.

'~

16

-v

...

.....::::;:~+12..C _

~
"'~

'"'l!!!I!oo.

•

(VOLTS)

12
SUPPL V VOLTAGE +V AND

15

,.

-v (VOLTSI

SUPPLY CURRENT VI
SUPPLY VOLTAGE
~

-

INPUT + RANGE ~C

0

>

~

w

w

§

II

-55°C

+25°C

+126°1::

+25°C

...

......".

~ +125'f-

-

~

,.
15

12
SUPPLY VOLTAGE +V AND

......".

-v

....

,.

•o

12
SUPPLY +V AND

(VOL TSI

15-11

15

18

-v (VOL TSI

1/86, Rev. A

--------I~ GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

TYPICAL PERFORMANCE CHARACTERISTICS

,.

AMPLIFIER "A" OR "B"
VOLTAGE TO CURRENT
TRANSFER FUNCTION
(VIN vs leH)

INPUT RANGE OF LOGIC
GROUND vs SUPPLY VOLTAGE
600

'4

~

I

-2

-6

-'0

"$!i!!;:,

-1-

"

~

~~i!;!,

-,.

+12 C
-400

"
'2

4

.4

-200

-~

-'4

..oi! ~

_5

r

+25°C

J I YI>'"

'5

-11/,

1..lP'

1

ACCEPTABLE GROUND
PIN POTENTIAL IS
BETWEEN SLIDE LINES

_55cC

i!!;:,

I

200

+2SOC
-55°C

-

1_55 C

+125°C
~

"" ....

r,,'' ' ' '

I

+25°C

400

II'"

~~~ dillO'

'0

~

+12SOC

iii"

~

V If

-I-' l.t:

~

"~t
8

1~!! .~

7

,.

-600
....06

-04

GAP_01

"

=~

-02

0.0

7

02

OUT

10011

04

06

VIN (VOLTS)

SUPPLY VOLTAGE +V AND -V (VOLTS)

COMPARATOR OFFSET VOLTAGE
vs TEMPERATURE

A AND B AMPLIFIERS OFFSET
VOLTAGEvsTEMPERATURE

-,

-2~-+--+--4--~--~-+--+-~

~r--+--+--4--~--~-+--+-~

TEMPERATURE fOCI

APPLICATION CIRCUITS
GAP-01 WITH POSITIVE AND NEGATIVE GAINS

GAP-01IN UNITY GAIN (+1) CONFIGURATION
CHANNELAI
CHANNEL B

0----,-------,

CHANNEL A

INPUT

R1

OUTPUT

>-1-1--+--<>

OUTPUT
CHANNEL A

INPUT
R,

CHANNEL B

INPUT

Rs

CHANNEL B

INPUT

R.
CHANNEL B

c,

.::;J; 1000pF
R,'" R3IR.

15-12

Rs= R21Rs

1/86, Rev. A

--------IIEMD

GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

APPLICATION CIRCUITS
ALTERNATE GAIN CONFIGURATION
R,

CHANNEL A

o----.....f - - - - - - - - ,

OUTPUT

>--.-+-*--<>

R,

A3 >2 41dl

R4>24kn

CH
CHANNEL B

l

0-------------'

1000PF

IF BOTH CHANNEL A AND CHANNEL B HAVE
THE SAME POSITIVE VOLTAGE GAIN, A SIN-

GLE VOL lAGE DIVIDER SETS THE GAIN FOR
BOTH CHANNELS

ABSOLUTE VALUE CIRCUIT WITH POLARITY PROGRAMMABLE OUTPUT
INPUT

---

+sv

27kn

4.7kn

10kn

1%

POLARITY

*
1iI:.I

II

CONTROL
SIGNAL

TRANSISTOR INCREASES COMPARATOR OAIN

Ioi:'::::'::'

L::::.I

I,i;,OI

~

'1

~

il'4

-- =-

~

TRACE 1 INPUT SIGNAL

~

I

IIIJ

1.::::.1
'

II!

~

II

TRACE 2" CHANNEL A/CHANNEL B
CONTROL SIGNAL

II-

'1 ~

:~

I·

:!'II

TRACE 3: OUTPUT POLARITY CONTROL
SIGNAL .. 0

~..
~, ~

:7'1

15-13

:!'II

:~

I! ~

~

•-

TRACE 1 INPUT SIGNAL

TRACE 2 CHANNEL AlCHANNEL B
CONTROL SIGNAL

••

TRACE 3- OUTPUT POLARITY CONTROL
SIGNAL = 1

1/86, Rev. A

--------t~ GAP-01 GENERAL·PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

TWO·CHANNEL SAMPLE/HOLD AMPLIFIER

00
+15Y

-15V

R,

INPUT Ao--.N>I'-+--..........!...H

,.n

INPUT B o----,MI'------"'ll-'

eH

4

,.

500PF~
CiiiNNErifCHANNEL B

13

0 - -.....--If',~'!;'0::0~

SAMPlElii0L6 0 - - - -......,.,.1.:.:'/
CHANNEL A GAIN ... -

CHANNEL B GAIN -1

~ = -10
+~ =2

TRACE 1: INPUT SIGNAL B (1V1DIY.)

TRACI! 2: INPUT SIGNAL A (O.5V/DIV.)
TRACE S. CHANNEL A/CHANNEL B
CONTROL SIGNAL (IVlDIY.)

TRACE 4' OUTPUT WITH SAMPLE/HOm
= "1" (5Y/DIY.)

TRACE l' INPUTSIQNAL B (1Y1DIV)

TRACE 2: SAMPLE/HOili (5V/DIY.)
CONTROL SIGNAL
TRACE 3' OUTPUT SIGNAL (2V1DIY)
~CHANNEL B = "1"

15-14

1/86, Rev_ A

---------I~ GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

APPLICATION CIRCUITS
LOGIC LEVEL TRANSLATION FOR
GAP-01 SINGLE SUPPLY OPERATION

DIGITAL GROUND CONNECTION FOR
SINGLE SUPPLY OPERATION

v+ '"
V+ - 16V

GAP-01
t----. TO
LOGIC

R" S.1kO

GAP-Ol

17
DIG
GNO

15V

R1 ., 3.3kO

CONTROL

'-

-14V

R2

OIl

3900

TT~>----+

.:s ~
.:s ~

~

R3 = 1kO

COLLECTOR

LOGIC HIGH VOLTAGE = 44V
LOGIC LOW VOLTAGE'"' 1.SSV

FOUR QUADRANT MULTIPLYING DAC

A INPUT

en
CH

Z

0

J

f::

u

Z
::J
~

81

g

8"

B INPUT

U

~
~

en

15-15

1/86, Rev. A

---------i~ GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

APPLICATION CIRCUITS
SYNCHRONOUS DEMODULATION OF LVDT SIGNAL
,Ol

e~

OV

Vc VOH
TRACE 2

ANALOG INPUT
NX)
l;.!s/DIV

APPLICATION INFORMATION

R1

R2

5

3.5 2.7K 6.2K

5

5.0 2.7K

.,

15

3.5 4.7K 1.5K

15

5.0 4.7K 2.4K

15

7.5 7.5K 7.5K

15

10.0 7.5K 15K

Vc

R,~--

Isink

",.",(v

c '_)

VOH

CAPACITOR RECOMMENDATIONS
The external capacitor (CH) serves as the compensation
capacitor and hold capacitor in sample/hold applications.
Stable operation requires a minimum value of 500pF. Larger
capacitors may be used to lower droop rate errors. but acquisition time will increase and bandwidth decrease.

The maximum comparator high output voltage (VOH) should
be constrained to: VO H (max) < V+ -2V

The capacitor should have very high insulation resistance
and low dielectric absorption. Fortemperatures below 85° C.
a polystyrene capacitor is recommended. while a Teflon
capacitor is recommended for high temperature environments.

CAPACITOR GUARDING AND GROUND LAYOUT
Ground planes are recommended to minimize ground path
resistance. Separate analog and digital grounds should be
used. The two ground systems are tied together only at the

Table I

15·17

1/86, Rev. A

II

--------I~ GAP-01 GENERAL-PURPOSE ANALOG SIGNAL PROCESSING SUBSYSTEM

common system ground. This avoids digital currents returning to the system ground through the analog ground
path.

r-----------.----ov+

The CHterminal (Pin 4) isa high-impedance point. To minimize gain errors and maintain the GAP-01's inherently low
droop rate, guarding Pin 4as shown in Figure 2 is recommended.
CHANNEL A

OR
CHANNEl B
DIGITAL
GROUND

,.
CURRENT TO

17

CONTROL MODES

Figure 3
16

15

ZERO-SCALE ERROR ADJUSTMENT
For sample/hold applications the zero-scale error (Vos plus
charge injection error) can be adjusted to zero. With the input
to each channel equal to zero, the GAP-01 is switched
between the sample mode (either channel A or channel 8
active) and the hold mode (channel A-1, channel 8=0). The
output is adjusted to read zero when the unit is in the hold
mode.

14
13

REPEAT ON
"COMPONENT SIDE" OF
PC BOARD IF POSSIBLE

12
11

10

The Vzs trim circuit is identical to the Vos trim circuit.

Figure 2

OFFSET VOLTAGE ERROR ADJUSTMENT Offset voltage
through either channel A or channel 8 may be nulled with an
external 100kO, potentiometer as shown below.

LOGIC CONTROL
The transconductance amplifier outputs are switched by the
digital logic signals applied at Channel A and Channel 8 pins.
Two signal paths through the GAP-01 are possible.

The logic threshold voltage is 1.4 volts when digital ground is
at zero volts. Other threshold voltages (V TH ) may be selected
by applying the formula:

v-

v-

VTH ~ 1.4V + Digital Ground Potential.
Figure 3 shows the simplified logic control circuit. For proper
operation, digital ground must always be at least 3.5V below
the positive supply and 2.5V above the negative supply. The
logic signals must always be at least 2.8V above the negative
supply.

13
A NULL

Operating the digital ground at other than zero volts does
influence the comparator output low voltage. The VOL level is
referenced to digital ground and will follow any changes in
digital ground potential:

B NULL

GAP-01

Figure 4

VOL ~ 0.2V + Digital Ground Potential.

15-18

1/86, Rev. A

PKD-Ol
MONOLITHIC PEAK DETECTOR
(WITH RESET-AND- HOLD MODE)
PrecIsion Mo])olithics Inc

FEATURES
• Monolithic Design for Reliability and Low Cost
• High Slew Rate .............................. O.5V1p.s
• Low Droop Rate
TA=25°C ............................... O.1mVlms
TA= 125°C .............................. 10mV/ms
• Low Zero-Scale Error • .. . • .. . . . . . . . .. . . .. .. . . .. .. 4mV
• Digitally Selected Hold and Reset Modes
• Reset to Positive or Negative Voltage Levels
• Logic Signals TTL and CMOS Compatible
• Uncommitted Comparator on Chip

Through the DET control pin, new peaks may either be
detected or ignored. Detected peaks are presented as pOSitive output levels. Positive or negative peaks may be detected
without additional active circuits since amplifier A can operate as an inverting or noninverting gain stage.
An uncommitted comparator provides many application
options. Status indication and logic shaping/shifting are typical examples.

PIN CONNECTIONS

ORDERING INFORMATIONt
25°C

PACKAGE

Yzs
(mY)

14-PIN DUAL-IN-LiNE PACKAGE
HERMETIC'
PLASTIC

4

7
4

7

PKD01AY'
PKD01BY'
PKD01EY
PKD01FY

4

7

PKD01EP
PKD01FP

14-PIN HERMETIC DIP
(V-Suffix)

OPERATING
TEMPERATURE
RANGE

EPOXVDIP
(P-Sufflx)

MIL
MIL
IND
IND
COM
COM

• For dev.ces processed on total compl.ance to MIL-STD-883, add /883 after
part number Consult factory for 883 data sheet
tAli commercial and Industnal temperature range parts are available With
burn-.n For orderong onformat.on see 1986 Data Book, Sect.on 2.

FUNCTIONAL DIAGRAM

+IN

-IN

OUTPUT

V+

v-

GENERAL DESCRIPTION
The PKD-Ol tracks an analog input signal until a maximum
amplitude is reached. The maximum value is then retained as
a peak voltage on a hold capacitor. Being a monolithic circuit,
the PKD-Ol offers significant performance and package density advantages over hybrid modules and discrete designs
without sacrificing system versatility. The matching characteristics attained in a monolithic circuit provide inherent
advantages when charge injection and droop rate error
reduction are primary goals.

LOGIC

13

GND

DET

OUTPUT

-'N

Innovative design techniques maximize the advantages of
monolithic technology. Transconductance (gm) amplifiers
were chosen over conventional voltage amplifier circuit
building blocks. The "gm" amplifiers simplify internal
frequency compensation, minimize acquisition time and
maximize circuit accuracy. Their outputs are easily switched
by low glitch current steering circuits. The steered outputs
are clamped to reduce charge injection errors upon entering
the hold mode or exiting the reset mode. The inherently low
zero-scale error is reduced further by active "Zener-Zap"
trimming to optimize overall accuracy.

+'N

-'N

II

+'N

RST

RST

o

The output buffer amplifier features an FET input stage to
reduce droop rate error during lengthy peak hold periods. A
bias current cancellation circuit minimizes droop error at
high ambient temperatures.

DEl OPERATIONAL MODE

0
1
1

PEAK DETECT
PEAK HOLD
RESET

o

Indeterminate

SWITCHES SHOWN FOR

RST="O", DET="O"

Manufactured under the following patent· 4,285,051

15-19

1/86, Rev. A

-------------I~ PKD-01 MONOLITHIC PEAK DETECTOR

ABSOLUTE MAXIMUM RATINGS (Note 2)
Supply Voltage ................................... ±18V
Power Dissipation .............................. 500mW
Input Voltage. . • . • • . . . •• •• •• • . •• Equal to Supply Voltage
Logic and Logic Ground
Voltage ...................... Equal to Supply Voltage
Output Short-Circuit Duration ••.•••••••••.•••• Indefinite
Amplifier A or B Differential Input Voltage •••..•••.• ±24V
Comparator Differential Input Voltage ••.•••••••••.• ±24V
Comparator Output Voltage
.. .. .. .. .. .. .. .. .... Equal to Positive Supply Voltage
Hold Capacitor Short-Circuit Duration ••.•••.•. Indefinite
Lead Temperature (Soldering, 60 sec) .••..•••...•. 300°C
Storage Temperature Range
PKD01AY, PKD01BY •..•••.•••..•••. -65°C to +150°C
PKD01 EY, PKD01 FY ................ -65° C to + 150° C
PKD01EP, PKD01FP ................ -65°Cto+125°C

ELECTRICAL CHARACTERISTICS
PARAMETER

SYMBOL

at Vs

Operating Temperature Range
PKD01AY, PKD01BY .•••••.•.•.••••• -55°C to +125°C
PKD01 EY, PKD01 FY ••••.••••.••••••• -25° C to +85° C
PKD01 EP, PKD01 FP .................... 0° C to + 70° C
Dice Junction Temperature •.••.••••.•. -65°C to +150°C

PACKAGE
(Note 1)

MAXIMUM AMBIENT
TEMPERATURE
FOR RATING

DERATE ABOVE
MAXIMUM AMBIENT
TEMPERATURE

14-Pin DIP (Y)

80'C

10mwrc

14-Pin DIP (P)

50'C

6mwrc

NOTES:
Maximum package power dissipation vs. ambient temperature.

2. Absolute ratings apply to both packaged parts and DICE. unless
otherwise noted.

= ±15V, CH = 1000pF, TA = 25° C.
PKD01A/E
MIN
TYP
MAX

CONDITIONS

MIN

PKD01B/F
TYP
MAX

UNITS

"9 m" AMPLIFIERS A, B
Zero-Scale Error

Vzs

2

4

3

Input Offset Voltage

Vas

2

3

3

Input Bias Current

18

80

150

80

250

Input Offset Current

los

20

40

20

75

Voltage Gain

Av

Open-Loop
Bandwidth

BW

Av= 1

Common-Mode
Rejection Ratio

CMRR

-10VSVCM S+l0V

80

90

PSRR

±9VSVs S±18V

86

Input Voltage Range

VCM

(Note 1)

±115

Slew Rate

SR

Power Supply
Rejection Ratio

Feedthrough Error
Acquisition Time to

0.1 % Accuracy
AcqUisition TIme to

001% Accuracy

18

25

10

nA
nA
VlmV

0.4

MHz

74

90

dB

96

76

96

dB

±12

±115

±12

V

0.5

Vlp.s

80

dB

05
66

mV

25

04

tiV ,N = 20V. DET = 1. RST = O. (Note 1)

mV

80

taq

20V Step. AVCL = + 1. (Note 1)

41

taq

20V Step, AVCL = +1, (Note 1)

45

66
70

41

70

45

p's
p's

COMPARATOR
I nput Offset Voltage

Vas

Input Bias Current

18

Input Offset Current

los

Voltage Gain

Av

2kO Pull-up Resistor to 5V. (Note 1)

Common-Mode
Rejection Ratio

CMRR

-10V S VCM S +10V

PSRR

±9VSVs S±18V

VCM

(Note 1)

Power Supply
Rejection Ratio

Input Voltage Range

82

NOTES:
1. Guaranteed by design
2. Due to limited productIon test tImes, the droop current corresponds to

05

15

3

mV

700

1000

700

1000

nA

75

300

75

300

nA

7.5

3.5

7.5

VlmV

106

82

106

dB

76

90

76

90

dB

±115

±125

±11.5

±12.5

V

Junction temperature (TJ) The droop current vs. tIme (after power-on)
curve clanfiesthis pOInt. Since most devices (in use) are on for more than

warmed-up (TA) droop current specification IS correlated to the junction
temperature (Tl ) value PMI hasa droop current cancellation clrcUltwhich
mmimizes droop current at high temperature. AmbIent (TAJ temperature
specIfIcations are not subject to production testing.

1 second, PMI specifies droop rate for amblenttemperature (TA) also The

DET= 1, RST =0.

15-20

1/86, Rev. A

~ PKD-Cl1 MONOLITHIC PEAK DETECTOR

ELECTRICAL CHARACTERISTICS

at Vs= ±15V. CH = 1000pF. TA = 25°C. (Continued)

PKD01A/E
PARAMETER

Low Output Voltage
"OFF" Output
Leakage Current

Output ShortCircuit Current
Response Time

SYMBOL

VOL

CONDITIONS

MIN

ISINK '; 5mA, Logic GND = 5V

TYP

-0.2

PKD018/F

MAX

MIN

TYP

MAX

0.t5

0.4

-0.2

0.15

0.4

V

80

25

80

~A

45

12

45

mA

IL

VOUT= 5V

25

Isc

VOUT= 5V

12

ts

5mV Overdrive, (Note 3)
2kO Pull-up Resistor to 5V

7

150

150

UNITS

ns

DIGITAL INPUTS-RST, DET (See Note 3)

2

Logic "I" Input Voltage

VH

Logic "0" Input Voltage

VL

Logic "I" Input Current

IINH

VH = 3.5V

002

Logic "0" Input Current

IINL

VL =0.4V

1.6

10

1.6

10

p.A

Droop Rate

VOA

TJ = 25"C,
(See Note 2)
TA =25"C

0.01
002

0.07
0.15

0.Q1
0.03

0.1
0.20

mV/ms

Output Voltage Swing:
Amplifier C

VOP

DET=1
RL = 2.5k

Short-Circuit Current:
Amplifier C

Isc

2

V

0.8

0.8
0.02

V
p.A

MISCELLANEOUS

±".5
7

Switch Aperture Time
Switch Switching Time
Slew Rate: Amplifier C

SR

RL = 2.5k

Power Supply Current

ISY

No Load

±12.5

±11

15

40

±12

V

15

40

mA

taE!

75

75

ts

50

50

ns

2.5

2.5

V/~s

ns

6

9

mA

ELECTRICAL CHARACTERISTICS atVs=±15V, CH= 1000pF. -55°C:5TA :5 125°C for PKD01AY. PKD01 BY, -25°C:5TA:585°C
for PKD01 EY, PKD01 FY and O°C:5 TA:5 70°C for PKD01 EP. PKD01 FP.
V)

PKD01A/E
PARAMETER

SYMBOL

CONDITIONS

MIN

TYP

PKD018/F

MAX

MIN

TYP

MAX

6

12

mV

10

mV

UNITS

"am" AMPLIFIERS A. B
Zero-Scale Error

Vzs

4

Input Offset Voltage

Vas

3

6

Average Input
Offset Drift

TCVos

-9

-24

-9

-24

p.V/"C

Input Bias Current

Is

160

250

160

500

nA

Input Offset Current

los

30

100

30

150

nA

Voltage Gain

Av

(Note 1)

7.5

9

9

VlmV

CMRR

-10V'; VCM '; +10V

74

82

72

Power Supply
Rejection Ratio

PSRR

±9VSVs:5±18V

80

90

Input Voltage Range

VCM

(Note 1)

±11

±12

Slew Rate

SR

Acquisition Time to
0.1% Accuracy

taq

20V Step, AVCL = +1, (Note 1)

80

dB

70

90

dB

±10.5

±12

V

0.4

0.4

VII'S

60

60

p.s

COMPARATOR

Input Offset Voltage

Vas
TCVos

Input Bias Current

Is

U

Z
::J
~

~

U

~

p.,
V)

Common-Mode
Rejection Ratio

Average Input
Offset Drift

Z

0
t==

2
(Note 1)

15·21

2.5

mV

-4

-6

-4

-6

p.V/"C

1000

2000

1100

2000

nA

1/86, Rev. A

II

~ PKD-01 MONOLITHIC PEAK DETECTOR

ELECTRICAL CHARACTERISTICS at Vs = ± 15V, C H = 1000pF, -55°C:s TA:S 125°C for PKD01 AY, PKD01 BY, -25°C:s TA:S 85°C
for PKD01 EY, PKD01 FY and O°C:s TA:S 70°C for PKD01 EP, PKD01 FP. (Continued)
PKD01A/E
PARAMETER

SYMBOL

CONDITIONS

MIN

Input Offset Current

los

Voltage Gain

Av

2kO Pull-up Resistor to 5V, (Note 1)

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio

PSRR

Input Voltage Range

VCM

(Note 1)

±11

Low Output Voltage

VOL

ISINKS 5mA, Logic GND = 5V

-0.2

"OFF" Output
Leakage Current

IL

Output ShortCircuit Current

Response Time

PKD01B/F

TYP

MAX

100

600

MIN

TYP

MAX

100

600

UNITS

nA

4

6.5

2.5

6.5

VlmV

-10V S VCM S + 10V

80

100

80

92

dB

±9VSVs S±18V

72

82

72

86

dB

±11

0.15

0.4

VOUT= 5V

25

100

Ise

VouT =5V

10

45

ts

5mV Overdrive,
2kO Pull-up Resistor to 5V

-0.2

6

200

V

0.15

0.4

V

100

180

p.A

10

45

mA
ns

200

DIGITAL INPUTS-RST, DET (See Note 3)

Logic "I" Input Voltage

VH

Logic "0" Input Voltage

VL

Logic "I" Input Current

IINH

VH =3.5V

0.02

IINL

VL = O.4V

2.5

15

2.5

15

p.A

Droop Rate

VOR

Tj = Max. Operating Temp
TA = Max. Operating Temp.
i5Ei' = 1. (Note 2)

1.2

10
20

3
6

15
20

mVlms

Output Voltage Swing'
AmplifierC

VoP

RL = 2.5k

Short-Circuit Current:
Amplifier C

Ise

Logic "0" Input Current

2

V

0.8

0.8

V
p.A

0.02

MISCELLANEOUS

Switch Aperture Time

tap

Slew Rate: Amplifier C

SR

Power Supply Current

ISY

2.4

±11

6

±12

12

±10.5

40

6

±12

12

V
40

mA

75

75

ns

RL =2.5k

2

2

VII's

No Load

5.5

6.5

10

mA

NOTES:

1. Guaranteed by design.
2. Due to limited production test times, the droop current corresponds to
junction temperature (TJ) The droop current vs. time (after power-on)
curve clarifies this pOint. Since most devices (in use) are on for more than
1 second, PMI specifies droop rate for ambient temperature (TA) also.

The warmed-up (TA) droop current speCifIcation is correlated to the
junction temperature (TJl value. PMI has a droop current cancellation
circuit which minimizes droop current at high temperature. Ambient (TAl
temperature specifications are not subject to production testIng.

3. DET=l, RST=O.

15-22

1/86, Rev. A

-------------IIEHD PKD·01

MONOLITHIC PEAK DETECTOR

DICE CHARACTERISTICS
1. RST (RESET CONTROL)

2. V+
3.
4.
5.
6.
7.
8.

OUTPUT
C H (HOLD CAPACITOR)
INVERTING INPUT (A)
NONINVERTING INPUT (A)
VNONINVERTING INPUT (B)

9.
10.
11.
12.
13.
14.

INVERTING INPUT (B)
COMPARATOR NONINVERTING INPUT
COMPARATOR INVERTING INPUT
COMPARATOR OUTPUT
LOGIC GROUND
DET (PEAK DETECT CONTROL)
A,B (A) NULL
C,D (B) NULL

DIE SIZE 0.090 x 0.100 Inch, 9000 sq. mils
(2.286 x 2.54mm, 5.8 sq mm)
For additional DICE Information refer to
1986 Data Book, Section 2.

WAFER TEST LIMITS at Vs = ±15V, C H = 1000pF, TA = 25°C.
PKD·01N
PARAMETER

SYMBOL

CONDITIONS

LIMIT

UNITS

"8m" AMPLIFIERS A, B
Zero-Scale Error

Vzs

Input Offset Voltage

Vos

mVMAX

Input Bias Current

6

mVMAX

250

nAMAX

Input Offset Current

lOS

75

nAMAX

Voltage Gain

Av

10

VlmV MIN

Common-Mode
Rejection Ratio

CMRR

-10V"VCM ,,+10V

74

dBMIN

Power Supply
Rejection Ratio

PSRR

±9V" Vs " ±18V

76

dBMIN

Input Voltage Range

(Note 1)

Feedthrough Error

tN'N

=20V. DET =1, RST =0, (Note 1)

±11.5

VMIN

66

dB MIN

COMPARATOR
Input Offset Voltage

Vos

Input Bias CUrrent
Input Offset Current

lOS

Voltage Gain

2kl1 Pull-up Resistor to 5V, (Note 1)

Common-Mode
Rejection Ratio

CMRR

Power Supply
Rejection Ratio

PSRR

Z

0
~

3

mVMAX

1000

nAMAX

300

nAMAX

3.5

VlmV MIN

82

dBMIN

U

76

dB MIN

rJ)

U

Z
::J
t:l...

~
~

Input Voltage Range

±9V"Vs ,,±18V
(Note 1)

Low Output Voltage

ISINK " 5mA, Logic GND

= 5V

"OFF" Output
Leakage Current
Output ShortCircUIt Current

rJ)

ISC

NOTES:
1. Guaranteed by design.
2. Due to limited production test times, the droop current corresponds to
junction temperature (T)). The droop current vs. time (after power-on)
curve clarifies this point. Since most devices ( in use) are on formere than
1 second, PMI specifies droop rate for ambient temperature (TA) also. The

±115

VMIN

0.4
-0.2

V MAX
VMIN

80

~AMAX

45
7

mAMAX
mAMIN

warmed-up (TA) droop current specification is correlated to the junction
temperature (TJ) value. PM! hasa droop current cancellation circuit which
minimizes droop current at high temperature. Ambient (TA) temperature
specifications are not subject to prod uction testing.
3. DET = 1, RST = O.

15-23

1/86, Rev. A

Pot

I

_ _ _ _ _ _ _ _ _ _ _ _--\~ PKD-01 MONOLITHIC PEAK DETECTOR

WAFER TEST LIMITS at Vs= ±15V, CH= 1000pF, TA = 25°C. (Continued)
PKD-01N
PARAMETER

SYMBOL

LIMIT

CONDITIONS

UNITS

DIGITAL INPUTS-RST, DET (See Note 3)
Logic "1" Input Voltage

2

VMIN

Logic "0" Input Voltage

0.8

VMAX

Logic "1" Input Current

p.AMAX
VL ~ 04V

Logic "0" Input Current

10

p.AMAX

T~ ~ 25'~ (See Note 2)

25'C

01
0.20

mV/ms MAX
mV/msMAX

AL

25k

±11

VMIN

40

mAMAX
mAMIN

9

mAMAX

MISCELLANEOUS
Droop Aate

VOA

Output Voltage Swing'
Amplif,er C

VOP

Short~Circuit

T,

~

Current

AmplifierC
Power Supply Current

Ise
ISY

No l.oad

NOTES:
1, Guaranteed by design.
2, Due to limited production test times, the droop current corresponds to
junction temperature (Tl ), The droop current vs. time (after power-on)
curve clarifies this pOint. Since most devices (in use) are on for more than

1 second, PMI specifies droop rate for ambient temperature (TAJ also. The
warmed~up (TAJ droop current specification is correlated to the junctIon
temperature (TJl value. PMI has a droop current cancellation circUit which
mmimizes droop current at high temperatures. Ambient (TAJ temperature
speCifications are not subject to production testing.
DET~ 1. AST ~O
Electrical tests are performed at wafer probe to the limits shown. Due to variations in assembly methods and normal yield loss, yield after packaging is not
guaranteed for standard product dice. Consult factory to negotiate specifications based on dice lot qualification through sample lot assembly and testing,

TYPICAL ELECTRICAL CHARACTERISTICS at Vs = ± 15V, CH = 1000pF, and TA = 25° C, unless otherwise noted.
PKD-01N
PARAMETER

SYMBOL

CONDITIONS

TYPICAL

UNITS

"gm" AMPLIFIERS A, B
Slew Aate

SA

0,5

VII'S

Acquisition Time

ta

0,1% Accuracy, 20V step, AVCL ~ 1. (Note 1)

41

p.s

Acquisition Time

ta

0.01% Accuracy, 20V step, AVCL ~ 1, (Note 1)

45

p.s

150

ns

COMPARATOR
5mV Overdrive,
2kn Pull-up Resistor to +5V

Response Time

MISCELLANEOUS
Switch Aperature Time

tap

75

ns

Switching Time

ts

50

ns

Buffer Slew Rate

SR

2.5

VII'S

AL ~2.5kn

15-24

1/86, Rev. A

--------------1lfMD PKD·01 MONOLITHIC PEAK DETECTOR
TYPICAL PERFORMANCE CHARACTERISTICS

~
~

-

~

....
z

-2

...

~I:o.
~
-8

a: -10 - V - SUPPLY

i[
! -14

-1.

30

-5S0C:5TA~+126°C

0

~
~

36

~NPUT + RANGE'" V+

10

a:

0

!

...

>

:}

i

+25°C
~ +125I1 C _

........

1J 20

~

-<;o'C

12

15

o
-50

18

-25

75

100

-75 -50

125

-25

0

25

50

VS ·+1SV
TA - +25 C C

-56'C

-10

-5
+5
+10
~~~~~~++~~~~

p.

VIN (VOLTS)

~

~

10

125 150

AMPLIFIER B CHARGE INJECTION
ERROR VB INPUT VOLTAGE
AND TEMPERATURE

:~I

10.,

100

+10

t=

100

75

TEMPERATURE (oCI

100

~

...
~

50

WIDEBAND NOISE
VB BANDWIDTH

E

z

25

TEMPERATURE (OC)

1000

~g

I-

10

INPUT SPOT NOISE
VB FREQUENCY

l!l0

" r-..... """t-

-4

...
4

15

-2

0

~

i'..

25

w

"

SUPPLY VOLTAGE +V AND -V (VOLTS)

~

A, BiOS vs TEMPERATURE
40

.A-"

14

:;a:

""

A AND B AMPLIFIERS OFFSET
VOLTAGE VB TEMPERATURE

A AND B INPUT RANGE
VB SUPPLY VOLTAGE

,.

-05

~

0:

iii
1.

100

10

01

10

FREQUENCY (Hz)

100

AMPLIFIER A CHARGE INJECTION
ERROR VB INPUT VOLTAGE
AND TEMPERATURE

,.

+10
C H '" 1000pF

POLAR lTV OF
ERRORMAV BE

TA " +25°C

POSITIVE OR
NEGATIVE

+05

~
0

ffi

-5

OUTPUT VOLTAGE SWING VB
SUPPLY VOLTAGE (DUAL
SUPPLY OPERATION)
t--RL

+5

f-- -,

10

f-- V+ SUPPLV
~

~

+10

"

....

10kn

14

~

0:

-10

10

1000

BANDWIDTH (kHz)

~

--=-r

--

~

....,::::;~+126'C_

~

+25°C

-6SoC

Z

+125"C

...~::>

+25°C

::>
0

VIN (VOLTS)

-05

~

-55°C

-2

...

-8

-10

ro=-..

... "'<:::

f

-6SoC

~

~V-SUPPLY

+2SoC

""=
....

~

I---

+tSOC=

-14

-1.

10

12

4

15

,.

SUPPLY VOLTAGE +V AND -V (VOLTS)

1/86, Rev. A

15·25

~~-----------~~-

-

~

--------------t!fMD PKD-G1 MONOLITHIC PEAK DETECTOR
TYPICAL PERFORMANCE CHARACTERISTICS

,.
12.
100

~
~
I-

-2 .•

~

-5.0
-7.5
-100

:>
0

+2SoC

-5S o C

"z
:>

12

II
II

10

17'

Jilll ~
2mV ERROR

1\

75
50
25

~

OUTPUT ERROR VI
FREQUENCY AND
INPUT VOLTAGE

OUTPUT VOLTAGE vs
LOAD RESISTANCE

\ 200mV ERROR

+125°C

lUI

-12.5

-15

J!!!
+25 C

-55°C

...... +12SoC

\

"'"+lj' CJ.·C
rt

L
I

10

ERROR

o
100

01

100

1k

LOAD RESISTOR TO GROUND (kill

PKD-01 SETTLING RESPONSE

-

2Jmv

-

10k

lOOk.

1M

FREQUENCY (Hz)

LARGE-SIGNAL
INVERTING RESPONSE

PKD-01 SETTLING RESPONSE

e>
+10V

.~

";:!

ov

cl

>
I-

:>

~

5
TIME (20p.s/DIV'

SETTLING TIME FOR -10V
TO OV STEP INPUT

LARGE-SIGNAL
NONINVERTING RESPONSE

ov

~

c

".
"

E

!!!

~

;:!

cl

0

>

>

I-

l-

11

~

I-

:>
0

:>
0
TIME (20ps/DIV )

~>

.

E

!!!

ov

SETTLING TIME FOR +10V
TO OV STEP INPUT

TIME (20p.s/DIV J

15-26

OV

TIME (20ps/OIV J

1/86, Rev. A

--------------II£MD PKD-01 MONOLITHIC PEAK DETECTOR
TYPICAL PERFORMANCE CHARACTERISTICS
SMALL-SIGNAL OPEN LOOP
GAIN/PHASE vs FREQUENCY

OFF ISOLATION
vs FREQUENCY

CHANNEL TO CHANNEL
ISOLATION vs FREQUENCY
100

120

90
TA =+25°C
RL = 10kU
C L = 30pF

TA = +2S"C

~

100

Z

'"

0

OD

§

j

45

e

3
!!l

80

z 30

90

":l

-'
w

z

60

135

~

Z

'"

"'""
0

40

0:

180

0:

f-

-30 1 -......_..1...._'---1.._-'-.................
10
100
lk
10k lOOk
1M
10M
FREQUENCY (Hz)

"'""

~

\

TEST CONDITION.

AMPrFIER

Z

z

20

i i
AND

60

-

A,AV=-1

CONjECTED1IN +1

1\I'J

lAIN

20

AMPLIFIER A{B} OFF, INPUT - 20V PK-PK

AMP~'FlER IB'BI O~' INPU~

=

OV I
I

o
1

10

laO

lk

10k

lOOk

o

1M

1

10M

10

100

lk

10k

lOOk

1M

10M

FREQUENCY (Hz)

ACQUISITION TIME vs EXTERNAL
HOLD CAPACITOR AND
ACQUISITION STEP

ACQUISITION TIME vs INPUT
VOLTAGE STEP SIZE
50.----,-----,----r---,

500.---,--,---,---r--,----.

T~

~
~

40

FREQUENCY (Hz)

DROOP RATE vs TIME AFTER
POWER ON

..........

~

C H = 1000pF

-'
w

A, A)= +1

.........

8,Av=±1

80

= +l'25'C'
CH = 1000pF

V

-"
w

~30f---+-

I

"z
~

~20~--~~~~~--+_--_1

ill

10 ~~~~--+--_+--~

00

9

10

TIME AFTER POWER APPLIED (MINUTES)

20
INPUT STEP (VOLTS)

HOLD CAPACITANCE (pF)

DROOP RATE

vs TEMPERATURE

ACQUISITION OF
SINEWAVE PEAK

ACQUISITION OF STEP INPUT

10000

+10V

"

-lOV

~~:~:~~TURE I

0

0

1
-100

ov

+10V

ov

.........::k':

+10V

OV

JUNCTlON-

-lOV

-lOV

TEMPERATURE~

-50

0

+50

+100

+150

TEMPERATURE (oC)

15-27

1/86, Rev. A

--------------l~ PKO-Oi MONOLITHIC PEAK DETECTOR

TYPICAL PERFORMANCE CHARACTERISTICS
COMPARATOR OUTPUT
RESPONSE TIME
(2kO PULL-UP RESISTOR, TA = +25°C)

COMPARATOR OUTPUT
RESPONSE TIME
(2kO PULL-UP RESISTOR, TA = +25°C)

INPUT LOGIC RANGE vs
SUPPLY VOLTAGE

18

~

;;

5

.§.

~
w

"~
!1

+5
3
-5

2

>-

"I!:
0"

0

~0

4

~

~

3

.,~

>

0

2

>

;;

5

~

w

~

.§.
+.

w

w

"~
0

:;

>-

,

"0

0

"1=

'4

-5

>
>-

~"

~

-2

9

-6

~

TIME (SOns/DIV.)

_55c C :5TA

FOR
:5 +125°C

L'"

w

"a:z
"
"(;

- ~VIN~V+

~

0

~

>-

TIME (SOns/DIV)

'0

........

-'0

r -5SoC

;: +2SOC

~

+126°C'-

'~ .J

-

l"~
~

V-

~~

-'4

-'8

INPUT RANGE OF LOGIC
GROUND vs SUPPLY VOLTAGE

~

•

LOGIC INPUT CURRENT vs
LOGIC INPUT VOLTAGE

'~

'5

12

SUPPLY VOL lAGE +V AND

18

-v (VOL T8)

SUPPLY CURRENT vs
SUPPLY VOLTAGE

'8

~0
~
0
Z

"0a:
"(;u
9
~

0

w

"z

:!

>-

ir

!!

,.

".

.

..-\~

~

"'"

.... 10"

..

-2

-5SoC

~

- ("

Ilo..
~~~

-'0

-,.
-'8

r

'!"S'l

_+2ScC

12

-

-2

~I"'l!

+125°(:

I

I

I

4

-1
LOGIC INPUT VOLTAGE (VOLTS)

SUPPLY VOLTAGE +V AND -V (VOLTS)

_55°C
+2SoC'

LOGIC GROUND'" OV

I

-2

18

... ~f-

I
I

LOGIC 0

'5

"j

+125°C

-3

•

,; I-"

II. ~ 1-"\

1

u

"9

+12 C

-5~OC
I

B-

PIN POTENTIAL IS
BETWEEN SLIDE LINES

+2SoC

,..._

>-

15a:

ACCEPTABLE GROUND

-550C

~E!o.

~

+12SQ C

+2SoC

0

iI'

...1-"1-'
r-r- .......
1-'
r-

LOGIC , _

I;I!"

V+

'0

I

I

~
~

o

12
SUPPLY +V AND

15

18

-v (VOLTS)

HOLD MODE POWER SUPPLY
REJECTION vs FREQUENCY
100
TA = +25°C
VIN = OV
CH = 1000pF

80

"'~"

POSITIVE SUPPLY
(+15V +lV SIN wTj

60

a:

z

0
j:: 40

lrl

~

20
CHANNEL A = 1
CHANNEL Boo 0

0
10

'00

1k

10k

100k

'M

FREQUENCY (Hz)

15-28

1/86, Rev_ A

--------------I~ PKD-01 MONOLITHIC PEAK DETECTOR

TYPICAL PERFORMANCE CHARACTERISTICS
COMPARATOR INPUT BIAS
CURRENT vs DIFFERENTIAL
INPUT VOLTAGE

COMPARATOR OFFSET VOLTAGE
vs TEMPERATURE

COMPARATOR lOS
vs TEMPERATURE
110

+3

1

1

;::

Vs = ±15V
TA = +2SQC

11

";':"'w +2

1

1

INPUT CURRENT

~~S~E:i TL~~~T;~A

'"

1DO

-

1 80

§

g"
::""

'i

;: +1

~

I-- OTHER
INPUT

OTHER
~ INPUT

""
""
:tl

AT -lOV

AT OV

...'"

-1

OTHER

11

"\

80

"'-

70

8'"

~r--+--+--+--+--+--+--+--1

INPUT

1\
"- .........

.........

60

......

AT +10V

"'

I

~1

-15

-10

-5

+5

+10

50
-75 -50

+15

INPUT VOLTAGE (VOLTS)

COMPARATOR 18
vs TEMPERATURE
1200

1000

<:

"- r--...

1
.!E 800

f'...

"

S600
~

8
400

"~"

8

""

~

"~
z

...

.........

::"

"0

200

--

-2

I
+125 O C -

,--src"""

~......a

~ 3

~+2SoC

o

<:

-25

0

25

50

75

100

125

I---- v-

+12S0 C

--

~

_
1

~_

~

,

10

U

~6 :52~n '-.

!\-

<:

...

g

I

"0w
""C;
0

RL = lkf2

TO+5V

15

10

"

05
05
10
INPUT VOLTAGE (mV)

18

1.5

1 " + 1 2 Soc--

~V

0

-5
-50

50

100

150

200

250

300

COMPARATOR RESPONSE
TIME vs TEMPERATURE

+~25°C
w

~

~~ 3r--l--l--\C\-~--+---+--j

_55°C~

DO

>~

J~

04

0.2

,,?
-02

0

IV

J~

TIME (ns)

+25°:~

>
~
>

\

!\

~!

11

~

08

0

INVERTING INPUT = VIN
NON INVERTING INPUT
= OV

I

150

+5

COMPARATOR OUTPUT VOLTAGE
vs OUTPUT CURRENT AND
TEMPERATURE

Q

I

1S

12

Vs = ±15V
TA = +2S C

I

T A "-55

SUPPLY VOLTAGE +V AND -V (VOLTS)

COMPARATOR TRANSFER
CHARACTERISTIC

125

0>
.'i'~

....

4

150

100

~/

-l~
j J

-

+2S"'C

-14

TEMPERATURE (OC)

T A =+2S"C

1

- ,.

1

RESISTiR "2k" /

-5So C

:::O:;~

-0
-10

~&;;Ii!iI'

V+,
~

-18
-75 -50

P~LL-UJ
-

~

~~

0
w

......

~~
~

...0"

75

COMPARATOR RESPONSE
TIME vs TEMPERATURE

~

-

50

OUTPUT SWING OF
COMPARATOR vs
SUPPLY VOLTAGE
14
10

25

TEMPERATURE (OC)

18

~
0

-25

TEMPERATURE fOCI

,P

h

~

~~

§

~V

2

I

w

...""

~"""

+S

~0>

...>S
11

20

10

12

10 - OUTPUT SINK CURRENT (rnA)

15-29

14

"'

-S
-SO

50

100

150

200

250

300

TIME (ns)

1/86, Rev. A

THEORY OF OPERATION
The typical peak detector uses voltage amplifiers and a diode
or an emitter follower to charge the hold capacitor, C H,
unidirectionally (Figure 1). The output impedance of A plus
0, 's dynamic impedance, rd, make up the resistance which
determines the feedback loop pole. The dynamic impedance

by providing the output buffer with an FET input stage. A
current cancellation circuit further reduces droop current
and minimizes the gate current's tendency to double for
every 10· C temperature change.

is rd=K!:.. Id is the capacitor charging current.
qld
The pole moves toward the origin of the S plane as Idgoes to
zero. The pole movement in itself will not significantly
lengthen the acquisition time since the pole is enclosed in the
system feedback loop.
When the moving pole is considered with the typical frequency compensation of voltage amplifiers there is however,
a loop stability problem. The necessary compensation can
increase the required acquisition time. PMI's approach
replaces the input voltage amplifier with a transconductance
amplifier; Figure 2.

Figure 1. Conventional Voltage Amplifier Peak Detector

The PKD-01 transfer function can be reduced to:
VOUT
1 + sC H
gm

Figure 2. Transconductance Amplifier Peak Detector

Where: gm ~ 1p.A/mV, ROUT ~ 20MO.
The diode in series with A's output (Figure 2) has no effect
because it is a resistance in series with a current source. In
addition to simplifying the system compensation, the input
transconductance amplifier output current is switched by
current steering. The steered output is clamped to reduce
and match any charge injection.
Fig.3 shows a simplified schematic of the reset "gm" amplifier, 8. In the track mode, 0, & 04 are ON and 02 & 0 3 are
OFF. A current of 21 passes through 0" I is summed at "8"
and passes through 0" and is summed with gmVIN. The
current sink can absorb only 31, thus, the current paSSing
through 02 can only be: 2K -gmVIN' The net current into the
hold capacitor node then, is gmVIN (C H=21-(21-g mV1N)' The
hold mode, 02 & 03 are ON while 0, & 04 are OFF. The net
current into the top of 0, is-I until D3turns ON. With 0, OFF,
the bottom of O2 is pulled up with a current I until 0 4 turns
ON, thus D, & D2 are reverse biased by ~0.6V and charge
injection is independent of input level.

A

~::::::t---t==:jl===---! }~~~~ROL
A> B '" PEAK DETECT
A B = PEAK DETECT
Aff~-<>VOUT

R,
VIN+

R,

VOUT

VIN

4

25kfl

poo.o,

lo'"

I

1 NULLRANGE::±Vs(';;)

C ::'.0

2MO

":'"

200

POD..,

-=

1. NULL RANGE

C =:'.0

2MO

I

NOTES:

R
1:0

eH'" 1000pF

4
-15V

R.

Vs+

-15V

NOTES'

R,

Vs-

1k0

=±Vs (~)(R1~1R3)

eH = 1000pF

As
1k0

"':'"

":'"

2. DISCONNECT ftc FROM eH AFTER AMPLIFIER A ADJUSTMENT.
3. REPEAT NULL CIRCUIT FOR RESET BUFFER AMPLIFIER B IF REQUIRED.

"':'"

2. DISCONNECT Rc FROM eH AFTER AMPLIFIER A ADJUSTMENT.
3. REPEAT NULL CIRCUIT FOR RESET BUFFER AMPLIFIER B IF REQUIRED.
RA, Ra AND Rc NOT NECESSARY FOR AMPLIFIER B ADJUSTMENT.

Figure B. Vos Null Circuit for Differential Peak Detector

Figure A. Vos Null Circuit for Unity Gain Positive Peak
Detector
Vs-

R,

GAIN=1+~

R,

R,

v,.

25kO

R,

Vs+

Vs-

>.,....-<~<>VOUT

R.

v,.

VOUT

25kO

4

POD·"
NOTES:
1. NULLRANGE:±Vs(';;)

I

POD·O,

C ::'.0

2Mn

R
1:0

CH,:dOOOpF

-:::-

4
-15V

-15V

I

NOTES'
1 NULL RANGE:

-=

±vs(~)

CH = 1000pF

C ::0,"

2MD

-:::-

R
8
1k0

-:::-

2, DISCONNECT Rc FROM CH AFTER AMPLIFIER A ADJUSTMENT
3 REPEAT NULL CIRCUIT FOR RESET BUFFER AMPLIFIER B IF REQUIRED.

2. DISCONNECT Rc FROM CH AFTER AMPLIFIER A ADJUSTMENT.
3. REPEAT NULL CIRCUIT FOR RESET BUFFER AMPLIFIER B IF REQUIRED

Figure D. Vos Null Circuit for Positive Peak Detector With
Gain

Figure C. Vos Null Circuit for Negative Peak Detector
15-31

1/86, Rev. A

I

---------------l\£MD PKD-ii'i MONOLiTHiC PEAK DETECTOR
PEAK HOLD CAPACITOR RECOMMENDATIONS
The hold capacitor (CH) serves as the peak memory element
and compensating capacitor. Stable operation requires a
minimum value of 1000pF. Larger capacitors may be used to
lower droop rate errors, but acquisition time will increase.

With the comparator in the low state (Vall, the output stage
will be required to sink a current approximately equal to
Vc /R 1'

t.V
ZS

Vc

COMPARATOR INPUT

Zero scale error is internally trimmed for CH = 1000pF. Other
CH values will cause a zero scale shift which can be approximated with the following equation.
(mV) = 1 x 103 (pC) -0.6mV
CH(nF)

The peak hold capacitor should have very high insulation
resistance and low dielectric absorption. For temperatures
below 85°C, a polystyrene capacitor is recommended, while
a Teflon capacitor is recommended for high temperature
environments.
CAPACITOR GUARDING AND GROUND LAYOUT
Ground planes are recommended to minimize ground path
resistance. Separate analog and digital grounds should be
used. The two ground systems are tied together only at the
common system ground. This avoids digital currents returning to the system ground through the analog ground path.

INVERTING
COMPARATOR INPUT

DIGITAL

v-

GNO

Figure 1
Table I.
Ve

The CH terminal (pin 4) is a high-impedance point. To
minimize gain errors and maintain the PKD-01's inherently
low droop rate, guarding Pin 4 as shown in Figure 2 is
recommended.

VOH

R1

R2

5

3.5 2.7K 6.2K

5

5.0 2.7K

15

3.5 4.7K 1.5K

15

5.0 4.7K 2.4K

15

7.5 7.5K 7.5K

15

10.0 7.5K 15K

R2~(V:
V

OH

-1

)

PEAK DETECTOR LOGIC CONTROL (RST, DET)
The transconductance amplifier outputs are controlled by
the digital logic signals RST and DET. The PKD-01 operational mode is selected by steering the current (1 1) through
01 and 02, thus providing high-speed switching and a predictable logic threshold. The logic threshold voltage is 1.4
volts when digital ground is at zero volts.
Other threshold voltages (VTH ) may be selected by applying
the formula:
VTH ~ 1.4V + Digital Ground Potential.
Figure 2. CH terminal (Pin 4) guarding. See text.
COMPARATOR
The comparator output high level (VOH ) is set by external
resistors. It's possible to optimize noise immunity while interfacing to all standard logic families - TTL, DTL, and CMOS.
Figure 1 shows the comparator output with external level
setting resistors. Table I gives typical R1 and R2 values for
common circuit conditions.

For proper operation, digital ground must always be at least
3.5V below the positive supply and 2.5V above the negative
supply. The RST or DET signal must always be at least 2.8V
above the negative supply.
Operating the digital ground at other than zero volts does
infl uence the comparator output low voltage. The VOL level is
referenced to digital ground and will follow any changes in
digital ground potential:
VOL ~ 0.2V + Digital Ground Potential.

The maximum comparator high output voltage (VOH ) should
be limited to:
VoH(maximum) < V+ -2.0V

15-32

1/86, Rev. A

_ _ _ _ _ _ _ _ _ _ _ _-\~ PKD-01 MONOLITHIC PEAK DETECTOR

PKD-01 LOGIC CONTROL

BURN-IN CIRCUIT
56kn

L

5%
, - - - - - -....- - - 0

v+

DET

~
PKD-01

2

+1BV

18kO

OR

AST
36kO

1

5%

5%

DIGITAL

-r12
-r-

4

.!!.....
~

~

GROUND

---------

J

r~

CURRENT TO
CONTROL MODES

13

3

-lav

TYPICAL CIRCUIT CONFIGURATIONS
UNITY GAIN POSITIVE PEAK DETECTOR
v-

v+

DET!RST

14

-+10V
PKO-01

INPUT

1I11111111111I

ov

INPUT

OUTPUT

+10V

-ov

OUTPUT

TIME (50.us/DIV)

CH

~1000PF

POSITIVE PEAK DETECTOR WITH GAIN
v-

v+

10kO
1%

INPUT

INPUT

14

OUTPUT

o-vv'Y'-+-+-+-I

(GAIN", +2)

4O.2kO
1%
RESET o-vv'Y'-+---'t-~
VOLTAGE= + 1V
(RESETS TO - 4V)

OUTPUT

TIME (SOMsiOlV)

OSTo--_ _ _-'

CH

,J,1000PF

15-33

1/86, Rev. A

I

--------------I~ i>KD-Ol MONOLITHIC PEAK DETECTOR

NEGATIVE PEAK DETECTOR WITH GAIN
v-

v+

DET/RST

20kn
_+2V

INPUT

1%

-ov

--5V

10kn 1%

o--NV'-+-......=t-1

INPUT
(GAIN = -2)

OUTPUT

>--t-I~-<)

-+10V
30.1kn

1%

-ov

10k01%

--4V

OUTPUT

--lOV

RESET
VOLTAGE:= -1V

TIME i50,us/DIV)

(RESETS TO - 4V)

'*'

RST

CN

1000pF

UNITY GAIN NEGATIVE PEAK DETECTOR
v+

I

I

INPUT

I I

TTl
I

I

I

10kO
1%

-ov
Y,N

v-

14
PKO-Ol

c>-'MMf-r-----'-I-I

>-..--t---<>--o VOUT

--lOV

-+10V

-ov

OUTPUT

TIME {50,us/01Vl
CN

,*1000PF

ALTERNATE GAIN CONFIGURATION

.,

.,

INPUT O-M~!-+--I

PKD-Ol

>-.-1r-~-o OUTPUT

IF BOTH INPUT SIGNAL (AMPLIFIER A INPUT) AND THE RESET
VOLTAGE (AMPLIFIER B INPUT HAVE THE SAME POSITIVE VOLTAGE GAIN THE GAIN CAN BE SET BY A SINGLE VOL rAGE DIVIDER
FOR BOTH INPUT AMPLIFIERS

..

NOTE

Rl. R2, R3 AND R4> 5kn

vo~i::~ O-M~!-+--I

15-34

1/86, Rev. A

--------------1~ PKD-01 MONOLITHIC PEAK DETECTOR

PEAK-TO-PEAK DETECTOR

PKD01

POSITive
PEAK
DETECTOR

10kO

v,"
10kO
PKD.()1

NEGATIVE
PEAK

DETECTOR
10kO

LOGIC SELECTABLE POSITIVE OR NEGATIVE PEAK DETECTOR

POS/f\iE"i'3
PEAK DETECTOR

+15V

U

105

k"

'6 /15

,.

" I,.

'3

y~~7

';7 L..
R

L-~J

I'

2

r~*
3

•

5

6

L~ OUTPUT

.......!

>----"-

SW-02

*511

'y: -y:

9

~

PKD-Ol

,!

7~
R

6,1
C

INPUT

H

I

PEAK DETECTOR
RESET

73
10aOpF
POLYSTYRENE

"::"

-15V

NOTES

1 DEVICE IS RESET TO 0 VOLTS
2 DETECTED PEAKS ARE PRESENTED AS POSITIVE OUTPUT LEVELS
3 R"'10kn

II
15-35

1/86, Rev. A

-----------------I1fM.!.) 1"1<0-01 MONOLiTHiC PEAK DETECTOR
PEAK READING AID CONVERTER

5.0V

PORT 0

2.7k.l1

pPROCESSOR
INPUT

SIGNAL

BIT 10

PORT 1

RST

RESET
VOLTAGE

POSITIVE PEAK DETECTOR WITH SELECTABLE RESET VOLTAGE

PKO..()l

VIN

0-----------+4-1

>4>--"1""---0

VOUT

ov

OV

13

OV

A1

/l2

A3o-----A4O------'
i'KI5E'f/RST o---I~~-----'
+15V

-15V

Loo
":" GND

15-36

NOTES

LOGIC

RESET VOLTAGE = -1 OV
TRACE 1 = 2V/DIV
TRACE 2 = 5V!DIV
TRACE 3 = 2V/DIV

GNO

1/86, Rev. A

-------------l~ PKD-01 MONOLITHIC PEAK DETECTOR

PROGRAMMABLE LOW FREQUENCY RAMP GENERATOR

AMPLITUDE {

SELECTION
LOGIC

>...,-.. . . . . . .

-0

BUFFERED

RAMP
OUTPUT

RAMP
AMPLITUDE

AST
RAMP

START

Jl

0------'

PULSE

AM~~~~DE

+---,--""'\.

RAMP
START
PULSE

NOTES

1 NEGATIVE SLOPE OF RAMP IS seT BV DAC-080UTPUT CURRENT
2. OAC-OSIS DIGITALLY CONTROLLED CURRENT GENERATOR
THE MAXIMUM FULL SCALE CURRENT MUST BE LESS THAN 0 SmA

15-37

1/86, Rev_ A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference. 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15

Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

JPMI)

COMMUNICATIONS
PRODUCTS

Precision Monolithics Inc.

16-3

Introduction

16-4

RPT-82/RPT-83
peM Repeaters

16-2

COMMUNICATIONS
PRODUCTS
PreCISion MOIlolnhlcs Inc

INTRODUCTION
The PMI repeater amplifiers, RPT-82 and RPT83, are monolithic integrated circuits that
perform all of the necessary active functions for
a PCM repeater. RPT-82 and RPT-83 are very
similar; the primary functional difference is the
incorporation of an automatic-clock-shutdown
circuit in the RPT-83. When the signal into the
RPT -83 becomes too low, the clock amplifier is
automatically shut down to prevent the
transmission of erroneous data. Both repeaters
are fully described and specified in this section
of the catalog.

PCM repeater circuits are used to regenerate
alternate-mark-inversion pulses in PCM carrier
systems operating at the 1.444-2.048 Mega-bitsper-second data rate. Information in a PCM
system is transmitted over cable pairs by the
presence or absence of pulses within specific
time slots. A repeater-circuit amplifies degraded
PCM pulses, sets an output flip-flop, and drives
an output transformer that connects to the PCM
cable pair. In addition, repeater amplifiers can
be used for clock-recovery circuits in high-datarate pulse transmission systems.

I
18-3

RPT-82/RPT-83
PCM

REPEATERS
Precision Monolithics Inc.

FEATURES
•
•
•
•

drive if the incoming signal is below the level where accurate
reconstruction is possible. This prevents noise or cross-talk
from appearing as a valid signal that would be retransmitted.

Automatic ALBO Function
Clock·Shutdown Circuit (RPT·S3)
Low·Power Operation (100mW)
Pin Compatible with XR·C277

PIN CONNECTIONS & ORDERING INFORMATION

GENERAL DESCRIPTION

ALBO
INPUT

The RPT-82/83 are integrated circuits that perform the active
functions required for regenerative PCM repeaters. They can
operate from less than 100kHz to greater than 3MHz. In PCM
systems. information is transmitted by the presence or absence
of bipolar pulses in specified time slots. The RPT-82/83 repeaters automatically adjust gain to optimize signal levels. determine if a pulse is present or not. and retransmit the reconstructed pulses.

PREAMP (

"

INPUTS

14

PREAMP (

ALBO FILTER

2
lC TANK
OSCILLATOR
CONTROL

4

OUTPUTS

RPT-82FQ
RPT-83FQ

12 ) PHASE

ANALOG GND

SHIFT
CAP

6

DIGITAL GND

7

DRIVER

B

OUTPUT

DRIVER

9

OUTPUT

16-PIN HERMETIC DUAL·IN·LINE
(O·Sufflx)

The difference between the RPT-82 and the RPT-83 is that the
RPT-83 contains a clock·shutdown circuit. This shutdown
circuit senses the incoming signal level and disables the clock

RPT-83 SIMPLIFIED SCHEMATIC
OSCILLATOR
CONTROL

LC

veel

TANK

DELAY

13

~

I)R35

01

R36

+--+--H:.~'3

1

J~40

,

QZ2AQZ2B

A3l

7~12

~

!r(Q~~~--::R:::46:--~---::R-::47:--,"",,----~--'fl~

I

R42

~7:9
~'7B

r-,?25A

~25B
~ 7[~05Q'4AQ'4B

I

R41

704

-~3

DE,\AY

12

R43

Q49

I I

R44

R45

~

R32

R34

t

R37

Yr-----t- +_-=:--,
Q1~
rtf
~
02~A-"RiV29'V-<1~R~'~7~--::::::ld-

~

1

021

050
Ra7 ~

DRIVE

QFa

OlSA 0288

010

Q30A 0308

QFBI

Q52

9

R63

Q~""'-'R86

. .~

pcp

R
_'-'

FILTER..............
1

-
Cl

0.8

1.1

V

002

015

V

005

50

p.A

~
Po.

30

50

ns

Z

10

60

ns

Output Pulse Rise-Time

Tos

(Note 2)

Output Pulse Fail-Time

Tol

(Note 2)

Output Pu Ise Width

Pw

At f

Pulse-Width Differential

PwD

(Note 2)

Bipolar Violations at Maximum Density

BV, MAX

0

Bipolar Violations with QuaSI-Random
I nput Pattern

BV A MAX

0

=1.544 MHz

U

ns

324
3

12

ns

15

p.A

V)

-

0

~

U

;
0

CLOCK CIRCUIT

=4 9V (Note 1)

Tank Emitter-Follower Base Current

ITS

Tank Input Impedance

Z,NT

Measured from pin 14to Pin 15

OSCillator Bias Current

losc

VPln 14

OSCillator Injection Current

'INJ

Delay CirCUit Resistor

Rd

Ipln 14. Vp,n 14

=3 9V (losc-I TS ) (Note 1)
Set Vp,n 4 - Vp,n 5 =± 1 4V. Vp,n 14 =3 9V
(I'NJ - losel

Measured from pin 11 or pin 12 to pin 15,
TA

=25°C

4

kO

300
10

30

50

p.A

60

160

190

p.A

3.2

40

48

kO

NOTES:
Vpm 2 =2 5V adjust VPIn 3 until Vpm 4 = Vpm 5
2 Sample tested
I

16-5

1/86, Rev. A

U

II

--------------I~ RPT-821RPT-83 PCM REPEATERS

ELECTRICAL CHARACTERISTICS at VCCI = 4.4V, VCC2 = 6.8V, -40°C:5 TA:5 +85°C, unless otherwise noted.
VPln 6 = Vpln 7 = Vpln 13 = GND. (Continued)
PARAMETER

RPT-82!RPT-83
MIN
TYP
MAX

SYMBOL

CONDITIONS

ALBa Threshold

VTA

and 5, required to activate the Peak Detector
TA = 25°C

Clock Th reshold

VTC

UNITS

MISCELLANEOUS
Differential voltage, measured between pinS 4

135

15

165

V

085

10

12

V

065

075

085

V

TA = 25°C

67

73

78

%

TA = 25°C

46

54

58

%

VOl6

Measured at pin 16,
[Vp• - V~51 = ALBa Threshold

10

17

25

V

ALBO OFF Voltage

VF16

Measured at pin 16 and pin 1
TA = 25°C (Note 1)

75

mV

Minimum ALBO Diode ReSistance

RD MIN

8

0

MaXimum ALBO Diode ReSistance

RD MAX

30

kO

48

dB

DlllerenMI voltage, measured between pins 4
and 5, required to activate the Clock Detector
TA = 25°C

Data Th reshold

VTL

Clock Threshold as % of ALBa Voltage

VTC
VTL

Data Threshold as % of ALBa Voltage
ALBa ON Voltage

ALBO Gain Range
NOTES:

DlllerenMI voltage, measured between pinS
and 5, reqUired to activate the Data Detector
TA = 25°C
4

36

(Note 3)

Am

Sample tested
3 Guaranteed by deSign

1 Vptn 2 = 2 5V, adjust Vpm 3 until VPIn 4 = Vpm 5

FUNCTIONAL DESCRIPTION
Bipolar-pulse transmission, the transmission of alternately
positive and negative pulses, is used on repeater lines to
remove the DC component present in unipolar PCM pulse
trains. This also places the principal energy components in the
0-1.544MHz band, as opposed to the 0-3,088MHz band for

unipolar pulse trains. The absence of a DC component in
bipolar pulse trains permits the repeater to be transformercoupled to the repeater line and helps prevent time-shifting of
the regenerator firing levels with variations in input pulse
density (see Figure 1),

ENERGY SPECTRA OF BIPOLAR AND UNIPOLAR PULSE TRAINS

/

PRINCIPAL FREQUENCY

PRINCIPAL FREQUENCY
COMPONENT = 1 544MHz

~
....
-' "-"'-' '- ..'

/

,

"

UNIPOLAR
PCM TRAIN

COMPONENT'" 772kHz

, ..

,

,.,............
' _'""

,
I

:b~~~:IN

AVERAGE DC
- LEVEL

722kHz

Figure 1

1544MHz

772kHz

3088MHz

FREOUENCY

1 544MHz

3088MHz

FREQUENCY

16-6

1/86, Rev. A

--------------I~ RPT-82/RPT-83 PCM REPEATERS

FUNCTIONAL BLOCK DIAGRAM

current to flow through the emitter follower and D1. In the
RPT-83, a low voltage at the ALBO filter enables the clockshutdown circuit when there is no input signal. The clockshutdown circuit turns off the clock amplifier so that neither the
regenerated clock, nor the strobe outputs, are sent to the flipflops. This prevents the RPT-83 from sending noise or crosstalk out as valid-appearing data pulses when the incoming data
level is too low.

The bipolar-PCM pulse train is transformer-coupled into the
preamplifier as shown in the functional block diagram (Figure
2). The secondary of the input transformer is loaded with the
proper terminating resistor, Rr , to match the line impedance.
One side of the transformer secondary is AC-coupled to ground
by capacitor C1; the other side of the secondary winding is in
series with resistance Rs. Resistor Rs and the RC network RA C A
are AC-coupled to the ALBO input by capacitor C2. The impedance of the ALBO (Automatic Line Build-Out) input to ground
is governed by the amount of current through the ALBO diode.
Rs , in series with RAC A, provides signal attenuation proportional to the current flowing through the ALBO diode. When
minimum current flows through the ALBO diode, C2 is effectively isolated from ground and the input signal attentuation is
minimal. The ALBO diode range of 80 to 30kO provides compensation for line losses of approximately 5dB to 41dB.

The clock detector output locks the oscillator to the input
frequency. The following amplifier stages shape the oscillator

Q

~
Po.
cJ)

Z

o.....

~

;
U

THRESHOLDS AND WAVEFORMS

The preamplifier stage amplifies the input signal and applies it
to the three comparators labeled data detector, clock detector,
and peak detector, respectively. Each comparator provides an
output whenever the signal exceeds the trip point on both
positive and negative pulses. Each comparator trips at a different threshold. The data detector is set to trip at the 54% point;
the clock detector trips at the 73% pOint; and the peak detector
trips at peak amplitude. Thresholds and waveforms are shown
in Figure 3.

~
u
::J

o

u
n

n

n n

n

n

...J LJ L-....J L

..J L...J l---J L

n

n

n

DATA DETECTOR
OUTPUT
CLOCK

DETECTOR

OUTPUT
PEAK

DETECTOR

---1L...JL--.JL. OUTPUT

Current pulses from the peak detector are integrated by the
capaCitor in the ALBO filter. This causes a relatively constant

Figure 3

16-7

1/86, Rev. A

---------------l~ RPT-82/RPT-83 PCM REPEATERS

output and shift it in time. The phase-shift capacitor is selected
to provide additional phase-shift so that the strobe pulses will
occur at the center of the incoming pulses. This provides
optimum timing for determining if a "1" or a "0" is present. A
o-to-30pF capacitor (10pF is typical at 1.544MHz) will optimize
the performance of the complete repeater.

The combination of Rs and RA, in parallel with both CA and the
series impedance ofthe ALBO diode, perform the following two
functions: 1) the automatic-gain-control function previously
described, and 2) the frequency/phase compensation for transmission-line losses.
FREQUENCY/PHASE COMPENSATION
Frequency/phase compensation is desirable for three reasons:

The delayed regenerated clock and the data-detector outputs
drive the input flip-flops and output transistors. The output
transistors are coupled to the transmission line through an
output transformer.

1.

If the bandwidth is wider than necessary, noise and crosstalk outside of the signal-frequency band will appear at the
threshold detectors. Out-of-band signals increase the probability that an incorrect logic decision will be made. These
incorrect logic decisions will increase the bit error rate.

2.

Nonlinear phase-shifts in the transmission line may cause
the signal to be distorted to the extent that bit errors occur.
Phase compensation in the repeater can partly correct for
this problem.

3.

Large phase-shifts in the preamplifier at high frequencies
can cause instability if not compensated for by the feedback network. (See Figures 4 and 5).

DETAILED DESCRIPTION
PREAMPLIFIER
The preamplifier performs two basic functions. The first is to
raise the level of the incoming signal to the correct level to trip
the comparators. The second is to provide frequency/gain
compensation to enhance the signal-to-noise ratio of the
incoming signal. The preamp is designed to be operated in a
near open-loop condition. A limited amount of feedback is used
to control the frequency response. The gain-phase relationship
of the preamp (see Figures 4 and 5) implies that the feedback
network must have 40dB attenuation or more at 20MHz and
above to ensure stability.

CLOCK DETECTOR
The clock detector drives the clock-tank circuit with a pulse
each time that the incoming signal is greater than 73% of the
average peak signal.

ALBO
To enable the preamp to operate open-loop with a wide range of
signal levels, the ALBO diode is connected between the
preamp input and ground. Since the ALBO-diode conductance
is directly proportional to the ALBO-diode current, and the
ALBO diode is driven by the peak detector, any signal in excess
of that required to trip the peak detector will be shunted to
ground through the ALBO diode. This automatic-gain-control
function maintains the signal at the optimum level to operate
the clock and data detectors.

PEAK DETECTOR
The peak detector drives the ALBO buffer and ALBO diode at
the peak of the amplified "1" bits. Whenever the preamp ACsignal-output exceeds about 1.5V peak-to-peak, the ALBO
buffer becomes forward biased and drives current into both the
ALBO diode and the ALBO filter. This closed-loop AGC action
maintains the preamp input signal at about 5mVp-p.

PREAMPLIFIER
FREQUENCY RESPONSE
so

r-

so

t-

40

~

1\

l-

t-

I+tIItI--- Iitttfft

20

10

0.1

r""
60

l-

30

o

PREAMPLIFIER
PHASE RESPONSE

~

10

100

h\ItIt- Ii111fft
300
360
0.1

1.

FREQUENCY (MHz)

1\

10
FREQUENCY (MHz)

Figure 4

100

"

Figure 5

16-8

1/86, Rev. A

- - - - - - - - - - - - - - I l m D RPT-82!RPT-83 PCM REPEATERS
RPT-82/83 IN TYPICAL 1.544MHz T1 REPEATER SYSTEM
ZD2

ZD'

R17

VCC2
6.BV

Vee1
4.4V

DELAY
CAPACITOR

T1
INPUT
TRANSFORMER

.----:-:::::---,

'3n
ALBD
SERIES
IMPEDANCE

RPT-821
RPT-83

R7

3."
C6

R11

10DpF

RB

5.10

II

RS

130U
C9

0.0039

5.5kn

NOTE'

Tl' BOURNS PIN 4260-1520

." 1
556n

T2' BOURNS PIN 4285-2007

ll' BOURNSP/N4265-1199

eOUALIZING
NETWORK

pF

c,o

Figure 6

*O"PF

When the ALBO-diode impedance is high. the ALBO series and
shunt impedances have very little effect. so the unattenuated
Signal is applied to the preamp input with only C1 affecting the
frequency response. When the ALBO-diode input impedance is
reduced by higher signal levels. more of the input signal is
shunted to ground through the ALBO shunt impedance.

APPLICATION
In a typical T1. 1.S44MHz repeater system (see Figure 6). the
repeater is placed in series with a twisted-pair transmission line
at distances of up to approximately 6000 feet. The power is
supplied by a constant current of 60mA that is sent commonmode down the transmission line. This constant current is
separated from the signal by input transformer T1 and output
transformer T2. and is converted to voltages VCC1 and VCC2 by
zener diodes ZD1 and ZD2' The signal is coupled into the input
network by T1. One end of T1 is held at AC ground by C2; and
the other end is terminated by the line-matching resistor R1.
The line-matching resistor is followed with a resistive attenuator
consisting of R2 and R3. and the ALBO series impedance R4.
The two resistors. R2 and R3, isolate the changing ALBO-diode
impedance from the transmission line such that the transmission
line is always correctly terminated. Resistor R4, in series with
the shunt ALBO-diode impedance. determines the amount of
attenuation provided at any given ALBO-diode current.
Capacitor C1 provides a shunt path to ground for signals that
are above the signal frequency.

The ALBO shunt impedance. C3 and RS. changes the input
attenuation vs. frequency such that the system has more high
frequency response at low signal levels. and less high frequency response at high signal levels. This change in bandwidth with signal level is intended to partially compensate for
the increased high-frequency losses that occur in long transmission lines.
The bias feedback components between pin 4 and pin 2. consisting of C7. R7. and C2. operate as a DC self-biasing network.
This C-R-C network prevents AC feedback and allows the
preamp to establish a balanced input-and-output DC bias of 2.S
to 2.6 volts. Resistor R11 provides the DC path for biasing
between pins Sand 3.

16-9

1/86, Rev. A

II

---------------l~ RPT-82/RPT-83 PCM REPEATERS

Resistor R11 and capacitor C6 provide an AC feedback path.
Resistors R10 and R11 act as an AC voltage divider that is
shunted by the variable impedance of the resonant circuit comprised of L2 and C9. This frequency-selective feedback path,
between pin 5 and pin 3, increases preamp gain at approximately 900kHz which further improves the system signal-tonoise ratio. The beneficial effect of the frequency-selective
network is shown in Figure 7. The lower trace is a typical input
signal (all 1's in this example) and the upper trace is the preamp output.

Figure 13 is a scope photograph of the signals as observed at
several locations in the system. All traces are DC coupled and
referenced to zero volts at the bottom graticule line. All signals,
except the output, are displayed at 1-volt-per-division. The output is shown at 2-volts-per-division. The signal is all 1's. The
phase relationships are tYPical for this type of repeater.

2.

The clock-tank at pin 14.

Figures 8 and 9 show the appearance of different preamp inputs
measured at pin 5. Figure 8 is typical of an all1's signal pattern
with very little cross-talk or noise. Figure 9 shows a normal
pattern of random 1's and O's.

3.

The phase-shifted clock at pin 11.

4.

The output signal at pin 8.

Due to the automatic-gain-control action of the ALBO circuitry,
the peak amplitude is held constant for line losses of approximately 5dB to greater than 36dB. These signals are superimposed on a DC level of approximately 2.5V.
The preamp output drives the clock detector (reference Figures
2 and 6) which drives the clock-tank circuitry (L 1, C8, and R12).
The signal at pin 14, a sine wave of 0.2 to 1.0Vp-p (depending
upon the percentage of 1-bits), drives the clock amplifier. The
phase-shift capacitor, C11, provides the additional phase shift
so that this integrated and phase-shifted Signal (Figure 10) will
strobe the output flip-flops at the optimum time to determine if a
1-bit is present. If a 1-bit is present, outputs from the data
detector and the strobe cause the flip-flops to drive alternate
output transistors. This signal is coupled through the output
transformer into the next section of transmission line (see Figure 11, all1's; and Figure 12, a random 1-0 pattern).

Figure 7

The signals shown are:
1.

The preamp output at pin 5.

R13, 14,15, and C12 control ringing and overshoot in the output
waveform.
The fault-locating winding with L3 and R16 is used in long-line
systems to determine which repeater, in a large series of repeaters, has become defective.
The RPT-82 and RPT-83 can be used in a variety of systems over
a wide range of frequencies. The low-frequency response is
limited by the difficulty in maintaining useable Q in the clocktank circuit and by transformer-coupling losses. At high frequencies, the major limitation is the output-pulse rise-and-fall
time.
The preamp is a high-gain, wide-bandwidth linear amplifier.
Analog circuits do not have the noise rejection that is common
with most digital circuits. To obtain best performance, certain
precautions should be observed.

Figure 9

Figure 8

IlfJJ.'
I
11
\ •. ,. .1,. '~I
.

,

! i.,'i

•. •.• •. i.I.,,'.\..

I

; .. ---1---:

,I·'

.•

I,~ill~jl
Figure 11

Figure 10

16-10

1/86, Rev. A

----------------I~ RPT-82/RPT-83 PCM REPEATERS

Figure 12

Figure 13

DATA THRESHOLD
The differential voltage measured between PinS 4 and 5 that is
required to activate the data detector.

Circuit layout techniques used for R F. amplifiers should be
followed. Use of double-sided boards with all unused circuitboard area made Into a ground plane is highly recommended
Keep input and output leads as far apart as possible, and signal
runs as short as possible. Locate the attenuator network and the
ALBO series impedance R4 as close to pin 2 as possible

DIFFERENTIAL OUTPUT VOLTAGE
The difference In voltage between the two outputs with a binary
"1" output of either polarity.

Power supply voltages VCC1 and VCC2 should be bypassed near
pins 10 and 15. A bypass capacitor between the VCC2 connection on T2 and pin 7 is also recommended.

EQUALIZING NETWORK
A network which compensates for the amplitude and phase
response of the cable over the operating bandwidth.

REPEATER DEFINITIONS

LINE BUILD-OUT
The attenuation added to the output of a short line to increase
the total attenuation

ALBO DIODE RESISTANCE
The small-signal resistance of the ALBO diode measured
between pins 1 and 6. The ALBO diode is a diode-connected
transistor whose current-resistance relationship is RD = 26/ID
where RD is the ALBO-diode resistance in ohms and ID is the
ALBO-diode current in mA.

MAXIMUM DENSITY
An input signal pattern consisting of all1's
MINIMUM DENSITY
A repeating signal pattern consisting of two 1's followed by
fourteen zeros

ALBO THRESHOLD
The differential voltage measured between pinS 4 and 5 that is
required to activate the internal peak detector

OUTPUT-PULSE RISE (FALL) TIME
Rise (Fall) time of regenerated pulse Measured from the 10% to
90% points.

AMI
Alternate Mark Inversion. A form of digital signal transmission
where each successive 1-bit is of opposite polarity.

OUTPUT-PULSE-WIDTH DIFFERENTIAL
In a T1 carrier system, a typical pulse width is 324nsec The
pulse-width differential is the difference in pulse width of two
successive outputs.

AUTOMATIC LINE BUILD-OUT
An automatic-gain-control circuit which operates by simulating
a line "build-out" or extension.

PREAMPLIFIER BANDWIDTH
3dB bandwidth of the preamplifier circuit.

BIPOLAR VIOLATION
The transmission of two consecutive pulses of the same polarity.

QRSS
QRSS is QuaSI-Random Signal Source: a signal consisting of
random 1's and O's.

CLOCK THRESHOLD
The differential voltage measu red between pinS 4 and 5 that is
required to activate the clock detector.

II
16-11

1/86, Rev. A

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrumentation Amplifiers 6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and-Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

PACKAGE

INFORMATION
Precision Monolithics Inc.

PMI Letter
DeSignator

38510
Applicable
Configuration

Package
Description

Package Dimensions H

J
K

Z
Y

a
X
R
V
T

8-Lead
Ceramic
14-Lead
Ceramic
16-Lead
Ceramic
18-Lead
Ceramic
20-Lead
Ceramic
24-Lead
Ceramic
28-Lead
Ceramic

oB'
XB'
RB'
VB'
TB'

A1

17-3

A2

17-3

M'
MB'

04-1

17-4

F'
FB'

DH

17-4

02-1

17-5

06-1

17-5

Ceramic DIPs

DIP

Package Dimensions -

DIP

RC'
08-1

17-6

03-1

17-7

DIP

F4-1
F4-2

17-19
17-19

FH
F1-2

17-19
17-19

F5-1
F5-2

17-20
17-20

F8-1
F8-2

17-20
17-20

Leadless Chip Carriers
C-2

17-21

C-4

17-22

'Special Order Only.

17-8
DIP

14-Lead
Side-Brazed
16-Lead
Side-Brazed
18-Lead
Side-Brazed
20-Lead
Side-Brazed
24-Lead
Side-Brazed
28-Lead
Side-Brazed

Dimensioning Symbols

Side-Brazed DIPs
01-3

17-9

02-3

17-9

06-3

17-10

08-3

17-10

03-3

17-11

The symbols to be used for dimensioning case outlines will
be as listed below. To deSignate the dimension as a diameter.
the lower-case Greek letter cP(phi) will be added in front of the
dimension symbol.

DIP
DIP

A
cPb
b
c

- Body dimensions.
- Terminal lead diameters.
- Terminal lead widths.
- Terminal lead thicknesses.
cPO - Body diameters.
o - Body lengths.
E - Body widths.
e - Terminal lead spacings.
F - Flange dimensions.
k - Index dimensions. length.
L - Terminal lead lengths.
Q - Standoff height. The height from the seating
plane or a reference plane parallel to the
seating plane.
R - Radius dimensions.
S - Distance between terminal leads and the
body end.
O! - Angular dimensions.

DIP
DIP
DIP

17-12
DIP

Epoxy DIPs

8-Lead Epoxy DIP
14-Lead Epoxy DIP
16-Lead Epoxy DIP
18-Lead Epoxy DIP
20-Lead Epoxy DIP
24-Lead Epoxy DIP

17-13
17-13
17-14
17-14
17-15
17-16

Package Dimensions - Cerpacks
L
M
F
N

20-Position
Chip Carrier
28-Position
Chip Carrier

TC'

DIP

Flatpacks

10-Lead Flatpack
10-Lead Flatpack.
Bottom-Brazed
14-Lead Flatpack
14-Lead Flatpack.
Bottom-Brazed
16-Lead Flatpack
16-Lead Flatpack.
Bottom-Brazed
24-Lead Flatpack
24-Lead Flatpack.
Bottom-Brazed

N'
NB'

DIP

Package Dimensions P
P
P
P
P
P

L'
LB'

38510
Applicable
Configuration Page

Package
Description

Package Dimensions -

17-3

DIP

Package Dimensions YB'

Metal Cans

6-Lead TO-78
Metal Can
8-Lead TO-99
Metal Can
10-Lead TO-100
Metal Can

Package Dimensions -

PMI Letter
Designator

Page

10-Lead Cerpack
14-Lead Cerpack
16-Lead Cerpack
24-Lead Cerpack

17-17
17-17
17-18
17-18

Standard lead finish is matte tin/lead. Other lead finishes per
M I L-M-38510 are available on special orders. Standard 883
product meets lead finish requirements per M I L-M-38510.

17-2

PACKAGE
INFORMATION
PreCISIon

Monolithics Inc.

PACKAGE DIMENSIONS - METAL CANS
6 & 8-Lead Can Dimensions

6-Lead TO-78 Metal Can (H-Sufflx)

INCHES

MILLIMETERS

SYMBOL

MIN

MAX

MIN

MAX

A

0.165

0.185

4.19

4.70

1/>0

0.016

0.019

0.41

0.48

0.016

0.021

0.41

0.53

0.335

0.370

8.51

9.40

0.305

0.335

7.75

8.51

0.110

0.160

2.79

4.06

e

0200 BSC

5.08 BSC

e,

0.100 BSC

2.54

L

0.027

0.034

0.69

0.027

0.045

0.69

1.14

0.500

0.750

12.70

19.05

0.250

..

4Reference Plane

n=t:~l4

0.86

Q

0.010

a

45°

",,"·ve".e_l-=_ ''<

1.27

0.045

asc

1.14

0.25
45°

asc

3

INCHES
MAX

MIN

MAX

A

0.165

0.185

4.19

4.70

0.016

0.D19

0.41

0.48

0.016

0.021

0.41

0.53

1/>0

0.335

0.370

8.51

9.40

1/>0,

0.305

0.335

7.75

8.51

0.110

0.160

2.79

4.06

0.230
0.115

10-Lead TO-100 Metal Can (K-Sufflx)

MILLIMETERB

MIN

e

asc
asc

5.84

asc

F

3
1.02

0.040

Reference Plane

0.027

0.034

0.69

0.027

0.045

0.66

1.14

0.500

0.750

12.70

19.05

"

~

L

0.050

el

V-~~

0.250
Q

a

0.010
36° BSC

NOTES

3

292 BSC

r~anae Se.llng Plane

IT fl
"'0"'01
lL +
_Q

6.35

SYMBOL

.D,

tj

2

10-Lead Can Dimensions

rtf ~slO-··,
rtf
A

3
1.02

0.050

8-Lead TO-99 Metal Can (J-Sufflx)

3

asc

0040

F

NOTES

0.66

2

1.27
6.35

0.045

0.25
36° BSC

1.14

3

NOTES:
1. (All leads) I/>b applies between L, and L2.l/>b, applies between L2
and 0.500 (12.70 mm) from the reference plane. Diameter is
uncontrolled In L, and beyond 0.500 (12.70 mm) from the
reference plane
2. Measured from the maximum diameter of the product
3. Leads haVing a maximum dlameterO.019 (0.48 mm) measured In
gaging plane 0.054 (1.37 mm) + 0.001 (0.03 mm) - 0.000
(0.00 mm) below the base plane of the product is Within 0.007
(0.18 mm) olthei r true position relative to a maxi mum width tab.

Icr=

~"'bl

17-3

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS -

CERAMIC DIPS

8-Lead Ceramic Dip

14-Lead Ceramic Dip

1r
IS
see_iSJs
SI

Nole 1

-1

.

t

4

r-E
--1

r-D~

e ~bl Sf c

----7,
TQ~
tf
----1.
A~

Seating
Plane

b-:1

W

INCHES
SYMBOL

MIN

A

MAX

El---'

---Ia

I

see--\\-Note 7

MAX

b

0.014

0.023

0.36

0.58

0.030

0070

0.76

1.78

0.008

0015

0.20

0.38

0.310

E1

0.290

0320

e

0.100 S5e

L

0.125

L1

0150

Q

0.015

5
0.005

ex

0°

A

b1

0.023

036

0.070

0.76

1.78

0008

0.015

0.20

0.38

0.785

D

E

0.220

0.310

737

813

E1

0290

0.320

3.18

0.060

038

0.100 S5e
0125

508

381

L1

0150

1.52

3

Q

0.015

1.35

6

5

6

51

0005

ex

0°

0.13
0°

15°

NOTES:
1. Index area; a notch or a lead one identification mark is located

NOTES

5.08

0.014

4

0.200

MAX

0030

4

5

MILLIMETERS
MIN

0.200

7.87

2.54 S5e

15°

MAX

10.29

0.055

51

MIN

5.59

0.405
0.220

SYMBOL

NOTES

5.08

b1

E

INCHES

MILLIMETERS
MIN

0.200

D

I

058

19.94

4

5.59

7.87

4

7.37

8.13

2.54 S5e
0200

3.18

0.060

038

5
5.08

3.81

0.098

1.52

3

2.49

6

0.13
15°

0°

6
15°

4. This dimension allows for off-center lid, meniscus and glass
overrun
5. The basic lead spacing is 0100 (2.54 mm) between centerlines.
6. Applies to all four corners.
7. Lead centerwhena ISOo . E1 shall be measured at thecenterline
of the leads

adjacent to lead one and IS within the shaded area shown

2. Themlnimum IimltfordlmenSlon b 1 may be 0.023 (0.58 mm) for
all four corner leads only

3. Dimension a shall be measured from the seating plane to the
base plane.

17-4

PACKAGE
INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS -

CERAMIC DIPS

16-Lead Ceramic Dip
(Q-Suffix)

INCHES
SYMBOL

MIN

MAX

MILLIMETERS
MIN

MAX

0.200

A

18-Lead Ceramic Dip
(X-Suffix)

SYMBOL

5.08

b

0.014

0.023

0.36

058

b,

0.030

0.070

0.76

1.78

c

0.008

0.015

0.20

0.38

0.840

D

INCHES
NOTES

MAX

MAX

0.200

A

2

MILLIMETERS
MIN

NOTES

5.08

b

0.014

0.023

0.36

b,

0.030

0.070

0.76

1.78

0008

0.015

0.20

0.38

0.58

24.38

4

E

0.220

0.310

5.59

7.87

E

0.220

0.310

5.59

7.87

4

E,

0.290

0.320

7.37

8.13

E,

0.290

0.320

7.37

8.13

e

0.100

e

0.100

L

0.125

L

0.125

L,

0.150

L,

0.150

Q

0.015

0.Q15

S

Bse

21.34

MIN

2.54
0.200

3.18

0.060

0.38

0.005

0<

0"

5
5.08

1.52

3

Q

2.03

6

S

6

s,

0.005

0<

0"

0.13
15"

0"

0.960

D

3.81

0.080

s,

Bse

4

15"

Bse

2.54
0.200

3.18

Bse
5.08

3.81
0.060

0.38

0.098

1.52

3

2.49

6

0.13
15"

0"

6
15"

NOTES:

4

1. Index area; a notch or a lead one identification mark is located
adjacent to lead one and IS within the shaded area shown.
2. The minimum limit lor dimension b, may be 0.023 (0.58 mm) lor
alilour corner leads only.
3. Dimension Q shali be measured Irom the seating plane to the

This dimension allows for off·center lid, meniscus and glass
overrun.

5. The basic lead spacing is 0100 (2.54 mm) between centerlines.
6. Applies to all four corners
7. Lead center when a isO°. E, shall be measured at thecenterline
01 the leads.

base plane.

17·5

I

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - CERAMIC DfPS

20-Lead Ceramic Dip
(R-Sufflx)

INCHES
SYMBOL

MIN

MAX

A

MAX

0.200
0.014

0.023

0.36

b,

0.030

0.070

0.76

1.78

c

0.008

0.015

0.20

0.38

0.58

1.060

4

0.220

0.310

5.59

7.87

0.290

0.320

7.37

8.13

0.100
0.125

L,

0.150

Q

0.015

5

sse

2.54
0.200

3.18

0.005

a

0'

5. The basic lead spacing is 0.100 (2.54 mm) between centerlines.
6. Applies to all four corners.
7. Lead center when a IsO'. E, shall be measured atthecenterline
of the leads.

sse
5.08

3.81
0.060

0.38

1.52
2.03

0.080

5,

overrun.
4

E

L

2

26.92

E,
e

NOTES

5.08

b

0

NOTES:
1. Index area; a notch or a lead one identification mark is located
adjacent to lead one and is within the shaded area shown.
2. The minimum IImitfordlmension b, may be 0.023 (0.58 mm)for
all four corner leads only.
3. Dimension Q shall be measured from the seating plane to the
base plane.
4. This dimension allows for off-center lid, meniscus and glass

MILLIMETERS
MIN

0.13
15'

0'

6
15'

17-6

------ ------

PACKAGE
INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS -

CERAMIC DIPS

24-Lead Ceramic Dip
(V-Suffix)

1 r-S1

-I I-s

hnnnnnnnnnnr'11

~

13

See

Not. 1'.
I

L!:::!JL

I'

12
LJ LJ LJ LJ LJ

LJ LJ LJ LJ LJI

D

.

j-E---j

Ai...
to~~'
~ --------I +
E,--

.~f.-

INCHES

• f.-

j~" ....,,---I.E
Plane

MIN

MAX

MIN

MAX

A

-

0.225

-

5.72

0.014

0.023

0.38

0.58

b,

0030

0070

076

178

2

c

0.008

0.015

0.20

0.38

-

b

NOTES

-

1.290

-

32.77

4

E

0.500

0.610

12.70

15.49

4

E,

0.590

0.620

14.99

15.75

7

e

0.100

L

0.120

0.200

3.05

5.08

L,

0.150

-

-

3.81

-

-

Q

0.015

0.075

0.38

1.91

3

0

5
5,

a

-

-

esc

2.54

esc

5

-

2.49

6

0.005

-

0.13

-

6

O'

15'

O'

15'

-

0.098

-I\--'

NOTES:
1. Index area; a notch or a lead one identification mark Is located
adjacent to lead one and is within the shaded area shown.
2. The minimum limitfordimension b, may be 0.023 (0.58 mm) for
all four corner leads only.
3. Dimension Q shall be measured from the seating plane to the
base plane.
4. This dimension allows for off-center lid, meniscus and glass
overrun.
5. The baSic lead spacing Is 0.100 (2.54 mm) between centerlines.
6. Applies to all four corners.
7. Lead center when a Is 0'. E, shall be measured althe centerline
of the leads.

MILLIMETERS

SYMBOL

Note 7

Z

0.....

~2
Z
.....
~

~
~
~

17-7

-

- ---- --------

PACKAGE

INFORMATION
PrecIsion Monolithics Inc.

PACKAGE DIMENSIONS -

CERAMIC DIPS

28-Lead Ceramic Dip
(T-Sufflx)

-I r-

Sl

n n n n n n n n n n n n n r+j,

se~-....

is

15

l\

Note 1

t

J::!.jLJLJLJLJLJLJO 0

I'

DO D

14

o

LJ.I

r-E~

D

TQ~~
tc
----- --I ~b1---r3: L"---1
b_ll_

INCHES
SYMBOL

Seeling
Plane

-leI--

MIN

MAX

NOTES

MAX

A

-

0.225

-

5.72

b

0.014

0.023

0.36

0.58

b,

0.030

0.070

0.76

1.78

2

c

0.006

0.D15

0.20

0.38

D

-

1.490

-

-

37.85

4

-

-

E

0.500

0.610

12.70

15.49

4

E,

0.590

0.620

14.99

15.75

7

e

0.100 BSC

L

0.120

0.200

3.05

5.08

-

L,

0.150

3.81

-

-

Q

0.015

0.075

0.38

1.91

3

S

-

0.098

-

2.49

8

S,

0005

-

6

CO<

O·

15·

-

2.54 BSC

-

15·

0.13

O·

See
Note 7

-\\-e

NOTES:
1. Index area; a notch or a lead one Identification mark Is located
adJacent to lead one and Is within the shaded area shown.
2. The minimum IImltfordlmenslon b, maybeO.023 (0.58 mm) for
all four corner leads only.
3. Dimension Q shall be measured from the seating plane to the
base plane.
4. ThIS dImenSIon allows lor oil-center lId, meniscus and glass

MILLIMETERS

MIN

----{a

overrun.
5. The basic lead spacing IsO.loo (2.54 mm) between centerllnes.
6. Applies to all four corners.
7. Lead center when CO< IsO·. E, shall be measured allhecenterline
of the leads.

5

17-8

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - SIDE-BRAZED DIPS

14-Lead Side-Brazed Dip
(VB-Suffix)

INCHES
SYMBOL

MIN

A

MAX

16-Lead Side-Brazed Dip
(OB-Sufflx)

MILLIMETERS
MIN

MAX

0.200

INCHES
SYMBOL

NOTES

5.08

MIN

A

MAX

MILLIMETERS
MIN

0.200

MAX

b

0.014

0.023

0.36

0.58

b

0.014

0.023

0.36

b,

0.030

0.070

0.76

1.78

b,

0.030

0.070

0.76

1.78

0.008

0.D15

0.20

0.38

0.008

0.015

0.20

0.38

0

0.785

E

0.220

0.310

E,

0.290

0.320

0.100 BSC
0.125

L,

0.150

a

0.D15

S

5.59
7.37
2.54

0.200

3.18

0005

a

0'

4

0

7.87

4

E

0.220

0.310

E,

0.290

0.320

e

0100 BSC

8.13

esc
5.08

3.81
0.060

0.38

0.098

S,

19.94

1.52
2.49

0.13
15'

0'

0.840

L

0.125

L,

0.150

a

0015

S

6

S,

0.005

a

0'

15'

NOTES:
Index area; a notch or a lead one identification mark is located
adjacent to lead one and is within the shaded area shown.
2. The minimum limit for dimension b, may be 0.023 (0 58 mm) for
all four corner leads only.
shall be measured from the seating plane to the
3. Dimension
base plane.

4

0.58

21.34

4

5.59

7.87

4

7.37

8.13

254 BSC
0.200

3.18

5.08

3.81
0.060

0.38

0.080

6

NOTES

5.08

1.52
2.03

0.13
15'

0'

3
6
6

15'

This dimension allows for off-center lid, meniscus and glass
overrun
5 The basIc lead spacing isO.100 (2 54 mm) between centerlmes.
6. Applies to all four corners
7. Lead center when Q' isO° E1 shall be measured atthecenterline
of the leads

a

17-9

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - SIDE-BRAZED DIPS

18-Lead Side-Brazed Dip
(XB-Sufflx)

-lr

-I

Sl

r-

20-Lead Side-Brazed Dip
(RB-Sufflx)

1r

s

A
:If
rr----3 r .~~
I·

0

I-E=1

·1

'1~~"
=r-

OJ

A

c

L

0

s

"'jJj IOU; 11 iU

j~

~ JUrn J1=r=;'

~t~-r-~ Q-~~

b---l~--I

1-,,-1

INCHES

r-

NO:~iS~O~]

N~~{)D]
Q

-1

Sl

~e

_

INCHES

MILLIMETERS

bl

Sealing
Plane

--f.1
t=E=:j
c

I--E--I
1

MILLIMETERS

SYMBOL

MIN

MAX

MIN

MAX

NOTES

SYMBOL

MIN

MAX

MIN

MAX

A

-

0.200

-

5.08

-

A

-

0.200

-

5.08

-

0.58

-

b

0.014

0.023

0.36

0.58

-

b

NOTES

b,

0.014

0.023

0.36

0.030

0.070

0.76

1.78

2

b,

0.030

0.070

0.76

1.78

2

c

0.008

0.015

0.20

0.38

-

c

0.008

0.015

0.20

0.38

4

0.960

-

24.38

4

D

1.060

-

26.92

E

0.220

0.310

5.59

7.87

4

E

0.220

0.310

5.59

7.87

4

E,

0.290

0.320

7.37

8.13

7

E,

0.290

0.320

7.37

8.13

7

e

0.100

5

e

0.100

L

0.125

5.08

-

L

0.125

5.08

L,

0.150

-

0

0.Q15

D

S

-

-

S,

0.005

ex

0'

sse

2.54
0.200

-

3.18

sse

-

3.81

-

-

L,

0.150

0.060

0.38

1.5~

3

0

0.015

0.098

-

2.49

6

S

0.13

-

6

S,

0.005

15'

-

ex

0'

15'

0'

NOTES:
1. Index area; a notch or a lead one identification mark is located
adjacent to lead one and is within the shaded area shown.
2. The minimum limit for dimension b , may be 0.023 (0.58 mm) for
all four corner leads only.
3. Dimension 0 shall be measured from the seating plane to the
base plane.

-

sse

2.54
0.200

-

3.18

sse

5

3.81

-

0.060

0.38

1.52

3

0.060

-

2.03

6

0.13

-

6

15'

-

15'

0'

4. This dimension allows for off-center lid, meniscus and glass

overrun.
5. The basic lead spacing is 0.100 (2.54 mm) between centerlines.
6. Applies to all four corners.
7. Lead center when ex is 0'. E, shall be measured atthecenterline
of the leads.

17·10

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - SIDE-BRAZED DIPS

24-Lead Side-Brazed Dip
(VB-Suffix)

INCHES
SYMBOL

MIN

MAX

b

0.014

bl

0.030

c

0.008

A

MAX

0.023

0.36

0.58

0.070

0.76

1.78

O.ot5

020

0.225

D

0.38

4

0.500

0.610

12.70

1549

0.590

0620

14.99

1575

0200

3.05

0.075

0.38

0120

Ll

0.150

a

0.015

S

esc

2.54

0.005

0<

0"

esc

5. The basIc lead spacing Is 0.100 (2.54 mm) batwean centerllnes.
6. Applies to all four corners.
Lead center when 0< IsO·. El shall be measuredatthecenterllne
of the leads.

5
5.08

3.81

0.098

SI

overrun.
4

E

0.100

a

32.77

El
e

NOTES

5.72

1.290

L

NOTES:
1. Index area; a notch or a lead one identification mark Is located
adjacent to lead one and Is within the shaded area shown.
2. The minimum lImit fordlmension bl may be 0.023 (0.58 mm) for
all four corner leads only.
3. DimenSIon shall be measured from the seating plane to the
base plane.
4. ThIS dImension allows for off-center lid, meniscus and glass

MILLIMETERS
MIN

1.91

3
6

2.49
0.13

15·

O·

6
15·

I
17-11

~

-

~~----~----

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - SIDE-BRAZED DIPS

28-Lead Side-Brazed Dip
(TB-Suffix)

Ii

I

sl

1"~:', '

D

)""<,,

See
Note

,;;~"

t.:::::J

oJ

is
15

14

'I

0

t=E=:=j

-t=.
I
~'
:::;~~~"l'

'

~El~

INCHES
SYMBOL

MIN

MIN

MAX

NOTES

0,225

-

5.72

-

b

0,014

0,023

0,36

0,58

-

b1

0,030

0,070

0,76

1.78

2

c

0,008

0.D15

0,20

0,38

-

-

37.85

4

A

D

-

-

1,490

E

0,500

0,610

12,70

15.49

4

E1

0,590

0,620

14,99

15,75

7

e

0,100 BSe

L

0,120

0,200

305

508

-

L1

0.150

Q

0.015

S

-

Sl

0.005

a

0°

NOTES:
1, Indexarea; a notch is ora lead one identification mark Is located
adjacent to lead one and is within the shaded area shown,
2, The minimum limitlor dimension b1 may be 0,023 (0,58 mm) for
all four corner leads only.
3, Dimension Q shall be measured from the seating plane to the
base plane,
4. This dimension allows for off-center lid, meniscus and glass
overrun.
5, The basic lead spacing isO,100 (2.54 mm) between centerlines,
6, Applies to all four corners,
7, Lead center when a isOo , E1 shall be measured at the centerline
of the leads,

MILLIMETERS

MAX

2,54 BSe

-

5

3.81

-

-

0.075

0.38

1.91

3

0.098

-

2.49

6

0.13

-

6

15°

-

15°

0°

17-12

PACKAGE
INFORMATION
Prt'CJSiOll Monolithics Inc

PACKAGE DIMENSIONS -

EPOXY DIPS

a-Lead Epoxy Dip
(P-Sufflx)

14-Lead Epoxy Dip
(P-Sufflx)

f4---D-----j

INCHES

MILLIMETERS

SYMBOL

MIN

MAX

MIN

MAX

A

0.148

0.152

3.76

386

INCHES
NOTES

SYMBOL

MAX

MIN

A

MILLIMETERS
MIN

MAX

0.200

b

0.016

0.020

0.406

0.508

b

0.014

0.023

0.36

b1

0.058

0.062

1.47

1.58

b1

0.030

0.070

0.76

1.78

c

0.008

0.012

0.203

0.304

c

0.008

0.015

0.20

0.38

D

0.370

0.382

947

9.70

D

0748

0.754

18.99

19.1

I:

0.246

0.254

6.25

6.45

I:

0.220

0.310

5.59

7.87

1:1

0.298

1:1

0.290

0.320

7.37

e

0100

0.302

esc

7.57
2.54

7.67

2

esc

0.100

esc

2.54

0.58

8.13

L

0128

0132

3.25

3.35

L

0125

0.146

0.152

3.76

3.86

L1

0.150

a

0.020 TYP

a

0.015

0.060

0.38

1.52

15·

Q

O·

15·

O·

15·

Q

o·

0.508 TYP
15·

O·

NOTES:

318

2

esc

L1

0200

NOTES

5.08

5.08

3.81

NOTES:
Minor changes In dimensions may occur without advance

1. Minor changes in dimensions may occur without advance
notice.

notice.

2. DimenSion "1:1' to center of leads when formed parallel.

2. Dimension "1:1' to center of leads when formed parallel.

Z
0

~

0~
Z

~

~
~
~

II
17-13

PACKAGE

INFORMATION
PrecisioJl Monolithics Inc.

PACKAGE DIMENSIONS - EPOXY DIPS

16-Lead Epoxy Dip
(P-Sufflx)

18-Lead Epoxy Dip
(P-Suffix)

A
L

INCHES
SYMBOL

MIN

MAX

A

MILLIMETERS
MIN

MAX

0.200

INCHES
NOTES

SYMBOL

MIN

MAX

A

5.08

MILLIMETERS
MIN

MAX

0.200

b

0.014

0.023

0.38

0.58

b

0.014

0.023

0.38

b1

0.030

0.070

0.76

178

b1

0.030

0.070

0.76

1.78

c

0.008

0,015

0.20

0.38

0.008

0.015

0.20

0.38

D

0.748

0754

18.99

19.1

D

0898

0.904

2281

22.91

E

0.220

0.310

5.59

7.87

E

0.220

0.310

5.59

7.87

E1

0.290

0.320

7.37

8.13

E1

0.290

0.320

7.37

8.13

e

0.100

e

0100

esc

254

L

0.125

L1

0.150

0.200

3.18

Q

0.015

0.060

0.38

a

O·

15·

O·

2

esc

esc

2.54

0.58

L

0.125
0.150

1.52

Q

0.015

0.060

0.38

1.52

15·

a

O·

15·

O·

15·

NOTES:
1. Minor changes .n dimensions may occur without advance
notice.
2. Dimension "E( to center of leads when formed parallel.

0.200

318

2

esc

L1

5.05

3.81

NOTES

5.08

506

3.81

NOTES:
1. Minor changes in dimensions may occur without advance
notice.

2. Dimension "E( to center of leads when formed parallel.

17-14

PACKAGE
INFORMATION
Precision MODolithics Inc.

PACKAGE DIMENSIONS -

EPOXY DIPS

20-Lead Epoxy Dip
(P-Suffix)

I: :::::::::II
INCHES
SYMBOL

MIN

MAX

MAX

b

0.014

0.023

0.36

0.58

b1

0.030

0.070

0.76

1.78

0.008

0.Q15

0.20

0.38
26.24

0.255

A

1.029

1.035

26.14

E

0.220

0.310

5.59

7.87

E1

0.290

0.320

7.37

8.13

e

0.100

2.54
0.200

1. Minor changes in dimensions may occur without advance
notice.
2. Dimension "E1" to center of leads when formed parallel.

NOTES

5.08

0

esc

NOTES:

MILLIMETERS
MIN

318

z

o

esc

L

0.125

L1

0.150

Q

0.015

0.060

0.38

1.52

a

O·

15·

O·

15·

~2

508

3.81

....z

17-15

PACKAGE
INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS -

EPOXY DIPS

24-Lead Epoxy Dip
(P-Sufflx)
24

13

1

~~~l
I'
0
'I

Ir=E1~

:±;~ti'-1~
INCHES
SYMBOL

MAX

MIN

A

MAX

0.225

b

0.014

0.023

0.36

0.030

0.070

0.76

1.78

0.008

0015

0.20

0.38

1.248

1254

31.7

31.8

E

0.538

0.542

13.67

13.77

E,

0.598

0.602

15.19

e

0.100

2.54
0.200

notice.

3.18

2. Dimension "E{ to center of leads when formed parallel.

0.58

0

0.125

1. Minor changes in dimensions may occur without advance

NOTES

5.08

b,

sse

NOTES:

MILLIMETERS
MIN

15.29

sse
5.08

L,

0.150

Q

O.Q1S

0.060

0.38

1.52

C/

0°

15°

0°

15°

3.81

17·16

PACKAGE
INFORMATION
Precision M01lo1ithics Inc

PACKAGE DIMENSIONS - CERPACKS

10-Lead Cerpack
(L-Sufflx)

k

14-Lead Cerpack
(M-Suffix)

S••

IT=~!~N°~·~E5f
.~~
~, ~'~601· ~~+T
~t!
Ll:
ff
AQ

I

.:rjt
~Ll:~~~~~
~F==L====;.14cI
11
~E

~c

I

I-E--I_L~l

----1·11---·-

AQ

INCHES
MIN

MAX

MIN

MAX

A

0.030

0.085

076

2.16

0.010

0,019

0.25

048

0.003

0.006

0.08

0.15

0.290

D

0.240

0.008

7.37

0.260

6.10

0.015

0.20

0.38

0.050 SSC

INCHES

MILLIMETERS

SYMBOL

E

L

1.27

NOTES

3

esc

MIN

MAX

MIN

MAX

A

0030

0085

0.76

2.16

0.010

0.019

025

048

0.003

0006

0.08

015

0.260

6.10

660

0,015

0.20

0.38

0.280

D

6.60

MILLIMETERS

SYMBOL

E

0.240
0050

4

0.008

esc

7.11

1.27

ssc

L

0.250

0.370

6.35

9.40

L

0.250

0.370

6.35

9.40

0.010

0.040

0.25

1.02

Q

0.010

0.040

0.25

1.02

1.14

S,

0005

0.13

S,

0.004

0.10

0.045

S,

0.005

0.13

5,6

NOTES:
Index Brea; a notch or a lead one identification mark IS located
adjacent to lead one and shall be located within the shaded area
shown. Alternatively, a tab (dim. k) may be used to Identify lead
one. This tab may be located on either side as shown.
2. Dimension Q shall be measured at the point of eXit of the lead
from the body.

1.

3

4

Q

S

NOTES

8

5,6

6. Dimension S, (See 40.3) may be 0.000 (0.00 mm) if corner leads

7.

bend toward the cavity of the package Within one lead's width
from the pOint of entry of the lead Into body.
Optional configuration If thiS configuration is used, no organiC
or polymeric matenals are molded to the bottom of the package
to cover the leads.

3. ThIS dimension allows for off-center lid, meniscus and glass

OptIonal, see note 1 If a lead one identification mark is used in
addition to this tab, the minimum limit of dimension k does not

overrun

apply.

4. The basic lead spacing IS 0.050 (1.27 mm) between centerlines.
5. Applies to a/l four corners

II
17-17

PACKAGE

INFORMATION
Precision MOllolithics Inc

PACKAGE DIMENSIONS - CERPACKS

24-Lead Cerpack
(N-Sufflx)

16-Lead Cerpack
(F-Suffix)

See Note
S2 L

b

~~
I

s[f

INCHES
MAX

MIN

MAX

~E-1-L-11

A

0.045

0.085

1.14

2.16

0.015

0.019

0.38

0.48

0.003

0.006

0.08

0.15

D
E

0.440
0.245
0.050
0.008

11.18

0.285

6.22

0.015

0.20

0.38

sse

1.27

0.250

0.370

6.35

9.40

0.010

0.040

0.23

1.02

S,

0.045
0.005

1.14
0.13

MIN

MAX

MIN

MAX

A

0.045

0.090

1.14

2.29

0.015

0.019

0.38

0.48

c

0.003

0.006

0.08

D

3

sse

L

MILLIMETERS

SYMBOL

7.24

Q

S

INCHES
NOTES

-.lc

I

I

tt

MILLIMETERS

MIN

1

II II

AQ

SYMBOL

1

24

1213

4

!-E-f--L--!f

=-r

III

e~'

-1c

I

1ark

0.15
10.92

0.430

E

0.245

e

0.050
0.008

0.285

6.22

0.Q15

0.20

0.38

sse

1.27

NOTES

3

7.24

sse

4

0.250

0.370

6.35

9.40

Q

0.010

0040

0.25

1.02

5

S

0.005

0.13

5,6

S2

0.004

0.10

5,6

NOTES:
1.

Index area; a notch or a lead one identification mark is located
adjacent to lead one and is located within the shaded area
shown. Alternatively, a tab (dim. k) may be used to identify lead
one. This tab may be located on either side as show~.
2. Dimension Q shall be measured at the point of exit of the lead
from the body.
3. This dimension allows for off~center lid, meniscus and glass
overrun.
4. The basic lead spaCing is 0.050 (1.27 mm) between centerlines.

6.

Dimension S, (See 40.3) may be 0.000 (0.00 mm) if corner leads
bend toward the cavity of the package within one lead's width
from the pOint of entry of the lead into body.

7.

Optional configuration. If this configuratIOn is used, no organic
or polymeric materials are molded to the bottom of the package
to cover the leads.
Optional, see note 1. If a lead one identification mark is used in
addition to this tab, the minimum limit of dimension k does not
apply.

8

5. Applies to all four corners.

17-18

PACKAGE
INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - FLATPACKS

14-Lead Flatpack (M-Suffix)

10-Lead Flatpack (L-Suffix)

See
k
NOle1':dre

L
1

sl L

7 8

t

Bottom-Brazed (MB-Suffix)

Bottom-Brazed (LB-Suffix)

LJ::_ _I~~I~
11 NoteSee7 7

1
--1.--P I---

A Q

INCHES
MIN

MAX

MIN

MAX

A

0.030

0.085

0.76

2.16

MILLIMETERS

SYMBOL

MIN

MAX

MIN

MAX

A

0.030

0.085

0.76

216

0.010

0.019

0.25

0.48

b

0.010

0.019

0.25

0.48

0.006

0.08

0.15

c

0.003

0.006

0.08

0.15

7.37

D

6.60

E

610

6.60

0.290
0.240

0.260

6.10

0280

E,

7.11

0.125

3.18

0.030

0.76

0.050 B5e

1.27

3

0.240

4

0.260
0.280

E,

0.20

038

8

E3

0.030

0.76

e

0050 B5e

127 B5e
0.015

L

0250

0.370

635

940

Q

0010

0.040

0.25

1.02

0.20

0.38

L

0.250

0.370

6.35

9.40

Q

0.010

0.040

0.25

1.02
1.14

5,

0.005

013

52

0.004

010

0.045
0.005

0.13

5,6

NOTES:
1. Index area; a notch or a lead one identification mark is located
adjacent to lead one and shall be located within the shaded area
shown. Alternatively, a tab (dim. k) may be used to identify lead
one. This tab may be located on Bither side as shown
2. Dimension Q shall be measured at the point of exit of the lead
from the body.
3. This dimension allows for off-center lid, meniscus and glass
overrun.
4. The basic lead spacing is 0.050 (1.27 mm) between centerlines.
5. Applies to all four corners.

4

0008

0.Q15

5

3

3

318

8

NOTES

7.11

0125

E2

esc

7.11

0280

0.008

5,

I-r
-I

0.003

D
E

NOTES

L

E3 E2 E3

INCHES

MILLIMETERS

SYMBOL

1

~

5,6

DimenSIOn 5, (See 40 3) may be 0.000 (O.OOmm) if corner leads
bend toward the cavity of the package wlthm one lead's width
from the point of entry of the lead mto body
7. Optional configuration If this configuration IS used, no organic
or polymeric matenals are molded to the bottom of the package
to cover the leads.
Optional, see note 1. If a lead one Identification mark IS used in
addition to this tab, the minimum limit of dimenSion k does not
apply.
6.

17-19

II

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - FLATPACKS

16-Lead Flatpack (F-Sufflx)

24-Lead Flatpack (N-Suffix)
See Note 1

k

See Note 1

b

=l"f
Tk

e

S2

f1

16

L

e~

D

T

~~11
9~

sL
f

=-r

L

1

s[f

BoUom-Brazed (FB-Suffix)

1

1213

Bottom-Brazed (NB-Suffix)

1r-~I-

4------+1" -----11

~-

11

See
A Q Note7

E3

INCHES
SYMBOL
A

MAX

MIN

MAX

0.045

0.085

1.14

2.16

A
b

b

0,015

0.019

0.38

0.48

c

0.003

0.006

0.08

0.15

0
E

0.440
0.245

0.285

E,
0.130

E3

0.030
0.050

6.22

SYMBOL

3

D

3

E,

E

7.24
7.75

0.76

sse

1.27
0,015

0.250
0.010

sse

MILLIMETERS

MIN

MAX

MIN

MAX

0.045

0.090

1.14

2.29

0.015

0.019

0.38

0.48

0.003

0.006

0.08

0.245

0.285

10.92

7.75

0.125

E3

0.030

0.76

sse

4

0050

8

0.008

0,015

1.27

0.370

6.35

9.40

0.040

0.25

1.02

6.35

9.40

0.250

0.23

1.02

2

Q

0.010

1.14

5

S

0.005

0.13

5,6

S2

0.004

0.10

6.

shown. Alternatively, a tab (dim. k) may be used to identify lead
one. This tab may be located on either side as shown.
2. Dimension Q shall be measured at the point of exit of the lead
from the body.
3. This dimension allows for off·center lid, meniscus and glass

7.

8

overrun.
4. The basic lead spacing is 0.050 (1.27 mm) between centerlines.
5. Applies to all four corners.

17·20

4
0.38

0.040

adjacent to lead one and is located within the shaded area

sse

0.20

0.370

NOTES:
1. Index area; a notch or a lead one identification mark is located

3

3.18

0.38

0.13

7.24

6.22

0.305

E2

NOTES

0.15

0.430

0.20

0.045
0.005

NOTES

3.30

0.008

S
S,

11.18

0.305

E2

Q

INCHES

MILLIMETERS

MIN

5,6

Dimension S, (See 40.3) may be 0.000 (0.00 mm) if corner leads
bend toward the cavity of the package within one lead's width
from the pOint of entry of the lead into body.
Optional configuration. If this configuration is used, no organic
or polymeric materials are molded to the bottom of the package
to cover the leads.
Optional, see note 1. If a lead one identification mark is used in
addition to this tab, the minimum limit of dimension k does not
apply.

PACKAGE
INFORMATION
Precision MOllolithics Inc.

PACKAGE DIMENSIONS - LEADLESS CHIP CARRIERS

20-Posltlon Chip Carrier
(RC-Sufflx)

Top View

INCHES

Side View

Bottom View

NOTES:

MILLIMETERS

1. A minimum clearance of 0.Q15" (0.381 mm) is maintained

SYMBOL

MIN

MAX

MIN

MAX

NOTES

A

0.064

0.100

1.63

2.54

5

Al

0.054

0.088

1.37

2.24

2. Electrical connection is required on plane 1. Metallization is
optional on plane 2. However, il plane 2 is metallized it must be

Bl

0.022

0.028

0.56

0.71

2

3. A minimum clearance of 0.020" (0.508 mm) is maintained

0

0.342

0.358

8.69

9.09

01

0.075 REF

1.91 REF

O2

0.200 REF

5.08 REF

03

0.100 REF

2.54 REF
0.358

04
05

0.150 BSe

E

0.342

between corner terminals.

electrically connected.

4.

9.09

5.
6.

3

3.81 BSe
0.358

8.69

El

0.075 REF

1.91 REF

E2

0.200 REF

5.08 REF

E3

0.100 REF

7.
9.09

1.27 BSe
0.38

~

I-Ll

9.09

0.050 BSe

Z

0
.....

0>.Lo
Z
.....

1.91 REF
0.358

E4

between overall dimensions D4 X E4 and all other features,
including metallization, chamfers and edges.
Non-electrical features for No.1 terminal identification, optical
orientation or handling purposes shall be within the shaded
area shown on plane 2.
Dimension A controls the overall package thickness.
Length of pad metallization may increase only toward package
periphery.
When space is available, the index corner may be metallized on
either or both planes 1 and 2. The package edge at the index
corner shall not be metallized.

el

0.015

Ll

0.045

0.055

1.14

L2

0.077

0.093

1.96

2.36

R

0.007

0.011

0.18

028

~
~
Poi

1.40
4

17-21

PACKAGE

INFORMATION
Precision Monolithics Inc.

PACKAGE DIMENSIONS - LEADLESS CHIP CARRIERS

28-Positlon Chip Carrier
(TC-Sufflx)

Side View

Top View

INCHES
SYMBOL

MIN

MAX

MAX

A

0.064

0.100

1.63

2.54

A1

0.054

0088

1.37

2.24

B1

0.022

0.028

0.56

0.71

0

0.442

0.458

11.23

11.63

01

0.Q75 REF

1.91 REF

O2

0.300 REF

7.62 REF

03

0.150 REF

04

0.200 BSC

optional on plane 2. However, if plane 2 is metallized it must be
electrically connected.
3. A minimum clearance of 0.020" (0.508 mm) is maintained
between overall dimensions 04 X E4 and all other features,
Including metallization, chamfers and edges.
4 Non-electrical features for No.1 terminal identification, optical
orientation or handling purposes shall be within the shaded
area shown on plane 2.
5. Dimension A controls the overall package thickness.
6. Length of pad metallization may Increase only toward package
periphery.
When space IS available, the Index corner may be metallized on
e.ther or both planes 1 and 2. The package edge at the index
corner shall not be metallized

11.63

E

0.442
0.Q75 REF

E2

0.300 REF

7.62 REF

E3

0150 REF

3.81 REF

e

1. A minimum clearance of 0.015" (0.381 mm) is maintained
between corner terminals.
2. Electrical connection is required on plane 1. Metallization is

508 BSC

E1

0.458

11.23

1163

191 REF

0.458

E4

NOTES

3.81 REF
0.458

05

NOTES:

MILLIMETERS
MIN

11.63

3

1.27

0.050

e1

0.015

L1

0.045

0055

1.14

1.40

L2

0077

0.093

1.96

2.36

R

0007

0011

0.18

028

BollomVlew

0.38

17-22

Table of Contents 1
Ordering Information 2
Product Assurance Program 3
Industry Cross Reference 4
Operational Amplifiers 5
Instrurp.entation Amplifiers .6
Voltage Followers/Buffers 7
Voltage Comparators 8
Matched Transistors 9
Voltage References 10
Digital-to-Analog Converters 11
Analog-to-Digital Converters 12
Analog Switches/Multiplexers 13
Sample-and -Hold Amplifiers 14
Special Functions 15
Communications Products 16
Package Information 17
Sales Offices, Representatives and Distributors 18

..

-

- .-. . . .

...

.--.--

~.---

SALES OFFICES
REPRESENTATIVES
PI t.'C 1 .... 1011 MOllOhtll1C ....

Inc

NORTH AMERICA
COLORADO

GEORGIA

MAINE

ATLANTA
EMA

Contact Boston. MA Office

620 ColOnial Park Or
Roswell, GA 30075-3746

MARYLAND

Santa Clar., CA 95052-8020

BOULDER
FRONT RANGE MARKETING
3100 Arapahoe Ave. SUite 404
Boulder, CO 80303-1082
(303) 443-4780 TWX 910-940-3442

(408) 727-9222 TWX 910-338-0218
TLX 172 070

LITTLETON
PMI SALES OFFICE

ALABAMA

7488 W Roxbury PI
Littleton, CO 80123-4738

CORPORATE
HEADQUARTERS
PMI
1500 Space Park Drive
P.O. Box 58020

HUNTSVILLE
EMA
309 Jordan Lane, Northwest
Huntsville, AL 35805-2621
(205) 830-4030. (800) 533-2920
TWX 810-726-2110

(303)979-8533

CONNECTICUT
UNIONVILLE
PMI SALES OFFICE
PO Box 43

SCOTTSDALE
PMI SALES OFFICE
8526 E Monterey Way
Scottsdale. AZ 85251-5926
(602) 941-1946

HAWAII

26 W Pennsylvania Ave
Baltimore, MO 21204~5058

Contact Los Angeles, CA OffIce

(301)296-2444

IDAHO

MASSACHUSETTS

Contact Bellevue, WA Office

BOSTON
PMI SALES OFFICE
209 W Central SI

CHICAGO
PMI SALES OFFICE

Natick, MA 01760-3716

(817)655-8900 TWX 710-386-0114

(203) 873-9995

1325 Remmgton Rd , SUIte "0"
Schaumburg, IL 60195~4813

MICHIGAN

DELAWARE

(312) 885-8440. (800) 323-8755
TWX 910-222-1808

DETROIT
PMI SALES OFFICE

Contact Bellevue, WA Office

ARIZONA

BALTIMORE
CONROY SALES

ILLINOIS

Unionville. CT 08085~OO43

ALASKA

(404)992-7240 TWX 810-726-2110

Contact Philadelphia. PA Office

INDIANA

DISTRICT OF COLUMBIA
Contact PhiladelphIa, PA OffIce

CARMEL
TECHNOLOGY MARKETING CORP

FLORIDA

599 Industnal Dr
Carmel, I N 46032~4207

(317) 844-8462 TWX 910-997-0194

722 Grand River Rd , SUite 6
Bnghton, MI 48116~1820

(313) 227-2190. (312) 885-8440
(800) 323-8755 TWX 910-222-1808
BRIGHTON
ELECTRONIC SOURCE. INC
8014 W Grand River, SUite 6
Brighton. MI 48118-9302
(313) 227-3598 TLX 298 269

SCOTTSDALE
SUMMIT SALES
7825 E Redfield Rd
Scottsdale, AZ 85260·3486
(602) 998-4850 TWX 910-950-1283

405 Douglas Ave
Altamonte Springs. FL 32714-2505

Ft wayne, IN 46804-2142

MINNESOTA

(305) 788-1403. (800) 223-6147
TWX 910-881-4079

(219) 432-5553 TWX 910-997-0195

ARKANSAS

MINNEAPOLIS
MEL FOSTER TECH SALES. INC

ALTAMONTE SPRINGS

IOWA

LAWRENCE ASSOCIATES. INC
711 Turnbull Ave
Altamonte Spnngs. FL 32701~6420

DAveNPORT
RUSH & WEST ASSOCIATES. INC

7611 Washington Ave, South
PO Box 35216
Edina. MN 55435-0216

Contact Ganzel Sales & Aseoe , Inc
Tulsa, OK Office

CALIFORNIA
LOS ANGELES
PMI SALES OFFICE
4025 E La Palma Ave
Anaheim, CA 92807-1716
(714) 666-0140. (213) 642-0142
TWX 910-328-6591

ALTAMONTE SPRINGS
PMI SALES OFFICE
Bldg II. SUIte 2505-5

FT. WAYNE
TECHNOLOGY MARKETING CORP
3428 W Taylor St

3 Columbia Ct , Northwest
Davenport, IA 52804~2416

(305)339-3855 TWX 510-953-7802
BOCA RATON
LAWRENCE ASSOCIATES. INC

(319) 326-3091

2151 N W 2nd, Suite 104
Boca Raton. FL 33431~7405

KANSAS

(612) 941-9790 TWX 910-578-2746

MISSISSIPPI
Contact EMA, Huntsville, AL OffIce

KANSAS CITY/WICHITA
RUSH & WEST ASSOCIATES. INC

(305) 368-7373 TWX 510-953-7802

BALLWIN
RUSH & WEST ASSOCIATES, INC

Olalhe. KS 66061-3690
(913) 764-2700 TWX 910-380-8110

720 W Manchester Rd • SUite 200
BallWin, MO 63011~3027

CLEARWATER
LAWRENCE ASSOCIATES. INC

SAN JOSE
QUAOREP INC.
2713 N First St.
San Jose, CA 95134-2099
(408)94&-4000 TWX 910-338-0207

MELBOURNE
LAWRENCE ASSOCIATES. INC

LOUISVILLE
TECHNOLOGY MARKETING CORP

1101 W HIbISCUS Blvd. SUite E115
W Melbourne, FL 32901-2745

8819 Roman Ct

(305) 724-8294 TWX 510-953-7802

LOUISVIlle, KY 40291-0147

1605 MIssourI Ave
Clearwater. FL 33516~1220

MISSOURI

107 N Chester St

SAN DIEGO
L & S ASSOCIATES
11772 Sorrento Valley Rd, SUite 235
San Diego. CA 92121-1017
(819) 455-0055 TWX 910-322-1730

(314) 394-7271 TLX 752 653

(813)584-8110 TWX 510-953-7602

KENTUCKY

PO Box 91147

(502)499-7808 TWX 810-535-3757

SANTA CLARA
PMI SALES OFFICE
4633 Old Ironsides Dr, Surte 312

LOUISIANA

Santa Clara, CA 95054-1806
(408) 727-6616 TWX 910-338-2102
TLX 172 070

Contact Houston, TX OffIce

18-2

MONTANA
Contact Littleton. CO Office

SALES OFFICES
REPRESENTATIVES
Pn:ClSlon Monolithics Inc.

NORTH AMERICA
NEBRASKA

OHIO

TENNESSEE

WISCONSIN (EASTERN)

Contact Rush & West ASSOCiates,
Davenport, IA Office

CLEVELAND
DEL STEFFEN & ASSOCIATES
69 Alpha Park
Cleveland, OH 44143-2296
(216) 461-8333 TWX 810-427-9272

DONELSON
EMA
118 Spring Valley
Donelson, TN 37214-2822
(615) 883-1414

Contact Chicago, IL Office

COLUMBUS
DEL STEFFEN & ASSOCIATES
355 W Main St
Lexington, OH 44904-9767
(419) 884-2313 TWX 810-427-9272

JONESBORO
EMA
Route 8, Dogwood Vlg
Jonesboro, TN 37659-9807
(615) 753-8081. (800) 633-2920

NEVADA
Contact Santa Clara, CA Office

NEW HAMPSHIRE
Contact Boston, MA Office

NEW JERSEY (NORTHERN)
Contact J-Square Marketing,
HickSVille, NY Office

NEW JERSEY (SOUTHERN)

TEXAS
DALLAS
PMI SALES OFFICE
11325 Pegasus St, SUite E-102
Dallas, TX 75238-3228
(214) 341-1742. (800) 223-6147
TWX 910-861-4079

OKLAHOMA
NEW MEXICO

TULSA
GENZEL SALES & ASSOC, INC
4135 S 100th East Ave, SUite 101
Tulsa, OK 7414&-3635
(918) 622-7744 TLX 312 394

HOUSTON
PMI SALES OFFICE
POBox 262423
Houston, TX 77207-2423
(713) 481-6460. (214) 341-1742
(800) 223-6147 TWX 910-861-4079

OREGON
NEW YORK
SYRACUSE
L-MAR ASSOCIATES. INC
216 Tilden Dr
East Syracuse, NY 13057-1630
(315) 437-7779
METRO NY, LONG ISLAND
J-SQUARE MARKETING. INC
161-C Levltown Pkwy
HickSVille, NY 11801-4421
(516) 935-3200 TWX 510-221-2136
ROCHESTER
L-MAR ASSOCIATES. INC
4515 Culver Rd
Rochester, NY 14622-1497
(716) 323-1000 TWX 510-253-0943

NORTH CAROLINA
PINEVILLE
EMA
8539 Glenway Ct
Pineville, NC 28134-8326
(704) 541-2628 TWX 810-726-2110

PORTLAND
NORTHWEST MARKETING
SUite 330
6975 S W Sandburg Rd
Portland, OR 97223-8010
(503) 620-0441 TWX 910-464-5157

UTAH
SALT LAKE CITY
FRONT RANGE MARKETING
2520 S State St, SUite 117
Salt Lake City, UT 84115-3110
(801) 364-6481 TWX 910-925-4117

PENNSYLVANIA
VERMONT

PHILADELPHIA
PMI SALES OFFICE
431 LakeSide Dr
Horsham, PA 19044-2320
(215) 675-7600 TWX 710-670-0021

Contact Boston, MA Office

VIRGINIA
Contact Conroy Sales,
Baltimore, MD Office

PITTSBURGH
DEL STEFFEN & ASSOCIATES
Carnegie Office Park
Bldg 1, Rm 116K
600 N. Bell Ave
Pittsburgh, PA 15106-4363
(412) 276-7366 TWX 810-427-9272

WASHINGTON
BELLEVUE
NORTHWEST MARKETING
SUite 330N
12835 Bellevue Redmond Rd
Bellevue, WA 98005-2625
(206) 455-5846 TWX 910-443-2445

SELLERSVILLE
TECH-COM MARKETING
PO Box 460
Sellersville, PA 18960-0460
(215) 453-0711

WINSTON SALEM
EMA
3463 Kernerville Rd
Winston Salem, NC 27107-1750
(919) 784-7304 TWX 810-726-2110

Contact Boston, MA Office

NORTH DAKOTA

SOUTH CAROLINA

Contact Mel FosterTech Sales. Inc ,
Edina, MN Office

Contact EMA, WInston Salem, NC
Office

WYOMING
Contact Llttfeton, CO Office

CANADA

DAYTON
DEL STEFFEN & ASSOCIATES
1201 E David Rd
Dayton, OH 45429-5701
(513) 293-3145 TWX 810-427-9272

Contact Philadelphia, PA Office

ALBUQUERQUE
BFA CORPORATION
1704 Moon, Northeast
Albuquerque, NM 87112-3936
(505) 292-1212 TWX 910-989-1157

WISCONSIN (WESTERN)
Contact Mel Foster Tech Sales, Inc,
Edina, MN Office

WEST VIRGINIA
Contact Del Steffen & Associates,
Pittsburg, PA Office

RHODE ISLAND

ALBERTA
HI-TECH SALES LIMITED
PO Box 115
339 10th Ave Southeast
Calgary. Alberta T2G OW2
(403) 239-3773
BRITISH COLUMBIA
HI-TECH SALES LIMITED
7510 B Kingsley
Burnaby, British Columbia V3N 3C2
(604) 524-2131
MANITOBA
HI-TECH SALES LIMITED
#102-902 St James St
Winnipeg, Manitoba R3G 3J7
(204) 786-3343
NEW BRUNSWICK
Contact Source Electronics Ltd,
Rexdale. Ontano Office
NEWFOUNDLAND
Contact Source Electronics Ltd.,
Rexdale. Ontario Office
NOVA SCOTIA
Contact Source Electromcs Ltd,
Rexdale, Ontario Office

ONTARIO
SOURCE ELECTRONICS LTD
PO Box 13235
Kanata, Ontario K2K 1X4
(613) 592-5392
ONTARIO
SOURCE ELECTRONICS LTD
83 Galaxy Blvd. Unit 9
Rexdale, Ontario M9W 5X6
(416) 675-6235
QUEBEC
Contact Source ElectronIcs Ltd ,
Rexdale, Ontano OffIce
SASKATCHEWAN
Contact HI-Tech Sales LImited
Winnipeg, Manitoba OffIce

SOUTH DAKOTA
Contact Mel Foster Tech Sales. Inc,
Edina, MN Office

18·3

I

AUTHORIZED
DISTRIBUTORS
1'1~l:l""l()n MOllOhthll.: ... ItH':

NORTH AMERICA
ALABAMA
HUNTSVILLE
HALL-MARK ELECTRONICS
4900 Bradford Dr
Huntsville, AL 35805-1951
(205) 837-8700 TWX 810-726-2187

CALIFORNIA continued
SACRAMENTO
BELL INDUSTRIES
500 Giuseppe Ot , SUite 6
Roseville, CA 95678-6305
(916) 969-3100

CONNECTICUT
NORWALK
PIONEER
112 Main 8t
Norwalk, CT 06851-4617
(203) 853-1515 TWX 710-488-3373

ILLINOIS
WOOD DALE
HALL-MARK ELECTRONICS
210 Mltel Dr
WOod Dale, IL 60191-1120
(312) 880-3800 TWX 910-651-0185

HUNTSVILLE
PIONEER
4825 Unrverslty Square
Huntsville. AL 35805-6041
(205) 837-9300 TWX 810-726-2197

SAN DIEGO
ANTHEM ELECTRONICS
9639 Carroll Park Dr
San D,ego, CA 92121-1406
(619) 453-9005 TWX 910-335-1515

WALLINGFORD
HALL-MARK ELECTRONICS
33 Village lane
Wlilingford, CT 06492-2428
(203) 269-0100 TLX 314-207

CHtCAGO
PIONEER
1551 Carmen Dr
Elk Grove Village, IL 60007-6581
(312) 437-9680 TWX 910-222-1834

ARIZONA
PHOENIX
HALL-MARK ELECTRONICS
4040 E Raymond
PhoenIx, AZ 85040-1983
(602) 437-1200 TWX 910-951-1325
TEMPE
ANTHEM ELECTRONICS

1727 E VVeber Dr
Tempe, AZ 85281-1841
(602) 966-6600 TWX 910-950-0110
TEMPE
BELL INDUSTRIES
1705 W 4th St
Tempe, AZ 85281-2403
(602) 966-7800 TWX 910-950-0133
CALIFORNIA
CANOGA PARK
HALL-MARK ELECTRONICS

8130 Remmel Ave
Canoga Park, CA 91304-4129
(818) 716-7300
CHATSWORTH
ANTHEM ELECTRONICS

20640 Bahama 8t
Chatsworth, CA 91311-6101
(618) 700-1000 TWX 910-493-2083
GARDEN GROVE
BELL INDUSTRIES
12322 Monarch St.
Garden Grove. CA 92641 ..2909
(714) 220-0681 TWX 910-596-2362
GARDENA
BELL INDUSTRIES

306 E Alondrs Blvd
Gardena, CA 90248-2810
(213) 515-1800 TWX 910-348-6336
IRVINE
ANTHEM ELECTRONICS
1 Oldfield Dr
Irvine. CA 9271&-2809
(714) 788-4444 TWX 910-595-1583
LOS ALAMITOS
SEMI DICE INC.
10981 Bloomfield 51.
Los Alamitos, CA 90720-2586
(213) 594-4831 TWX 910-341-7710

SAN DIEGO
BELL INDUSTRIES
7450 Ronson Rd
San Diego, CA 92111-1508
(619) 268-1277

FLORIDA
CLEARWATER
HALL-MARK ELECTRONICS
15301 Roosevelt Blvd, SUite 303
Clearwater, FL 33520-3561
(813) 530-4543

SAN DtEGO
HALL-MARK ELECTRONICS
3878 Ruffm Ad , SUite 10B
Ssn Diego, CA 92123-1849
(619) 266-1201

FT. LAUDERDALE
PIONEER
674 S MIlitary Trail
Deerfield Beach, Fl 33442-3023
(305) 428-8877 TWX 51D-955-9653

SAN JOSE
ANTHEM ELECTRONICS
1040 E Brokaw Rd
San Jose, CA 95131-2309
(408) 295-4200 TWX 910-338-2038

FT. LAUDERDALE
HALL-MARK ELECTRONICS
3161 S W 15th St
Pompano Beach, FL 33069-4806
(305) 971-9280 TWX 510-956-9720

SAN JOSE
HALL-MARK ELECTRONICS
1110 Ringwood Ct
San Jose, CA 95131-1726
(408) 946-0900 TWX 910-339-9505

ORLANDO
HALL-MARK ELECTRONICS
7648 Southland Blvd, SUite 100
Orlando, Fl32809-6993
(305) 855-4020 TWX 810-850-0105

SUNNYVALE
BELL INOUSTRIES
1161 N Falroaks Ave
Sunnyvale, CA 94089-2102
(408) 734-8570 TWX 910-339-9378

ORLANDO
PIONEER
221 N Lake Blvd
Altamonte Sprrngs, Fl 32701-4399
(305) 834-9090, (600) 432-6064
TWX 810-853-0284

THOUSAND OAKS
BELL INDUSTRIES
1829A De Havilland Dr
Thousand Oaks, CA 91320-1702
(805) 499-8821 TWX 910-321-3799

GEORGIA
NORCROSS
HALL-MARK ELECTRONICS
6410 AtlantiC Blvd, SUite 115
Norcross, GA 30071-1241
(404) 447-6000 TWX 810-786-4510

TORRANCE
HALL-MARK ELECTRONICS
19220 S Normandle
Torrance, CA 90502-1011
(213) 217-8400

NORCROSS (ATLANTA)
PIONEER
Peachtree Crossing Bus Park
5835B Peachtree Corners, East
Norcross, GA 30092-3404
(404) 448-1711 TWX 810-786-4515

TUSTIN
HALL-MARK ELECTRONICS
14831 Franklin Ave
Tustin, CA 92680-7292
(714) 889-4700 TWX 910-997-0523
COLORADO
DENVER
ANTHEM ELECTRONICS
8200 S Akron St.
Englewood, CO 80112-3505
(303) 790-4500

INDIANA
INDIANAPOLIS
HALL-MARK ELECTRONICS
4275 W 96th St
Indianapolis, IN 46268-1113
(317) 872-8875
INDIANAPOLIS
PIONEER
6408 Castleplace Dr
IndianapoliS, IN 46250-1914
(317) 849-7300 TWX 810-260-1794
KANSAS
LENEXA (KANSAS CITy)
HALL-MARK ELECTRONICS
10815 Lakeview Ave
Lenexa, KS 66219-1329
(913) 888-4747 TWX 910-749-6820
MARYLAND
BALTIMORE
HALL-MARK ELECTRONICS
10240 Old Columbia Rd.
Columbia, MO 21046-1218
(301) 986-9600 TWX 710-862-1907
GAITHERSBURG
PIONEER
9100 Gaither Rd
Gaithersburg, MD 20877-1422
(301) 921-0660 TWX 710-828-0545
MASSACHUSETTS
LEXINGTON
PIONEER
44 Hartwell Ave
LeXington, MA 02173-3103
(817) 861-9200 TWX 710-326-8817
NORWOOD
GERBER ELECTRONICS
128 Carnegie Row
Norwood, MA 02082-5010
(617) 769-6000 TWX 710-336-1987
WESTBOROUGH
FUTURE ELECTRONICS
133 Flanders Rd
Westborough, MA 01581-1005
(617) 366-2400 TWX 710-390-0374

DENVER
HALL-MARK ELECTRONICS
6950 S. Tucson Way, Suite G
Englewood, CO 80112-3922
(303) 790-1662 TWX 910-931-0472

WOBURN
HALL-MARK ELECTRONICS
6 Cook St.
Billerica, MA 01821-6036
(617) 935-9777 TLX 929 419

WHEAT RIDGE
BELL INDUSTRIES
8155 W. 48th Ave.
Wheat Rldgo, CO 80033-3199
(303) 424-1985 TWX 910-938-0393

18·4

AUTHORIZED
DISTRIBUTORS
1'1(.·<.:'..,1011 l'\'lonollthI<.:'" Inc

NORTH AMERICA
MICHIGAN

NEW MEXICO

OKLAHOMA

UTAH

LIVONIA
PIONEER
13485 Stamford
livonia, MI 48150-1598
(313)525-1800 TWX 810-242-3271

ALBUOUERQUE
BELL INDUSTRIES
11728 linn, Northeast
Albuquerque, NM 87123-2943
(505) 292-2700 TWX 910-989-0625

TULSA
HALL-MARK ELECTRONICS
5460 S 103rd East Ave
Tulsa, OK 74146-5815
(918) 685-3200 TWX 910-845-2290

SALT LAKE CITY
ANTHEM ELECTRONICS
1615 W 2200, South, Suite A
Salt Lake City, UT 84119-1456
(801)973-8555 TWX 910-925-5273

MINNESOTA

NEW YORK

OREGON

BLOOMINGTON
HALL-MARK ELECTRONICS
783812th Ave, South
Bloommgton, MN 55420-1477
(812) 854-3223 TWX 910-578-3167

BINGHAMTON
PIONEER
1806 VIstal Pkwy, East
vestal, NY 13850-1942
(607) 748-8211 TWX 510-252-0893

LAKE OSWEGO
ANTHEM ELECTRONICS, INC
15812 S W Upper Boones Ferry Rd
Lake Oswego, OR 97034-4066
(503) 884-2881

WASHINGTON

MINNEAPOLIS
MERIT ELECTRONICS CORP
warehouse SUile 210
2525 Nevada Ave , North
Minneapolis, MN 55427-3862
(812) 548-5363

FAIRPORT
PIONEER
840 Fairport Park
Fairport, NY 14450-2012
(716) 381-7070 TWX 510-253-7001

PORTLAND
BELL INDUSTRIES
8024 S W Jean Rd
lake Oswego, OR 97034-5390
(503) 241-4115 TWX 910-455-8177

REDMONO
ANTHEM ELECTRONICS
5020 148th Ave, Northeast
Redmond, WA 98052-5171
(206) 881-0850

LONG ISLAND
HALL-MARK ELECTRONICS

PENNSYLVANIA

MINNETONKA (TWIN CITIES)
PIONEER
10203 Bren Rd , East
Minnetonka, MN 55343-9072
(612) 935-5444 TWX 910-576-2738

MISSOURI
EARTH CITY
HALL-MARK ELECTRONICS
13750 Shoreline Or
Earth City, MO 63045-1224
(314) 291-5350, (800) 325-1021
TWX 910-762-0672

NEW JERSEY
FAIRFIELD
HALL-MARK ELECTRONICS
107 Fairfield Rd
Fairfield, NJ 07006-2412
(201) 575-4415 TWX 710-734-4409
FAIRFIELD
NU-HORIZONS ELECTRONICS
CORP
258 Route 46
Fairfield, NJ 07008-2324
(201) 682-8300
MT. LAUREL
HALL-MARK ELECTRONICS
1000 Mldlantlc Or
Mt Laurel, NJ 08054-1511
(609) 235-1900 TWX 710-940-0660
PINE BROOK
PIONEER
45 Route 46
Pme Brook, NJ 07058-9607
(201) 575-3510 TWX 710-734-4382

1 Comae Loop

HORSHAM
PIONEER
261 Glbralter Rd
Horsham, PA 19044-2377
(215) 874-4000 TWX 510-685-8778

Ronkonkoma, NY 11779·6816
(516) 737-0600 TWX 510-222-0162
NO, AMITYVILLE
NU-HORIZONS ELECTRONICS
CORP
6000 New HOrizons Blvd
No. AmitYVille, NY 11701-1130
(516) 226-6000, (800) 645-9222
TLX 221 228

PITTSBURGH
PIONEER
259 Kappa Or
Pittsburgh, PA 15238-2817
(412) 782-2300 TWX 710-795-3122

WOODBURY (LONG ISLAND)
PIONEER
60 Crossways Park, West
Woodbury, NY 11797-2019
(516) 921-8700 TWX 510-221-2184

SALT LAKE CITY
BELL INDUSTRIES
3639 W 2150, South
Salt Lake City, UT 84120-1286
(801)972-6989 TWX 910-925-5886

SEATTLE
BELL INDUSTRIES
1900 132nd Ave, Northwest
Bellevue, WA 98005-2288
(206) 747-1515 TWX 910-443-2482

WISCONSIN
NEW.ERLIN
HALL-MARK ELECTRONICS
16255 W Lincoln Ave.
New Berlin, WI 53151-2834
(414) 797-7844, (600) 242-5252
TLX 323 082

TEXAS

NORTH CAROLINA

AUSTIN
HALL-MARK ELECTRON ICS
12211 Technology Blvd
Austin, TX 78727-8102
(512) 258-8848 TWX 910-847-2031

CHARLOTTE
PIONEER
9801-A Southern Pine Blvd
Charlotte, Ne 28210-5562
(704) 527-8188 TWX 810-621-0366

AUSTIN
PIONEER
9901 Burnet Rd.
Austin, TX 78758-5239
(512) 835-4000 TWX 910-874-1323

RALEIGH
HALL-MARK ELECTRONICS
5237 North Blvd
Raleigh, Ne 27604-2925
(919) 872-0712 TWX 510-928-1831

DALLAl
HALL-MARK ELECTRONICS
10375 Brookwood Rd
Dallas, TX 75238-1856
(214)341-1147 TWX 910-887-4775
DALLAl
PIONEER
13710 Omega Rd.
Dallas, TX 75244-4518
(214) 388-7300, (BOO) 492-9027
TWX 910-880-5583

OHIO
CLEVELAND
PIONEER
4800 E 131.t St
Cleveland, OH 44105-7132
(216) 587-3600 TWX 810-422-2210

CANADA
ALBERTA
FUTURE ELECTRONICS, INC
5809 Macleod Trail South, Unit 109
Calgary, Alberta T2H OJ9
(403) 259-6408 TWX 610-821-1927
BRITISH COLUMBIA
FUTURE ELECTRONICS, INC
3070 Klngsway
Vancouver, British Columbia
V5R 5J7
(604) 438-5545 TWX 610-922-1688
BRITISH COLUMBIA
INTEK ELECTRONICS LTD
838551 George St -10
Vancouver, British Columbia
V5X 4P3
(604) 324-6831 TWX 610-922-5032
TLX 04-507578
ONTARIO
FUTURE ELECTRONICS, INC
Baxter Centre, 1050 Baxter Rd
Ottawa, Ontano K2C 3P2
(613) 820-8313 TWX 610-563-1697

HOUSTON
HALL-MARK ELECTRONICS
8000 Westglen
P,O. Box 42190
Houston, TX 77083-8485
(713) 781-8100 TWX 910-881-2711

DAYTON
PIONEER
4433 Interpolnl Blvd
PO Box 291
Dayton, OH 45424-5708
(513) 236-9900
TWX 810-459-162211623

ONTARIO
FUTURE ELECTRONICS, INC
82 51 RegiS Crescent N
Oownsvl8w, Ontano M3J 1Z3
(416) 638-4771 TWX 610-491-1470

HOUSTON
PIONEER
5853 POint West Or
Houston, TX 77036-2811
(713) 988-5555 TWX 910-881-1806

SOLON
HALL-MARK ELECTRONICS
5821 Harper Ad
Solon, OH 44139-1832
(216) 349-4632

QUEBEC (MONTREAL)
FUTURE ELECTRONICS, INC
237 Hymus Blvd
POinte Claire, Quebec H9R 5C7
(514) 694-7710 TWX 610-421-3251

WORTHINGTON
HALL-MARK ELECTRONICS
SUite "5"
400 e Wilson Bridge Ad
'Northington, OH 430as-.2321
(614) 888-3313

18-5

----- -

~-

~-

---~

-

~~

~-~

-

SALES OFFICES
REPRESENTATIVES
I'llll'>IOIl

\lollollthl(~'" lilt.:

INTERNATIONAL
EUROPEAN
HEADQUARTERS
BOURNSAG
ZUIII·,.t'.... 74

8340 a•• r
Swltzertand

Phone: 042-33 33 33
Telex: 888 722

ARGENTINA
NOise SAL
V Cevallos 239
(1077) Buenos Aires
ArgentIna
Phone 46-5776/0628/0864/2214
Telex 22892 NOise AR

AUSTRALIA
VICTORIA
RIFA Ply Ltd

202 Sell St
Preston, Victoria 3072
Phone (03) 480 1211
Telex AA 31001
Cable RIFMEL
NEW SOUTH WALES
RIFA Pty Ltd
2 Cross St
HurstvllJe, 2220 N S W
Phone (02) 510-6122
Telex 22515

AUSTRIA
Ing Otto Folger
Elektronlsche Geraete
Bllndengasse 36
1080 Vienna
Phone 0222-43 26 39
Telex 131 882

BENELUX
Sourns Benelux B V
Van Tuyl van Serooskerkestr 81-85

FINLAND

JAPAN

Insele ov
Kumpulantle 1
00520 HelSinki 52
Phone 90-75 06 00
Telex 122217

SPAIN

Nippon PMI Corporation
Haratetsu BUilding
4-1-11, Kudan Klta
Chlyoda-ku, Tokyo
102 Japan
Phone (03) 234-141'
Telex 781 J 27632

Selco SA
Paseo de la Habana, 190
28036 Madnd
Phone 01-40542 13
Telex 45458

FRANCE
Bourns OHMIC S A

21123 rue des Ardennes

KOREA

75019 Pans
Phone 01-42039633
Telex 230008

Yeoml & Company, Ltd
#498-5 Dapslpn-Oong
Oongdaemoon-Ku, Seoul
Phone (02) 244-1492
Telex K24123 YEONIL

GERMANY
Bourns GmbH
Postfach 1155
Brelte Strasse 2
7000 Stuttgart 1
Phone 0711-22930
Telex 721 656
Bourns GmbH
Bahnhofstrasse 4b
8057 Echlng
Phone 089-319 12 70

ISRAEL

Phone 02-986333

AVIV ElectrOnics Ltd
12, Kehllat Venecla St
PO B 24190
Tel-AVIV 61241
Phone 03-49 44 50
Telex 33572 mavlv II
Fax 3-494065

EASTERN EUROPE
Dlpl Ing
Gerhard StOlte
Nordbahnstrasse 44/15
1020 Vienna
Austria
Phone 0222-24 71 37
Telex 134 171

Dynamar International Ltd
12, Lorong Bakar Batu, #65-11
Kolam Ayer Industrial Park
Singapore 1534
Phone 747-6188
Telex RS26283 Dynama

Components Agent Ltd.
Unit 2301 C-2, Nan Fung Centre
298 Castle Peak Road, N T
Phone 0-4992688
Telex 30398 Comag HX

2880 Bagsvaerd-Copenhagen
Telex 37350

SINGAPORE

HONG KONG

SUJata Sales & Electromcs
141-A Mlltal Ct
Narlman POint
Bombay 40021
Phone 222-999
Telex 011 3855

KrogshoJV8J 51

Rua Rodrigo da Fonseca 103
Lisbon 1
Phone 19-68 60 72
Telex 42827

Germanls Co
Trade of ElectrOnic Gear
Anstotelous 51 47-49
PO Box 8209
10010 Athens
Phone 01-821 5825
Telex 219179

2270 AA Voorburg
Phone 070-87 54 04

E Frus-Mlkkelsen AlS

PORTUGAL
TeJectra S A R L

GREECE

INDIA

DENMARK

AlS Klell Bakke
Ovre Raellngsvel 20
PO Box 27
2001 Lillestrom
Phone 02-83 02 20
Telex 19407

REGIONAL OFFICE

PO Box 37
Telex 32023

NORWAY

REGIONAL OFFICES
BARCELONA
Selco SA
Gran via de las cortes catalanas,
1176 bls
080020 Barcelona
Phone 03-3147411
BILBAO
Selco SA
Rodnguez Arias, 71 bls
48013 Bilbao
Phone 04-442 46 00

SWEDEN
Bexab Elektronlk AB
PO Box 516
18325 Taeby
Phone 08-7680560
Telex 10912

SWITZERLAND
Bourns (Schwelz) AG
Zugerstrasse 74
6340 Baar
Phone 042-33 33 33
Telex 868 722

TAIWAN
Mornhan International Corporation
9F No 176 Fu-Hslng N Rd
Taipei, Taiwan, ROC
Telex (785) 20422 Mornhan

SOUTH AFRICA
JOHANNESBURG
ASSOCiated Electronics (Ply) Ltd
POBox 31094, Braamfonteln 2017
ASSOCiated House, 150 Caroline St
Bnxton, Johannesburg
Phone 011-839 18 24
Telex 425 586
DUNSWART
Allied electronic Components
(Pty) lid
PO Box6387
Dunswart 1508
Phone 52-86 61
Telex 425559

TURKEY
NEL Elektronlk
Suemer Sokak No 42/1
Yemsehlr-Ankara
Phone 041-30 15 10
Telex 42229

REGIONAL OFFICE
NEL Elektronlk
Inonu Cad Dumen Sokak 1/15
Takslm-Istanbul
Phone 01-144 06 36
Telex 24549

UNITED KINGDOM
Bourns Electronics Ltd
Hodford House
17/27 High St
Hounslow, Middlesex TW3 1TE
Phone 01-5726531
Telex 264485

ITALY
TECHNIC S r L
Via Brembo 21
20139 Milan
Phone 02-569 57 46
Telex 316651

YUGOSLAVIA
Jugomlneral
Sektor Inozemna zastupstva
1llca 34/11
PO Box376
41000 Zagreb
Phone 041-423746
Telex 21 194

REGIONAL OFFICE
TECHNIC Sr L
Via Ippomo 2
00183 Rome
Phone 06-77 83 94

18-6

AUTHORIZED
DISTRIBUTORS
I'n·CI .... I01l

l\'lonollthtcs Inc

INTERNATIONAL
BELGIUM
Aunema BelgIum
rue Brognl9zstraat 172 a

1070 Brussels
Phone 02-523 82 95

Telex' 216 48
FRANCE
MAISONS-LAFITTE

I SA Electromqu8
50/52, rue d'Acheres
78600 Mal80ns-lafltte
Phone 1-39 122452
Telex 695 877
SURESNES
ISC
International Semi-Conductor

Corp France
28/30 rue de 18 ProcesSion
B P 118

92153 Suresnes
Phone 1-45 06 42 75

Telex, 614 596
VERRIERES LE BUISSON
HYBRITECH (for dice only)
Z.I des Godets
Aoute de Bua
CE421

91374 V8meres Ie BUisson
Phone. Hl9 20 22 10
Telex 300 315
CHATILLON CEDEX
BANELEC
90. rue Pierre Semard
B P 83
92322 Chatillon Cedex

Phone 1-46 55 43 43
Telex 204 874
CLiCHY CEDEX
DIMACEL
11, rue Jean d'Asmeres
B P 280
92113 Cllchy Cedex
Phone 1-4730 15 15
Telex 610 652
BONNEUIL CEDEX
SYSCOM
Z A des Petlts-Carreaux
12, av des Coquehcots
94385 BonneUlI Cedex
Phone 1-4377 84 88

Telex. 231 568
LILLE
DIMACEL

78, rue

Boucher~de~Perthes

59800 lille
Phone 20 30 85 80
Telex 110 173
STRASBOURG
DIMACEL
17, Boulevard de Nancy
6700 Strasbourg
Phone 882207 19
Telex 880 372

FRANCE continued

GERMANY continued
BREMEN
Assmy & Bottger
Ingolstadterstrasse 1/3
2800 Bremen 1
Phone 0421-38941
Telex' 244 363

BORDEAUX CEDEX
S,C T. Toutelectrlc
80/83, qual de Queyrles
33072 Bordeaux Cedex
Phone: 58 88 50 31
Telex 550 988
SAINT PRIEST
DIMACEL
Censloz 2-Tour 5
32bls, Boulevard des Roses
69800 Saint Priest
Phone' 78 21 3721
Telex 380 010

HAMBURG
Walter Kulxen GmbH
Nordkanalstrasse 52
2000 Hamburg 1
Phone. 040-2 37 01 0
Telex, 2 162074
HANNOVER
Elkose GmbH
Vahrenwalderstr8sse 205/207
3000 Hannover 1
Phone: 0511-6 78 05 0
Telex 921 501

SAINT MARTIN d'HERES CEDEX
DIMACEL
21, rue Beal
Z I Sud
B P 155
38404 Saint Martin d'Heres Cedex

DORTMUND
Elkose GmbH
llndenhorsterstrasse 38
4600 Dortmund
Phone 0231-8 40 50
Telex, 8 227 882

LYON CEDEX
RADIALEX
74, rue Vendome
B P 6003
69411 lyon Cedex 06
Phone 78 89 45 45
Telex 300 238

DREIEleM
Spoerle ElectrOnic KG
M8x-Planck~Str8sse 1/3
6072 Drelelch
Phone 06103-3 04 0
Telex 417972

GRENOBLE
RADIALEX
6, rue Georges Jacquet
B P 866
38036 Grenoble
Phone 76 49 49 92
Telex 320 378

MOGLINGEN
Elkose GmbH
BahnhofstraS8e 44
7141 Moghngen
Phone 07141-48 70
Telex 7 264 472

TOULOUSE CEDEX
S C, T Toutelectrlc
37, av Emile Dewoltme
B P 2167
31022 Toulouse Cedex
Phone. 61 220422
Telex 530 219

STUTTGART
DACOM Electronic Vertnebs GmbH
Huttenelchenweg 10
7000 Stuttgart 80
Phone 0711-741021
Telex 7 255 309

TOULOUSE
DIMACEL
284, route de St Simon
31100 Toulouse
Phone 61 40 96 50
Telex, 521 364

MUNICH
Astronlc
Wlnzererstrasse 47 0
8000 Munchen 40
Phone 089-30 90 31
Telex 5218 187

LESMILLES
DIMACEL
64, rue Georges Claude
Z I Alx en Provence
13763 les Milies
Phone 42 39 85 50
Telex 441 569

KIRCHHEIM
MBS ElectrOnic
Benzstrasse 1
8011 Klrchhelm
Phone 089-903 85 51
Telex 5 215 555

TOULON CEDEX
DIMEL
"le MarinO"
Avenue Claude Farrere
BP 1153
83058 Toulon Cedex
Phone 94 41 49 63
Telex. 430 093

RENNES
DIMACEL
5, rue loUis Turban
35100 Rennes
Phone 99 50 25 92
Telex 950 466

GERMANY
BERLIN
Spoerle ElectrOniC KG
Gnelsenaustrasse 2
1000 Berhn 61
Phone 030-693 40 90
Telex 186 029

BORDEAUX
DIMACEL
137/139, rue Crolx-de-Seguey
33000 Bordeaux
Phone 56 81 14 40
Telex 540 579

BERLIN
Distron GmbH & Co
BehSlmstrasse 3
1000 Berlin 10
Phone 030-342 10 41 45
Telex 185 478

PUTZBRUNN
Sasco GmbH
Hermann-Oberth-Strasse 16
8011 Putzbrunn
Phone 089-461 10
Telex 529 504
NUERNBERG
Semtech GmbH
Hallerwelherstrasse 5
8500 Nuernberg 50
Phone 0911-831003
Telex 623 960
ISRAEL
AVIV ElectrOniCs ltd
12, Kehllat Veneela St
P.O Box 24190
Tel-AVIV 61241
Phone 03-49 44 50
Telex 33572

18-7

ITALY
MILAN
ComprelS,PA
Vlale F, Testl, 115
20092 Clnlsello Balsamo
Milan
Phone 02-6120641/5
Phone 02~612 66 41/5
Telex, 332 484
ROME
Comprel SPA
Via Farne8lna, 224
00194 Rome
Phone' 08-328 67 39
Phone 08-328 88 89
YICENZA
Comprel S PA
Via Paslm 18
36100 Vlcenza
Phone, 0444-3 39 12
NETHERLANDS
HAAKSBERGEN
Texlm ElectrOniCs B V
Albert Cuyptaan 4
7482 JA Haaksbergen
Phone 05427-333 33
Telex, 448 08
STADSKANAAL
Ellncomb B V
Oosterkade 33,
PO Box 248
9500 AE Stadskanaal
Phone 05990-14830
Telex 53378
SWEDEN
Elfa RadiO and TelevIsion AB
17117Solna
Phone 08-730 07 00
SWITZERLAND
Distrelec AG
Hardturmstrasse 131
8037 ZUrich
Phone 01-276 22 11
Telex 823 045
UNITED KINGDOM
CAMBRIDGE
HlwTek ElectrOnics Ltd
Ditton Walk
Cambndge CB5 8CD
Phone 0223-21 33 33
Telex 817 347
BEDFORD
RR ElectrOniCs ltd
SI MartinS Way Industrial Estate
Cambndge Road
Bedford M K42 OlF
Phone 0234-47 188
Telex 826 251
KENT
Jermyn Distribution Ltd
Vestry Estate
Sevenoaks
Kent TN 14 5EU
Phone 0732-4501 44
Telex, 95 142
ESSEX
STC Electronics Services
Edinburgh Way
Harlow
Essex CM20 2DF
Phone 0279-26 777
Telex 81525

II

Production
Team
Text Coordination and Production: Kaz Hamano
Rose Kimball
Judy Sharp
Cover Design: Kaz Hamano
Typesetting: Alana Haas
Proofreading: Sarah Baker
Project Management: Jean-Marie DeRousse



Source Exif Data:
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.3
Linearized                      : No
XMP Toolkit                     : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:56:37
Create Date                     : 2013:08:22 11:52:40-08:00
Modify Date                     : 2013:08:22 12:51:45-07:00
Metadata Date                   : 2013:08:22 12:51:45-07:00
Producer                        : Adobe Acrobat 9.55 Paper Capture Plug-in
Format                          : application/pdf
Document ID                     : uuid:def004b6-5b18-2d49-945e-ff4cd95bf958
Instance ID                     : uuid:2192088d-1cf4-524d-8680-6dd588e597b9
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 899
EXIF Metadata provided by
EXIF.tools

Navigation menu