Ngspice User Manual V27

User Manual: Pdf

Open the PDF directly: View PDF PDF.
Page Count: 624

DownloadNgspice User Manual Ngspice-manual V27
Open PDF In BrowserView PDF
Ngspice Users Manual
Version 27
(Describes ngspice-27 release version)
Holger Vogt, Marcel Hendrix, Paolo Nenzi
August 31, 2017

2

Locations
The project and download pages of ngspice may be found at
Ngspice home page http://ngspice.sourceforge.net/
Project page at sourceforge http://sourceforge.net/projects/ngspice/
Download page at sourceforge http://sourceforge.net/projects/ngspice/files/
Git source download http://sourceforge.net/scm/?type=cvs&group_id=38962

Status
This manual is a work in progress. Some to-dos are listed in the following. More is surely
needed. You are invited to report bugs, missing items, wrongly described items, bad English
style etc.

To Do
1. Review of Chapt. 1.3
2. hfet1,2 model descriptions

How to use this manual
The manual is a ‘work in progress’. It may accompany a specific ngspice release, e.g. ngspice24 as manual version 24. If its name contains ‘Version xxplus’, it describes the actual code
status, found at the date of issue in the Git Source Code Management (SCM) tool. The manual is
intended to provide a complete description of the ngspice functionality, its features, commands,
or procedures. It is not a book about learning SPICE usage, but the novice user may find some
hints how to start using ngspice. Chapter 21.1 gives a short introduction how to set up and
simulate a small circuit. Chapter 32 is about compiling and installing ngspice from a tarball or
the actual Git source code, which you may find on the ngspice web pages. If you are running a
specific Linux distribution, you may check if it provides ngspice as part of the package. Some
are listed here.

Part I
Ngspice User Manual

3

Contents
I

Ngspice User Manual

1

Introduction

35

1.1

Simulation Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

36

1.1.1

Analog Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . .

36

1.1.2

Digital Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . .

37

1.1.3

Mixed-Mode Simulation . . . . . . . . . . . . . . . . . . . . . . . . .

37

1.1.4

Mixed-Level Simulation . . . . . . . . . . . . . . . . . . . . . . . . .

38

Supported Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39

1.2.1

DC Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39

1.2.2

AC Small-Signal Analysis . . . . . . . . . . . . . . . . . . . . . . . .

40

1.2.3

Transient Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . .

40

1.2.4

Pole-Zero Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . .

40

1.2.5

Small-Signal Distortion Analysis . . . . . . . . . . . . . . . . . . . .

41

1.2.6

Sensitivity Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . .

41

1.2.7

Noise Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

42

1.2.8

Periodic Steady State Analysis

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

42

1.3

Analysis at Different Temperatures . . . . . . . . . . . . . . . . . . . . . . . .

42

1.4

Convergence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

44

1.4.1

Voltage convergence criterion . . . . . . . . . . . . . . . . . . . . . .

44

1.4.2

Current convergence criterion . . . . . . . . . . . . . . . . . . . . . .

44

1.4.3

Convergence failure . . . . . . . . . . . . . . . . . . . . . . . . . . .

45

1.2

2

3

Circuit Description

47

2.1

General Structure and Conventions . . . . . . . . . . . . . . . . . . . . . . . .

47

2.1.1

Input file structure . . . . . . . . . . . . . . . . . . . . . . . . . . . .

47

2.1.2

Circuit elements (device instances)

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

47

2.1.3

Some naming conventions . . . . . . . . . . . . . . . . . . . . . . . .

49

5

6

CONTENTS
2.2

Basic lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

50

2.2.1

.TITLE line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

50

2.2.2

.END Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

50

2.2.3

Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51

2.2.4

End-of-line comments . . . . . . . . . . . . . . . . . . . . . . . . . .

51

2.3

.MODEL Device Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51

2.4

.SUBCKT Subcircuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

52

2.4.1

.SUBCKT Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

53

2.4.2

.ENDS Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

54

2.4.3

Subcircuit Calls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

54

2.5

.GLOBAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

54

2.6

.INCLUDE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

55

2.7

.LIB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

55

2.8

.PARAM Parametric netlists . . . . . . . . . . . . . . . . . . . . . . . . . . .

55

2.8.1

.param line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

56

2.8.2

Brace expressions in circuit elements: . . . . . . . . . . . . . . . . . .

56

2.8.3

Subcircuit parameters . . . . . . . . . . . . . . . . . . . . . . . . . . .

57

2.8.4

Symbol scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

58

2.8.5

Syntax of expressions . . . . . . . . . . . . . . . . . . . . . . . . . .

58

2.8.6

Reserved words

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

61

2.8.7

Alternative syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . .

61

.FUNC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

61

2.10 .CSPARAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

62

2.11 .TEMP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

63

2.12 .IF Condition-Controlled Netlist . . . . . . . . . . . . . . . . . . . . . . . . .

63

2.13 Parameters, functions, expressions, and command scripts . . . . . . . . . . . .

65

2.13.1 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

65

2.13.2 Nonlinear sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

65

2.13.3 Control commands, Command scripts . . . . . . . . . . . . . . . . . .

65

2.9

3

Circuit Elements and Models

67

3.1

General options and information . . . . . . . . . . . . . . . . . . . . . . . . .

67

3.1.1

Paralleling devices with multiplier m . . . . . . . . . . . . . . . . . .

67

3.1.2

Technology scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . .

69

3.1.3

Model binning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

69

CONTENTS
3.1.4
3.2

4

7
Initial conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

69

Elementary Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

70

3.2.1

Resistors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

70

3.2.2

Semiconductor Resistors . . . . . . . . . . . . . . . . . . . . . . . . .

71

3.2.3

Semiconductor Resistor Model (R) . . . . . . . . . . . . . . . . . . .

72

3.2.4

Resistors, dependent on expressions (behavioral resistor) . . . . . . . .

73

3.2.5

Capacitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

74

3.2.6

Semiconductor Capacitors . . . . . . . . . . . . . . . . . . . . . . . .

75

3.2.7

Semiconductor Capacitor Model (C) . . . . . . . . . . . . . . . . . . .

75

3.2.8

Capacitors, dependent on expressions (behavioral capacitor) . . . . . .

77

3.2.9

Inductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

78

3.2.10 Inductor model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

78

3.2.11 Coupled (Mutual) Inductors . . . . . . . . . . . . . . . . . . . . . . .

80

3.2.12 Inductors, dependent on expressions (behavioral inductor) . . . . . . .

80

3.2.13 Capacitor or inductor with initial conditions

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

81

3.2.14 Switches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

82

3.2.15 Switch Model (SW/CSW) . . . . . . . . . . . . . . . . . . . . . . . .

83

Voltage and Current Sources

85

4.1

Independent Sources for Voltage or Current . . . . . . . . . . . . . . . . . . .

85

4.1.1

Pulse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

86

4.1.2

Sinusoidal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

87

4.1.3

Exponential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

87

4.1.4

Piece-Wise Linear . . . . . . . . . . . . . . . . . . . . . . . . . . . .

88

4.1.5

Single-Frequency FM . . . . . . . . . . . . . . . . . . . . . . . . . .

88

4.1.6

Amplitude modulated source (AM) . . . . . . . . . . . . . . . . . . .

89

4.1.7

Transient noise source . . . . . . . . . . . . . . . . . . . . . . . . . .

90

4.1.8

Random voltage source . . . . . . . . . . . . . . . . . . . . . . . . . .

91

4.1.9

External voltage or current input . . . . . . . . . . . . . . . . . . . . .

91

4.1.10 Arbitrary Phase Sources . . . . . . . . . . . . . . . . . . . . . . . . .

92

Linear Dependent Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . .

92

4.2.1

Gxxxx: Linear Voltage-Controlled Current Sources (VCCS) . . . . . .

92

4.2.2

Exxxx: Linear Voltage-Controlled Voltage Sources (VCVS) . . . . . .

93

4.2.3

Fxxxx: Linear Current-Controlled Current Sources (CCCS) . . . . . .

93

4.2.4

Hxxxx: Linear Current-Controlled Voltage Sources (CCVS) . . . . . .

93

4.2.5

Polynomial Source Compatibility . . . . . . . . . . . . . . . . . . . .

94

4.2

8
5

CONTENTS
Non-linear Dependent Sources (Behavioral Sources)

95

5.1

Bxxxx: Nonlinear dependent source (ASRC) . . . . . . . . . . . . . . . . . .

95

5.1.1

Syntax and usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

95

5.1.2

Special B-Source Variables time, temper, hertz . . . . . . . . . . . . .

98

5.1.3

par(’expression’) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

99

5.1.4

Piecewise Linear Function: pwl . . . . . . . . . . . . . . . . . . . . .

99

5.2

5.3

5.4
6

Exxxx: non-linear voltage source1 . . . . . . . . . . . . . . . . . . . . . . . . 101
5.2.1

VOL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

5.2.2

VALUE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

5.2.3

TABLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

5.2.4

POLY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

5.2.5

LAPLACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Gxxxx: non-linear current source2 . . . . . . . . . . . . . . . . . . . . . . . . 104
5.3.1

CUR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

5.3.2

VALUE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

5.3.3

TABLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

5.3.4

POLY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

5.3.5

LAPLACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

5.3.6

Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Debugging a behavioral source . . . . . . . . . . . . . . . . . . . . . . . . . . 106

Transmission Lines
6.1

Lossless Transmission Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

6.2

Lossy Transmission Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6.2.1

6.3
6.4

Lossy Transmission Line Model (LTRA) . . . . . . . . . . . . . . . . 110

Uniform Distributed RC Lines . . . . . . . . . . . . . . . . . . . . . . . . . . 112
6.3.1

7

109

Uniform Distributed RC Model (URC) . . . . . . . . . . . . . . . . . 112

KSPICE Lossy Transmission Lines . . . . . . . . . . . . . . . . . . . . . . . . 113
6.4.1

Single Lossy Transmission Line (TXL) . . . . . . . . . . . . . . . . . 113

6.4.2

Coupled Multiconductor Line (CPL) . . . . . . . . . . . . . . . . . . . 114

Diodes

117

7.1

Junction Diodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

7.2

Diode Model (D) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

7.3

Diode Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

1 To get this functionality, the compatibility mode has to be set in spinit or
2 To get this functionality, the compatibility mode has to be set in spinit or

.spiceinit by set ngbehavior=all.
.spiceinit by set ngbehavior=all.

CONTENTS
8

9

BJTs

9
125

8.1

Bipolar Junction Transistors (BJTs) . . . . . . . . . . . . . . . . . . . . . . . 125

8.2

BJT Models (NPN/PNP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

JFETs

131

9.1

Junction Field-Effect Transistors (JFETs) . . . . . . . . . . . . . . . . . . . . 131

9.2

JFET Models (NJF/PJF) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
9.2.1

JFET level 1 model with Parker Skellern modification . . . . . . . . . 131

9.2.2

JFET level 2 Parker Skellern model . . . . . . . . . . . . . . . . . . . 133

10 MESFETs

135

10.1 MESFETs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
10.2 MESFET Models (NMF/PMF) . . . . . . . . . . . . . . . . . . . . . . . . . . 135
10.2.1 Model by Statz e.a. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
10.2.2 Model by Ytterdal e.a. . . . . . . . . . . . . . . . . . . . . . . . . . . 136
10.2.3 hfet1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
10.2.4 hfet2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
11 MOSFETs

137

11.1 MOSFET devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
11.2 MOSFET models (NMOS/PMOS) . . . . . . . . . . . . . . . . . . . . . . . . 138
11.2.1 MOS Level 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
11.2.2 MOS Level 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
11.2.3 MOS Level 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
11.2.4 MOS Level 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
11.2.5 Notes on Level 1-6 models . . . . . . . . . . . . . . . . . . . . . . . . 140
11.2.6 BSIM Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
11.2.7 BSIM1 model (level 4) . . . . . . . . . . . . . . . . . . . . . . . . . . 143
11.2.8 BSIM2 model (level 5) . . . . . . . . . . . . . . . . . . . . . . . . . . 145
11.2.9 BSIM3 model (levels 8, 49) . . . . . . . . . . . . . . . . . . . . . . . 145
11.2.10 BSIM4 model (levels 14, 54) . . . . . . . . . . . . . . . . . . . . . . . 146
11.2.11 EKV model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
11.2.12 BSIMSOI models (levels 10, 58, 55, 56, 57) . . . . . . . . . . . . . . . 147
11.2.13 SOI3 model (level 60) . . . . . . . . . . . . . . . . . . . . . . . . . . 147
11.2.14 HiSIM models of the University of Hiroshima . . . . . . . . . . . . . . 147

10

CONTENTS

12 Mixed-Mode and Behavioral Modeling with XSPICE
12.1 Code Model Element & .MODEL Cards

149

. . . . . . . . . . . . . . . . . . . . 149

12.1.1 Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
12.1.2 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
12.1.3 Search path for file input . . . . . . . . . . . . . . . . . . . . . . . . . 153
12.2 Analog Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
12.2.1 Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
12.2.2 Summer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
12.2.3 Multiplier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
12.2.4 Divider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
12.2.5 Limiter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
12.2.6 Controlled Limiter . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
12.2.7 PWL Controlled Source . . . . . . . . . . . . . . . . . . . . . . . . . 161
12.2.8 Filesource . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
12.2.9 multi_input_pwl block . . . . . . . . . . . . . . . . . . . . . . . . . . 165
12.2.10 Analog Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
12.2.11 Zener Diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
12.2.12 Current Limiter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
12.2.13 Hysteresis Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
12.2.14 Differentiator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
12.2.15 Integrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
12.2.16 S-Domain Transfer Function . . . . . . . . . . . . . . . . . . . . . . . 176
12.2.17 Slew Rate Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
12.2.18 Inductive Coupling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
12.2.19 Magnetic Core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
12.2.20 Controlled Sine Wave Oscillator . . . . . . . . . . . . . . . . . . . . . 184
12.2.21 Controlled Triangle Wave Oscillator . . . . . . . . . . . . . . . . . . . 185
12.2.22 Controlled Square Wave Oscillator . . . . . . . . . . . . . . . . . . . . 186
12.2.23 Controlled One-Shot . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
12.2.24 Capacitance Meter . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
12.2.25 Inductance Meter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
12.2.26 Memristor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
12.2.27 2D table model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
12.2.28 3D table model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
12.3 Hybrid Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

CONTENTS

11

12.3.1 Digital-to-Analog Node Bridge . . . . . . . . . . . . . . . . . . . . . 197
12.3.2 Analog-to-Digital Node Bridge . . . . . . . . . . . . . . . . . . . . . 198
12.3.3 Controlled Digital Oscillator . . . . . . . . . . . . . . . . . . . . . . . 200
12.3.4 Node bridge from digital to real with enable . . . . . . . . . . . . . . . 201
12.3.5 A Z**-1 block working on real data . . . . . . . . . . . . . . . . . . . 201
12.3.6 A gain block for event-driven real data . . . . . . . . . . . . . . . . . . 202
12.3.7 Node bridge from real to analog voltage . . . . . . . . . . . . . . . . . 203
12.4 Digital Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
12.4.1 Buffer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
12.4.2 Inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
12.4.3 And . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
12.4.4 Nand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
12.4.5 Or . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
12.4.6 Nor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
12.4.7 Xor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
12.4.8 Xnor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
12.4.9 Tristate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
12.4.10 Pullup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
12.4.11 Pulldown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
12.4.12 D Flip Flop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
12.4.13 JK Flip Flop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
12.4.14 Toggle Flip Flop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
12.4.15 Set-Reset Flip Flop . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
12.4.16 D Latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
12.4.17 Set-Reset Latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
12.4.18 State Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
12.4.19 Frequency Divider . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
12.4.20 RAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
12.4.21 Digital Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
12.4.22 LUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
12.4.23 General LUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
12.5 Predefined Node Types for event driven simulation . . . . . . . . . . . . . . . 239
12.5.1 Digital Node Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
12.5.2 Real Node Type

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239

12.5.3 Int Node Type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
12.5.4 (Digital) Input/Output . . . . . . . . . . . . . . . . . . . . . . . . . . 239

12

CONTENTS

13 Verilog A Device models

241

13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
13.2 adms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
13.3 How to integrate a Verilog-A model into ngspice . . . . . . . . . . . . . . . . 241
13.3.1 How to setup a *.va model for ngspice . . . . . . . . . . . . . . . . . . 241
13.3.2 Adding admsXml to your build environment . . . . . . . . . . . . . . 241
14 Mixed-Level Simulation (ngspice with TCAD)

243

14.1 Cider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
14.2 GSS, Genius . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
15 Analyses and Output Control (batch mode)

245

15.1 Simulator Variables (.options) . . . . . . . . . . . . . . . . . . . . . . . . . . 245
15.1.1 General Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
15.1.2 DC Solution Options . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
15.1.3 AC Solution Options . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
15.1.4 Transient Analysis Options . . . . . . . . . . . . . . . . . . . . . . . . 248
15.1.5 ELEMENT Specific options . . . . . . . . . . . . . . . . . . . . . . . 249
15.1.6 Transmission Lines Specific Options . . . . . . . . . . . . . . . . . . . 250
15.1.7 Precedence of option and .options commands . . . . . . . . . . . . . . 250
15.2 Initial Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
15.2.1 .NODESET: Specify Initial Node Voltage Guesses . . . . . . . . . . . 250
15.2.2 .IC: Set Initial Conditions . . . . . . . . . . . . . . . . . . . . . . . . 251
15.3 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
15.3.1 .AC: Small-Signal AC Analysis . . . . . . . . . . . . . . . . . . . . . 252
15.3.2 .DC: DC Transfer Function . . . . . . . . . . . . . . . . . . . . . . . . 253
15.3.3 .DISTO: Distortion Analysis . . . . . . . . . . . . . . . . . . . . . . . 253
15.3.4 .NOISE: Noise Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 255
15.3.5 .OP: Operating Point Analysis . . . . . . . . . . . . . . . . . . . . . . 255
15.3.6 .PZ: Pole-Zero Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 256
15.3.7 .SENS: DC or Small-Signal AC Sensitivity Analysis . . . . . . . . . . 257
15.3.8 .TF: Transfer Function Analysis . . . . . . . . . . . . . . . . . . . . . 257
15.3.9 .TRAN: Transient Analysis . . . . . . . . . . . . . . . . . . . . . . . . 258
15.3.10 Transient noise analysis (at low frequency) . . . . . . . . . . . . . . . 258
15.3.11 .PSS: Periodic Steady State Analysis . . . . . . . . . . . . . . . . . . 262
15.4 Measurements after AC, DC and Transient Analysis . . . . . . . . . . . . . . . 263

CONTENTS

13

15.4.1 .meas(ure) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
15.4.2 batch versus interactive mode . . . . . . . . . . . . . . . . . . . . . . 263
15.4.3 General remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
15.4.4 Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
15.4.5 Trig Targ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
15.4.6 Find ... When . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
15.4.7 AVG|MIN|MAX|PP|RMS|MIN_AT|MAX_AT . . . . . . . . . . . . . . . . 267
15.4.8 Integ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
15.4.9 param . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
15.4.10 par(’expression’) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
15.4.11 Deriv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
15.4.12 More examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
15.5 Safe Operating Area (SOA) warning messages . . . . . . . . . . . . . . . . . . 270
15.5.1 Resistor and Capacitor SOA model parameters . . . . . . . . . . . . . 271
15.5.2 Diode SOA model parameter . . . . . . . . . . . . . . . . . . . . . . . 271
15.5.3 BJT SOA model parameter . . . . . . . . . . . . . . . . . . . . . . . . 271
15.5.4 MOS SOA model parameter . . . . . . . . . . . . . . . . . . . . . . . 271
15.6 Batch Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
15.6.1 .SAVE: Name vector(s) to be saved in raw file . . . . . . . . . . . . . . 272
15.6.2 .PRINT Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
15.6.3 .PLOT Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
15.6.4 .FOUR: Fourier Analysis of Transient Analysis Output . . . . . . . . . 274
15.6.5 .PROBE: Name vector(s) to be saved in raw file . . . . . . . . . . . . . 275
15.6.6 par(’expression’): Algebraic expressions for output . . . . . . . . . . . 275
15.6.7 .width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
15.7 Measuring current through device terminals . . . . . . . . . . . . . . . . . . . 276
15.7.1 Adding a voltage source in series . . . . . . . . . . . . . . . . . . . . 276
15.7.2 Using option ’savecurrents’ . . . . . . . . . . . . . . . . . . . . . . . 276
16 Starting ngspice

279

16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
16.2 Where to obtain ngspice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
16.3 Command line options for starting ngspice and ngnutmeg . . . . . . . . . . . . 280
16.4 Starting options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
16.4.1 Batch mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

14

CONTENTS
16.4.2 Interactive mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
16.4.3 Control mode (Interactive mode with control file or control section) . . 283
16.5 Standard configuration file spinit . . . . . . . . . . . . . . . . . . . . . . . . . 284
16.6 User defined configuration file .spiceinit . . . . . . . . . . . . . . . . . . . . . 285
16.7 Environmental variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
16.7.1 Ngspice specific variables . . . . . . . . . . . . . . . . . . . . . . . . 285
16.7.2 Common environment variables . . . . . . . . . . . . . . . . . . . . . 286
16.8 Memory usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
16.9 Simulation time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
16.10Ngspice on multi-core processors using OpenMP . . . . . . . . . . . . . . . . 287
16.10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
16.10.2 Some results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
16.10.3 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
16.10.4 Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
16.11Server mode option -s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
16.12Ngspice control via input, output fifos . . . . . . . . . . . . . . . . . . . . . . 291
16.13Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
16.13.1 Compatibility mode . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
16.13.2 Missing functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
16.13.3 Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
16.13.4 Controls and commands . . . . . . . . . . . . . . . . . . . . . . . . . 294
16.14Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
16.15Reporting bugs and errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296

17 Interactive Interpreter

297

17.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
17.2 Expressions, Functions, and Constants . . . . . . . . . . . . . . . . . . . . . . 298
17.3 Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
17.4 Command Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
17.4.1 On the console . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
17.4.2 Scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
17.4.3 Add-on to circuit file . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
17.5 Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
17.5.1 Ac*: Perform an AC, small-signal frequency response analysis . . . . . 304
17.5.2 Alias: Create an alias for a command . . . . . . . . . . . . . . . . . . 305

CONTENTS

15

17.5.3 Alter*: Change a device or model parameter . . . . . . . . . . . . . . 305
17.5.4 Altermod*: Change model parameter(s)

. . . . . . . . . . . . . . . . 306

17.5.5 Asciiplot: Plot values using old-style character plots . . . . . . . . . . 308
17.5.6 Aspice*: Asynchronous ngspice run . . . . . . . . . . . . . . . . . . . 308
17.5.7 Bug: Mail a bug report . . . . . . . . . . . . . . . . . . . . . . . . . . 308
17.5.8 Cd: Change directory . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
17.5.9 Cdump: Dump the control flow to the screen . . . . . . . . . . . . . . 309
17.5.10 Circbyline*: Enter a circuit line by line . . . . . . . . . . . . . . . . . 309
17.5.11 Codemodel*: Load an XSPICE code model library . . . . . . . . . . . 310
17.5.12 Compose: Compose a vector . . . . . . . . . . . . . . . . . . . . . . . 310
17.5.13 Dc*: Perform a DC-sweep analysis . . . . . . . . . . . . . . . . . . . 311
17.5.14 Define: Define a function . . . . . . . . . . . . . . . . . . . . . . . . . 311
17.5.15 Deftype: Define a new type for a vector or plot . . . . . . . . . . . . . 311
17.5.16 Delete*: Remove a trace or breakpoint . . . . . . . . . . . . . . . . . . 312
17.5.17 Destroy: Delete an output data set . . . . . . . . . . . . . . . . . . . . 312
17.5.18 Devhelp: information on available devices . . . . . . . . . . . . . . . . 312
17.5.19 Diff: Compare vectors . . . . . . . . . . . . . . . . . . . . . . . . . . 313
17.5.20 Display: List known vectors and types . . . . . . . . . . . . . . . . . . 313
17.5.21 Echo: Print text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
17.5.22 Edit*: Edit the current circuit . . . . . . . . . . . . . . . . . . . . . . 313
17.5.23 Edisplay: Print a list of all the event nodes

. . . . . . . . . . . . . . . 314

17.5.24 Eprint: Print an event driven node . . . . . . . . . . . . . . . . . . . . 314
17.5.25 Eprvcd: Dump event nodes in VCD format . . . . . . . . . . . . . . . 314
17.5.26 FFT: fast Fourier transform of vectors . . . . . . . . . . . . . . . . . . 314
17.5.27 Fourier: Perform a Fourier transform . . . . . . . . . . . . . . . . . . 316
17.5.28 Gnuplot: Graphics output via gnuplot . . . . . . . . . . . . . . . . . . 317
17.5.29 Hardcopy: Save a plot to a file for printing . . . . . . . . . . . . . . . 317
17.5.30 Help: Print summaries of Ngspice commands

. . . . . . . . . . . . . 318

17.5.31 History: Review previous commands . . . . . . . . . . . . . . . . . . 318
17.5.32 Inventory: Print circuit inventory . . . . . . . . . . . . . . . . . . . . . 321
17.5.33 Iplot*: Incremental plot . . . . . . . . . . . . . . . . . . . . . . . . . 321
17.5.34 Jobs*: List active asynchronous ngspice runs . . . . . . . . . . . . . . 321
17.5.35 Let: Assign a value to a vector . . . . . . . . . . . . . . . . . . . . . . 321
17.5.36 Linearize*: Interpolate to a linear scale . . . . . . . . . . . . . . . . . 322
17.5.37 Listing*: Print a listing of the current circuit . . . . . . . . . . . . . . 323

16

CONTENTS
17.5.38 Load: Load rawfile data . . . . . . . . . . . . . . . . . . . . . . . . . 323
17.5.39 Meas*: Measurements on simulation data . . . . . . . . . . . . . . . . 323
17.5.40 Mdump*: Dump the matrix values to a file (or to console) . . . . . . . 324
17.5.41 Mrdump*: Dump the matrix right hand side values to a file (or to console)324
17.5.42 Noise*: Noise analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 324
17.5.43 Op*: Perform an operating point analysis . . . . . . . . . . . . . . . . 325
17.5.44 Option*: Set a ngspice option . . . . . . . . . . . . . . . . . . . . . . 325
17.5.45 Plot: Plot vectors on the display . . . . . . . . . . . . . . . . . . . . . 326
17.5.46 Pre_: execute commands prior to parsing the circuit . . . . 327
17.5.47 Print: Print values . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328
17.5.48 Quit: Leave Ngspice or Nutmeg . . . . . . . . . . . . . . . . . . . . . 328
17.5.49 Rehash: Reset internal hash tables . . . . . . . . . . . . . . . . . . . . 329
17.5.50 Remcirc*: Remove the current circuit . . . . . . . . . . . . . . . . . . 329
17.5.51 Reset*: Reset an analysis . . . . . . . . . . . . . . . . . . . . . . . . . 329
17.5.52 Reshape: Alter the dimensionality or dimensions of a vector . . . . . . 329
17.5.53 Resume*: Continue a simulation after a stop . . . . . . . . . . . . . . 330
17.5.54 Rspice*: Remote ngspice submission . . . . . . . . . . . . . . . . . . 330
17.5.55 Run*: Run analysis from the input file . . . . . . . . . . . . . . . . . . 330
17.5.56 Rusage: Resource usage . . . . . . . . . . . . . . . . . . . . . . . . . 331
17.5.57 Save*: Save a set of outputs . . . . . . . . . . . . . . . . . . . . . . . 332
17.5.58 Sens*: Run a sensitivity analysis . . . . . . . . . . . . . . . . . . . . . 333
17.5.59 Set: Set the value of a variable . . . . . . . . . . . . . . . . . . . . . . 333
17.5.60 Setcirc*: Change the current circuit . . . . . . . . . . . . . . . . . . . 333
17.5.61 Setplot: Switch the current set of vectors . . . . . . . . . . . . . . . . 334
17.5.62 Setscale: Set the scale vector for the current plot . . . . . . . . . . . . 334
17.5.63 Settype: Set the type of a vector . . . . . . . . . . . . . . . . . . . . . 334
17.5.64 Shell: Call the command interpreter . . . . . . . . . . . . . . . . . . . 335
17.5.65 Shift: Alter a list variable . . . . . . . . . . . . . . . . . . . . . . . . . 335
17.5.66 Show*: List device state . . . . . . . . . . . . . . . . . . . . . . . . . 335
17.5.67 Showmod*: List model parameter values . . . . . . . . . . . . . . . . 335
17.5.68 Snload*: Load the snapshot file . . . . . . . . . . . . . . . . . . . . . 336
17.5.69 Snsave*: Save a snapshot file . . . . . . . . . . . . . . . . . . . . . . 336
17.5.70 Source: Read a ngspice input file . . . . . . . . . . . . . . . . . . . . 337
17.5.71 Spec: Create a frequency domain plot . . . . . . . . . . . . . . . . . . 338
17.5.72 Status*: Display breakpoint information . . . . . . . . . . . . . . . . . 338

CONTENTS

17

17.5.73 Step*: Run a fixed number of time-points . . . . . . . . . . . . . . . . 338
17.5.74 Stop*: Set a breakpoint . . . . . . . . . . . . . . . . . . . . . . . . . . 339
17.5.75 Strcmp: Compare two strings . . . . . . . . . . . . . . . . . . . . . . 339
17.5.76 Sysinfo*: Print system information . . . . . . . . . . . . . . . . . . . 340
17.5.77 Tf*: Run a Transfer Function analysis . . . . . . . . . . . . . . . . . . 340
17.5.78 Trace*: Trace nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
17.5.79 Tran*: Perform a transient analysis . . . . . . . . . . . . . . . . . . . 341
17.5.80 Transpose: Swap the elements in a multi-dimensional data set . . . . . 342
17.5.81 Unalias: Retract an alias . . . . . . . . . . . . . . . . . . . . . . . . . 342
17.5.82 Undefine: Retract a definition . . . . . . . . . . . . . . . . . . . . . . 342
17.5.83 Unlet: Delete the specified vector(s) . . . . . . . . . . . . . . . . . . . 342
17.5.84 Unset: Clear a variable . . . . . . . . . . . . . . . . . . . . . . . . . . 343
17.5.85 Version: Print the version of ngspice . . . . . . . . . . . . . . . . . . . 343
17.5.86 Where*: Identify troublesome node or device . . . . . . . . . . . . . . 344
17.5.87 Wrdata: Write data to a file (simple table) . . . . . . . . . . . . . . . . 345
17.5.88 Write: Write data to a file (Spice3f5 format) . . . . . . . . . . . . . . . 345
17.5.89 Wrs2p: Write scattering parameters to file (Touchstone® format) . . . 346
17.5.90 Xgraph: use the xgraph(1) program for plotting. . . . . . . . . . . . . 346
17.6 Control Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
17.6.1 While - End . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
17.6.2 Repeat - End . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
17.6.3 Dowhile - End . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
17.6.4 Foreach - End . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
17.6.5 If - Then - Else . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
17.6.6 Label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
17.6.7 Goto

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348

17.6.8 Continue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
17.6.9 Break . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
17.7 Internally predefined variables . . . . . . . . . . . . . . . . . . . . . . . . . . 349
17.8 Scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
17.8.1 Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
17.8.2 Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
17.8.3 Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
17.8.4 control structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
17.8.5 Example script ’spectrum’ . . . . . . . . . . . . . . . . . . . . . . . . 359

18

CONTENTS
17.8.6 Example script for random numbers . . . . . . . . . . . . . . . . . . . 361
17.8.7 Parameter sweep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
17.8.8 Output redirection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
17.9 Scattering parameters (s-parameters) . . . . . . . . . . . . . . . . . . . . . . . 364
17.9.1 Intro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
17.9.2 S-parameter measurement basics . . . . . . . . . . . . . . . . . . . . . 364
17.9.3 Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
17.10MISCELLANEOUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
17.11Bugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367

18 Ngspice User Interfaces

369

18.1 MS Windows Graphical User Interface . . . . . . . . . . . . . . . . . . . . . . 369
18.2 MS Windows Console . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
18.3 Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
18.4 CygWin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
18.5 Error handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
18.6 Postscript printing options . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
18.7

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374

18.8 Integration with CAD software and ‘third party’ GUIs . . . . . . . . . . . . . . 374
18.8.1 KJWaves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
18.8.2 GNU Spice GUI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
18.8.3 XCircuit

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374

18.8.4 GEDA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
18.8.5 CppSim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
18.8.6 NGSPICE Online . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
18.8.7 Spicy Schematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
18.8.8 MSEspice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
18.8.9 PartSim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
19 ngspice as shared library or dynamic link library

377

19.1 Compile options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
19.1.1 How to get the sources . . . . . . . . . . . . . . . . . . . . . . . . . . 377
19.1.2 Linux, MINGW, CYGWIN . . . . . . . . . . . . . . . . . . . . . . . 377
19.1.3 MS Visual Studio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377
19.2 Linking shared ngspice to a calling application . . . . . . . . . . . . . . . . . 378
19.2.1 Linking during creating the caller . . . . . . . . . . . . . . . . . . . . 378

CONTENTS

19

19.2.2 Loading at runtime . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
19.3 Shared ngspice API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
19.3.1 structs and types defined for transporting data . . . . . . . . . . . . . . 378
19.3.2 Exported functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
19.3.3 Callback functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
19.4 General remarks on using the API . . . . . . . . . . . . . . . . . . . . . . . . 384
19.4.1 Loading a netlist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
19.4.2 Running the simulation . . . . . . . . . . . . . . . . . . . . . . . . . . 385
19.4.3 Accessing data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
19.4.4 Altering model or device parameters . . . . . . . . . . . . . . . . . . . 386
19.4.5 Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
19.4.6 Error handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
19.5 Example applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
19.6 ngspice parallel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
19.6.1 Go parallel! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
19.6.2 Additional exported functions . . . . . . . . . . . . . . . . . . . . . . 389
19.6.3 Additional callback functions . . . . . . . . . . . . . . . . . . . . . . 390
19.6.4 Parallel ngspice example . . . . . . . . . . . . . . . . . . . . . . . . . 391
20 TCLspice

393

20.1 tclspice framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
20.2 tclspice documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
20.3 spicetoblt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
20.4 Running TCLspice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
20.5 examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
20.5.1 Active capacitor measurement . . . . . . . . . . . . . . . . . . . . . . 394
20.5.2 Optimization of a linearization circuit for a Thermistor . . . . . . . . . 397
20.5.3 Progressive display . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401
20.6 Compiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
20.6.1 Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
20.6.2 MS Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
20.7 MS Windows 32 Bit binaries . . . . . . . . . . . . . . . . . . . . . . . . . . . 403

20

CONTENTS

21 Example Circuits

405

21.1 AC coupled transistor amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . 405
21.2 Differential Pair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
21.3 MOSFET Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
21.4 RTL Inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
21.5 Four-Bit Binary Adder (Bipolar) . . . . . . . . . . . . . . . . . . . . . . . . . 412
21.6 Four-Bit Binary Adder (MOS) . . . . . . . . . . . . . . . . . . . . . . . . . . 414
21.7 Transmission-Line Inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415
22 Statistical circuit analysis

417

22.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
22.2 Using random param(eters) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
22.3 Behavioral sources (B, E, G, R, L, C) with random control . . . . . . . . . . . 419
22.4 ngspice scripting language . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
22.5 Monte-Carlo Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
22.5.1 Example 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
22.5.2 Example 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
22.5.3 Example 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
22.5.4 Example 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
22.6 Data evaluation with . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
23 Circuit optimization with ngspice

427

23.1 Optimization of a circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
23.2 ngspice optimizer using ngspice scripts . . . . . . . . . . . . . . . . . . . . . 428
23.3 ngspice optimizer using tclspice . . . . . . . . . . . . . . . . . . . . . . . . . 428
23.4 ngspice optimizer using a Python script . . . . . . . . . . . . . . . . . . . . . 428
23.5 ngspice optimizer using ASCO . . . . . . . . . . . . . . . . . . . . . . . . . . 428
23.5.1 Three stage operational amplifier . . . . . . . . . . . . . . . . . . . . . 429
23.5.2 Digital inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 430
23.5.3 Bandpass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
23.5.4 Class-E power amplifier . . . . . . . . . . . . . . . . . . . . . . . . . 433
24 Notes

435

24.1 Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
24.2 Acronyms and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . 436

CONTENTS

II

21

XSPICE Software User’s Manual

25 XSPICE Basics

439
441

25.1 ngspice with the XSPICE option . . . . . . . . . . . . . . . . . . . . . . . . . 441
25.2 The XSPICE Code Model Subsystem . . . . . . . . . . . . . . . . . . . . . . 441
25.3 XSPICE Top-Level Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
26 Execution Procedures

443

26.1 Simulation and Modeling Overview . . . . . . . . . . . . . . . . . . . . . . . 443
26.1.1 Describing the Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . 443
26.2 Circuit Description Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
26.2.1 XSPICE Syntax Extensions . . . . . . . . . . . . . . . . . . . . . . . 449
26.3 How to create code models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451
27 Example circuits

455

27.1 Amplifier with XSPICE model ‘gain’ . . . . . . . . . . . . . . . . . . . . . . 455
27.2 XSPICE advanced usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
27.2.1 Circuit example C3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
27.2.2 Running example C3 . . . . . . . . . . . . . . . . . . . . . . . . . . . 460
28 Code Models and User-Defined Nodes

465

28.1 Code Model Data Type Definitions . . . . . . . . . . . . . . . . . . . . . . . . 466
28.2 Creating Code Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
28.3 Creating User-Defined Nodes . . . . . . . . . . . . . . . . . . . . . . . . . . . 467
28.4 Adding a new code model library . . . . . . . . . . . . . . . . . . . . . . . . . 468
28.5 Compiling and loading the new code model (library) . . . . . . . . . . . . . . 468
28.6 Interface Specification File . . . . . . . . . . . . . . . . . . . . . . . . . . . . 469
28.6.1 The Name Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
28.6.2 The Port Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
28.6.3 The Parameter Table . . . . . . . . . . . . . . . . . . . . . . . . . . . 473
28.6.4 Static Variable Table . . . . . . . . . . . . . . . . . . . . . . . . . . . 474
28.7 Model Definition File

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 476

28.7.1 Macros . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 476
28.7.2 Function Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485
28.8 User-Defined Node Definition File . . . . . . . . . . . . . . . . . . . . . . . . 492
28.8.1 Macros . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493
28.8.2 Function Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493
28.8.3 Example UDN Definition File . . . . . . . . . . . . . . . . . . . . . . 496

22

CONTENTS

29 Error Messages

501

29.1 Preprocessor Error Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . 501
29.2 Simulator Error Messages

. . . . . . . . . . . . . . . . . . . . . . . . . . . . 506

29.3 Code Model Error Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . 507
29.3.1 Code Model aswitch . . . . . . . . . . . . . . . . . . . . . . . . . . . 507
29.3.2 Code Model climit . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508
29.3.3 Code Model core . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508
29.3.4 Code Model d_osc . . . . . . . . . . . . . . . . . . . . . . . . . . . . 508
29.3.5 Code Model d_source . . . . . . . . . . . . . . . . . . . . . . . . . . 509
29.3.6 Code Model d_state . . . . . . . . . . . . . . . . . . . . . . . . . . . 509
29.3.7 Code Model oneshot . . . . . . . . . . . . . . . . . . . . . . . . . . . 510
29.3.8 Code Model pwl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 510
29.3.9 Code Model s_xfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . 510
29.3.10 Code Model sine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 511
29.3.11 Code Model square

. . . . . . . . . . . . . . . . . . . . . . . . . . . 511

29.3.12 Code Model triangle . . . . . . . . . . . . . . . . . . . . . . . . . . . 512

III

CIDER

30 CIDER User’s Manual

513
515

30.1 SPECIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 515
30.1.1 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 516
30.2 BOUNDARY, INTERFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . 517
30.2.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517

30.2.2 PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518
30.2.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518
30.3 COMMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 518
30.3.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519

30.3.2 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
30.4 CONTACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
30.4.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519

30.4.2 PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
30.4.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 519
30.4.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
30.5 DOMAIN, REGION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520

CONTENTS

23

30.5.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520

30.5.2 PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
30.5.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
30.5.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521
30.6 DOPING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521
30.6.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 521

30.6.2 PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 524
30.6.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 524
30.6.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525
30.7 ELECTRODE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525
30.7.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525

30.7.2 PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
30.7.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
30.7.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526
30.8 END

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526

30.8.1 DESCRIPTION
30.9 MATERIAL

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527

30.9.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 527

30.9.2 PARAMETERS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528
30.9.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528
30.9.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 528
30.10METHOD

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529

30.10.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529

30.10.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529
30.10.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 529
30.11Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 530
30.11.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 530
30.11.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 531
30.11.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 531
30.11.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 531
30.11.5 BUGS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532

30.12MODELS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532
30.12.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532

30.12.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532
30.12.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 532

24

CONTENTS
30.12.4 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533
30.12.5 Bugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533
30.13OPTIONS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533

30.13.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533

30.13.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
30.13.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
30.13.4 See also . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 534
30.14OUTPUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535
30.14.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535

30.14.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 536
30.14.3 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 536
30.14.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
30.15TITLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
30.15.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537

30.15.2 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
30.15.3 BUGS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537

30.16X.MESH, Y.MESH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
30.16.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 538

30.16.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539
30.16.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539
30.16.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 539
30.17NUMD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540
30.17.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 540

30.17.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541
30.17.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541
30.17.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542
30.17.5 BUGS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542

30.18NBJT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542
30.18.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 542

30.18.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543
30.18.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543
30.18.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 544
30.18.5 BUGS
30.19NUMOS

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 544

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 544

30.19.1 DESCRIPTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 544

CONTENTS

25

30.19.2 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545
30.19.3 EXAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545
30.19.4 SEE ALSO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546
30.20Cider examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 546

IV

Appendices

31 Model and Device Parameters

547
549

31.1 Accessing internal device parameters . . . . . . . . . . . . . . . . . . . . . . . 549
31.2 Elementary Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551
31.2.1 Resistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551
31.2.2 Capacitor - Fixed capacitor . . . . . . . . . . . . . . . . . . . . . . . . 553
31.2.3 Inductor - Fixed inductor . . . . . . . . . . . . . . . . . . . . . . . . . 554
31.2.4 Mutual - Mutual Inductor . . . . . . . . . . . . . . . . . . . . . . . . . 555
31.3 Voltage and current sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . 556
31.3.1 ASRC - Arbitrary source . . . . . . . . . . . . . . . . . . . . . . . . . 556
31.3.2 Isource - Independent current source . . . . . . . . . . . . . . . . . . . 557
31.3.3 Vsource - Independent voltage source . . . . . . . . . . . . . . . . . . 558
31.3.4 CCCS - Current controlled current source . . . . . . . . . . . . . . . . 559
31.3.5 CCVS - Current controlled voltage source . . . . . . . . . . . . . . . . 559
31.3.6 VCCS - Voltage controlled current source . . . . . . . . . . . . . . . . 560
31.3.7 VCVS - Voltage controlled voltage source . . . . . . . . . . . . . . . . 560
31.4 Transmission Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561
31.4.1 CplLines - Simple Coupled Multiconductor Lines . . . . . . . . . . . . 561
31.4.2 LTRA - Lossy transmission line . . . . . . . . . . . . . . . . . . . . . 562
31.4.3 Tranline - Lossless transmission line . . . . . . . . . . . . . . . . . . . 563
31.4.4 TransLine - Simple Lossy Transmission Line . . . . . . . . . . . . . . 564
31.4.5 URC - Uniform R. C. line . . . . . . . . . . . . . . . . . . . . . . . . 565
31.5 BJTs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 566
31.5.1 BJT - Bipolar Junction Transistor . . . . . . . . . . . . . . . . . . . . 566
31.5.2 BJT - Bipolar Junction Transistor Level 2 . . . . . . . . . . . . . . . . 569
31.5.3 VBIC - Vertical Bipolar Inter-Company Model . . . . . . . . . . . . . 572
31.6 MOSFETs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 576
31.6.1 MOS1 - Level 1 MOSFET model with Meyer capacitance model . . . . 576
31.6.2 MOS2 - Level 2 MOSFET model with Meyer capacitance model . . . . 579

26

CONTENTS
31.6.3 MOS3 - Level 3 MOSFET model with Meyer capacitance model

. . . 583

31.6.4 MOS6 - Level 6 MOSFET model with Meyer capacitance model

. . . 587

31.6.5 MOS9 - Modified Level 3 MOSFET model . . . . . . . . . . . . . . . 590
31.6.6 BSIM1 - Berkeley Short Channel IGFET Model . . . . . . . . . . . . 594
31.6.7 BSIM2 - Berkeley Short Channel IGFET Model . . . . . . . . . . . . 597
31.6.8 BSIM3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 601
31.6.9 BSIM4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602
32 Compilation notes

605

32.1 Ngspice Installation under Linux (and other ’UNIXes’) . . . . . . . . . . . . . 605
32.1.1 Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 605
32.1.2 Install from Git . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 605
32.1.3 Install from a tarball, e.g. ngspice-rework-27.tgz . . . . . . . . . . . . 607
32.1.4 Compilation using an user defined directory tree for object files . . . . 607
32.1.5 Advanced Install . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607
32.1.6 Compilers and Options . . . . . . . . . . . . . . . . . . . . . . . . . . 609
32.1.7 Compiling For Multiple Architectures . . . . . . . . . . . . . . . . . . 610
32.1.8 Installation Names . . . . . . . . . . . . . . . . . . . . . . . . . . . . 610
32.1.9 Optional Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 610
32.1.10 Specifying the System Type . . . . . . . . . . . . . . . . . . . . . . . 611
32.1.11 Sharing Defaults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 611
32.1.12 Operation Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . 611
32.2 Ngspice Compilation under Windows OS . . . . . . . . . . . . . . . . . . . . 612
32.2.1 Compile ngspice with MS Visual Studio 2015 or 2017 . . . . . . . . . 612
32.2.2 How to make ngspice with MINGW and MSYS

. . . . . . . . . . . . 614

32.2.3 64 Bit executables with MINGW-w64 . . . . . . . . . . . . . . . . . . 616
32.2.4 make ngspice with pure CYGWIN . . . . . . . . . . . . . . . . . . . . 618
32.2.5 ngspice mingw or cygwin console executable w/o graphics . . . . . . . 618
32.2.6 ngspice for MS Windows, cross compiled from Linux . . . . . . . . . 618
32.2.7 make ngspice with CYGWIN and external MINGW32 . . . . . . . . . 619
32.2.8 make ngspice with CYGWIN and internal MINGW32 (use config.h
made above) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619
32.3 Reporting errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619

CONTENTS
33 Copyrights and licenses

27
621

33.1 Documentation license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621
33.1.1 Spice documentation copyright . . . . . . . . . . . . . . . . . . . . . . 621
33.1.2 XSPICE SOFTWARE (documentation) copyright . . . . . . . . . . . . 621
33.1.3 CIDER RESEARCH SOFTWARE AGREEMENT (superseded by 33.2.1)622
33.2 ngspice license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 622
33.2.1 ‘Modified’ BSD license . . . . . . . . . . . . . . . . . . . . . . . . . 623
33.2.2 XSPICE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 624
33.2.3 tclspice, numparam . . . . . . . . . . . . . . . . . . . . . . . . . . . . 624
33.2.4 Linking to GPLd libraries (e.g. readline, fftw, table.cm): . . . . . . . . 624

28

CONTENTS

Prefaces
Preface to the first edition
This manual has been assembled from different sources:
1. The spice3f5 manual,
2. the XSPICE user’s manual,
3. the CIDER user’s manual
and some original material needed to describe the new features and the newly implemented
models. This cut and paste approach, while not being orthodox, allowed ngspice to have a full
manual in a fraction of the time that writing a completely new text would have required. The use
of LaTex and Lyx instead of TeXinfo, which was the original encoding for the manual, further
helped to reduce the writing effort and improved the quality of the result, at the expense of an
on-line version of the manual but, due to the complexity of the software I hardly think that users
will ever want to read an on-line text version.
In writing this text I followed the spice3f5 manual, both in the chapter sequence and presentation
of material, mostly because that was already the user manual of SPICE.
Ngspice is an open source software, users can download the source code, compile, and run it.
This manual has an entire chapter describing program compilation and available options to help
users in building ngspice (see Chapt. 32). The source package already comes with all ‘safe’
options enabled by default, and activating the others can produce unpredictable results and thus
is recommended to expert users only. This is the first ngspice manual and I have removed all
the historical material that described the differences between ngspice and spice3, since it was
of no use for the user and not so useful for the developer who can look for it in the Changelogs
of in the revision control system.
I want to acknowledge the work done by Emmanuel Rouat and Arno W. Peters for converting the
original spice3f documentation to TEXinfo. Their effort gave ngspice users the only available
documentation that described the changes for many years. A good source of ideas for this
manual came from the on-line spice3f manual written by Charles D.H. Williams (Spice3f5
User Guide), constantly updated and useful for its many insights.
As always, errors, omissions and unreadable phrases are only my fault.
Paolo Nenzi
Roma, March 24th 2001

29

30

CONTENTS
Indeed. At the end of the day, this is engineering, and one learns to live
within the limitations of the tools.

Kevin Aylward, Warden of the King’s Ale

Preface to the actual edition (as of January 2013)
Due to the wealth of new material and options in ngspice the actual order of chapters has been
revised. Several new chapters have been added. The LYX text processor has allowed adding
internal cross references. The PDF format has become the standard format for distribution of
the manual. Within each new ngspice distribution (starting with ngspice-21) a manual edition
is provided reflecting the ngspice status at the time of distribution. At the same time, located
at ngspice manuals, the manual is constantly updated. Every new ngspice feature should enter
this manual as soon as it has been made available in the Git source code.
Holger Vogt
Mülheim, 2013

Acknowledgments
ngspice contributors
Spice was originally written at The University of California at Berkeley (USA).
XSPICE has been provided by Georgia Institute of Technology, Atlanta (USA).
Since then, there have been many people working on the software, most of them releasing
patches to the original code through the Internet.
The following people have contributed in some way:
Vera Albrecht,
Cecil Aswell,
Giles C. Billingsley,
Phil Barker,
Steven Borley,
Stuart Brorson,
Mansun Chan,
Wayne A. Christopher,
Al Davis,
Glao S. Dezai,
Jon Engelbert,
Daniele Foci,
Noah Friedman,
David A. Gates,
Alan Gillespie,
John Heidemann,
Marcel Hendrix,
Jeffrey M. Hsu,
JianHui Huang,
S. Hwang,
Chris Inbody,
Gordon M. Jacobs,
Min-Chie Jeng,
Beorn Johnson,
Stefan Jones,
Kenneth H. Keller,
Francesco Lannutti,
Robert Larice,

31

32

CONTENTS
Mathew Lew,
Robert Lindsell,
Weidong Liu,
Kartikeya Mayaram,
Richard D. McRoberts,
Manfred Metzger,
Wolfgang Muees,
Paolo Nenzi,
Gary W. Ng,
Hong June Park,
Stefano Perticaroli,
Arno Peters,
Serban-Mihai Popescu,
Georg Post,
Thomas L. Quarles,
Emmanuel Rouat,
Jean-Marc Routure,
Jaijeet S. Roychowdhury,
Lionel Sainte Cluque,
Takayasu Sakurai,
Amakawa Shuhei,
Kanwar Jit Singh,
Bill Swartz,
Hitoshi Tanaka,
Steve Tell,
Andrew Tuckey,
Andreas Unger,
Holger Vogt,
Dietmar Warning,
Michael Widlok,
Charles D.H. Williams,
Antony Wilson,

and many others...
If someone helped in the development and has not been inserted in this list then this omission was unintentional. If you feel you should be on this list then please write to . Do not be shy, we would like to make a list as complete as
possible.

XSPICE
The XSPICE simulator, developed at Georgia Tech., is based on the SPICE3 program developed
by the Electronics Research Laboratory, Department of Electrical Engineering and Computer
Sciences, University of California at Berkeley. The authors of XSPICE gratefully acknowledge
UC Berkeley’s development and distribution of this software, and their licensing policies which
promote further improvements to simulation technology. They also gratefully acknowledge the

CONTENTS

33

participation and support of their U.S. Air Force sponsors, the Aeronautical Systems Center and
the Warner Robins Air Logistics Command, without which the development of XSPICE would
not have been possible.

34

CONTENTS

Chapter 1
Introduction
Ngspice is a general-purpose circuit simulation program for nonlinear and linear analyses. Circuits may contain resistors, capacitors, inductors, mutual inductors, independent or dependent
voltage and current sources, loss-less and lossy transmission lines, switches, uniform distributed
RC lines, and the five most common semiconductor devices: diodes, BJTs, JFETs, MESFETs,
and MOSFETs.
Some introductory remarks on how to use ngspice may be found in Chapt. 21.
Ngspice is an update of Spice3f5, the last Berkeley’s release of Spice3 simulator family. Ngspice is being developed to include new features to existing Spice3f5 and to fix its bugs. Improving a complex software like a circuit simulator is a very hard task and, while some improvements have been made, most of the work has been done on bug fixing and code refactoring.
Ngspice has built-in models for the semiconductor devices, and the user need specify only the
pertinent model parameter values. There are three models for bipolar junction transistors, all
based on the integral-charge model of Gummel and Poon; however, if the Gummel-Poon parameters are not specified, the basic model (BJT) reduces to the simpler Ebers-Moll model.
In either case and in either models, charge storage effects, ohmic resistances, and a currentdependent output conductance may be included. The second bipolar model BJT2 adds dc current computation in the substrate diode. The third model (VBIC) contains further enhancements
for advanced bipolar devices.
The semiconductor diode model can be used for either junction diodes or Schottky barrier diodes. There are two models for JFET: the first (JFET) is based on the model of Shichman and
Hodges, the second (JFET2) is based on the Parker-Skellern model. All the original six MOSFET models are implemented: MOS1 is described by a square-law I-V characteristic, MOS2 [1]
is an analytical model, while MOS3 [1] is a semi-empirical model; MOS6 [2] is a simple analytic model accurate in the short channel region; MOS9, is a slightly modified Level 3 MOSFET
model - not to confuse with Philips level 9; BSIM 1 [3, 4]; BSIM2 [5] are the old BSIM (Berkeley Short-channel IGFET Model) models. MOS2, MOS3, and BSIM include second-order
effects such as channel-length modulation, subthreshold conduction, scattering-limited velocity
saturation, small-size effects, and charge controlled capacitances. The recent MOS models for
submicron devices are the BSIM3 (Berkeley BSIM3 web page) and BSIM4 (Berkeley BSIM4
web page) models. Silicon-on-insulator MOS transistors are described by the SOI models from
the BSIMSOI family (Berkeley BSIMSOI web page) and the STAG [18] one. There is partial
support for a couple of HFET models and one model for MESA devices.

35

36

CHAPTER 1. INTRODUCTION

Ngspice supports mixed-level simulation and provides a direct link between technology parameters and circuit performance. A mixed-level circuit and device simulator can provide greater
simulation accuracy than a stand-alone circuit or device simulator by numerically modeling the
critical devices in a circuit. Compact models can be used for noncritical devices. The mixedlevel extensions to ngspice are two:
• CIDER: a mixed-level circuit and device simulator integrated into ngspice code. CIDER
was originally the name of the mixed-level extension made to spice3f5.
• GSS: GSS (now called GENIUS) TCAD is a 2D simulator developed independently from
ngspice. The device simulator itself is free and not included into ngspice, but a socket
interface is provided.
Ngspice supports mixed-signal simulation through the integration of XSPICE code into it.
XSPICE software, developed as an extension to Spice3C1 from GeorgiaTech, has been ported to ngspice to provide ‘board’ level and mixed-signal simulation.
New devices can be added to ngspice by two means: the XSPICE code-model interface and the
ADMS interface based on Verilog-A and XML.
Finally, numerous small bugs have been discovered and fixed, and the program has been ported
to a wider variety of computing platforms.

1.1

Simulation Algorithms

Computer-based circuit simulation is often used as a tool by designers, test engineers, and
others who want to analyze the operation of a design without examining the physical circuit.
Simulation allows you to change quickly the parameters of many of the circuit elements to
determine how they affect the circuit response. Often it is difficult or impossible to change
these parameters in a physical circuit.
However, to be practical, a simulator must execute in a reasonable amount of time. The key to
efficient execution is choosing the proper level of modeling abstraction for a given problem. To
support a given modeling abstraction, the simulator must provide appropriate algorithms.
Historically, circuit simulators have supported either an analog simulation algorithm or a digital
simulation algorithm. Ngspice inherits the XSPICE framework and supports both analog and
digital algorithms and is a ‘mixed-mode’ simulator.

1.1.1

Analog Simulation

Analog simulation focuses on the linear and non-linear behavior of a circuit over a continuous
time or frequency interval. The circuit response is obtained by iteratively solving Kirchhoff’s
Laws for the circuit at time steps selected to ensure the solution has converged to a stable value
and that numerical approximations of integrations are sufficiently accurate. Since Kirchhoff’s
laws form a set of simultaneous equations, the simulator operates by solving a matrix of equations at each time point. This matrix processing generally results in slower simulation times
when compared to digital circuit simulators.

1.1. SIMULATION ALGORITHMS

37

The response of a circuit is a function of the applied sources. Ngspice offers a variety of
source types including DC, sine-wave, and pulse. In addition to specifying sources, the user
must define the type of simulation to be run. This is termed the ‘mode of analysis’. Analysis
modes include DC analysis, AC analysis, and transient analysis. For DC analysis, the timevarying behavior of reactive elements is neglected and the simulator calculates the DC solution
of the circuit. Swept DC analysis may also be accomplished with ngspice. This is simply the
repeated application of DC analysis over a range of DC levels for the input sources. For AC
analysis, the simulator determines the response of the circuit, including reactive elements to
small-signal sinusoidal inputs over a range of frequencies. The simulator output in this case
includes amplitudes and phases as a function of frequency. For transient analysis, the circuit
response, including reactive elements, is analyzed to calculate the behavior of the circuit as a
function of time.

1.1.2

Digital Simulation

Digital circuit simulation differs from analog circuit simulation in several respects. A primary
difference is that a solution of Kirchhoff’s laws is not required. Instead, the simulator must only
determine whether a change in the logic state of a node has occurred and propagate this change
to connected elements. Such a change is called an ‘event’.
When an event occurs, the simulator examines only those circuit elements that are affected by
the event. As a result, matrix analysis is not required in digital simulators. By comparison,
analog simulators must iteratively solve for the behavior of the entire circuit because of the
forward and reverse transmission properties of analog components. This difference results in
a considerable computational advantage for digital circuit simulators, which is reflected in the
significantly greater speed of digital simulations.

1.1.3

Mixed-Mode Simulation

Modern circuits often contain a mix of analog and digital circuits. To simulate such circuits
efficiently and accurately a mix of analog and digital simulation techniques is required. When
analog simulation algorithms are combined with digital simulation algorithms, the result is
termed ‘mixed-mode simulation’.
Two basic methods of implementing mixed-mode simulation used in practice are the ‘native
mode’ and ‘glued mode’ approaches. Native mode simulators implement both an analog algorithm and a digital algorithm in the same executable. Glued mode simulators actually use two
simulators, one of which is analog and the other digital. This type of simulator must define an
input/output protocol so that the two executables can communicate with each other effectively.
The communication constraints tend to reduce the speed, and sometimes the accuracy, of the
complete simulator. On the other hand, the use of a glued mode simulator allows the component
models developed for the separate executables to be used without modification.
Ngspice is a native mode simulator providing both analog and event-based simulation in the
same executable. The underlying algorithms of ngspice (coming from XSPICE and its Code
Model Subsystem) allow use of all the standard SPICE models, provide a pre-defined collection
of the most common analog and digital functions, and provide an extensible base on which to
build additional models.

38
1.1.3.1

CHAPTER 1. INTRODUCTION
User-Defined Nodes

Ngspice supports creation of ‘User-Defined Node’ types. User-Defined Node types allow you
to specify nodes that propagate data other than voltages, currents, and digital states. Like digital
nodes, User-Defined Nodes use event-driven simulation, but the state value may be an arbitrary
data type. A simple example application of User-Defined Nodes is the simulation of a digital
signal processing filter algorithm. In this application, each node could assume a real or integer
value. More complex applications may define types that involve complex data such as digital
data vectors or even non-electronic data.
Ngspice digital simulation is actually implemented as a special case of this User-Defined Node
capability where the digital state is defined by a data structure that holds a Boolean logic state
and a strength value.

1.1.4

Mixed-Level Simulation

Ngspice can simulate numerical device models for diodes and transistors in two different ways,
either through the integrated DSIM simulator or interfacing to GSS TCAD system. DSIM is an
internal C-based device simulator that is part of the CIDER simulator, the mixed-level simulator
based on SPICE3f5. CIDER within ngspice provides circuit analyses, compact models for
semiconductor devices, and one- or two-dimensional numerical device models.
1.1.4.1

CIDER (DSIM)

DSIM provides accurate, one- and two-dimensional numerical device models based on the solution of Poisson’s equation, and the electron and hole current-continuity equations. DSIM
incorporates many of the same basic physical models found in the Stanford two-dimensional
device simulator PISCES. Input to CIDER consists of a SPICE-like description of the circuit
and its compact models, and PISCES-like descriptions of the structures of numerically modeled
devices. As a result, CIDER should seem familiar to designers already accustomed to these
two tools. CIDER is based on the mixed-level circuit and device simulator CODECS, and is a
replacement for this program. The basic algorithms of the two programs are the same. Some of
the differences between CIDER and CODECS are described below. The CIDER input format
has greater flexibility and allows increased access to physical model parameters. New physical
models have been added to allow simulation of state-of-the-art devices. These include transverse field mobility degradation important in scaled-down MOSFETs and a polysilicon model
for poly-emitter bipolar transistors. Temperature dependence has been included over the range
from -50C to 150C. The numerical models can be used to simulate all the basic types of semiconductor devices: resistors, MOS capacitors, diodes, BJTs, JFETs and MOSFETs. BJTs
and JFETs can be modeled with or without a substrate contact. Support has been added for the
management of device internal states. Post-processing of device states can be performed using
the ngnutmeg user interface.
1.1.4.2

GSS TCAD

GSS is a TCAD software that enables two-dimensional numerical simulation of semiconductor
device with well-known drift-diffusion and hydrodynamic method. GSS has Basic DDM (driftdiffusion method) solver, Lattice Temperature Corrected DDM solver, EBM (energy balance

1.2. SUPPORTED ANALYSES

39

method) solver and Quantum corrected DDM solver based on density-gradient theory. The GSS
program is directed via input statements by a user specified disk file. Supports triangle mesh
generation and adaptive mesh refinement. Employs PMI (physical model interface) to support
various materials, including compound semiconductor materials such as SiGe and AlGaAs.
Supports DC sweep, transient and AC sweep calculations. The device can be stimulated by
voltage or current source(s).
GSS is no longer updated, but is still available as open source as a limited edition of the commercial GENIUS TCAD tool.

1.2

Supported Analyses

The ngspice simulator supports the following different types of analysis:
1. DC Analysis (Operating Point and DC Sweep)
2. AC Small-Signal Analysis
3. Transient Analysis
4. Pole-Zero Analysis
5. Small-Signal Distortion Analysis
6. Sensitivity Analysis
7. Noise Analysis
Applications that are exclusively analog can make use of all analysis modes with the exception
of Code Model subsystem that do not implements Pole-Zero, Distortion, Sensitivity and Noise
analyses. Event-driven applications that include digital and User-Defined Node types may make
use of DC (operating point and DC sweep) and Transient only.
In order to understand the relationship between the different analyses and the two underlying
simulation algorithms of ngspice, it is important to understand what is meant by each analysis
type. This is detailed below.

1.2.1

DC Analyses

The dc analysis portion of ngspice determines the dc operating point of the circuit with inductors
shorted and capacitors opened. The dc analysis options are specified on the .DC, .TF, and .OP
control lines.
There is assumed to be no time dependence on any of the sources within the system description.
The simulator algorithm subdivides the circuit into those portions that require the analog simulator algorithm and such that require the event-driven algorithm. Each subsystem block is then
iterated to solution, with the interfaces between analog nodes and event-driven nodes iterated
for consistency across the entire system.
Once stable values are obtained for all nodes in the system, the analysis halts and the results
may be displayed or printed out as you request them.

40

CHAPTER 1. INTRODUCTION

A dc analysis is automatically performed prior to a transient analysis to determine the transient
initial conditions, and prior to an ac small-signal analysis to determine the linearized, smallsignal models for nonlinear devices. If requested, the dc small-signal value of a transfer function
(ratio of output variable to input source), input resistance, and output resistance is also computed
as a part of the dc solution. The dc analysis can also be used to generate dc transfer curves: a
specified independent voltage, current source, resistor or temperature1 is stepped over a userspecified range and the dc output variables are stored for each sequential source value.

1.2.2

AC Small-Signal Analysis

AC analysis is limited to analog nodes and represents the small signal, sinusoidal solution of the
analog system described at a particular frequency or set of frequencies. This analysis is similar
to the DC analysis in that it represents the steady-state behavior of the described system with a
single input node at a given set of stimulus frequencies.
The program first computes the dc operating point of the circuit and determines linearized,
small-signal models for all of the nonlinear devices in the circuit. The resultant linear circuit
is then analyzed over a user-specified range of frequencies. The desired output of an ac smallsignal analysis is usually a transfer function (voltage gain, transimpedance, etc). If the circuit
has only one ac input, it is convenient to set that input to unity and zero phase, so that output
variables have the same value as the transfer function of the output variable with respect to the
input.

1.2.3

Transient Analysis

Transient analysis is an extension of DC analysis to the time domain. A transient analysis begins by obtaining a DC solution to provide a point of departure for simulating time-varying
behavior. Once the DC solution is obtained, the time-dependent aspects of the system are reintroduced, and the two simulator algorithms incrementally solve for the time varying behavior of
the entire system. Inconsistencies in node values are resolved by the two simulation algorithms
such that the time-dependent waveforms created by the analysis are consistent across the entire
simulated time interval. Resulting time-varying descriptions of node behavior for the specified
time interval are accessible to you.
All sources that are not time dependent (for example, power supplies) are set to their dc value.
The transient time interval is specified on a .TRAN control line.

1.2.4

Pole-Zero Analysis

The pole-zero analysis portion of Ngspice computes the poles and/or zeros in the small-signal
ac transfer function. The program first computes the dc operating point and then determines
the linearized, small-signal models for all the nonlinear devices in the circuit. This circuit is
then used to find the poles and zeros of the transfer function. Two types of transfer functions
are allowed: one of the form (output voltage)/(input voltage) and the other of the form (output
voltage)/(input current). These two types of transfer functions cover all the cases and one can
1 Temperature

(TEMP) and resistance sweeps have been introduced in Ngspice, they were not available in the
original code of Spice3f5.

1.2. SUPPORTED ANALYSES

41

find the poles/zeros of functions like input/output impedance and voltage gain. The input and
output ports are specified as two pairs of nodes. The pole-zero analysis works with resistors,
capacitors, inductors, linear-controlled sources, independent sources, BJTs, MOSFETs, JFETs
and diodes. Transmission lines are not supported. The method used in the analysis is a suboptimal numerical search. For large circuits it may take a considerable time or fail to find all
poles and zeros. For some circuits, the method becomes ‘lost’ and finds an excessive number
of poles or zeros.

1.2.5

Small-Signal Distortion Analysis

The distortion analysis portion of Ngspice computes steady-state harmonic and intermodulation
products for small input signal magnitudes. If signals of a single frequency are specified as the
input to the circuit, the complex values of the second and third harmonics are determined at
every point in the circuit. If there are signals of two frequencies input to the circuit, the analysis
finds out the complex values of the circuit variables at the sum and difference of the input
frequencies, and at the difference of the smaller frequency from the second harmonic of the
larger frequency. Distortion analysis is supported for the following nonlinear devices:
• Diodes (DIO),
• BJT,
• JFET (level 1),
• MOSFETs (levels 1, 2, 3, 9, and BSIM1),
• MESFET (level 1).
All linear devices are automatically supported by distortion analysis. If there are switches
present in the circuit, the analysis continues to be accurate provided the switches do not change
state under the small excitations used for distortion calculations.
If a device model does not support direct small signal distortion analysis, please use the Fourier
statement and evaluate the output per scripting.

1.2.6

Sensitivity Analysis

Ngspice will calculate either the DC operating-point sensitivity or the AC small-signal sensitivity of an output variable with respect to all circuit variables, including model parameters.
Ngspice calculates the difference in an output variable (either a node voltage or a branch current)
by perturbing each parameter of each device independently. Since the method is a numerical
approximation, the results may demonstrate second order effects in highly sensitive parameters,
or may fail to show very low but non-zero sensitivity. Further, since each variable is perturb
by a small fraction of its value, zero-valued parameters are not analyzed (this has the benefit of
reducing what is usually a very large amount of data).

42

1.2.7

CHAPTER 1. INTRODUCTION

Noise Analysis

The noise analysis portion of Ngspice gives the device-generated noise for a given circuit. When
provided with an input source and an output port, the analysis calculates the noise contributions
of each device, and each noise generator within the device, to the output port voltage. It also
calculates the equivalent input noise of the circuit, based on the output noise. This is done for
every frequency point in a specified range - the calculated value of the noise corresponds to
the spectral density of the circuit variable viewed as a stationary Gaussian stochastic process.
After calculating the spectral densities, noise analysis integrates these values over the specified frequency range to arrive at the total noise voltage and current over this frequency range.
The calculated values correspond to the variance of the circuit variables viewed as stationary
Gaussian processes.

1.2.8

Periodic Steady State Analysis

Experimental code, not yet made publicly available.
PSS is a radio frequency periodical large-signal dedicated analysis. The implementation is
based on a time domain shooting method that make use of transient analysis. As it is in early
development stage, PSS performs analysis only on autonomous circuits, meaning that it is able
to predict fundamental frequency and (harmonic) amplitude(s) for oscillators, VCOs, etc.. The
algorithm is based on a search of the minimum error vector defined as the difference of RHS
vectors between two occurrences of an estimated period. Convergence is reached when the
mean of this error vector decreases below a given threshold parameter. Results of PSS are the
basis of periodical large-signal analyses like PAC or PNoise.

1.3

Analysis at Different Temperatures

Temperature, in ngspice, is a property associated to the entire circuit, rather than an analysis option. Circuit temperature has a default (nominal) value of 27°C (300.15 K) that can be changed
using the TEMP option in an .option control line (see 15.1.1) or by the .TEMP line (see 2.11),
which has precedence over the .option TEMP line. All analyses are, thus, performed at circuit
temperature, and if you want to simulate circuit behavior at different temperatures you should
prepare a netlist for each temperature.
All input data for ngspice is assumed to have been measured at the circuit nominal temperature.
This value can further be overridden for any device that models temperature effects by specifying the TNOM parameter on the .model itself. Individual instances may further override the
circuit temperature through the specification of TEMP and DTEMP parameters on the instance.
The two options are not independent even if you can specify both on the instance line, the TEMP
option overrides DTEMP. The algorithm to compute instance temperature is described below:
IF TEMP is specified THEN
instance_temperature = TEMP
ELSE IF
instance_temperature = circuit_temperature + DTEMP
END IF
Algorithm 1: Instance temperature computation

1.3. ANALYSIS AT DIFFERENT TEMPERATURES

43

Temperature dependent support is provided for all devices except voltage and current sources
(either independent and controlled) and BSIM models. BSIM MOSFETs have an alternate
temperature dependency scheme that adjusts all of the model parameters before input to ngspice.
For details of the BSIM temperature adjustment, see [6] and [7]. Temperature appears explicitly
in the exponential terms of the BJT and diode model equations. In addition, saturation currents
have a built-in temperature dependence. The temperature dependence of the saturation current
in the BJT models is determined by:


T1
IS (T1 ) = IS (T0 )
T0

XT I



Eg q (T1 T0 )
exp
k (T1 − T0 )


(1.1)

where k is Boltzmann’s constant, q is the electronic charge, Eg is the energy gap model parameter, and XT I is the saturation current temperature exponent (also a model parameter, and
usually equal to 3).
The temperature dependence of forward and reverse beta is according to the formula:


T1
B (T1 ) = B (T0 )
T0

XT B
(1.2)

where T0 and T1 are in degrees Kelvin, and XT B is a user-supplied model parameter. Temperature effects on beta are carried out by appropriate adjustment to the values of BF , ISE , BR , and
ISC (SPICE model parameters BF, ISE, BR, and ISC, respectively).
Temperature dependence of the saturation current in the junction diode model is determined by:


T1
IS (T1 ) = IS (T0 )
T0

 XT I
N



Eg q (T1 T0 )
exp
Nk (T1 − T0 )


(1.3)

where N is the emission coefficient model parameter, and the other symbols have the same
meaning as above. Note that for Schottky barrier diodes, the value of the saturation current
temperature exponent, XT I, is usually 2. Temperature appears explicitly in the value of junction
potential, U (in Ngspice PHI), for all the device models.
The temperature dependence is determined by:
kT
U (T ) =
ln
q

Na Nd
Ni (T )2

!
(1.4)

where k is Boltzmann’s constant, q is the electronic charge, Na is the acceptor impurity density, Nd is the donor impurity density, Ni is the intrinsic carrier concentration, and Eg is the
energy gap. Temperature appears explicitly in the value of surface mobility, M0 (or U0 ), for the
MOSFET model.
The temperature dependence is determined by:
M0 (T0 )
M0 (T ) =  1.5
T
T0

(1.5)

44

CHAPTER 1. INTRODUCTION

The effects of temperature on resistors, capacitor and inductors is modeled by the formula:
h
i
2
R (T ) = R (T0 ) 1 + TC1 (T − T0 ) + TC2 (T − T0 )

(1.6)

where T is the circuit temperature, T0 is the nominal temperature, and TC1 and TC2 are the first
and second order temperature coefficients.

1.4

Convergence

Ngspice uses the Newton-Raphson algorithm to solve nonlinear equations arising from circuit
description. The NR algorithm is interactive and terminates when both of the following conditions hold:
1. The nonlinear branch currents converge to within a tolerance of 0.1% or 1 picoamp (1.0e12 Amp), whichever is larger.
2. The node voltages converge to within a tolerance of 0.1% or 1 microvolt (1.0e-6 Volt),
whichever is larger.

1.4.1

Voltage convergence criterion

The algorithm has reached convergence when the difference between the last iteration k and the
current one (k + 1)
(k+1)

vn

(k)

− vn

≤ RELTOL vnmax + VNTOL,

(1.7)

where
vnmax = max



(k+1)
vn

,

(k)
vn



.

(1.8)

The RELTOL (RELative TOLerance) parameter, which default value is 10−3 , specifies how small
the solution update must be, relative to the node voltage, to consider the solution to have converged. The VNTOL (absolute convergence) parameter, which has 1µV as default value, becomes
important when node voltages have near zero values. The relative parameter alone, in such
case, would need too strict tolerances, perhaps lower than computer round-off error, and thus
convergence would never be achieved. VNTOL forces the algorithm to consider as converged any
node whose solution update is lower than its value.

1.4.2

Current convergence criterion

Ngspice checks the convergence on the non-linear functions that describe the non-linear branches in circuit elements. In semiconductor devices the functions defines currents through the
device and thus the name of the criterion.

1.4. CONVERGENCE

45

Ngspice computes the difference between the value of the nonlinear function computed for the
last voltage and the linear approximation of the same current computed with the actual voltage
\
(k+1)
(k)
ibranch − ibranch ≤ RELTOL ibrmax + ABSTOL,

(1.9)

where
ibrmax



\
(k+1) (k)
= max ibranch , ibranch .

(1.10)

In the two expressions above, the i\
branch indicates the linear approximation of the current.

1.4.3

Convergence failure

Although the algorithm used in ngspice has been found to be very reliable, in some cases it fails
to converge to a solution. When this failure occurs, the program terminates the job. Failure
to converge in dc analysis is usually due to an error in specifying circuit connections, element
values, or model parameter values. Regenerative switching circuits or circuits with positive
feedback probably will not converge in the dc analysis unless the OFF option is used for some
of the devices in the feedback path, .nodeset control line is used to force the circuit to converge
to the desired state.

46

CHAPTER 1. INTRODUCTION

Chapter 2
Circuit Description
2.1
2.1.1

General Structure and Conventions
Input file structure

The circuit to be analyzed is described to ngspice by a set of element instance lines, which
define the circuit topology and element instance values, and a set of control lines, which define
the model parameters and the run controls. All lines are assembled in an input file to be read by
ngspice. Two lines are essential:
• The first line in the input file must be the title, which is the only comment line that does
not need any special character in the first place.
• The last line must be .end.
The order of the remaining lines is arbitrary (except, of course, that continuation lines must
immediately follow the line being continued). This feature in the ngspice input language dates back to the punched card times where elements were written on separate cards (and cards
frequently fell off). Leading white spaces in a line are ignored, as well as empty lines.
The lines decribed in sections 2.1 to 2.12 are typically used in the core of the input file, outside
of a .control section (see 16.4.3). An exception is the .include includefile line (2.6)
that may be placed anywhere in the input file. The contents of includefile will be inserted
exactly in place of the .include line.

2.1.2

Circuit elements (device instances)

Each element in the circuit is a device instance specified by an instance line that contains:
• the element instance name,
• the circuit nodes to which the element is connected,
• and the values of the parameters that determine the electrical characteristics of the element.

47

48

CHAPTER 2. CIRCUIT DESCRIPTION

The first letter of the element instance name specifies the element type. The format for the
ngspice element types is given in the following manual chapters. In the rest of the manual, the
strings XXXXXXX, YYYYYYY, and ZZZZZZZ denote arbitrary alphanumeric strings.
For example, a resistor instance name must begin with the letter R and can contain one or more
characters. Hence, R, R1, RSE, ROUT, and R3AC2ZY are valid resistor names. Details of each
type of device are supplied in a following section 3. Table 2.1 lists the element types available
in ngspice, sorted by their first letter.

First letter

Element description

A

XSPICE code model

B
C
D

Behavioral (arbitrary) source
Capacitor
Diode

E

Voltage-controlled voltage source (VCVS)

F

Current-controlled current source (CCCs)

G

Voltage-controlled current source (VCCS)

H
I
J
K
L

Current-controlled voltage source (CCVS)
Current source
Junction field effect transistor (JFET)
Coupled (Mutual) Inductors
Inductor

M

Metal oxide field effect transistor (MOSFET)

N
O
P
Q
R
S
T
U
V
W
X
Y
Z

Numerical device for GSS
Lossy transmission line
Coupled multiconductor line (CPL)
Bipolar junction transistor (BJT)
Resistor
Switch (voltage-controlled)
Lossless transmission line
Uniformly distributed RC line
Voltage source
Switch (current-controlled)
Subcircuit
Single lossy transmission line (TXL)
Metal semiconductor field effect transistor (MESFET)
Table 2.1: ngspice element types

Comments, links
12
analog (12.2)
digital (12.4)
mixed signal (12.3)
5.1
3.2.5
7
linear (4.2.2),
non-linear (5.2)
linear (4.2.3)
linear (4.2.1),
non-linear (5.3)
linear (4.2.4)
4.1
9
3.2.11
3.2.9
11
BSIM3 (11.2.9)
BSIM4 (11.2.10)
14.2
6.2
6.4.2
8
3.2.1
3.2.14
6.1
6.3
4.1
3.2.14
2.4.3
6.4.1
10

2.1. GENERAL STRUCTURE AND CONVENTIONS

2.1.3

49

Some naming conventions

Fields on a line are separated by one or more blanks, a comma, an equal (=) sign, or a left or
right parenthesis; extra spaces are ignored. A line may be continued by entering a ‘+’ (plus) in
column 1 of the following line; ngspice continues reading beginning with column 2. A name
field must begin with a letter (A through Z) and cannot contain any delimiters. A number field
may be an integer field (12, -44), a floating point field (3.14159), either an integer or floating
point number followed by an integer exponent (1e-14, 2.65e3), or either an integer or a floating
point number followed by one of the following scale factors:
Suffix
T
G
Meg
K
mil
m
u
n
p
f

Name
Tera
Giga
Mega
Kilo
Mil
milli
micro
nano
pico
femto

Factor
1012
109
106
103
25.4 × 10−6
10−3
10−6
10−9
10−12
10−15

Table 2.2: Ngspice scale factors

Letters immediately following a number that are not scale factors are ignored, and letters immediately following a scale factor are ignored. Hence, 10, 10V, 10Volts, and 10Hz all represent
the same number, and M, MA, MSec, and MMhos all represent the same scale factor. Note that
1000, 1000.0, 1000Hz, 1e3, 1.0e3, 1kHz, and 1k all represent the same number. Note that ‘M’
or ‘m’ denote ‘milli’, i.e. 10−3 . Suffix meg has to be used for 106 .
Nodes names may be arbitrary character strings and are case insensitive, if ngspice is used in
batch mode (16.4.1). If in interactive (16.4.2) or control (16.4.3) mode, node names may either
be plain numbers or arbitrary character strings, not starting with a number. The ground node
must be named ‘0’ (zero). For compatibility reason gnd is accepted as ground node, and will
internally be treated as a global node and be converted to ‘0’. Each circuit has to have a ground
node (gnd or 0)! Note the difference in ngspice where the nodes are treated as character strings
and not evaluated as numbers, thus ‘0’ and 00 are distinct nodes in ngspice but not in SPICE2.
Ngspice requires that the following topological constraints are satisfied:
• The circuit cannot contain a loop of voltage sources and/or inductors and cannot contain
a cut-set of current sources and/or capacitors.
• Each node in the circuit must have a dc path to ground.
• Every node must have at least two connections except for transmission line nodes (to
permit unterminated transmission lines) and MOSFET substrate nodes (which have two
internal connections anyway).

50

CHAPTER 2. CIRCUIT DESCRIPTION

2.2

Basic lines

2.2.1

.TITLE line

Examples:
POWER AMPLIFIER CIRCUIT
* additional lines following
*...
Test of CAM cell
* additional lines following
*...
The title line must be the first in the input file. Its contents are printed verbatim as the heading
for each section of output.
As an alternative you may place a .TITLE  line anywhere in your input deck.
The first line of your input deck will be overridden by the contents of this line following the
.TITLE statement.
.TITLE line example:
******************************
* additional lines following
*...
.TITLE Test of CAM cell
* additional lines following
*...
will internally be replaced by
Internal input deck:
Test of CAM cell
* additional lines following
*...
*TITLE Test of CAM cell
* additional lines following
*...

2.2.2

.END Line

Examples:
.end
The .end line must always be the last in the input file. Note that the period is an integral part
of the name.

2.3. .MODEL DEVICE MODELS

2.2.3

51

Comments

General Form:
* 
Examples:
* RF =1K Gain should be 100
* Check open -loop gain and phase margin
The asterisk in the first column indicates that this line is a comment line. Comment lines may
be placed anywhere in the circuit description.

2.2.4

End-of-line comments

General Form:
 $ 
Examples:
RF2 =1K $ Gain should be 100
C1 =10p $ Check open -loop gain and phase margin
.param n1 =1 // new value
ngspice supports comments that begin with double characters ‘$ ’ (dollar plus space) or ‘//’.
For readability you should precede each comment character with a space. ngspice will accept
the single character ‘$’.
Please note that in .control sections the ‘;’ character means ‘continuation’ and can be used
to put more than one statement on a line.

2.3

.MODEL Device Models

General form:
.model mname type( pname1 =pval1 pname2 =pval2 ... )
Examples:
.model MOD1 npn (bf =50 is=1e -13 vbf =50)

52

CHAPTER 2. CIRCUIT DESCRIPTION
Code
R
C
L
SW
CSW
URC
LTRA
D
NPN
PNP
NJF
PJF
NMOS
PMOS
NMF
PMF

Model Type
Semiconductor resistor model
Semiconductor capacitor model
Inductor model
Voltage controlled switch
Current controlled switch
Uniform distributed RC model
Lossy transmission line model
Diode model
NPN BJT model
PNP BJT model
N-channel JFET model
P-channel JFET model
N-channel MOSFET model
P-channel MOSFET model
N-channel MESFET model
P-channel MESFET model

Table 2.3: Ngspice model types
Most simple circuit elements typically require only a few parameter values. However, some devices (semiconductor devices in particular) that are included in ngspice require many parameter
values. Often, many devices in a circuit are defined by the same set of device model parameters.
For these reasons, a set of device model parameters is defined on a separate .model line and
assigned a unique model name. The device element lines in ngspice then refer to the model
name.
For these more complex device types, each device element line contains the device name, the
nodes the device is connected to, and the device model name. In addition, other optional parameters may be specified for some devices: geometric factors and an initial condition (see the
following section on Transistors (8 to 11) and Diodes (7) for more details). mname in the above
is the model name, and type is one of the following fifteen types:
Parameter values are defined by appending the parameter name followed by an equal sign and
the parameter value. Model parameters that are not given a value are assigned the default values
given below for each model type. Models are listed in the section on each device along with
the description of device element lines. Model parameters and their default values are given in
Chapt. 31.

2.4

.SUBCKT Subcircuits

A subcircuit that consists of ngspice elements can be defined and referenced in a fashion similar
to device models. Subcircuits are the way ngspice implements hierarchical modeling, but this is
not entirely true because each subcircuit instance is flattened during parsing, and thus ngspice
is not a hierarchical simulator.
The subcircuit is defined in the input deck by a grouping of element cards delimited by the
.subckt and the .ends cards (or the keywords defined by the substart and subend options

2.4. .SUBCKT SUBCIRCUITS

53

(see 17.7)); the program then automatically inserts the defined group of elements wherever the
subcircuit is referenced. Instances of subcircuits within a larger circuit are defined through the
use of an instance card that begins with the letter ‘X’. A complete example of all three of these
cards follows:
Example:
* The following is the instance card:
*
xdiv1 10 7 0 vdivide
* The following are the subcircuit definition cards:
*
. subckt vdivide 1 2 3
r1 1 2 10K
r2 2 3 5K
.ends
The above specifies a subcircuit with ports numbered ‘1’, ‘2’ and ‘3’:
• Resistor ‘R1’ is connected from port ‘1’ to port ‘2’, and has value 10 kOhms.
• Resistor ‘R2’ is connected from port ‘2’ to port ‘3’, and has value 5 kOhms.
The instance card, when placed in an ngspice deck, will cause subcircuit port ‘1’ to be equated
to circuit node ‘10’, while port ‘2’ will be equated to node ‘7’ and port ‘3’ will equated to node
‘0’.
There is no limit on the size or complexity of subcircuits, and subcircuits may contain other
subcircuits. An example of subcircuit usage is given in Chapt. 21.6.

2.4.1

.SUBCKT Line

General form:
. SUBCKT subnam N1 
Examples:
. SUBCKT OPAMP 1 2 3 4
A circuit definition is begun with a .SUBCKT line. subnam is the subcircuit name, and N1, N2,
... are the external nodes, which cannot be zero. The group of element lines that immediately
follow the .SUBCKT line define the subcircuit. The last line in a subcircuit definition is the
.ENDS line (see below). Control lines may not appear within a subcircuit definition; however,
subcircuit definitions may contain anything else, including other subcircuit definitions, device
models, and subcircuit calls (see below). Note that any device models or subcircuit definitions
included as part of a subcircuit definition are strictly local (i.e., such models and definitions

54

CHAPTER 2. CIRCUIT DESCRIPTION

are not known outside the subcircuit definition). Also, any element nodes not included on the
.SUBCKT line are strictly local, with the exception of 0 (ground) that is always global. If you
use parameters, the .SUBCKT line will be extended (see 2.8.3).

2.4.2

.ENDS Line

General form:
.ENDS 
Examples:
.ENDS OPAMP
The .ENDS line must be the last one for any subcircuit definition. The subcircuit name, if
included, indicates which subcircuit definition is being terminated; if omitted, all subcircuits
being defined are terminated. The name is needed only when nested subcircuit definitions are
being made.

2.4.3

Subcircuit Calls

General form:
XYYYYYYY N1  SUBNAM
Examples:
X1 2 4 17 3 1 MULTI
Subcircuits are used in ngspice by specifying pseudo-elements beginning with the letter X,
followed by the circuit nodes to be used in expanding the subcircuit. If you use parameters, the
subcircuit call will be modified (see 2.8.3).

2.5

.GLOBAL

General form:
. GLOBAL nodename
Examples:
. GLOBAL gnd vcc

2.6. .INCLUDE

55

Nodes defined in the .GLOBAL statement are available to all circuit and subcircuit blocks independently from any circuit hierarchy. After parsing the circuit, these nodes are accessible from
top level.

2.6

.INCLUDE

General form:
. INCLUDE filename
Examples:
. INCLUDE / users / spice / common /bsim3 -param.mod
Frequently, portions of circuit descriptions will be reused in several input files, particularly with
common models and subcircuits. In any ngspice input file, the .INCLUDE line may be used to
copy some other file as if that second file appeared in place of the .INCLUDE line in the original
file.
There is no restriction on the file name imposed by ngspice beyond those imposed by the local
operating system.

2.7

.LIB

General form:
.LIB filename libname
Examples:
.LIB / users / spice / common / mosfets .lib mos1
The .LIB statement allows to include library descriptions into the input file. Inside the *.lib
file a library libname will be selected. The statements of each library inside the *.lib file are
enclosed in .LIB libname <...> .ENDL statements.
If the compatibility mode (16.13) is set to ’ps’ by set ngbehavior=ps (17.7) in spinit (16.5)
or .spiceinit (16.6), then a simplified syntax .LIB filename is available: a warning is issued
and filename is simply included as described in Chapt. 2.6.

2.8

.PARAM Parametric netlists

Ngspice allows for the definition of parametric attributes in the netlists. This is an enhancement
of the ngspice front-end that adds arithmetic functionality to the circuit description language.

56

2.8.1

CHAPTER 2. CIRCUIT DESCRIPTION

.param line

General form:
.param  = 

 =  ...

Examples:
.param
.param
.param
.param
.param

pippo =5
po =6 pp =7.8 pap ={ AGAUSS (pippo , 1, 1.67)}
pippp ={ pippo + pp}
p={ pp}
pop=’pp+p’

This line assigns numerical values to identifiers. More than one assignment per line is possible
using a separating space. Parameter identifier names must begin with an alphabetic character.
The other characters must be either alphabetic, a number, or ! # $ % [ ] _ as special characters. The variables time, temper, and hertz (see 5.1.1) are not valid identifier names. Other
restrictions on naming conventions apply as well, see 2.8.6.
The .param lines inside subcircuits are copied per call, like any other line. All assignments
are executed sequentially through the expanded circuit. Before its first use, a parameter name
must have been assigned a value. Expressions defining a parameter should be put within braces
{p+p2}, or alternatively within single quotes ’AGAUSS(pippo, 1, 1.67)’. An assignment
cannot be self-referential, something like .param pip = ’pip+3’ will not work.
The current ngspice version does not always need quotes or braces in expressions, especially
when spaces are used sparingly. However, it is recommended to do so, as the following examples demonstrate.
.param
.param
.param
.param
.param

2.8.2

a
a
c
c
c

=
=
=
=
=

123 * 3
b = sqrt (9) $ doesn ’t work , a <= 123
’123 * 3’ b = sqrt (9) $ ok.
a + 123
$ won ’t work
’a + 123 ’ $ ok.
a+123
$ ok.

Brace expressions in circuit elements:

General form:
{  }
Examples:
These are allowed in .model lines and in device lines. A SPICE number is a floating point
number with an optional scaling suffix, immediately glued to the numeric tokens (see Chapt.
2.8.5). Brace expressions ({..}) cannot be used to parametrize node names or parts of names.

2.8. .PARAM PARAMETRIC NETLISTS

57

All identifiers used within an  must have known values at the time when the line is
evaluated, else an error is flagged.

2.8.3

Subcircuit parameters

General form:
. subckt  node node ...

= = ...

Examples:
. subckt myfilter in out rval =100k cval =100 nF

 is the name of the subcircuit given by the user. node is an integer number or an
identifier, for one of the external nodes. The first = introduces an optional
section of the line. Each  is a formal parameter, and each  is either a SPICE
number or a brace expression. Inside the .subckt ... .ends context, each formal parameter
may be used like any identifier that was defined on a .param control line. The  parts
are supposed to be default values of the parameters. However, in the current version of , they
are not used and each invocation of the subcircuit must supply the _exact_ number of actual
parameters.
The syntax of a subcircuit call (invocation) is:
General form:
X node node ...  = = ...
Examples:
X1 input output myfilter rval =1k cval =1n

Here  is the symbolic name given to that instance of the subcircuit,  is the
name of a subcircuit defined beforehand. node node ... is the list of actual nodes where the
subcircuit is connected.  is either a SPICE number or a brace expression {  }
. The sequence of  items on the X line must exactly match the number and the order of
formal parameters of the subcircuit.

58

CHAPTER 2. CIRCUIT DESCRIPTION

Subcircuit example with parameters:
* Param - example
.param amplitude = 1V
*
. subckt myfilter in out rval =100k cval =100 nF
Ra in p1
{2* rval}
Rb p1 out {2* rval}
C1 p1 0
{2* cval}
Ca in p2
{cval}
Cb p2 out {cval}
R1 p2 0
{rval}
.ends myfilter
*
X1 input output myfilter rval =1k cval =1n
V1 input 0 AC { amplitude }
.end

2.8.4

Symbol scope

All subcircuit and model names are considered global and must be unique. The .param symbols
that are defined outside of any .subckt ... .ends section are global. Inside such a section, the
pertaining params: symbols and any .param assignments are considered local: they mask any
global identical names, until the .ends line is encountered. You cannot reassign to a global
number inside a .subckt, a local copy is created instead. Scope nesting works up to a level of
10. For example, if the main circuit calls A that has a formal parameter xx, A calls B that has a
param. xx, and B calls C that also has a formal param. xx, there will be three versions of ‘xx’
in the symbol table but only the most local one - belonging to C - is visible.

2.8.5

Syntax of expressions
 ( optional parts within [...] )

An expression may be one of:
 where  is either a spice number or an identifier
 
 (  [ ,  ...] )
  
(  )
As expected, atoms, built-in function calls and stuff within parentheses are evaluated before
the other operators. The operators are evaluated following a list of precedence close to the one
of the C language. For equal precedence binary ops, evaluation goes left to right. Functions
operate on real values only!

2.8. .PARAM PARAMETRIC NETLISTS
Operator
!
**
*
/
%
\
+
==
!=
<=
>=
<
>
&&
||
c?x:y

Alias

^

<>

Precedence
1
1
2
3
3
3
3
4
4
5
5
5
5
5
5
6
7
8

59

Description
unary unary not
power, like pwr
multiply
divide
modulo
integer divide
add
subtract
equality
non-equal
less or equal
greater or equal
less than
greater than
boolean and
boolean or
ternary operator

The number zero is used to represent boolean False. Any other number represents boolean True.
The result of logical operators is 1 or 0. An example input file is shown below:

Example input file with logical operators:
* Logical operators
v1or
v1and
v1not
v1mod
v1div
v0not

1
2
3
4
5
6

0
0
0
0
0
0

. control
op
print allv
.endc
.end

{1
{1
{!
{5
{5
{!

|| 0}
&& 0}
1}
% 3}
\ 3}
0}

60

CHAPTER 2. CIRCUIT DESCRIPTION

Built-in function
sqr(x)
sqrt(x)
sin(x), cos(x), tan(x)
sinh(x), cosh(x), tanh(x)
asin(x), acos(x), atan(x)
asinh(x), acosh(x), atanh(x)
arctan(x)
exp(x)
ln(x), log(x)
abs(x)
nint(x)
int(x)
floor(x)
ceil(x)
pow(x,y)
pwr(x,y)
min(x, y)
max(x, y)
sgn(x)
ternary_fcn(x, y, z)
gauss(nom, rvar, sigma)

agauss(nom, avar, sigma)

unif(nom, rvar)
aunif(nom, avar)
limit(nom, avar)

Notes
y = x * x
y = sqrt(x)

atan(x), kept for compatibility

Nearest integer, half integers towards even
Nearest integer rounded towards 0
Nearest integer rounded towards -∞
Nearest integer rounded towards +∞
x raised to the power of y (pow from C runtime library)
pow(fabs(x), y)

1.0 for x > 0, 0.0 for x == 0, -1.0 for x < 0
x ? y : z
nominal value plus variation drawn from Gaussian
distribution with mean 0 and standard deviation rvar
(relative to nominal), divided by sigma
nominal value plus variation drawn from Gaussian
distribution with mean 0 and standard deviation avar
(absolute), divided by sigma
nominal value plus relative variation (to nominal)
uniformly distributed between +/-rvar
nominal value plus absolute variation uniformly distributed
between +/-avar
nominal value +/-avar, depending on random number in
[-1, 1) being > 0 or < 0

The scaling suffixes (any decorative alphanumeric string may follow):
suffix
g
meg
k
m
u
n
p
f

value
1e9
1e6
1e3
1e-3
1e-6
1e-9
1e-12
1e-15

Note: there are intentional redundancies in expression syntax, e.g. x^y , x**y and pwr(x,y)
all have nearly the same result.

2.9. .FUNC

2.8.6

61

Reserved words

In addition to the above function names and to the verbose operators ( not and or div mod
), other words are reserved and cannot be used as parameter names: or, defined, sqr, sqrt,
sin, cos, exp, ln, log, log10, arctan, abs, pwr, time, temper, hertz.

2.8.7

Alternative syntax

The & sign is tolerated to provide some ‘historical’ parameter notation: & as the first character
of a line is equivalent to: .param.
Inside a line, the notation &(....) is equivalent to {....}, and &identifier means the same
thing as {identifier} .
Comments in the style of C++ line trailers (//) are detected and erased.
Warning: this is not possible in embedded .control parts of a source file, these lines are
outside of this scope.
Confusion may arise in ngspice because of its multiple numerical expression features. The
.param lines and the brace expressions (see Chapt. 2.9) are evaluated in the front-end, that
is, just after the subcircuit expansion. (Technically, the X lines are kept as comments in the
expanded circuit so that the actual parameters can be correctly substituted). Therefore, after the
netlist expansion and before the internal data setup, all number attributes in the circuit are known
constants. However, there are circuit elements in Spice that accept arithmetic expressions not
evaluated at this point, but only later during circuit analysis. These are the arbitrary current
and voltage sources (B-sources, 5), as well as E- and G-sources and R-, L-, or C-devices.
The syntactic difference is that ‘compile-time’ expressions are within braces, but ‘run-time’
expressions have no braces. To make things more complicated, the back-end ngspice scripting
language accepts arithmetic/logic expressions that operate only on its own scalar or vector data
sets (17.2). Please see Chapt. 2.13.
It would be desirable to have the same expression syntax, operator and function set, and precedence rules, for the three contexts mentioned above. In the current Numparam implementation,
that goal is not achieved.

2.9

.FUNC

This keyword defines a function. The syntax of the expression is the same as for a .param
(2.8.5).

62

CHAPTER 2. CIRCUIT DESCRIPTION

General form:
.func  {  }
.func  = {  }
Examples:
.func icos(x) {cos(x) - 1}
.func f(x,y) {x*y}
.func foo(a,b) = {a + b}
.func will initiate a replacement operation. After reading the input files, and before parameters
are evaluated, all occurrences of the icos(x) function will be replaced by cos(x)-1. All
occurrences of f(x,y) will be replaced by x*y. Function statements may be nested to a depth
of t.b.d..

2.10

.CSPARAM

Create a constant vector (see 17.8.2) from a parameter in plot (17.3) const.
General form:
. csparam  = 
Examples:
.param pippo =5
.param pp =6
. csparam pippp ={ pippo + pp}
.param p={ pp}
. csparam pap=’pp+p’
In the example shown, vectors pippp, and pap are added to the constants that already reside
in plot const, having length one and real values. These vectors are generated during circuit
parsing and thus cannot be changed later (same as with ordinary parameters). They may be used
in ngspice scripts and .control sections (see Chapt. 17).
The use of .csparam is still experimental and has to be tested. A simple usage is shown below.
* test csparam
.param TEMPS = 27
.csparam newt = {3*TEMPS}
.csparam mytemp = ’2 + TEMPS’
.control
echo $&newt $&mytemp
.endc
.end

2.11. .TEMP

2.11

63

.TEMP

Sets the circuit temperature in degrees Celsius.

General form:
.temp value
Examples:
.temp 27

This card overrides the circuit temperature given in an .option line (15.1.1).

2.12

.IF Condition-Controlled Netlist

A simple .IF-.ELSE(IF) block allows condition-controlling of the netlist. boolean expression
is any expression according to Chapt. 2.8.5 that evaluates parameters and returns a boolean 1
or 0. The netlist block in between the .if ... .endif statements may contain device instances or
.model cards that are selected according to the logic condition.

64

CHAPTER 2. CIRCUIT DESCRIPTION

General form:
.if( boolean expression )
...
. elseif ( boolean expression )
...
.else
...
.endif
Example 1:
* device instance in IF -ELSE block
.param ok =0 ok2 =1
v1 1 0 1
R1 1 0 2
.if (ok && ok2)
R11 1 0 2
.else
R11 1 0 0.5
$ <-- selected
.endif
Example 2:
* .model in IF -ELSE block
.param m0 =0 m1 =1
M1 1 2 3 4 N1 W=1 L=0.5
.if(m0 ==1)
.model N1 NMOS level =49 Version =3.1
. elseif (m1 ==1)
.model N1 NMOS level =49 Version =3.2.4
.else
.model N1 NMOS level =49 Version =3.3.0
.endif

$ <-- selected

For now this is a very restricted version of an .IF-.ELSE(IF) block, so several netlist components are currently not supported within the .IF-.ENDIF block: .SUBCKT, .INC, .LIB, and
.PARAM. Nesting of .IF-.ELSE(IF) blocks is not possible. Only one .elseif is allowed per
block.

2.13. PARAMETERS, FUNCTIONS, EXPRESSIONS, AND COMMAND SCRIPTS

2.13

65

Parameters, functions, expressions, and command scripts

In ngspice there are several ways to describe functional dependencies. In fact there are three
independent function parsers, being active before, during, and after the simulation. So it might
be due to have a few words on their interdependence.

2.13.1

Parameters

Parameters (Chapt. 2.8.1) and functions, either defined within the .param statement or with
the .func statement (Chapt. 2.9) are evaluated before any simulation is started, that is during
the setup of the input and the circuit. Therefore these statements may not contain any simulation output (voltage or current vectors), because it is simply not yet available. The syntax is
described in Chapt. 2.8.5. During the circuit setup all functions are evaluated, all parameters
are replaced by their resulting numerical values. Thus it will not be possible to get feedback
from a later stage (during or after simulation) to change any of the parameters.

2.13.2

Nonlinear sources

During the simulation, the B source (Chapt. 5) and their associated E and G sources, as well
as some devices (R, C, L) may contain expressions. These expressions may contain parameters
from above (evaluated immediately upon ngspice start up), numerical data, predefined functions, but also node voltages and branch currents resulting from the simulation. The source or
device values are continuously updated during the simulation. Therefore the sources are powerful tools to define non-linear behavior, you may even create new ‘devices’ by yourself.
Unfortunately the expression syntax (see Chapt. 5.1) and the predefined functions may deviate
from the ones for parameters listed in 2.8.1.

2.13.3

Control commands, Command scripts

Commands, as described in detail in Chapt. 17.5, may be used interactively, but also as a command script enclosed in .control ... .endc lines. The scripts may contain expressions
(see Chapt. 17.2). The expressions may work upon simulation output vectors (of node voltages, branch currents), as well as upon predefined or user defined vectors and variables, and are
invoked after the simulation. Parameters from 2.8.1 defined by the .param statement are not
allowed in these expressions. However you may define such parameters with .csparam (2.10).
Again the expression syntax (see Chapt. 17.2) will deviate from the one for parameters or B
sources listed in 2.8.1 and 5.1.
If you want to use parameters from 2.8.1 inside your control script, you may use .csparam
(2.10) or apply a trick by defining a voltage source with the parameter as its value, and then
have it available as a vector (e.g. after a transient simulation) with a then constant output (the
parameter). A feedback from here back into parameters (2.13.1) is never possible. Also you
cannot access non-linear sources of the preceding simulation. However you may start a first
simulation inside your control script, then evaluate its output using expressions, change some of
the element or model parameters with the alter and altermod statements (see Chapt. 17.5.3)
and then automatically start a new simulation.

66

CHAPTER 2. CIRCUIT DESCRIPTION

Expressions and scripting are powerful tools within ngspice, and we will enhance the examples
given in Chapt. 21 continuously to describe these features.

Chapter 3
Circuit Elements and Models
Data fields that are enclosed in less-than and greater-than signs (‘< >’) are optional. All indicated punctuation (parentheses, equal signs, etc.) is optional but indicate the presence of any
delimiter. Further, future implementations may require the punctuation as stated. A consistent style adhering to the punctuation shown here makes the input easier to understand. With
respect to branch voltages and currents, ngspice uniformly uses the associated reference convention (current flows in the direction of voltage drop).

3.1

General options and information

3.1.1

Paralleling devices with multiplier m

When it is needed to simulate several devices of the same kind in parallel, use the ‘m’ (parallel
multiplier) instance parameter available for the devices listed in Table 3.1. This multiplies the
value of the element’s matrix stamp with m’s value. The netlist below shows how to correctly
use the parallel multiplier:
Multiple device example:
d1 2 0 mydiode m=10
d01 1 0 mydiode
d02 1 0 mydiode
d03 1 0 mydiode
d04 1 0 mydiode
d05 1 0 mydiode
d06 1 0 mydiode
d07 1 0 mydiode
d08 1 0 mydiode
d09 1 0 mydiode
d10 1 0 mydiode
...
The d1 instance connected between nodes 2 and 0 is equivalent to the 10 parallel devices
d01-d10 connected between nodes 1 and 0.

67

68

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

The following devices support the multiplier m:
First letter
C
D
F
G
I
J
L
M
Q
R
X
Z

Element description
Capacitor
Diode
Current-controlled current source (CCCs)
Voltage-controlled current source (VCCS)
Current source
Junction field effect transistor (JFET)
Inductor
Metal oxide field effect transistor (MOSFET)
Bipolar junction transistor (BJT)
Resistor
Subcircuit (for details see below)
Metal semiconductor field effect transistor (MESFET)

Table 3.1: ngspice elements supporting multiplier ’m’

When the X line (e.g. x1 a b sub1 m=5) contains the token m=value (as shown) or m=expression,
subcircuit invocation is done in a special way. If an instance line of the subcircuit sub1 contains
any of the elements shown in table 3.1, then these elements are instantiated with the additional
parameter m (in this example having the value 5). If such an element already has an m multiplier parameter, the element m is multiplied with the m derived from the X line. This works
recursively, meaning that if a subcircuit contains another subcircuit (a nested X line), then the
latter m parameter will be multiplied by the former one, and so on.
Example 1:
.param madd = 6
X1 a b sub1 m=5
. subckt sub1 a1 b1
Cs1 a1 b1 C=5p m=’madd -2’
.ends
In example 1, the capacitance between nodes a and b will be C = 5pF*(madd-2)*5 = 100pF.
Example 2:
.param madd = 4
X1 a b sub1 m=3
. subckt sub1 a1 b1
X2 a1 b1 sub2 m=’madd -2’
.ends
. subckt sub2 a2 b2
Cs2 a2 b2 3p m=2
.ends
In example 2, the capacitance between nodes a and b is C = 3pF*2*(madd-2)*3 = 36pF.

3.1. GENERAL OPTIONS AND INFORMATION

69

Using m may fail to correctly describe geometrical properties for real devices like MOS transistors.
M1 d g s nmos W=0.3u L=0.18u m=20
is probably not be the same as
M1 d g s nmos W=6u L=0.18u
because the former may suffer from small width (or edge) effects, whereas the latter is simply
a wide transistor.

3.1.2

Technology scaling

Still to be implemented and written.

3.1.3

Model binning

Binning is a kind of range partitioning for geometry dependent models like MOSFET’s. The
purpose is to cover larger geometry ranges (Width and Length) with higher accuracy then the
model built-in geometry formulas. Each size range described by the additional model parameters LMIN, LMAX, WMIN and WMAX has its own model parameter set. These model cards
are defined by a number extension, like ‘nch.1’. NGSPICE has a algorithm to choose the right
model card by the requested W and L.
This is implemented for BSIM3 (11.2.9) and BSIM4 (11.2.10) models.

3.1.4

Initial conditions

Two different forms of initial conditions may be specified for some devices. The first form
is included to improve the dc convergence for circuits that contain more than one stable state.
If a device is specified OFF, the dc operating point is determined with the terminal voltages
for that device set to zero. After convergence is obtained, the program continues to iterate to
obtain the exact value for the terminal voltages. If a circuit has more than one dc stable state,
the OFF option can be used to force the solution to correspond to a desired state. If a device
is specified OFF when in reality the device is conducting, the program still obtains the correct
solution (assuming the solutions converge) but more iterations are required since the program
must independently converge to two separate solutions.
The .NODESET control line (see Chapt. 15.2.1) serves a similar purpose as the OFF option. The
.NODESET option is easier to apply and is the preferred means to aid convergence. The second
form of initial conditions are specified for use with the transient analysis. These are true ‘initial
conditions’ as opposed to the convergence aids above. See the description of the .IC control
line (Chapt. 15.2.2) and the .TRAN control line (Chapt. 15.3.9) for a detailed explanation of
initial conditions.

70

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

3.2

Elementary Devices

3.2.1

Resistors

General form:
RXXXXXXX n+ n- value  
+     
+ 
Examples:
R1 1 2
RC1 12
R2 5 7
RL 1 4

100
17 1K
1K ac =2K
2K m=2

Ngspice has a fairly complex model for resistors. It can simulate both discrete and semiconductor resistors. Semiconductor resistors in ngspice means: resistors described by geometrical
parameters. So, do not expect detailed modeling of semiconductor effects.
n+ and n- are the two element nodes, value is the resistance (in ohms) and may be positive or
negative1 but not zero.
Simulating small valued resistors: If you need to simulate very small resistors (0.001 Ohm or less), you should use CCVS (transresistance), it is less
efficient but improves overall numerical accuracy. Think about that a small
resistance is a large conductance.
Ngspice can assign a resistor instance a different value for AC analysis, specified using the
ac keyword. This value must not be zero as described above. The AC resistance is used in
AC analysis only (neither Pole-Zero nor Noise). If you do not specify the ac parameter, it is
defaulted to value.
Ngspice calculates the nominal resistance as

Rnom =

VALUE scale
m

Racnom =

ac scale
m .

(3.1)

If you want to simulate temperature dependence of a resistor, you need to specify its temperature
coefficients, using a .model line or as instance parameters, like in the examples below:
1A

negative resistor modeling an active element can cause convergence problems, please avoid it.

3.2. ELEMENTARY DEVICES

71

Examples:
RE1 1 2 800 newres dtemp =5
.MODEL newres R tc1 =0.001
RE2 a b 1.4k tc1 =2m tc2 =1.4u
RE3 n1 n2 1Meg tce =700m
The temperature coefficients tc1 and tc2 describe a quadratic temperature dependence (see
equation 1.6) of the resistance. If given in the instance line (the R... line) their values will
override the tc1 and tc2 of the .model line (3.2.3). Ngspice has an additional temperature
model equation 3.2 parameterised by tce given in model or instance line. If all parameters are
given (quadratic and exponential) the exponential temperature model is choosen.
h
i
R (T ) = R (T0 ) 1.01TCE·(T −T0 )

(3.2)

where T is the circuit temperature, T0 is the nominal temperature, and TCE is the exponential
temperature coefficients.
Instance temperature is useful even if resistance does not vary with it, since the thermal noise
generated by a resistor depends on its absolute temperature. Resistors in ngspice generates two
different noises: thermal and flicker. While thermal noise is always generated in the resistor, to
add a flicker noise2 source you have to add a .model card defining the flicker noise parameters.
It is possible to simulate resistors that do not generate any kind of noise using the noisy (or
noise) keyword and assigning zero to it, as in the following example:
Example:
Rmd 134 57 1.5k noisy =0
If you are interested in temperature effects or noise equations, read the next section on semiconductor resistors.

3.2.2

Semiconductor Resistors

General form:
RXXXXXXX n+ n-    
+     
+ 
Examples:
RLOAD 2 10 10K
RMOD 3 7 RMODEL L=10u W=1u
2 Flicker

noise can be used to model carbon resistors.

72

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

This is the more general form of the resistor presented before (3.2.1) and allows the modeling of
temperature effects and for the calculation of the actual resistance value from strictly geometric
information and the specifications of the process. If value is specified, it overrides the geometric information and defines the resistance. If mname is specified, then the resistance may be
calculated from the process information in the model mname and the given length and width.
If value is not specified, then mname and length must be specified. If width is not specified,
then it is taken from the default width given in the model.
The (optional) temp value is the temperature at which this device is to operate, and overrides
the temperature specification on the .option control line and the value specified in dtemp.

3.2.3

Semiconductor Resistor Model (R)

The resistor model consists of process-related device data that allow the resistance to be calculated from geometric information and to be corrected for temperature. The parameters available
are:
Name
Parameter
Units Default Example
Ω/◦C
TC1
first order temperature coeff.
0.0
◦
2
Ω/ C
TC2
second order temperature coeff.
0.0
Ω
RSH
sheet resistance
/
50
DEFW
default width
m
1e-6
2e-6
NARROW
narrowing due to side etching
m
0.0
1e-7
SHORT
shortening due to side etching
m
0.0
1e-7
◦
C
TNOM
parameter measurement temperature
27
50
KF
flicker noise coefficient
0.0
1e-25
AF
flicker noise exponent
0.0
1.0
WF
flicker noise width exponent
1.0
LF
flicker noise length exponent
1.0
EF
flicker noise frequnecy exponent
1.0
R (RES) default value if element value not given
Ω
1000
The sheet resistance is used with the narrowing parameter and l and w from the resistor device
to determine the nominal resistance by the formula:
Rnom = rsh

l − SHORT
w − NARROW

(3.3)

DEFW is used to supply a default value for w if one is not specified for the device. If either rsh
or l is not specified, then the standard default resistance value of 1 mOhm is used. TNOM is used
to override the circuit-wide value given on the .options control line where the parameters
of this model have been measured at a different temperature. After the nominal resistance is
calculated, it is adjusted for temperature by the formula:


R(T ) = R(TNOM) 1 + TC1 (T − TNOM) + TC2 (T − TNOM)2

(3.4)

where R(TNOM) = Rnom |Racnom . In the above formula, ‘T ’ represents the instance temperature,
which can be explicitly set using the temp keyword or calculated using the circuit temperature
and dtemp, if present. If both temp and dtemp are specified, the latter is ignored. Ngspice

3.2. ELEMENTARY DEVICES

73

improves SPICE’s resistors noise model, adding flicker noise (1/ f ) to it and the noisy (or
noise) keyword to simulate noiseless resistors. The thermal noise in resistors is modeled
according to the equation:
4kT
∆f
i¯2R =
R

(3.5)

where ‘k’ is the Boltzmann’s constant, and ‘T ’ the instance temperature.
Flicker noise model is:

i2R¯f n =

KFIRAF
∆f
W W F LLF f EF

(3.6)

A small list of sheet resistances (in Ω/) for conductors is shown below. The table represents
typical values for MOS processes in the 0.5 - 1 um
range. The table is taken from: N. Weste, K. Eshraghian - Principles of CMOS VLSI Design
2nd Edition, Addison Wesley.
Material
Inter-metal (metal1 - metal2)
Top-metal (metal3)
Polysilicon (poly)
Silicide
Diffusion (n+, p+)
Silicided diffusion
n-well

3.2.4

Min.
0.005
0.003
15
2
10
2
1000

Typ.
0.007
0.004
20
3
25
4
2000

Max.
0.1
0.05
30
6
100
10
5000

Resistors, dependent on expressions (behavioral resistor)

General form:
RXXXXXXX n+ n- R = ’expression ’  
RXXXXXXX n+ n- ’expression ’  
Examples:
R1 rr 0 r = ’V(rr) < {Vt} ? {R0} : {2* R0}’ tc1 =2e -03 tc2 =3.3e -06
R2 r2 rr r = {5k + 50* TEMPER }
Expression may be an equation or an expression containing node voltages or branch currents
(in the form of i(vm)) and any other terms as given for the B source and described in Chapt. 5.1.
It may contain parameters (2.8.1) and the special variables time, temper, and hertz (5.1.2).
An example file is given below.

74

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

Example input file for non-linear resistor:
Non - linear resistor
.param R0 =1k Vi =1 Vt =0.5
* resistor depending on control voltage V(rr)
R1 rr 0 r = ’V(rr) < {Vt} ? {R0} : {2* R0}’
* control voltage
V1 rr 0 PWL (0 0 100u {Vi})
. control
unset askquit
tran 100n 100u uic
plot i(V1)
.endc
.end

3.2.5

Capacitors

General form:
CXXXXXXX n+ n-     
+    
Examples:
CBYP 13 0 1UF
COSC 17 23 10U IC =3V
Ngspice provides a detailed model for capacitors. Capacitors in the netlist can be specified
giving their capacitance or their geometrical and physical characteristics. Following the original
SPICE3 ‘convention’, capacitors specified by their geometrical or physical characteristics are
called ‘semiconductor capacitors’ and are described in the next section.
In this first form n+ and n- are the positive and negative element nodes, respectively and value
is the capacitance in Farads.
Capacitance can be specified in the instance line as in the examples above or in a .model line,
as in the example below:
C1 15 5 cstd
C2 2 7 cstd
.model cstd C cap =3n
Both capacitors have a capacitance of 3nF.
If you want to simulate temperature dependence of a capacitor, you need to specify its temperature coefficients, using a .model line, like in the example below:

3.2. ELEMENTARY DEVICES

75

CEB 1 2 1u cap1 dtemp =5
.MODEL cap1 C tc1 =0.001
The (optional) initial condition is the initial (time zero) value of capacitor voltage (in Volts).
Note that the initial conditions (if any) apply only if the uic option is specified on the .tran
control line.
Ngspice calculates the nominal capacitance as described below:

Cnom = value · scale · m

(3.7)

The temperature coefficients tc1 and tc2 describe a quadratic temperature dependence (see
equation17.12) of the capacitance. If given in the instance line (the C... line) their values will
override the tc1 and tc2 of the .model line (3.2.7).

3.2.6

Semiconductor Capacitors

General form:
CXXXXXXX n+ n-     
+    
Examples:
CLOAD 2 10 10P
CMOD 3 7 CMODEL L=10u W=1u
This is the more general form of the Capacitor presented in section (3.2.5), and allows for the
calculation of the actual capacitance value from strictly geometric information and the specifications of the process. If value is specified, it defines the capacitance and both process and
geometrical information are discarded. If value is not specified, the capacitance is calculated from information contained model mname and the given length and width (l, w keywords,
respectively).
It is possible to specify mname only, without geometrical dimensions and set the capacitance in
the .model line (3.2.5).

3.2.7

Semiconductor Capacitor Model (C)

The capacitor model contains process information that may be used to compute the capacitance
from strictly geometric information.

76

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

Name
CAP
CJ
CJSW
DEFW
DEFL
NARROW
SHORT
TC1
TC2
TNOM
DI
THICK

Parameter
model capacitance
junction bottom capacitance
junction sidewall capacitance
default device width
default device length
narrowing due to side etching
shortening due to side etching
first order temperature coeff.
second order temperature coeff.
parameter measurement temperature
relative dielectric constant
insulator thickness

Units
F
F/m2
F/m
m
m
m
m
F/◦C
F/◦C2
◦C
F/m
m

Default
0.0
1e-6
0.0
0.0
0.0
0.0
0.0
27
0.0

Example
1e-6
5e-5
2e-11
2e-6
1e-6
1e-7
1e-7
0.001
0.0001
50
1
1e-9

The capacitor has a capacitance computed as:
If value is specified on the instance line then
Cnom = value · scale · m

(3.8)

If model capacitance is specified then
Cnom = CAP · scale · m

(3.9)

If neither value nor CAP are specified, then geometrical and physical parameters are take into
account:

C0 = CJ(l − SHORT)(w − NARROW) + 2CJSW(l − SHORT + w − NARROW)

(3.10)

CJ can be explicitly given on the .model line or calculated by physical parameters. When CJ
is not given, is calculated as:
If THICK is not zero:
CJ =

DI ε0
THICK

CJ =

εSiO2
THICK

if DI is specified,
(3.11)
otherwise.

If the relative dielectric constant is not specified the one for SiO2 is used. The values of the
F
F
constants are: ε0 = 8.854214871e − 12 m
and εSiO2 = 3.4531479969e − 11 m
. The nominal
capacitance is then computed as:
Cnom = C0 scale m

(3.12)

After the nominal capacitance is calculated, it is adjusted for temperature by the formula:


C(T ) = C(TNOM) 1 + TC1 (T − TNOM) + TC2 (T − TNOM)2

(3.13)

3.2. ELEMENTARY DEVICES

77

where C(TNOM) = Cnom .
In the above formula, ‘T ’ represents the instance temperature, which can be explicitly set using
the temp keyword or calculated using the circuit temperature and dtemp, if present.

3.2.8

Capacitors, dependent on expressions (behavioral capacitor)

General form:
CXXXXXXX n+ n- C = ’expression ’  
CXXXXXXX n+ n- ’expression ’  
Examples:
C1 cc 0 c = ’V(cc) < {Vt} ? {C1} : {Ch}’ tc1 =-1e -03 tc2 =1.3e -05
Expression may be an equation or an expression containing node voltages or branch currents
(in the form of i(vm)) and any other terms as given for the B source and described in Chapt. 5.1.
It may contain parameters (2.8.1) and the special variables time, temper, and hertz (5.1.2).
Example input file:
Behavioral Capacitor
.param Cl =5n Ch =1n Vt =1m Il =100n
.ic v(cc) = 0
v(cc2) = 0
* capacitor depending on control voltage V(cc)
C1 cc 0 c = ’V(cc) < {Vt} ? {Cl} : {Ch}’
*C1 cc 0 c ={ Ch}
I1 0 1 {Il}
Exxx n1 -copy n2 n2 cc2 1
Cxxx n1 -copy n2 1
Bxxx cc2 n2 I = ’(V(cc2) < {Vt} ? {Cl} : {Ch})’ * i(Exxx)
I2 n2 22 {Il}
vn2 n2 0 DC 0
* measure charge by integrating current
aint1 %id (1 cc) 2 time_count
aint2 %id (22 cc2) 3 time_count
.model time_count int( in_offset =0.0 gain =1.0
+ out_lower_limit =-1 e12 out_upper_limit =1 e12
+ limit_range =1e -9 out_ic =0.0)
. control
unset askquit
tran 100n 100u
plot v(2)
plot v(cc) v(cc2)
.endc
.end

78

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

3.2.9

Inductors

General form:
LYYYYYYY n+ n-    
+    
+  
Examples:
LLINK 42 69 1UH
LSHUNT 23 51 10U IC =15.7 MA
The inductor device implemented into ngspice has many enhancements over the original one.n+
and n- are the positive and negative element nodes, respectively. value is the inductance in
Henry. Inductance can be specified in the instance line as in the examples above or in a .model
line, as in the example below:
L1 15 5 indmod1
L2 2 7 indmod1
.model indmod1 L ind =3n
Both inductors have an inductance of 3nH.
The nt is used in conjunction with a .model line, and is used to specify the number of turns
of the inductor. If you want to simulate temperature dependence of an inductor, you need to
specify its temperature coefficients, using a .model line, like in the example below:
Lload 1 2 1u ind1 dtemp =5
.MODEL ind1 L tc1 =0.001
The (optional) initial condition is the initial (time zero) value of inductor current (in Amps) that
flows from n+, through the inductor, to n-. Note that the initial conditions (if any) apply only if
the UIC option is specified on the .tran analysis line.
Ngspice calculates the nominal inductance as described below:

Lnom =

3.2.10

value scale
m

(3.14)

Inductor model

The inductor model contains physical and geometrical information that may be used to compute
the inductance of some common topologies like solenoids and toroids, wound in air or other
material with constant magnetic permeability.

3.2. ELEMENTARY DEVICES
Name
IND
CSECT
LENGTH
TC1
TC2
TNOM
NT
MU

79

Parameter
model inductance
cross section
length
first order temperature coeff.
second order temperature coeff.
parameter measurement temperature
number of turns
relative magnetic permeability

Units
H
m2
m
H/◦C
H/◦C2
◦C
H/m

Default
0.0
0.0
0.0
0.0
0.0
27
0.0
0.0

Example
1e-3
1e-3
1e-2
0.001
0.0001
50
10
-

The inductor has an inductance computed as:
If value is specified on the instance line then

Lnom =

value scale
m

(3.15)

Lnom =

IND scale
m

(3.16)

If model inductance is specified then

If neither value nor IND are specified, then geometrical and physical parameters are take into
account. In the following formulas
NT refers to both instance and model parameter (instance parameter overrides model parameter):
If LENGTH is not zero:

(
Lnom =
Lnom =

MU µ0 NT2 CSECT
LENGTH
µ0 NT2 CSECT
LENGTH

if MU is specified,

(3.17)

otherwise.

with µ0 = 1.25663706143592 µH
m . After the nominal inductance is calculated, it is adjusted for
temperature by the formula



L(T ) = L(TNOM) 1 + TC1 (T − TNOM) + TC2 (T − TNOM)2 ,

(3.18)

where L(TNOM) = Lnom . In the above formula, ‘T ’ represents the instance temperature, which
can be explicitly set using the temp keyword or calculated using the circuit temperature and
dtemp, if present.

80

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

3.2.11

Coupled (Mutual) Inductors

General form:
KXXXXXXX LYYYYYYY LZZZZZZZ value
Examples:
K43 LAA LBB 0.999
KXFRMR L1 L2 0.87

LYYYYYYY and LZZZZZZZ are the names of the two coupled inductors, and value is the
coefficient of coupling, K, which must be greater than 0 and less than or equal to 1. Using the
‘dot’ convention, place a ‘dot’ on the first node of each inductor.

3.2.12

Inductors, dependent on expressions (behavioral inductor)

General form:
LXXXXXXX n+ n- L = ’expression ’  
LXXXXXXX n+ n- ’expression ’  
Examples:
L1 l2 lll L = ’i(Vm) < {It} ? {Ll} : {Lh}’ tc1 =-4e -03 tc2 =6e -05

Expression may be an equation or an expression containing node voltages or branch currents
(in the form of i(vm)) and any other terms as given for the B source and described in Chapt. 5.1.
It may contain parameters (2.8.1) and the special variables time, temper, and hertz (5.1.2).

3.2. ELEMENTARY DEVICES

81

Example input file:
Variable inductor
.param Ll =0.5m Lh =5m It =50u Vi=2m
.ic v( int21 ) = 0
* variable inductor depending on control current i(Vm)
L1 l2 lll L = ’i(Vm) < {It} ? {Ll} : {Lh}’
* measure current through inductor
vm lll 0 dc 0
* voltage on inductor
V1 l2 0 {Vi}
* fixed inductor
L3 33 331 {Ll}
* measure current through inductor
vm33 331 0 dc 0
* voltage on inductor
V3 33 0 {Vi}
* non linear inductor ( discrete setup)
F21 int21 0 B21 -1
L21 int21 0 1
B21 n1 n2 V = ’(i(Vm21) < {It} ? {Ll} : {Lh})’ * v(int21)
* measure current through inductor
vm21 n2 0 dc 0
V21 n1 0 {Vi}
. control
unset askquit
tran 1u 100u uic
plot i(Vm) i(vm33)
plot i(vm21) i(vm33)
plot i(vm)-i(vm21)
.endc
.end

3.2.13

Capacitor or inductor with initial conditions

The simulator supports the specification of voltage and current initial conditions on capacitor and inductor models, respectively. These models are not the standard ones supplied with
SPICE3, but are in fact code models that can be substituted for the SPICE models when realistic initial conditions are required. For details please refer to Chapter 12. A XSPICE deck
example using these models is shown below:
*
* This circuit contains a capacitor and an inductor with

82

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS
* initial conditions on them. Each of the components
* has a parallel resistor so that an exponential decay
* of the initial condition occurs with a time constant of
* 1 second.
*
a1 1 0 cap
.model cap capacitor (c=1000uf ic=1)
r1 1 0 1k
*
a2 2 0 ind
.model ind inductor (l=1H ic=1)
r2 2 0 1.0
*
.control
tran 0.01 3
plot v(1) v(2)
.endc
.end

3.2.14

Switches

Two types of switches are available: a voltage controlled switch (type SXXXXXX, model SW)
and a current controlled switch (type WXXXXXXX, model CSW). A switching hysteresis may
be defined, as well as on- and off-resistances (0 < R < ∞).
General form:
SXXXXXXX N+ N- NC+ NC - MODEL 
WYYYYYYY N+ N- VNAM MODEL 
Examples:
s1 1 2 3 4 switch1 ON
s2 5 6 3 0 sm2 off
Switch1 1 2 10 0 smodel1
w1 1 2 vclock switchmod1
W2 3 0 vramp sm1 ON
wreset 5 6 vclck lossyswitch OFF
Nodes 1 and 2 are the nodes between which the switch terminals are connected. The model
name is mandatory while the initial conditions are optional. For the voltage controlled switch,
nodes 3 and 4 are the positive and negative controlling nodes respectively. For the current
controlled switch, the controlling current is that through the specified voltage source. The
direction of positive controlling current flow is from the positive node, through the source, to
the negative node.
The instance parameters ON or OFF are required, when the controlling voltage (current) starts
inside the range of the hysteresis loop (different outputs during forward vs. backward voltage
or current ramp). Then ON or OFF determine the initial state of the switch.

3.2. ELEMENTARY DEVICES

3.2.15

83

Switch Model (SW/CSW)

The switch model allows an almost ideal switch to be described in ngspice. The switch is not
quite ideal, in that the resistance can not change from 0 to infinity, but must always have a finite
positive value. By proper selection of the on and off resistances, they can be effectively zero
and infinity in comparison to other circuit elements. The parameters available are:
Name
VT
IT
VH
IH
RON
ROFF

Parameter
threshold voltage
threshold current
hysteresis voltage
hysteresis current
on resistance
off resistance

Units
V
A
V
A
Ω
Ω

Default
0.0
0.0
0.0
0.0
1.0
1.0e+12 (*)

Switch model
SW
CSW
SW
CSW
SW,CSW
SW,CSW

(*) Or 1/GMIN, if you have set GMIN to any other value, see the .OPTIONS control line
(15.1.2) for a description of GMIN, its default value results in an off-resistance of 1.0e+12
ohms.
The use of an ideal element that is highly nonlinear such as a switch can cause large discontinuities to occur in the circuit node voltages. A rapid change such as that associated with a switch
changing state can cause numerical round-off or tolerance problems leading to erroneous results
or time step difficulties. The user of switches can improve the situation by taking the following
steps:

• First, it is wise to set ideal switch impedances just high or low enough to be negligible
with respect to other circuit elements. Using switch impedances that are close to ‘ideal’
in all cases aggravates the problem of discontinuities mentioned above. Of course, when
modeling real devices such as MOSFETS, the on resistance should be adjusted to a realistic level depending on the size of the device being modeled.

• If a wide range of ON to OFF resistance must be used in the switches (ROFF/RON >
1e+12), then the tolerance on errors allowed during transient analysis should be decreased
by using the .OPTIONS control line and specifying TRTOL to be less than the default value
of 7.0.

• When switches are placed around capacitors, then the option CHGTOL should also be reduced. Suggested values for these two options are 1.0 and 1e-16 respectively. These
changes inform ngspice to be more careful around the switch points so that no errors are
made due to the rapid change in the circuit.

84

CHAPTER 3. CIRCUIT ELEMENTS AND MODELS

Example input file:
Switch test
.tran 2us 5ms
* switch control voltage
v1 1 0 DC 0.0 PWL (0 0 2e-3 2 4e-3 0)
* switch control voltage starting inside hysteresis window
* please note influence of instance parameters ON , OFF
v2 2 0 DC 0.0 PWL (0 0.9 2e-3 2 4e-3 0.4)
* switch control current
i3 3 0 DC 0.0 PWL (0 0 2e-3 2m 4e-3 0) $ <--- switch control current
*load voltage
v4 4 0 DC 2.0
*input load for current source i3
r3 3 33 10k
vm3 33 0 dc 0 $ <--- measure the current
* ouput load resistors
r10 4 10 10k
r20 4 20 10k
r30 4 30 10k
r40 4 40 10k
*
s1 10 0 1 0 switch1 OFF
s2 20 0 2 0 switch1 OFF
s3 30 0 2 0 switch1 ON
.model switch1 sw vt =1 vh =0.2 ron =1 roff =10k
*
w1 40 0 vm3 wswitch1 off
.model wswitch1 csw it=1m ih =0.2m ron =1 roff =10k
*
. control
run
plot v(1) v(10)
plot v(10) vs v(1) $ <-- get hysteresis loop
plot v(2) v(20) $ <--- different initial values
plot v(20) vs v(2) $ <-- get hysteresis loop
plot v(2) v(30) $ <--- different initial values
plot v(30) vs v(2) $ <-- get hysteresis loop
plot v(40) vs vm3# branch $ <--- current controlled switch hysteresis
.endc
.end

Chapter 4
Voltage and Current Sources
4.1

Independent Sources for Voltage or Current

General form:
VXXXXXXX N+ N- < DC/TRAN VALUE > >>
+ >> >>
IYYYYYYY N+ N- < DC/TRAN VALUE > >>
+ >> >>
Examples:
VCC 10 0 DC 6
VIN 13 2 0.001 AC 1 SIN (0 1 1MEG)
ISRC 23 21 AC 0.333 45.0 SFFM (0 1 10K 5 1K)
VMEAS 12 9
VCARRIER 1 0 DISTOF1 0.1 -90.0
VMODULATOR 2 0 DISTOF2 0.01
IIN1 1 5 AC 1 DISTOF1 DISTOF2 0.001
n+ and n- are the positive and negative nodes, respectively. Note that voltage sources need not
be grounded. Positive current is assumed to flow from the positive node, through the source, to
the negative node. A current source of positive value forces current to flow out of the n+ node,
through the source, and into the n- node. Voltage sources, in addition to being used for circuit
excitation, are the ‘ammeters’ for ngspice, that is, zero valued voltage sources may be inserted
into the circuit for the purpose of measuring current. They of course have no effect on circuit
operation since they represent short-circuits.
DC/TRAN is the dc and transient analysis value of the source. If the source value is zero both for
dc and transient analyses, this value may be omitted. If the source value is time-invariant (e.g.,
a power supply), then the value may optionally be preceded by the letters DC.
ACMAG is the ac magnitude and ACPHASE is the ac phase. The source is set to this value in the
ac analysis. If ACMAG is omitted following the keyword AC, a value of unity is assumed. If
ACPHASE is omitted, a value of zero is assumed. If the source is not an ac small-signal input,
the keyword AC and the ac values are omitted.

85

86

CHAPTER 4. VOLTAGE AND CURRENT SOURCES

DISTOF1 and DISTOF2 are the keywords that specify that the independent source has distortion
inputs at the frequencies F1 and F2 respectively (see the description of the .DISTO control line).
The keywords may be followed by an optional magnitude and phase. The default values of the
magnitude and phase are 1.0 and 0.0 respectively.
Any independent source can be assigned a time-dependent value for transient analysis. If a
source is assigned a time-dependent value, the time-zero value is used for dc analysis. There
are nine independent source functions:
• pulse,
• exponential,
• sinusoidal,
• piece-wise linear,
• single-frequency FM
• AM
• transient noise
• random voltages or currents
• and external data (only with ngspice shared library).
If parameters other than source values are omitted or set to zero, the default values shown are
assumed. TSTEP is the printing increment and TSTOP is the final time – see the .TRAN control
line for an explanation.

4.1.1

Pulse

General form (the PHASE parameter is only possible when XSPICE is enabled):
PULSE (V1 V2 TD TR TF PW PER PHASE)
Examples:
VIN 3 0 PULSE (-1 1 2NS 2NS 2NS 50NS 100 NS)
Name
V1
V2
TD
TR
TF
PW
PER
PHASE

Parameter
Initial value
Pulsed value
Delay time
Rise time
Fall time
Pulse width
Period
Phase

Default Value
0.0
TSTEP
TSTEP
TSTOP
TSTOP
0.0

Units
V, A
V, A
sec
sec
sec
sec
sec
degrees

4.1. INDEPENDENT SOURCES FOR VOLTAGE OR CURRENT

87

A single pulse, without phase offset, is described by the following table:
Time
0
TD
TD+TR
TD+TR+PW
TD+TR+PW+TF
TSTOP

Value
V1
V1
V2
V2
V1
V1

Intermediate points are determined by linear interpolation.

4.1.2

Sinusoidal

General form (the PHASE parameter is only possible when XSPICE is enabled):
SIN(VO VA FREQ TD THETA PHASE)
Examples:
VIN 3 0 SIN (0 1 100 MEG 1NS 1E10)
Name
VO
VA
FREQ
TD
THETA
PHASE

Parameter
Offset
Amplitude
Frequency
Delay
Damping factor
Phase

Default Value
1/T ST OP
0.0
0.0
0.0

Units
V, A
V, A
Hz
sec
1/sec
degrees

The shape of the waveform is described by the following formula:
(
V0
if 0 ≤ t < T D
V (t) =
V 0 +VA e−(t−T D)T HETA sin (2π · FREQ · (t − T D) + PHASE) if T D ≤ t < T ST OP.
(4.1)

4.1.3

Exponential

General Form:
EXP(V1 V2 TD1 TAU1 TD2 TAU2)
Examples:
VIN 3 0 EXP (-4 -1 2NS 30NS 60NS 40NS)

88

CHAPTER 4. VOLTAGE AND CURRENT SOURCES

Name
V1
V2
TD1
TAU1
TD2
TAU2

Parameter
Initial value
pulsed value
rise delay time
rise time constant
fall delay time
fall time constant

Default Value
0.0
TSTEP
TD1+TSTEP
TSTEP

Units
V, A
V, A
sec
sec
sec
sec

The shape of the waveform is described by the following formula:
Let V 21 = V 2 −V 1,V 12 = V 1 −V 2:

V1
if 0 ≤ t < T D1,





(t−T D1)
− TAU1
if T D1 ≤ t < T D2,
V (t) = V 1 +V 21 1 − e





(t−T D1)
(t−T D2)

V 1 +V 21 1 − e− TAU1 +V 12 1 − e− TAU2
if T D2 ≤ t < T ST OP.

4.1.4

(4.2)

Piece-Wise Linear

General Form:
PWL(T1 V1 )  
Examples:
VCLOCK 7 5 PWL (0 -7 10NS -7 11NS -3 17NS -3 18NS -7 50NS -7)
+ r=0 td =15 NS
Each pair of values (Ti , Vi ) specifies that the value of the source is Vi (in Volts or Amps) at
time = Ti . The value of the source at intermediate values of time is determined by using linear
interpolation on the input values. The parameter r determines a repeat time point. If r is not
given, the whole sequence of values (Ti , Vi ) is issued once, then the output stays at its final
value. If r = 0, the whole sequence from time 0 to time Tn is repeated forever. If r = 10ns, the
sequence between 10ns and 50ns is repeated forever. the r value has to be one of the time points
T1 to Tn of the PWL sequence. If td is given, the whole PWL sequence is delayed by the value
of td.

4.1.5

Single-Frequency FM

General Form (the PHASE parameters are only possible when XSPICE is enabled):
SFFM(VO VA FC MDI FS PHASEC PHASES )
Examples:
V1 12 0 SFFM (0 1M 20K 5 1K)

4.1. INDEPENDENT SOURCES FOR VOLTAGE OR CURRENT
Name
VO
VA
FC
MDI
FS
PHASEC
PHASES

Parameter
Offset
Amplitude
Carrier frequency
Modulation index
Signal frequency
carrier phase
signal phase

Default value
1/T ST OP
1/T ST OP
0
0

89

Units
V, A
V, A
Hz
Hz
degrees
degrees

The shape of the waveform is described by the following equation:

V (t) = VO +VA sin (2π · FC · t + MDI sin (2π · FS · t + PHASES) + PHASEC)

4.1.6

(4.3)

Amplitude modulated source (AM)

General Form (the PHASE parameter is only possible when XSPICE is enabled):
AM(VA VO MF FC TD PHASES )
Examples:
V1 12 0 AM (0.5 1 20K 5MEG 1m)

Name
VA
VO
MF
FC
TD
PHASES

Parameter
Amplitude
Offset
Modulating frequency
Carrier frequency
Signal delay
Phase

Default value
1/T ST OP
0.0

Units
V, A
V, A
Hz
Hz
s
degrees

The shape of the waveform is described by the following equation:

V (t) = VA (VO + sin (2π · MF · t) + PHASES) sin (2π · FC · t + PHASES)

(4.4)

90

4.1.7

CHAPTER 4. VOLTAGE AND CURRENT SOURCES

Transient noise source

General Form:
TRNOISE (NA NT NALPHA NAMP RTSAM RTSCAPT RTSEMT )
Examples:
VNoiw 1 0 DC 0 TRNOISE (20n 0.5n 0 0)
VNoi1of 1 0 DC 0 TRNOISE (0 10p 1.1 12p)
VNoiw1of 1 0 DC 0 TRNOISE (20 10p 1.1 12p)
IALL 10 0 DC 0 trnoise (1m 1u 1.0 0.1m 15m

$ white
$ 1/f
$ white and 1/f
22u 50u)
$ white , 1/f, RTS

Transient noise is an experimental feature allowing (low frequency) transient noise injection
and analysis. See Chapt. 15.3.10 for a detailed description. NA is the Gaussian noise rms
voltage amplitude, NT is the time between sample values (breakpoints will be enforced on multiples of this value). NALPHA (exponent to the frequency dependency), NAMP (rms voltage or
current amplitude) are the parameters for 1/f noise, RTSAM the random telegraph signal amplitude, RTSCAPT the mean of the exponential distribution of the trap capture time, and RTSEMT
its emission time mean. White Gaussian, 1/f, and RTS noise may be combined into a single
statement.
Name
Parameter
Default value Units
NA
Rms noise amplitude (Gaussian)
V, A
NT
Time step
sec
NALPHA
1/f exponent
0<α <2
NAMP
Amplitude (1/f)
V, A
RTSAM
Amplitude
V, A
RTSCAPT
Trap capture time
sec
RTSEMT
Trap emission time
sec
If you set NT and RTSAM to 0, the noise option TRNOISE ... is ignored. Thus you may switch off
the noise contribution of an individual voltage source VNOI by the command
alter @vnoi[trnoise] = [ 0 0 0 0 ] $ no noise
alter @vrts[trnoise] = [ 0 0 0 0 0 0 0] $ no noise
See Chapt. 17.5.3 for the alter command.
You may switch off all TRNOISE noise sources by setting
set notrnoise
to your .spiceinit file (for all your simulations) or into your control section in front of the next
run or tran command (for this specific and all following simulations). The command
unset notrnoise
will reinstate all noise sources.
The noise generators are implemented into the independent voltage (vsrc) and current (isrc)
sources.

4.1. INDEPENDENT SOURCES FOR VOLTAGE OR CURRENT

4.1.8

91

Random voltage source

The TRRANDOM option yields statistically distributed voltage values, derived from the ngspice
random number generator. These values may be used in the transient simulation directly within
a circuit, e.g. for generating a specific noise voltage, but especially they may be used in the
control of behavioral sources (B, E, G sources 5, voltage controllable A sources 12, capacitors
3.2.8, inductors 3.2.12, or resistors 3.2.4) to simulate the circuit dependence on statistically varying device parameters. A Monte-Carlo simulation may thus be handled in a single simulation
run.
General Form:

TRRANDOM (TYPE TS >>)

Examples:

VR1 r1

0 dc 0 trrandom (2 10m 0 1) $ Gaussian

TYPE determines the random variates generated: 1 is uniformly distributed, 2 Gaussian, 3 exponential, 4 Poisson. TS is the duration of an individual voltage value. TD is a time delay with
0 V output before the random voltage values start up. PARAM1 and PARAM2 depend on the type
selected.
TYPE
1
2
3
4

4.1.9

description
Uniform
Gaussian
Exponential
Poisson

PARAM1
Range
Standard Dev.
Mean
Lambda

default
1
1
1
1

PARAM2
Offset
Mean
Offset
Offset

default
0
0
0
0

External voltage or current input

General Form:

EXTERNAL

Examples:

Vex 1 0 dc 0 external
Iex i1 i2 dc 0 external 

Voltages or currents may be set from the calling process, if ngspice is compiled as a shared
library and loaded by the process. See Chapt. 19.6.3 for an explanation.

92

CHAPTER 4. VOLTAGE AND CURRENT SOURCES

4.1.10

Arbitrary Phase Sources

The XSPICE option supports arbitrary phase independent sources that output at TIME=0.0 a
value corresponding to some specified phase shift. Other versions of SPICE use the TD (delay
time) parameter to set phase-shifted sources to their time-zero value until the delay time has
elapsed. The XSPICE phase parameter is specified in degrees and is included after the SPICE3
parameters normally used to specify an independent source. Partial XSPICE deck examples of
usage for pulse and sine waveforms are shown below:
* Phase shift is specified after Berkeley defined parameters
* on the independent source cards. Phase shift for both of the
* following is specified as +45 degrees
*
v1 1 0 0.0 sin(0 1 1k 0 0 45.0)
r1 1 0 1k
*
v2 2 0 0.0 pulse(-1 1 0 1e-5 1e-5 5e-4 1e-3 45.0)
r2 2 0 1k
*

4.2

Linear Dependent Sources

Ngspice allows circuits to contain linear dependent sources characterized by any of the four
equations
i = gv

v = ev

i = fi

v = hi

where g, e, f , and h are constants representing transconductance, voltage gain, current gain,
and transresistance, respectively. Non-linear dependent sources for voltages or currents (B, E,
G) are described in Chapt. 5.

4.2.1

Gxxxx: Linear Voltage-Controlled Current Sources (VCCS)

General form:
GXXXXXXX N+ N- NC+ NC - VALUE 
Examples:
G1 2 0 5 0 0.1
n+ and n- are the positive and negative nodes, respectively. Current flow is from the positive
node, through the source, to the negative
node. nc+ and nc- are the positive and negative controlling nodes, respectively. value is the
transconductance (in mhos). m is an optional multiplier to the output current. val may be a
numerical value or an expression according to 2.8.5 containing references to other parameters.

4.2. LINEAR DEPENDENT SOURCES

4.2.2

93

Exxxx: Linear Voltage-Controlled Voltage Sources (VCVS)

General form:
EXXXXXXX N+ N- NC+ NC - VALUE
Examples:
E1 2 3 14 1 2.0
n+ is the positive node, and n- is the negative node. nc+ and nc- are the positive and negative
controlling nodes, respectively. value is the voltage gain.

4.2.3

Fxxxx: Linear Current-Controlled Current Sources (CCCS)

General form:
FXXXXXXX N+ N- VNAM VALUE 
Examples:
F1 13 5 VSENS 5 m=2
n+ and n- are the positive and negative nodes, respectively. Current flow is from the positive
node, through the source, to the negative node. vnam is the name of a voltage source through
which the controlling current flows. The direction of positive controlling current flow is from
the positive node, through the source, to the negative node of vnam. value is the current gain.
m is an optional multiplier to the output current.

4.2.4

Hxxxx: Linear Current-Controlled Voltage Sources (CCVS)

General form:
HXXXXXXX n+ n- vnam value
Examples:
HX 5 17 VZ 0.5K
n+ and n- are the positive and negative nodes, respectively. vnam is the name of a voltage source
through which the controlling current flows. The direction of positive controlling current flow
is from the positive node, through the source, to the negative node of vnam. value is the
transresistance (in ohms).

94

4.2.5

CHAPTER 4. VOLTAGE AND CURRENT SOURCES

Polynomial Source Compatibility

Dependent polynomial sources available in SPICE2G6 are fully supported in ngspice using the
XSPICE extension (25.1). The form used to specify these sources is shown in Table 4.1. For
details on its usage please see Chapt. 5.2.4.

Source Type
POLYNOMIAL VCVS
POLYNOMIAL VCCS
POLYNOMIAL CCCS
POLYNOMIAL CCVS

Dependent Polynomial Sources
Instance Card
EXXXXXXX N+ N- POLY(ND) NC1+ NC1- P0 (P1...)
GXXXXXXX N+ N- POLY(ND) NC1+ NC1- P0 (P1...)
FXXXXXXX N+ N- POLY(ND) VNAM1 !VNAM2...? P0 (P1...)
HXXXXXXX N+ N- POLY(ND) VNAM1 !VNAM2...? P0 (P1...)
Table 4.1: Dependent Polynomial Sources

Chapter 5
Non-linear Dependent Sources (Behavioral
Sources)
The non-linear dependent sources B ( see Chapt. 5.1), E (see 5.2), G see (5.3) described in
this chapter allow to generate voltages or currents that result from evaluating a mathematical
expression. Internally E and G sources are converted to the more general B source. All three
sources may be used to introduce behavioral modeling and analysis.

5.1

Bxxxx: Nonlinear dependent source (ASRC)

5.1.1

Syntax and usage

General form:
BXXXXXXX n+ n-    
+  
Examples:
B1
B2
B3
B4
B5
+

0 1 I=cos(v (1))+ sin(v(2))
0 1 V=ln(cos(log(v(1 ,2)^2))) -v (3)^4+ v(2)^v(1)
3 4 I=17
3 4 V=exp(pi^i(vdd ))
2 0 V = V(1) < {Vlow} ? {Vlow} :
V(1) > { Vhigh } ? { Vhigh} : V(1)

n+ is the positive node, and n- is the negative node. The values of the V and I parameters
determine the voltages and currents across and through the device, respectively. If I is given
then the device is a current source, and if V is given the device is a voltage source. One and only
one of these parameters must be given.
A simple model is implemented for temperature behavior by the formula:


I(T ) = I(TNOM) 1 + TC1 (T − TNOM) + TC2 (T − TNOM)2

95

(5.1)

96

CHAPTER 5. NON-LINEAR DEPENDENT SOURCES (BEHAVIORAL SOURCES)

or



V (T ) = V (TNOM) 1 + TC1 (T − TNOM) + TC2 (T − TNOM)2

(5.2)

In the above formula, ‘T ’ represents the instance temperature, which can be explicitly set using
the temp keyword or calculated using the circuit temperature and dtemp, if present. If both
temp and dtemp are specified, the latter is ignored.
The small-signal AC behavior of the nonlinear source is a linear dependent source (or sources)
with a proportionality constant equal to the derivative (or derivatives) of the source at the DC
operating point. The expressions given for V and I may be any function of voltages and currents
through voltage sources in the system.
The following functions of a single real variable are defined:

Trigonometric functions: cos, sin, tan, acos, asin, atan
Hyperbolic functions: cosh, sinh, acosh, asinh, atanh
Exponential and logarithmic: exp, ln, log, log10 (ln, log with base e, log10 with base 10)
Other: abs, sqrt, u, u2, uramp, floor, ceil
Functions of two variables are: min, max, pow
Functions of three variables are: a ? b:c

The function ‘u’ is the unit step function, with a value of one for arguments greater than zero
and a value of zero for arguments less than zero. The function ‘u2’ returns a value of zero
for arguments less than zero, one for arguments greater than one and assumes the value of the
argument between these limits. The function ‘uramp’ is the integral of the unit step: for an
input x, the value is zero if x is less than zero, or if x is greater than zero the value is x. These
three functions are useful in synthesizing piece-wise non-linear functions, though convergence
may be adversely affected.
The following standard operators are defined: +, -, *, /, ^, unary Logical operators are !=, <>, >=, <=, ==, >, <, ||, &&, !

.

A ternary function is defined as a ? b : c , which means IF a, THEN b, ELSE c. Be
sure to place a space in front of ‘?’ to allow the parser distinguishing it from other tokens.

5.1. BXXXX: NONLINEAR DEPENDENT SOURCE (ASRC)

97

Example: Ternary function
* B source test Clamped voltage source
* C. P. Basso "Switched -mode power supplies ", New York , 2008
.param Vhigh = 4.6
.param Vlow = 0.4
Vin1 1 0 DC 0 PWL (0 0 1u 5)
Bcl 2 0 V = V(1) < Vlow ? Vlow : V(1) > Vhigh ? Vhigh : V(1)
. control
unset askquit
tran 5n 1u
plot V(2) vs V(1)
.endc
.end
If the argument of log, ln, or sqrt becomes less than zero, the absolute value of the argument is
used. If a divisor becomes zero or the argument of log or ln becomes zero, an error will result.
Other problems may occur when the argument for a function in a partial derivative enters a
region where that function is undefined.
Parameters may be used like {Vlow} shown in the example above. Parameters will be evaluated
upon set up of the circuit, vectors like V(1) will be evaluated during the simulation.
To get time into the expression you can integrate the current from a constant current source
with a capacitor and use the resulting voltage (don’t forget to set the initial voltage across the
capacitor).
Non-linear resistors, capacitors, and inductors may be synthesized with the nonlinear dependent
source. Nonlinear resistors, capacitors and inductors are implemented with their linear counterparts by a change of variables implemented with the nonlinear dependent source. The following
subcircuit will implement a nonlinear capacitor:
Example: Non linear capacitor
. Subckt nlcap pos neg
* Bx: calculate f( input voltage )
Bx 1 0 v = f(v(pos ,neg ))
* Cx: linear capacitance
Cx 2 0 1
* Vx: Ammeter to measure current into the capacitor
Vx 2 1 DC 0 Volts
* Drive the current through Cx back into the circuit
Fx pos neg Vx 1
.ends
Example for f(v(pos,neg)):
Bx 1 0 V = v(pos ,neg )*v(pos ,neg)
Non-linear resistors or inductors may be described in a similar manner. An example for a
nonlinear resistor using this template is shown below.

98

CHAPTER 5. NON-LINEAR DEPENDENT SOURCES (BEHAVIORAL SOURCES)

Example: Non linear resistor
* use of ’hertz ’ variable in nonlinear resistor
*. param rbase =1k
* some tests
B1 1 0 V = hertz *v(33)
B2 2 0 V = v (33)* hertz
b3 3 0 V = 6.283 e3 /( hertz +6.283 e3)*v(33)
V1 33 0 DC 0 AC 1
*** Translate R1 10 0 R=’1k/sqrt(HERTZ)’ to B source ***
. Subckt nlres pos neg rb=rbase
* Bx: calculate f( input voltage )
Bx
1
0
v = -1 / {rb} / sqrt(HERTZ) * v(pos , neg)
* Rx: linear resistance
Rx
2
0
1
Example: Non linear resistor (continued)
* Vx: Ammeter to measure current into the resistor
Vx
2
1
DC 0 Volts
* Drive the current through Rx back into the circuit
Fx
pos neg Vx 1
.ends
Xres 33 10 nlres rb =1k
*Rres 33 10 1k
Vres 10 0 DC 0
. control
define check (a,b) vecmax (abs(a - b))
ac lin 10 100 1k
* some checks
print v(1) v(2) v(3)
if check (v(1) , frequency ) < 1e -12
echo "INFO: ok"
end
plot vres# branch
.endc
.end

5.1.2

Special B-Source Variables time, temper, hertz

The special variables time and temper are available in a transient analysis, reflecting the actual
simulation time and circuit temperature. temper returns the circuit temperature, given in degree
C (see 2.11). The variable hertz is available in an AC analysis. time is zero in the AC analysis,
hertz is zero during transient analysis. Using the variable hertz may cost some CPU time if
you have a large circuit, because for each frequency the operating point has to be determined
before calculating the AC response.

5.1. BXXXX: NONLINEAR DEPENDENT SOURCE (ASRC)

5.1.3

99

par(’expression’)

The B source syntax may also be used in output lines like .plot as algebraic expressions for
output (see Chapt.15.6.6 ).

5.1.4

Piecewise Linear Function: pwl

Both B source types may contain a piece-wise linear dependency of one network variable:

Example: pwl_current
Bdio 1 0 I = pwl(v(A), 0,0, 33 ,10m, 100 ,33m, 200 ,50m)

v(A) is the independent variable x. Each pair of values following describes the x,y functional
relation: In this example at node A voltage of 0V the current of 0A is generated - next pair gives
10mA flowing from ground to node 1 at 33V on node A and so forth.

The same is possible for voltage sources:

Example: pwl_voltage
Blimit b 0 V = pwl(v(1), -4,0, -2,2, 2,4, 4,5, 6,5)

Monotony of the independent variable in the pwl definition is checked - non-monotonic x entries
will stop the program execution. v(1) may be replaced by a controlling current source. v(1) may
also be replaced by an expression, e.g. −2 i(Vin ). The value pairs may also be parameters, and
have to be predefined by a .param statement. An example for the pwl function using all of
these options is shown below.

100

CHAPTER 5. NON-LINEAR DEPENDENT SOURCES (BEHAVIORAL SOURCES)

Example: pwl function in B source
Demonstrates usage of the pwl function in an B source (ASRC)
* Also emulates the TABLE function with limits
.param
.param
.param
.param
.param
.param

x0=-4 y0 =0
x1=-2 y1 =2
x2 =2 y2=-2
x3 =4 y3 =1
xx0=x0 -1
xx3=x3 +1

Vin
1 0
R 1 0 2

DC =0V

* no limits outside of the tabulated x values
* ( continues linearily )
Btest2 2 0 I = pwl(v(1),’x0 ’,’y0 ’,’x1 ’,’y1 ’,’x2 ’,’y2 ’,’x3 ’,’y3 ’)
* like TABLE function with limits :
Btest3 3 0
I = (v(1) < ’x0 ’) ? ’y0 ’ :
(v(1) < ’x3 ’) ?
+ pwl(v(1),’x0 ’,’y0 ’,’x1 ’,’y1 ’,’x2 ’,’y2 ’,’x3 ’,’y3 ’) : ’y3 ’
* more efficient and elegant TABLE function with limits
*( voltage controlled ):
Btest4 4 0
I = pwl(v(1),
+ ’xx0 ’,’y0 ’, ’x0 ’,’y0 ’,
+
’x1 ’,’y1 ’,
+
’x2 ’,’y2 ’,
+
’x3 ’,’y3 ’, ’xx3 ’,’y3 ’)
*
* more efficient and elegant TABLE function with limits
* ( controlled by current ):
Btest5 5 0
I = pwl ( -2*i(Vin),
+ ’xx0 ’,’y0 ’, ’x0 ’,’y0 ’,
+
’x1 ’,’y1 ’,
+
’x2 ’,’y2 ’,
+
’x3 ’,’y3 ’, ’xx3 ’,’y3 ’)
Rint2 2 0 1
Rint3 3 0 1
Rint4 4 0 1
Rint5 5 0 1
. control
dc Vin -6 6 0.2
plot v(2) v(3) v(4) -0.5 v (5)+0.5
.endc
.end

5.2. EXXXX: NON-LINEAR VOLTAGE SOURCE2

5.2
5.2.1

101

Exxxx: non-linear voltage source1
VOL

General form:
EXXXXXXX n+ n- vol=’expr ’
Examples:
E41 4 0 vol = ’V(3)*V(3)-Offs ’
Expression may be an equation or an expression containing node voltages or branch currents
(in the form of i(vm)) and any other terms as given for the B source and described in Chapt.
5.1. It may contain parameters (2.8.1) and the special variables time, temper, hertz (5.1.2).
’ or { } may be used to delimit the function.

5.2.2

VALUE

Optional syntax:
EXXXXXXX n+ n- value ={ expr}
Examples:
E41 4 0 value = {V(3)*V(3)- Offs}
The ’=’ sign is optional.

5.2.3

TABLE

Data may be entered from the listings of a data table similar to the pwl B-Source (5.1.4). Data
are grouped into x, y pairs. Expression may be an equation or an expression containing node
voltages or branch currents (in the form of i(vm)) and any other terms as given for the B source
and described in Chapt. 5.1. It may contain parameters (2.8.1). ’ or { } may be used to delimit
the function. Expression delivers the x-value, which is used to generate a corresponding yvalue according to the tabulated value pairs, using linear interpolation. If the x-value is below
x0 , y0 is returned, above x2 y2 is returned (limiting function). The value pairs have to be real
numbers, parameters are not allowed.
1 To get this functionality, the compatibility mode has to be set in spinit or

.spiceinit by set ngbehavior=all.

102

CHAPTER 5. NON-LINEAR DEPENDENT SOURCES (BEHAVIORAL SOURCES)

Syntax for data entry from table:
Exxx n1 n2 TABLE { expression } = (x0 , y0) (x1 , y1) (x2 , y2)
Example (simple comparator):
ECMP 11 0 TABLE {V(10 ,9)} = (-5mV , 0V) (5mV , 5V)
An ’=’ sign may follow the keyword TABLE.

5.2.4

POLY

Polynomial sources are only available when the XSPICE option (see 32) is enabled.
General form:
EXXXX N+ N- POLY(ND) NC1+ NC1 - (NC2+ NC2 -...) P0 (P1 ...)
Example:
ENONLIN 100 101 POLY (2) 3 0 4 0 0.0 13.6 0.2 0.005
POLY(ND) Specifies the number of dimensions of the polynomial. The number of pairs of
controlling nodes must be equal to the number of dimensions.
(N+) and (N-) nodes are output nodes. Positive current flows from the (+) node through the
source to the (-) node.
The  and  are in pairs and define a set of controlling voltages. A particular node
can appear more than once, and the output and controlling nodes need not be different.
The example yields a voltage output controlled by two input voltages v(3,0) and v(4,0). Four
polynomial coefficients are given. The equivalent function to generate the output is:
0 + 13.6 * v(3) + 0.2 * v(4) + 0.005 * v(3) * v(3)
Generally you will set the equation according to
POLY(1) y = p0 +
POLY(2) y = p0 +
+
+
+
POLY(3) y = p0 +
+
+

k1*X1 + p2*X1*X1
+ p3*X1*X1*X1 + ...
p1*X1
+ p2*X2 +
p3*X1*X1
+ p4*X2*X1
+ p5*X2*X2
+
p6*X1*X1*X1
+ p7*X2*X1*X1 + p8*X2*X2*X1 +
p9*X2*X2*X2
+ ...
p1*X1
+ p2*X2
+ p3*X3
+
p4*X1*X1 + p5*X2*X1 + p6*X3*X1 +
p7*X2*X2 + p8*X2*X3 + p9*X3*X3 + ...

where X1 is the voltage difference of the first input node pair, X2 of the second pair and so on.
Keeping track of all polynomial coefficient is rather tedious for large polynomials.

5.2. EXXXX: NON-LINEAR VOLTAGE SOURCE4

5.2.5

103

LAPLACE

Currently ngspice does not offer a direct E-Source element with the LAPLACE option. There
is however a XSPICE code model equivalent called s_xfer (see Chapt. 12.2.16), which you
may invoke manually. The XSPICE option has to be enabled (32.1). AC (15.3.1) and transient
analysis (15.3.9) is supported.

The following E-Source:
ELOPASS 4 0 LAPLACE {V(1)}
+
{5 * (s/100 + 1) / (s ^2/42000 + s/60 + 1)}
may be replaced by:
AELOPASS 1 int_4 filter1
.model filter1 s_xfer (gain =5
+
num_coeff =[{1/100} 1]
+
den_coeff =[{1/42000} {1/60} 1]
+
int_ic =[0 0])
ELOPASS 4 0 int_4 0 1

where you have the voltage of node 1 as input, an intermediate output node int_4 and an Esource as buffer to keep the name ‘ELOPASS’ available if further processing is required.

If the controlling expression is more complex than just a voltage node, you may add a B-Source
(5.1) for evaluating the expression before entering the A-device.

E-Source with complex controlling expression:
ELOPASS 4 0 LAPLACE {V(1)*v(2)} {10 / (s/6800 + 1)}
may be replaced by:
BELOPASS int_1 0 V=V(1)*v(2)
AELOPASS int_1 int_4 filter1
.model filter1 s_xfer (gain =10
+
num_coeff =[1]
+
den_coeff =[{1/6800} 1]
+
int_ic =[0])
ELOPASS 4 0 int_4 0 1

104

5.3
5.3.1

CHAPTER 5. NON-LINEAR DEPENDENT SOURCES (BEHAVIORAL SOURCES)

Gxxxx: non-linear current source3
CUR

General form:
GXXXXXXX n+ n- cur=’expr ’ 
Examples:
G51 55 225 cur = ’V(3)*V(3)-Offs ’
Expression may be an equation or an expression containing node voltages or branch currents
(in the form of i(vm)) and any other terms as given for the B source and described in Chapt.
5.1. It may contain parameters (2.8.1) and special variables (5.1.2). m is an optional multiplier
to the output current. val may be a numerical value or an expression according to 2.8.5 containing only references to other parameters (no node voltages or branch currents!), because it is
evaluated before the simulation commences.

5.3.2

VALUE

Optional syntax:
GXXXXXXX n+ n- value =’expr ’ 
Examples:
G51 55 225 value = ’V(3)*V(3)-Offs ’
The ’=’ sign is optional.

5.3.3

TABLE

A data entry by a tabulated listing is available with syntax similar to the E-Source (see Chapt.
5.2.3).
3 To get this functionality, the compatibility mode has to be set in spinit or

.spiceinit by set ngbehavior=all.

5.3. GXXXX: NON-LINEAR CURRENT SOURCE5

105

Syntax for data entry from table:
Gxxx n1 n2 TABLE { expression } =
+ (x0 , y0) (x1 , y1) (x2 , y2) 
Example (simple comparator with current output and voltage control):
GCMP 0 11 TABLE {V(10 ,9)} = (-5MV , 0V) (5MV , 5V)
R 11 0 1k

m is an optional multiplier to the output current. val may be a numerical value or an expression
according to 2.8.5 containing only references to other parameters (no node voltages or branch
currents!), because it is evaluated before the simulation commences. An ’=’ sign may follow
the keyword TABLE.

5.3.4

POLY

see E-Source at Chapt. 5.2.4.

5.3.5

LAPLACE

See E-Source, Chapt. 5.2.5 , for an equivalent code model replacement.

5.3.6

Example

An example file is given below.

106

CHAPTER 5. NON-LINEAR DEPENDENT SOURCES (BEHAVIORAL SOURCES)

Example input file:
VCCS , VCVS , non - linear dependency
.param Vi =1
.param Offs = ’0.01*Vi ’
* VCCS depending on V(3)
B21 int1 0 V = V(3)*V(3)
G1 21 22 int1 0 1
* measure current through VCCS
vm 22 0 dc 0
R21 21 0 1
* new VCCS depending on V(3)
G51 55 225 cur = ’V(3)*V(3)-Offs ’
* measure current through VCCS
vm5 225 0 dc 0
R51 55 0 1
* VCVS depending on V(3)
B31 int2 0 V = V(3)*V(3)
E1 1 0 int2 0 1
R1 1 0 1
* new VCVS depending on V(3)
E41 4 0 vol = ’V(3)*V(3)-Offs ’
R4 4 0 1
* control voltage
V1 3 0 PWL (0 0 100u {Vi})
. control
unset askquit
tran 10n 100u uic
plot i(E1) i(E41)
plot i(vm) i(vm5)
.endc
.end

5.4

Debugging a behavioral source

The B, E, G, sources and the behavioral R, C, L elements are powerful tools to set up user
defined models. Unfortunately debugging these models is not very comfortable.

5.4. DEBUGGING A BEHAVIORAL SOURCE

107

Example input file with bug (log(-2)):
B source debugging
V1 1 0 1
V2 2 0 -2
E41 4 0 vol = ’V(1)* log(V(2)) ’
. control
tran 1 1
.endc
.end
The input file given above results in an error message:
Error:

-2 out of range for log

In this trivial example, the reason and location for the bug is obvious. However, if you have
several equations using behavioral sources, and several occurrences of the log function, then
debugging is nearly impossible.
However, if the variable ngdebug (see 17.7) is set (e.g. in file .spiceinit), a more distinctive
error message is issued that (after some closer investigation) will reveal the location and value
of the buggy parameter.
Detailed error message for input file with bug (log(-2)):
Error : -2 out of range for log
calling PTeval , tree =
(v0) * (log (v1))
d / d v0 : log (v1)
d / d v1 : (v0) * ((0.434294) / (v1))
values :
var0 = 1
var1 = -2
If variable strict_errorhandling (see 17.7) is set, ngspice exits after this message. If not,
gmin and source stepping may be started, typically without success.

108

CHAPTER 5. NON-LINEAR DEPENDENT SOURCES (BEHAVIORAL SOURCES)

Chapter 6
Transmission Lines
Ngspice implements both the original SPICE3f5 transmission lines models and the one introduced with KSPICE. The latter provide an improved transient analysis of lossy transmission lines.
Unlike SPICE models that use the state-based approach to simulate lossy transmission lines,
KSPICE simulates lossy transmission lines and coupled multiconductor line systems using the
recursive convolution method. The impulse response of an arbitrary transfer function can be
determined by deriving a recursive convolution from the Pade approximations of the function.
We use this approach for simulating each transmission line’s characteristics and each multiconductor line’s modal functions. This method of lossy transmission line simulation has been
proved to give a speedup of one to two orders of magnitude over SPICE3f5.

6.1

Lossless Transmission Lines

General form:
TXXXXXXX N1 N2 N3 N4 Z0=VALUE 
+ > 
Examples:
T1 1 0 2 0 Z0 =50 TD =10 NS
n1 and n2 are the nodes at port 1; n3 and n4 are the nodes at port 2. z0 is the characteristic
impedance. The length of the line may be expressed in either of two forms. The transmission
delay, td, may be specified directly (as td=10ns, for example). Alternatively, a frequency f
may be given, together with nl, the normalized electrical length of the transmission line with
respect to the wavelength in the line at the frequency ‘f’. If a frequency is specified but nl is
omitted, 0.25 is assumed (that is, the frequency is assumed to be the quarter-wave frequency).
Note that although both forms for expressing the line length are indicated as optional, one of
the two must be specified.
Note that this element models only one propagating mode. If all four nodes are distinct in the actual circuit, then two modes may be excited. To simulate such a situation, two transmission-line
elements are required. (see the example in Chapt. 21.7 for further clarification.) The (optional)

109

110

CHAPTER 6. TRANSMISSION LINES

initial condition specification consists of the voltage and current at each of the transmission line
ports. Note that the initial conditions (if any) apply only if the UIC option is specified on the
.TRAN control line.
Note that a lossy transmission line (see below) with zero loss may be more accurate than the
lossless transmission line due to implementation details.

6.2

Lossy Transmission Lines

General form:
OXXXXXXX n1 n2 n3 n4 mname
Examples:
O23 1 0 2 0 LOSSYMOD
OCONNECT 10 5 20 5 INTERCONNECT

This is a two-port convolution model for single conductor lossy transmission lines. n1 and n2
are the nodes at port 1; n3 and n4 are the nodes at port 2. Note that a lossy transmission line
with zero loss may be more accurate than the lossless transmission line due to implementation
details.

6.2.1

Lossy Transmission Line Model (LTRA)

The uniform RLC/RC/LC/RG transmission line model (referred to as the LTRA model henceforth) models a uniform constant-parameter distributed transmission line. The RC and LC
cases may also be modeled using the URC and TRA models; however, the newer LTRA model
is usually faster and more accurate than the others. The operation of the LTRA model is based
on the convolution of the transmission line’s impulse responses with its inputs (see [8]). The
LTRA model takes a number of parameters, some of which must be given and some of which
are optional.

6.2. LOSSY TRANSMISSION LINES
Name
R
L
G
C
LEN
REL
ABS
NOSTEPLIMIT
NO CONTROL
LININTERP
MIXEDINTERP
COMPACTREL
COMPACTABS
TRUNCNR
TRUNCDONTCUT

Parameter
resistance/length
inductance/length
conductance/length
capacitance/length
length of line
breakpoint control
breakpoint control
don’t limit time-step to less
than line delay
don’t do complex time-step
control
use linear interpolation
use linear when quadratic
seems bad
special reltol for history
compaction
special abstol for history
compaction
use Newton-Raphson method
for time-step control
don’t limit time-step to keep
impulse-response errors low

111
Units/Type
Ω/unit
H/unit
mhos/unit
F/unit
unit
arbitrary unit
flag

Default
0.0
0.0
0.0
0.0
no default
1
1
not set

Example
0.2
9.13e-9
0.0
3.65e-12
1.0
0.5
5
set

flag

not set

set

flag
flag

not set
not set

set
set

RELTOL

1.0e-3

ABSTOL

1.0e-9

flag

not set

set

flag

not set

set

The following types of lines have been implemented so far:
• RLC (uniform transmission line with series loss only),
• RC (uniform RC line),
• LC (lossless transmission line),
• RG (distributed series resistance and parallel conductance only).
Any other combination will yield erroneous results and should not be tried. The length LEN
of the line must be specified. NOSTEPLIMIT is a flag that will remove the default restriction
of limiting time-steps to less than the line delay in the RLC case. NO CONTROL is a flag that
prevents the default limiting of the time-step based on convolution error criteria in the RLC and
RC cases. This speeds up simulation but may in some cases reduce the accuracy of results.
LININTERP is a flag that, when specified, will use linear interpolation instead of the default
quadratic interpolation for calculating delayed signals. MIXEDINTERP is a flag that, when specified, uses a metric for judging whether quadratic interpolation is not applicable and if so uses
linear interpolation; otherwise it uses the default quadratic interpolation. TRUNCDONTCUT is a
flag that removes the default cutting of the time-step to limit errors in the actual calculation of
impulse-response related quantities. COMPACTREL and COMPACTABS are quantities that control
the compaction of the past history of values stored for convolution. Larger values of these lower
accuracy but usually increase simulation speed. These are to be used with the TRYTOCOMPACT
option, described in the .OPTIONS section. TRUNCNR is a flag that turns on the use of NewtonRaphson iterations to determine an appropriate time-step in the time-step control routines. The

112

CHAPTER 6. TRANSMISSION LINES

default is a trial and error procedure by cutting the previous time-step in half. REL and ABS are
quantities that control the setting of breakpoints.
The option most worth experimenting with for increasing the speed of simulation is REL. The
default value of 1 is usually safe from the point of view of accuracy but occasionally increases
computation time. A value greater than 2 eliminates all breakpoints and may be worth trying
depending on the nature of the rest of the circuit, keeping in mind that it might not be safe from
the viewpoint of accuracy.
Breakpoints may usually be entirely eliminated if it is expected the circuit will not display
sharp discontinuities. Values between 0 and 1 are usually not required but may be used for
setting many breakpoints.
COMPACTREL may also be experimented with when the option TRYTOCOMPACT is specified in
a .OPTIONS card. The legal range is between 0 and 1. Larger values usually decrease the
accuracy of the simulation but in some cases improve speed. If TRYTOCOMPACT is not specified
on a .OPTIONS card, history compaction is not attempted and accuracy is high.
NO CONTROL, TRUNCDONTCUT and NOSTEPLIMIT also tend to increase speed at the expense of
accuracy.

6.3

Uniform Distributed RC Lines

General form:
UXXXXXXX n1 n2 n3 mname l=len 
Examples:
U1 1 2 0 URCMOD L=50U
URC2 1 12 2 UMODL l=1 MIL N=6
n1 and n2 are the two element nodes the RC line connects, while n3 is the node the capacitances
are connected to. mname is the model name, len is the length of the RC line in meters. lumps,
if specified, is the number of lumped segments to use in modeling the RC line (see the model
description for the action taken if this parameter is omitted).

6.3.1

Uniform Distributed RC Model (URC)

The URC model is derived from a model proposed by L. Gertzberg in 1974. The model is
accomplished by a subcircuit type expansion of the URC line into a network of lumped RC
segments with internally generated nodes. The RC segments are in a geometric progression,
increasing toward the middle of the URC line, with K as a proportionality constant. The number of lumped segments used, if not specified for the URC line device, is determined by the
following formula:
log Fmax RL CL 2πL2
N=

log K

(K−1) 2
K

(6.1)

6.4. KSPICE LOSSY TRANSMISSION LINES

113

The URC line is made up strictly of resistor and capacitor segments unless the ISPERL parameter is given a nonzero value, in which case the capacitors are replaced with reverse biased diodes
with a zero-bias junction capacitance equivalent to the capacitance replaced, and with a saturation current of ISPERL amps per meter of transmission line and an optional series resistance
equivalent to RSPERL ohms per meter.
Name
K
FMAX
RPERL
CPERL
ISPERL
RSPERL

6.4

Parameter
Propagation Constant
Maximum Frequency of interest
Resistance per unit length
Capacitance per unit length
Saturation Current per unit length
Diode Resistance per unit length

Units
Hz
Ω/m
F/m
A/m
Ω/m

Default
2.0
1.0 G
1000
10e-15
0
0

Example
1.2
6.5 Meg
10
1p
-

Area
-

KSPICE Lossy Transmission Lines

Unlike SPICE3, which uses the state-based approach to simulate lossy transmission lines,
KSPICE simulates lossy transmission lines and coupled multiconductor line systems using the
recursive convolution method. The impulse response of an arbitrary transfer function can be
determined by deriving a recursive convolution from the Pade approximations of the function.
NGSPICE is using this approach for simulating each transmission line’s characteristics and each
multiconductor line’s modal functions. This method of lossy transmission line simulation has
shown to give a speedup of one to two orders of magnitude over SPICE3E. Please note that the
following two models will support only transient simulation, no ac.
Additional Documentation Available:
• S. Lin and E. S. Kuh, ‘Pade Approximation Applied to Transient Simulation of Lossy
Coupled Transmission Lines,’ Proc. IEEE Multi-Chip Module Conference, 1992, pp.
52-55.
• S. Lin, M. Marek-Sadowska, and E. S. Kuh, ‘SWEC: A StepWise Equivalent Conductance Timing Simulator for CMOS VLSI Circuits,’ European Design Automation Conf.,
February 1991, pp. 142-148.
• S. Lin and E. S. Kuh, ‘Transient Simulation of Lossy Interconnect,’ Proc. Design Automation Conference, Anaheim, CA, June 1992, pp. 81-86.

6.4.1

Single Lossy Transmission Line (TXL)

General form:
YXXXXXXX N1 0 N2 0 mname 
Example:
Y1 1 0 2 0 ymod LEN =2
.MODEL ymod txl R =12.45 L =8.972e-9 G=0 C =0.468e -12 length =16

114

CHAPTER 6. TRANSMISSION LINES

n1 and n2 are the nodes of the two ports. The optional instance parameter len is the length of
the line and may be expressed in multiples of [unit]. Typically unit is given in meters. len will
override the model parameter length for the specific instance only.
The TXL model takes a number of parameters:
Name
R
L
G
C
LENGTH

Parameter
resistance/length
inductance/length
conductance/length
capacitance/length
length of line

Units/Type
Ω/unit
H/unit
mhos/unit
F/unit
unit

Default
0.0
0.0
0.0
0.0
no default

Example
0.2
9.13e-9
0.0
3.65e-12
1.0

Model parameter length must be specified as a multiple of unit. Typically unit is given in [m].
For transient simulation only.

6.4.2

Coupled Multiconductor Line (CPL)

The CPL multiconductor line model is in theory similar to the RLGC model, but without frequency dependent loss (neither skin effect nor frequency-dependent dielectric loss). Up to 8
coupled lines are supported in NGSPICE.
General form:
PXXXXXXX NI1 NI2 ... NIX GND1 NO1 NO2 ... NOX GND2 mname 
Example:
P1 in1 in2 0 b1 b2 0 PLINE
.model PLINE CPL length ={ Len}
+R=1 0 1
+L={ L11} {L12} {L22}
+G=0 0 0
+C={ C11} {C12} {C22}
.param Len =1 Rs =0
+ C11 =9.143579E -11 C12 = -9.78265E -12 C22 =9.143578E -11
+ L11 =3.83572E -7 L12 =8.26253E-8 L22 =3.83572E-7
ni1 ... nix are the nodes at port 1 with gnd1; no1 ... nox are the nodes at port 2 with gnd2.
The optional instance parameter len is the length of the line and may be expressed in multiples
of [unit]. Typically unit is given in meters. len will override the model parameter length for
the specific instance only.
The CPL model takes a number of parameters:
Name
R
L
G
C
LENGTH

Parameter
resistance/length
inductance/length
conductance/length
capacitance/length
length of line

Units/Type
Ω/unit
H/unit
mhos/unit
F/unit
unit

Default
0.0
0.0
0.0
0.0
no default

Example
0.2
9.13e-9
0.0
3.65e-12
1.0

6.4. KSPICE LOSSY TRANSMISSION LINES

115

All RLGC parameters are given in Maxwell matrix form. For the R and G matrices the diagonal
elements must be specified, for L and C matrices the lower or upper triangular elements must
specified. The parameter LENGTH is a scalar and is mandatory. For transient simulation only.

116

CHAPTER 6. TRANSMISSION LINES

Chapter 7
Diodes
7.1

Junction Diodes

General form:
DXXXXXXX n+ n- mname    
+
  
Examples:
DBRIDGE 2 10 DIODE1
DCLMP 3 7 DMOD AREA =3.0 IC =0.2
The pn junction (diode) implemented in ngspice expands the one found in SPICE3f5. Perimeter
effects and high injection level have been introduced into the original model and temperature
dependence of some parameters has been added. n+ and n- are the positive and negative nodes,
respectively. mname is the model name. Instance parameters may follow, dedicated to only
the diode described on the respective line. area is the area scale factor, which may scale
the saturation current given by the model parameters (and others, see table below). pj is the
perimeter scale factor, scaling the sidewall saturation current and its associated capacitance. m
is a multiplier of area and perimeter, and off indicates an (optional) starting condition on the
device for dc analysis. If the area factor is omitted, a value of 1.0 is assumed. The (optional)
initial condition specification using ic is intended for use with the uic option on the .tran
control line, when a transient analysis is desired starting from other than the quiescent operating
point. You should supply the initial voltage across the diode there. The (optional) temp value
is the temperature at which this device is to operate, and overrides the temperature specification
on the .option control line. The temperature of each instance can be specified as an offset to
the circuit temperature with the dtemp option.

7.2

Diode Model (D)

The dc characteristics of the diode are determined by the parameters is and n. An ohmic
resistance, rs, is included. Charge storage effects are modeled by a transit time, tt, and a

117

118

CHAPTER 7. DIODES

nonlinear depletion layer capacitance that is determined by the parameters cjo, vj, and m. The
temperature dependence of the saturation current is defined by the parameters eg, the energy,
and xti, the saturation current temperature exponent. The nominal temperature where these
parameters were measured is tnom, which defaults to the circuit-wide value specified on the
.options control line. Reverse breakdown is modeled by an exponential increase in the reverse
diode current and is determined by the parameters bv and ibv (both of which are positive
numbers).

Junction DC parameters

Name
BV
IBV
IK (IKF)
IKR
IS (JS)
JSW
N
RS

Parameter
Reverse breakdown voltage
Current at breakdown voltage
Forward knee current
Reverse knee current
Saturation current
Sidewall saturation current
Emission coefficient
Ohmic resistance

Units
V
A
A
A
A
A
Ω

Default
∞
1.0e-3
1.0e-3
1.0e-3
1.0e-14
1.0e-14
1
0.0

Example
40
1.0e-4
1.0e-6
1.0e-6
1.0e-16
1.0e-15
1.5
100

Scale factor

Parameter
Zero-bias junction bottom-wall
capacitance
Zero-bias junction sidewall
capacitance
Coefficient for forward-bias
depletion bottom-wall capacitance
formula
Coefficient for forward-bias
depletion sidewall capacitance
formula
Area junction grading coefficient
Periphery junction grading
coefficient
Junction potential
Periphery junction potential
Transit-time

Units
F

Default
0.0

Example
2pF

F

0.0

.1pF

-

0.5

-

-

0.5

-

-

0.5
0.33

0.5
0.5

V
V
sec

1
1
0

0.6
0.6
0.1ns

area
perimeter
1/area

Junction capacitance parameters

Name
CJO (CJ0)
CJP (CJSW)
FC

FCS

M (MJ)
MJSW
VJ (PB)
PHP
TT

Scale factor
area
perimeter

7.3. DIODE EQUATIONS

119

Temperature effects
Name

Parameter

Units

Default

EG

Activation energy

eV

1.11

TM1
TM2
TNOM (TREF)
TRS1 (TRS)
TRS2
TM1
TM2
TTT1
TTT2

1st order tempco for MJ
2nd order tempco for MJ
Parameter measurement temperature
1st order tempco for RS
2nd order tempco for RS
1st order tempco for MJ
2nd order tempco for MJ
1st order tempco for TT
2nd order tempco for TT

1/◦C

1/◦C2

0.0
0.0
27
0.0
0.0
0.0
0.0
0.0
0.0

XTI

Saturation current temperature exponent

-

3.0

TLEV
TLEVC
CTA (CTC)
CTP
TCV

Diode temperature equation selector
Diode capac. temperature equation selector
Area junct. cap. temperature coefficient
Perimeter junct. cap. temperature coefficient
Breakdown voltage temperature coefficient

1/◦C
1/◦C
1/◦C

0
0
0.0
0.0
0.0

◦C 2

1/
◦C

1/◦C
1/◦C2
1/◦C
1/◦C2
1/◦C

Example
1.11 Si
0.69 Sbd
0.67 Ge
50
3.0 pn
2.0 Sbd

-

Noise modeling
Name
KF
AF

Parameter
Flicker noise coefficient
Flicker noise exponent

Units
-

Default
0
1

Example

Scale factor

Diode models may be described in the input file (or an file included by .inc) according to the
following example:
General form:
.model mname type( pname1 =pval1 pname2 =pval2 ... )
Examples:
.model DMOD D (bf =50 is=1e -13 vbf =50)

7.3

Diode Equations

The junction diode is the basic semiconductor device and the simplest one in ngspice, but its
model is quite complex, even when not all the physical phenomena affecting a pn junction are
handled. The diode is modeled in three different regions:

120

CHAPTER 7. DIODES
• Forward bias: the anode is more positive than the cathode, the diode is ‘on’ and can
conduct large currents. To avoid convergence problems and unrealistic high current, it is
prudent to specify a series resistance to limit current with the rs model parameter.
• Reverse bias: the cathode is more positive than the anode and the diode is ‘off’. A reverse
bias diode conducts a small leakage current.
• Breakdown: the breakdown region is modeled only if the bv model parameter is given.
When a diode enters breakdown the current increases exponentially (remember to limit
it); bv is a positive value.

Parameters Scaling
Model parameters are scaled using the unit-less parameters area and pj and the multiplier m as
depicted below:
AREAe f f = AREA m
PJe f f = PJ m
ISe f f = IS AREAe f f + JSW PJe f f
IBVe f f = IBV AREAe f f
IKe f f = IK AREAe f f
IKRe f f = IKR AREAe f f
CJe f f = CJ0 AREAe f f
CJPe f f = CJP PJe f f
Diode DC, Transient and AC model equations

qVD

NkT
if VD ≥ −3 NkT

q
ISe f f (e − 1) +VD · GMIN,
3NkT 3
ID = −ISe f f [1 + ( qVD e ) ] +VD · GMIN, if − BVe f f < VD < −3 NkT
q

−q(BVe f f +VD )


NkT
−ISe f f (e
) +VD · GMIN, if VD ≤ −BVe f f

(7.1)

The breakdown region must be described with more depth since the breakdown is not modeled
physically. As written before, the breakdown modeling is based on two model parameters: the
‘nominal breakdown voltage’ bv and the current at the onset of breakdown ibv. For the diode
model to be consistent, the current value cannot be arbitrarily chosen, since the reverse bias and
breakdown regions must match. When the diode enters breakdown region from reverse bias,
the current is calculated using the formula1 :
Ibdwn = −ISe f f (e

−qBV
NkT

− 1)

(7.2)

The computed current is necessary to adjust the breakdown voltage making the two regions
match. The algorithm is a little bit convoluted and only a brief description is given here:
1 if

you look at the source code in file diotemp.c you will discover that the exponential relation is replaced
with a first order Taylor series expansion.

7.3. DIODE EQUATIONS

121

if IBVe f f < Ibdwn then
IBVe f f = Ibdwn
BVe f f = BV
else
IBVe f f
BVe f f = BV − NVt ln( Ibdwn
)
Algorithm 2: Diode breakdown current calculation
Most real diodes shows a current increase that, at high current levels, does not follow the exponential relationship given above. This behavior is due to high level of carriers injected into the
junction. High injection effects (as they are called) are modeled with ik and ikr.

IDe f f =



rID
,


 1+ IKID

if VD ≥ −3 NkT
q

ef f

rID



 1+ IKRID

, otherwise.

(7.3)

ef f

Diode capacitance is divided into two different terms:
• Depletion capacitance
• Diffusion capacitance
Depletion capacitance is composed by two different contributes, one associated to the bottom
of the junction (bottom-wall depletion capacitance) and the other to the periphery (sidewall
depletion capacitance). The basic equations are:
CDiode = Cdi f f usion +Cdepletion
Where the depletion capacitance is defined as:
Cdepletion = Cdeplbw +Cdeplsw
The diffusion capacitance, due to the injected minority carriers, is modeled with the transit time
tt:
Cdi f f usion = TT

∂ IDe f f
∂VD

The depletion capacitance is more complex to model, since the function used to approximate it
diverges when the diode voltage become greater than the junction built-in potential. To avoid
function divergence, the capacitance function is approximated with a linear extrapolation for
applied voltage greater than a fraction of the junction built-in potential.

Cdeplbw =


CJe f f (1 − VD )−MJ ,
VJ

if VD < FC · VJ
V

D
VJ
CJe f f 1−FC(1+MJI)+MJ
(1−FC)(1+MJ)

, otherwise.

(7.4)

122

CHAPTER 7. DIODES

Cdeplsw =


CJPe f f (1 − VD )−MJSW ,
PHP
CJPe f f

if VD < FCS · PHP

VD
1−FCS(1+MJSW)+MJSW· PHP
(1+MJSW)
(1−FCS)

, otherwise.

(7.5)

Temperature dependence
The temperature affects many of the parameters in the equations above, and the following equations show how. One of the most significant parameters that varies with the temperature for a
semiconductor is the band-gap energy:
EGnom = 1.16 − 7.02e−4

TNOM2
TNOM + 1108.0

(7.6)

−4

T2
TNOM + 1108.0

(7.7)

EG(T ) = 1.16 − 7.02e

The leakage current temperature’s dependence is:
IS(T ) = IS e

log f actor
N

JSW (T ) = JSW e

log f actor
N

(7.8)

(7.9)

where ‘logfactor’ is defined as
log f actor =

EG
T
EG
−
+ XTI ln(
)
Vt (TNOM) Vt (T )
TNOM

(7.10)

The contact potentials (bottom-wall an sidewall) temperature dependence is:


T
T
EGnom
EG(T)
V J(T ) = VJ(
) −Vt (T ) 3 · ln(
)+
−
TNOM
TNOM
Vt (TNOM) Vt (T )

(7.11)



T
T
EGnom
EG(T)
PHP(T ) = PHP(
) −Vt (T ) 3 · ln(
)+
−
TNOM
TNOM
Vt (TNOM) Vt (T )

(7.12)

The depletion capacitances temperature dependence is:


V J(T )
−4
CJ(T ) = CJ 1 + MJ(4.0e (T − TNOM) −
+ 1)
VJ


PHP(T )
−4
CJSW (T ) = CJSW 1 + MJSW(4.0e (T − TNOM) −
+ 1)
PHP

(7.13)

(7.14)

The transit time temperature dependence is:
T T (T ) = TT(1 + TTT1(T − TNOM) + TTT2(T − TNOM)2 )

(7.15)

7.3. DIODE EQUATIONS

123

The junction grading coefficient temperature dependence is:
MJ(T ) = MJ(1 + TM1(T − TNOM) + TM2(T − TNOM)2 )

(7.16)

The series resistance temperature dependence is:
RS(T ) = RS(1 + TRS(T − TNOM) + TRS2(T − TNOM)2 )

(7.17)

Noise model
The diode has three noise contribution, one due to the presence of the parasitic resistance rs
and the other two (shot and flicker) due to the pn junction.
The thermal noise due to the parasitic resistance is:
i2RS =

4kT ∆ f
RS

(7.18)

The shot and flicker noise contributions are:
i2d = 2qID ∆ f +

KF · IDAF
∆f
f

(7.19)

124

CHAPTER 7. DIODES

Chapter 8
BJTs
8.1

Bipolar Junction Transistors (BJTs)

General form:
QXXXXXXX nc nb ne  mname  
+     
+ 
Examples:
Q23 10 24 13 QMOD IC =0.6 , 5.0
Q50A 11 26 4 20 MOD1
nc, nb, and ne are the collector, base, and emitter nodes, respectively. ns is the (optional) substrate node. When unspecified, ground is used. mname is the model name, area, areab, areac
are the area factors (emitter, base and collector respectively), and off indicates an (optional)
initial condition on the device for the dc analysis. If the area factor is omitted, a value of 1.0 is
assumed.
The (optional) initial condition specification using ic=vbe,vce is intended for use with the
uic option on a .tran control line, when a transient analysis is desired to start from other
than the quiescent operating point. See the .ic control line description for a better way to set
transient initial conditions. The (optional) temp value is the temperature where this device is
to operate, and overrides the temperature specification on the .option control line. Using the
dtemp option one can specify the instance’s temperature relative to the circuit temperature.

8.2

BJT Models (NPN/PNP)

Ngspice provides three BJT device models, which are selected by the .model card.
.model QMOD1 BJT level=2
This is the minimal version, further optional parameters listed in the table below may replace
the ngspice default parameters. The level keyword specifies the model to be used:

125

126

CHAPTER 8. BJTS
• level=1 : This is the original SPICE BJT model, and it is the default model if the level
keyword is not specified on the .model line.
• level=2 : This is a modified version of the original SPICE BJT that models both vertical
and lateral devices and includes temperature corrections of collector, emitter and base
resistors.
• level=4: Advanced VBIC model (see http://www.designers-guide.org/VBIC/ for details)

The bipolar junction transistor model in ngspice is an adaptation of the integral charge control
model of Gummel and Poon. This modified Gummel-Poon model extends the original model
to include several effects at high bias levels. The model automatically simplifies to the simpler
Ebers-Moll model when certain parameters are not specified. The parameter names used in the
modified Gummel-Poon model have been chosen to be more easily understood by the user, and
to reflect better both physical and circuit design thinking.
The dc model is defined by the parameters is, bf, nf, ise, ikf, and ne, which determine
the forward current gain characteristics, is, br, nr, isc, ikr, and nc, which determine the
reverse current gain characteristics, and vaf and var, which determine the output conductance
for forward and reverse regions.
The level 1 model has among the standard temperature parameters an extension compatible with
most foundry provided process design kits (see parameter table below tlev).
The level 1 and 2 models include the substrate saturation current iss. Three ohmic resistances
rb, rc, and re are included, where rb can be high current dependent. Base charge storage is
modeled by forward and reverse transit times, tf and tr, where the forward transit time tf can
be bias dependent if desired. Nonlinear depletion layer capacitances are defined with cje, vje,
and nje for the B-E junction, cjc, vjc, and njc for the B-C junction and cjs, vjs, and mjs
for the C-S (collector-substrate) junction.
The level 1 and 2 model support a substrate capacitance that is connected to the device’s base or
collector, to model lateral or vertical devices dependent on the parameter subs. The temperature
dependence of the saturation currents, is and iss (for the level 2 model), is determined by the
energy-gap, eg, and the saturation current temperature exponent, xti.
In the new model, additional base current temperature dependence is modeled by the beta temperature exponent xtb. The values specified are assumed to have been measured at the temperature tnom, which can be specified on the .options control line or overridden by a specification
on the .model line.
The level 4 model (VBIC) has the following improvements beyond the GP models: improved Early effect modeling, quasi-saturation modeling, parasitic substrate transistor modeling,
parasitic fixed (oxide) capacitance modeling, includes an avalanche multiplication model, improved temperature modeling, base current is decoupled from collector current, electrothermal
modeling, smooth and continuous mode.
The BJT parameters used in the modified Gummel-Poon model are listed below. The parameter
names used in earlier versions of SPICE2 are still accepted.
Gummel-Poon BJT Parameters (incl. model extensions)

8.2. BJT MODELS (NPN/PNP)

Name
SUBS

IS
ISS

BF
NF
VAF (VA)
IKF
NKF
ISE
NE
BR
NR
VAR (VB)
IKR
ISC

NC
RB
IRB
RBM
RE
RC
CJE
VJE (PE)
MJE (ME)
TF
XTF
VTF
ITF

Parameters
Substrate connection: for vertical
geometry, -1 for lateral geometry
(level 2 only).
Transport saturation current.
Reverse saturation current,
substrate-to-collector for vertical
device or substrate-to-base for
lateral (level 2 only).
Ideal maximum forward beta.
Forward current emission
coefficient.
Forward Early voltage.
Corner for forward beta current
roll-off.
High current Beta rolloff exponent
B-E leakage saturation current.
B-E leakage emission coefficient.
Ideal maximum reverse beta.
Reverse current emission
coefficient.
Reverse Early voltage.
Corner for reverse beta high
current roll-off.
B-C leakage saturation current
(area is ‘areab’ for vertical devices
and ‘areac’ for lateral).
B-C leakage emission coefficient.
Zero bias base resistance.
Current where base resistance falls
halfway to its min value.
Minimum base resistance at high
currents.
Emitter resistance.
Collector resistance.
B-E zero-bias depletion
capacitance.
B-E built-in potential.
B-E junction exponential factor.
Ideal forward transit time.
Coefficient for bias dependence of
TF.
Voltage describing VBC
dependence of TF.
High-current parameter for effect
on TF.

127

Units

Default
1

Example

Scale factor

A
A

1.0e-16
1.0e-16

1.0e-15
1.0e-15

area
area

-

100
1.0

100
1

V
A

∞
∞

200
0.01

A
-

0.5
0.0
1.5
1
1

0.58
1e-13
2
0.1
1

V
A

∞
∞

200
0.01

area

A

0.0

1e-13

area

Ω
A

2
0
∞

1.5
100
0.1

area
area

Ω

RB

10

area

Ω
Ω
F

0
0
0

1
10
2pF

area
area
area

V
sec
-

0.75
0.33
0
0

0.6
0.33
0.1ns

V

∞

A

0

-

area

area

area

128

CHAPTER 8. BJTS

PTF
CJC

VJC (PC)
MJC
XCJC

TR
CJS

VJS (PS)
MJS (MS)
XTB
EG
XTI
KF
AF
FC
TNOM (TREF)
TLEV
TLEVC
TRE1
TRE2
TRC1
TRC2
TRB1
TRB2

1
Hz.
2πT F
B-C zero-bias depletion
capacitance (area is ‘areab’ for
vertical devices and ‘areac’ for
lateral).
B-C built-in potential.
B-C junction exponential factor.
Fraction of B-C depletion
capacitance connected to internal
base node.
Ideal reverse transit time.
Zero-bias collector-substrate
capacitance (area is ‘areac’ for
vertical devices and ‘areab’ for
lateral).
Substrate junction built-in
potential.
Substrate junction exponential
factor.
Forward and reverse beta
temperature exponent.
Energy gap for temperature effect
on IS.
Temperature exponent for effect on
IS.
Flicker-noise coefficient.
Flicker-noise exponent.
Coefficient for forward-bias
depletion capacitance formula.
Parameter measurement
temperature.
BJT temperature equation selector
BJT capac. temperature equation
selector
1st order temperature coefficient
for RE.
2nd order temperature coefficient
for RE.
1st order temperature coefficient
for RC .
2nd order temperature coefficient
for RC.
1st order temperature coefficient
for RB.
2nd order temperature coefficient
for RB.
Excess phase at freq=

deg

0

F

0

2pF

V
-

0.75
0.33
1

0.5
0.5

sec
F

0
0

10ns
2pF

V

0.75

-

0

-

0

eV

1.11

-

3

-

0
1
0.5

0

◦C

27

50

-

0
0

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

0.5

area

area

8.2. BJT MODELS (NPN/PNP)

TRBM1
TRBM2
TBF1
TBF2
TBR1
TBR2
TIKF1
TIKF2
TIKR1
TIKR2
TIRB1
TIRB2
TNC1
TNC2
TNE1
TNE2
TNF1
TNF2
TNR1
TNR2
TVAF1
TVAF2
TVAR1

1st order temperature coefficient
for RBM
2nd order temperature coefficient
for RBM
1st order temperature coefficient
for BF
2nd order temperature coefficient
for BF
1st order temperature coefficient
for BR
2nd order temperature coefficient
for BR
1st order temperature coefficient
for IKF
2nd order temperature coefficient
for IKF
1st order temperature coefficient
for IKR
2nd order temperature coefficient
for IKR
1st order temperature coefficient
for IRB
2nd order temperature coefficient
for IRB
1st order temperature coefficient
for NC
2nd order temperature coefficient
for NC
1st order temperature coefficient
for NE
2nd order temperature coefficient
for NE
1st order temperature coefficient
for NF
2nd order temperature coefficient
for NF
1st order temperature coefficient
for IKF
2nd order temperature coefficient
for IKF
1st order temperature coefficient
for VAF
2nd order temperature coefficient
for VAF
1st order temperature coefficient
for VAR

129

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

130

CHAPTER 8. BJTS

TVAR2
CTC
CTE
CTS
TVJC
TVJE
TITF1
TITF2
TTF1
TTF2
TTR1
TTR2
TMJE1
TMJE2
TMJC1
TMJC2

2nd order temperature coefficient
for VAR
1st order temperature coefficient
for CJC
1st order temperature coefficient
for CJE
1st order temperature coefficient
for CJS
1st order temperature coefficient
for VJC
1st order temperature coefficient
for VJE
1st order temperature coefficient
for ITF
2nd order temperature coefficient
for ITF
1st order temperature coefficient
for TF
2nd order temperature coefficient
for TF
1st order temperature coefficient
for TR
2nd order temperature coefficient
for TR
1st order temperature coefficient
for MJE
2nd order temperature coefficient
for MJE
1st order temperature coefficient
for MJC
2nd order temperature coefficient
for MJC

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C

0.0

1e-3

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

1/◦C

0.0

1e-3

1/◦C2

0.0

1e-5

Chapter 9
JFETs
9.1

Junction Field-Effect Transistors (JFETs)

General form:
JXXXXXXX nd ng ns mname    
Examples:
J1 7 2 3 JM1 OFF
nd, ng, and ns are the drain, gate, and source nodes, respectively. mname is the model name,
area is the area factor, and off indicates an (optional) initial condition on the device for dc
analysis. If the area factor is omitted, a value of 1.0 is assumed. The (optional) initial condition
specification, using ic=VDS,VGS is intended for use with the uic option on the .TRAN control
line, when a transient analysis is desired starting from other than the quiescent operating point.
See the .ic control line for a better way to set initial conditions. The (optional) temp value is
the temperature where this device is to operate, and overrides the temperature specification on
the .option control line.

9.2
9.2.1

JFET Models (NJF/PJF)
JFET level 1 model with Parker Skellern modification

The level 1 JFET model is derived from the FET model of Shichman and Hodges. The dc
characteristics are defined by the parameters VTO and BETA, which determine the variation
of drain current with gate voltage, LAMBDA, which determines the output conductance, and
IS, the saturation current of the two gate junctions. Two ohmic resistances, RD and RS, are
included.
vgst = vgs −V T O

131

(9.1)

132

CHAPTER 9. JFETS

β p = BETA (1 + LAMBDA vds)
b f ac =

1−B
PB −V T O

(9.2)

(9.3)



if vgst ≤ 0
vds · GMIN,
IDrain = β p vds (vds (b f ac vds − B) vgst (2B + 3b f ac (vgst − vds))) + vds · GMIN, if vgst ≥ vds


β p vgst 2 (B + vgst b f ac) + vds · GMIN,
if vgst < vds
(9.4)
Note that in Spice3f and later, the fitting parameter B has been added by Parker and Skellern.
For details, see [9]. If parameter B is set to 1 equation above simplifies to


if vgst ≤ 0
vds · GMIN,
IDrain = β p vds (2vgst − vds) + vds · GMIN, if vgst ≥ vds


β p vgst 2 + vds · GMIN,
if vgst < vds

(9.5)

Charge storage is modeled by nonlinear depletion layer capacitances for both gate junctions,
which vary as the −1/2 power of junction voltage and are defined by the parameters CGS, CGD,
and PB.
Name
Parameter
Units Default Example Scaling factor
VTO
Threshold voltage VT 0
V
-2.0
-2.0
A/V ”
BETA
Transconductance parameter (β )
1.0e-4
1.0e-3
area
1/V
LAMBDA
Channel-length modulation
0
1.0e-4
parameter (λ )
RD
Drain ohmic resistance
Ω
0
100
area
RS
Source ohmic resistance
Ω
0
100
area
CGS
Zero-bias G-S junction capacitance
F
0
5pF
area
Cgs
CGD
Zero-bias G-D junction
F
0
1pF
area
capacitance Cgd
PB
Gate junction potential
V
1
0.6
IS
Gate saturation current IS
A
1.0e-14 1.0e-14
area
B
Doping tail parameter
1
1.1
KF
Flicker noise coefficient
0
AF
Flicker noise exponent
1
NLEV
Noise equation selector
1
3
GDSNOI
Channel noise coefficient for
1.0
2.0
nlev=3
FC
Coefficient for forward-bias
0.5
depletion capacitance formula
◦C
TNOM
Parameter measurement
27
50
temperature
1/°C
TCV
Threshold voltage temperature
0.0
0.1
coefficient
BEX
Mobility temperature exponent
0.0
1.1

9.2. JFET MODELS (NJF/PJF)

133

Additional to the standard thermal and flicker noise model an alternative thermal channel noise
model is implemented and is selectable by setting NLEV parameter to 3. This follows in a
correct channel thermal noise in the linear region.

Snoise =

(1 + α + α 2 )
2
4kT · BETA ·V gst
GDSNOI
3
1+α

(9.6)

with

α=

9.2.2

(
vds
1 − vgs−V
T O , if vgs −V T O ≥ vds
0,

else

(9.7)

JFET level 2 Parker Skellern model

The level 2 model is an improvement to level 1. Details are available from Macquarie University. Some important items are:

• The description maintains strict continuity in its high-order derivatives, which is essential
for prediction of distortion and intermodulation.
• Frequency dependence of output conductance and transconductance is described as a
function of bias.
• Both drain-gate and source-gate potentials modulate the pinch-off potential, which is consistent with S-parameter and pulsed-bias measurements.
• Self-heating varies with frequency.
• Extreme operating regions - subthreshold, forward gate bias, controlled resistance, and
breakdown regions - are included.
• Parameters provide independent fitting to all operating regions. It is not necessary to
compromise one region in favor of another.
• Strict drain-source symmetry is maintained. The transition during drain-source potential
reversal is smooth and continuous.

The model equations are described in this pdf document and in [19].

134
Name
ID
ACGAM
BETA
CGD
CGS
DELTA
FC
HFETA
HFE1
HFE2
HFGAM
HFG1
HFG2
IBD
IS
LFGAM
LFG1
LFG2
MVST
N
P
Q
RS
RD
TAUD
TAUG
VBD
VBI
VST
VTO
XC
XI
Z
RG
LG
LS
LD
CDSS
AFAC
NFING
TNOM
TEMP

CHAPTER 9. JFETS
Description
Device IDText
Capacitance modulation
Linear-region transconductance scale
Zero-bias gate-source capacitance
Zero-bias gate-drain capacitance
Thermal reduction coefficient
Forward bias capacitance parameter
High-frequency VGS feedback parameter
HFGAM modulation by VGD
HFGAM modulation by VGS
High-frequency VGD feedback parameter
HFGAM modulation by VSG
HFGAM modulation by VDG
Gate-junction breakdown current
Gate-junction saturation current
Low-frequency feedback parameter
LFGAM modulation by VSG
LFGAM modulation by VDG
Subthreshold modulation
Gate-junction ideality factor
Linear-region power-law exponent
Saturated-region power-law exponent
Source ohmic resistance
Drain ohmic resistance
Relaxation time for thermal reduction
Relaxation time for gamma feedback
Gate-junction breakdown potential
Gate-junction potential
Subthreshold potential
Threshold voltage
Capacitance pinch-off reduction factor
Saturation-knee potential factor
Knee transition parameter
Gate ohmic resistance
Gate inductance
Source inductance
Drain inductance
Fixed Drain-source capacitance
Gate-width scale factor
Number of gate fingers scale factor
Nominal Temperature (Not implemented)
Temperature

Unit Type
Text
None
None
Capacitance
Capacitance
None
None
None
None
None
None
None
None
Current
Current
None
None
None
None
None
None
None
Resistance
Resistance
Time
Time
Voltage
Voltage
Voltage
Voltage
None
None
None
Resistance
Inductance
Inductance
Inductance
Capacitance
None
None
Temperature
Temperature

Default
PF1
0
10−4
0F
0F
0W
0.5
0
0V −1
0 V−1
0
0 V−1
0 V−1
0A
10−14A
0
0 V−1
0 V−1
0 V−1
1
2
2
0 Ohm
0 Ohm
0s
0s
1V
1V
0V
-2.0 V
0
1000
0.5
0 Ohm
0H
0H
0H
0F
1
1
300 K
300 K

Chapter 10
MESFETs
10.1

MESFETs

General form:
ZXXXXXXX ND NG NS MNAME   
Examples:
Z1 7 2 3 ZM1 OFF

10.2

MESFET Models (NMF/PMF)

10.2.1

Model by Statz e.a.

The MESFET model level 1 is derived from the GaAs FET model of Statz et al. as described in
[11]. The dc characteristics are defined by the parameters VTO, B, and BETA, which determine
the variation of drain current with gate voltage, ALPHA, which determines saturation voltage,
and LAMBDA, which determines the output conductance. The formula are given by:

Id =





B(Vgs −VT )2
1+b(Vgs −VT )




B(Vgs −VT )2
1+b(Vgs −VT ) (1 + LVds )

1 − 1 − A V3ds

3

(1 + LVds ) for 0 < Vds <
for V >

3
A

(10.1)

3
A

Two ohmic resistances, rd and rs, are included. Charge storage is modeled by total gate charge
as a function of gate-drain and gate-source voltages and is defined by the parameters cgs, cgd,
and pb.

135

136

CHAPTER 10. MESFETS

Name
VTO
BETA
B
ALPHA
LAMBDA
RD
RS
CGS
CGD
PB
KF
AF
FC

Parameter
Pinch-off voltage
Transconductance parameter
Doping tail extending parameter
Saturation voltage parameter
Channel-length modulation parameter
Drain ohmic resistance
Source ohmic resistance
Zero-bias G-S junction capacitance
Zero-bias G-D junction capacitance
Gate junction potential
Flicker noise coefficient
Flicker noise exponent
Coefficient for forward-bias depletion
capacitance formula

Units
V
A/V 2
1/V
1/V
1/V
Ω
Ω
F
F
V
-

Default
-2.0
1.0e-4
0.3
2
0
0
0
0
0
1
0
1
0.5

Example
-2.0
1.0e-3
0.3
2
1.0e-4
100
100
5pF
1pF
0.6

Area
*
*
*
*
*
*
*

Device instance:
z1 2 3 0 mesmod area =1.4
Model:
.model mesmod nmf level =1 rd =46 rs =46 vt0 = -1.3
+ lambda =0.03 alpha =3 beta =1.4e-3

10.2.2

Model by Ytterdal e.a.

level 2 (and levels 3,4) Copyright 1993: T. Ytterdal, K. Lee, M. Shur and T. A. Fjeldly
to be written
M. Shur, T.A. Fjeldly, T. Ytterdal, K. Lee, "Unified GaAs MESFET Model for Circuit Simulation", Int. Journal of High Speed Electronics, vol. 3, no. 2, pp. 201-233, 1992

10.2.3

hfet1

level 5
to be written
no documentation available

10.2.4

hfet2

level6
to be written
no documentation available

Chapter 11
MOSFETs
Ngspice supports all the original mosfet models present in SPICE3f5 and almost all the newer
ones that have been published and made open-source. Both bulk and SOI (Silicon on Insulator) models are available. When compiled with the cider option, ngspice implements the four
terminals numerical model that can be used to simulate a MOSFET (please refer to numerical
modeling documentation for additional information and examples).

11.1

MOSFET devices

General form:
MXXXXXXX nd ng ns nb mname   
+     
+    
Examples:
M1 24 2 0 20 TYPE1
M31 2 17 6 10 MOSN L=5U W=2U
M1 2 9 3 0 MOSP L=10U W=5U AD =100P AS =100P PD =40U PS =40U
Note the suffixes in the example: the suffix ‘u’ specifies microns (1e-6 m) and ‘p’ sq-microns
(1e-12 m2 ).
The instance card for MOS devices starts with the letter ’M’. nd, ng, ns, and nb are the drain,
gate, source, and bulk (substrate) nodes, respectively. mname is the model name and m is the
multiplicity parameter, which simulates ‘m’ paralleled devices. All MOS models support the
‘m’ multiplier parameter. Instance parameters l and w, channel length and width respectively,
are expressed in meters. The areas of drain and source diffusions: ad and as, in squared meters
(m2 ).
If any of l, w, ad, or as are not specified, default values are used. The use of defaults simplifies
input file preparation, as well as the editing required if device geometries are to be changed. pd
and ps are the perimeters of the drain and source junctions, in meters. nrd and nrs designate
the equivalent number of squares of the drain and source diffusions; these values multiply the

137

138

CHAPTER 11. MOSFETS

sheet resistance rsh specified on the .model control line for an accurate representation of the
parasitic series drain and source resistance of each transistor. pd and ps default to 0.0 while nrd
and nrs to 1.0. off indicates an (optional) initial condition on the device for dc analysis. The
(optional) initial condition specification using ic=vds,vgs,vbs is intended for use with the
uic option on the .tran control line, when a transient analysis is desired starting from other
than the quiescent operating point. See the .ic control line for a better and more convenient
way to specify transient initial conditions. The (optional) temp value is the temperature at
which this device is to operate, and overrides the temperature specification on the .option
control line.
The temperature specification is ONLY valid for level 1, 2, 3, and 6 MOSFETs, not for level 4
or 5 (BSIM) devices.
BSIM3.2 and BSIM4 (v4.7 and v4.8) models are also supporting the instance parameter delvto
and mulu0 for local mismatch and NBTI (negative bias temperature instability) modeling:
Name
delvto (delvt0)
mulu0

11.2

Parameter
Threshold voltage shift
Low-field mobility multiplier (U0)

Units
V
-

Default
0.0
1.0

Example
0.07
0.9

MOSFET models (NMOS/PMOS)

MOSFET models are the central part of ngspice, probably because they are the most widely
used devices in the electronics world. Ngspice provides all the MOSFETs implemented in the
original Spice3f and adds several models developed by UC Berkeley’s Device Group and other
independent groups.
Each model is invoked with a .model card. A minimal version is:
.model MOSN NMOS level=8 version=3.3.0
The model name MOSN corresponds to the model name in the instance card (see 11.1). Parameter NMOS selects an n-channel device, PMOS would point to a p-channel transistor. The
level and version parameters select the specific model. Further model parameters are optional and replace ngspice default values. Due to the large number of parameters (more than
100 for modern models), model cards may be stored in extra files and loaded into the netlist by
the .include (2.6) command. Model cards are specific for a an IC manufacturing process and
are typically provided by the IC foundry. Some generic parameter sets, not linked to a specific
process, are made available by the model developers, e.g. UC Berkeley’s Device Group for
BSIM4 and BSIMSOI.
Ngspice provides several MOSFET device models, which differ in the formulation of the I-V
characteristic, and are of varying complexity. Models available are listed in table 11.1. Current
models for IC design are BSIM3 (11.2.9, down to channel length of 0.25 µm), BSIM4 (11.2.10,
below 0.25 µm), BSIMSOI (11.2.12, silicon-on-insulator devices), HiSIM2 and HiSIM_HV
(11.2.14, surface potential models for standard and high voltage/high power MOS devices).

11.2.1

MOS Level 1

This model is also known as the ‘Shichman-Hodges’ model. This is the first model written and
the one often described in the introductory textbooks for electronics. This model is applicable

Level
1
2
3
4
5
6
9
8, 49
8, 49
8, 49
8, 49
10, 58
14, 54
14, 54
14, 54
14, 54
44
45
55
56
57
60
68
73

Name
MOS1
MOS2
MOS3
BSIM1
BSIM2
MOS6
MOS9
BSIM3v0
BSIM3v1
BSIM3v32
BSIM3
B4SOI
BSIM4v5
BSIM4v6
BSIM4v7
BSIM4
EKV
PSP
B3SOIFD
B3SOIDD
B3SOIPD
STAG
HiSIM2
HiSIM_HV

Model
Shichman-Hodges
Grove-Frohman

Table 11.1: MOSFET model summary
SOI3
2.8.0
1.2.4/2.2.0

1.0.2

3.0
3.1
3.2 - 3.2.4
3.3.0
4.3.1
4.0 - 4.5
4.6.5
4.7.0
4.8.0

Version
-

Developer
Berkeley
Berkeley
Berkeley
Berkeley
Berkeley
Berkeley
Alan Gillespie
Berkeley
Berkeley
Berkeley
Berkeley
Berkeley
Berkeley
Berkeley
Berkeley
Berkeley
EPFL
Gildenblatt
Berkeley
Berkeley
Berkeley
Southampton
Hiroshima
Hiroshima

References

High Voltage Version for LDMOS

adms configured
adms configured

Multi version code

extensions by Alan Gillespie
extensions by Serban Popescu
Multi version code
Described in [13]

Notes
This is the classical quadratic model.
Described in [2]
A semi-empirical model (see [1])
Described in [3]
Described in [5]
Described in [2]

11.2. MOSFET MODELS (NMOS/PMOS)
139

140

CHAPTER 11. MOSFETS

only to long channel devices. The use of Meyer’s model for the C-V part makes it non charge
conserving.

11.2.2

MOS Level 2

This model tries to overcome the limitations of the Level 1 model addressing several shortchannel effects, like velocity saturation. The implementation of this model is complicated and
this leads to many convergence problems. C-V calculations can be done with the original Meyer
model (non charge conserving).

11.2.3

MOS Level 3

This is a semi-empirical model derived from the Level 2 model. In the 80s this model has often
been used for digital design and, over the years, has proved to be robust. A discontinuity in the
model with respect to the KAPPA parameter has been detected (see [10]). The supplied fix has
been implemented in Spice3f2 and later. Since this fix may affect parameter fitting, the option
badmos3 may be set to use the old implementation (see the section on simulation variables and
the .options line). Ngspice level 3 implementation takes into account length and width mask
adjustments (xl and xw) and device width narrowing due to diffusion (wd).

11.2.4

MOS Level 6

This model is described in [2]. The model can express the current characteristics of shortchannel MOSFETs at least down to 0.25 µm channel-length, GaAs FET, and resistance inserted
MOSFETs. The model evaluation time is about 1/3 of the evaluation time of the SPICE3 mos
level 3 model. The model also enables analytical treatments of circuits in short-channel region
and makes up for a missing link between a complicated MOSFET current characteristics and
circuit behaviors in the deep submicron region.

11.2.5

Notes on Level 1-6 models

The dc characteristics of the level 1 through level 3 MOSFETs are defined by the device parameters vto, kp, lambda, phi and gamma. These parameters are computed by ngspice if process
parameters (nsub, tox, ...) are given, but users specified values always override. vto is positive (negative) for enhancement mode and negative (positive) for depletion mode N-channel
(P-channel) devices.
Charge storage is modeled by three constant capacitors, cgso, cgdo, and cgbo, which represent
overlap capacitances, by the nonlinear thin-oxide capacitance that is distributed among the gate,
source, drain, and bulk regions, and by the nonlinear depletion-layer capacitances for both
substrate junctions divided into bottom and periphery, which vary as the mj and mjsw power
of junction voltage respectively, and are determined by the parameters cbd, cbs, cj, cjsw, mj,
mjsw and pb.
Charge storage effects are modeled by the piecewise linear voltages-dependent capacitance model proposed by Meyer. The thin-oxide charge-storage effects are treated slightly different for

11.2. MOSFET MODELS (NMOS/PMOS)

141

the level 1 model. These voltage-dependent capacitances are included only if tox is specified
in the input description and they are represented using Meyer’s formulation.
There is some overlap among the parameters describing the junctions, e.g. the reverse current
can be input either as is (in A) or as js (in A/m2 ). Whereas the first is an absolute value the
second is multiplied by ad and as to give the reverse current of the drain and source junctions
respectively.
This methodology has been chosen since there is no sense in relating always junction characteristics with ad and as entered on the device line; the areas can be defaulted. The same idea
applies also to the zero-bias junction capacitances cbd and cbs (in F) on one hand, and cj (in
F/m2 ) on the other.
The parasitic drain and source series resistance can be expressed as either rd and rs (in ohms)
or rsh (in ohms/sq.), the latter being multiplied by the number of squares nrd and nrs input on
the device line.
NGSPICE level 1, 2, 3 and 6 parameters

Name
LEVEL
VTO
KP
GAMMA
PHI
LAMBDA

RD
RS
CBD
CBS
IS
PB
CGSO

CGDO

CGBO

Parameter
Model index
Zero-bias threshold voltage
(VT 0 )
Transconductance
parameter
Bulk threshold parameter
Surface potential (U)
Channel length modulation
(MOS1 and MOS2 only)
(λ )
Drain ohmic resistance
Source ohmic resistance
Zero-bias B-D junction
capacitance
Zero-bias B-S junction
capacitance
Bulk junction saturation
current (IS )
Bulk junction potential
Gate-source overlap
capacitance per meter
channel width
Gate-drain overlap
capacitance per meter
channel width
Gate-bulk overlap
capacitance per meter
channel width

Units
V

Default
1
0.0

Example

A/V 2

2.0e-5

3.1e-5

√
V
V
1/V

0.0
0.6
0.0

0.37
0.65
0.02

Ω
Ω
F

0.0
0.0
0.0

1.0
1.0
20fF

F

0.0

20fF

A

1.0e-14

1.0e-15

V
F/m

0.8
0.0

0.87
4.0e-11

F/m

0.0

4.0e-11

F/m

0.0

2.0e-11

1.0

142

CHAPTER 11. MOSFETS
Name
RSH
CJ

MJ
CJSW

MJSW
JS
TOX
NSUB
NSS
NFS
TPG

XJ
LD
UO
UCRIT
UEXP

UTRA

VMAX
NEFF

KF
AF
FC

DELTA
THETA

Parameter
Drain and source diffusion
sheet resistance
Zero-bias bulk junction
bottom cap. per sq-meter of
junction area
Bulk junction bottom
grading coeff.
Zero-bias bulk junction
sidewall cap. per meter of
junction perimeter

Units
Ω/

Default
0.0

Example
10

F/m2

0.0

2.0e-4

-

0.5

0.5

F/m

0.0

1.0e-9

Bulk junction sidewall
grading coeff.
Bulk junction saturation
current
Oxide thickness
Substrate doping
Surface state density
Fast surface state density
Type of gate material: +1
opp. to substrate, -1 same as
substrate, 0 Al gate
Metallurgical junction depth
Lateral diffusion
Surface mobility
Critical field for mobility
degradation (MOS2 only)
Critical field exponent in
mobility degradation
(MOS2 only)
Transverse field coeff.
(mobility) (deleted for
MOS2)
Maximum drift velocity of
carriers
Total channel-charge (fixed
and mobile) coefficient
(MOS2 only)
Flicker noise coefficient
Flicker noise exponent
Coefficient for forward-bias
depletion capacitance
formula
Width effect on threshold
voltage (MOS2 and MOS3)
Mobility modulation
(MOS3 only)

-

0.50 (level1)
0.33 (level2, 3)

m
cm−3
cm−2
cm−2
-

1.0e-7
0.0
0.0
0.0
1.0

1.0e-7
4.0e15
1.0e10
1.0e10

m
m
cm2/V ·sec
V/cm

0.0
0.0
600
1.0e4

1M
0.8M
700
1.0e4

-

0.0

0.1

-

0.0

0.3

m/s

0.0

5.0e4

-

1.0

5.0

-

0.0
1.0
0.5

1.0e-26
1.2

-

0.0

1.0

1/V

0.0

0.1

11.2. MOSFET MODELS (NMOS/PMOS)
Name
ETA
KAPPA
TNOM

11.2.6

143

Parameter
Static feedback (MOS3
only)
Saturation field factor
(MOS3 only)
Parameter measurement
temperature

Units
-

Default
0.0

Example
1.0

-

0.2

0.5

◦C

27

50

BSIM Models

Ngspice implements many of the BSIM models developed by Berkeley’s BSIM group. BSIM
stands for Berkeley Short-Channel IGFET Model and groups a class of models that is continuously updated. In general, all parameters of BSIM models are obtained from process characterization, in particular level 4 and level 5 (BSIM1 and BSIM2) parameters are can be generated
automatically. J. Pierret [4] describes a means of generating a ‘process’ file, and the program
ngproc2mod provided with ngspice converts this file into a sequence of BSIM1 .model lines
suitable for inclusion in an ngspice input file.
Parameters marked below with an * in the l/w column also have corresponding parameters with
a length and width dependency. For example, vfb is the basic parameter with units of Volts,
and lvfb and wvfb also exist and have units of Volt-meter.
The formula
P = P0 +

PL
Leffective

+

PW
Weffective

(11.1)

is used to evaluate the parameter for the actual device specified with
Leffective = Linput − DL

(11.2)

Weffective = Winput − DW

(11.3)

Note that unlike the other models in ngspice, the BSIM models are designed for use with a
process characterization system that provides all the parameters, thus there are no defaults for
the parameters, and leaving one out is considered an error. For an example set of parameters
and the format of a process file, see the SPICE2 implementation notes[3]. For more information
on BSIM2, see reference [5]. BSIM3 (11.2.9) and BSIM4 (11.2.10) represent state of the art
for submicron and deep submicron IC design.

11.2.7

BSIM1 model (level 4)

BSIM1 model (the first is a long series) is an empirical model. Developers placed less emphasis on device physics and based the model on parametrical polynomial equations to model
the various physical effects. This approach pays in terms of circuit simulation behavior but the
accuracy degrades in the submicron region. A known problem of this model is the negative output conductance and the convergence problems, both related to poor behavior of the polynomial
equations.

144

CHAPTER 11. MOSFETS

Ngspice BSIM (level 4) parameters
Name
VFB
PHI
K1
K2
ETA
MUZ
DL
DW
U0
U1
X2MZ
X2E
X3E
X2U0
X2U1
MUS
X2MS
X3MS
X3U1
TOX
TEMP
VDD
CGDO
CGSO
CGBO
XPART
N0
NB
ND
RSH
JS
PB
MJ

Parameter
Flat-band voltage
Surface inversion potential
Body effect coefficient
Drain/source depletion charge-sharing
coefficient
Zero-bias drain-induced barrier-lowering
coefficient
Zero-bias mobility
Shortening of channel
Narrowing of channel
Zero-bias transverse-field mobility degradation
coefficient
Zero-bias velocity saturation coefficient
Sens. of mobility to substrate bias at v=0
Sens. of drain-induced barrier lowering effect
to substrate bias
Sens. of drain-induced barrier lowering effect
to drain bias at Vds = Vdd
Sens. of transverse field mobility degradation
effect to substrate bias
Sens. of velocity saturation effect to substrate
bias
Mobility at zero substrate bias and at Vds = Vdd
Sens. of mobility to substrate bias at Vds = Vdd
Sens. of mobility to drain bias at Vds = Vdd
Sens. of velocity saturation effect on drain bias
at Vds=Vdd
Gate oxide thickness
Temperature where parameters were measured
Measurement bias range
Gate-drain overlap capacitance per meter
channel width
Gate-source overlap capacitance per meter
channel width
Gate-bulk overlap capacitance per meter
channel length
Gate-oxide capacitance-charge model flag
Zero-bias subthreshold slope coefficient
Sens. of subthreshold slope to substrate bias
Sens. of subthreshold slope to drain bias
Drain and source diffusion sheet resistance
Source drain junction current density
Built in potential of source drain junction
Grading coefficient of source drain junction

Units
V
√V
V
-

l/w
*
*
*
*

-

*

cm2/V ·sec

µm
µm
1/V
µ/V
cm2/V 2 ·sec

*

1/V

*
*
*

1/V

*

1/V 2

*

µm/V 2

*

cm2/V 2 sec
cm2/V 2 sec
cm2/V 2 sec
µm/V 2

*
*
*

µm
◦C
V
F/m
F/m
F/m

Ω/
A/m2
V
-

*
*
*

11.2. MOSFET MODELS (NMOS/PMOS)
Name
PBSW
MJSW
CJ
CJSW
WDF
DELL

145

Parameter
Built in potential of source, drain junction
sidewall
Grading coefficient of source drain junction
sidewall
Source drain junction capacitance per unit area
source drain junction sidewall capacitance per
unit length
Source drain junction default width
Source drain junction length reduction

Units
V

l/w

F/m2
F/m

m
m

xpart = 0 selects a 40/60 drain/source charge partition in saturation, while xpart=1 selects
a 0/100 drain/source charge partition. nd, ng, and ns are the drain, gate, and source nodes,
respectively. mname is the model name, area is the area factor, and off indicates an (optional)
initial condition on the device for dc analysis. If the area factor is omitted, a value of 1.0 is
assumed. The (optional) initial condition specification, using ic=vds,vgs is intended for use
with the uic option on the .tran control line, when a transient analysis is desired starting from
other than the quiescent operating point. See the .ic control line for a better way to set initial
conditions.

11.2.8

BSIM2 model (level 5)

This model contains many improvements over BSIM1 and is suitable for analog simulation.
Nevertheless, even BSIM2 breaks transistor operation into several distinct regions and this leads
to discontinuities in the first derivative in C-V and I-V characteristics that can cause numerical
problems during simulation.

11.2.9

BSIM3 model (levels 8, 49)

BSIM3 solves the numerical problems of previous models with the introduction of smoothing
functions. It adopts a single equation to describe device characteristics in the operating regions.
This approach eliminates the discontinuities in the I-V and C-V characteristics. The original model, BSIM3 evolved through three versions: BSIM3v1, BSIM3v2 and BSIM3v3. Both
BSIM3v1 and BSIM3v2 had suffered from many mathematical problems and were replaced by
BSIM3v3. The latter is the only surviving release and has itself a long revision history.
The following table summarizes the story of this model:
Release
BSIM3v3.0
BSIM3v3.1
BSIM3v3.2

Date
10/30/1995
12/09/1996
06/16/1998

BSIM3v3.3

07/29/2005

Notes

Revisions available: BSIM3v3.2.2,
BSIM3v3.2.3, and BSIM3v3.2.4
Parallel processing with OpenMP is available
for BSIM3v3.2.4.
Parallel processing with OpenMP is available
for this model.

Version flag
3.0
3.1
3.2, 3.2.2,
3.2.3, 3.2.4

3.3.0

146

CHAPTER 11. MOSFETS

BSIM3v2 and 3v3 models has proved for accurate use in 0.18 µm technologies. The model is
publicly available as source code form from University of California, Berkeley.
A detailed description is given in the user’s manual available from here .
We recommend that you use only the most recent BSIM3 model (version 3.3.0), because it
contains corrections to all known bugs. To achieve that, change the version parameter in your
modelcard files to
VERSION = 3.3.0.
If no version number is given in the .model card, this (newest) version is selected as the default.
The older models will not be supported, they are made available for reference only.
BSIM3v3.2.4 supports the extra model parameter lmlt on channel length scaling.

11.2.10

BSIM4 model (levels 14, 54)

This is the newest class of the BSIM family and introduces noise modeling and extrinsic parasitics. BSIM4, as the extension of BSIM3 model, addresses the MOSFET physical effects into
sub-100nm regime. It is a physics-based, accurate, scalable, robust and predictive MOSFET
SPICE model for circuit simulation and CMOS technology development. It is developed by
the BSIM Research Group in the Department of Electrical Engineering and Computer Sciences
(EECS) at the University of California, Berkeley (see BSIM4 home page). BSIM4 has a long
revision history, which is summarized below.
Release
BSIM4.0.0
BSIM4.1.0
BSIM4.2.0
BSIM4.2.1
BSIM4.3.0
BSIM4.4.0
BSIM4.5.0
BSIM4.6.0
...
BSIM4.6.5
BSIM4.7.0
BSIM4.8.0

Date
03/24/2000
10/11/2000
04/06/2001
10/05/2001
05/09/2003
03/04/2004
07/29/2005
12/13/2006

Notes

Version flag

*
*
*
* **

4.2.1
4.3.0
4.4.0
4.5.0

09/09/2009
04/08/2011
30/10/2013

* **
* **
* **

4.6.5
4.7
4.8

*) supported in ngspice, using e.g. the version= flag in the parameter file.
**) Parallel processing using OpenMP support is available for this model.
Details of any revision are to be found in the Berkeley user’s manuals, a pdf download of the
most recent edition is to be found here .
We recommend that you use only the most recent BSIM4 model (version 4.8.0), because it
contains corrections to all known bugs. To achieve that, change the version parameter in your
modelcard files to
VERSION = 4.8.
If no version number is given in the .model card, this (newest) version is selected as the default.
The older models will typically not be supported, they are made available for reference only.

11.2. MOSFET MODELS (NMOS/PMOS)

11.2.11

147

EKV model

Level 44 model (EKV) is not available in the standard distribution since it is not released in
source form by the EKV group. To obtain the code please refer to the (EKV model home page,
EKV group home page). A verilog-A version is available contributed by Ivan Riis Nielsen
11/2006.

11.2.12

BSIMSOI models (levels 10, 58, 55, 56, 57)

BSIMSOI is a SPICE compact model for SOI (Silicon-On-Insulator) circuit design, created by
University of California at Berkeley . This model is formulated on top of the BSIM3 framework. It shares the same basic equations with the bulk model so that the physical nature and
smoothness of BSIM3v3 are retained. Four models are supported in ngspice, those based on
BSIM3 and modeling fully depleted (FD, level 55), partially depleted (PD, level 57) and both
(DD, level 56), as well as the modern BSIMSOI version 4 model (levels 10, 58). Detailed descriptions are beyond the scope of this manual, but see e.g. BSIMSOIv4.4 User Manual for a
very extensive description of the recent model version. OpenMP support is available for levels
10, 58, version 4.4.

11.2.13

SOI3 model (level 60)

see literature citation [18] for a description.

11.2.14

HiSIM models of the University of Hiroshima

There are two model implementations available - see also HiSIM Research Center:
1. HiSIM2 model: Surface-Potential-Based MOSFET Model for Circuit Simulation version
2.8.0 - level 68 (see link to HiSIM2 for source code and manual).
2. HiSIM_HV model: Surface-Potential-Based HV/LD-MOSFET Model for Circuit Simulation version 1.2.4 and 2.2.0 - level 73 (see link to HiSIM_HV for source code and
manual).

148

CHAPTER 11. MOSFETS

Chapter 12
Mixed-Mode and Behavioral Modeling
with XSPICE
Ngspice implements XSPICE extensions for behavioral and mixed-mode (analog and digital)
modeling. In the XSPICE framework this is referred to as code level modeling. Behavioral
modeling may benefit dramatically because XSPICE offers a means to add analog functionality
programmed in C. Many examples (amplifiers, oscillators, filters ...) are presented in the following. Even more flexibility is available because you may define your own models and use them
in addition and in combination with all the already existing ngspice functionality. Mixed mode
simulation is speeded up significantly by simulating the digital part in an event driven manner,
in that state equations use only a few allowed states and are evaluated only during switching,
and not continuously in time and signal as in a pure analog simulator.
This chapter describes the predefined models available in ngspice, stemming from the original
XSPICE simulator. The instructions for writing new code models are given in Chapt. 28.
To make use of the XSPICE extensions, you need to compile them in. Linux, CYGWIN,
MINGW and other users may add the flag --enable-xspice to their ./configure command and then recompile. The prebuilt ngspice for Windows distribution has XSPICE already
enabled. For detailed compiling instructions see Chapt. 32.1.

12.1

Code Model Element & .MODEL Cards

12.1.1

Syntax

Ngspice includes a library of predefined ‘Code Models’ that can be placed within any circuit
description in a manner similar to that used to place standard device models. Code model instance cards always begin with the letter ‘A’, and always make use of a .MODEL card to describe
the code model desired. Section 28 of this document goes into greater detail as to how a code
model similar to the predefined models may be developed, but once any model is created and
linked into the simulator it may be placed using one instance card and one .MODEL card (note
here we conform to the SPICE custom of referring to a single logical line of information as a
‘card’). As an example, the following uses a predefined ‘gain’ code model taking as an input
some value on node 1, multiplies it by a gain of 5.0, and outputs the new value to node 2.
Note that, by convention, input ports are specified first on code models. Output ports follow the
inputs.

149

150

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

Example:
a1 1 2 amp
. model amp gain(gain =5.0)
In this example the numerical values picked up from single-ended (i.e. ground referenced)
input node 1 and output to single-ended output node 2 will be voltages, since in the Interface
Specification File for this code model (i.e., gain), the default port type is specified as a voltage
(more on this later). However, if you didn’t know this, the following modifications to the
instance card could be used to insure it:

Example:
a1 %v(1) %v(2) amp
. model amp gain(gain =5.0)
The specification %v preceding the input and output node numbers of the instance card indicate
to the simulator that the inputs to the model should be single-ended voltage values. Other
possibilities exist, as described later.
Some of the other features of the instance and .MODEL cards are worth noting. Of particular
interest is the portion of the .MODEL card that specifies gain=5.0. This portion of the card
assigns a value to a parameter of the ‘gain’ model. There are other parameters that can be assigned values for this model, and in general code models will have several. In addition to numeric
values, code model parameters can take non-numeric values (such as TRUE and FALSE), and
even vector values. All of these topics will be discussed at length in the following pages. In
general, however, the instance and .MODEL cards that define a code model will follow the abstract form described below. This form illustrates that the number of inputs and outputs and the
number of parameters that can be specified is relatively open-ended and can be interpreted in a
variety of ways (note that angle-brackets ‘<’ and ‘>’ enclose optional inputs):

Example:
AXXXXXXX <%v ,%i ,%vd ,%id ,%g,%gd ,%h ,%hd , or %d>
+ <[> <~><%v ,%i ,%vd ,%id ,%g,%gd ,%h,%hd , or %d>
+ 
+ <~>...< NIN2 .. <]> >
+ <%v ,%i ,%vd ,%id ,%g,%gd ,%h,%hd ,%d or %vnam >
+ <[> <~><%v ,%i ,%vd ,%id ,%g,%gd ,%h,%hd ,
or %d>< NOUT1 or +NOUT1 -NOUT1 >
+ <~>...< NOUT2 .. <]>>
+ MODELNAME
. MODEL MODELNAME MODELTYPE
+ <( PARAMNAME1 = <[> VAL1 > PARAMNAME2 ..>)>

12.1. CODE MODEL ELEMENT & .MODEL CARDS

151

Square brackets ([ ]) are used to enclose vector input nodes. In addition, these brackets are used
to delineate vectors of parameters.
The literal string ‘null’, when included in a node list, is interpreted as no connection at that input
to the model. ‘Null’ is not allowed as the name of a model’s input or output if the model only
has one input or one output. Also, ‘null’ should only be used to indicate a missing connection
for a code model; use on other XSPICE component is not interpreted as a missing connection,
but will be interpreted as an actual node name.
The tilde, ‘~’, when prepended to a digital node name, specifies that the logical value of that
node be inverted prior to being passed to the code model. This allows for simple inversion of
input and output polarities of a digital model in order to handle logically equivalent cases and
others that frequently arise in digital system design. The following example defines a NAND
gate, one input of which is inverted:

a1 [~1 2] 3 nand1
. model nand1 d_nand ( rise_delay =0.1 fall_delay =0.2)
The optional symbols %v, %i, %vd, etc. specify the type of port the simulator is to expect for
the subsequent port or port vector. The meaning of each symbol is given in Table 12.1.
The symbols described in Table 12.1 may be omitted if the default port type for the model is
desired. Note that non-default port types for multi-input or multi-output (vector) ports must be
specified by placing one of the symbols in front of EACH vector port. On the other hand, if all
ports of a vector port are to be declared as having the same non-default type, then a symbol may
be specified immediately prior to the opening bracket of the vector. The following examples
should make this clear:
Example 1: - Specifies two differential voltage connections, one
to nodes 1 & 2, and one to nodes 3 & 4.
%vd [1 2 3 4]
Example 2: - Specifies two single-ended connections to node 1 and
at node 2, and one differential connection to
nodes 3 & 4.
%v [1 2 %vd 3 4]
Example 3: - Identical to the previous example...parenthesis
are added for additional clarity.
%v [1 2 %vd(3 4)]
Example 4: - Specifies that the node numbers are to be treated in the
default fashion for the particular model.
If this model had ‘%v” as a default for this
port, then this notation would represent four single-ended
voltage connections.
[1 2 3 4]

152

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

Port Type Modifiers
Modifier
%v
%i
%g

%h

%d
%vnam

%vd
%id
%gd
%hd

Interpretation
represents a single-ended voltage port - one node name or number is expected
for each port.
represents a single-ended current port - one node name or number is expected
for each port.
represents a single-ended voltage-input, current-output (VCCS) port - one
node name or number is expected for each port. This type of port is automatically an input/output.
represents a single-ended current-input, voltage-output (CCVS) port - one
node name or number is expected for each port. This type of port is automatically an input/output.
represents a digital port - one node name or number is expected for each port.
This type of port may be either an input or an output.
represents the name of a voltage source, the current through which is taken as
an input. This notation is provided primarily in order to allow models defined
using SPICE2G6 syntax to operate properly in XSPICE.
represents a differential voltage port - two node names or numbers are expected for each port.
represents a differential current port - two node names or numbers are expected for each port.
represents a differential VCCS port - two node names or numbers are expected
for each port.
represents a differential CCVS port - two node names or numbers are expected
for each port.
Table 12.1: Port Type Modifiers

12.1. CODE MODEL ELEMENT & .MODEL CARDS

153

The parameter names listed on the .MODEL card must be identical to those named in the code
model itself. The parameters for each predefined code model are described in detail in Sections
12.2 (analog), 12.3 (Hybrid, A/D) and 12.4 (digital) . The steps required in order to specify
parameters for user-defined models are described in Chapter 28.

12.1.2

Examples

The following is a list of instance card and associated .MODEL card examples showing use of
predefined models within an XSPICE deck:

a1 1 2 amp
.model amp gain(in_offset=0.1 gain=5.0 out_offset=-0.01)
a2 %i[1 2] 3 sum1
.model sum1 summer(in_offset=[0.1 -0.2] in_gain=[2.0 1.0]
+ out_gain=5.0 out_offset=-0.01)
a21 %i[1 %vd(2 5) 7 10] 3 sum2
.model sum2 summer(out_gain=10.0)
a5 1 2 limit5
.model limit5 limit(in_offset=0.1 gain=2.5
+ out_lower.limit=-5.0 out_upper_limit=5.0 limit_domain=0.10
+ fraction=FALSE)
a7 2 %id(4 7) xfer.cntl1
.model xfer_cntl1 pwl(x_array=[-2.0 -1.0 2.0 4.0 5.0]
+ y_array=[-0.2 -0.2 0.1 2.0 10.0]
+ input_domain=0.05 fraction=TRUE)
a8 3 %gd(6 7) switch3
.model switch3 aswitch(cntl_off=0.0 cntl_on=5.0 r_off=1e6
+ r_on=10.0 log=TRUE)

12.1.3

Search path for file input

Several code models (filesource 12.2.8, d_source 12.4.21, d_state 12.4.18) call additional
files for supply of input data. A call to file="path/filename" (or input_file=, state_file=)
in the .model card will start a search sequence for finding the file. path may be an absolute
path. If path is omitted or is a relative path, filename is looked for according to the following
search list:
Infile_Path/ (Infile_Path is the path of the input file *.sp containing the
netlist)
NGSPICE_INPUT_DIR/ (where an additional path is set by the environmental variable)
 (where the search is relativ to the current directory (OS dependant))

154

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

12.2

Analog Models

The following analog models are supplied with XSPICE. The descriptions included consist
of the model Interface Specification File and a description of the model’s operation. This is
followed by an example of a simulator-deck placement of the model, including the .MODEL
card and the specification of all available parameters.

12.2.1

Gain

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:

cm_gain
gain
"A simple gain block"

PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector.Bounds:
Null.Allowed:

in
"input"
in
v
[v,vd,i,id,vnam]
no
no

PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

in_offset
"input offset"
real
0.0
no
yes

out
"output"
out
v
[v,vd,i,id]
no
no

gain
"gain"
real
1.0
no
yes

out_offset
"output offset"
real
0.0
no
yes

Description: This function is a simple gain block with optional offsets on the input and the
output. The input offset is added to the input, the sum is then multiplied by the gain, and
the result is produced by adding the output offset. This model will operate in DC, AC,
and Transient analysis modes.

Example:

a1 1 2 amp
. model amp gain( in_offset =0.1 gain =5.0
+ out_offset = -0.01)

12.2. ANALOG MODELS

12.2.2

155

Summer

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:

cm_summer
summer
"A summer block"

PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:

in
"input vector"
in
v
[v,vd,i,id,vnam]
yes
no

PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

in_offset
"input offset vector"
real
0.0
yes
in
yes

in_gain
"input gain vector"
real
1.0
yes
in
yes

PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

out_gain
"output gain"
real
1.0
no
yes

out_offset
"output offset"
real
0.0
no
yes

out
"output"
out
v
[v,vd,i,id]
no
no

Description: This function is a summer block with 2-to-N input ports. Individual gains and
offsets can be applied to each input and to the output. Each input is added to its respective
offset and then multiplied by its gain. The results are then summed, multiplied by the
output gain and added to the output offset. This model will operate in DC, AC, and
Transient analysis modes.
Example usage:
a2 [1 2] 3 sum1
.model sum1 summer ( in_offset =[0.1 -0.2] in_gain =[2.0 1.0]
+ out_gain =5.0 out_offset = -0.01)

156

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

12.2.3

Multiplier

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

cm_mult
mult
"multiplier block"
in
"input vector"
in
v
[v,vd,i,id,vnam]
yes
[2 -]
no

out
"output"
out
v
[v,vd,i,id]
no
no

in_offset
"input offset vector"
real
0.0
yes
in
yes

in_gain
"input gain vector"
real
1.0
yes
in
yes

out_gain
"output gain"
real
1.0
no
yes

out_offset
"output offset"
real
0.0
no
yes

Description: This function is a multiplier block with 2-to-N input ports. Individual gains and
offsets can be applied to each input and to the output. Each input is added to its respective
offset and then multiplied by its gain. The results are multiplied along with the output
gain and are added to the output offset. This model will operate in DC, AC, and Transient
analysis modes. However, in ac analysis it is important to remember that results are
invalid unless only one input of the multiplier is connected to a node that i connected to
an AC signal (this is exemplified by the use of a multiplier to perform a potentiometer
function: one input is DC, the other carries the AC signal).
Example SPICE Usage:
a3 [1 2 3] 4 sigmult
. model sigmult mult( in_offset =[0.1 0.1 -0.1]
+ in_gain =[10.0 10.0 10.0] out_gain =5.0 out_offset =0.05)

12.2. ANALOG MODELS

12.2.4

157

Divider

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:

cm_divide
divide
"divider block"
num
"numerator"
in
v
[v,vd,i,id,vnam]
no
no

den
"denominator"
in
v
[v,vd,i,id,vnam]
no
no

num_offset
"numerator offset"
real
0.0
no
yes

num_gain
"numerator gain"
real
1.0
no
yes

den_offset
"denominator offset"
real
0.0
no
yes

den_gain
"denominator gain"
real
1.0
no
yes

den_lower_limit
"denominator lower limit"
real
1.0e-10
no
yes
den_domain
"denominator smoothing domain"
real
1.0e-10
no

out
"output"
out
v
[v,vd,i,id]
no
no

158

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

yes
fraction
"smoothing fraction/absolute value switch"
boolean
false
no
yes
out_gain
"output gain"
real
1.0
no
yes

out_offset
"output offset"
real
0.0
no
yes

Description: This function is a two-quadrant divider. It takes two inputs; num (numerator) and
den (denominator). Divide offsets its inputs, multiplies them by their respective gains,
divides the results, multiplies the quotient by the output gain, and offsets the result. The
denominator is limited to a value above zero via a user specified lower limit. This limit
is approached through a quadratic smoothing function, the domain of which may be specified as a fraction of the lower limit value (default), or as an absolute value. This model
will operate in DC, AC and Transient analysis modes. However, in ac analysis it is important to remember that results are invalid unless only one input of the divider is connected
to a node that is connected to an ac signal (this is exemplified by the use of the divider to
perform a potentiometer function: one input is dc, the other carries the ac signal).
Example SPICE Usage:
a4 1 2 4 divider
.model divider divide(num_offset=0.1 num_gain=2.5 den_offset=-0.1
+ den_gain=5.0 den_lower.limit=1e-5 den_domain=1e-6
+ fraction=FALSE out_gain=1.0 out_offset=0.0)

12.2.5

Limiter

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_limit
limit
"limit block"
in
"input"
in

out
"output"
out

12.2. ANALOG MODELS
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

159
v
[v,vd,i,id]
no
no

v
[v,vd,i,id]
no
no

in_offset
"input offset"
real
0.0
no
yes

gain
"gain"
real
1.0
no
yes

out_lower_limit
"output lower limit"
real
0.0
no
yes

out_upper_limit
"output upper limit"
real
1.0
no
yes

limit_range
"upper & lower smoothing range"
real
1.0e-6
no
yes
fraction
"smoothing fraction/absolute value switch"
boolean
FALSE
no
yes

Description: The Limiter is a single input, single output function similar to the Gain Block.
However, the output of the Limiter function is restricted to the range specified by the
output lower and upper limits. This model will operate in DC, AC and Transient analysis
modes. Note that the limit range is the value below the upper limit and above the lower
limit at which smoothing of the output begins. For this model, then, the limit range
represents the delta with respect to the output level at which smoothing occurs. Thus, for

160

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
an input gain of 2.0 and output limits of 1.0 and -1.0 volts, the output will begin to smooth
out at ±0.9 volts, which occurs when the input value is at ±0.4.
Example SPICE Usage:
a5 1 2 limit5
.model limit5 limit(in_offset=0.1 gain=2.5 out_lower_limit=-5.0
+ out_upper_limit=5.0 limit_range=0.10 fraction=FALSE)

12.2.6

Controlled Limiter

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:

cm_climit
climit
"controlled limiter block"
in
"input"
in
v
[v,vd,i,id,vnam]
no
no

cntl_upper
"upper lim. control input"
in
v
[v,vd,i,id,vnam]
no
no

cntl_lower
"lower limit control input"
in
v
[v,vd,i,id,vnam]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

in_offset
"input offset"
real
0.0
no
yes

gain
"gain"
real
1.0
no
yes

upper_delta
"output upper delta"
real
0.0
no

lower_delta
"output lower delta"
real
0.0
no

12.2. ANALOG MODELS
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

161
yes

yes

limit_range
"upper & lower sm. range"
real
1.0e-6
no
yes

fraction
"smoothing %/abs switch"
boolean
FALSE
no
yes

Description: The Controlled Limiter is a single input, single output function similar to the Gain
Block. However, the output of the Limiter function is restricted to the range specified by
the output lower and upper limits. This model will operate in DC, AC, and Transient
analysis modes. Note that the limit range is the value below the cntl_upper limit and
above the cntl_lower limit at which smoothing of the output begins (minimum positive
value of voltage must exist between the cntl_upper input and the cntl_lower input at
all times). For this model, then, the limit range represents the delta with respect to the
output level at which smoothing occurs. Thus, for an input gain of 2.0 and output limits
of 1.0 and -1.0 volts, the output will begin to smooth out at ±0.9 volts, which occurs
when the input value is at ±0.4. Note also that the Controlled Limiter code tests the
input values of cntl_upper and cntl_lower to make sure that they are spaced far enough
apart to guarantee the existence of a linear range between them. The range is calculated
as the difference between (cntl_upper − upper_delta − limit_range) and (cntl_lower +
lower_delta + limit_range) and must be greater than or equal to zero. Note that when
the limit range is specified as a fractional value, the limit range used in the above is taken
as the calculated fraction of the difference between cntl_upper and cntl_lower. Still, the
potential exists for too great a limit range value to be specified for proper operation, in
which case the model will return an error message.
Example SPICE Usage:
a6 3 6 8 4 varlimit
.
.
.model varlimit climit(in_offset=0.1 gain=2.5 upper_delta=0.0
+ lower_delta=0.0 limit_range=0.10 fraction=FALSE)

12.2.7

PWL Controlled Source

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:

cm_pwl
pwl
"piecewise linear controlled source"
in
"input"
in

out
"output"
out

162

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
STATIC_VAR_TABLE:
Static_Var_Name:
Data_Type:
Description:

v
[v,vd,i,id,vnam]
no
no

v
[v,vd,i,id]
no
no

x_array
"x-element array"
real
yes
[2 -]
no

y_array
"y-element array"
real
yes
[2 -]
no

input_domain
"input sm. domain"
real
0.01
[1e-12 0.5]
no
yes

fraction
"smoothing %/abs switch"
boolean
TRUE
no
yes

last_x_value
pointer
"iteration holding variable for limiting"

Description: The Piece-Wise Linear Controlled Source is a single input, single output function
similar to the Gain Block. However, the output of the PWL Source is not necessarily linear for all values of input. Instead, it follows an I/O relationship specified by you via the
x_array and y_array coordinates. This is detailed below.
The x_array and y_array values represent vectors of coordinate points on the x and
y axes, respectively. The x_array values are progressively increasing input coordinate
points, and the associated y_array values represent the outputs at those points. There
may be as few as two (x_array[n], y_array[n]) pairs specified, or as many as memory
and simulation speed allow. This permits you to very finely approximate a non-linear
function by capturing multiple input-output coordinate points.
Two aspects of the PWL Controlled Source warrant special attention. These are the handling of endpoints and the smoothing of the described transfer function near coordinate
points.
In order to fully specify outputs for values of in outside of the bounds of the PWL
function (i.e., less than x_array[0] or greater than x_array[n], where n is the largest
user-specified coordinate index), the PWL Controlled Source model extends the slope
found between the lowest two coordinate pairs and the highest two coordinate pairs.
This has the effect of making the transfer function completely linear for in less than
x_array[0] and in greater than x_array[n]. It also has the potentially subtle effect of
unrealistically causing an output to reach a very large or small value for large inputs. You

12.2. ANALOG MODELS

163

should thus keep in mind that the PWL Source does not inherently provide a limiting
capability.
In order to diminish the potential for non-convergence of simulations when using the
PWL block, a form of smoothing around the x_array, y_array coordinate points is necessary. This is due to the iterative nature of the simulator and its reliance on smooth first
derivatives of transfer functions in order to arrive at a matrix solution. Consequently, the
input_domain and fraction parameters are included to allow you some control over
the amount and nature of the smoothing performed.
Fraction is a switch that is either TRUE or FALSE. When TRUE (the default setting),
the simulator assumes that the specified input domain value is to be interpreted as a fractional figure. Otherwise, it is interpreted as an absolute value. Thus, if fraction=TRUE
and input_domain=0.10, The simulator assumes that the smoothing radius about each
coordinate point is to be set equal to 10% of the length of either the x_array segment
above each coordinate point, or the x_array segment below each coordinate point. The
specific segment length chosen will be the smallest of these two for each coordinate point.
On the other hand, if fraction=FALSE and input=0.10, then the simulator will begin
smoothing the transfer function at 0.10 volts (or amperes) below each x_array coordinate and will continue the smoothing process for another 0.10 volts (or amperes) above
each x_array coordinate point. Since the overlap of smoothing domains is not allowed,
checking is done by the model to ensure that the specified input domain value is not excessive.
One subtle consequence of the use of the fraction=TRUE feature of the PWL Controlled Source is that, in certain cases, you may inadvertently create extreme smoothing
of functions by choosing inappropriate coordinate value points. This can be demonstrated by considering a function described by three coordinate pairs, such as (-1,-1), (1,1),
and (2,1). In this case, with a 10% input_domain value specified (fraction=TRUE,
input_domain=0.10), you would expect to see rounding occur between in=0.9 and
in=1.1, and nowhere else. On the other hand, if you were to specify the same function
using the coordinate pairs (-100,-100), (1,1) and (201,1), you would find that rounding
occurs between in=-19 and in=21. Clearly in the latter case the smoothing might cause
an excessive divergence from the intended linearity above and below in=1.
Example SPICE Usage:
a7 2 4 xfer_cntl1
.
.
.model xfer_cntl1 pwl(x_array=[-2.0 -1.0 2.0 4.0 5.0]
+
y_array=[-0.2 -0.2 0.1 2.0 10.0]
+
input_domain=0.05 fraction=TRUE)

12.2.8

Filesource

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:

cm_filesource
filesource
"File Source"
out

164

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

"output"
out
v
[v,vd,i,id]
yes
[1 -]
no
timeoffset
"time offset"
real
0.0
no
yes

timescale
"timescale"
real
1.0
no
yes

timerelative
"relative time"
boolean
FALSE
no
yes

amplstep
"step amplitude"
boolean
FALSE
no
yes

amploffset
"ampl offset"
real
yes
[1 -]
yes

amplscale
"amplscale"
real
yes
[1 -]
yes

file
"file name"
string
"filesource.txt"
no
yes

Description: The File Source is similar to the Piece-Wise Linear Source, except that the waveform data is read from a file instead of being taken from parameter vectors. The file
format is line oriented ASCII. ‘#’ and ‘;’ are comment characters; all characters from
a comment character until the end of the line are ignored. Each line consists of two or

12.2. ANALOG MODELS

165

more real values. The first value is the time; subsequent values correspond to the outputs.
Values are separated by spaces. Time values are absolute and must be monotonically increasing, unless timerelative is set to TRUE, in which case the values specify the interval
between two samples and must be positive. Waveforms may be scaled and shifted in the
time dimension by setting timescale and timeoffset.
Amplitudes can also be scaled and shifted using amplscale and amploffset. Amplitudes
are normally interpolated between two samples, unless amplstep is set to TRUE.
Note: The file named by the parameter filename in file="filename" is sought after according to a search list described in12.1.3.
Example SPICE Usage:
a8 %vd([1 0 2 0]) filesrc
.
.
.model filesrc filesource (file="sine.m" amploffset=[0 0] amplscale=[1 1]
+
timeoffset=0 timescale=1
+
timerelative=false amplstep=false)

Example input file:
# name: sine.m
# two output ports
# column 1: time
# columns 2, 3: values
0 0 1
3.90625e-09 0.02454122852291229 0.9996988186962042
7.8125e-09 0.04906767432741801 0.9987954562051724
1.171875e-08 0.07356456359966743 0.9972904566786902
...

12.2.9

multi_input_pwl block

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:

cm_multi_input_pwl
multi_input_pwl
"multi_input_pwl block"
in
"input array"
in
vd
[vd,id]
yes
[2 -]
no

out
"output"
out
vd
[vd,id]
no
no

166

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

x
"x array"
real
0.0
yes
[2 -]
no

y
"y array"
real
0.0
yes
[2 -]
no

model
"model type"
string
"and"
no
yes

Description: Multi-input gate voltage controlled voltage source that supports and or or gating.
The x’s and y’s represent the piecewise linear variation of output (y) as a function of input
(x). The type of gate is selectable by the parameter model. In case the model is and, the
smallest input determines the output value (i.e. the and function). In case the model is or,
the largest input determines the output value (i.e. the or function). The inverse of these
functions (i.e. nand and nor) is constructed by complementing the y array.
Example SPICE Usage:
a82 [1 0 2 0 3 0] 7 0 pwlm
.
.
.model pwlm multi_input_pwl((x=[-2.0 -1.0 2.0 4.0 5.0]
+
y=[-0.2 -0.2 0.1 2.0 10.0]
+
model="and")

12.2.10

Analog Switch

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:

cm_aswitch
aswitch
"analog switch"
cntl_in
"input"
in
v
[v,vd,i,id]
no
no

out
"resistive output"
out
gd
[gd]
no
no

12.2. ANALOG MODELS
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

167

cntl_off
cntl_on
"control ‘off’ value" "control ‘on’ value"
real
real
0.0
1.0
no
no
yes
yes
r_off
"off resistance"
real
1.0e12
no
yes

log
"log/linear switch"
boolean
TRUE
no
yes

r_on
"on resistance"
real
1.0
no
yes

Description: The Analog Switch is a resistor that varies either logarithmically or linearly between specified values of a controlling input voltage or current. Note that the input is
not internally limited. Therefore, if the controlling signal exceeds the specified OFF state
or ON state value, the resistance may become excessively large or excessively small (in
the case of logarithmic dependence), or may become negative (in the case of linear dependence). For the experienced user, these excursions may prove valuable for modeling
certain devices, but in most cases you are advised to add limiting of the controlling input
if the possibility of excessive control value variation exists.
Example SPICE Usage:
a8 3 %gd(6 7) switch3
.
.
.model switch3 aswitch(cntl_off=0.0 cntl_on=5.0 r_off=1e6
+
r_on=10.0 log=TRUE)

12.2.11

Zener Diode

NAME_TABLE:
C_Function_Name:

cm_zener

168

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
STATIC_VAR_TABLE:
Static_Var_Name:
Data_Type:
Description:

zener
"zener diode"
z
"zener"
inout
gd
[gd]
no
no
v_breakdown
"breakdown voltage"
real
[1.0e-6 1.0e6]
no
no

i_breakdown
"breakdown current"
real
2.0e-2
[1.0e-9 -]
no
yes

i_sat
"saturation current"
real
1.0e-12
[1.0e-15 -]
no
yes

n_forward
"forward emission coefficient"
real
1.0
[0.1 10]
no
yes

limit_switch
"switch for on-board limiting (convergence aid)"
boolean
FALSE
no
yes
previous_voltage
pointer
"iteration holding variable for limiting"

Description: The Zener Diode models the DC characteristics of most zeners. This model
differs from the Diode/Rectifier by providing a user-defined dynamic resistance in the
reverse breakdown region. The forward characteristic is defined by only a single point,
since most data sheets for zener diodes do not give detailed characteristics in the forward
region.

12.2. ANALOG MODELS

169

The first three parameters define the DC characteristics of the zener in the breakdown
region and are usually explicitly given on the data sheet.
The saturation current refers to the relatively constant reverse current that is produced
when the voltage across the zener is negative, but breakdown has not been reached. The
reverse leakage current determines the slight increase in reverse current as the voltage
across the zener becomes more negative. It is modeled as a resistance parallel to the
zener with value v breakdown / i rev.
Note that the limit switch parameter engages an internal limiting function for the zener.
This can, in some cases, prevent the simulator from converging to an unrealistic solution
if the voltage across or current into the device is excessive. If use of this feature fails to
yield acceptable results, the convlimit option should be tried (add the following statement
to the SPICE input deck: .options convlimit)
Example SPICE Usage:
a9 3 4 vref10
.
.
.model vref10 zener(v_breakdown=10.0 i_breakdown=0.02
+
r_breakdown=1.0 i_rev=1e-6 i_sat=1e-12)

12.2.12

Current Limiter

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:

cm_ilimit
ilimit
"current limiter block"
in
"input"
in
v
[v,vd]
no
no

pos_pwr
"positive power supply"
inout
g
[g,gd]
no
yes

neg_pwr
"negative power supply"
inout
g
[g,gd]
no
yes

out
"output"
inout
g
[g,gd]
no
no

in_offset
"input offset"
real

gain
"gain"
real

170

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:

0.0
no
yes

1.0
no
yes

r_out_source
"sourcing resistance"
real
1.0
[1.0e-9 1.0e9]
no
yes

r_out_sink
"sinking resistance"
real
1.0
[1.0e-9 1.0e9]
no
yes

i_limit_source
"current sourcing limit"
real
[1.0e-12 -]
no
yes
i_limit_sink
"current sinking limit"
real
[1.0e-12 -]
no
yes
v_pwr_range
"upper & lower power
supply smoothing range"
real
1.0e-6
[1.0e-15 -]
no
yes

i_source_range
"sourcing current
smoothing range"
real
1.0e-9
[1.0e-15 -]
no
yes

i_sink_range
"sinking current smoothing range"
real
1.0e-9
[1.0e-15 -]

12.2. ANALOG MODELS
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

171
no
yes
r_out_domain
"internal/external voltage delta smoothing range"
real
1.0e-9
[1.0e-15 -]
no
yes

Description: The Current Limiter models the behavior of an operational amplifier or comparator device at a high level of abstraction. All of its pins act as inputs; three of the four
also act as outputs. The model takes as input a voltage value from the in connector. It
then applies an offset and a gain, and derives from it an equivalent internal voltage (veq),
which it limits to fall between pos_pwr and neg_pwr. If veq is greater than the output
voltage seen on the out connector, a sourcing current will flow from the output pin. Conversely, if the voltage is less than vout, a sinking current will flow into the output pin.
Depending on the polarity of the current flow, either a sourcing or a sinking resistance
value (r_out_source, r_out_sink) is applied to govern the vout/i_out relationship.
The chosen resistance will continue to control the output current until it reaches a maximum value specified by either i_limit_source or i_limit_sink. The latter mimics
the current limiting behavior of many operational amplifier output stages.
During all operation, the output current is reflected either in the pos_pwr connector current or the neg_pwr current, depending on the polarity of i_out. Thus, realistic power
consumption as seen in the supply rails is included in the model.
The user-specified smoothing parameters relate to model operation as follows: v_pwr_range
controls the voltage below vpos_pwr and above vneg_pwr inputs beyond which veq =
gain(vin+vo f f set ) is smoothed; i_source_range specifies the current below i_limit_source
at which smoothing begins, as well as specifying the current increment above i_out=0.0
at which i_pos_pwr begins to transition to zero; i_sink_range serves the same purpose with respect to i_limit_sink and i_neg_pwr that i_source_range serves for
i_limit_source and i_pos_pwr; r_out_domain specifies the incremental value above
and below (veq-vout)=0.0 at which r_out will be set to r_out_source and r_out_sink,
respectively. For values of (veq-vout) less than r_out_domain and greater than -r_out_domain,
r_out is interpolated smoothly between r_out_source and r_out_sink.
Example SPICE Usage:
a10 3 10 20 4 amp3
.
.
.model amp3 ilimit(in_offset=0.0 gain=16.0 r_out_source=1.0
+
r_out_sink=1.0 i_limit_source=1e-3
+
i_limit_sink=10e-3 v_pwr_range=0.2
+
i_source_range=1e-6 i_sink_range=1e-6
+
r_out_domain=1e-6)

172

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

12.2.13

Hysteresis Block

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:

cm_hyst
hyst
"hysteresis block"
in
"input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

in_low
"input low value"
real
0.0
no
yes

in_high
"input high value"
real
1.0
no
yes

hyst
"hysteresis"
real
0.1
[0.0 -]
no
yes

out_lower_limit
"output lower limit"
real
0.0
no
yes

out_upper_limit
"output upper limit"
real
1.0
no
yes

input_domain
"input smoothing domain"
real
0.01
no
yes

fraction
"smoothing fraction/absolute value switch"
boolean
TRUE
no

12.2. ANALOG MODELS
Vector_Bounds:
Null_Allowed:

173
yes

Description: The Hysteresis block is a simple buffer stage that provides hysteresis of the output
with respect to the input. The in low and in high parameter values specify the center
voltage or current inputs about which the hysteresis effect operates. The output values
are limited to out lower limit and out upper limit. The value of hyst is added to the in
low and in high points in order to specify the points at which the slope of the hysteresis
function would normally change abruptly as the input transitions from a low to a high
value. Likewise, the value of hyst is subtracted from the in high and in low values in
order to specify the points at which the slope of the hysteresis function would normally
change abruptly as the input transitions from a high to a low value. In fact, the slope of the
hysteresis function is never allowed to change abruptly but is smoothly varied whenever
the input domain smoothing parameter is set greater than zero.
Example SPICE Usage:
a11 1 2 schmitt1
.
.
.model schmitt1 hyst(in_low=0.7 in_high=2.4 hyst=0.5
+
out_lower_limit=0.5 out_upper_limit=3.0
+
input_domain=0.01 fraction=TRUE)

12.2.14

Differentiator

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:

cm_d_dt
d_dt
"time-derivative block"
in
"input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

gain
"gain"
real
1.0
no
yes

out_offset
"output offset"
real
0.0
no
yes

174

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

out_lower_limit
"output lower limit"
real
no
yes

out_upper_limit
"output upper limit"
real
no
yes

limit_range
"upper & lower limit smoothing range"
real
1.0e-6
no
yes

Description: The Differentiator block is a simple derivative stage that approximates the time
derivative of an input signal by calculating the incremental slope of that signal since the
previous time point. The block also includes gain and output offset parameters to allow
for tailoring of the required signal, and output upper and lower limits to prevent convergence errors resulting from excessively large output values. The incremental value of
output below the output upper limit and above the output lower limit at which smoothing
begins is specified via the limit range parameter. In AC analysis, the value returned is
equal to the radian frequency of analysis multiplied by the gain.
Note that since truncation error checking is not included in the d_dt block, it is not recommended that the model be used to provide an integration function through the use
of a feedback loop. Such an arrangement could produce erroneous results. Instead, you
should make use of the "integrate" model, which does include truncation error checking
for enhanced accuracy.
Example SPICE Usage:
a12 7 12 slope_gen
.
.
.model slope_gen d_dt(out_offset=0.0 gain=1.0
+
out_lower_limit=1e-12 out_upper_limit=1e12
+
limit_range=1e-9)

12.2.15

Integrator

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:

cm_int
int
"time-integration block"
in

out

12.2. ANALOG MODELS
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

175
"input"
in
v
[v,vd,i,id]
no
no

"output"
out
v
[v,vd,i,id]
no
no

in_offset
"input offset"
real
0.0
no
yes

gain
"gain"
real
1.0
no
yes

out_lower_limit
"output lower limit"
real
no
yes

out_upper_limit
"output upper limit"
real
no
yes

limit_range
"upper & lower limit smoothing range"
real
1.0e-6
no
yes
out_ic
"output initial condition"
real
0.0
no
yes

Description: The Integrator block is a simple integration stage that approximates the integral
with respect to time of an input signal. The block also includes gain and input offset
parameters to allow for tailoring of the required signal, and output upper and lower limits
to prevent convergence errors resulting from excessively large output values. Note that

176

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
these limits specify integrator behavior similar to that found in an operational amplifierbased integration stage, in that once a limit is reached, additional storage does not occur.
Thus, the input of a negative value to an integrator that is currently driving at the out
upper limit level will immediately cause a drop in the output, regardless of how long
the integrator was previously summing positive inputs. The incremental value of output
below the output upper limit and above the output lower limit at which smoothing begins
is specified via the limit range parameter. In AC analysis, the value returned is equal to
the gain divided by the radian frequency of analysis.
Note that truncation error checking is included in the int block. This should provide
for a more accurate simulation of the time integration function, since the model will
inherently request smaller time increments between simulation points if truncation errors
would otherwise be excessive.
Example SPICE Usage:
a13 7 12 time_count
.
.
.model time_count int(in_offset=0.0 gain=1.0
+
out_lower_limit=-1e12 out_upper_limit=1e12
+
limit_range=1e-9 out_ic=0.0)

12.2.16

S-Domain Transfer Function

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:

cm_s_xfer
s_xfer
"s-domain transfer function"
in
"input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

in_offset
"input offset"
real
0.0
no
yes

gain
"gain"
real
1.0
no
yes

num_coeff
"numerator polynomial coefficients"

12.2. ANALOG MODELS
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

177
real
yes
[1 -]
no
den_coeff
"denominator polynomial coefficients"
real
yes
[1 -]
no
int_ic
"integrator stage initial conditions"
real
0.0
yes
den_coeff
yes
denormalized_freq
"denorm. corner freq.(radians) for 1 rad/s coeffs"
real
1.0
no
yes

Description: The s-domain transfer function is a single input, single output transfer function
in the Laplace transform variable ‘s’ that allows for flexible modulation of the frequency
domain characteristics of a signal. Ac and transient simulations are supported. The code
model may be configured to produce an arbitrary s-domain transfer function with the
following restrictions:
1. The degree of the numerator polynomial cannot exceed that
of the denominator polynomial in the variable "s".
2. The coefficients for a polynomial must be stated
explicitly. That is, if a coefficient is zero, it must be
included as an input to the num coeff or den coeff vector.
The order of the coefficient parameters is from that associated with the highest-powered term
decreasing to that of the lowest. Thus, for the coefficient parameters specified below, the equation in ‘s’ is shown:

178

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
.model filter s_xfer(gain=0.139713
+ num_coeff=[1.0 0.0 0.7464102]
+ den_coeff=[1.0 0.998942 0.001170077]
+ int_ic=[0 0])
...specifies a transfer function of the form...
2

s +0.7464102
N(s) = 0.139713 · s2 +0.998942s+0.00117077

The s-domain transfer function includes gain and in_offset (input offset) parameters to allow
for tailoring of the required signal. There are no limits on the internal signal values or on
the output value of the s-domain transfer function, so you are cautioned to specify gain and
coefficient values that will not cause the model to produce excessively large values. In AC
analysis, the value returned is equal to the real and imaginary components of the total s-domain
transfer function at each frequency of interest.
The denormalized_freq term allows you to specify coefficients for a normalized filter (i.e. one
in which the frequency of interest is 1 rad/s). Once these coefficients are included, specifying
the denormalized frequency value ‘shifts’ the corner frequency to the actual one of interest. As
an example, the following transfer function describes a Chebyshev low-pass filter with a corner
(pass-band) frequency of 1 rad/s:
1.0
N(s) = 0.139713 · s2 +1.09773s+1.10251

In order to define an s_xfer model for the above, but with the corner frequency equal to 1500
rad/s (9425 Hz), the following instance and model lines would be needed:
a12 node1 node2 cheby1
.model cheby1 s_xfer(num_coeff=[1] den_coeff=[1 1.09773 1.10251]
+
int_ic=[0 0] denormalized_freq=1500)
In the above, you add the normalized coefficients and scale the filter through the use of the
denormalized freq parameter. Similar results could have been achieved by performing the denormalization prior to specification of the coefficients, and setting denormalized freq to the
value 1.0 (or not specifying the frequency, as the default is 1.0 rad/s) Note in the above that
frequencies are always specified as radians/second.
Truncation error checking is included in the s-domain transfer block. This should provide for
more accurate simulations, since the model will inherently request smaller time increments
between simulation points if truncation errors would otherwise be excessive.
The int_ic parameter is an array that must be of size one less as the array of values specified for
the den_coeff parameter. Even if a 0 start value is required, you have to add the specific int_ic
vector to the set of coefficients (see the examples above and below).
Example SPICE Usage:
a14 9 22 cheby_LP_3kHz
.
.
.model cheby_LP_3kHz s_xfer(in_offset=0.0 gain=1.0 int_ic=[0 0]
+
num_coeff=[1.0]
+
den_coeff=[1.0 1.42562 1.51620])

12.2. ANALOG MODELS

12.2.17

179

Slew Rate Block

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

cm_slew
slew
"A simple slew rate follower block"
in
"input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

rise_slope
"maximum rising slope value"
real
1.0e9
no
yes
fall_slope
"maximum falling slope value"
real
1.0e9
no
yes
range
"smoothing range"
real
0.1
no
yes

Description: This function is a simple slew rate block that limits the absolute slope of the
output with respect to time to some maximum or value. The actual slew rate effects of
over-driving an amplifier circuit can thus be accurately modeled by cascading the amplifier with this model. The units used to describe the maximum rising and falling slope
values are expressed in volts or amperes per second. Thus a desired slew rate of 0.5 V/µs
will be expressed as 0.5e+6, etc.

180

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
The slew rate block will continue to raise or lower its output until the difference between
the input and the output values is zero. Thereafter, it will resume following the input signal, unless the slope again exceeds its rise or fall slope limits. The range input specifies
a smoothing region above or below the input value. Whenever the model is slewing and
the output comes to within the input + or - the range value, the partial derivative of the
output with respect to the input will begin to smoothly transition from 0.0 to 1.0. When
the model is no longer slewing (output = input), dout/din will equal 1.0.
Example SPICE Usage:
a15 1 2 slew1
.model slew1 slew(rise_slope=0.5e6 fall_slope=0.5e6)

12.2.18

Inductive Coupling

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

cm_lcouple
lcouple
"inductive coupling (for use with ’core’ model)"
l
"inductor"
inout
hd
[h,hd]
no
no

mmf_out
"mmf output (in ampere-turns)"
inout
hd
[hd]
no
no

num_turns
"number of inductor turns"
real
1.0
no
yes

Description: This function is a conceptual model that is used as a building block to create a
wide variety of inductive and magnetic circuit models. This function is normally used in
conjunction with the core model, but can also be used with resistors, hysteresis blocks,
etc. to build up systems that mock the behavior of linear and nonlinear components.
The lcouple takes as an input (on the ‘l’ port), a current. This current value is multiplied
by the num_turns value, N, to produce an output value (a voltage value that appears on the
mmf_out port). The mmf_out acts similar to a magnetomotive force in a magnetic circuit;
when the lcouple is connected to the core model, or to some other resistive device, a
current will flow. This current value (which is modulated by whatever the lcouple is
connected to) is then used by the lcouple to calculate a voltage ‘seen’ at the l port. The
voltage is a function of the derivative with respect to time of the current value seen at

12.2. ANALOG MODELS

181

mmf_out.
The most common use for lcouples will be as a building block in the construction of
transformer models. To create a transformer with a single input and a single output, you
would require two lcouple models plus one core model. The process of building up
such a transformer is described under the description of the core model, below.
Example SPICE Usage:
a150 (7 0) (9 10) lcouple1
.model lcouple1 lcouple(num_turns=10.0)

12.2.19

Magnetic Core

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector: no
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:

cm_core
core
"magnetic core"
mc
"magnetic core"
inout
gd
[g,gd]
no
H_array
"magnetic field array"
real
yes
[2 -]
no

B_array
"flux density array"
real
yes
[2 -]
no

area
"cross-sectional area"
real
no
no

length
"core length"
real
no
no

input_domain
"input sm. domain"
real
0.01

182

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:

[1e-12 0.5]
no
yes
fraction
"smoothing fraction/abs switch"
boolean
TRUE
no
yes
mode
"mode switch (1 = pwl, 2 = hyst)"
int
1
[1 2]
no
yes
in_low
"input low value"
real
0.0
no
yes

in_high
"input high value"
real
1.0
no
yes

hyst
"hysteresis"
real
0.1
[0 -]
no
yes

out_lower_limit
"output lower limit"
real
0.0
no
yes

out_upper_limit
"output upper limit"
real
1.0
no
-

12.2. ANALOG MODELS
Null_Allowed:

183
yes

Description: This function is a conceptual model that is used as a building block to create
a wide variety of inductive and magnetic circuit models. This function is almost always
expected to be used in conjunction with the lcouple model to build up systems that mock
the behavior of linear and nonlinear magnetic components. There are two fundamental
modes of operation for the core model. These are the pwl mode (which is the default, and
which is the most likely to be of use to you) and the hysteresis mode. These are detailed
below.
PWL Mode (mode = 1)
The core model in PWL mode takes as input a voltage that it treats as a magnetomotive force
(mmf) value. This value is divided by the total effective length of the core to produce a value
for the Magnetic Field Intensity, H. This value of H is then used to find the corresponding Flux
Density, B, using the piecewise linear relationship described by you in the H array / B array
coordinate pairs. B is then multiplied by the cross-sectional area of the core to find the Flux
value, which is output as a current. The pertinent mathematical equations are listed below:
H = mmf =L, where L = Length
Here H, the Magnetic Field Intensity, is expressed in ampere-turns/meter.
B = f (H)
The B value is derived from a piecewise linear transfer function described to the model via
the (H_array[],B_array[]) parameter coordinate pairs. This transfer function does not include
hysteretic effects; for that, you would need to substitute a HYST model for the core.
φ = BA, where A = Area
The final current allowed to flow through the core is equal to φ . This value in turn is used by
the "lcouple" code model to obtain a value for the voltage reflected back across its terminals to
the driving electrical circuit.
The following example code shows the use of two lcouple models and one core model to
produce a simple primary/secondary transformer.
Example SPICE Usage:
a1 (2 0) (3 0) primary
.model primary lcouple (num_turns = 155)
a2 (3 4) iron_core
.model iron_core core (H_array = [-1000 -500 -375 -250 -188 -125 -63 0
+
63 125 188 250 375 500 1000]
+
B_array = [-3.13e-3 -2.63e-3 -2.33e-3 -1.93e-3
+
-1.5e-3 -6.25e-4 -2.5e-4 0 2.5e-4
+
6.25e-4 1.5e-3 1.93e-3 2.33e-3
+
2.63e-3 3.13e-3]
+
area = 0.01 length = 0.01)
a3 (5 0) (4 0) secondary
.model secondary lcouple (num_turns = 310)

184

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
HYSTERESIS Mode (mode = 2)

The core model in HYSTERESIS mode takes as input a voltage that it treats as a magnetomotive
force (mmf) value. This value is used as input to the equivalent of a hysteresis code model block.
The parameters defining the input low and high values, the output low and high values, and the
amount of hysteresis are as in that model. The output from this mode, as in PWL mode, is a
current value that is seen across the mc port. An example of the core model used in this fashion
is shown below:
Example SPICE Usage:
a1 (2 0) (3 0) primary
.model primary lcouple (num_turns = 155)
a2 (3 4) iron_core
.model iron_core core (mode = 2 in_low=-7.0 in_high=7.0
+
out_lower_limit=-2.5e-4 out_upper_limit=2.5e-4
+
hyst = 2.3 )
a3 (5 0) (4 0) secondary
.model secondary lcouple (num_turns = 310)
One final note to be made about the two core model nodes is that certain parameters are available in one mode, but not in the other. In particular, the in_low, in_high, out_lower_limit,
out_upper_limit, and hysteresis parameters are not available in PWL mode. Likewise, the
H_array, B_array, area, and length values are unavailable in HYSTERESIS mode. The input
domain and fraction parameters are common to both modes (though their behavior is somewhat
different; for explanation of the input domain and fraction values for the HYSTERESIS mode,
you should refer to the hysteresis code model discussion).

12.2.20

Controlled Sine Wave Oscillator

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:

cm_sine
sine
"controlled sine wave oscillator"
cntl_in
"control input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

cntl_array
"control array"
real
0.0

freq_array
"frequency array"
real
1.0e3

12.2. ANALOG MODELS
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

185
yes
[2 -]
no

[0 -]
yes
cntl_array
no

out_low
out_high
"output peak low value" "output peak high value"
real
real
-1.0
1.0
no
no
yes
yes

Description: This function is a controlled sine wave oscillator with parametrizable values of
low and high peak output. It takes an input voltage or current value. This value is used as
the independent variable in the piecewise linear curve described by the coordinate points
of the cntl array and freq array pairs. From the curve, a frequency value is determined,
and the oscillator will output a sine wave at that frequency. From the above, it is easy
to see that array sizes of 2 for both the cntl array and the freq array will yield a linear
variation of the frequency with respect to the control input. Any sizes greater than 2 will
yield a piecewise linear transfer characteristic. For more detail, refer to the description of
the piecewise linear controlled source, which uses a similar method to derive an output
value given a control input.
Example SPICE Usage:
asine 1 2 in_sine
.model in_sine sine(cntl_array = [-1 0 5 6]
+
freq_array=[10 10 1000 1000] out_low = -5.0
+
out_high = 5.0)

12.2.21

Controlled Triangle Wave Oscillator

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:

cm_triangle
triangle
"controlled triangle wave oscillator"
cntl_in
"control input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

cntl_array

freq_array

186

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

"control array"
real
0.0
yes
[2 -]
no

"frequency array"
real
1.0e3
[0 -]
yes
cntl_array
no

out_low
out_high
"output peak low value" "output peak high value"
real
real
-1.0
1.0
no
no
yes
yes
duty_cycle
"rise time duty cycle"
real
0.5
[1e-10 0.999999999]
no
yes

Description: This function is a controlled triangle/ramp wave oscillator with parametrizable
values of low and high peak output and rise time duty cycle. It takes an input voltage or
current value. This value is used as the independent variable in the piecewise linear curve
described by the coordinate points of the cntl_array and freq_array pairs.
From the curve, a frequency value is determined, and the oscillator will output a triangle
wave at that frequency. From the above, it is easy to see that array sizes of 2 for both the
cntl_array and the freq_array will yield a linear variation of the frequency with respect to
the control input. Any sizes greater than 2 will yield a piecewise linear transfer characteristic. For more detail, refer to the description of the piecewise linear controlled source,
which uses a similar method to derive an output value given a control input.
Example SPICE Usage:
ain 1 2 ramp1
.model ramp1 triangle(cntl_array = [-1 0 5 6]
+
freq_array=[10 10 1000 1000] out_low = -5.0
+
out_high = 5.0 duty_cycle = 0.9)

12.2.22

Controlled Square Wave Oscillator

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:

cm_square
square

12.2. ANALOG MODELS
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER.TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector: no
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

187
"controlled square wave oscillator"
cntl_in
"control input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

cntl_array
"control array"
real
0.0
yes
[2 -]
no

freq_array
"frequency array"
real
1.0e3
[0 -]
yes
cntl_array
no

out_low
out_high
"output peak low value" "output peak high value"
real
real
-1.0
1.0
no
no
yes
yes
duty_cycle
"duty cycle"
real
0.5
[1e-6 0.999999]

rise_time
"output rise time"
real
1.0e-9
-

yes

yes

fall_time
"output fall time"
real
1.0e-9
no
yes

Description: This function is a controlled square wave oscillator with parametrizable values
of low and high peak output, duty cycle, rise time, and fall time. It takes an input voltage

188

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
or current value. This value is used as the independent variable in the piecewise linear
curve described by the coordinate points of the cntl_array and freq_array pairs. From the
curve, a frequency value is determined, and the oscillator will output a square wave at
that frequency.
From the above, it is easy to see that array sizes of 2 for both the cntl_array and the
freq_array will yield a linear variation of the frequency with respect to the control input.
Any sizes greater than 2 will yield a piecewise linear transfer characteristic. For more
detail, refer to the description of the piecewise linear controlled source, which uses a
similar method to derive an output value given a control input.
Example SPICE Usage:
ain 1 2 pulse1
.model pulse1 square(cntl_array = [-1 0 5 6]
+
freq_array=[10 10 1000 1000] out_low = 0.0
+
out_high = 4.5 duty_cycle = 0.2
+
rise_time = 1e-6 fall_time = 2e-6)

12.2.23

Controlled One-Shot

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:

cm_oneshot
oneshot
"controlled one-shot"
clk
"clock input"
in
v
[v,vd,i,id]
no
no

cntl_in
"control input"
in
v
[v,vd,i,id]
no
yes

clear
"clear signal"
in
v
[v,vd,i,id]
no
yes

out
"output"
out
v
[v,vd,i,id]
no
no

clk_trig
"clock trigger value"
real
0.5
no

retrig
"retrigger switch"
boolean
FALSE
no

12.2. ANALOG MODELS
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:

189
no

yes

pos_edge_trig
"positive/negative edge trigger switch"
boolean
TRUE
no
no
cntl_array
"control array"
real
0.0
yes
yes

pw_array
"pulse width array"
real
1.0e-6
[0.00 -]
yes
cntl_array
yes

out_low
"output low value"
real
0.0
no
yes

out_high
"output high value"
real
1.0
no
yes

fall_time
"output fall time"
real
1.0e-9
no
yes

rise_time
"output rise time"
real
1.0e-9
no
yes

rise_delay
"output delay from trigger"
real
1.0e-9
no
yes

190

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

fall_delay
"output delay from pw"
real
1.0e-9
no
yes

Description: This function is a controlled oneshot with parametrizable values of low and high
peak output, input trigger value level, delay, and output rise and fall times. It takes an
input voltage or current value. This value is used as the independent variable in the
piecewise linear curve described by the coordinate points of the cntl_array and pw_array
pairs. From the curve, a pulse width value is determined. The one-shot will output a
pulse of that width, triggered by the clock signal (rising or falling edge), delayed by the
delay value, and with specified rise and fall times. A positive slope on the clear input will
immediately terminate the pulse, which resets with its fall time.
From the above, it is easy to see that array sizes of 2 for both the cntl_array and the
pw_array will yield a linear variation of the pulse width with respect to the control input.
Any sizes greater than 2 will yield a piecewise linear transfer characteristic. For more
detail, refer to the description of the piecewise linear controlled source, which uses a
similar method to derive an output value given a control input.
Example SPICE Usage:
ain 1 2 3 4 pulse2
.model pulse2 oneshot(cntl_array = [-1 0 10 11]
+
pw_array=[1e-6 1e-6 1e-4 1e-4]
+
clk_trig = 0.9 pos_edge_trig = FALSE
+
out_low = 0.0 out_high = 4.5
+
rise_delay = 20.0-9 fall_delay = 35.0e-9)

12.2.24

Capacitance Meter

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:

cm_cmeter
cmeter
"capacitance meter"
in
"input"
in
v
[v,vd,i,id]
no
no
gain

out
"output"
out
v
[v,vd,i,id]
no
no

12.2. ANALOG MODELS
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

191
"gain"
real
1.0
no
yes

Description: The capacitance meter is a sensing device that is attached to a circuit node and
produces as an output a scaled value equal to the total capacitance seen on its input multiplied by the gain parameter. This model is primarily intended as a building block for
other models that must sense a capacitance value and alter their behavior based upon it.
Example SPICE Usage:
atest1 1 2 ctest
.model ctest cmeter(gain=1.0e12)

12.2.25

Inductance Meter

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

cm_lmeter
lmeter
"inductance meter"
in
"input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
v
[v,vd,i,id]
no
no

gain
"gain"
real
1.0
no
yes

Description: The inductance meter is a sensing device that is attached to a circuit node and
produces as an output a scaled value equal to the total inductance seen on its input multiplied by the gain parameter. This model is primarily intended as a building block for
other models that must sense an inductance value and alter their behavior based upon it.

192

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Example SPICE Usage:
atest2 1 2 ltest
.model ltest lmeter(gain=1.0e6)

12.2.26

Memristor

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

cm_memristor
memristor
"Memristor Interface"
memris
"memristor terminals"
inout
gd
[gd]
no
no
rmin
"minimum resistance"
real
10.0
no
no

rmax
"maximum resistance"
real
10000.0
no
no

rinit
"initial resistance"
real
7000.0
no
no

vt
"threshold"
real
0.0
no
no

alpha
"model parameter 1"
real
0.0
no
no

beta
"model parameter 2"
real
1.0
no
no

Description: The memristor is a two-terminal resistor with memory, whose resistance depends

12.2. ANALOG MODELS

193

on the time integral of the voltage across its terminals. rmin and rmax provide the lower
and upper limits of the resistance, rinit is its starting value (no voltage applied so far).
The voltage has to be above a threshold vt to become effective in changing the resistance.
alpha and beta are two model parameters. The memristor code model is derived from a
SPICE subcircuit published in [23].
Example SPICE Usage:
amen 1 2 memr
.model memr memristor (rmin=1k rmax=10k rinit=7k
+ alpha=0 beta=2e13 vt=1.6)

12.2.27

2D table model

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:

cm_table2D
table2D
"2D table model"
inx
"inputx"
in
v
[v,vd,i,id,vnam]
no
no

iny
"inputy"
in
v
[v,vd,i,id,vnam]
no
no

order
"order"
int
3
no
yes

verbose
"verbose"
int
0
no
yes

offset
"offset"
real
0.0
no
yes

gain
"gain"
real
1.0
no
yes

file
"file name"
string

out
"output"
out
i
[v,vd,i,id]
no
no

194

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

"2D-table-model.txt"
no
yes

Description: The 2D table model reads a matrix from file "file name" (default 2D-tablemodel.txt) which has x columns and y rows. Each x,y pair, addressed by inx and iny,
yields an output value out. Linear interpolation is used for out, eno (essentially non
oscillating) interpolation for its derivatives. Parameters offset (default 0) and gain (default 1) modify the output table values according to o f f set + gain out. Parameter order
(default 3) influences the calculation of the derivatives. Parameter verbose (default 0)
yields test outputs, if set to 1 or 2. The table format is shown below. Be careful to include
the data point inx = 0, iny = 0 into your table, because ngspice uses these during .OP
computations. The x horizontal and y vertical address values have to increase monotonically. The usage example consists of two input voltages referenced to ground and a
current source output with two floating nodes.
Table Example:
* table source
* number of columns (x)
8
* number of rows (y)
9
* x horizontal (column) address values (real numbers)
-1 0 1 2 3 4 5 6
* y vertical (row) address values (real numbers)
-0.6 0 0.6 1.2 1.8 2.4 3.0 3.6 4.2
* table with output data (horizontally addressed by x, vertically by y)
1 0.9 0.8 0.7 0.6 0.5 0.4 0.3
1 1 1 1 1 1 1 1
1 1.2 1.4 1.6 1.8 2 2.2 2.4
1 1.5 2 2.5 3 3.5 4 4.5
1 2 3 4 5 6 7 8
1 2.5 4 5.5 7 8.5 10 11.5
1 3 5 7 9 11 13 15
1 3.5 6 8.5 11 13.5 16 18.5
1 4 7 10 13 16 19 22
Example SPICE Usage:
atab inx iny %id(out1 out2) tabmod
.model tabmod table2d (offset=0.0 gain=1 order=3 file="table-simple.txt")

12.2.28

3D table model

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:

cm_table3D
table3D
"3D table model"

12.2. ANALOG MODELS
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

195

inx
"inputx"
in
v
[v,vd,i,id,vnam]
no
no

iny
"inputy"
in
v
[v,vd,i,id,vnam]
no
no

inz
"inputz"
in
v
[v,vd,i,id,vnam]
no
no

out
"output"
out
i
[v,vd,i,id]
no
no
order
"order"
int
3
no
yes
offset
"offset"
real
0.0
no
yes

verbose
"verbose"
int
0
no
yes
gain
"gain"
real
1.0
no
yes

file
"file name"
string
"3D-table-model.txt"
no
yes

Description: The 3D table model reads a matrix from file "file name" (default 3D-tablemodel.txt) which has x columns, y rows per table and z tables. Each x,y,z triple, addressed by inx, iny, and inz, yields an output value out. Linear interpolation is used

196

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
for out, eno (essentially non oscillating) interpolation for its derivatives. Parameters
offset (default 0) and gain (default 1) modify the output table values according to
o f f set + gain out. Parameter order (default 3) influences the calculation of the derivatives. Parameter verbose (default 0) yields test outputs, if set to 1 or 2. The table
format is shown below. Be careful to include the data point inx = 0, iny = 0, inz = 0 into
your table, because ngspice needs these to for the .OP calculation. The x horizontal, y
vertical, and z table address values have to increase monotonically. The usage example
simulates a NMOS transistor with independent drain, gate and bulk nodes, referenced to
source. Parameter gain may be used to emulate transistor width, with respect to the table
transistor.

Table Example:
* 3D table for nmos bsim 4, W=10um, L=0.13um
*x
39
*y
39
*z
11
*x (drain voltage)
-0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 ...
*y (gate voltage)
-0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 ...
*z (substrate voltage)
-1.8 -1.6 -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2
*table -1.8
-4.50688E-10 -4.50613E-10 -4.50601E-10 -4.50599E-10 ...
-4.49622E-10 -4.49267E-10 -4.4921E-10 -4.49202E-10 ...
-4.50672E-10 -4.49099E-10 -4.48838E-10 -4.48795E-10 ...
-4.55575E-10 -4.4953E-10 -4.48435E-10 -4.48217E-10 ...
...
*table -1.6
-3.10015E-10 -3.09767E-10 -3.0973E-10 -3.09724E-10 ...
-3.09748E-10 -3.08524E-10 -3.08339E-10 -3.08312E-10 ...
...
*table -1.4
-2.04848E-10 -2.04008E-10 -2.03882E-10 ...
-2.07275E-10 -2.03117E-10 -2.02491E-10 ...
...
Example SPICE Usage:
amos1 %vd(d s) %vd(g s) %vd(b s) %id(d s) mostable1
.model mostable1 table3d (offset=0.0 gain=0.5 order=3
+ verbose=1 file="table-3D-bsim4n.txt")

12.3. HYBRID MODELS

12.3

197

Hybrid Models

The following hybrid models are supplied with XSPICE. The descriptions included below consist of the model Interface Specification File and a description of the model’s operation. This
is followed by an example of a simulator-deck placement of the model, including the .MODEL
card and the specification of all available parameters.
A note should be made with respect to the use of hybrid models for other than simple digital-toanalog and analog-to-digital translations. The hybrid models represented in this section address
that specific need, but in the development of user-defined nodes you may find a need to translate
not only between digital and analog nodes, but also between real and digital, real and int, etc.
In most cases such translations will not need to be as involved or as detailed as shown in the
following.

12.3.1

Digital-to-Analog Node Bridge

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:

cm_dac_bridge
dac_bridge
"digital-to-analog node bridge"
in
"input"
in
d
[d]
yes
no

out
"output"
out
v
[v,vd,i,id,d]
yes
no

out_low
"0-valued analog output"
real
0.0
no
yes
out_high
"1-valued analog output"
real
1.0
no
yes

198

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

out_undef
"U-valued analog output"
real
0.5
no
yes

input_load
"input load (F)"
real
1.0e-12
no
yes

t_rise
"rise time 0->1"
real
1.0e-9
no
yes

t_fall
"fall time 1->0"
real
1.0e-9
no
yes

Description: The dac_bridge is the first of two node bridge devices designed to allow for
the ready transfer of digital information to analog values and back again. The second
device is the adc_bridge (which takes an analog value and maps it to a digital one).The
dac_bridge takes as input a digital value from a digital node. This value by definition
may take on only one of the values ‘0’, ‘1’ or ‘U’. The dac_bridge then outputs the value
out_low, out_high or out_undef, or ramps linearly toward one of these ‘final’ values
from its current analog output level. The speed at which this ramping occurs depends
on the values of t_rise and t_fall. These parameters are interpreted by the model
such that the rise or fall slope generated is always constant. Note that the dac_bridge
includes test code in its cfunc.mod file for determining the presence of the out_undef parameter. If this parameter is not specified by you, and if out_high and out_low values are
specified, then out_undef is assigned the value of the arithmetic mean of out_high and
out_low. This simplifies coding of output buffers, where typically a logic family will
include an out_low and out_high voltage, but not an out_undef value. This model
also posts an input load value (in farads) based on the parameter input load.
Example SPICE Usage:
abridge1 [7] [2] dac1
.model dac1 dac_bridge(out_low = 0.7 out_high = 3.5 out_undef = 2.2
+
input_load = 5.0e-12 t_rise = 50e-9
+
t_fall = 20e-9)

12.3.2

Analog-to-Digital Node Bridge

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:

cm_adc_bridge
adc_bridge
"analog-to-digital node bridge"
in

out

12.3. HYBRID MODELS
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

199
"input"
in
v
[v,vd,i,id,d]
yes
no

"output"
out
d
[d]
yes
no

in_low
"maximum 0-valued analog input"
real
1.0
no
yes
in_high
"minimum 1-valued analog input"
real
2.0
no
yes
rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

Description: The adc_bridge is one of two node bridge devices designed to allow for the
ready transfer of analog information to digital values and back again. The second device
is the dac_bridge (which takes a digital value and maps it to an analog one). The
adc_bridge takes as input an analog value from an analog node. This value by definition
may be in the form of a voltage, or a current. If the input value is less than or equal to
in_low, then a digital output value of ‘0’ is generated. If the input is greater than or equal
to in_high, a digital output value of ‘1’ is generated. If neither of these is true, then
a digital ‘UNKNOWN’ value is output. Note that unlike the case of the dac_bridge,
no ramping time or delay is associated with the adc_bridge. Rather, the continuous
ramping of the input value provides for any associated delays in the digitized signal.
Example SPICE Usage:
abridge2 [1] [8] adc_buff
.model adc_buff adc_bridge(in_low = 0.3 in_high = 3.5)

200

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

12.3.3

Controlled Digital Oscillator

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

cm_d_osc
d_osc
"controlled digital oscillator"
cntl_in
"control input"
in
v
[v,vd,i,id]
no
no

out
"output"
out
d
[d]
no
no

cntl_array
"control array"
real
0.0
yes
[2 -]
no

freq_array
"frequency array"
real
1.0e6
[0 -]
yes
cntl_array
no

duty_cycle
"duty cycle"
real
0.5
[1e-6 0.999999]
no
yes

init_phase
"initial phase of output"
real
0
[-180.0 +360.0]
no
yes

rise_delay
"rise delay"
real
1e-9
[0 -]
no
yes

fall_delay
"fall delay"
real
1e-9
[0 -]
no
yes

Description: The digital oscillator is a hybrid model that accepts as input a voltage or current.
This input is compared to the voltage-to-frequency transfer characteristic specified by the
cntl_array/freq_array coordinate pairs, and a frequency is obtained that represents
a linear interpolation or extrapolation based on those pairs. A digital time-varying signal
is then produced with this fundamental frequency.
The output waveform, which is the equivalent of a digital clock signal, has rise and fall

12.3. HYBRID MODELS

201

delays that can be specified independently. In addition, the duty cycle and the phase of
the waveform are also variable and can be set by you.
Example SPICE Usage:
a5 1 8 var_clock
.model var_clock d_osc(cntl_array
+
freq_array
+
duty_cycle
+
rise_delay

12.3.4

=
=
=
=

[-2 -1 1 2]
[1e3 1e3 10e3 10e3]
0.4 init_phase = 180.0
10e-9 fall_delay=8e-9)

Node bridge from digital to real with enable

NAME_TABLE:
Spice_Model_Name: d_to_real
C_Function_Name: ucm_d_to_real
Description: "Node bridge from digital to real
PORT_TABLE:
Port_Name:
in
enable
out
Description:
"input"
"enable"
"output"
Direction:
in
in
out
Default_Type:
d
d
real
Allowed_Types: [d]
[d]
[real]
Vector:
no
no
no
Vector_Bounds: Null_Allowed:
no
yes
no
PARAMETER_TABLE:
Parameter_Name: zero
one
Description:
"value for 0"
"value for 1"
Data_Type:
real
real
Default_Value: 0.0
1.0
Limits:
Vector:
no
no
Vector_Bounds: Null_Allowed:
yes
yes

12.3.5

with enable"

delay
"delay"
real
1e-9
[1e-15 -]
no
yes

A Z**-1 block working on real data

NAME_TABLE:
Spice_Model_Name: real_delay
C_Function_Name: ucm_real_delay
Description: "A Z ** -1 block working on real data"
PORT_TABLE:
Port_Name:
in
clk
out
Description:
"input"
"clock"
"output"
Direction:
in
in
out
Default_Type:
real
d
real
Allowed_Types:
[real]
[d]
[real]

202

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

12.3.6

no
no

no
no

no
no

delay
"delay from clk to out"
real
1e-9
[1e-15 -]
no
yes

A gain block for event-driven real data

NAME_TABLE:
Spice_Model_Name:
C_Function_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

real_gain
ucm_real_gain
"A gain block for event-driven real data"
in
"input"
in
real
[real]
no
no

out
"output"
out
real
[real]
no
no

in_offset
"input offset"
real
0.0
no
yes

gain
"gain"
real
1.0
no
yes

delay
"delay"
real
1.0e-9
no
yes

ic
"initial condition"
real
0.0
no
yes

out_offset
"output offset"
real
0.0
no
yes

12.4. DIGITAL MODELS

12.3.7

203

Node bridge from real to analog voltage

NAME_TABLE:
Spice_Model_Name:
C_Function_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

12.4

real_to_v
ucm_real_to_v
"Node bridge from real to analog voltage"
in
"input"
in
real
[real]
no
no

out
"output"
out
v
[v, vd, i, id]
no
no

gain
"gain"
real
1.0
no
yes

transition_time
"output transition time"
real
1e-9
[1e-15 -]
no
yes

Digital Models

The following digital models are supplied with XSPICE. The descriptions included below consist of an example model Interface Specification File and a description of the model’s operation. This is followed by an example of a simulator-deck placement of the model, including the
.MODEL card and the specification of all available parameters. Note that these models have not
been finalized at this time.
Some information common to all digital models and/or digital nodes is included here. The following are general rules that should make working with digital nodes and models more straightforward:
1. All digital nodes are initialized to ZERO at the start of a simulation (i.e., when INIT=TRUE).
This means that a model need not post an explicit value to an output node upon initialization if its output would normally be a ZERO (although posting such would certainly
cause no harm).

12.4.1

Buffer

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:

cm_d_buffer
d_buffer
"digital one-bit-wide buffer"

204

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

in
"input"
in
d
[d]
no
no

out
"output"
out
d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The buffer is a single-input, single-output digital buffer that produces as output
a time-delayed copy of its input. The delays associated with an output rise and those
associated with an output fall may be different. The model also posts an input load value
(in farads) based on the parameter input load. The output of this model does not, however,
respond to the total loading it sees on its output; it will always drive the output strongly
with the specified delays.
Example SPICE Usage:
a6 1 8 buff1
.model buff1 d_buffer(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.2

Inverter

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:

cm_d_inverter
d_inverter
"digital one-bit-wide inverter"
in

out

12.4. DIGITAL MODELS
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

205
"input"
in
d
[d]
no
no

"output"
out
d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The inverter is a single-input, single-output digital inverter that produces as output an inverted, time-delayed copy of its input. The delays associated with an output rise
and those associated with an output fall may be specified independently. The model also
posts an input load value (in farads) based on the parameter input load. The output of this
model does not, however, respond to the total loading it sees on its output; it will always
drive the output strongly with the specified delays.
Example SPICE Usage:
a6 1 8 inv1
.model inv1 d_inverter(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.3

And

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_d_and
d_and
"digital ‘and’ gate"
in
"input"
in

out
"output"
out

206

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

d
[d]
yes
[2 -]
no

d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The digital and gate is an n-input, single-output and gate that produces an active
‘1’ value if, and only if, all of its inputs are also ‘1’ values. If ANY of the inputs is a
‘0’, the output will also be a ‘0’; if neither of these conditions holds, the output will be
unknown. The delays associated with an output rise and those associated with an output
fall may be specified independently. The model also posts an input load value (in farads)
based on the parameter input load. The output of this model does not, however, respond
to the total loading it sees on its output; it will always drive the output strongly with the
specified delays.
Example SPICE Usage:
a6 [1 2] 8 and1
.model and1 d_and(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.4

Nand

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_d_nand
d_nand
"digital ‘nand’ gate"
in
"input"
in

out
"output"
out

12.4. DIGITAL MODELS
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

207
d
[d]
yes
[2 -]
no

d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The digital nand gate is an n-input, single-output nand gate that produces an
active ‘0’ value if and only if all of its inputs are ‘1’ values. If ANY of the inputs is a ‘0’,
the output will be a ‘1’; if neither of these conditions holds, the output will be unknown.
The delays associated with an output rise and those associated with an output fall may be
specified independently. The model also posts an input load value (in farads) based on
the parameter input load. The output of this model does not, however, respond to the total
loading it sees on its output; it will always drive the output strongly with the specified
delays.
Example SPICE Usage:
a6 [1 2 3] 8 nand1
.model nand1 d_nand(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.5

Or

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_d_or
d_or
"digital ‘or’ gate"
in
"input"
in

out
"output"
out

208

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

d
[d]
yes
[2 -]
no

d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The digital or gate is an n-input, single-output or gate that produces an active
‘1’ value if at least one of its inputs is a ‘1’ value. The gate produces a ‘0’ value if
all inputs are ‘0’; if neither of these two conditions holds, the output is unknown. The
delays associated with an output rise and those associated with an output fall may be
specified independently. The model also posts an input load value (in farads) based on
the parameter input load. The output of this model does not, however, respond to the total
loading it sees on its output; it will always drive the output strongly with the specified
delays.
Example SPICE Usage:
a6 [1 2 3] 8 or1
.model or1 d_or(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.6

Nor

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_d_nor
d_nor
"digital ‘nor’ gate"
in
"input"
in

out
"output"
out

12.4. DIGITAL MODELS
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

209
d
[d]
yes
[2 -]
no

d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The digital nor gate is an n-input, single-output nor gate that produces an active
‘0’ value if at least one of its inputs is a ‘1’ value. The gate produces a ‘0’ value if
all inputs are ‘0’; if neither of these two conditions holds, the output is unknown. The
delays associated with an output rise and those associated with an output fall may be
specified independently. The model also posts an input load value (in farads) based on
the parameter input load. The output of this model does not, however, respond to the total
loading it sees on its output; it will always drive the output strongly with the specified
delays.
Example SPICE Usage:
anor12 [1 2 3 4] 8 nor12
.model nor12 d_or(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.7

Xor

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_d_xor
d_xor
"digital exclusive-or gate"
in
"input"
in

out
"output"
out

210

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

d
[d]
yes
[2 -]
no

d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The digital xor gate is an n-input, single-output xor gate that produces an active
‘1’ value if an odd number of its inputs are also ‘1’ values. The delays associated with an
output rise and those associated with an output fall may be specified independently.
The model also posts an input load value (in farads) based on the parameter input load.
The output of this model does not, however, respond to the total loading it sees on its
output; it will always drive the output strongly with the specified delays. Note also that
to maintain the technology-independence of the model, any UNKNOWN input, or any
floating input causes the output to also go UNKNOWN.
Example SPICE Usage:
a9 [1 2] 8 xor3
.model xor3 d_xor(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.8

Xnor

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_d_xnor
d_xnor
"digital exclusive-nor gate"
in
"input"
in

out
"output"
out

12.4. DIGITAL MODELS
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

211
d
[d]
yes
[2 -]
no

d
[d]
no
no

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

input_load
"input load value (F)"
real
1.0e-12
no
yes

Description: The digital xnor gate is an n-input, single-output xnor gate that produces an
active ‘0’ value if an odd number of its inputs are also ‘1’ values. It produces a ‘1’ output
when an even number of ‘1’ values occurs on its inputs. The delays associated with an
output rise and those associated with an output fall may be specified independently. The
model also posts an input load value (in farads) based on the parameter input load. The
output of this model does not, however, respond to the total loading it sees on its output; it
will always drive the output strongly with the specified delays. Note also that to maintain
the technology-independence of the model, any UNKNOWN input, or any floating input
causes the output to also go UNKNOWN.
Example SPICE Usage:
a9 [1 2] 8 xnor3
.model xnor3 d_xnor(rise_delay = 0.5e-9 fall_delay = 0.3e-9
+
input_load = 0.5e-12)

12.4.9

Tristate

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:

cm_d_tristate
d_tristate
"digital tristate buffer"
in
"input"

enable
"enable"

out
"output"

212

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

in
d
[d]
no
no

in
d
[d]
no
no

out
d
[d]
no
no

delay
"delay"
real
1.0e-9
[1.0e-12 -]
no
yes
input_load
"input load value (F)"
real
1.0e-12
no
yes
enable_load
"enable load value (F)"
real
1.0e-12
no
yes

Description: The digital tristate is a simple tristate gate that can be configured to allow for
open-collector behavior, as well as standard tristate behavior. The state seen on the input
line is reflected in the output. The state seen on the enable line determines the strength of
the output. Thus, a ONE forces the output to its state with a STRONG strength. A ZERO
forces the output to go to a HI_IMPEDANCE strength. The delays associated with an
output state or strength change cannot be specified independently, nor may they be specified independently for rise or fall conditions; other gate models may be used to provide
such delays if needed. The model posts input and enable load values (in farads) based
on the parameters input load and enable. The output of this model does not, however,
respond to the total loading it sees on its output; it will always drive the output with the
specified delay. Note also that to maintain the technology-independence of the model,
any UNKNOWN input, or any floating input causes the output to also go UNKNOWN.
Likewise, any UNKNOWN input on the enable line causes the output to go to an UNDETERMINED strength value.

12.4. DIGITAL MODELS

213

Example SPICE Usage:
a9 1 2 8 tri7
.model tri7 d_tristate(delay = 0.5e-9 input_load = 0.5e-12
+
enable_load = 0.5e-12)

12.4.10

Pullup

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

cm_d_pullup
d_pullup
"digital pullup resistor"
out
"output"
out
d
[d]
no
no
load
"load value (F)"
real
1.0e-12
no
yes

Description: The digital pullup resistor is a device that emulates the behavior of an analog
resistance value tied to a high voltage level. The pullup may be used in conjunction
with tristate buffers to provide open-collector wired or constructs, or any other logical
constructs that rely on a resistive pullup common to many tristated output devices. The
model posts an input load value (in farads) based on the parameter load.
Example SPICE Usage:
a2 9 pullup1
.model pullup1 d_pullup(load = 20.0e-12)

12.4.11

Pulldown

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:

cm_d_pulldown
d_pulldown
"digital pulldown resistor"

214

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

out
"output"
out
d
[d]
no
no
load
"load value (F)"
real
1.0e-12
no
yes

Description: The digital pulldown resistor is a device that emulates the behavior of an analog
resistance value tied to a low voltage level. The pulldown may be used in conjunction
with tristate buffers to provide open-collector wired or constructs, or any other logical
constructs that rely on a resistive pulldown common to many tristated output devices.
The model posts an input load value (in farads) based on the parameter load.
Example SPICE Usage:
a4 9 pulldown1
.model pulldown1 d_pulldown(load = 20.0e-12)

12.4.12

D Flip Flop

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:

cm_d_dff
d_dff
"digital d-type flip flop"
data
"input data"
in
d
[d]
no
no
set
"asynch. set"
in
d

clk
"clock"
in
d
[d]
no
no
reset
"asynch. reset"
in
d

12.4. DIGITAL MODELS
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector.Bounds:

215
[d]
no
yes

[d]
no
yes

out
"data output"
out
d
[d]
no
yes

Nout
"inverted data output"
out
d
[d]
no
yes

clk_delay
"delay from clk"
real
1.0e-9
[1.0e-12 -]
no
yes

set_delay
"delay from set"
real
1.0e-9
[1.0e-12 -]
no
yes

reset_delay
"delay from reset"
real
1.0e-9
[1.0e-12 -]
no
yes

ic
"output initial state"
int
0
[0 2]
no
yes

data_load
"data load value (F)"
real
1.0e-12
no
yes

clk_load
"clk load value (F)"
real
1.0e-12
no
yes

set_load
"set load value (F)"
real
1.0e-12
no
-

reset_load
"reset load (F)"
real
1.0e-12
no
-

216

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

yes

yes

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

Description: The digital d-type flip flop is a one-bit, edge-triggered storage element that will
store data whenever the clk input line transitions from low to high (ZERO to ONE). In
addition, asynchronous set and reset signals exist, and each of the three methods of changing the stored output of the d_dff have separate load values and delays associated with
them. Additionally, you may specify separate rise and fall delay values that are added to
those specified for the input lines; these allow for more faithful reproduction of the output
characteristics of different IC fabrication technologies.
Note that any UNKNOWN input on the set or reset lines immediately results in an
UNKNOWN output.
Example SPICE Usage:
a7 1 2 3 4 5 6 flop1
.model flop1 d_dff(clk_delay = 13.0e-9 set_delay = 25.0e-9
+
reset_delay = 27.0e-9 ic = 2 rise_delay = 10.0e-9
+
fall_delay = 3e-9)

12.4.13

JK Flip Flop

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:

cm_d_jkff
d_jkff
"digital jk-type flip flop"
j
"j input"
in
d
[d]
no
no
clk
"clock"
in
d
[d]

k
"k input"
in
d
[d]
no
no

12.4. DIGITAL MODELS
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

217
no
no
set
"asynchronous set"
in
d
[d]
no
yes

reset
"asynchronous reset"
in
d
[d]
no
yes

out
"data output"
out
d
[d]
no
yes

Nout
"inverted data output"
out
d
[d]
no
yes

clk_delay
"delay from clk"
real
1.0e-9
[1.0e-12 -]
no
yes

set_delay
"delay from set"
real
1.0e-9
[1.0e-12 -]
no
yes

reset_delay
"delay from reset"
real
1.0e-9
[1.0e-12 -]
no
yes

ic
"output initial state"
int
0
[0 2]
no
yes

jk_load
"j,k load values (F)"
real
1.0e-12
no
yes

clk_load
"clk load value (F)"
real
1.0e-12
no
yes

218

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

set_load
"set load value (F)"
real
1.0e-12
no
yes

reset_load
"reset load (F)"
real
1.0e-12
no
yes

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

Description: The digital jk-type flip flop is a one-bit, edge-triggered storage element that will
store data whenever the clk input line transitions from low to high (ZERO to ONE).
In addition, asynchronous set and reset signals exist, and each of the three methods of
changing the stored output of the d_jkff have separate load values and delays associated
with them. Additionally, you may specify separate rise and fall delay values that are
added to those specified for the input lines; these allow for more faithful reproduction of
the output characteristics of different IC fabrication technologies.
Note that any UNKNOWN inputs other than j or k cause the output to go UNKNOWN
automatically.
Example SPICE Usage:
a8 1 2 3 4 5 6 7 flop2
.model flop2 d_jkff(clk_delay = 13.0e-9 set_delay = 25.0e-9
+
reset_delay = 27.0e-9 ic = 2 rise_delay = 10.0e-9
+
fall_delay = 3e-9)

12.4.14

Toggle Flip Flop

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:

cm_d_tff
d_tff
"digital toggle flip flop"
t
"toggle input"
in
d
[d]
no

clk
"clock"
in
d
[d]
no

12.4. DIGITAL MODELS
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT.TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:

219
no

no

set
"set"
in
d
[d]
no
yes

reset
"reset"
in
d
[d]
no
yes

out
"data output"
out
d
[d]
no
yes

Nout
"inverted data output"
out
d
[d]
no
yes

clk_delay
"delay from clk"
real
1.0e-9
[1.0e-12 -]
no
yes

set_delay
"delay from set"
real
1.0e-9
[1.0e-12 -]
no
yes

reset_delay
"delay from reset"
real
1.0e-9
[1.0e-12 -]
no
yes

ic
"output initial state"
int
0
[0 2]
no
yes

t_load
clk_load
"toggle load value (F)" "clk load value (F)"
real
real
1.0e-12
1.0e-12
no
no
yes
yes

220

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Parameter_Name:
Description:
Data_Type:
Default.Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

set_load
"set load value (F)"
real
1.0e-12
no
yes

reset_load
"reset load (F)"
real
1.0e-12
no
yes

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

Description: The digital toggle-type flip flop is a one-bit, edge-triggered storage element that
will toggle its current state whenever the clk input line transitions from low to high (ZERO
to ONE). In addition, asynchronous set and reset signals exist, and each of the three methods of changing the stored output of the d_tff have separate load values and delays
associated with them. Additionally, you may specify separate rise and fall delay values
that are added to those specified for the input lines; these allow for more faithful reproduction of the output characteristics of different IC fabrication technologies.
Note that any UNKNOWN inputs other than t immediately cause the output to go UNKNOWN.
Example SPICE Usage:
a8 2 12 4 5 6 3 flop3
.model flop3 d_tff(clk_delay = 13.0e-9 set_delay = 25.0e-9
+
reset_delay = 27.0e-9 ic = 2 rise_delay = 10.0e-9
+
fall_delay = 3e-9 t_load = 0.2e-12)

12.4.15

Set-Reset Flip Flop

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:

cm_d_srff
d_srff
"digital set-reset flip flop"
s
"set input"
in
d
[d]
no
no

r
"reset input"
in
d
[d]
no
no

12.4. DIGITAL MODELS
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:

221

clk
"clock"
in
d
[d]
no
no
set
"asynchronous set"
in
d
[d]
no
yes

reset
"asynchronous reset"
in
d
[d]
no
yes

out
"data output"
out
d
[d]
no
yes

Nout
"inverted data output"
out
d
[d]
no
yes

clk_delay
"delay from clk"
real
1.0e-9
[1.0e-12 -]
no
yes

set_delay
"delay from set"
real
1.0e-9
[1.0e-12 -]
no
yes

reset_delay
"delay from reset"
real
1.0e-9
[1.0e-12 -]
no
yes

ic
"output initial state"
int
0
[0 2]
no
yes

sr_load
"set/reset loads (F)"

clk_load
"clk load value (F)"

222

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

real
1.0e-12
no
yes

real
1.0e-12
no
yes

set_load
"set load value (F)"
real
1.0e-12
no
yes

reset_load
"reset load (F)"
real
1.0e-12
no
yes

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

Description: The digital sr-type flip flop is a one-bit, edge-triggered storage element that will
store data whenever the clk input line transitions from low to high (ZERO to ONE). The
value stored (i.e., the out value) will depend on the s and r input pin values, and will be:
out=ONE
out=ZERO
out=previous value
out=UNKNOWN

if
if
if
if

s=ONE and r=ZERO;
s=ZERO and r=ONE;
s=ZERO and r=ZERO;
s=ONE and r=ONE;

In addition, asynchronous set and reset signals exist, and each of the three methods of changing
the stored output of the d_srff have separate load values and delays associated with them. You
may also specify separate rise and fall delay values that are added to those specified for the
input lines; these allow for more faithful reproduction of the output characteristics of different
IC fabrication technologies.
Note that any UNKNOWN inputs other than s and r immediately cause the output to go UNKNOWN.
Example SPICE Usage:
a8 2 12 4 5 6 3 14 flop7
.model flop7 d_srff(clk_delay = 13.0e-9 set_delay = 25.0e-9
+
reset_delay = 27.0e-9 ic = 2 rise_delay = 10.0e-9
+
fall_delay = 3e-9)

12.4. DIGITAL MODELS

12.4.16

223

D Latch

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:

cm_d_dlatch
d_dlatch
"digital d-type latch"
data
"input data"
in
d
[d]
no
no

enable
"enable input"
in
d
[d]
no
no

set
"set"
in
d
[d]
no
yes

reset
"reset"
in
d
[d]
no
yes

out
"data output"
out
d
[d]
no
no

Nout
"inverter data output"
out
d
[d]
no
no

data_delay
"delay from data"
real
1.0e-9
[1.0e-12 -]
no
yes
enable_delay
"delay from enable"
real
1.0e-9
[1.0e-12 -]
no

set_delay
"delay from SET"
real
1.0e-9
[1.0e-12 -]
no

224

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

yes

yes

reset_delay
"delay from RESET"
real
1.0e-9
[1.0e-12 -]
no
yes

ic
"output initial state"
boolean
0
no
yes

data_load
"data load (F)"
real
1.0e-12
no
yes

enable_load
"enable load value (F)"
real
1.0e-12
no
yes

set_load
"set load value (F)"
real
1.0e-12
no
yes

reset_load
"reset load (F)"
real
1.0e-12
no
yes

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

Description: The digital d-type latch is a one-bit, level-sensitive storage element that will output the value on the data line whenever the enable input line is high (ONE). The value on
the data line is stored (i.e., held on the out line) whenever the enable line is low (ZERO).
In addition, asynchronous set and reset signals exist, and each of the four methods of
changing the stored output of the d_dlatch (i.e., data changing with enable=ONE, enable
changing to ONE from ZERO with a new value on data, raising set and raising reset) have
separate delays associated with them. You may also specify separate rise and fall delay
values that are added to those specified for the input lines; these allow for more faithful
reproduction of the output characteristics of different IC fabrication technologies.

12.4. DIGITAL MODELS

225

Note that any UNKNOWN inputs other than on the data line when enable=ZERO immediately cause the output to go UNKNOWN.
Example SPICE Usage:
a4 12 4 5 6 3 14 latch1
.model latch1 d_dlatch(data_delay = 13.0e-9 enable_delay = 22.0e-9
+
set_delay = 25.0e-9
+
reset_delay = 27.0e-9 ic = 2
+
rise_delay = 10.0e-9 fall_delay = 3e-9)

12.4.17

Set-Reset Latch

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:

cm_d_srlatch
d_srlatch
"digital sr-type latch"
s
"set"
in
d
[d]
no
no

r
"reset"
in
d
[d]
no
no

enable
"enable"
in
d
[d]
no
no
set
"set"
in
d
[d]
no
yes

reset
"reset"
in
d
[d]
no
yes

out
"data output"
out
d

Nout
"inverted data output"
out
d

226

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Allowed_Types:
Vector: no no
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:

[d]

[d]

no

no

sr_delay
"delay from s or r input change"
real
1.0e-9
[1.0e-12 -]
no
yes
enable_delay
"delay from enable"
real
1.0e-9
[1.0e-12 -]
no
yes

set_delay
"delay from SET"
real
1.0e-9
[1.0e-12 -]
no
yes

reset_delay
"delay from RESET"
real
1.0e-9
[1.0e-12 -]
no
yes

ic
"output initial state"
boolean
0
no
yes

sr_load
enable_load
"s & r input loads (F)" "enable load value (F)"
real
real
1.0e-12
1.0e-12
no
no
yes
yes
set_load
"set load value (F)"
real
1.0e-12
no
-

reset_load
"reset load (F)"
real
1.0e-12
no
-

12.4. DIGITAL MODELS

227

Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

yes

yes

rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
no
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
no
yes

Description: The digital sr-type latch is a one-bit, level-sensitive storage element that will
output the value dictated by the state of the s and r pins whenever the enable input line
is high (ONE). This value is stored (i.e., held on the out line) whenever the enable line is
low (ZERO). The particular value chosen is as shown below:

s=ZERO, r=ZERO =>
s=ZERO, r=ONE
s=ONE, r=ZERO
s=ONE, r=ONE

out=current value (i.e., not change in output)
=> out=ZERO
=> out=ONE
=> out=UNKNOWN

Asynchronous set and reset signals exist, and each of the four methods of changing the stored output of the d srlatch (i.e., s/r combination changing with enable=ONE, enable changing
to ONE from ZERO with an output-changing combination of s and r, raising set and raising
reset) have separate delays associated with them. You may also specify separate rise and fall
delay values that are added to those specified for the input lines; these allow for more faithful
reproduction of the output characteristics of different IC fabrication technologies.
Note that any UNKNOWN inputs other than on the s and r lines when enable=ZERO immediately cause the output to go UNKNOWN.
Example SPICE Usage:
a4 12 4 5 6 3 14 16 latch2
.model latch2 d_srlatch(sr_delay = 13.0e-9 enable_delay = 22.0e-9
+
set_delay = 25.0e-9
+
reset_delay = 27.0e-9 ic = 2
+
rise_delay = 10.0e-9 fall_delay = 3e-9)

12.4.18

State Machine

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:

cm_d_state
d_state
"digital state machine"
in
"input"

clk
"clock"

228

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:

in
d
[d]
yes
[1 -]
yes

in
d
[d]
no
no

reset
"reset"
in
d
[d]
no
yes

out
"output"
out
d
[d]
yes
[1 -]
no

clk_delay
reset_delay
"delay from CLK"
"delay from RESET"
real
real
1.0e-9
1.0e-9
[1.0e-12 -]
[1.0e-12 -]
no
no
yes
yes
Parameter_Name:
state_file
"state transition specification file name"
string
"state.txt"
no
no
reset_state
"default state on RESET & at DC"
int
0
no
no
input_load
"input loading capacitance (F)"
real
1.0e-12
no

12.4. DIGITAL MODELS
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

229
yes
clk_load
"clock loading capacitance (F)"
real
1.0e-12
no
yes
reset_load
"reset loading capacitance (F)"
real
1.0e-12
no
yes

Description: The digital state machine provides for straightforward descriptions of clocked
combinatorial logic blocks with a variable number of inputs and outputs and with an
unlimited number of possible states. The model can be configured to behave as virtually
any type of counter or clocked combinatorial logic block and can be used to replace very
large digital circuit schematics with an identically functional but faster representation.
The d state model is configured through the use of a state definition file (state.in) that
resides in a directory of your choosing. The file defines all states to be understood by the
model, plus input bit combinations that trigger changes in state. An example state.in file
is shown below:
----------- begin file ------------* This is an example state.in file. This file
* defines a simple 2-bit counter with one input. The
* value of this input determines whether the counter counts
* up (in = 1) or down (in = 0).
0 0s 0s 0 -> 3
1 -> 1
1 0s 1z 0 -> 0
1 -> 2
2 1z 0s 0 -> 1
1 -> 3
3 1z 1z 0 -> 2
3 1z 1z 1 -> 0
------------------ end file --------------Several attributes of the above file structure should be noted. First, all lines in the file must be
one of four types. These are:

230

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

1. A comment, beginning with a ‘*’ in the first column.
2. A header line, which is a complete description of the current state, the outputs corresponding to that state, an input value, and the state that the model will assume should that
input be encountered. The first line of a state definition must always be a header line.
3. A continuation line, which is a partial description of a state, consisting of an input value
and the state that the model will assume should that input be encountered. Note that
continuation lines may only be used after the initial header line definition for a state.
4. A line containing nothing but white-spaces (space, form-feed, newline, carriage return,
tab, vertical tab).
A line that is not one of the above will cause a file-loading error. Note that in the example
shown, whitespace (any combination of blanks, tabs, commas) is used to separate values, and
that the character -> is used to underline the state transition implied by the input preceding it.
This particular character is not critical in of itself, and can be replaced with any other character
or non-broken combination of characters that you prefer (e.g. ==>, >>, ‘:’, resolves_to, etc.)
The order of the output and input bits in the file is important; the first column is always interpreted to refer to the ‘zeroth’ bit of input and output. Thus, in the file above, the output from
state 1 sets out[0] to 0s, and out[1] to 1z.
The state numbers need not be in any particular order, but a state definition (which consists of
the sum total of all lines that define the state, its outputs, and all methods by which a state can
be exited) must be made on contiguous line numbers; a state definition cannot be broken into
sub-blocks and distributed randomly throughout the file. On the other hand, the state definition
can be broken up by as many comment lines as you desire.
Header files may be used throughout the state.in file, and continuation lines can be discarded
completely if you so choose: continuation lines are primarily provided as a convenience.
Example SPICE Usage:
a4 [2 3 4 5] 1 12 [22 23 24 25 26 27 28 29] state1
.model state1 d_state(clk_delay = 13.0e-9 reset_delay = 27.0e-9
+
state_file = "newstate.txt" reset_state = 2)
Note: The file named by the parameter filename in state_file="filename" is sought after
according to a search list described in12.1.3.

12.4.19

Frequency Divider

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:

cm_d_fdiv
d_fdiv
"digital frequency divider"
freq_in
"frequency input"
in

freq_out
"frequency output"
out

12.4. DIGITAL MODELS
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

231
d
[d]
no
no

d
[d]
no
no

div_factor
"divide factor"
int
2
[1 -]
no
yes

high_cycles
"# of cycles for high out"
int
1
[1 div_factor-1]
no
yes

i_count
"divider initial count value"
int
0
no
yes
rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
yes
in
yes

fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
yes
in
yes

freq_in_load
"freq_in load value (F)"
real
1.0e-12
no
yes

Description: The digital frequency divider is a programmable step-down divider that accepts
an arbitrary divisor (div_factor), a duty-cycle term (high_cycles), and an initial count
value (i_count). The generated output is synchronized to the rising edges of the input
signal. Rise delay and fall delay on the outputs may also be specified independently.
Example SPICE Usage:

232

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
a4 3 7 divider
.model divider d_fdiv(div_factor = 5 high_cycles = 3
+
i_count = 4 rise_delay = 23e-9
+
fall_delay = 9e-9)

12.4.20

RAM

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:

cm_d_ram
d_ram
"digital random-access memory"
data_in
"data input line(s)"
in
d
[d]
yes
[1 -]
no

data_out
"data output line(s)"
out
d
[d]
yes
data_in
no

address
write_en
"address input line(s)" "write enable line"
in
in
d
d
[d]
[d]
yes
no
[1 -]
no
no
select
"chip select line(s)"
in
d
[d]
yes
[1 16]
no
select_value
"decimal active value for select line comparison"
int
1
[0 32767]
no
yes

12.4. DIGITAL MODELS
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

233
ic
"initial bit state @ dc"
int
2
[0 2]
no
yes
read_delay
"read delay from address/select/write.en active"
real
100.0e-9
[1.0e-12 -]
no
yes
data_load
address_load
"data_in load value (F)" "addr. load value (F)"
real
real
1.0e-12
1.0e-12
no
no
yes
yes
select_load
"select load value (F)"
real
1.0e-12
no
yes
enable_load
"enable line load value (F)"
real
1.0e-12
no
yes

Description: The digital RAM is an M-wide, N-deep random access memory element with
programmable select lines, tristated data out lines, and a single write/~read line. The
width of the RAM words (M) is set through the use of the word width parameter. The

234

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
depth of the RAM (N) is set by the number of address lines input to the device. The value
of N is related to the number of address input lines (P) by the following equation:
2P = N
There is no reset line into the device. However, an initial value for all bits may be specified
by setting the ic parameter to either 0 or 1. In reading a word from the ram, the read delay
value is invoked, and output will not appear until that delay has been satisfied. Separate
rise and fall delays are not supported for this device.
Note that UNKNOWN inputs on the address lines are not allowed during a write. In the
event that an address line does indeed go unknown during a write, the entire contents
of the ram will be set to unknown. This is in contrast to the data in lines being set to
unknown during a write; in that case, only the selected word will be corrupted, and this is
corrected once the data lines settle back to a known value. Note that protection is added
to the write en line such that extended UNKNOWN values on that line are interpreted as
ZERO values. This is the equivalent of a read operation and will not corrupt the contents
of the RAM. A similar mechanism exists for the select lines. If they are unknown, then it
is assumed that the chip is not selected.
Detailed timing-checking routines are not provided in this model, other than for the enable
delay and select delay restrictions on read operations. You are advised, therefore, to
carefully check the timing into and out of the RAM for correct read and write cycle
times, setup and hold times, etc. for the particular device they are attempting to model.
Example SPICE Usage:
a4 [3 4 5 6] [3 4 5 6] [12 13 14 15 16 17 18 19] 30 [22 23 24] ram2
.model ram2 d_ram(select_value = 2 ic = 2 read_delay = 80e-9)

12.4.21

Digital Source

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:

cm_d_source
d_source
"digital signal source"
out
"output"
out
d
[d]
yes
no
input_file
"digital input vector filename"
string
"source.txt"
-

12.4. DIGITAL MODELS
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

235
no
no
input_load
"input loading capacitance (F)"
real
1.0e-12
no
no

Description: The digital source provides for straightforward descriptions of digital signal vectors in a tabular format. The model reads input from the input file and, at the times
specified in the file, generates the inputs along with the strengths listed. The format of
the input file is as shown below. Note that comment lines are delineated through the use
of a single ‘*’ character in the first column of a line. This is similar to the way the SPICE
program handles comments.
* T
* i
* m
* e
*
0.0000
1.234e-9
1.376e-9
2.5e-7
2.5006e-7
5.0e-7

c
l
o
c
k
Uu
0s
0s
1s
1s
0s

n
o
d
e
a
Uu
1s
0s
0s
1s
1s

n
o
d
e
b
Us
1s
1s
1s
1s
1s

n . . .
o . . .
d . . .
e . . .
c . . .
Uu . . .
0z . . .
0z . . .
0z . . .
0z . . .
0z . . .

Note that in the example shown, whitespace (any combination of blanks, tabs, commas) is used
to separate the time and state/strength tokens. The order of the input columns is important; the
first column is always interpreted to mean ‘time’. The second through the N’th columns map
to the out[0] through out[N-2] output nodes. A non-commented line that does not contain
enough tokens to completely define all outputs for the digital source will cause an error. Also,
time values must increase monotonically or an error will result in reading the source file.
Errors will also occur if a line exists in source.txt that is neither a comment nor vector line.
The only exception to this is in the case of a line that is completely blank; this is treated as
a comment (note that such lines often occur at the end of text within a file; ignoring these in
particular prevents nuisance errors on the part of the simulator).
Example SPICE Usage:
a3 [2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17] input_vector
.model input_vector d_source(input_file = "source_simple.text")
Note: The file named by the parameter filename in input_file="filename" is sought after
according to a search list described in12.1.3.

236

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

12.4.22

LUT

NAME_TABLE:
C_Function_Name: cm_d_lut
Spice_Model_Name: d_lut
Description:
"digital n-input look-up table gate"
PORT_TABLE:
Port_Name:
in out
Description:
"input"
"output"
Direction:
in
out
Default_Type:
d
d
Allowed_Types:
[d]
[d]
Vector:
yes
no
Vector_Bounds:
[1 -]
Null_Allowed:
no
no
PARAMETER_TABLE:
Parameter_Name:
rise_delay
fall_delay
Description:
"rise delay" "fall delay"
Data_Type:
real
real
Default_Value:
1.0e-9
1.0e-9
Limits:
[1.0e-12 -]
[1.0e-12 -]
Vector:
no
no
Vector_Bounds:
Null_Allowed:
yes
yes
PARAMETER_TABLE:
Parameter_Name: input_load
Description:
"input load value (F)"
Data_Type:
real
Default_Value:
1.0e-12
Limits:
Vector:
no
Vector_Bounds:
Null_Allowed:
yes
PARAMETER_TABLE:
Parameter_Name: table_values
Description:
"lookup table values"
Data_Type:
string
Default_Value:
"0"
Limits:
Vector:
no
Vector_Bounds:
Null_Allowed:
no

12.4. DIGITAL MODELS

237

Description: The lookup table provides a way to map any arbitrary n-input, 1-output combinational logic block to XSPICE. The inputs are mapped to the output using a string of length
2^n. The string may contain values "0", "1" or "X", corresponding to an output of low, high,
or unknown, respectively. The outputs are only mapped for inputs which are valid logic levels. Any unknown bit in the input vector will always produce an unknown output. The first
character of the string table_values corresponds to all inputs value zero, and the last (2^n)
character corresponds to all inputs value one, with the first signal in the input vector being the
least significant bit. For example, a 2-input lookup table representing the function (A * B)
(that is, A AND B), with input vector [A B] can be constructed with a table_values string of
"0001"; function (~A * B) with input vector [A B] can be constructed with a table_values
string of "0010". The delays associated with an output rise and those associated with an output
fall may be specified independently. The model also posts an input load value (in farads) based
on the parameter input_load. The output of this model does not respond to the total loading
it sees on the output; it will always drive the output strongly with the specified delays.
Example SPICE Usage:
* LUT encoding 3-bit parity function
a4 [1 2 3] 5 lut_pty3_1
.model lut_pty3_1 d_lut(table_values = "01101001"
+ input_load 2.0e-12)

12.4.23

General LUT

NAME_TABLE:
C_Function_Name:
Spice_Model_Name:
Description:
PORT_TABLE:
Port_Name:
Description:
Direction:
Default_Type:
Allowed_Types:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:

cm_d_genlut
d_genlut
"digital n-input x m-output look-up table gate"
in
"input"
in
d
[d]
yes
no
rise_delay
"rise delay"
real
1.0e-9
[1.0e-12 -]
yes
yes

out
"output"
out
d
[d]
yes
no
fall_delay
"fall delay"
real
1.0e-9
[1.0e-12 -]
yes
yes

input_load
"input load value (F)"

input_delay
"input delay"

238

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:
PARAMETER_TABLE:
Parameter_Name:
Description:
Data_Type:
Default_Value:
Limits:
Vector:
Vector_Bounds:
Null_Allowed:

real
1.0e-12
yes
yes

real
0.0
yes
yes

table_values
"lookup table values"
string
"0"
no
no

Description: The lookup table provides a way to map any arbitrary n-input, m-output combinational logic block to XSPICE. The inputs are mapped to the output using a string of length m
* (2^n). The string may contain values "0", "1", "X", or "Z", corresponding to an output of low,
high, unknown, or high-impedance, respectively. The outputs are only mapped for inputs which
are valid logic levels. Any unknown bit in the input vector will always produce an unknown
output. The character string is in groups of (2^n) characters, one group corresponding to each
output pin, in order. The first character of a group in the string table_values corresponds to
all inputs value zero, and the last (2^n) character in the group corresponds to all inputs value
one, with the first signal in the input vector being the least significant bit. For example, a 2-input
lookup table representing the function (A * B) (that is, A AND B), with input vector [A B] can
be constructed with a table_values string of "0001"; function (~A * B) with input vector
[A B] can be constructed with a "table_values" string of "0010". The delays associated with
each output pin’s rise and those associated with each output pin’s fall may be specified independently. The model also posts independent input load values per input pin (in farads) based on
the parameter input_load. The parameter input_delay provides a way to specify additional
delay between each input pin and the output. This delay is added to the rise- or fall-time of the
output. The output of this model does not respond to the total loading it sees on the output; it
will always drive the output strongly with the specified delays.
Example SPICE Usage:
* LUT encoding 3-bit parity function
a4 [1 2 3] [5] lut_pty3_1
.model lut_pty3_1 d_genlut(table_values = "01101001"
+ input_load [2.0e-12])
* LUT encoding a tristate inverter function (en in out)
a2 [1 2] [3] lut_triinv_1
.model lut_triinv_1 d_genlut(table_values = "Z1Z0")
* LUT encoding a half-adder function (A B Carry Sum)
a8 [1 2] [3 4] lut_halfadd_1
.model lut_halfadd_1 d_genlut(table_values = "00010110"
+ rise_delay [ 1.5e-9 1.0e-9 ] fall_delay [ 1.5e-9 1.0e-9 ])

12.5. PREDEFINED NODE TYPES FOR EVENT DRIVEN SIMULATION

12.5

239

Predefined Node Types for event driven simulation

The following prewritten node types are included with the XSPICE simulator. These should
provide you not only with valuable event-driven modeling capabilities, but also with examples
to use for guidance in creating new UDN (user defined node) types. You may access these node
data by the plot (17.5.45) or eprint (17.5.24) commands.

12.5.1

Digital Node Type

The ‘digital’ node type is directly built into the simulator. 12 digital node values are available.
They are described by a two character string (the state/strength token). The first character (0,
1, or U) gives the state of the node (logic zero, logic one, or unknown logic state). The second
character (s, r, z, u) gives the "strength" of the logic state (strong, resistive, hi-impedance, or
undetermined). So these are the values we have: 0s, 1s, Us, 0r, 1r, Ur, 0z, 1z, Uz, 0u, 1u, Uu.

12.5.2

Real Node Type

The ‘real’ node type provides for event-driven simulation with double-precision floating point
data. This type is useful for evaluating sampled-data filters and systems. The type implements
all optional functions for User-Defined Nodes, including inversion and node resolution. For
inversion, the sign of the value is reversed. For node resolution, the resultant value at a node is
the sum of all values output to that node. The node is implemented as a user defined node in
ngspice/src/xspice/icm/xtraevt/real.

12.5.3

Int Node Type

The ‘int’ node type provides for event-driven simulation with integer data. This type is useful
for evaluating round-off error effects in sampled-data systems. The type implements all optional
functions for User-Defined Nodes, including inversion and node resolution. For inversion, the
sign of the integer value is reversed. For node resolution, the resultant value at a node is the
sum of all values output to that node. The node is implemented as a user defined node in
ngspice/src/xspice/icm/xtraevt/int.

12.5.4

(Digital) Input/Output

The analog code models use the standard (analog) nodes provided by ngspice and thus are using
all the commands for sourcing, storing, printing, and plotting data.
I/O for event nodes (digital, real, int, and UDNs) currently is much more limited to a few tools.
For output you may use the plot (17.5.45) or eprint (17.5.24) commands. For input, you may
create a test bench with existing code models (oscillator (12.3.3), frequency divider (12.4.19),
state machine (12.4.18) etc.). Reading data from a file is offered by d_source (12.4.21). Some
comments and hints have been provided by Sdaau. You may also use the analog input from file,
(filesource 12.2.8) and convert its analog input to the digital type by the adc_bridge (12.3.2).
If you want reading data from a VCD file, you may have a look at ngspice-users forum and
apply a python script provided by Sdaau to translate the VCD data to d_source or filesource
input.

240

CHAPTER 12. MIXED-MODE AND BEHAVIORAL MODELING WITH XSPICE

Chapter 13
Verilog A Device models
13.1

Introduction

The ngspice-adms interface will implement extra HICUM level0 and level2 (HICUM model
web page), MEXTRAM(MEXTRAM model web page), EKV(EKV model web page) and
PSP(NXP MOS model 9 web page) models written in Verilog-A behavior language.

13.2

adms

To compile Verilog-A compact models into ngspice-ready C models the the program admsXml
is required. Details of this software are described in adms home page.

13.3

How to integrate a Verilog-A model into ngspice

13.3.1

How to setup a *.va model for ngspice

The root entry for new Verilog-A models is \src\spicelib\devices\adms. Below the modelname entry the Verilog-A code should reside in folder admsva
(e.g.: ngspice\src\spicelib\devices\adms\ekv\admsva\ekv.va). The file extension is fixed
to .va.
Certain files must modified to create the interface to ngspice - see the guideline README.adms
in the ngspice root.

13.3.2

Adding admsXml to your build environment

To facilitate the installation of adms, a source code package has been assembled for use with ngspice, available as a zip file for download. It is based on adms source code from the subversion
repository downloaded on August 1st, 2010, and has been slightly modified (see ChangeLog).
Under OS Linux (tested with SUSE 11.2, 64 bit) you may expand the zip file and run
./autogen_lin.sh, followed by ’make’ and ’make install’.

241

242

CHAPTER 13. VERILOG A DEVICE MODELS

Under OS CYGWIN (tested with actual CYGWIN on MS Windows 7, 64 bit), please use
./autogen_cyg.sh, followed by ’make’ and ’make install’.
Under OS MINGW, a direct compilation would require the additional installation of perl module
XML-LibXML, which is not as straightforward as it should be. However you may start with a
CYGWIN compile as described above. If you then go to your MSYS window, cd to the adms
top directory and start ./mingw-compile.sh, you will obtain admsXml.exe, copied to MSYS
/bin, and you are ready to go. To facilitate installation under MS Windows, a admsXml.exe
zipped binary is available. Just copy it to MSYS /bin directory and start working on your verilog
models.
A short test of a successful installation is:
$ admsXml -v
$ [usage..] release name="admsXml" version="2.3.0" date="Aug 4 2010"
time="10:24:18"
Compilation of admsXml with MS Visual Studio is not possible, because the source code has
variable declarations not only at the top of a block, but deliberately also in the following lines.
This is ok by the C99 standard, but not supported by MS Visual Studio.

Chapter 14
Mixed-Level Simulation (ngspice with
TCAD)
14.1

Cider

Ngspice implements mixed-level simulation through the merging of its code with CIDER (details see Chapt. 30).
CIDER is a mixed-level circuit and device simulator that provides a direct link between technology parameters and circuit performance. A mixed-level circuit and device simulator can provide greater simulation accuracy than a stand-alone circuit or device simulator by numerically
modeling the critical devices in a circuit. Compact models can be used for noncritical devices.
CIDER couples the latest version of SPICE3 (version 3F.2) [JOHN92] to a internal C-based
device simulator, DSIM. SPICE3 provides circuit analyses, compact models for semiconductor
devices, and an interactive user interface. DSIM provides accurate, one- and two-dimensional
numerical device models based on the solution of Poisson’s equation, and the electron and
hole current-continuity equations. DSIM incorporates many of the same basic physical models
found in the the Stanford two-dimensional device simulator PISCES [PINT85]. Input to CIDER
consists of a SPICE-like description of the circuit and its compact models, and PISCES-like
descriptions of the structures of numerically modeled devices. As a result, CIDER should seem
familiar to designers already accustomed to these two tools. For example, SPICE3F.2 input files
should run without modification, producing identical results.
CIDER is based on the mixed-level circuit and device simulator CODECS [MAYA88] and is a
replacement for this program. The basic algorithms of the two programs are the same. Some of
the differences between CIDER and CODECS are described below. The CIDER input format
has greater flexibility and allows increased access to physical model parameters. New physical
models have been added to allow simulation of state-of-the-art devices. These include transverse field mobility degradation [GATE90] that is important in scaled-down MOSFETs and a
polysilicon model for poly-emitter bipolar transistors. Temperature dependence has been included for most physical models over the range from -50°C to 150°C. The numerical models can
be used to simulate all the basic types of semiconductor devices: resistors, MOS capacitors, diodes, BJTs, JFETs and MOSFETs. BJTs and JFETs can be modeled with or without a substrate
contact. Support has been added for the management of device internal states. Post-processing
of device states can be performed using the NUTMEG user interface of SPICE3. Previously

243

244

CHAPTER 14. MIXED-LEVEL SIMULATION (NGSPICE WITH TCAD)

computed states can be loaded into the program to provide accurate initial guesses for subsequent analyses. Finally, numerous small bugs have been discovered and fixed, and the program
has been ported to a wider variety of computing platforms.
Berkeley tradition calls for the naming of new versions of programs by affixing a (number,
letter, number) triplet to the end of the program name. Under this scheme, CIDER should
instead be named CODECS2A.l. However, tradition has been broken in this case because major
incompatibilities exist between the two programs and because it was observed that the acronym
CODECS is already used in the analog design community to refer to coder-decoder circuits.
Details of the basic semiconductor equations and the physical models used by CIDER are not
provided in this manual. Unfortunately, no other single source exists that describes all of the
relevant background material. Comprehensive reviews of device simulation can be found in
[PINT90] and the book [SELB84]. CODECS and its inversion-layer mobility model are described in [MAYA88] and LGATE90], respectively. PISCES and its models are described in
[PINT85]. Temperature dependencies for the PISCES models used by CIDER are available in
[SOLL90].

14.2

GSS, Genius

For Linux users the cooperation of the TCAD software GSS with ngspice might be of interest,
see http://ngspice.sourceforge.net/gss.html. This project is no longer maintained however, but
has moved into the Genius simulator, still available as open source cogenda genius.

Chapter 15
Analyses and Output Control (batch
mode)
The command lines described in this chapter are specifying analyses and outputs within the
circuit description file. They start with a ‘.’ (dot commands). Specifying analyses and plots
(or tables) in the input file with dot commands is used with batch runs. Batch mode is entered
when either the -b option is given upon starting ngspice
ngspice -b -r rawfile.raw circuitfile.cir
or when the default input source is redirected from a file (see also Chapt. 16.4.1).
ngspice < circuitfile.cir
In batch mode, the analyses specified by the control lines in the input file (e.g. .ac, .tran, etc.)
are immediately executed. If the -r rawfile option is given then all data generated is written to
a ngspice rawfile. The rawfile may later be read by the interactive mode of ngspice using the
load command (see 17.5.38). In this case, the .save line (see 15.6) may be used to record the
value of internal device variables (see Appendix, Chapt. 31).
If a rawfile is not specified, then output plots (in ‘line-printer’ form) and tables can be printed
according to the .print, .plot, and .four control lines, described in Chapt. 15.6.
If ngspice is started in interactive mode (see Chapt. 16.4.2), like
ngspice circuitfile.cir
and no control section (.control ... .endc, see 16.4.3) is provided in the circuit file, the dot
commands are not executed immediately, but are waiting for manually receiving the command
run.

15.1

Simulator Variables (.options)

Various parameters of the simulations available in Ngspice can be altered to control the accuracy, speed, or default values for some devices. These parameters may be changed via the
option command (described in Chapt. 17.5.44) or via the .options line:

245

246

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

General form:
. options opt1 opt2 ... (or opt= optval ...)
Examples:
. options reltol =.005 trtol =8
The options line allows the user to reset program control and user options for specific simulation
purposes. Options specified to Ngspice via the option command (see Chapt. ) are also passed
on as if specified on a .options line. Any combination of the following options may be
included, in any order. ‘x’ (below) represents some positive number.

15.1.1

General Options

ACCT causes accounting and run time statistics to be printed.
NOACCT no printing of statistics, no printing of the Initial Transient Solution.
NOINIT suppresses only printing of the Initial Transient Solution, maybe combined with
ACCT.
LIST causes the summary listing of the input data to be printed.
NOMOD suppresses the printout of the model parameters.
NOPAGE suppresses page ejects.
NODE causes the printing of the node table.
OPTS causes the option values to be printed.
TEMP=x Resets the operating temperature of the circuit. The default value is 27 ◦C (300K).
TEMP can be overridden per device by a temperature specification on any temperature
dependent instance. May also be generally overridden by a .TEMP card (2.11).
TNOM=x resets the nominal temperature at which device parameters are measured. The default value is 27 ◦C (300 deg K). TNOM can be overridden by a specification on any
temperature dependent device model.
WARN=1|0 enables or turns of SOA (Safe Operating Area) voltage warning messages (default:
0).
MAXWARNS=x specifies the maximum number of SOA (Safe Operating Area) warning messages per model (default: 5).
SAVECURRENTS save currents through all terminals of the following devices: M, J, Q, D,
R, C, L, B, F, G, W, S, I (see 2.1.2). Recommended only for small circuits, because
otherwise memory requirements explode and simulation speed suffers. See 15.7 for more
details.

15.1. SIMULATOR VARIABLES (.OPTIONS)

15.1.2

247

DC Solution Options

The following options controls properties pertaining to DC analysis and algorithms. Since
transient analysis is based on DC many of the options affect the latter one.
ABSTOL=x resets the absolute current error tolerance of the program. The default value is 1
pA.
GMIN=x resets the value of GMIN, the minimum conductance allowed by the program. The
default value is 1.0e-12.
ITL1=x resets the dc iteration limit. The default is 100.
ITL2=x resets the dc transfer curve iteration limit. The default is 50.
KEEPOPINFO Retain the operating point information when either an AC, Distortion, or PoleZero analysis is run. This is particularly useful if the circuit is large and you do not want
to run a (redundant) .OP analysis.
PIVREL=x resets the relative ratio between the largest column entry and an acceptable pivot
value. The default value is 1.0e-3. In the numerical pivoting algorithm the allowed minimum pivot value is determined by EPSREL = AMAX1(PIVREL · MAXVAL, PIVTOL) where
MAXVAL is the maximum element in the column where a pivot is sought (partial pivoting).
PIVTOL=x resets the absolute minimum value for a matrix entry to be accepted as a pivot.
The default value is 1.0e-13.
RELTOL=x resets the relative error tolerance of the program. The default value is 0.001
(0.1%).
RSHUNT=x introduces a resistor from each analog node to ground. The value of the resistor
should be high enough to not interfere with circuit operations. The XSPICE option has to
be enabled (see 32.1.5) .
VNTOL=x resets the absolute voltage error tolerance of the program. The default value is 1
µV .
15.1.2.1

Matrix Conditioning info

In most SPICE-based simulators, problems can arise with certain circuit topologies. One of
the most common problems is the absence of a DC path to ground at some node. This may
happen, for example, when two capacitors are connected in series with no other connection at
the common node or when certain code models are cascaded. The result is an ill-conditioned
or nearly singular matrix that prevents the simulation from completing. The XSPICE option
introduces the rshunt option to help eliminate this problem. When used, this option inserts
resistors to ground at all the analog nodes in the circuit. In general, the value of rshunt should
be set to some very high resistance (e.g. 1000 Meg Ohms or greater) so that the operation of the
circuit is essentially unaffected, but the matrix problems are corrected. If you should encounter
a ‘no DC path to ground’ or a ‘matrix is nearly singular’ error message with your circuit, you
should try adding the following .option card to your circuit description deck.

248

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)
.option rshunt = 1.0e12

Usually a value of 1.0e12 is sufficient to correct the matrix problems. However, if you still have
problems, you may wish to try lowering this value to 1.0e10 or 1.0e9.
Another matrix conditioning problem might occur if you try to place an inductor in parallel to
a voltage source. An ac simulation will fail, because it is preceded by an op analysis. Option
noopac (15.1.3) will help if the circuit is linear. If the circuit is non-linear, you will need the
op analysis. Then adding a small resistor (e.g. 1e-4 Ohms) in series to the inductor will help to
obtain convergence.
.option rseries = 1.0e-4
will add a series resistor to each inductor in the circuit. Be careful if you use behavioral inductors (see 3.2.12), because the result may become unpredictable.

15.1.3

AC Solution Options

NOOPAC Do not do an operating point (OP) analysis before the AC analysis. To become
valid, this option requires that the circuit is linear, thus consists only of R, L, and C
devices, independent V, I sources and linear dependent E, G, H, and F sources (without
poly statement, non-behavioral). If a non-linear device is detected, the OP analysis will
be executed automatically. This option is of interest for example in nested LC circuits,
where there is no series resistance for the L device given, which during OP analysis may
result in an ill formed matrix, yields an error message and aborts the simulation.

15.1.4

Transient Analysis Options

AUTOSTOP stops a transient analysis after successfully calculating all measure functions
(15.4) specified with the dot command .meas. Autostop is not available with meas
(17.5.39) used in control mode.
CHGTOL=x resets the charge tolerance of the program. The default value is 1.0e-14.
CONVSTEP=x relative step limit applied to code models.
CONVABSSTEP=x absolute step limit applied to code models.
GMINSTEPS=x [*] sets number of Gmin steps to be attempted. If the value is set to zero, the
gmin stepping algorithm is disabled. In such case the source stepping algorithm becomes
the standard when the standard procedure fails to converge to a solution.
INTERP interpolates output data onto fixed time steps, detemined by TSTEP (15.3.9). Uses
linear interpolation between previous and next time value. Simulation itself is not influenced by this option. May be used in all simulation modes (batch, control or interactive, 16.4). This option may drastically reduce memory requirements in control mode
or file size in batch mode, but be careful not to choose a too large TSTEP value, otherwise your output data may be corrupted by undersampling. See command ’linearize’
(17.5.36) in control or interactive mode to achieve similar outputs by post-processing of
data. See ngspice/examples/xspice/delta-sigma/delta-sigma-1.cir how INTERP will
reduce memory requirements and speeds up plotting.

15.1. SIMULATOR VARIABLES (.OPTIONS)

249

ITL3=x resets the lower transient analysis iteration limit. the default value is 4. (Note: not
implemented in Spice3).
ITL4=x resets the transient analysis time-point iteration limit. the default is 10.
ITL5=x resets the transient analysis total iteration limit. the default is 5000. Set ITL5=0 to
omit this test. (Note: not implemented in Spice3).
ITL6=x [*] synonym for SRCSTEPS.
MAXEVITER=x sets the number of event iterations that are allowed at an analysis point
MAXOPALTER=x specifies the maximum number of analog/event alternations that the simulator can use in solving a hybrid circuit.
MAXORD=x [*] specifies the maximum order for the numerical integration method used by
SPICE. Possible values for the Gear method are from 2 (the default) to 6. Using the value
1 with the trapezoidal method specifies backward Euler integration.
METHOD=name sets the numerical integration method used by SPICE. Possible names are
‘Gear’ or ‘trapezoidal’ (or just ‘trap’). The default is trapezoidal.
NOOPALTER=TRUE|FALSE if set to false alternations between analog/event are enabled.
RAMPTIME=x this options sets the rate of change of independent supplies and code model
inductors and capacitors with initial conditions specified.
SRCSTEPS=x [*] a non-zero value causes SPICE to use a source-stepping method to find the
DC operating point. Its value specifies the number of steps.
TRTOL=x resets the transient error tolerance. The default value is 7. This parameter is an estimate of the factor by which ngspice overestimates the actual truncation error. If XSPICE
is enabled and ’A’ devices included, the value is internally set to 1 for higher precision.
This will cost a factor of two in CPU time during transient analysis.
XMU=x sets a damping factor for trapezoidal integration. The default value is XMU=0.5. A
value < 0.5 may be chosen. Even a small reduction, e.g. to 0.495, may suppress trap
ringing. The reduction has to be set carefully in order not to excessively damp circuits
that are prone to ringing, and lead the simulation (and the user) to believe that the circuit
is stable.

15.1.5

ELEMENT Specific options

BADMOS3 Use the older version of the MOS3 model with the ‘kappa’ discontinuity.
DEFAD=x resets the value for MOS drain diffusion area; the default is 0.0.
DEFAS=x resets the value for MOS source diffusion area; the default is 0.0.
DEFL=x resets the value for MOS channel length; the default is 100.0 µm.
DEFW=x resets the value for MOS channel width; the default is 100.0 µm.

250

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

SCALE=x set the element scaling factor for geometric element parameters whose default unit
is meters. As an example: scale=1u and a MOSFET instance parameter W=10 will result
in a width of 10µm for this device. An area parameter AD=20 will result in 20e-12 m2 .
Following instance parameters are scaled:
• Resistors and Capacitors: W, L
• Diodes: W, L, Area
• JFET, MESFET: W, L, Area
• MOSFET: W, L, AS, AD, PS, PD, SA, SB, SC, SD

15.1.6

Transmission Lines Specific Options

TRYTOCOMPACT Applicable only to the LTRA model (see 6.2.1). When specified, the
simulator tries to condense LTRA transmission line’s past history of input voltages and
currents.

15.1.7

Precedence of option and .options commands

There are various ways to set the above mentioned options in Ngspice. If no option or
.options lines are set by the user, internal default values are given for each of the simulator variables.
You may set options in the init files spinit or .spiceinit via the option command (see Chapt.
17.5.44). The values given here will supersede the default values. If you set options via the
.options line in your input file, their values will supersede the default and init file data. Finally
if you set options inside a .control ... .endc section, these values will supersede any values
of the respective simulator variables given so far.

15.2

Initial Conditions

15.2.1

.NODESET: Specify Initial Node Voltage Guesses

General form:
. NODESET V( NODNUM )= VAL V( NODNUM )= VAL ...
. NODESET ALL=VAL
Examples:
. NODESET V (12)=4.5 V (4)=2.23
. NODESET ALL =1.5

15.2. INITIAL CONDITIONS

251

The .nodeset line helps the program find the dc or initial transient solution by making a
preliminary pass with the specified nodes held to the given voltages. The restriction is then
released and the iteration continues to the true solution. The .nodeset line may be necessary
for convergence on bistable or a-stable circuits. .nodeset all=val allows to set all starting
node voltages (except for the ground node) in a single line. In general, the .nodeset line
should not be necessary.

15.2.2

.IC: Set Initial Conditions

General form:
.ic v( nodnum )= val v( nodnum )= val ...
Examples:
.ic v (11)=5 v(4)= -5 v (2)=2.2
The .ic line is for setting transient initial conditions. It has two different interpretations, depending on whether the uic parameter is specified on the .tran control line. Also, one should
not confuse this line with the .nodeset line. The .nodeset line is only to help dc convergence, and does not affect the final bias solution (except for multi-stable circuits). The two
interpretations of this line are as follows:

1. When the uic parameter is specified on the .tran line, then the node voltages specified
on the .ic control line are used to compute the capacitor, diode, BJT, JFET, and MOSFET
initial conditions. This is equivalent to specifying the ic=... parameter on each device
line, but is much more convenient. The ic=... parameter can still be specified and takes
precedence over the .ic values. Since no dc bias (initial transient) solution is computed
before the transient analysis, one should take care to specify all dc source voltages on the
.ic control line if they are to be used to compute device initial conditions.

2. When the uic parameter is not specified on the .tran control line, the dc bias (initial
transient) solution is computed before the transient analysis. In this case, the node voltages specified on the .ic control lines are forced to the desired initial values during the
bias solution. During transient analysis, the constraint on these node voltages is removed.
This is the preferred method since it allows ngspice to compute a consistent dc solution.

252

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

15.3

Analyses

15.3.1

.AC: Small-Signal AC Analysis

General form:
.ac dec nd fstart fstop
.ac oct no fstart fstop
.ac lin np fstart fstop
Examples:
.ac dec 10 1 10K
.ac dec 10 1K 100 MEG
.ac lin 100 1 100 HZ
dec stands for decade variation, and nd is the number of points per decade. oct stands for
octave variation, and no is the number of points per octave. lin stands for linear variation, and
np is the number of points. fstart is the starting frequency, and fstop is the final frequency.
If this line is included in the input file, ngspice performs an AC analysis of the circuit over the
specified frequency range. Note that in order for this analysis to be meaningful, at least one
independent source must have been specified with an ac value. Typically it does not make much
sense to specify more than one ac source. If you do, the result will be a superposition of all
sources, thus difficult to interpret.
Example:
Basic RC circuit
r 1 2 1.0
c 2 0 1.0
vin 1 0 dc 0 ac 1 $ <--- the ac source
. options noacct
.ac dec 10 .01 10
.plot ac vdb (2) xlog
.end
In this ac (or ’small signal’) analysis all non-linear devices are linearized around their actual dc
operating point. All Ls and Cs get their imaginary value, depending on the actual frequency
step. Each output vector will be calculated relative to the input voltage (current) given by the ac
value (Vin equals to 1 in the example above). The resulting node voltages (and branch currents)
are complex vectors. Therefore you have to be careful using the plot command. Especially
you may use the variants of vxx(node) described in Chapt. 15.6.2 like vdb(2) (see example
above).

15.3. ANALYSES

15.3.2

253

.DC: DC Transfer Function

General form:
.dc srcnam vstart vstop vincr [src2 start2 stop2 incr2]
Examples:

.dc
.dc
.dc
.dc
.dc

VIN 0.25
VDS 0 10
VCE 0 10
RLoad 1k
TEMP -15

5.0 0.25
.5 VGS 0 5 1
.25 IB 0 10u 1u
2k 100
75 5

The .dc line defines the dc transfer curve source and sweep limits (again with capacitors open
and inductors shorted). srcnam is the name of an independent voltage or current source, a
resistor or the circuit temperature. vstart, vstop, and vincr are the starting, final, and incrementing values respectively. The first example causes the value of the voltage source VIN
to be swept from 0.25 Volts to 5.0 Volts in increments of 0.25 Volts. A second source (src2)
may optionally be specified with associated sweep parameters. In this case, the first source is
swept over its range for each value of the second source. This option can be useful for obtaining
semiconductor device output characteristics. See the example circuit description on transistor
characteristics (21.3).

15.3.3

.DISTO: Distortion Analysis

General form:
.disto dec nd fstart fstop 
.disto oct no fstart fstop 
.disto lin np fstart fstop 
Examples:
.disto dec 10 1kHz 100 MEG
.disto dec 10 1kHz 100 MEG 0.9
The .disto line does a small-signal distortion analysis of the circuit. A multi-dimensional Volterra series analysis is done using multi-dimensional Taylor series to represent the nonlinearities
at the operating point. Terms of up to third order are used in the series expansions.
If the optional parameter f2overf1 is not specified, .disto does a harmonic analysis - i.e.,
it analyses distortion in the circuit using only a single input frequency F1 , which is swept as
specified by arguments of the .disto command exactly as in the .ac command. Inputs at this
frequency may be present at more than one input source, and their magnitudes and phases are
specified by the arguments of the distof1 keyword in the input file lines for the input sources

254

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

(see the description for independent sources). (The arguments of the distof2 keyword are not
relevant in this case).
The analysis produces information about the AC values of all node voltages and branch currents
at the harmonic frequencies 2F1 and , vs. the input frequency F1 as it is swept. (A value of 1
(as a complex distortion output) signifies cos(2π(2F1 )t) at 2F1 and cos(2π(3F1 )t) at 3F1 , using
the convention that 1 at the input fundamental frequency is equivalent to cos(2πF1t).) The
distortion component desired (2F1 or 3F1 ) can be selected using commands in ngnutmeg, and
then printed or plotted. (Normally, one is interested primarily in the magnitude of the harmonic
components, so the magnitude of the AC distortion value is looked at). It should be noted that
these are the AC values of the actual harmonic components, and are not equal to HD2 and HD3.
To obtain HD2 and HD3, one must divide by the corresponding AC values at F1 , obtained from
an .ac line. This division can be done using ngnutmeg commands.
If the optional f2overf1 parameter is specified, it should be a real number between (and not
equal to) 0.0 and 1.0; in this case, .disto does a spectral analysis. It considers the circuit with
sinusoidal inputs at two different frequencies F1 and F2 . F1 is swept according to the .disto
control line options exactly as in the .ac control line. F2 is kept fixed at a single frequency
as F1 sweeps - the value at which it is kept fixed is equal to f2overf1 times fstart. Each
independent source in the circuit may potentially have two (superimposed) sinusoidal inputs
for distortion, at the frequencies F1 and F2 . The magnitude and phase of the F1 component are
specified by the arguments of the distof1 keyword in the source’s input line (see the description of independent sources); the magnitude and phase of the F2 component are specified by the
arguments of the distof2 keyword. The analysis produces plots of all node voltages/branch
currents at the intermodulation product frequencies F1 + F2 , F1 − F2 , and (2F1 ) − F2 , vs the
swept frequency F1 . The IM product of interest may be selected using the setplot command,
and displayed with the print and plot commands. It is to be noted as in the harmonic analysis
case, the results are the actual AC voltages and currents at the intermodulation frequencies, and
need to be normalized with respect to .ac values to obtain the IM parameters.
If the distof1 or distof2 keywords are missing from the description of an independent
source, then that source is assumed to have no input at the corresponding frequency. The default
values of the magnitude and phase are 1.0 and 0.0 respectively. The phase should be specified
in degrees.
It should be carefully noted that the number f2overf1 should ideally be an irrational number,
and that since this is not possible in practice, efforts should be made to keep the denominator
in its fractional representation as large as possible, certainly above 3, for accurate results (i.e.,
if f2overf1 is represented as a fraction A/B, where A and B are integers with no common
factors, B should be as large as possible; note that A < B because f2overf1 is constrained
to be < 1). To illustrate why, consider the cases where f2overf1 is 49/100 and 1/2. In a
spectral analysis, the outputs produced are at F1 + F2 , F1 − F2 and 2F1 − F2 . In the latter case,
F1 − F2 = F2 , so the result at the F1 − F2 component is erroneous because there is the strong
fundamental F2 component at the same frequency. Also, F1 + F2 = 2F1 − F2 in the latter case,
and each result is erroneous individually. This problem is not there in the case where f2overf1
= 49/100, because F1 − F2 = 51/100 F1 <> 49/100 F1 = F2 . In this case, there are two very
closely spaced frequency components at F2 and F1 − F2 . One of the advantages of the Volterra
series technique is that it computes distortions at mix frequencies expressed symbolically (i.e.
nF1 + mF2 ), therefore one is able to obtain the strengths of distortion components accurately
even if the separation between them is very small, as opposed to transient analysis for example.
The disadvantage is of course that if two of the mix frequencies coincide, the results are not

15.3. ANALYSES

255

merged together and presented (though this could presumably be done as a postprocessing step).
Currently, the interested user should keep track of the mix frequencies himself or herself and
add the distortions at coinciding mix frequencies together should it be necessary.
Only a subset of the ngspice nonlinear device models supports distortion analysis. These are
• Diodes (DIO),
• BJT,
• JFET (level 1),
• MOSFETs (levels 1, 2, 3, 9, and BSIM1),
• MESFET (level 1).

15.3.4

.NOISE: Noise Analysis

General form:
.noise v( output <,ref >) src ( dec | lin | oct ) pts fstart fstop
+ 
Examples:
.noise v(5) VIN dec 10 1kHz 100 MEG
.noise v(5 ,3) V1 oct 8 1.0 1.0 e6 1
The .noise line does a noise analysis of the circuit. output is the node at which the total
output noise is desired; if ref is specified, then the noise voltage v(output) - v(ref) is
calculated. By default, ref is assumed to be ground. src is the name of an independent source
to which input noise is referred. pts, fstart and fstop are .ac type parameters that specify
the frequency range over which plots are desired. pts_per_summary is an optional integer; if
specified, the noise contributions of each noise generator is produced every pts_per_summary
frequency points. The .noise control line produces two plots:
√
Hz

1. one for the Noise Spectral Density (in V/

√
Hz

or A/

) curves and

2. one for the total Integrated Noise (in V or A) over the specified frequency range.

15.3.5

.OP: Operating Point Analysis

General form:
.op
The inclusion of this line in an input file directs ngspice to determine the dc operating point of
the circuit with inductors shorted and capacitors opened.

256

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

Note: a DC analysis is automatically performed prior to a transient analysis to determine the
transient initial conditions, and prior to an AC small-signal, Noise, and Pole-Zero analysis
to determine the linearized, small-signal models for nonlinear devices (see the KEEPOPINFO
variable 15.1.2).

15.3.6

.PZ: Pole-Zero Analysis

General form:
.pz
.pz
.pz
.pz
.pz
.pz

node1
node1
node1
node1
node1
node1

node2
node2
node2
node2
node2
node2

node3
node3
node3
node3
NODE3
node3

node4
node4
node4
node4
node4
node4

cur
cur
cur
vol
vol
vol

pol
zer
pz
pol
zer
pz

Examples:
.pz 1 0 3 0 cur pol
.pz 2 3 5 0 vol zer
.pz 4 1 4 1 cur pz

cur stands for a transfer function of the type (output voltage)/(input current) while vol stands
for a transfer function of the type (output voltage)/(input voltage). pol stands for pole analysis
only, zer for zero analysis only and pz for both. This feature is provided mainly because
if there is a nonconvergence in finding poles or zeros, then, at least the other can be found.
Finally, node1 and node2 are the two input nodes and node3 and node4 are the two output
nodes. Thus, there is complete freedom regarding the output and input ports and the type of
transfer function.
In interactive mode, the command syntax is the same except that the first field is pz instead of
.pz. To print the results, one should use the command print all.

15.3. ANALYSES

15.3.7

257

.SENS: DC or Small-Signal AC Sensitivity Analysis

General form:
.SENS
.SENS
.SENS
.SENS

OUTVAR
OUTVAR AC DEC ND FSTART FSTOP
OUTVAR AC OCT NO FSTART FSTOP
OUTVAR AC LIN NP FSTART FSTOP

Examples:
.SENS V(1, OUT)
.SENS V(OUT) AC DEC 10 100 100k
.SENS I( VTEST )
The sensitivity of OUTVAR to all non-zero device parameters is calculated when the SENS
analysis is specified. OUTVAR is a circuit variable (node voltage or voltage-source branch
current). The first form calculates sensitivity of the DC operating-point value of OUTVAR.
The second form calculates sensitivity of the AC values of OUTVAR. The parameters listed
for AC sensitivity are the same as in an AC analysis (see .AC above). The output values are in
dimensions of change in output per unit change of input (as opposed to percent change in output
or per percent change of input).

15.3.8

.TF: Transfer Function Analysis

General form:
.tf outvar insrc
Examples:
.tf v(5, 3) VIN
.tf i( VLOAD ) VIN
The .tf line defines the small-signal output and input for the dc small-signal analysis. outvar
is the small signal output variable and insrc is the small-signal input source. If this line is
included, ngspice computes the dc small-signal value of the transfer function (output/input),
input resistance, and output resistance. For the first example, ngspice would compute the ratio
of V(5, 3) to VIN, the small-signal input resistance at VIN, and the small signal output resistance
measured across nodes 5 and 3.

258

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

15.3.9

.TRAN: Transient Analysis

General form:
.tran tstep tstop > 
Examples:
.tran 1ns 100 ns
.tran 1ns 1000 ns 500 ns
.tran 10 ns 1us
tstep is the printing or plotting increment for line-printer output. For use with the postprocessor, tstep is the suggested computing increment. tstop is the final time, and tstart
is the initial time. If tstart is omitted, it is assumed to be zero. The transient analysis always
begins at time zero. In the interval , the circuit is analyzed (to reach a steady
state), but no outputs are stored. In the interval , the circuit is analyzed and
outputs are stored. tmax is the maximum stepsize that ngspice uses; for default, the program
chooses either tstep or (tstop-tstart)/50.0, whichever is smaller. tmax is useful when one
wishes to guarantee a computing interval that is smaller than the printer increment, tstep.
An initial transient operating point at time zero is calculated according to the following procedure: all independent voltages and currents are applied with their time zero values, all capacitances are opened, inductances are shorted, the non linear device equations are solved iteratively.
uic (use initial conditions) is an optional keyword that indicates that the user does not want
ngspice to solve for the quiescent operating point before beginning the transient analysis. If this
keyword is specified, ngspice uses the values specified using IC=... on the various elements as
the initial transient condition and proceeds with the analysis. If the .ic control line has been
specified (see 15.2.2), then the node voltages on the .ic line are used to compute the initial
conditions for the devices. IC=... will take precedence over the values given in the .ic control
line. If neither IC=... nor the .ic control line is given for a specific node, node voltage zero is
assumed.
Look at the description on the .ic control line (15.2.2) for its interpretation when uic is not
specified.

15.3.10

Transient noise analysis (at low frequency)

In contrast to the analysis types described above the transient noise simulation (noise current or
voltage versus time) is not implemented as a dot command, but is integrated with the independent voltage source vsrc (isrc not yet available) (see 4.1.7) and used in combination with the
.tran transient analysis (15.3.9).
Transient noise analysis deals with noise currents or voltages added to your circuits as a time
dependent signal of randomly generated voltage excursion on top of a fixed dc voltage. The
sequence of voltage values has random amplitude, but equidistant time intervals, selectable by
the user (parameter NT). The resulting voltage waveform is differentiable and thus does not
require any modifications of the matrix solving algorithms.

15.3. ANALYSES

259

White noise is generated by the ngspice random number generator, applying the Box-Muller
transform. Values are generated on the fly, each time when a breakpoint is hit.
The 1/f noise is generated with an algorithm provided by N. J. Kasdin (‘Discrete simulation of
colored noise and stochastic processes and 1/ f a power law noise generation’, Proceedings of
the IEEE, Volume 83, Issue 5, May 1995 Page(s):802–827). The noise sequence (one for each
voltage/current source with 1/f selected) is generated upon start up of the simulator and stored
for later use. The number of points is determined by the total simulation time divided by NT,
rounded up the the nearest power of 2. Each time a breakpoint (n ? NT , relevant to the noise
signal) is hit, the next value is retrieved from the sequence.
If you want a random, but reproducible sequence, you may select a seed value for the random
number generator by adding
set rndseed=nn
to the spinit or .spiceinit file, nn being a positive integer number.
The transient noise analysis will allow the simulation of the three most important noise sources.
Thermal noise is described by the Gaussian white noise. Flicker noise (pink noise or 1 over
f noise) with an exponent between 0 and 2 is provided as well. Shot noise is dependent on
the current flowing through a device and may be simulated by applying a non-linear source as
demonstrated in the following example:
Example:
* Shot noise test with B source , diode
* voltage on device (diode , forward )
Vdev out 0 DC 0 PULSE (0.4 0.45 10u)
* diode , forward direction , to be modeled with noise
D1 mess 0 DMOD
.model DMOD D IS =1e -14 N=1
X1 0 mess out ishot
* device between 1 and 2
* new output terminals of device including noise: 1 and 3
. subckt ishot 1 2 3
* white noise source with rms 1V
* 20000 sample points
VNG 0 11 DC 0 TRNOISE (1 1n 0 0)
* measure the current i(v1)
V1 2 3 DC 0
* calculate the shot noise
* sqrt (2* current *q* bandwidth )
BI 1 3 I=sqrt (2* abs(i(v1 ))*1.6e -19*1 e7)*v(11)
.ends ishot
.tran 1n 20u
. control
run
plot ( -1)*i(vdev)
.endc
.end

260

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

The selection of the delta time step (NT) is worth discussing. Gaussian white noise has unlimited
bandwidth and thus unlimited energy content. This is unrealistic. The bandwidth of real noise
is limited, but it is still called ‘White’ if it is the same level throughout the frequency range
of interest, e.g. the bandwidth of your system. Thus you may select NT to be a factor of 10
smaller than the frequency limit of your circuit. A thorough analysis is still needed to clarify the
appropriate factor. The transient method is probably most suited to circuits including switches,
which are not amenable to the small signal .NOISE analysis (Chapt. 15.3.4).

There is a price you have to pay for transient noise analysis: the number of required time steps,
and thus the simulation time, increases.

In addition to white and 1/f noise the independent voltage and current sources offer a random
telegraph signal (RTS) noise source, also known as burst noise or popcorn noise, again for
transient analysis. For each voltage (current) source offering RTS noise an individual noise
amplitude is required for input, as well as a mean capture time and a mean emission time.
The amplitude resembles the influence of a single trap on the current or voltage. The capture
and emission times emulate the filling and emptying of the trap, typically following a Poisson
process. They are generated from an random exponential distribution with respective mean
values given by the user. To simulate an ensemble of traps, you may combine several current or
voltage sources with different parameters.

All three sources (white, 1/f, and RTS) may be combined in a single command line.

15.3. ANALYSES

261

RTS noise example:
* white noise , 1/f noise , RTS noise
* voltage source
VRTS2 13 12 DC 0 trnoise (0 0 0 0 5m 18u 30u)
VRTS3 11 0 DC 0 trnoise (0 0 0 0 10m 20u 40u)
VALL 12 11 DC 0 trnoise (1m 1u 1.0 0.1m 15m 22u 50u)
VW1of 21 0 DC

trnoise (1m 1u 1.0 0.1m)

* current source
IRTS2 10 0 DC 0 trnoise (0 0 0 0 5m 18u 30u)
IRTS3 10 0 DC 0 trnoise (0 0 0 0 10m 20u 40u)
IALL 10 0 DC 0 trnoise (1m 1u 1.0 0.1m 15m 22u 50u)
R10 10 0 1
IW1of 9 0 DC
Rall 9 0 1

trnoise (1m 1u 1.0 0.1m)

* sample points
.tran 1u 500u
. control
run
plot v(13) v(21)
plot v(10) v(9)
.endc
.end
Some details on RTS noise modeling are available in a recent article [20], available here.
This transient noise feature is still experimental.
The following questions (among others) are to be solved:
• clarify the theoretical background
• noise limit of plain ngspice (numerical solver, fft etc.)
• time step (NT) selection
• calibration of noise spectral density
• how to generate noise from a transistor model
• application benefits and limits

262

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

15.3.11

.PSS: Periodic Steady State Analysis

Experimental code, not yet made publicly available.
General form:
.pss gfreq tstab oscnob psspoints harms sciter steadycoeff 
Examples:
.pss 150 200e -3 2 1024 11 50 5e-3 uic
.pss 624 e6 1u v_plus 1024 10 150 5e-3 uic
.pss 624 e6 500n bout 1024 10 100 5e-3 uic
gfreq is guessed frequency of fundamental suggested by user. When performing transient
analysis the PSS algorithm tries to infer a new rough guess rgfreq on the fundamental. If
gfreq is out of ±10% with respect to rgfreq then gfreq is discarded.
tstab is stabilization time before the shooting begin to search for the PSS. It has to be noticed
that this parameter heavily influence the possibility to reach the PSS. Thus is a good practice to
ensure a circuit to have a right tstab, e.g. performing a separate TRAN analysis before to run
PSS analysis.
oscnob is the node or branch where the oscillation dynamic is expected. PSS analysis will give
a brief report of harmonic content at this node or branch.
psspoints is number of step in evaluating predicted period after convergence is reached. It
is useful only in Time Domain plots. However this number should be higher than 2 times the
requested harms. Otherwise the PSS analysis will properly adjust it.
harms number of harmonics to be calculated as requested by the user.
sciter number of allowed shooting cycle iterations. Default is 50.
steady_coeff is the weighting coefficient for calculating the Global Convergence Error (GCE),
which is the reference value in order to infer is convergence is reached. The lower steady_coeff
is set, the higher the accuracy of predicted frequency can be reached but at longer analysis time
and sciter number. Default is 1e-3.
uic (use initial conditions) is an optional keyword that indicates that the user does not want
ngspice to solve for the quiescent operating point before beginning the transient analysis. If this
keyword is specified, ngspice uses the values specified using IC=... on the various elements as
the initial transient condition and proceeds with the analysis. If the .ic control line has been
specified, then the node voltages on the .ic line are used to compute the initial conditions for
the devices. Look at the description on the .ic control line for its interpretation when uic is
not specified.

15.4. MEASUREMENTS AFTER AC, DC AND TRANSIENT ANALYSIS

15.4

Measurements after AC, DC and Transient Analysis

15.4.1

.meas(ure)

263

The .meas or .measure statement (and its equivalent meas command, see Chapt. 17.5.39)
are used to analyze the output data of a tran, ac, or dc simulation. The command is executed
immediately after the simulation has finished.

15.4.2

batch versus interactive mode

.meas analysis may not be used in batch mode (-b command line option), if an output file
(rawfile) is given at the same time (-r rawfile command line option). In this batch mode
ngspice will write its simulation output data directly to the output file. The data is not kept in
memory, thus is no longer available for further analysis. This is made to allow a very large
output stream with only a relatively small memory usage. For .meas to be active you need to
run the batch mode with a .plot or .print command. A better alternative may be to start
ngspice in interactive mode.
If you need batch like operation, you may add a .control ...
file:

.endc section to the input

Example:
*input file
...
.tran 1ns 1000 ns
...
*********************************
. control
run
write outputfile data
.endc
*********************************
.end
and start ngspice in interactive mode, e.g. by running the command
ngspice inputfile .
.meas then prints its user-defined data analysis to the standard output. The analysis
includes propagation, delay, rise time, fall time, peak-to-peak voltage, minimum or maximum
voltage, the integral or derivative over a specified period and several other user defined values.

15.4.3

General remarks

The measure type {DC|AC|TRAN|SP} depends on the data that is to be evaluated, either originating from a dc analysis, an ac analysis, or a transient simulation. The type SP to analyze a
spectrum from the spec or fft commands is only available when executed in a meas command,
see 17.5.39.

264

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

result will be a vector containing the result of the measurement. trig_variable, targ_variable,
and out_variable are vectors stemming from the simulation, e.g. a voltage vector v(out).
VAL=val expects a real number val. It may be as well a parameter delimited by ” or {}
expanding to a real number.
TD=td and AT=time expect a time value if measure type is tran. For ac and sp AT will be a
frequency value, TD is ignored. For dc analysis AT is a voltage (or current), TD is ignored as
well.
CROSS=# requires an integer number #. CROSS=LAST is possible as well. The same is expected
by RISE and FALL.
Frequency and time values may start at 0 and extend to positive real numbers. Voltage (or
current) inputs for the independent (scale) axis in a dc analysis may start or end at arbitrary real
valued numbers.
Please note that not all of the .measure commands have been implemented.

15.4.4

Input

In the following lines you will get some explanation on the .measure commands. A simple
simulation file with two sines of different frequencies may serve as an example. The transient
simulation delivers time as the independent variable and two voltages as output (dependent
variables).
Input file:
File: simple -meas -tran.sp
* Simple . measure examples
* transient simulation of two sine
* signals with different frequencies
vac1 1 0 DC 0 sin (0 1 1k 0 0)
vac2 2 0 DC 0 sin (0 1.2 0.9k 0 0)
.tran 10u 5m
*
. measure tran ... $ for the different inputs see below!
*
. control
run
plot v(1) v(2)
.endc
.end
After displaying the general syntax of the .measure statement, some examples are posted,
referring to the input file given above.

15.4.5

Trig Targ

.measure according to general form 1 measures the difference in dc voltage, frequency or time
between two points selected from one or two output vectors. The current examples all are using

15.4. MEASUREMENTS AFTER AC, DC AND TRANSIENT ANALYSIS

265

transient simulation. Measurements for tran analysis start after a delay time td. If you run
other examples with ac simulation or spectrum analysis, time may be replaced by frequency,
after a dc simulation the independent variable may become a voltage or current.
General form 1:
. MEASURE {DC|AC|TRAN|SP} result TRIG trig_variable VAL=
val
+   
+   TARG
targ_variable
+ VAL=val     
Measure statement example (for use in the input file given above):
.measure tran tdiff TRIG v(1) VAL=0.5 RISE=1 TARG v(1) VAL=0.5 RISE=2
measures the time difference between v(1) reaching 0.5 V for the first time on its first rising
slope (TRIG) versus reaching 0.5 V again on its second rising slope (TARG), i.e. it measures
the signal period.
Output:
tdiff = 1.000000e-003 targ= 1.083343e-003 trig= 8.334295e-005
Measure statement example:
.measure tran tdiff TRIG v(1) VAL=0.5 RISE=1 TARG v(1) VAL=0.5 RISE=3
measures the time difference between v(1) reaching 0.5 V for the first time on its rising slope
versus reaching 0.5 V on its rising slope for the third time (i.e. two periods).
Measure statement:
.measure tran tdiff TRIG v(1) VAL=0.5 RISE=1 TARG v(1) VAL=0.5 FALL=1
measures the time difference between v(1) reaching 0.5V for the first time on its rising slope
versus reaching 0.5 V on its first falling slope.
Measure statement:
.measure tran tdiff TRIG v(1) VAL=0 FALL=3 TARG v(2) VAL=0 FALL=3
measures the time difference between v(1) reaching 0V its third falling slope versus v(2) reaching 0 V on its third falling slope.
Measure statement:
.measure tran tdiff TRIG v(1) VAL=-0.6 CROSS=1 TARG v(2) VAL=-0.8 CROSS=1
measures the time difference between v(1) crossing -0.6 V for the first time (any slope) versus
v(2) crossing -0.8 V for the first time (any slope).
Measure statement:
.measure tran tdiff TRIG AT=1m TARG v(2) VAL=-0.8 CROSS=3
measures the time difference between the time point 1ms versus the time when v(2) crosses -0.8
V for the third time (any slope).

266

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

15.4.6

Find ... When

The FIND and WHEN functions allow to measure any dependent or independent time, frequency,
or dc parameter, when two signals cross each other or a signal crosses a given value. Measurements start after a delay TD and may be restricted to a range between FROM and TO.
General form 2:
. MEASURE {DC|AC|TRAN|SP} result WHEN out_variable =val
+    
+  
Measure statement:
.measure tran teval WHEN v(2)=0.7 CROSS=LAST
measures the time point when v(2) crosses 0.7 V for the last time (any slope).
General form 3:
. MEASURE {DC|AC|TRAN|SP} result
+ WHEN out_variable = out_variable2
+    
+  
Measure statement:
.measure tran teval WHEN v(2)=v(1) RISE=LAST
measures the time point when v(2) and v(1) are equal, v(2) rising for the last time.
General form 4:
. MEASURE {DC|AC|TRAN|SP} result FIND out_variable
+ WHEN out_variable2 =val   
+  
+ 
Measure statement:
.measure tran yeval FIND v(2) WHEN v(1)=-0.4 FALL=LAST
returns the dependent (y) variable drawn from v(2) at the time point when v(1) equals a value
of -0.4, v(1) falling for the last time.
General form 5:
. MEASURE {DC|AC|TRAN|SP} result FIND out_variable
+ WHEN out_variable2 = out_variable3 
+ 
+  
Measure statement:
.measure tran yeval FIND v(2) WHEN v(1)=v(3) FALL=2

15.4. MEASUREMENTS AFTER AC, DC AND TRANSIENT ANALYSIS

267

returns the dependent (y) variable drawn from v(2) at the time point when v(1) crosses v(3),
v(1) falling for the second time.
General form 6:
. MEASURE {DC|AC|TRAN|SP} result FIND out_variable AT=
val
Measure statement:
.measure tran yeval FIND v(2) AT=2m
returns the dependent (y) variable drawn from v(2) at the time point 2 ms (given by AT=time).

15.4.7

AVG|MIN|MAX|PP|RMS|MIN_AT|MAX_AT

General form 7:
. MEASURE {DC|AC|TRAN|SP} result
+ {AVG|MIN|MAX|PP|RMS| MIN_AT | MAX_AT }
+ out_variable   
Measure statements:
.measure tran ymax MAX v(2) from=2m to=3m
returns the maximum value of v(2) inside the time interval between 2 ms and 3 ms.
.measure tran tymax MAX_AT v(2) from=2m to=3m
returns the time point of the maximum value of v(2) inside the time interval between 2 ms and
3 ms.
.measure tran ypp PP v(1) from=2m to=4m
returns the peak to peak value of v(1) inside the time interval between 2 ms and 4 ms.
.measure tran yrms RMS v(1) from=2m to=4m
returns the root mean square value of v(1) inside the time interval between 2 ms and 4 ms.
.measure tran yavg AVG v(1) from=2m to=4m
returns the average value of v(1) inside the time interval between 2 ms and 4 ms.

15.4.8

Integ

General form 8:
. MEASURE {DC|AC|TRAN|SP} result INTEG  out_variable
+   
Measure statement:
.measure tran yint INTEG v(2) from=2m to=3m
returns the area under v(2) inside the time interval between 2 ms and 3 ms.

268

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

15.4.9

param

General form 9:
. MEASURE {DC|AC|TRAN|SP} result

param=’expression ’

Measure statement:
.param fval=5
.measure tran yadd param=’fval + 7’
will evaluate the given expression fval + 7 and return the value 12.
’Expression’ is evaluated according to the rules given in Chapt. 2.8.5 during start up of ngspice.
It may contain parameters defined with the .param statement. It may also contain parameters
resulting from preceding .meas statements.
.param vout_diff=50u
...
.measure tran tdiff TRIG AT=1m TARG v(2) VAL=-0.8 CROSS=3
.meas tran bw_chk param=’(tdiff < vout_diff) ?

1 :

0’

will evaluate the given ternary function and return the value 1 in bw_chk, if tdiff measured is
smaller than parameter vout_diff.
The expression may not contain vectors like v(10), e.g. anything resulting directly from a
simulation. This may be handled with the following .meas command option.

15.4.10

par(’expression’)

The par(’expression’) option (15.6.6) allows to use algebraic expressions on the .measure
lines. Every out_variable may be replaced by par(’expression’) using the general forms 1. . . 9
described above. Internally par(’expression’) is substituted by a vector according to the rules
of the B source (Chapt. 5.1). A typical example of the general form is shown below:
General form 10:
. MEASURE {DC|TRAN|AC|SP} result
+ FIND par(’ expression ’) AT=val
The measure statement
.measure tran vtest find par(’(v(2)*v(1))’) AT=2.3m
returns the product of the two voltages at time point 2.3 ms.
Note that a B-source, and therefore the par(’...’) feature, operates on values of type complex
in AC analysis mode.

15.4. MEASUREMENTS AFTER AC, DC AND TRANSIENT ANALYSIS

15.4.11

269

Deriv

General form:
. MEASURE {DC|AC|TRAN|SP} result DERIV 
out_variable
+ AT=val
. MEASURE {DC|AC|TRAN|SP} result DERIV 
out_variable
+ WHEN out_variable2 =val 
+  
+ 
. MEASURE {DC|AC|TRAN|SP} result DERIV 
out_variable
+ WHEN out_variable2 = out_variable3
+  
+  

15.4.12

More examples

Some other examples, also showing the use of parameters, are given below. Corresponding
demonstration input files are distributed with ngspice in folder /examples/measure.

270

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

Other examples:
.meas tran inv_delay2 trig v(in) val=’vp/2’ td=1n fall
=1
+
targ v(out) val=’vp/2’ rise =1
.meas tran test_data1 trig AT = 1n targ v(out)
+
val=’vp/2’ rise =3
.meas tran out_slew trig v(out) val = ’0.2*vp ’ rise =2
+
targ v(out) val = ’0.8*vp ’ rise =2
.meas tran delay_chk param =’( inv_delay < 100 ps) ? 1 :
0’
.meas tran skew when v(out)=0.6
.meas tran skew2 when v(out)= skew_meas
.meas tran skew3 when v(out)= skew_meas fall =2
.meas tran skew4 when v(out)= skew_meas fall=LAST
.meas tran skew5 FIND v(out) AT=2n
.meas tran v0_min min i(v0)
+
from=’dfall ’ to=’dfall+period ’
.meas tran v0_avg avg i(v0)
+
from=’dfall ’ to=’dfall+period ’
.meas tran v0_integ integ i(v0)
+
from=’dfall ’ to=’dfall+period ’
.meas tran v0_rms rms i(v0)
+
from=’dfall ’ to=’dfall+period ’
.meas dc is_at FIND i(vs) AT=1
.meas dc is_max max i(vs) from =0 to =3.5
.meas dc vds_at when i(vs)=0.01
.meas ac vout_at FIND v(out) AT=1 MEG
.meas ac vout_atd FIND vdb(out) AT=1 MEG
.meas ac vout_max max v(out) from =1k to =10 MEG
.meas ac freq_at when v(out)=0.1
.meas ac vout_diff trig v(out) val =0.1 rise =1 targ v(
out)
+
val =0.1 fall =1
.meas ac fixed_diff trig AT = 10k targ v(out)
+
val =0.1 rise =1
.meas ac vout_avg
avg
v(out) from =10k to=1 MEG
.meas ac vout_integ integ v(out) from =20k to =500k
.meas ac freq_at2 when v(out)=0.1 fall=LAST
.meas ac bw_chk param =’( vout_diff < 100k) ? 1 : 0’
.meas ac vout_rms rms v(out) from =10 to=1G

15.5

Safe Operating Area (SOA) warning messages

By setting .option warn=1 the Safe Operation Area check algorithm is enabled. In this case
for .op, .dc and .tran analysis warning messages are issued if the branch voltages of devices

15.5. SAFE OPERATING AREA (SOA) WARNING MESSAGES

271

(Resistors, Capacitors, Diodes, BJTs and MOSFETs) exceed limits that are specified by model
parameters. All these parameters are positive with default value of infinity.
The check is executed after Newton-Raphson iteration is finished i.e. in transient analysis in
each time step. The user can specify an additional .option maxwarns (default: 5) to limit the
count of messages.
The output goes on default to stdout or alternatively to a file specified by command line option
--soa-log=filename.

15.5.1

Resistor and Capacitor SOA model parameters

1. Bv_max:

15.5.2

if |Vr| or |Vc| exceed Bv_max, SOA warning is issued.

Diode SOA model parameter

1. Bv_max:

if |Vj| exceeds Bv_max, SOA warning is issued.

2. Fv_max:

if |Vf| exceeds Fv_max, SOA warning is issued.

15.5.3

BJT SOA model parameter

1. Vbe_max:

if |Vbe| exceeds Vbe_max, SOA warning is issued.

2. Vbc_max:

if |Vbc| exceeds Vbc_max, SOA warning is issued.

3. Vce_max:

if |Vce| exceeds Vce_max, SOA warning is issued.

4. Vcs_max:

if |Vcs| exceeds Vcs_max, SOA warning is issued.

15.5.4

MOS SOA model parameter

1. Vgs_max:

if |Vgs| exceeds Vgs_max, SOA warning is issued.

2. Vgd_max:

if |Vgd| exceeds Vgd_max, SOA warning is issued.

3. Vgb_max:

if |Vgb| exceeds Vgb_max, SOA warning is issued.

4. Vds_max:

if |Vds| exceeds Vds_max, SOA warning is issued.

5. Vbs_max:

if |Vbs| exceeds Vbs_max, SOA warning is issued.

6. Vbd_max:

if |Vbd| exceeds Vbd_max, SOA warning is issued.

272

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

15.6

Batch Output

The following commands .print (15.6.2), .plot (15.6.3) and .four (15.6.4) are valid only
if ngspice is started in batch mode (see 16.4.1), whereas .save and the equivalent .probe are
aknowledged in all operating modes.
If you start ngspice in batch mode using the -b command line option, the outputs of .print,
.plot, and .four are printed to the console output. You may use the output redirection of your
shell to direct this printout into a file (not available with MS Windows GUI). As an alternative
you may extend the ngspice command by specifying an output file:
ngspice -b -o output.log input.cir
If you however add the command line option -r to create a rawfile, .print and .plot are
ignored. If you want to involve the graphics plot output of ngspice, use the control mode
(16.4.3) instead of the -b batch mode option.

15.6.1

.SAVE: Name vector(s) to be saved in raw file

General form:
.save vector vector vector ...
Examples:
.save i(vin) node1 v( node2)
.save @m1[id] vsource # branch
.save all @m2[ vdsat ]
The vectors listed on the .SAVE line are recorded in the rawfile for use later with ngspice or ngnutmeg (ngnutmeg is just the data-analysis half of ngspice, without the ability to simulate). The
standard vector names are accepted. Node voltages may be saved by giving the nodename or
v(nodename). Currents through an independent voltage source are given by i(sourcename)
or sourcename#branch. Internal device data are accepted as @dev[param].
If no .SAVE line is given, then the default set of vectors is saved (node voltages and voltage
source branch currents). If .SAVE lines are given, only those vectors specified are saved. For
more discussion on internal device data, e.g. @m1[id], see Appendix, Chapt. 31.1. If you want
to save internal data in addition to the default vector set, add the parameter all to the additional
vectors to be saved. If the command .save vm(out) is given, and you store the data in a rawfile, only the original data v(out) are stored. The request for storing the magnitude is ignored,
because this may be added later during rawfile data evaluation with ngnutmeg or ngspice. See
also the section on the interactive command interpreter (Chapt. 17.5) for information on how to
use the rawfile.

15.6. BATCH OUTPUT

15.6.2

273

.PRINT Lines

General form:
.print prtype ov1 
Examples:
.print tran v(4) i(vin)
.print dc v(2) i(vsrc) v(23, 17)
.print ac vm(4, 2) vr (7) vp(8, 3)
The .print line defines the contents of a tabular listing of one to eight output variables. prtype
is the type of the analysis (DC, AC, TRAN, NOISE, or DISTO) for which the specified outputs are
desired. The form for voltage or current output variables is the same as given in the previous
section for the print command; Spice2 restricts the output variable to the following forms
(though this restriction is not enforced by ngspice):
V(N1<,N2>)

I(VXXXXXXX)

specifies the voltage difference between nodes N1 and N2.
If N2 (and the preceding comma) is omitted, ground (0) is
assumed. See the print command in the previous section
for more details. For compatibility with SPICE2, the
following five additional values can be accessed for the ac
analysis by replacing the ‘V’ in V(N1,N2) with:
VR
Real part
VI
Imaginary part
VM
Magnitude
VP
Phase
VDB 20log10(magnitude)
specifies the current flowing in the independent voltage
source named VXXXXXXX. Positive current flows from
the positive node, through the source, to the negative node.
(Not yet implemented: For the ac analysis, the
corresponding replacements for the letter I may be made
in the same way as described for voltage outputs.)

Output variables for the noise and distortion analyses have a different general form from that of
the other analyses. There is no limit on the number of .print lines for each type of analysis.
The par(’expression’) option (15.6.6) allows to use algebraic expressions in the .print lines.
.width (15.6.7) selects the maximum number of characters per line.

15.6.3

.PLOT Lines

.plot creates a printer plot output.

274

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

General form:
.plot pltype ov1 <(plo1 , phi1)>  ... ov8 >
Examples:
.plot
.plot
.plot
.plot
.plot

dc v(4) v(5) v(1)
tran v(17 , 5) (2, 5) i(vin) v(17) (1, 9)
ac vm (5) vm (31 , 24) vdb (5) vp (5)
disto hd2 hd3(R) sim2
tran v(5, 3) v(4) (0, 5) v(7) (0, 10)

The .plot line defines the contents of one plot of from one to eight output variables. pltype is
the type of analysis (DC, AC, TRAN, NOISE, or DISTO) for which the specified outputs are desired.
The syntax for the ovi is identical to that for the .print line and for the plot command in the
interactive mode.
The overlap of two or more traces on any plot is indicated by the letter ‘X’. When more than
one output variable appears on the same plot, the first variable specified is printed as well
as plotted. If a printout of all variables is desired, then a companion .print line should be
included. There is no limit on the number of .plot lines specified for each type of analysis.
The par(’expression’) option (15.6.6) allows to use algebraic expressions in the .plot lines.

15.6.4

.FOUR: Fourier Analysis of Transient Analysis Output

General form:
.four freq ov1 
Examples:
.four 100K v(5)
The .four (or Fourier) line controls whether ngspice performs a Fourier analysis as a part of
the transient analysis. freq is the fundamental frequency, and ov1 is the desired vector to
be analyzed. The Fourier analysis is performed over the interval ,
where TSTOP is the final time specified for the transient analysis, and period is one period of
the fundamental frequency. The dc component and the first nine harmonics are determined. For
maximum accuracy, TMAX (see the .tran line) should be set to period/100.0 (or less for very
high-Q circuits). The par(’expression’) option (15.6.6) allows to use algebraic expressions in
the .four lines.

15.6. BATCH OUTPUT

15.6.5

275

.PROBE: Name vector(s) to be saved in raw file

General form:
.probe vector 
Examples:
.probe i(vin) input output
.probe @m1[id]
Same as .SAVE (see Chapt. 15.6.1).

15.6.6

par(’expression’): Algebraic expressions for output

General form:
par(’ expression ’)
output =par(’ expression ’)

$ not in . measure ac

Examples:
.four 1001 sq1=par(’v(1)*v(1) ’)
. measure tran vtest find par (’(v(2)*v(1)) ’) AT =2.3m
.print tran output =par(’v(1)/v(2) ’) v(1) v(2)
.plot dc v(1) diff=par (’(v(4)-v (2))/0.01 ’) out222
With the output lines .four, .plot, .print, .save and in .measure evaluation it is possible to add algebraic expressions for output, in addition to vectors. All of these output lines
accept par(’expression’), where expression is any expression valid for a B source (see Chapt.
5.1). Thus expression may contain predefined functions, numerical values, constants, simulator output like v(n1) or i(vdb), parameters predefined by a .param statement, and the variables
hertz, temper, and time. Note that a B-source, and therefore the par(’...’) feature, operates on values of type complex in AC analysis mode.
Internally the expression is replaced by a generated voltage node that is the output of a B source,
one node, and the B source implementing par(’...’). Several par(’...’) are allowed in each line,
up to 99 per input file. The internal nodes are named pa_00 to pa_99. An error will occur if
the input file contains any of these reserved node names.
In .four, .plot, .print, .save, but not in .measure, an alternative syntax
output=par(’expression’) is possible. par(’expression’) may be used as described above.
output is the name of the new node to replace the expression. So output has to be unique and
a valid node name.
The syntax of output=par(expression) is strict, no spaces between par and (’, or between (
and ’ are allowed, (’ and ’) both are required. Also there is not much error checking on your
input, if there is a typo, for example, an error may pop up at an unexpected place.

276

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

15.6.7

.width

Set the width of a print-out or plot with the following card:
.with out = 256
Parameter out yields the maximum number of characters plotted in a row, if printing in columns
or an ASCII-plot is selected.

15.7

Measuring current through device terminals

15.7.1

Adding a voltage source in series

Originally the ngspice matrix solver delivers node voltages and currents through independent
voltage sources. So to measure the currents through a resistor you may add a voltage source in
series with dc voltage 0.
Current measurement with series voltage source
* measure current through R1
V1 1 0 1
R1 1 0 5
R2 1 0 10
* will become
V1 1 0 1
R1 1 11 5
Vmess 11 0 dc 0
R2 1 0 10

15.7.2

Using option ’savecurrents’

Current measurement with series voltage source
* measure current through R1 and R2
V1 1 0 1
R1 1 0 5
R2 1 0 10
. options savecurrents
The option savecurrents will add .save lines (15.6.1) like
.save @r1[i]
.save @r2[i]
to your input file information read during circuit parsing. These newly created vectors contain
the terminal currents of the devices R1 and R2.
You will find information of the nomenclature in Chapt. 31, also how to plot these vectors. The
following devices are supported: M, J, Q, D, R, C, L, B, F, G, W, S, I (see 2.1.2). For M only

15.7. MEASURING CURRENT THROUGH DEVICE TERMINALS

277

MOSFET models MOS1 to MOS9 are included so far. Devices in subcircuits are supported as
well. Be careful when choosing this option in larger circuits, because 1 to 4 additional output
vectors are created per device and this may consume lots of memory.

278

CHAPTER 15. ANALYSES AND OUTPUT CONTROL (BATCH MODE)

Chapter 16
Starting ngspice
16.1

Introduction

Ngspice consists of the simulator and a front-end for data analysis and plotting. Input to the
simulator is a netlist file, including commands for circuit analysis and output control. Interactive
ngspice can plot data from a simulation on a PC or a workstation display.
Ngspice on Linux (and OSs like Cygwin, BCD, Solaris ...) uses the X Window System for
plotting (see Chapt. 18.3) if the environment variable DISPLAY is available. Otherwise, a console mode (non-graphical) interface is used. If you are using X on a workstation, the DISPLAY
variable should already be set; if you want to display graphics on a system different from the
one you are running ngspice or ngutmeg on, DISPLAY should be of the form machine:0.0. See
the appropriate documentation on the X Window System for more details.
The MS Windows versions of ngspice and ngnutmeg will have a native graphics interface (see
Chapt. 18.1).
The front-end may be run as a separate ‘stand-alone’ program under the name ngnutmeg. ngnutmeg is a subset of ngspice dedicated to data evaluation, still made available for historical
reasons. Ngnutmeg will read in the ‘raw’ data output file created by ngspice -r or by the write
command during an interactive ngspice session.

16.2

Where to obtain ngspice

The actual distribution of ngspice may be downloaded from the ngspice download web page.
The installation for Linux or MS Windows is described in the file INSTALL to be found in
the top level directory. You may also have a look at Chapt. 32 of this manual for compiling
instructions.
If you want to check out the source code that is actually under development, you may have a
look at the ngspice source code repository, which is stored using the Git Source Code Management (SCM) tool. The Git repository may be browsed on the Git web page, also useful for
downloading individual files. You may however download (or clone) the complete repository
including all source code trees from the console window (Linux, CYGWIN or MSYS/MINGW)
by issuing the command (in a single line)

279

280

CHAPTER 16. STARTING NGSPICE

git clone git://ngspice.git.sourceforge.net/gitroot/ngspice/ngspice

You need to have Git installed, which is available for all three OSs. The whole source tree
is then available in /ngspice. Compilation and local installation is again
described in INSTALL (or Chapt. 32). If you later want to update your files and download the
recent changes from sourceforge into your local repository, cd into the ngspice directory and
just type

git pull

git pull will deny to overwrite modified files in your working directory. To drop your local
changes first, you can run

git reset --hard

To learn more about git, which can be both powerful and difficult to master, please consult
http://git-scm.com/, especially: http://git-scm.com/documentation, which has pointers to documentation and tutorials.

16.3

Command line options for starting ngspice and ngnutmeg

Command Synopsis:
ngspice [ -o logfile ] [ -r rawfile ] [ -b ] [ -i ] [ input files ]
ngnutmeg [ - ] [ datafile ... ]

Options are:

16.3. COMMAND LINE OPTIONS FOR STARTING NGSPICE AND NGNUTMEG
Option
-

Long option

-n

--no-spiceinit

-t TERM

--terminal=TERM

-b

--batch

-s

--server

-i

--interactive

-r FILE

--rawfile=FILE

-p

--pipe

-o FILE

--output=FILE

-h
-v
-a

--help
--version
--autorun

--soa-log=FILE

281

Meaning
Don’t try to load the default data file ("rawspice.raw") if no
other files are given (ngnutmeg only).
Don’t try to source the file .spiceinit upon start-up.
Normally ngspice and ngnutmeg try to find the file in the
current directory, and if it is not found then in the user’s
home directory (obsolete).
The program is being run on a terminal with mfb name
term (obsolete).
Run in batch mode. Ngspice reads the default input source
(e.g. keyboard) or reads the given input file and performs
the analyses specified; output is either Spice2-like
line-printer plots ("ascii plots") or a ngspice rawfile. See
the following section for details. Note that if the input
source is not a terminal (e.g. using the IO redirection
notation of "<") ngspice defaults to batch mode (-i
overrides). This option is valid for ngspice only.
Run in server mode. This is like batch mode, except that a
temporary rawfile is used and then written to the standard
output, preceded by a line with a single "@", after the
simulation is done. This mode is used by the ngspice
daemon. This option is valid for ngspice only.
Example for using pipes from the console window:
cat adder.cir|ngspice -s|more
Run in interactive mode. This is useful if the standard input
is not a terminal but interactive mode is desired. Command
completion is not available unless the standard input is a
terminal, however. This option is valid for ngspice only.
Use rawfile as the default file into which the results of the
simulation are saved. This option is valid for ngspice only.
Allow a program (e.g., xcircuit) to act as a GUI frontend
for ngspice through a pipe. Thus ngspice will assume that
the input pipe is a tty and allows to run in interactive mode.
All logs generated during a batch run (-b) will be saved in
outfile.
A short help statement of the command line syntax.
Prints a version information.
Start simulation immediately, as if a control section
.control
run
.endc
had been added to the input file.
output from Safe Operating Area (SOA) check

Further arguments to ngspice are taken to be ngspice input files, which are read and saved (if
running in batch mode then they are run immediately). Ngspice accepts Spice3 (and also most
Spice2) input files, and outputs ASCII plots, Fourier analyses, and node printouts as specified
in .plot, .four, and .print cards. If an out parameter is given on a .width card (15.6.7),

282

CHAPTER 16. STARTING NGSPICE

the effect is the same as set width = .... Since ngspice ASCII plots do not use multiple ranges,
however, if vectors together on a .plot card have different ranges they do not provide as much
information as they do in a scalable graphics plot.
For ngnutmeg, further arguments are taken to be data files in binary or ASCII raw file format
(generated with -r in batch mode or the write (see 17.5.88) command) that are loaded into ngnutmeg. If the file is in binary format, it may be only partially completed (useful for examining
output before the simulation is finished). One file may contain any number of data sets from
different analyses.

16.4

Starting options

16.4.1

Batch mode

Let’s take as an example the Four-Bit binary adder MOS circuit shown in Chapt. 21.6, stored
in a file adder-mos.cir. You may start the simulation immediately by calling
ngspice -b -r adder.raw -o adder.log adder-mos.cir
ngspice will start, simulate according to the .tran command and store the output data in a
rawfile adder.raw. Comments, warnings and infos go to log file adder.log. Commands for batch
mode operation are described in Chapt. 15.

16.4.2

Interactive mode

If you call
ngspice
ngspice will start, load spinit (16.5) and .spiceinit (16.6, if available), and then waits for your
manual input. Any of the commands described in 17.5 may be chosen, but many of them are
useful only after a circuit has been loaded by
ngspice 1 -> source adder-mos.cir
others require the simulation being done already (e.g. plot):
ngspice 2 ->run
ngspice 3 ->plot allv
If you call ngspice from the command line with a circuit file as parameter:
ngspice adder-mos.cir
ngspice will start, load the circuit file, parse the circuit (same circuit file as above, containing
only dot commands (see Chapt. 15) for analysis and output control). ngspice then just waits for
your input. You may start the simulation by issuing the run command. Following completion
of the simulation you may analyze the data by any of the commands given in Chapt. 17.5.

16.4. STARTING OPTIONS

16.4.3

283

Control mode (Interactive mode with control file or control section)

If you add the following control section to your input file adder-mos.cir, you may call
ngspice adder-mos.cir
from the command line and see ngspice starting, simulating and then plotting immediately.
Control section:
* ADDER - 4 BIT ALL -NAND -GATE BINARY ADDER
. control
unset askquit
save vcc# branch
run
plot vcc# branch
rusage all
.endc
Any suitable command listed in Chapt. 17.5 may be added to the control section, as well as
control structures described in Chapt. 17.6. Batch-like behavior may be obtained by changing
the control section to
Control section with batch-like behavior:
* ADDER - 4 BIT ALL -NAND -GATE BINARY ADDER
. control
unset askquit
save vcc# branch
run
write adder .raw vcc# branch
quit
.endc
If you put this control section into a file, say adder-start.sp, you may just add the line
.include adder-start.sp
to your input file adder-mos.cir to obtain the batch-like behavior. In the following example
the line .tran ... from the input file is overridden by the tran command given in the control
section.
Control section overriding the .tran command:
* ADDER - 4 BIT ALL -NAND -GATE BINARY ADDER
. control
unset askquit
save vcc# branch
tran 1n 500n
plot vcc# branch
rusage all
.endc

284

CHAPTER 16. STARTING NGSPICE

The commands within the .control section are executed in the order they are listed and only
after the circuit has been read in and parsed. If you want to have a command being executed
before circuit parsing, you may use the prefix pre_ (17.5.46) to the command.
A warning is due however: If your circuit file contains such a control section (.control ...
.endc), you should not start ngspice in batch mode (with -b as parameter). The outcome may
be unpredictable!

16.5

Standard configuration file spinit

At startup ngspice reads its configuration file spinit. spinit may be found in a path relative to
the location of the ngspice executable
..\share\ngspice\scripts. The path may be overridden by setting the environmental variable
SPICE_SCRIPTS to a path where spinit is located. Ngspice for Windows will additionally
search for spinit in the directory where ngspice.exe resides. If spinit is not found a warning
message is issued, but ngspice continues.
Standard spinit contents:
* Standard ngspice init file
alias exit quit
alias acct rusage all
** set the number of threads in openmp
** default (if compiled with --enable - openmp ) is: 2
set num_threads =4
if $? sharedmode
unset interactive
unset moremode
else
set interactive
set x11lineararcs
end
strcmp __flag $program " ngspice "
if $__flag = 0
codemodel
codemodel
codemodel
codemodel
codemodel
codemodel

../ lib/ spice/ spice2poly .cm
../ lib/ spice/ analog .cm
../ lib/ spice/ digital .cm
../ lib/ spice/ xtradev .cm
../ lib/ spice/ xtraevt .cm
../ lib/ spice/table.cm

end
unset __flag
spinit contains a script, made of commands from chapt. 17.5, that is run upon start up of

16.6. USER DEFINED CONFIGURATION FILE .SPICEINIT

285

ngspice. Aliases (name equivalences) can be set. The asterisk ‘*’ comments out a line. If
used by ngspice, spinit will then load the XSPICE code models from a path relative to the
current directory where the ngspice executable resides. You may also define absolute paths.
The compatibility mode of ngspice can be set in spinit with set ngbehavior=all.
If the standard path for the libraries (see standard spinit above or /usr/local/lib/spice under CYGWIN and Linux) is not adequate, you can add the ./configure options --prefix=/usr
--libdir=/usr/lib64 to set the codemodel search path to /usr/lib64/spice. Besides the
standard lib only lib64 is acknowledged.
Special care has to be taken when using the ngspice shared library. If you apply ngspice.dll
under Windows OS, the standard is to use relative paths for the code models as shown above.
However, the path is relative to the calling program, not to the dll. This is fine when ngspice.dll
and the calling program reside in the same directory. If ngspice.dll is placed in a different
directory, please check chapt. 32.2.
The Linux shared library ... t.b.d.

16.6

User defined configuration file .spiceinit

In addition to spinit you may define a (personal) file .spiceinit and put it into the current directory or in your home directory. The typical search sequence for .spiceinit is: current directory, HOME (Linux) and then USERPROFILE (Windows). USERPROFILE is typically
C:\Users\. This file will be read in and executed after spinit, but before any
other input file is read. It may contain further scripts, set variables, or issue commands from
chapt.17.5 to override commands given in spinit. For example set filetype=ascii will
yield ASCII output in the output data file (rawfile), instead of the compact binary format that is
used by default. set ngdebug will yield a lot of additional debug output. Any other contents
of the script, e.g. plotting preferences, may be included here also. If the command line option
-n is used upon ngspice start up, this file will be ignored.
.spiceinit may contain:
* User defined ngspice init file
set filetype = ascii
*set ngdebug
set numthreads = 8
*set outputpath =C:\ Spice64 \out

16.7

Environmental variables

16.7.1

Ngspice specific variables

SPICE_LIB_DIR default: /usr/local/share/ngspice (Linux, CYGWIN), C:\Spice\share\ngspice
(Windows)
SPICE_EXEC_DIR default: /usr/local/bin (Linux, CYGWIN), C:\Spice\bin (Windows)

286

CHAPTER 16. STARTING NGSPICE

SPICE_BUGADDR default: http://ngspice.sourceforge.net/bugrep.html
Where to send bug reports on ngspice.
SPICE_EDITOR default: vi (Linux, CYGWIN), notepad.exe (MINGW, Visual Studio)
Set the editor called in the edit command. Always overrides the EDITOR env. variable.
SPICE_ASCIIRAWFILE default: 0
Format of the rawfile. 0 for binary, and 1 for ascii.
SPICE_NEWS default: $SPICE_LIB_DIR/news
A file that is copied verbatim to stdout when ngspice starts in interactive mode.
SPICE_HELP_DIR default: $SPICE_LIB_DIR/helpdir
Help directory, not used in Windows mode
SPICE_HOST default: empty string
Used in the rspice command (probably obsolete, to be documented)
SPICE_SCRIPTS default: $SPICE_LIB_DIR/scripts
In this directory the spinit file will be searched.
SPICE_PATH default: $SPICE_EXEC_DIR/ngspice
Used in the aspice command (probably obsolete, to be documented)
NGSPICE_MEAS_PRECISION default: 5
Sets the number of digits if output values are printed by the meas(ure) command.
SPICE_NO_DATASEG_CHECK default: undefined
If defined, will suppress memory resource info (probably obsolete, not used on Windows
or where the /proc information system is available.)
NGSPICE_INPUT_DIR default: undefined
If defined, using a valid directory name, will add the given directory to the search path
when looking for input files (*.cir, *.inc, *.lib).

16.7.2

Common environment variables

TERM LINES COLS DISPLAY HOME PATH EDITOR SHELL POSIXLY_CORRECT

16.8

Memory usage

Ngspice started with batch option (-b) and rawfile output (-r rawfile) will store all simulation
data immediately into the rawfile without keeping them in memory. Thus very large circuits
may be simulated, the memory requested upon ngspice start up will depend on the circuit size,
but will not increase during simulation.
If you start ngspice in interactive mode or interactively with control section, all data will be kept
in memory, to be available for later evaluation. A large circuit may outgrow even Gigabytes of
memory. The same may happen after a very long simulation run with many vectors and many
time steps to be stored. Issuing the save  command will help to reduce memory
requirements by saving only the data defined by the command. You may alos choose option
INTERP (15.1.4) to reduce memory usage.

16.9. SIMULATION TIME

16.9

287

Simulation time

Simulating large circuits may take an considerable amount of CPU time. If this is of importance,
you should compile ngspice with the flags for optimum speed, set during configuring ngspice
compilation. Under Linux, MINGW, and CYGWIN you should select the following option to
disable the debug mode, which slows down ngspice:
./configure --disable-debug
Adding --disable-debug will set the -O2 optimization flag for compiling and linking.
Under MS Visual Studio, you will have to select the release version, which includes optimization for speed.
If you have selected XSPICE (see Chapt. 12 and II) as part of your compilation configuration
(by adding the option --enable-xspice to your ./configure command), the value of trtol
(see 15.1.4) is set internally to 1 (instead of default 7) for higher precision if XSPICE code
model ’A’ devices included in the circuit. This may double or even triple the CPU time needed
for any transient simulation, because the amount of time steps and thus iteration steps is more
than doubled. For MS Visual Studio compilation there is currently no simple way to exclude
XSPICE during compilation.
You may enforce higher speed during XSPICE usage by setting the variable xtrtol in your
.spiceinit initialization file or in the .control section in front of the tran command (via set
xtrtol=2 using the set command 17.5.59) and override the above trtol reduction. Beware
however of precision or convergence issues if you use XSPICE ’A’ devices, especially if xtrtol
is set to values larger than 2.
If your circuit comprises mostly of MOS transistors, and you have a multi-core processor at
hand, you may benefit from OpenMP parallel processing, as described next (16.10).

16.10

Ngspice on multi-core processors using OpenMP

16.10.1

Introduction

Today’s computers typically come with CPUs having more than one core. It will thus be useful
to enhance ngspice to make use of such multi-core processors.
Using circuits comprising mostly of transistors and e.g. the BSIM3 model, around 2/3 of the
CPU time is spent in evaluating the model equations (e.g. in the BSIM3Load() function). The
same happens with other advanced transistor models. Thus this function should be paralleled, if
possible. Resulting from that the parallel processing has to be within a dedicated device model.
Interestingly solving the matrix takes only about 10% of the CPU time, so paralleling the matrix
solver is of secondary interest here!
A recent publication [1] has described a way to exactly do that using OpenMP, which is available
on many platforms and is easy to use, especially if you want to parallel processing of a for-loop.
To explain the implemented approach BSIM3 version 3.3.0 model was chosen, located in the
BSIM3 directory, as the first example. The BSIM3load() function in b3ld.c contains two nested
for-loops using linked lists (models and instances, e.g. individual transistors). Unfortunately

288

CHAPTER 16. STARTING NGSPICE

Table 16.1: OpenMP performance
Threads
CPU time [s] CPU time [s]
Windows
Linux
1 (standard)
167
165
1 (OpenMP)
174
167
2
110
110
3
95
94-120
4
83
107
6
94
90
8
93
91
OpenMP requires a loop with an integer index. So in file B3set.c an array is defined, filled with
pointers to all instances of BSIM3 and stored in model->BSIM3InstanceArray.
BSIM3load() is now a wrapper function, calling the for-loop, which runs through functions
BSIM3LoadOMP(), once per instance. Inside BSIM3LoadOMP() the model equations are calculated.
Typically need it is needed to synchronize the activities, in that storing the results into the
matrix has to be guarded. The trick offered by the authors now is that the storage is moved out
of the BSIM3LoadOMP() function. Inside BSIM3LoadOMP() the updated data are stored in
extra locations locally per instance, defined in bsim3def.h. Only after the complete for-loop
is exercised, the update to the matrix is done in an extra function BSIM3LoadRhsMat() in the
main thread after the paralleled loop. No extra synchronization is required.
Then the thread programming needed is only a single line!!
#pragma omp parallel for
introducing the for-loop over the device instances.
This of course is made possible only thanks to the OpenMP guys and the clever trick on no
synchronization introduced by the above cited authors.
The time-measuring function getrusage() used with Linux or Cygwin to determine the CPU
time usage (with the rusage option enabled) counts tics from every core, adds them up, and
thus reports a CPU time value enlarged by a factor of 8 if 8 threads have been chosen. So now
ngspice is forced to use ftime for time measuring if OpenMP is selected.

16.10.2

Some results

Some results on an inverter chain with 627 CMOS inverters, running for 200ns, compiled with
Visual Studio professional 2008 on Windows 7 (full optimization) or gcc 4.4, SUSE Linux 11.2,
-O2, on a i7 860 machine with four real cores (and 4 virtuals using hyperthreading) are shown
in table 16.1.
So we see a ngspice speed up of nearly a factor of two! Even on an older notebook with a dual
core processor, more than 1.5x improvement using two threads was attained. Similar results are
to be expected from BSIM4.

16.11. SERVER MODE OPTION -S

16.10.3

289

Usage

To state it clearly: OpenMP is installed inside the model equations of a particular model. So for
the moment it is available only in BSIM3 version 3.3.0, not in version 3.2.4 nor in any other
BSIM3 model, in BSIM4 versions 4.6.5 or 4.8, not in any other BSIM4 model, and in B4SOI,
version 4.3.1, not in any other SOI model. Older parameter files of version 4.6.x (x any number
up to 5) are accepted, you have to check for compatibility.
Under Linux you may run
./autogen.sh
./configure ...

--enable-openmp

make install
The same has been tested under MS Windows with CYGWIN and MINGW as well and delivers similar results.
Under MS Windows with Visual Studio Professional you have to place an additional preprocessor flag USE_OMP, and then enable /openmp. Visual Studio Express allone is not sufficient
due to lack of OpenMP support. But OpenMP is provided after installation of the Microsoft
SDK version 7.1. Even Visual Studio Professional lacks debugging support for OpenMP.
The number of threads has to be set manually by placing
set num_threads=4
into spinit or .spiceinit or in the control section of the SPICE input file. If OpenMP is enabled,
but num_threads not set, a default value num_threads=2 is set internally.
If you run a circuit, please keep in mind to select BSIM3 (levels 8, 49) version 3.3.0 (11.2.9),
by placing this version number into your parameter files, BSIM4 (levels 14, 54) version 4.6.5
or 4.8 (11.2.10), or B4SOI (levels 10, 58) version 4.3.1 (11.2.12).
If you run ./configure without --enable-openmp (or without USE_OMP preprocessor flag under MS Windows), you will get the standard, not paralleled BSIM3 and BSIM4 model, as has
been available from Berkeley. If OpenMP is selected and the number of threads set to 1, there
will be only a very slight CPU time disadvantage (typ. 3%) compared to the standard, non
OpenMP build.

16.10.4

Literature

[1] R.K. Perng, T.-H. Weng, and K.-C. Li: "On Performance Enhancement of Circuit Simulation
Using Multithreaded Techniques", IEEE International Conference on Computational Science
and Engineering, 2009, pp. 158-165

16.11

Server mode option -s

A program may write the SPICE input to the console. This output is redirected to ngspice via
‘|’. ngspice called with the -s option writes its output to the console, which again is redirected
to a receiving program by ‘|’. In the following simple example cat reads the input file and

290

CHAPTER 16. STARTING NGSPICE

prints it content to the console, which is redirected to ngspice by a first pipe, ngspice transfers
its output (similar to a raw file, see below) to less via another pipe.
Example command line:
cat input .cir| ngspice -s|less
Under MS Windows you will need to compile ngspice as a console application (see Chapt.
32.2.5) for this server mode usage.
Example input file:
test -s
v1 1 0 1
r1 1 0 2k
. options filetype = ascii
.save i(v1)
.dc v1 -1 1 0.5
.end
If you start ngspice console with
ngspice -s
you may type in the above circuit line by line (not to forget the first line, which is a title and
will be ignored). If you close your input with ctrl Z, and return, you will get the following
output (this is valid for MINGW only) on the console, like a raw file:
Circuit: test -s
Doing analysis at TEMP = 27.000000 and TNOM = 27.000000
Title: test -s
Date: Sun Jan 15 18:57:13 2012
Plotname: DC transfer characteristic
Flags: real
No. Variables: 2
No. Points: 0
Variables:
No. of Data Columns : 2
0 v(v-sweep) voltage
1 i(v1) current
Values:
0
-1.000000000000000e+000
5.000000000000000e-004
1
-5.000000000000000e-001
2.500000000000000e-004
2
0.000000000000000e+000
0.000000000000000e+000

16.12. NGSPICE CONTROL VIA INPUT, OUTPUT FIFOS

291

3

5.000000000000000e-001
-2.500000000000000e-004
4
1.000000000000000e+000
-5.000000000000000e-004
@@@ 122 5

The number 5 of the last line @@@ 122 5 shows the number of data points, which is missing in
the above line No. Points: 0 because at the time of writing to the console it has not yet
been available.

ctrl Z is not usable here in Linux, a patch to install ctrl D instead is being evaluated.

16.12

Ngspice control via input, output fifos

The following bash script (under Linux)

- launches ngspice in another thread.

- writes some commands in ngspice input

- reads the output and prints them on the console.

292

CHAPTER 16. STARTING NGSPICE

Example:
#!/ usr/bin/env bash
NGSPICE_COMMAND =" ngspice "
rm input .fifo
rm output .fifo
mkfifo input .fifo
mkfifo output .fifo
$NGSPICE_COMMAND

-p -i output .fifo &

exec 3> input .fifo
echo "I can write to input.fifo"
echo " Start processing ..."
echo ""
echo
echo
echo
echo
echo
echo

" source circuit .cir" >&3
" unset askquit " >&3
"set nobreak " >&3
"tran 0.01 ms 0.1 ms">&3
" print n0" >&3
"quit" >&3

echo "Try to open output .fifo ..."
exec 4< output .fifo
echo "I can read from output .fifo"
echo " Ready to read ..."
while read output
do
echo $output
done <&4
exec 3>&exec 4>&echo "End processing "

The input file for SPICE is:

16.13. COMPATIBILITY

293

Circuit.cir:
* Circuit .cir
V1 n0 0 SIN (0 10 1kHz)
C1 n1 n0 3.3 nF
R1 0 n1 1k
.end

16.13

Compatibility

ngspice is a direct derivative of spice3f5 from UC Berkeley and thus inherits all of the commands available in its predecessor. Thanks to the open source policy of UCB (original spice3
from 1994 is still available here), several commercial variants have sprung off, either being more
dedicated to IC design or more concentrating on simulating discrete and board level electronics.
None of the commercial and almost none of the freely downloadable SPICE providers publishes
the source code. All of them have proceeded with the development, by adding functionality, or
by adding a more dedicated user interface. Some have kept the original SPICE syntax for their
netlist description, others have quickly changed some if not many of the commands, functions
and procedures. Thus it is difficult, if not impossible, to offer a simulator that acknowledges
all of these netlist dialects. ngspice includes some features that enhance compatibility that are
included automatically. This selection may be controlled to some extend by setting the compatibility mode. Others may be invoked by the user by small additions to the netlist input file.
Some of them are listed in this chapter, some will be integrated into ngspice at a later stage,
others will be added if they are reported by users.

16.13.1

Compatibility mode

The variable (17.7) ngbehavior sets the compatibility mode. ’all’ is set as the default value.
’spice3’ as invoked by the command
set ngbehavior=spice3
in spinit or .spiceinit will disable some of the advanced ngspice features. ’ps’ will enable
including a library by a simple .lib  statement that is not compatible to the
more comfortable library handling described in Chapt. 2.7.

16.13.2

Missing functions

You may add one or more function definitions to your input file, as listed below.
.func
.func
.func
.func

LIMIT(x,a,b) {min(max(x, a), b)}
PWR(x,a) {abs(x) ** a}
PWRS(x,a) {sgn(x) * PWR(x,a)}
stp(x) {u(x)}

294

CHAPTER 16. STARTING NGSPICE

16.13.3

Devices

16.13.3.1

E Source with LAPLACE

see Chapt. 5.2.5.

16.13.3.2

VSwitch

The VSwitch
S1 2 3 11 0 SW
.MODEL SW VSWITCH(VON=5V VOFF=0V RON=0.1 ROFF=100K)
may become
a1 %v(11) %gd(2 3) sw
.MODEL SW aswitch(cntl_off=0.0 cntl_on=5.0 r_off=1e5
+ r_on=0.1 log=TRUE)
The XSPICE option has to be enabled.

16.13.4

Controls and commands

16.13.4.1

.lib

The ngspice .lib command (see 2.7) requires two parameters, a file name followed by a library
name. If no library name is given, the line
.lib filename
should be replaced by
.inc filename
Alternatively, the compatibility mode (16.13.1) may be set to ’ps’.

16.13.4.2

.step

Repeated analysis in ngspice if offered by a short script inside of a .control section (see
Chapt. 17.8.7) added to the input file. A simple application (multiple dc sweeps) is shown
below.

16.14. TESTS

295

Input file with parameter sweep
parameter sweep
* resistive divider , R1 swept from start_r to stop_r
* replaces .STEP R1 1k 10k 1k
R1 1 2 1k
R2 2 0 1k
VDD 1 0 DC 1
.dc VDD 0 1 .1
. control
let start_r = 1k
let stop_r = 10k
let delta_r = 1k
let r_act = start_r
* loop
while r_act le stop_r
alter r1 r_act
run
write dc - sweep .out v(2)
set appendwrite
let r_act = r_act + delta_r
end
plot dc1.v(2) dc2.v(2) dc3.v(2) dc4.v(2) dc5.v(2)
+ dc6.v(2) dc7.v(2) dc8.v(2) dc9.v(2) dc10.v(2)
.endc
.end

16.14

Tests

The ngspice distribution is accompanied by a suite of test input and output files, located in the
directory ngspice/tests. Originally this suite was meant to see if ngspice with all models was
made and installed properly. It is started by
$ make check
from within your compilation and development shell. A sequence of simulations is thus started,
its outputs compared to given output files by comparisons string by string. This feature is
momentarily used only to check for the BSIM3 model (11.2.9) and the XSPICE extension (12).
Several other input files located in directory ngspice/tests may serve as light-weight examples
for invoking devices and simple circuits.
Today’s very complex device models (BSIM4 (see 11.2.10), HiSIM (see 11.2.14) and others)
require a different strategy for verification. Under development for ngspice is the CMC Regression test by Colin McAndrew, which accompanies every new model. These tests cover a large

296

CHAPTER 16. STARTING NGSPICE

range of different DC, AC and noise simulations with different geometry ranges and operating
conditions and are more meaningful the transient simulations with their step size dependencies.
A major advantage is the scalability of the diff comparisons, which check for equality within a
given tolerance. A set of Perl modules cares for input, output and comparisons of the models.
Currently BSIM3, BSIM4, BSIMSOI4, HiSIM, and HiSIM_HV models implement the new
test. You may invoke it by running the command given above or by
$ make -i check 2>&1 | tee results
-i will make make to ignore any errors, tee will provide console output as well as printing to
file ’results’. Be aware that under MS Windows you will need the console binary (see 32.2.5)
to run the CMC tests, and you have to have Perl installed!

16.15

Reporting bugs and errors

Ngspice is a complex piece of software. The source code contains over 1500 files. Various
models and simulation procedures are provided, some of them not used and tested intensively.
Therefore errors may be found, some still evolving from the original spice3f5 code, others
introduced during the ongoing code enhancements.
If you happen to experience an error during the usage of ngspice, please send a report to the
development team. Ngspice is hosted on sourceforge, the preferred place to post a bug report is
the ngspice bug tracker. We would prefer to have your bug tested against the actual source code
available at Git, but of course a report using the most recent ngspice release is welcome! Please
provide the following information with your report:
Ngspice version
Operating system
Small input file to reproduce the bug
Actual output versus the expected output

Chapter 17
Interactive Interpreter
17.1

Introduction

The simulation flow in ngspice (input, simulation, output) may be controlled by dot commands
(see Chapt. 15 and 16.4.1) in batch mode. There is, however, a much more powerful control
scheme available in ngspice, traditionally coined ‘Interactive Interpreter’, but being much more
than just that. In fact there are several ways to use this feature, truly interactively by typing
commands to the input, but also running command sequences as scripts or as part of your input
deck in a quasi batch mode.
You may type in expressions, functions (17.2) or commands (17.5) into the input console to
elaborate on data already achieved from the interactive simulation session.
Sequences of commands, functions and control structures (17.6) may be assembled as a script
(17.8) into a file, and then activated by just typing the file name into the console input of an
interactive ngspice session.
Finally, and most useful, is it to add a script to the input file, in addition the the netlist and dot
commands. This is achieved by enclosing the script into .control ... .endc (see 16.4.3,
and 17.8.7 for an example). This feature enables a wealth of control options. You may set
internal (17.7) and other variables, start a simulation, evaluate the simulation output, start a new
simulation based on these data, and finally make use of many options for outputting the data
(graphically or into output files).
Historical note: The final releases of Berkeley Spice introduced a command shell and scripting
possibilities. The former releases were not interactive. The choice for the scripting language
was an early version of ‘csh’, the C-shell, which was en vogue back then as an improvement
over the ubiquitous Bourne Shell. Berkely Spice incorporated a modified csh source code that,
instead of invoking the unix ‘exec’ system call, executed internal SPICE C subroutines. Apart
from bug fixes, this is still how ngspice works.
The csh-like scripting language is active in .control sections. It works on ‘strings’, and does
string substitution of ‘environment’ variables. You see the csh at work in ngspice with set foo
= "bar"; set baz = "bar$foo", and in if, repeat, for, ... constructs. However, ngspice
processes mainly numerical data, and support for this was not available in the c-sh implementation. Therefore, Berkeley implemented an additional type of variables, with different syntax, to
access double and complex double vectors (possibly of length 1). This new variable type is modified with let, and can be used without special syntax in places where a numerical expression

297

298

CHAPTER 17. INTERACTIVE INTERPRETER

is expected: let bar = 4 * 5; let zoo = bar * 4 works. Unfortunately, occasionally
one has to cross the boundary between the numeric and the string domain. For this purpose the
$& construct is available – it queries a variable in the numerical let domain, and expands it to
a c-sh string denoting the value. This lets you do do something like set another = "this
is $&bar". It is important to remember that set can only operate on (c-sh) strings, and that
let operates only on numeric data. Convert from numeric to string with $&, and from string to
numeric with $.

17.2

Expressions, Functions, and Constants

Ngspice and ngnutmeg store data in the form of vectors: time, voltage, etc. Each vector has a
type, and vectors can be operated on and combined algebraically in ways consistent with their
types. Vectors are normally created as the output of a simulation, or when a data file (output raw
file) is read in again (ngspice, ngnutmeg, see the load command 17.5.38), or when the initial
data-file is loaded directly into ngnutmeg. They can also be created with the let command
817.5.35).
An expression is an algebraic formula involving vectors and scalars (a scalar is a vector of
length 1) and the following operations:

+ - * / ^ % ,

% is the modulo operator, and the comma operator has two meanings: if it is present in the
argument list of a user definable function, it serves to separate the arguments. Otherwise, the
term x , y is synonymous with x + j(y). Also available are the logical operations & (and),
| (or), ! (not), and the relational operations <, >, >=, <=, =, and <> (not equal). If used in an
algebraic expression they work like they would in C, producing values of 0 or 1. The relational
operators have the following synonyms:
Operator
gt
lt
ge
le
ne
and
or
not
eq

Synonym
>
<
>=
<=
<>
&
|
!
=

The operators are useful when < and > might be confused with the internal IO redirection (see
17.4, which is almost always happening). It is however safe to use < and > with the define
command (17.5.14).
The following functions are available:

17.2. EXPRESSIONS, FUNCTIONS, AND CONSTANTS
Name
mag(vector)
ph(vector)
cph(vector)
unwrap(vector)
j(vector)
real(vector
imag(vector)
db(vector)
log10(vector)
ln(vector)
exp(vector)
abs(vector)
sqrt(vector)
sin(vector)
cos(vector)
tan(vector)
atan(vector)
sinh(vector)
cosh(vector)
tanh(vector)
floor(vector)
ceil(vector)
norm(vector)
mean(vector)

avg(vector)

stddev(vector)
group_delay(vector)

vector(number)

unitvec(number)

Function
Magnitude of vector (same as abs(vector)).
Phase of vector.
Phase of vector. Continuous values, no discontinuity at ±π.
Phase of vector. Continuous values, no discontinuity at ±π.
Real phase vector in degrees as input.
i(sqrt(-1)) times vector.
The real component of vector.
The imaginary part of vector.
20 log10(mag(vector)).
The logarithm (base 10) of vector.
The natural logarithm (base e) of vector.
e to the vector power.
The absolute value of vector (same as mag).
The square root of vector.
The sine of vector.
The cosine of vector.
The tangent of vector.
The inverse tangent of vector.
The hyperbolic sine of vector.
The hyperbolic cosine of vector.
The hyperbolic tangent of vector.
Largest integer that is less than or equal to vector.
Smallest integer that is greater than or equal to vector.
The vector normalized to 1 (i.e, the largest magnitude of
any component is 1).
The result is a scalar (a length 1 vector) that is the mean of
the elements of vector (elements values added, divided by
number of elements).
The average of a vector.
Returns a vector where each element is the mean of the
preceding elements of the input vector (including the
actual element).
The result is a scalar (a length 1 vector) that is the standard
deviation of the elements of vector .
Calculates the group delay -dphase[rad]/dω[rad/s]. Input is
the complex vector of a system transfer function versus
frequency, resembling damping and phase per frequency
value. Output is a vector of group delay values (real values
of delay times) versus frequency.
The result is a vector of length number, with elements 0, 1,
... number - 1. If number is a vector then just the first
element is taken, and if it isn’t an integer then the floor of
the magnitude is used.
The result is a vector of length number, all elements having
a value 1.

299

300

CHAPTER 17. INTERACTIVE INTERPRETER

Name
length(vector)
interpolate(plot.vector)

deriv(vector)

vecd(vector)
vecmin(vector)
minimum(vector)
vecmax(vector)
maximum(vector)
fft(vector)
ifft(vector)
sortorder(vector)

timer(vector)
clock(vector)

Function
The length of vector.
The result of interpolating the named vector onto the scale
of the current plot. This function uses the variable
polydegree to determine the degree of interpolation.
Calculates the derivative of the given vector. This uses
numeric differentiation by interpolating a polynomial and
may not produce satisfactory results (particularly with
iterated differentiation). The implementation only
calculates the derivative with respect to the real component
of that vector’s scale.
Compute the differential of a vector.
Returns the value of the vector element with minimum
value. Same as minimum.
Returns the value of the vector element with minimum
value. Same as vecmin.
Returns the value of the vector element with maximum
value. Same as maximum.
Returns the value of the vector element with maximum
value. Same as vecmax.
fast fourier transform (17.5.26)
inverse fast fourier transform (17.5.26)
Returns a vector with the positions of the elements in a real
vector after they have been sorted into increasing order
using a stable method (qsort).
Returns CPU-time minus the value of the first vector
element.
Returns wall-time minus the value of the first vector
element.

Several functions offering statistical procedures are listed in the following table:

17.2. EXPRESSIONS, FUNCTIONS, AND CONSTANTS
Name
rnd(vector)

sgauss(vector)

sunif(vector)

poisson(vector)

exponential(vector)

301

Function
A vector with each component a random integer between 0
and the absolute value of the input vector’s corresponding
integer element value.
Returns a vector of random numbers drawn from a
Gaussian distribution (real value, mean = 0 , standard
deviation = 1). The length of the vector returned is
determined by the input vector. The contents of the input
vector will not be used. A call to sgauss(0) will return a
single value of a random number as a vector of length 1..
Returns a vector of random real numbers uniformly
distributed in the interval [-1 .. 1[. The length of the vector
returned is determined by the input vector. The contents of
the input vector will not be used. A call to sunif(0) will
return a single value of a random number as a vector of
length 1.
Returns a vector with its elements being integers drawn
from a Poisson distribution. The elements of the input
vector (real numbers) are the expected numbers λ.
Complex vectors are allowed, real and imaginary values
are treated separately.
Returns a vector with its elements (real numbers) drawn
from an exponential distribution. The elements of the input
vector are the respective mean values (real numbers).
Complex vectors are allowed, real and imaginary values
are treated separately.

An input vector may be either the name of a vector already defined or a floating-point number
(a scalar). A scalar will result in an output vector of length 1. A number may be written in
any format acceptable to ngspice, such as 14.6Meg or -1.231e-4. Note that you can either use
scientific notation or one of the abbreviations like MEG or G, but not both. As with ngspice, a
number may have trailing alphabetic characters.

The notation expr [num] denotes the num’th element of expr. For multi-dimensional vectors,
a vector of one less dimension is returned. Also for multi-dimensional vectors, the notation
expr[m][n] will return the nth element of the mth subvector. To get a subrange of a vector, use
the form expr[lower, upper]. To reference vectors in a plot that is not the current plot (see the
setplot command, below), the notation plotname.vecname can be used. Either a plotname or
a vector name may be the wildcard all. If the plotname is all, matching vectors from all plots
are specified, and if the vector name is all, all vectors in the specified plots are referenced. Note
that you may not use binary operations on expressions involving wildcards - it is not obvious
what all + all should denote, for instance. Thus some (contrived) examples of expressions are:

302

CHAPTER 17. INTERACTIVE INTERPRETER

Expressions examples:
cos(TIME) + db(v(3))
sin(cos(log ([1 2 3 4 5 6 7 8 9 10])))
TIME * rnd(v(9)) - 15 * cos(vin# branch ) ^ [7.9 e5 8]
not (( ac3.FREQ [32] & tran1.TIME [10]) gt 3)
(sunif (0) ge 0) ? 1.0 : 2.0
mag(fft(v (18)))
Vector names in ngspice may look like @dname[param], where dname is either the name of
a device instance or of a device model. The vector contains the value of the parameter of the
device or model. See Appendix, Chapt. 31 for details of which parameters are available. The
returned value is a vector of length 1. Please note that finding the value of device and device
model parameters can also be done with the show command (e.g. show v1 : dc).
There are a number of pre-defined constants in ngspice, which you may use by their name. They
are stored in plot (17.3) const and are listed in the table below:
Name
pi
e
c
i
kelvin
echarge
boltz
planck
yes
no
TRUE
FALSE

Description
π
e (the base of natural logarithms)
c (the speed of light)
i (the square root of -1)
(absolute zero in centigrade)
q (the charge of an electron)
k (Boltzmann’s constant)
h (Planck’s constant)
boolean
boolean
boolean
boolean

Value
3.14159...
2.71828...
m/sec
299,792,500
√
−1
-273.15◦C
1.60219e-19 C
1.38062e-23J/K
6.62620e-34
1
0
1
0

These constants are all given in MKS units. If you define another variable with a name that
conflicts with one of these then it takes precedence.
Additional constants may be generated during circuit setup (see .csparam, 2.10).

17.3

Plots

The output vectors of any analysis are stored in plots, a traditional SPICE notion. A plot is a
group of vectors. A first tran command will generate several vectors within a plot tran1. A
subsequent tran command will store their vectors in tran2. Then a linearize command will
linearize all vectors from tran2 and store them in tran3, which then becomes the current plot. A
fft will generate a plot spec1, again now the current plot. The display command always will
show all vectors in the current plot. Echo $plots followed by Return lists all plots generated
so far. Setplot followed by Return will show all plots and ask for a (new) plot to become
current. A simple Return will end the command. Setplot name will change the current plot
to ’name’ (e.g. setplot tran2 will make tran2 the current plot). A sequence name.vector
may be used to access the vector from a foreign plot.

17.4. COMMAND INTERPRETATION

303

You may generate plots by yourself: setplot new will generate a new plot named unknown1,
set curplottitle=”a new plot” will set a title, set curplotname=myplot will set its
name as a short description, set curplotdate=”Sat Aug 28 10:49:42 2010” will set its
date. Note that strings with spaces have to be given with double quotes.
Of course the notion ’plot’ will be used by this manual also in its more common meaning,
denoting a graphics plot or being a plot command. Be careful to get the correct meaning.

17.4

Command Interpretation

17.4.1

On the console

On the ngspice console window (or into the Windows GUI) you may directly type in any command from 17.5. Within a command sequence Input/output redirection is available (see Chapt.
17.8.8 for an example) - the symbols >, >>, >&, >>&, and < have the same effects as in the
C-shell. This I/O-redirection is internal to ngspice commands, and should not be mixed up with
the ‘external’ I/O-redirection offered by the usual shells (Linux, MSYS etc.), see 17.5.64. You
may type multiple commands on one line, separated by semicolons.

17.4.2

Scripts

If a word is typed as a command, and there is no built-in command with that name, the directories in the sourcepath list are searched in order for a file with the name given by the word. If
it is found, it is read in as a command file (as if it were sourced). Before it is read, however, the
variables argc and argv are set to the number of words following the file-name on the command line, and a list of those words respectively. After the file is finished, these variables are
unset. Note that if a command file calls another, it must save its argv and argc since they are
altered. Also, command files may not be re-entrant since there are no local variables. Of course,
the procedures may explicitly manipulate a stack.... This way one can write scripts analogous
to shell scripts for ngnutmeg and ngspice.
Note that for the script to work with ngspice, it must begin with a blank line (or whatever else,
since it is thrown away) and then a line with .control on it. This is an unfortunate result
of the source command being used for both circuit input and command file execution. Note
also that this allows the user to merely type the name of a circuit file as a command and it is
automatically run. The commands are executed immediately, without running any analyses that
may be specified in the circuit (to execute the analyses before the script executes, include a run
command in the script).
There are various command scripts installed in /usr/local/lib/spice/scripts (or whatever the path is on your machine), and the default sourcepath includes this directory, so you
can use these command files (almost) like built-in commands.

17.4.3

Add-on to circuit file

The problably most common way to invoke the commands described in the following Chapt.
17.5 is to add a .control ... .endc section to the circuit input file (see 16.4.3).

304

CHAPTER 17. INTERACTIVE INTERPRETER

Example:
. control
pre_set strict_errorhandling
unset ngdebug
*save outputs and specials
save x1.x1.x1 .7 V(9) V(10) V(11) V(12) V(13)
run
display
* plot the inputs , use offset to plot on top of each other
plot v(1) v (2)+4 v (3)+8 v (4)+12 v (5)+16 v (6)+20 v (7)+24 v (8)+28
* plot the outputs , use offset to plot on top of each other
plot v(9) v (10)+4 v (11)+8 v (12)+12 v (13)+16
.endc

17.5

Commands

Commands marked with a * are only available in ngspice, not in ngnutmeg.

17.5.1

Ac*: Perform an AC, small-signal frequency response analysis

General Form:
ac ( DEC | OCT | LIN ) N Fstart Fstop
Do an small signal ac analysis (see also Chapt. 15.3.1) over the specified frequency range.
DEC decade variation, and N is the number of points per decade.
OCT stands for octave variation, and N is the number of points per octave.
LIN stands for linear variation, and N is the number of points.
fstart is the starting frequency, and fstop is the final frequency.
Note that in order for this analysis to be meaningful, at least one independent source must have
been specified with an ac value.
In this ac analysis all non-linear devices are linearized around their actual dc operating point.
All Ls and Cs get their imaginary value, depending on the actual frequency step. Each output
vector will be calculated relative to the input voltage (current) given by the ac value (Iin equals
to 1 in the example below). The resulting node voltages (and branch currents) are complex
vectors. Therefore you have to be careful using the plot command.

17.5. COMMANDS

305

Example:
* AC test
Iin 1 0 AC 1
R1 1 2 100
L1 2 0 1
. control
AC LIN 101 10 10K
plot v(2)
$ real part !
plot mag(v(2)) $ magnitude
plot db(v(2))
$ same as vdb (2)
plot imag(v(2)) $ imaginary part of v(2)
plot real(v(2)) $ same as plot v(2)
plot phase (v(2)) $ phase in rad
plot cph(v(2)) $ phase in rad , continuous beyond pi
plot 180/ PI* phase (v(2)) $ phase in deg
.endc
.end
In addition to the plot examples given above you may use the variants of vxx(node) described in
Chapt. 15.6.2 like vdb(2). An option to suppress OP analysis before AC may be set for linear
circuits (15.1.3).

17.5.2

Alias: Create an alias for a command

General Form:
alias [word] [text ...]
Causes word to be aliased to text. History substitutions may be used, as in C-shell aliases.

17.5.3

Alter*: Change a device or model parameter

Alter changes the value for a device or a specified parameter of a device or model.
General Form:
alter dev = 
alter dev param = 
alter @dev[ param ] = 
 must be real (complex isn’t handled right now, integer is fine though, but no
strings. For booleans, use 0/1).

306

CHAPTER 17. INTERACTIVE INTERPRETER

Old style (pre 3f4):
alter device value
alter device parameter value [ parameter value ]
Using the old style, its first form is used by simple devices that have one principal value (resistors, capacitors, etc.) where the second form is for more complex devices (bjt’s, etc.). Model
parameters can be changed with the second form if the name contains a ‘#’. For specifying a
list of parameters as values, start it with ‘[’, followed by the values in the list, and end with ‘]’.
Be sure to place a space between each of the values and before and after the ‘[’ and ‘]’.
Some examples are given below:
Examples (Spice3f4 style):
alter
alter
alter
alter
alter

vd = 0.1
vg dc = 0.6
@m1[w]= 15e -06
@vg[sin] [ -1 1.5 2MEG ]
@Vi[pwl] = [ 0 1.2 100p 0 ]

alter may have vectors (17.8.2) or variables (17.8.1) as parameters.
Examples (vector or variable in parameter list):
let newfreq = 10k
alter @vg[sin] [ -1 1.5 $& newfreq ] $ vector
set newperiod = 150u
alter @Vi[pwl] = [ 0 1.2 $newperiod 0 ] $ variable
You may change a parameter of a device residing in a subcircuit, e.g. of MOS transistor msub1
in subcircuit xm1 (see also Chapt. 31.1).
Examples (parameter of device in subcircuit):
alter m.xm1. msub1 w = 20u
alter @m.xm1. msub1 [w] = 20u

17.5.4

Altermod*: Change model parameter(s)

General form:
altermod mod param = 
altermod @mod[ param ] = 
Example:
altermod nc1 tox = 10e-9
altermod @nc1[tox] = 10e-9

17.5. COMMANDS

307

Altermod operates on models and is used to change model parameters. The above example
will change the parameter tox in all devices using the model nc1, which is defined as
*** BSIM3v3 model
.MODEL nc1 nmos LEVEL=8 version = 3.2.2
+ acm = 2 mobmod = 1 capmod = 1 noimod = 1
+ rs = 2.84E+03 rd = 2.84E+03 rsh = 45
+ tox = 20E-9 xj = 0.25E-6 nch = 1.7E+17
+ ...

If you invoke the model by the MOS device
M1 d g s b nc1 w=10u l=1u
you might also insert the device name M1 for mod as in
altermod M1 tox = 10e-9
The model parameter tox will be modified, however not only for device M1, but for all devices
using the associated MOS model nc1!
If you want to run corner simulations within a single simulation flow, the following option of
altermod may be of help. The parameter set with name modn may be overrun by the altermod
command specifying a model file. All parameter values fitting to the existing model modn will
be modified. As usual the ’reset’ command (see 17.5.51) restores the original values. The model
file (see 2.3) has to use the standard specifications for an input file, the .model section is the
relevant part. However the first line in the model file will be ignored by the input parser, so it
should contain only some title information. The .model statement should appear then in the
second or any later line. More than one .model section may reside in the file.
General form:
altermod mod1 [mod2 .. mod15] file = 
altermod mod1 [mod2 .. mod15] file 
Example:
altermod nch file = BSIM3_nmos .mod
altermod pch nch file BSIM4_mos .mod
Be careful that the new model file corresponds to the existing model selected by modn. The existing models are defined during circuit setup at start up of ngspice. Models have been included
by .model statements (2.3) in your input file or included by the .include command. In the
example given above, the models nch (or nch and pch) have to be already available before calling altermod. If they are not found in the active circuit, ngspice will terminate with an error
message. There is no checking however of the version and level parameters! So you have to
be responsible for offering model data of the same model level (e.g. level 8 for BSIM3). Thus
no new model is selectable by altermod, but the parameters of the existing model(s) may be
changed (partially, completely, temporarily).

308

CHAPTER 17. INTERACTIVE INTERPRETER

17.5.5

Asciiplot: Plot values using old-style character plots

General Form:
asciiplot plotargs
Produce a line printer plot of the vectors. The plot is sent to the standard output, or you can
put it into a file with asciiplot args ... > file. The set options width, height, and nobreak
determine the width and height of the plot, and whether there are page breaks, respectively.
The ’more’ mode is the standard mode if printing to the screen, that is after a number of lines
given by height, and after a page break printing stops with request for answering the prompt
by , ’c’ or ’q’. If everything shall be printed without stopping, put the command set
nomoremode into .spiceinit 16.6 (or spinit 16.5). Note that you will have problems if you try
to asciiplot something with an X-scale that isn’t monotonic (i.e, something like sin(TIME)
), because asciiplot uses a simple-minded linear interpolation. The asciiplot command
doesn’t deal with log scales or the delta keywords.

17.5.6

Aspice*: Asynchronous ngspice run

General Form:
aspice input -file [output -file]
Start an ngspice run, and when it is finished load the resulting data. The raw data is kept in
a temporary file. If output-file is specified then the diagnostic output is directed into that file,
otherwise it is thrown away.

17.5.7

Bug: Mail a bug report

General Form:
bug
Send a bug report. Please include a short summary of the problem, the version number and
name of the operating system that you are running, the version of ngspice that you are running,
and the relevant ngspice input file. (If you have defined BUGADDR, the mail is delivered to there.)

17.5.8

Cd: Change directory

General Form:
cd [ directory ]
Change the current working directory to directory, or to the user’s home directory if none is
given.

17.5. COMMANDS

17.5.9

309

Cdump: Dump the control flow to the screen

General Form:
cdump
Dumps the control sequence to the screen (all statements inside the .control ... .endc structure before the line with cdump). Indentations show the structure of the sequence. The example
below is printed if you add cdump to /examples/Monte_Carlo/MonteCarlo.sp.
Example (abbreviated):
let mc_runs =5
let run =0
...
define agauss (nom , avar , sig) (nom + avar/sig * sgauss (0))
define limit (nom , avar) (nom + (( sgauss (0) >=0) ? avar : -avar ))
dowhile run < mc_runs
alter c1=unif (1e -09 , 0.1)
...
ac oct 100 250k 10 meg
meas ac bw trig vdb(out) val =-10 rise =1 targ vdb(out)
+ val = -10 fall =1
set run ="$&run"
...
let run=run + 1
end
plot db ({ $scratch }. allv)
echo
print { $scratch }. bwh
cdump

17.5.10

Circbyline*: Enter a circuit line by line

General Form:
circbyline line
Enter a circuit line by line. line is any circuit line, as found in the *.cir ngspice input files. The
first line is a title line. The entry will be finished by entering .end. Circuit parsing is then
started automatically.

310

CHAPTER 17. INTERACTIVE INTERPRETER

Example:
circbyline
circbyline
circbyline
circbyline
circbyline
run
plot i(v1)

17.5.11

test circuit
v1 1 0 1
r1 1 0 1
.dc v1 0.5 1.5 0.1
.end

Codemodel*: Load an XSPICE code model library

General Form:
codemodel [ library file]
Load a XSPICE code model shared library file (e.g. analog.cm ...). Only available if ngspice
is compiled with the XSPICE option (--enable-xspice) or with the Windows executable
distributed since ngspice21. This command has to be called from spinit (see Chapt. 16.5) (or
.spiceinit for personal code models, 16.6).

17.5.12

Compose: Compose a vector

General Form:
compose name values value1 [ value2 ... ]
compose name param = val [ param = val ... ]
The first form takes the values and creates a new vector, where the values may be arbitrary
expressions.
The second form has the following possible parameters:
start
stop
step
lin
log
dec
center
span
gauss
sd
sigma
random
rvar

The value of name[0]
The last value of name
The difference between successive elements of the created vector
How many linearly spaced elements the new vector should have
The number of points, logarithmically spaced (not working)
The number of points per decade, logarithmically spaced (not working)
Where to center the range of points (not working)
The size of the range of points (not working)
The nominal value for the used gaussian distribution
The standard deviation for the used gaussian distribution
The sigma for the used gaussian distribution
The nominal value for a uniform random distribution
The percentage variation for the uniform random distribution

17.5. COMMANDS

17.5.13

311

Dc*: Perform a DC-sweep analysis

General Form:
dc Source Vstart Vstop Vincr [ Source2 Vstart2 Vstop2 Vincr2 ]
Do a dc transfer curve analysis. See the previous Chapt. 15.3.2 for more details. Several options
may be set (15.1.2).

17.5.14

Define: Define a function

General Form:
define function (arg1 , arg2 , ...) expression
Define the function with the name function and arguments arg1, arg2, ... to be expression,
which may involve the arguments. When the function is later used, the arguments it is given
are substituted for the formal arguments when it was parsed. If expression is not present, any
existing definition for function is printed, and if there are no arguments then all expressions for
all currently active definitions are printed. Note that you may have different functions defined
with the same name but different arities. Some useful definitions are:
Example:
define max(x,y) (x > y) * x + (x <= y) * y
define min(x,y) (x < y) * x + (x >= y) * y
define limit (nom , avar) (nom + (( sgauss (0) >= 0) ? avar : -avar ))

17.5.15

Deftype: Define a new type for a vector or plot

General Form:
deftype [v | p] typename abbrev
defines types for vectors and plots. abbrev will be used to parse things like abbrev(name) and
to label axes with M, instead of numbers. It may be omitted. Also, the command
‘deftype p plottype pattern ...’ will assign plottype as the name to any plot with one of the
patterns in its Name: field.
Example:
deftype v capacitance F
settype capacitance moscap
plot moscap vs v(cc)

312

CHAPTER 17. INTERACTIVE INTERPRETER

17.5.16

Delete*: Remove a trace or breakpoint

General Form:
delete [ debug - number ... ]
Delete the specified saved nodes and parameters, breakpoints and traces. The debug numbers
are those shown by the status command (unless you do status > file, in which case the debug
numbers are not printed).

17.5.17

Destroy: Delete an output data set

General Form:
destroy [ plotnames | all]
Release the memory holding the output data (the given plot or all plots) for the specified runs.

17.5.18

Devhelp: information on available devices

General Form:
devhelp [[- csv] device_name [ parameter ]]
Devhelp command shows the user information about the devices available in the simulator. If
called without arguments, it simply displays the list of available devices in the simulator. The
name of the device is the name used inside the simulator to access that device. If the user specifies a device name, then all the parameters of that device (model and instance parameters)
will be printed. Parameter description includes the internal ID of the parameter (id#), the name
used in the model card or on the instance line (Name), the direction (Dir) and the description
of the parameter (Description). All the fields are self-explanatory, except the ‘direction’. Direction can be in, out or inout and corresponds to a ‘write-only’, ‘read-only’ or a ‘read/write’
parameter. Read-only parameters can be read but not set, write only can be set but not read and
read/write can be both set and read by the user.
The -csv option prints the fields separated by a comma, for direct import into a spreadsheet.
This option is used to generate the simulator documentation.
Example:
devhelp
devhelp resistor
devhelp capacitor ic

17.5. COMMANDS

17.5.19

313

Diff: Compare vectors

General Form:
diff plot1 plot2 [vec ...]
Compare all the vectors in the specified plots, or only the named vectors if any are given. If
there are different vectors in the two plots, or any values in the vectors differ significantly,
the difference is reported. The variables diff_abstol, diff_reltol, and diff_vntol are used to
determine a significant difference.

17.5.20

Display: List known vectors and types

General Form:
display [ varname ...]
Prints a summary of currently defined vectors, or of the names specified. The vectors are sorted
by name unless the variable nosort is set. The information given is the name of the vector, the
length, the type of the vector, and whether it is real or complex data. Additionally, one vector
is labeled [scale]. When a command such as plot is given without a vs argument, this scale is
used for the X-axis. It is always the first vector in a rawfile, or the first vector defined in a new
plot. If you undefine the scale (i.e, let TIME = []), one of the remaining vectors becomes the
new scale (which one is unpredictable). You may set the scale to another vector of the plot with
the command setscale (17.5.62).

17.5.21

Echo: Print text

General Form:
echo [text ...] [ $variable ] ["$& vector "]
Echos the given text, variable or vector to the screen. echo without parameters issues a blank
line.

17.5.22

Edit*: Edit the current circuit

General Form:
edit [ file -name ]
Print the current ngspice input file into a file, call up the editor on that file and allow the user to
modify it, and then read it back in, replacing the original file. If a file-name is given, then edit
that file and load it, making the circuit the current one. The editor may be defined in .spiceinit
or spinit by a command line like
set editor=emacs

314

CHAPTER 17. INTERACTIVE INTERPRETER

Using MS Windows, to allow the edit command calling an editor, you will have to add the
editor’s path to the PATH variable of the command prompt windows (see here). edit then calls
cmd.exe with e.g. notepad++ and file-name as parameter, if you have set
set editor=notepad++.exe
in .spiceinit or spinit.

17.5.23

Edisplay: Print a list of all the event nodes

General Form:
edisplay
Print the names of all event driven nodes generated or used by XSPICE ’A’ devices. See
eprint, eprvcd, and 27.2.2 for an example.

17.5.24

Eprint: Print an event driven node

General Form:
eprint node [node]
eprint node [node] > nodeout .txt $ output redirected
Print an event driven node generated or used by an XSPICE ’A’ device. These nodes are vectors
not organized in plots. See edisplay, eprvcd, and Chapt. 27.2.2 for an example. Output
redirection into a file is available.

17.5.25

Eprvcd: Dump event nodes in VCD format

General Form:
eprvcd node1 node2 .. noden [ > filename ]
Dump the data of the specified event driven nodes to a .vcd file. Such files may be viewed
with an vcd viewer, for example ’gtkwave.’ See edisplay, eprint, eprvcd, and 27.2.2 for
an example.

17.5.26

FFT: fast Fourier transform of vectors

General Form:
fft vector1 [ vector2 ] ...
This analysis provides a fast Fourier transform of the input vector(s) in forward direction. fft
is much faster than spec (17.5.71) (about a factor of 50 to 100 for larger vectors).

17.5. COMMANDS

315

The fft command will create a new plot consisting of the Fourier transforms of the vectors
given on the command line. Each vector given should be a transient analysis result, i.e. it
should have time as a scale. You will have got these vectors by the tran Tstep Tstop
Tstart command.
The vector should have a linear equidistant time scale. Therefore linearization using the linearize
command is recommended before running fft. Be careful selecting a Tstep value small
enough for good interpolation, e.g. much smaller than any signal period to be resolved by fft
(see linearize command). The Fast Fourier Transform will be computed using a window
function as given with the specwindow variable. A new plot named specx will be generated
with a new vector (having the same name as the input vector, see command above) containing
the transformed data.
Ngspice has two FFT implementations:
1. Standard code is based on the FFT function provided by John Green ‘FFTs for RISC 2.0‘,
downloaded 2012, now to be found here. These are a power-of-two routines for fft and
ifft. If the input size doesn’t fit this requirement the remaining data will be zero padded
upto the next 2N field size. You have to take care of the correlated change in the scale
vector.
2. If available on the operating system (see Chapter 32) ngspice can be linked to the famous
FFTW-3 package, found here. This high performance package has advantages in speed
and accuracy compared to most of the freely available FFT libraries. It makes arbitrary
size transforms for even and odd data.
How to compute the fft from a transient simulation output:
ngspice
ngspice
ngspice
ngspice
ngspice

8 -> setplot tran1
9 -> linearize V(2)
9 -> set specwindow = blackman
10 -> fft V(2)
11 -> plot mag(V(2))

Linearize will create a new vector V(2) in a new plot tran2. The command fft V(2) will
create a new plot spec1 with vector V(2) holding the resulting data.
The variables listed in the following table control operation of the fft command. Each can be
set with the set command before calling fft.
specwindow: This variable is set to one of the following strings, which will determine the
type of windowing used for the Fourier transform in the spec and fft command. If not set, the
default is hanning.
none No windowing
rectangular Rectangular window
bartlet Bartlett (also triangle) window
blackman Blackman window

316

CHAPTER 17. INTERACTIVE INTERPRETER

hanning Hanning (also hann or cosine) window
hamming Hamming window
gaussian Gaussian window
flattop Flat top window

Figure 17.1: Spec and FFT window functions (Gaussian order = 4)
specwindoworder: This can be set to an integer in the range 2-8. This sets the order when
the Gaussian window is used in the spec and fft commands. If not set, order 2 is used.

17.5.27

Fourier: Perform a Fourier transform

General Form:
fourier fundamental_frequency [ expression ...]
Fourier is used to analyse the output vector(s) of a preceeding transient analysis (see 17.5.79).
It does a Fourier analysis of each of the given values, using the first 10 multiples of the fundamental frequency (or the first nfreqs multiples, if that variable is set - see 17.7). The printed
output is like that of the .four ngspice line (Chapt. 15.6.4). The expressions may be any valid
expression (see 17.2), e.g. v(2). The evaluated expression values are interpolated onto a fixedspace grid with the number of points given by the fourgridsize variable, or 200 if it is not set.
The interpolation is of degree polydegree if that variable is set, or 1. If polydegree is 0, then no
interpolation is done. This is likely to give erroneous results if the time scale is not monotonic,
though.
The fourier command not only issues a printout, but also generates vectors, one per expression.
The size of the vector is 3 x nfreqs (per default 3 x 10). The name of the new vector is fouriermn,

17.5. COMMANDS

317

where m is set by the mth call to the fourier command, n is the nth expression given in the actual
fourier command. fouriermn[0] is the vector of the 10 (nfreqs) frequency values, fouriermn[1]
contains the 10 (nfreqs) magnitude values, fouriermn[2] the 10 (nfreqs) phase values of the
result.
Example:
* do the transient analysis
tran 1n 1m
* do the fourier analysis
fourier 3.34 e6 v(2) v(3) $ first call
fourier 100 e6 v(2) v(3) $ second call
* get individual values
let newt1 = fourier11 [0][1]
let newt2 = fourier11 [1][1]
let newt3 = fourier11 [2][1]
let newt4 = fourier12 [0][4]
let newt5 = fourier12 [1][4]
let newt6 = fourier12 [2][4]
* plot magnitude of second expression (v(3))
* from first call versus frequency
plot fourier12 [1] vs fourier12 [0]
The plot command from the example plots the vector of the magnitude values, obtained by
the first call to fourier and evaluating the first expression in this call, against the vector of the
frequency values.

17.5.28

Gnuplot: Graphics output via gnuplot

General Form:
gnuplot file plotargs
Like plot, but using for graphics output and further data manipulation. ngspice creates a file
called file.plt containing the command sequence, a file called file.data containing the data to be
plotted, and a file called either file.eps (Postscript, this is the default) or file.png (the compressed binary png format, when the variable gnuplot_terminal is set to png). It is possible to
suppress the latter hardcopy file by using a file name that starts with ’np_’. On Linux gnuplot
is called via xterm, and offers a console to manipulate the data. On Windows a plot window is
opened and the command console window is available with a mouse click. Of course you have
to have gnuplot installed on your system.

17.5.29

Hardcopy: Save a plot to a file for printing

General Form:
hardcopy file plotargs

318

CHAPTER 17. INTERACTIVE INTERPRETER

Just like plot, except that it creates a file called file containing the plot. The file is a postscript
image. As an alternative the plot(5) format is available by setting the hcopydevtype variable
to plot5, and can be printed by either the plot(1) program or lpr with the -g flag. See also
Chapt. 18.6 for more details (color etc.).

17.5.30

Help: Print summaries of Ngspice commands

Prints help. This help information, however, is spice3f5-like, stemming from 1991 and thus
is outdated. If commands are given, descriptions of those commands are printed. Otherwise
help for only a few major commands is printed. On Windows this help command is no longer
available. Spice3f5 compatible help may be found in the Spice 3 User manual. For ngspice
please use this manual.

17.5.31

History: Review previous commands

General Form:
history [-r] [ number ]
Print out the history, or the last (first if -r is specified) number commands typed at the keyboard.
A history substitution enables you to reuse a portion of a previous command as you type the
current command. History substitutions save typing. A history substitution normally starts
with a ’!’. A history substitution has three parts: an event that specifies a previous command,
a selector that selects one or more words of the event, and some modifiers that modify the
selected words. The selector and modifiers are optional. A history substitution has the form
![event][:]selector[:modifier] . . . ] The event is required unless it is followed by a
selector that does not start with a digit. The ’:’ can be omitted before the selector if this
selector does not begin with a digit. History substitutions are interpreted before anything else
 even before quotations and command substitutions. The only way to quote the ’!’ of a
history substitution is to escape it with a preceding backslash. A ’!’ need not be escaped if it
is followed by whitespace, ’=’, or ’(’.
Ngspice saves each command that you type on a history list, provided that the command contains at least one word. The commands on the history list are called events. The events are
numbered, with the first command that you issue when you start Ngspice being number one.
The history variable specifies how many events are retained on the history list.
These are the forms of an event in a history substitution:
!!
!n
!-n
!str
!?str[?]

The preceding event. Typing ’!!’ is an easy way to reissue the previous command.
Event number n.
The nth previous event. For example, !-1 refers to the immediately preceding event and
is equivalent to !!.
The unique previous event whose name starts with str.
The unique previous event containing the string str. The closing ’?’ can be omitted if it
is followed by a newline.

17.5. COMMANDS

319

You can modify the words of an event by attaching one or more modifiers. Each modifier must
be preceded by a colon. The following modifiers assume that the first selected word is a file
name:
:r
:h
:t
:e
:p
s/old/new

&
g a
G

Removes the trailing .str extension from the first selected word.
Removes a trailing path name component from the first selected word.
Removes all leading path name components from the first selected word.
Remove all but the trailing suffix.
Print the new command but do not execute it.
Substitute new for the first occurrence of old in the event line. Any delimiter may be
used in place of ‘/’. The delimiter may be quoted in old and new with a single backslash.
If ‘&’ appears in new, it is replaced by old. A single backslash will quote the ‘&’. The
final delimiter is optional if it is the last character on the input line.
Repeat the previous substitution.
Cause changes to be applied over the entire event line. Used in conjunction with ‘s’, as
in gs/old/new/, or with ‘&’.
Apply the following ‘s’ modifier once to each word in the event.

For example, if the command ls /usr/elsa/toys.txt has just been executed, then the command
echo !!^:r !!^:h !!^:t !!^:t:r produces the output /usr/elsa/toys /usr/elsa toys.txt toys . The ’^’
command is explained in the table below.
You can select a subset of the words of an event by attaching a selector to the event. A history
substitution without a selector includes all of the words of the event. These are the possible
selectors for selecting words of the event:
:0
[:]^
[:]$
:n
:n1-n2
[:]*
:x*
:x[:]-x
[:]%

The command name
The first argument
The last argument
The nth argument (n ≥ 1)
Words n1 through n2
Words 1 through $
Words x through $
Words x through ($ - 1)
Words 0 through x
The word matched by the last ?str? search used

The colon preceding a selector can be omitted if the selector does not start with a digit.
The following additional special conventions provide abbreviations for commonly used forms
of history substitution:
• An event specification can be omitted from a history substitution if it is followed by a
selector that does not start with a digit. In this case the event is taken to be the event
used in the most recent history reference on the same line if there is one, or the preceding
event otherwise. For example, the command echo !?qucs?^ !$ echoes the first and last
arguments of the most recent command containing the string qucs .
• If the first nonblank character of an input line is ’^’, the ’^’ is taken as an abbreviation
for !:s^ . This form provides a convenient way to correct a simple spelling error in the

320

CHAPTER 17. INTERACTIVE INTERPRETER
previous line. For example, if by mistake you typed the command cat /etc/lasswd you
could re-execute the command with lasswd changed to passwd by typing ^l^p .
• You can enclose a history substitution in braces to prevent it from absorbing the following
characters. In this case the entire substitution except for the starting ’!’ must be within
the braces. For example, suppose that you previously issued the command cp accounts
../money . Then the command !cps looks for a previous command starting with cps
while the command !{cp}s turns into a command cp accounts ../moneys .

Some characters are handled specially as follows:
~
*
?
[]
^

Expands to the home directory
Matches any string of characters in a filename
Matches any single character in a filename
Matches any of the characters enclosed in a filename
Used within [] to specify a range of characters. For example, [b-k] matches on any
character between and including ‘b’ through to ‘k’.
If the ^ is included within [] as the first character, then it negates the following characters
matching on anything but those. For example, [^agm] would match on anything other
than ‘a’, ‘g’ and ‘m’. [^a-zA-Z] would match on anything other than an alphabetic
character.

The wildcard characters *, ?, [, and ] can be used, but only if you unset noglob first. This
makes them rather useless for typing algebraic expressions, so you should set noglob again
after you are done with wildcard expansion.
When the environment variable HOME exists (on Unix, Linux, or CYGWIN), history permanently stores previous command lines in the file $HOME/._ngspice_history. When this
variable does not exist (typically on Windows when the readline library is not officially installed), the history file is called .history and put in the current working directory.
The history command is part of the readline or editline package. The readline program provides a command line editor that is configurable through the file .inputrc. The path to this
configuration file is either found in the shell variable INPUTRC, or it is (on Unix/Linux/CYGWIN) the file ~/.inputrc in the user’s home directory. On Windows systems the configuration
file is /Users//.inputrc, unless the readline library was officially installed. In that
case the filename is taken from the Windows registry and points to a location that the user specified during installation. See https://cnswww.cns.cwru.edu/php/chet/readline/rltop.html for
detailed documentation. Some useful commands are:
Left/Right arrow
Home/End
Up/Down arrow
C-_C-r
TAB
C-ak
C-y
C-u

Move one character to the left or right
Move to beginning or end of line
Cycle through the history buffer
Undo last editing command
Incremental search backward
completion of a file name
Erase the command line (kill)
Retrieve last kill (yank)
Erase from cursor to start of line

17.5. COMMANDS

17.5.32

321

Inventory: Print circuit inventory

General Form:
inventory
This commands accepts no argument and simply prints the number of instances of a particular
device in a loaded netlist.

17.5.33

Iplot*: Incremental plot

General Form:
iplot [ node ...]
Incrementally plot the values of the nodes while ngspice runs. The iplot command can be used
with the where command to find trouble spots in a transient simulation.
The @name[param] notation (31.1) might not work yet.

17.5.34

Jobs*: List active asynchronous ngspice runs

General Form:
jobs
Report on the asynchronous ngspice jobs currently running. Ngnutmeg checks to see if the
jobs are finished every time you execute a command. If it is done then the data is loaded and
becomes available.

17.5.35

Let: Assign a value to a vector

General Form:
let name = expr
Creates a new vector called name with the value specified by expr, an expression as described
above. If expr is [] (a zero-length vector) then the vector becomes undefined. Individual elements of a vector may be modified by appending a subscript to name (ex. name[0]). If there are
no arguments, let is the same as display.
The command let creates a vector in the current plot. Use setplot (17.5.61) to create a new plot.
There is no straightforward way to initialize a new vector. In general, one might want let
initialize a slice (i.e. name[4:4,21:23] = [ 1 2 3 ]) of a multi-dimensional matrix of arbitrary
type (i.e. real, complex ..), where all values and indexes are arbitrary expressions. This will
fail. The procedure is to first allocate a real vector of the appropriate size with either vector(),
unitvec(), or [ n1 n2 n3 ... ]. The second step is to optionally change the type of the

322

CHAPTER 17. INTERACTIVE INTERPRETER

new vector (to complex) with the j() function. The third step reshapes the dimensions, and
the final step (re)initializes the contents, like so:
let a = j(vector(10))
reshape a [2][5]
let a[0][0] = (pi,pi)
Initialization of real vectors can be done quite efficiently with compose:
compose a values (pi, pi) (1,1) (2,sqrt(7)) (boltz,e)
reshape a [2][2]
See also unlet (17.5.83), compose (17.5.12).

17.5.36

Linearize*: Interpolate to a linear scale

General Form:
linearize vec ...
Create a new plot with all of the vectors in the current plot, or only those mentioned as arguments to the command, all data linearized onto an equidistant time scale.
How to compute the fft from a transient simulation output:
ngspice
ngspice
ngspice
ngspice
ngspice

8 -> setplot tran1
9 -> linearize V(2)
9 -> set specwindow = blackman
10 -> fft V(2)
11 -> plot mag(V(2)) tstep

Linearize will redo the vectors vec or renew all vectors of the current plot (e.g. tran3) if no
arguments are given and store them into a new plot (e.g. tran4). The new vectors are interpolated
onto a linear time scale, which is determined by the values of tstep, tstart, and tstop in
the currently active transient analysis. The currently loaded input file must include a transient
analysis (a tran command may be run interactively before the last reset, alternately), and
the current plot must be from this transient analysis. The length of the new vector is (tstop
- tstart) / tstep + 1.5. This command is needed for example if you want to do a fft
analysis (17.5.26). Please note that the parameter tstep of your transient analysis (see Chapt.
15.3.9) has to be small enough to get adequate resolution, otherwise the command linearize
will do sub-sampling of your signal. If no circuit is loaded and the data have been aquired by the
load (17.5.38) command, Linearize will take time data from transient analysis scale vector.

17.5. COMMANDS

17.5.37

323

Listing*: Print a listing of the current circuit

General Form:
listing [ logical ] [ physical ] [deck] [ expand ] [param]
If the logical argument is given, the listing is with all continuation lines collapsed into one line,
and if the physical argument is given the lines are printed out as they were found in the file. The
default is logical. A deck listing is just like the physical listing, except without the line numbers
it recreates the input file verbatim (except that it does not preserve case). If the word expand is
present, the circuit is printed with all subcircuits expanded. The option param allows to print
all parameters and their actual values.

17.5.38

Load: Load rawfile data

General Form:
load [ filename ] ...
Loads either binary or ascii format rawfile data from the files named. The default file-name is
rawspice.raw, or the argument to the -r flag if there was one.

17.5.39

Meas*: Measurements on simulation data

General Form (example):
MEAS {DC|AC|TRAN|SP} result TRIG trig_variable VAL=val 
  
 TARG targ_variable VAL=val 
 
 
Most of the input forms found in 15.4 may be used here with the command meas instead of
.meas(ure). Using meas inside the .control ... .endc section offers additional features
compared to the .meas use. meas will print the results as usual, but in addition will store
its measurement result (typically the token result given in the command line) in a vector.
This vector may be used in following command lines of the script as an input value of another
command. For details of the command see Chapt. 15.4. The measurement type SP is only
available here, because a fft command will prepare the data for SP measurement. Option
autostop (15.1.4) is not available.
Unfortunately par(’expression’) (15.6.6) will not work here, i.e. inside the .control section.
You may use an expression by the let command instead, giving let vec_new = expression.

324

CHAPTER 17. INTERACTIVE INTERPRETER

Replacement for par(’expression’) in meas inside the .control section
let vdiff = v(n1)-v(n0)
meas tran vtest find vdiff at =0.04e-3
*the following will not do here:
*meas tran vtest find par(’v(n1)-v(n0)’) at =0.04e-3

17.5.40

Mdump*: Dump the matrix values to a file (or to console)

General Form:
mdump 
If  is given, the output will be stored in file , otherwise dumped to
your console.

17.5.41

Mrdump*: Dump the matrix right hand side values to a file (or
to console)

General Form:
mrdump 
If  is given, the output will be appended to file , otherwise dumped to
your console.
Example usage after ngspice has started:
* Dump matrix and RHS values after 10 and 20 steps
* of a transient simulation
source rc.cir
step 10
mdump m1.txt
mrdump mr1.txt
step 10
mdump m2.txt
mrdump mr2.txt
* just to continue to the end
step 10000
You may create a loop using the control structures (Chapt. 17.6).

17.5.42

Noise*: Noise analysis

See the .NOISE analysis (15.3.4) for details.

17.5. COMMANDS

325

The noise command will generate two plots (typically named noise1 and noise2) with Noise
Spectral Density Curves and Integrated Noise data. To write these data into output file(s), you
may use the following command sequence:
Command sequence for writing noise data to file(s):
. control
tran 1e -6 1e -3
write test_tran .raw
noise V(out) vinp dec 333 1 1e8 16
print inoise_total onoise_total
*first option to get all of the output (two files)
setplot noise1
write test_noise1 .raw all
setplot noise2
write test_noise2 .raw all
* second option (all in one raw -file)
write testall .raw noise1 .all noise2 .all
.endc

17.5.43

Op*: Perform an operating point analysis

General Form:
op
Do an operating point analysis. See Chapt. 15.3.5 for more details.

17.5.44

Option*: Set a ngspice option

General Form:
option [ option =val] [ option =val] ...
Set any of the simulator variables as listed in Chapt. 15.1. See this chapter also for more
information on the available options. The option command without any argument lists the
actual options set in the simulator (to be verified). Multiple options may be set in a single line.
The following example demonstrates a control section, which may be added to your circuit file
to test the influence of variable trtol on the number of iterations and on the simulation time.

326

CHAPTER 17. INTERACTIVE INTERPRETER

Command sequence for testing option trtol:
. control
set noinit
option trtol =1
echo
echo trtol =1
run
rusage traniter trantime
reset
option trtol =3
echo
echo trtol =3
run
rusage traniter trantime
reset
option trtol =5
echo
echo trtol =5
run
rusage traniter trantime
reset
option trtol =7
echo
echo trtol =7
run
rusage traniter trantime
plot tran1 .v( out25 ) tran1.v(out50) v(out25)
.endc

17.5.45

v(out50)

Plot: Plot vectors on the display

General Form:
plot exprs [ ylimit ylo yhi] [ xlimit xlo xhi] [ xindices xilo xihi]
[ xcompress comp] [ xdelta xdel] [ ydelta ydel]
[xlog] [ylog] [ loglog ] [ nogrid ] [vs xname_expr ]
[ linplot ] [ combplot ] [ pointplot ] [ nointerp ]
[ xlabel word] [ ylabel word] [title word] [samep] [ linear ]
Plot the given vectors or exprs on the screen (if you are on a graphics terminal). The xlimit
and ylimit arguments determine the high and low x- and y-limits of the axes, respectively. The
xindices arguments determine what range of points are to be plotted - everything between the
xilo’th point and the xihi’th point is plotted. The xcompress argument specifies that only
one out of every comp points should be plotted. If an xdelta or a ydelta parameter is present,
it specifies the spacing between grid lines on the X- and Y-axis. These parameter names may

17.5. COMMANDS

327

be abbreviated to xl, yl, xind, xcomp, xdel, and ydel respectively.
The xname_expr argument is an expression to use as the scale on the x-axis. If xlog or ylog are
present then the X or Y scale, respectively, are logarithmic (loglog is the same as specifying
both). The xlabel and ylabel arguments cause the specified labels to be used for the X and
Y axes, respectively.
If samep is given, the values of the other parameters (other than xname_expr) from the previous
plot, hardcopy, or asciiplot command are used unless re-defined on the command line.
The title argument is used in the headline of the plot window and replaces the default text,
which is ‘actual plot: first line of input file’.
The linear keyword is used to override a default logscale plot (as in the output for an AC
analysis).
The keywords linplot, combplot and pointplot select different plot styles. The keyword
nointerp turns of interpolation of the vector data, nogrid suppresses the drawing of gridlines.
Finally, the keyword polar generates a polar plot. To produce a smith plot, use the keyword
smith. Note that the data is transformed, so for smith plots you will see the data transformed
by the function (x-1)/(x+1). To produce a polar plot with a smith grid but without performing
the smith transform, use the keyword smithgrid.
If you specify plot all, all vectors (including the scale vector) are plotted versus the scale
vector (see commands display (17.5.20) or setscale (17.5.62) on viewing the vectors of the
current plot). The command plot ally will not plot the scale vector, but all other ’real’ y
values. The command plot alli selects all current vectors, the command plot allv all
voltage vectors.
If the vector name to be plotted contains - , / or other tokens that may be taken for operators of an expression, and plotting fails, try enclosing the name in double quotes, e.g. plot
“/vout”.
Plotting of complex vectors, as may occur after an ac simulation, requires special considerations. Please see Chapt. 17.5.1 for details.

17.5.46

Pre_: execute commands prior to parsing the circuit

General Form:
pre_ 
All commands in a .control ... .endc section are executed after the circuit has been parsed.
If you need command execution before circuit parsing, you may add these commands to the
general spinit or local .spiceinit files. Another possibility is adding a leading pre_ to a command within the .control section of an ordinary input file, which forces the command to be
executed before circuit parsing. Basically  may be any command listed in Chapt.
17.5, however only a few commands are indeed useful here. Some examples are given below:

328

CHAPTER 17. INTERACTIVE INTERPRETER

Examples:
pre_unset ngdebug
pre_set strict_errorhandling
pre_codemodel mymod .cm
pre_ is available only in the .control mode (see 16.4.3), not in interactive mode,
where the user may determine herself when a circuit is to be parsed, using the source command
(17.5.70) .

17.5.47

Print: Print values

General Form:
print [col] [line] expr ...
Prints the vector(s) described by the expression expr. If the col argument is present, print the
vectors named side by side. If line is given, the vectors are printed horizontally. col is the
default, unless all the vectors named have a length of one, in which case line is the default.
The options width (default 80) and height (default 24) are effective for this command (see
asciiplot 17.5.5). The ’more’ mode is the standard mode if printing to the screen, that is after
a number of lines given by height, and after a page break printing stops with request for answering the prompt by  (print next page), ’c’ (print rest) or ’q’ (quit printing). If everything
shall be printed without stopping, put the command set nomoremode into .spiceinit 16.6 (or
spinit 16.5). If the expression is all, all of the vectors available are printed. Thus print col
all > filename prints everything into the file filename in SPICE2 format. The scale vector
(time, frequency) is always in the first column unless the variable noprintscale is true. You
may use the vectors alli, allv, ally with the print command, but then the scale vector
will not be printed.
Examples:
print all
set width =300
print v(1) > outfile .out

17.5.48

Quit: Leave Ngspice or Nutmeg

General Form:
quit
quit [ exitcode ]
Quit ngnutmeg or ngspice. Ngspice will ask for an acknowledgment if parameters have not
been saved. If unset askquit is specified, ngspice will terminate immediately.
The optional parameter exitcode is an integer that sets the exit code for ngspice. This is useful
to return a success/fail value to the operating system.

17.5. COMMANDS

17.5.49

329

Rehash: Reset internal hash tables

General Form:
rehash
Recalculate the internal hash tables used when looking up UNIX commands, and make all
UNIX commands in the user’s PATH available for command completion. This is useless unless
you have set unixcom first (see above).

17.5.50

Remcirc*: Remove the current circuit

General Form:
remcirc
This command removes the current circuit from the list of circuits sourced into ngspice. To select a specific circuit, use setcirc (17.5.60). To load another circuit, refer to source (17.5.70).
The new actual circuit will be the circuit on top of the list of the remaining circuits.

17.5.51

Reset*: Reset an analysis

General Form:
reset
Throw out any intermediate data in the circuit (e.g, after a breakpoint or after one or more
analyses have been done), and re-parse the input file. The circuit can then be re-run from it’s
initial state, overriding the effect of any set or alter commands.
Reset may be required in simulation loops preceding any run (or tran ...) command.

17.5.52

Reshape: Alter the dimensionality or dimensions of a vector

General Form:
reshape vector vector ...
or
reshape vector vector ... [ dimension , dimension , ... ]
or
reshape vector vector ... [ dimension ][ dimension ] ...
This command changes the dimensions of a vector or a set of vectors. The final dimension
may be left off and it will be filled in automatically. If no dimensions are specified, then the
dimensions of the first vector are copied to the other vectors. An error message of the form
’dimensions of x were inconsistent’ can be ignored.

330

CHAPTER 17. INTERACTIVE INTERPRETER

Example:
* generate vector with all (here 30) elements
let newvec = vector (30)
* reshape vector to format 3 x 10
reshape newvec [3][10]
* access elements of the reshaped vector
print newvec [0][9]
print newvec [1][5]
let newt = newvec [2][4]

17.5.53

Resume*: Continue a simulation after a stop

General Form:
resume
Resume a simulation after a stop or interruption (control-C).

17.5.54

Rspice*: Remote ngspice submission

General Form:
rspice 
Runs a ngspice remotely taking the input file as a ngspice input file, or the current circuit if
no argument is given. Ngnutmeg or ngspice waits for the job to complete, and passes output
from the remote job to the user’s standard output. When the job is finished the data is loaded
in as with aspice. If the variable rhost is set, ngnutmeg connects to this host instead of the
default remote ngspice server machine. This command uses the rsh command and thereby
requires authentication via a .rhosts file or other equivalent method. Note that rsh refers to
the ‘remote shell’ program, which may be remsh on your system; to override the default name
of rsh, set the variable remote_shell. If the variable rprogram is set, then rspice uses this
as the pathname to the program to run on the remote system.
Note: rspice will not acknowledge elements that have been changed via the alter or altermod
commands.

17.5.55

Run*: Run analysis from the input file

General Form:
run [ rawfile ]
Run the simulation as specified in the input file. If there were any of the control lines .ac, .op,
.tran, or .dc, they are executed. The output is put in rawfile if it was given, in addition to
being available interactively.

17.5. COMMANDS

17.5.56

331

Rusage: Resource usage

General Form:
rusage [ resource ...]
Print resource usage statistics. If any resources are given, just print the usage of that resource.
Most resources require that a circuit be loaded. Currently valid resources are:
decklineno Number of lines in deck
netloadtime Nelist loading time
netparsetime Netlist parsing time
elapsed The amount of time elapsed since the last rusage elapsed call.
faults Number of page faults and context switches (BSD only).
space Data space used.
time CPU time used so far.
temp Operating temperature.
tnom Temperature at which device parameters were measured.
equations Circuit Equations
time Total Analysis Time
totiter Total iterations
accept Accepted time-points
rejected Rejected time-points
loadtime Time spent loading the circuit matrix and RHS.
reordertime Matrix reordering time
lutime L-U decomposition time
solvetime Matrix solve time
trantime Transient analysis time
tranpoints Transient time-points
traniter Transient iterations
trancuriters Transient iterations for the last time point*
tranlutime Transient L-U decomposition time
transolvetime Transient matrix solve time
everything All of the above.
* listed incorrectly as ‘Transient iterations per point’.

332

CHAPTER 17. INTERACTIVE INTERPRETER

17.5.57

Save*: Save a set of outputs

General Form:
save [all | outvec ...]
Save a set of outputs, discarding the rest (if not keyword all is given). Maybe used to dramatically reduce memory (RAM) requirements if only a few useful node voltages or branch currents
are saved.
Node voltages may be saved by giving the nodename or v(nodename). Currents through an
independent voltage source are given by i(sourcename) or sourcename#branch. Internal device data (31.1) are accepted as @dev[param]. The syntax is identical to the .save command
(15.6.1).
Note: In the .control ....
mand to become effective.

.endc section save must occur before the run or tran com-

If a node has been mentioned in a save command, it appears in the working plot after a run has
completed, or in the rawfile written by the write (17.5.88) command. For backward compatibility, if there are no save commands given, all outputs are saved. If you want to trace (17.5.78)
or plot (17.5.45) a node, you have to save it explicitly, except for all given or no save command
at all.
When the keyword all appears in the save command, all node voltages, voltage source currents
and inductor currents are saved in addition to any other vectors listed.
Save voltage and current:
save vd_node vs# branch v( vs_node ) i(vs2)
Save allows to store and later access internal device parameters. e.g. in a command like
Save internal parameters:
save all @mn1[gm]
saves all standard analysis output data plus gm of transistor mn1 to internal memory (see also
31.1).
save may store data from nodes or devices residing inside of a subcircuit:
Save voltage on node 3 (top level), node 8 (from inside subcircuit x2) and current through vmeas
(from subcircuit x1):
save 3 x1.x2.x1.x2 .8 v.x1.x1.x1.vmeas# branch
Save internal parameters within subcircuit:
save @m. xmos3 .mn1[gm]
Use commands listing expand (17.5.37, before the simulation) or display (17.5.20, after simulation) to obtain a list of all nodes and currents available. Please see Chapt. 31 for an
explanation of the syntax for internal parameters.

17.5. COMMANDS

333

Entering several save lines in a single .control section will accumulate the nodes and parameters to be saved. If you want to exclude a node, you have to get its number by calling status
(17.5.72) and then calling delete number (17.5.16).

17.5.58

Sens*: Run a sensitivity analysis

General Form:
sens output_variable
sens output_variable ac ( DEC | OCT | LIN ) N Fstart Fstop
Perform a Sensitivity analysis. output_variable is either a node voltage (ex. v(1) or
v(A,out)) or a current through a voltage source (e.g. i(vtest)). The first form calculates DC sensitivities, the second form AC sensitivities. The output values are in dimensions of
change in output per unit change of input (as opposed to percent change in output or per percent
change of input).

17.5.59

Set: Set the value of a variable

General Form:
set [word]
set [word = value ] ...
Set the value of word to value, if it is present. You can set any word to be any value, numeric or
string. If no value is given then the value is the Boolean ‘true’. If you enter a string, you have
to enclose it in double quotes. Set save the lower case version of a word string.
The value of word may be inserted into a command by writing $word. If a variable is set to
a list of values that are enclosed in parentheses (which must be separated from their values by
white space), the value of the variable is the list.
The variables used by ngspice are listed in section 17.7.
Set entered without any parameter will list all variables set, and their values, if applicable.
Be advised that set sets the lower case variant of word. If this is not desired, please use setcf
(see ?? below).

17.5.60

Setcirc*: Change the current circuit

General Form:
setcirc [ circuit number ]
The current circuit is the one that is used for the simulation commands below. When a circuit
is loaded with the source command (see below, 17.5.70) it becomes the current circuit.
Setcirc followed by ’return’ without any parameters lists all circuits loaded.

334

CHAPTER 17. INTERACTIVE INTERPRETER

17.5.61

Setplot: Switch the current set of vectors

General Form:
setplot [ plotname ]
Set the current plot to the plot with the given name, or if no name is given, prompt the user
with a menu. (Note that the plots are named as they are loaded, with names like tran1 or op2.
These names are shown by the setplot and display commands and are used by diff, below.)
If the ‘New plot’ item is selected, the current plot becomes one with no vectors defined.
Note that here the word plot refers to a group of vectors that are the result of one ngspice run.
When more than one file is loaded in, or more than one plot is present in one file, ngspice keeps
them separate and only shows you the vectors in the current plot.

17.5.62

Setscale: Set the scale vector for the current plot

General Form:
setscale [ vector ]
Defines the scale vector for the current plot. If no argument is given, the current scale vector is
printed. The scale vector delivers the values for the x-axis in a 2D plot.

17.5.63

Settype: Set the type of a vector

General Form:
settype type vector ...
Change the type of the named vectors to type. Type names can be found in the following table.
Type
notype
time
frequency
voltage
current
onoise-spectrum
onoise-integrated
inoise-spectrum
inoise-integrated

Unit
s
Hz
V
A √
(V or A)/ Hz
V or A
√
(V or A)/ Hz
V or A

Type
pole
zero
s-param
temp-sweep
res-sweep
impedance
admittance
power
phase
decibel

Unit

Celsius
Ohms
Ohms
Mhos
W
Degree
dB

17.5. COMMANDS

17.5.64

335

Shell: Call the command interpreter

General Form:
shell [ command ]
Call the operating system’s command interpreter; execute the specified command or call for
interactive use.

17.5.65

Shift: Alter a list variable

General Form:
shift [ varname ] [ number ]
If varname is the name of a list variable, it is shifted to the left by number elements (i.e, the
number leftmost elements are removed). The default varname is argv, and the default number
is 1.

17.5.66

Show*: List device state

General Form:
show devices [ : parameters ] , ...
The show command prints out tables summarizing the operating condition of selected devices.
If devices is missing, a default set of devices are listed, if devices is a single letter, devices
of that type are listed. A device’s full name may be specified to list only that device. Finally,
devices may be selected by model by using the form #modelname.
If no parameters are specified, the values for a standard set of parameters are listed. If the list of
parameters contains a ‘+’, the default set of parameters is listed along with any other specified
parameters.
For both devices and parameters, the word all has the obvious meaning.
Note: there must be spaces separating the ‘:’ that divides the device list from the parameter list.

17.5.67

Showmod*: List model parameter values

General Form:
showmod models [ : parameters ] , ...
The showmod command operates like the show command (above) but prints out model parameter values. The applicable forms for models are a single letter specifying the device type letter
(e.g. m, or c), a device name (e.g. m.xbuf22.m4b), or #modelname (e.g. #p1).

336

CHAPTER 17. INTERACTIVE INTERPRETER

17.5.68

Snload*: Load the snapshot file

General Form:
snload circuit -file file
snload reads the snapshot file generated by snsave (17.5.69). circuit-file is the original circuit
input file. After reading, the simulation may be continued by resume (17.5.53).
An input script for loading circuit and intermediate data, resuming simulation and plotting is
shown below:
Typical usage:
* SCRIPT : ADDER - 4 BIT BINARY
* script to reload circuit and continue the simulation
* begin with editing the file location
* to be started with ’ngspice adder_snload .script ’
. control
* cd to where all files are located
cd D:\ Spice_general \ ngspice \ examples \ snapshot
* load circuit and snpashot file
snload adder_mos_circ .cir adder500 .snap
* continue simulation
resume
* plot some node voltages
plot v(10) v(11) v(12)
.endc
Due to a bug we currently need the term ’script’ in the title line (first line) of the script.

17.5.69

Snsave*: Save a snapshot file

General Form:
snsave file
If you run a transient simulation and interrupt it by e.g. a stop breakpoint (17.5.74), you may
resume simulation immediately (17.5.53) or store the intermediate status in a snapshot file by
snsave for resuming simulation later (using snload (17.5.68)), even with a new instance of
ngspice.

17.5. COMMANDS

337

Typical usage:
Example input file for snsave
* load a circuit ( including transistor models and .tran command )
* starts transient simulation until stop point
* store intermediate data to file
* begin with editing the file location
* to be run with ’ngspice adder_mos .cir ’
. include adder_mos_circ .cir
. control
*cd to where all files are located
cd D:\ Spice_general \ ngspice \ examples \ snapshot
unset askquit
set noinit
* interrupt condition for the simulation
stop when time > 500n
* simulate
run
* store snapshot to file
snsave adder500 .snap
quit
.endc
.END
adder_mos_circ.cir is a circuit input file, including the netlist, .model and .tran statements.
Unfortunately snsave/snload will not work if you have XSPICE devices (or V/I sources with
polynomial statements) in your input deck.

17.5.70

Source: Read a ngspice input file

General Form:
source infile
For ngspice: read the ngspice input file infile, containing a circuit netlist. Ngnutmeg and ngspice
commands may be included in the file, and must be enclosed between the lines .control and
.endc. These commands are executed immediately after the circuit is loaded, so a control
line of ac ... works the same as the corresponding .ac card. The first line in any input file
is considered a title line and not parsed but kept as the name of the circuit. Thus, a ngspice
command script in infile must begin with a blank line and then with a .control line. Also,
any line starting with the string ‘*#’ is considered as a control line (.control and .endc is
placed around this line automatically.). The exception to these rules are the files spinit (16.5)
and .spiceinit (16.6).

338

CHAPTER 17. INTERACTIVE INTERPRETER

For ngutmeg: reads commands from the file infile. Lines beginning with the character ‘*’ are
considered comments and are ignored.
The following search path is executed to find infile: current directory (OS dependent), /share/ngspice/scripts, env. variable NGSPICE_INPUT_DIR (if defined), see 16.7. This
sequence may be overriden by setting the internal sourcepath variable (see 17.7) before calling source infile.

17.5.71

Spec: Create a frequency domain plot

General Form:
spec start_freq stop_freq step_freq vector [ vector ...]
Calculates a new complex vector containing the Fourier transform of the input vector (typically the linearized result of a transient analysis). The default behavior is to use a Hanning
window, but this can be changed by setting the variables specwindow and specwindoworder
appropriately.
Typical usage:
ngspice
ngspice
ngspice
ngspice

13
14
15
16

->
->
->
->

linearize
set specwindow = " blackman "
spec 10 1000000 1000 v(out)
plot mag(v(out ))

Possible values for specwindow are: none, hanning, cosine, rectangular, hamming, triangle,
bartlet, blackman, gaussian and flattop. In the case of a gaussian window specwindoworder
is a number specifying its order. For a list of window functions see 17.5.26.

17.5.72

Status*: Display breakpoint information

General Form:
status
Display all of the saved nodes and parameters, traces and breakpoints currently in effect.

17.5.73

Step*: Run a fixed number of time-points

General Form:
step [ number ]
Iterate number times, or once, and then stop.

17.5. COMMANDS

17.5.74

339

Stop*: Set a breakpoint

General Form:
stop [ after n] [ when value cond value ] ...
Set a breakpoint. The argument after n means stop after iteration number ‘n’, and the argument
when value cond value means stop when the first value is in the given relation with the
second value, the possible relations being
Symbol
=
<>
>
<
>=
<=

Alias
eq
ne
gt
lt
ge
le

Meaning
equal to
not equal
greater than
less than
greater than or equal to
less than or equal to

Symbol or alias may be used alternatively. All stop commands have to be given in the control
flow before the run command. The values above may be node names in the running circuit, or
real values. If more than one condition is given, e.g.
stop after 4 when v(1) > 4 when v(2) < 2,
the conjunction of the conditions is implied. If the condition is met, the simulation and control
flow are interrupted, and ngspice waits for user input.
In a transient simulation the ‘=’ or eq will only work with vector time in commands like
stop when time = 200n.
Internally a breakpoint will be set at the time requested. Multiple breakpoints may be set. If the
first stop condition is met, the simulation is interrupted, the commands following run or tran
(e.g. alter or altermod) are executed, then the simulation may continue at the first resume
command. The next breakpoint requires another resume to continue automatically. Otherwise
the simulation stops and ngspice waits for user input.
If you try to stop at
stop when V(1) eq 1
(or similar) during a transient simulation, you probably will miss this point, because it is not
very likely that at any time step the vector v(1) will have the exact value of 1. Then ngspice
simply will not stop.

17.5.75

Strcmp: Compare two strings

General Form:
strcmp _flag $string1 " string2 "
The command compares two strings, either given by a variable (string1) or as a string in quotes
(‘string2’). _flag is set as an output variable to ’0’, if both strings are equal. A value greater
than zero indicates that the first character that does not match has a greater value in str1 than in
str2; and a value less than zero indicates the opposite (like the C strcmp function).

340

CHAPTER 17. INTERACTIVE INTERPRETER

17.5.76

Sysinfo*: Print system information

General Form:
sysinfo
The command prints system information useful for sending bug report to developers. Information consists of:
• Name of the operating system,
• CPU type,
• Number of physical processors (not available under Windows OS), number of logical
processors,
• Total amount of DRAM available,
• DRAM currently available.
The example below shows the use of this command.
ngspice 1 -> sysinfo
OS: CYGWIN_NT -5.1 1.5.25(0.156/4/2) 2008 -06 -12 19:34
CPU: Intel (R) Pentium (R) 4 CPU 3.40 GHz
Logical processors : 2
Total DRAM available = 1535.480469 MB.
DRAM currently available = 984.683594 MB.
ngspice 2 ->
This command has been tested under Windows OS and Linux. It may not be available in your
operating system environment.

17.5.77

Tf*: Run a Transfer Function analysis

General Form:
tf output_node input_source
The tf command performs a transfer function analysis, returning:
• the transfer function (output/input),
• output resistance,
• and input resistance

17.5. COMMANDS

341

between the given output node and the given input source. The analysis assumes a small-signal
DC (slowly varying) input. The following example file
Example input file:
* Tf test circuit
vs
1
0
dc 5
r1
1
2
100
r2
2
3
50
r3
3
0
150
r4
2
0
200
. control
tf v(3 ,5) vs
print all
.endc
.end
will yield the following output:
transfer_function = 3.750000e-001
output_impedance_at_v(3,5) = 6.662500e+001
vs#input_impedance = 2.000000e+002

17.5.78

Trace*: Trace nodes

General Form:
trace [ node ...]
For every step of an analysis, the value of the node is printed. Several traces may be active at
once. Tracing is not applicable for all analyses. To remove a trace, use the delete (17.5.16)
command.

17.5.79

Tran*: Perform a transient analysis

General Form:
tran Tstep Tstop [ Tstart [ Tmax ] ] [ UIC ]
Perform a transient analysis. See Chapt. 15.3.9 of this manual for more details.
An interactive transient analysis may be interrupted by issuing a ctrl-c (control-C) command.
The analysis then can be resumed by the resume command (17.5.53). Several options may be
set to control the simulation (15.1.4).

342

CHAPTER 17. INTERACTIVE INTERPRETER

17.5.80

Transpose: Swap the elements in a multi-dimensional data set

General Form:
transpose vector vector ...
This command transposes a multidimensional vector. No analysis in ngspice produces multidimensional vectors, although the DC transfer curve may be run with two varying sources. You
must use the reshape command to reform the one-dimensional vectors into two dimensional
vectors. In addition, the default scale is incorrect for plotting. You must plot versus the vector corresponding to the second source, but you must also refer only to the first segment of
this second source vector. For example (circuit to produce the transfer characteristic of a MOS
transistor):
How to produce the transfer characteristic of a MOS transistor:
ngspice
ngspice
ngspice
ngspice
ngspice

17.5.81

>
>
>
>
>

dc vgg 0 5 1 vdd 0 5 1
plot i(vdd)
reshape all [6 ,6]
transpose i(vdd) v(drain)
plot i(vdd) vs v(drain )[0]

Unalias: Retract an alias

General Form:
unalias [word ...]
Removes any aliases present for the words.

17.5.82

Undefine: Retract a definition

General Form:
undefine function
Definitions for the named user-defined functions are deleted.

17.5.83

Unlet: Delete the specified vector(s)

General Form:
unlet vector [ vector ... ]
Delete the specified vector(s). See also let (17.5.35).

17.5. COMMANDS

17.5.84

343

Unset: Clear a variable

General Form:
unset [word ...]
Clear the value of the specified variable(s) (word).

17.5.85

Version: Print the version of ngspice

General Form:
version [-s | -f | ]
Print out the version of ngnutmeg that is running, if invoked without argument or with -s or -f.
If the argument is a  (any string different from -s or -f is considered a 
), the command checks to make sure that the arguments match the current version of ngspice.
(This is mainly used as a Command: line in rawfiles.)
Options description:

• No option: The output of the command is the message you can see when running ngspice
from the command line, no more no less.

• -s(hort): A shorter version of the message you see when calling ngspice from the command line.

• -f(ull): You may want to use this option if you want to know what extensions are included
into the simulator and what compilation switches are active. A list of compilation options
and included extensions is appended to the normal (not short) message. May be useful
when sending bug reports.

The following example shows what the command returns in some situations:

344

CHAPTER 17. INTERACTIVE INTERPRETER

Use of the version command:
ngspice 10 -> version
******
** ngspice -24 : Circuit level simulation program
** The U. C. Berkeley CAD Group
** Copyright 1985 -1994 , Regents of the University of California .
** Please get your ngspice manual from
http :// ngspice . sourceforge .net/docs.html
** Please file your bug - reports at
http :// ngspice . sourceforge .net/ bugrep .html
** Creation Date: Jan 1 2011
13:36:34
******
ngspice 2 ->
ngspice 11 -> version 14
Note: rawfile is version 14 ( current version is 24)
ngspice 12 -> version 24
ngspice 13 ->

Note for developers: The option listing returned when version is called with the
-f flag is built at compile time using #ifdef blocks. When new compile switches
are added, if you want them to appear on the list, you have to modify the code in
misccoms.c.

17.5.86

Where*: Identify troublesome node or device

General Form:
where
When performing a transient or operating point analysis, the name of the last node or device to
cause non-convergence is saved. The where command prints out this information so that you
can examine the circuit and either correct the problem or generate a bug report. You may do this
either in the middle of a run or after the simulator has given up on the analysis. For transient
simulation, the iplot command can be used to monitor the progress of the analysis. When the
analysis slows down severely or hangs, interrupt the simulator (with control-C) and issue the
where command. Note that only one node or device is printed; there may be problems with
more than one node.

17.5. COMMANDS

17.5.87

345

Wrdata: Write data to a file (simple table)

General Form:



Source Exif Data:
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.7
Linearized                      : Yes
Page Mode                       : UseOutlines
XMP Toolkit                     : Adobe XMP Core 4.0-c316 44.253921, Sun Oct 01 2006 17:14:39
Create Date                     : 2017:08:31 15:02:04+02:00
Creator Tool                    : LaTeX with hyperref package
Modify Date                     : 2017:08:31 15:03:02+02:00
Metadata Date                   : 2017:08:31 15:03:02+02:00
Format                          : application/pdf
Description                     : ngspice circuit simulator
Creator                         : Paolo Nenzi, Holger Vogt, Dietmar Warning e.a.
Title                           : ngspice user manual
Producer                        : pdfTeX-1.40.18
Trapped                         : False
PTEX Fullbanner                 : This is MiKTeX-pdfTeX 2.9.6362 (1.40.18)
Document ID                     : uuid:1d07fe46-625a-426d-b623-b6b412012b5b
Instance ID                     : uuid:b0349b4b-edd9-432e-ac58-e819995c0aa4
Page Count                      : 624
Page Layout                     : OneColumn
Subject                         : ngspice circuit simulator
Author                          : Paolo Nenzi, Holger Vogt, Dietmar Warning e.a.
Keywords                        : ngspice, spice
PTEX Fullbanner                 : This is MiKTeX-pdfTeX 2.9.6362 (1.40.18)
EXIF Metadata provided by EXIF.tools

Navigation menu