Service_Bulletins Service Bulletins

User Manual: Service_Bulletins

Open the PDF directly: View PDF .
Page Count: 11

~

---..

.-

...

~

.~::..

:1.

•

\0.01'

I - .... ' !

TX

NO

1110

TO:

M Abrahams

D

Baum

p.

Eruer

R

Burton

....

Carson

l

C Cobb

D Cowdin

W

Garrison

R

Geyer

R

Gutosky

T

DeLaFuente

RED

R

Havlick

E

Heston

-

Hoy

B

Hull

A

Jager

B

Keller

D

Leek

H

Leggett

B.

rvey

C

~arrone

B

HcNeese

L I G H T

C

McNutt

E.

Miguel

G

Orlowski

A

Powers

C

Sheldon

R

Sundius

B

Wicks

B

Williams

G

Wipperfuerth

B

~·right

B

Zinsmeister

ALE

R T 0561

EFFECTIVITY:

Fujitisu

Eagle

Disk

Drives

PROBLEM:

Local

cc:

J

Bogdan

D

Calderara

D

Carver

D

Corts

B

Daley

R

Diodati

H

Johnson

W

May

tan

D

Petucci

L

Stevens

G

Whittaker

R

Poland

J

Hill

The

disk

will

not

power

up when

the

system

is

first

received.

SOLUTION:

-

Reset

connector

CJN502 on

the

head/deckassembly.

To

gain

access

to

the

connector,

the

card

cage

on

the

left

side

of

the

disk

must

be

lifted"out.

The

connector

is

on

the

bottom

of

the

HDA.

Seating

of

this

connector

is

apparently

disturbed

while

the

vibration

damping

foam

block

is

installed

or

removed.

CONCLUSION:

This

information

will

be

contained

in

the

service

manual

in

the

black

book.

BEn

Wright

Qantel

4."

f5.L

SE~Uc)

tJl\

DATA

!?J6--1

/J

If.

t)-'U

.,

re..

,,/,.

111-

Tt..sr

~SSt',6/~

JI~·Ad

S.dt,t:.-{-

.

Prv,6J~WJ

{<.<.f./

~

t:-Co.

f{;;Al·~J

13

~J$.

fossi

.J,/~

/3J

Dr

f);r-

t7

H~J

If

"SSi

.tIJ/.f.

CIt"',

J-I

..

iA.~

~t.ep<"

tAw·

n4,--

To

CD.,

.r.

...

,.,

LD"J

Hu/

A,..,.

i?

(1\.-,

LJR

1ft

744-

~..,

~c".

#'-1"c.

~

.

DI"Se.

Of(

F.E..

k-l-AY~II\.

fo!r;6/(

S't.

.....

t.10

PJ/I.~l'

FI';'~YC

0"

fos;hDlle,..

.

GtAt-;"

If

Co

..

.p,.

~-fltAcJ..

.oi

t'K'Ehc/OS~Y

-

\

/{4!f'/#o

c.-

A~l!~,I,I'1

IV

'O.tr"

;',."

0", 57ii

TCAS

Ru."

-Yf;~uf

Or

U5..(..

(Kano

fo

1",it,"Ji

zc,

OnJ{ 'I

tJ.L

J

S~c.

.,,,,

)~-S+<'''1

1'1

/!..£-

RE

STci'()

I+

FrD""

.

i3F1LF

Il

:

Lta\JC,.

Sf-u

'"

r..,se

....

V

l<y~t".~J

,--.-.IS(!I'Aflr.

f6;

lv.rL:

I<IlJO/H..f-

A~soc.

B,",ls.

.D·ft",J;

"1

{'h

srATlAf:

~fA."

M,'c

r

".

f/?°Co

A"'''/~r

*Flx~£cr

rJ

BFILFIL

All

/l.;f

t3

..

c<

s(cA",.

T"

T~

.It..

Rc."

Foy"".{-

DISc...

rf

r-~,J~J..:

,r<...~DIAC~

..

:

DiSc.

E,.,<.I.

;"rc~

A/f~Yl'ttd.(.

~5~doy

If

1flXSEc

lJAS

t.d"t',l ;

fro~/""7

S,c..-:.ft-

.....

IA-I",

;~

BAD

R~f!".e

...

lJj5

(..

t",c!cs.t,v

..D

...

:

0<.

/.s

()~""1V4-:0-...J

(3i,c

r$"ctf7lV'",tE

IS

",,,t-:

"

fr{)~

~

til

",,.,.

(~

r,. A

1.

L

ar~~.'

M-.

j

'"

p".o~

I

~tt1

SCOL-.f...".

!/lc,I S

If]

<..

R.A

J!.IA

:

,.cJ

M.<

'f

A~s,s

Wi4~

('(7

II.

,

, .

Sc.lsr..tG-r

A

......

AS:

Co~iYO

II~~

Ow.

(j

..

J..

Su-~:

(;

...

1,.,

C(J.~r

I/OC~hs

i

'CA.tII.f..

'O~;II'(

",. '

l-E..10'1

1-°1;"',

a..4

lt

CD

...

n..o

'/t",

7i1Mt~

Ser,,,,,s

o~

f../I«

....

C

...

~

/(l./,

I,.

(-t.

f,

W-t.V'"

5

&A.fr'

y

TtM-<'

O,..IIJ

~

""'" II

C"lo/'$

()"

1;.,

Vo/~,.f.

[)~fJ

$

rSllh.

-<-

To

t)1\.A,.

13

,..J,

-r

/?tflloc

...

r+

r t

O"(~Q

II

Lo,.J

It71.....

f,.oA)~

IYI

~

cA~ck

C",,..,.,,,11;,..;

S~Hi"1·(tr1Jr?)

~(f'''t.

-<...

,fo~

..

,...SIA('

./

f~>f~c-r

Asuh7-6li.(,r:

j).tu. J 2-

1.).( t/.

J.(

2.

T()~·

J.{~

..tr~J

114dr

l3~r

~J/<-

foss;

),/011..

P4

Sif;

017.f.",

Fit;

14

r'-{

/Jelf'Ti.,riDt1

(e.,ll-<~

~.ffI4(~

Ad

j

/A;d

It-

/(,,1"";

~

WAtfrJ,

NG

.','

{<t.fl""

~

"l}s;.f;Olt

~rl

j)}SK

£'t1t:./".I~

r

Disk

failures

generally

appear

to

the

customer

as

two

major

types:

1.

System

"BOOTS"

but

gets

Disk

Errors.

2.

System

"Can't

Boot".

These

two

major

catagories

are

your

first

nhalf-split".

DRIVE "BOOTS"

BUT

GETS

ERRORS

A

drive

that

boots

will

provide

many more

clues

to

aid

repair

than

one

that

doesn't.

The

ability

to

use

the

Operating

System

as

a

tool

is

a

definite

advantage.

Drives

that

can

be

booted

are

not

necessarilly

easier

to

repair,

but

require

much

less

discussion

HERE

since

the

"CAN

BOOT

FLOWCHART"

covers

the

wide

range

of

possibilities

much

better.

Refer

to

the

flowchart

for

a

logical

approach

to

diagnosing

this

type

of

failure.

CAN'T

BOOT

A

system

that

"can't

boot"

is

a

whole

new

ball

game.

A

major

dis-

advantage

is,

in

the

case

of

most

Dynex

systems,

you

don't

have

any

way

to

load

test

programs.

Also

a

customer

may

not

have

a

good

Backup.

In

these

cases,

You

must

rely

on

Macro

Programs

to

help

diagnose

the

failure.

Listed

below

are

a

few

Macro

Programs

that

can

be

adapted

to

suite

your

particular

situation.

Refer

to

the

"CAN'T

BOOT

FLOWCHART".

Many

of

the

boxes

require

macro

programs.

These

listed

are

only

one

of

many.ways

the

tasks

may

be

·"accomplished.

The

Flowchart

is

a

series

of

"half-splits"

that

attempts

,,·to

isolate

the

problem

into

one

of

FOUR

MAJOR

CATAGORIES:

POWER,

SELECT,

SERVO,

or

READ/WRITE. .

ACCESS

OTHER

DEVICES?

Is

the

problem

in

the

disk

subsystem

OR

is

it

somewhere

else?

(CPU,

Memory, CRT,

Power,

or

another

Controller

hanging

"up

the

backplane)

•

.I;"

>1~1:4:'

,.\..

.

__

.~

.

..;....,

;'

ECHO

MEMORY:

0010F2100000

OOlOF21000BO 0006A7

(xmitl~T'

Enter

16

characters

and

xmi

t.

The

CRT

s'creen

should

fill

with

the

character

entered.

If

this

fails,

remove

all

controllers

except

for

'the

CRT

and

try

.again.

If

it

still

fails,

you

have

eliminate~

the

Disk

AND

3/4

of

the

System.

The

above

test

is

a

very

,basic

test

but

is

a

good

starting

point.

The

next

step

may

be

attempting

to

boot

the

Tape

Dr~ve

(if

there

is

one).

If

you

can

boot

the

tape,

your

problem

IS

most

likely

the

Disk

~l1bsvstem.

-

--

-~

, ,

READ

STATUS

ZERO

RSO:

4d009D 000lF30017BO 00aaA8

(xmit)

An

IOU-24

should

return

1174"

(immediatly

following

IPL).

An

IOU-32

or

42

should

return

n04".

Status'

F4,

E4,

94,

or

84

indicate

serious

failures.

Follow

the

nREADY"

branch

to

NO.

If

status

is

good,

follow

YES

branch.

~"'--.

l"!-~~"""

;

'~

.. '

~:

... I

..

,:_

S~EX

&

FILL:

C0056C

J~CGA8

xt3x3~3~3x3x

(x~i~)

Choose

2

LARGE

seek

address

and

execute

with

the

drive

covers

off.

Watch

for

~he

positioner

movement.

If

the

drive

can

seek

and

it

is

possible

to

change

Packs,

mount

your

ATP

pack

and

attempt

to

boot

i~.

If

it

boots,

it

is

very

possible

that

the

drive

is

O~

but

Sector

0

data

is

blown.

The

Operating

System

may

need

a ne'-l

LOADER.

This

is

accomplished

by

running

*LDLD

(BEST

utility).

DO

NOT

use

the

loader

program

on

your

ATP

pack.

ATP's

are

an

incompatable

language.

CAN

IT

READ

ALTERNATE

SEEKS,

READ,

&

REPORT

STATUS:

0020A7 <Q0020>

(cr)

_

006080

0300F2l0000d

4d009D OOOlF30017BO

OO~OA7

<Q0040>

(er)

008080

0300F220000d

4d009D

OOOlF30017BO 002DA7 <Q0060>

(er)

xd3x3x3x3x3x

<Q0080>

(er)

xd3x3x3x3x3x

(xmit)

Good

status

is

"04".

n74n

Status

indicates

a

failure

to

Read

properly.

However,

the

problem

may

still

be

in

the

SERVO

ckts.

(heads

not

positioned

directly

over

data

track).

,

problem

may

also

be

Head-a

is

bad

or

need~

·al~lgnment.

The

above

~,!program

"should

be

run

on

Head

O,first,

then

",try,"it

again

using

~,

,,:>

,:,I:,l>differeritheads.

If

other

heads"return

n74·,,~status,

the

next

step'is

.

ha'I~;:s~1~f~:

:b~~ween

SE~VO:c'~i?c~il

.

·.Ji~~,t'

;:f~~~;\,

.

"Y:i!j~;ij;~.·

IBIT

,TRACK

VERIFY·

Write

Controls:'"in

'ron't

'"of

EACH·

seeK"&

'.

ructlon~':

Ex.

ld019D

006080

'etc.

• Th*e":{statns'

:retllrne'd:\b()w.':~"

'1,'itl

be~::aItf:er~ent.An

D04

a

status'~:;indicate~tbe"-d(iv'e

1s

O'n"tllE(\'irong

·:~~t"i~rc1C~~:t~~'ec't~'servo"

qkts.")

'. A

D24~~,1I3411,

or

'1I,54:t1';;~:'~E;t:a~Us

'point$:a't'

the

a~a";~ckY~:~~~"Beawa're,

however,

th'e"r'e

is

a

'g'

ray

3

~i:lr"ea:'wb·ere!·:.a

'Poorly

us:t~~r<!(c~~~~vocan

make

the

'DATA'

ckts.

A~P'E~::?Ci,~':~-

I~il;~~;\i;:~l~':~j~:

" .

",::t:

:

MI

s~cy:~~:~~~~s:

. .

.,

;l¥~

l,l,~,R<,~,·;.{:·~ir~·

ri:t~fr·

·

r.

"'I

:i·~'~~.

:",!>.~

..

,

;.,:_,

These

may

'used

anytime

you

want

to

"theck

the

VJRIT'E;::~Ipabili{l~s'

of

a

drive~

CAUTION:

BE

CAREFUL

WHERE

YOU

WRITE!!

On

seaiedmediadisks,

,

'use

the

F

.E.

TEST

AREA.

On

Dynex

I

use

a

'sector

number,

greater

than

8000.

Thisltlay

will

not

destroy"

customer

data.,

"

It

is

highly

recommended

that

the

vlRITE

FAULT

test

point

be

monitored

as

this

program

executes.

SEEK

&

WRITE

0020A7

(Q0020)

0006F3l00000

004080

0006F21000Bd

0020A8

(Q0040),

(cr)

'··xd3x3x3x3x3x

(xmit)

Clear

screen

and

enter

six

bytes

of

dat'a~(xmit)

** To

loop

on

WRITE,

substitute

0029A7

for

0020A8.

SEEK

&

READ

To

verify

the

data

just

written,

the

following

program

will

dump

what

you

just

wrote

back

to

the

CRT.

0020A7 <Q0020)

(cr)

004080

0006F220000d

0006F32000BO 0020A8 <Q0040)

(cr)

xd3x3x3x3x3x

(xmit)

EMS

MESSAGE

NUMBER

156

026/00053

15:24

03/12

HERK

(CJC133:REDLT,LISVC,MEND)

TX

NO

133

TO:

M Abrahams

D

Saum

R

Bruer

R

Burton

T

Carson

C

Cobb

D.

Cowdin

W

Garrison

R Geyer

R

Gutosky

T DeLaFuente

RED

R

Havlicek

E

Heston

T

Hoy

B

Hull

A

Jager

B.

Keller

D Leek

H

Leggett

B.

Ivey

C Marrone

B McNeese

L I G H T

C McNutt

E.

Miguel

G

Orlowski

A Powers

C

Sheldon

R

Sundius

A.

Rodman

B

Williams

G

Wipperfuerth

B

Wright

B

Zinsmeister

R.

Biegun

'

ALE

R T

11564

EFFECTIVITY:

SYSTEM

64

+5V

POWER

SUPPLIES

PROBLEM:

Local

cc:

J Bogdan

D

Calderara

D

Carver

D

Corts

B

Daley

R

Diodati

H

Johnson

W

May

tan

D

Petucci

L

Stevens

G

Whittaker

R

Poland

J

Hill

Please

be

advised

that

we

have

two

different

versions

of

the

3/64

5VDC/150

AMP

power

supply

(042646501)

which

could

cause

sparing

problems

for

the

field.

They

are

designated,

but

may

not

be marked

Rev

2 and

Rev

A.

Rev

2

supplies

must

have

an

adapter

cable

(042681201),

but

Rev

A

does

not

need

one.

SOLUTION:

Attach

a

warning

label

to

all

Rev. 2

supplies

CONCLUSIONS:

~

determine

if

a

supply

nee,ds

the

adapter

l

turn

off

the

main power

~i

tch

and

measure

the

resistance

between

pin

4

of

the"

Molex'~'

~~nnector

and

ground..

It

should

measure

90 to,

110

ohms

if·

adapter

is

~N"OT

needed

(Rev A

supply).·

:'With

power

"bn',

$hortin.g~'·

Pin

6

to-

"

~:

'~~und

should

cause

tne,

supply

to

shut

doWn.

Wi

thout

,.

the

adapter

{~~t268120

1) , Rev. 2

~upplies

will

measure

90

to

100·

at-uns

at

Pin;

2,

i~~'

Pin

3

grounded

w~ll

shut

it

down~'

;

';

~,~~;:..

. t'

:t

,N}:V:~R

separate

the

adapter

cable

frOll a

Rev.

2

system

64

+5V

supply.

f)j"~r

warning

labels'

'from

Techr!lical -

Support·

in

Hayward,

(4'1~)

887-1111,

Ext:

3702

whenever"you~:',r1nd

a Rev. 2

supply

without

'::lIftrning

label.

;;,.

i

..

:":

",'10

CONNECTOR

COHFIGURAT~QN

~~Sr

1.

Pre-release

Revision

level

2

has

the

control

outputs

provided

via

6-pin

MOLEX

connector

with

pin-outs

of:

PIN

II

-~---

1

2

3

4

5

6

2.

, Prod

uction

PIN

II

-----

1

2

3

4

5

6

FUNCTION

--------

+

SENSE

-

SENSE

ON/OFF

POWER

FAIL

HI-MARGIN

LOW-MARGIN

Release

Level

FUNCTION

--------

+

SENSE

HI-MARGIN

LOW-MARGIN

-

SENSE

POWER

FAIL

INHIBIT

A

has

the

COMMENT

NOT

USED

NOT

USED

NOT

USED

pin-out

of:

COMMENT

--------

------

NOT

USED

NOT

USED

------

------

NOT

USED

This

information

will

be

included

in

a

CSB.

George

Steffner

Qantel

A

EMS

MESSAGE

NUMBER

268

008/00016

11:26

03/26

HERK

(CJC102:REDLT,LISVC,MEND)

TX

NO

102

TO:

M Abrahams

D

Baum

R

Bruer

R

Burton

T

Carson

C

Cobb

D.

Cowdin

W

Garrison

R Geyer

R

Gutosky

T DeLaFuente

RED

R

Havlicek

E

Heston

T

Hoy

B

Hull

A

Jager

B.

Keller

D Leek

H

Leggett

B.

Ivey

C Marrone

B McNeese

T.

Elder

L I G H T

C McNutt

E.

Miguel

G

Orlowski

A Powers

C

Sheldon

R

Sundius

A.

Rodman

B

Williams

G

Wipperfuerth

D.

Crandall

B

Zinsmeister

R.

Biegun

ALE

R T

11566

EFFECTIVITY:

All

Qantel

System

64's

PROBLEM:

Local

cc:

J Bogdan

D

Calderara

D

Carver

D

Corts

B

Daley

R

Diodati

H

Johnson

W

May

tan

D

Petucci

L

Stevens

G

Whittaker

R

Poland

J

Hill

A

load

base

list

fault

will

not

be

detected

by 64

ALU

firmware

below

Revision

C.

Distortion

of

memory

data,

and

consequently

disk

files,

can

result

if

the

fault

occurs.

SOLUTION:

Replace

processor

firmware

in

the

ALU

board

as

soon

as

possible.

PARTS:

System

64

ALU

PROM

Set,

PN

042716901

SPECIAL

CONSIDERATIONS:

Each

ALU

board's

serial

number

must

be sUDDlied when

the

PROM's

are

ordered.

Factory

effectivity

of

this·

revision

begins

3/25/84.

This

will

be

followed

up

by

a eSB.

George

Steffner

Qantel

Tech

Support

,..

EMS

MESSAGE

NUMBER

197

013/00012

16:20

04/02

EAST

<CJC221:REDLT,LISVC,MEND)

TX

NO

221

April

2,

1984

TO:

M Abrahams R

Havlicek

C McNutt

D

Baum

E

Heston

E.

Miguel

R

Bruer

T

Hoy

G

Orlowski

R

Burton

B

Hull

A Powers

T

Carson

A

Jager

C

Sheldon

C Cobb B.

Keller

R

Sundius

D.

Cowdin D Leek

A.

Rodman

W

Garrison

H

Leggett

B

Williams

R

Geyer

B.

Ivey

G

Wipperfuerth

R

Gutosky

C Marrone

D.

Crandall

T DeLaFuente B McNeese B

Zinsmeister

T.

Elder

R.

Biegun

RED

L I G H T

ALE

R T

11568

EFFECTIVITY:

Qantel

System

64

PROBLEM:

Local

cc:

.

J Bogdan

D

Calderara

D

Carver

D

Corts

B

Daley

R

Diodati

H

Johnson

W

May

tan

D

Petucci

L

Stevens

G

Whittaker

R

Poland

J

Hill

Mandatory

board

revision

levels

are

not

published

for

the

field.

SOLUTION:

A

revision

list

is

provided

in

this

RLA.

SPECIAL

CIRCUMSTANCES:

The

following

is

the

current

reV1Slon

levels

for

the

System

64

Processor

Boards.

If

any

new

units

arrive

with

boards

below

these

revisions,

please

contact

TFO

in

Cherry

Hill,

New

Jersey,

or

Technical

Support

in

Hayward,

California.

This

list

will

be

updated

as

necessary.

Q64

Revision

Levels

as

of

3/15/84

Board

Latest

Rev

Necessary

-----

---------- ----------

ALU

D C

ALU-R

A A

AUX

J E

...

1\1'\

lW

Il1.lJ

D C

MAD-R

A A

MDA

H H

MDA-R

B B

MEM

64

A C B

MEM

64 B A A

PROMS·

c

C

A

A

A

C

A

A

RED

L I G H T

ALE

R T 0567

EFFECTIVITY:

Qantel

IOU39Q

CRT

Controller

PROBLEM:

An

excessive

time

addressed

"0"

can

Data

transmission

will

be

stalled.

streaming

mode.

SOLUTION:

out

delay

on

the

IOU39Q

CRT

controller

cause

too

much

delay

during

a

streaming

backup.

to

the

streaming

tape

drive,

on

he

System

64,

The

tape

drive

will

then

ramp

down

out

of

The

standard

time

out

delay

for

the

IOU39Q

is

fifteen

mili-seconds.

Do

not

extend

the

delay

for

the

IOU39Q

addressed

"0".

Extended

delays

should

only

be

necessary

for

remote

communications

over

modems.

SPECIAL

CONSIDERATIONS:

It

has

been

reported

that

some

error

30's

and

string

hang

problems

were

relieved

by

increasing

the

IOU39Q

time

out

delay.

That

delay

can

slow

operation

of

the

system

as

well

as

interfere

with

streaming

backup.

CONCLUSION:

All

non-remote

operating

IOU39Q

CRT

controllers

should

have

minimum

time

out

delay

(15

mili-seconds).

If

failures

occur

at

that

setting,

please

notify

Product

Support

in

Hayward,

CA

and

TFO

by

WINC.

George

Steffner

Qantel

Tech

Support

A