26 Feb
2002
26 Feb
'02
12:25 a.m.
I have a DSD440 dual RX02 unit, and the first drive abruptly quit.
The disk access LED just flashes. When you first power it on, you can
hear a slight clicking noise when it first starts spinning. It does spin
though. Right after this noise, the LED starts flashing, and the drive
won't function. Does anyone have any ideas on what might be wrong and
what I can do about it?
Also, from the ODT prompt on my 11/23, how do you get it to boot
from the other (DY1) RX02?
Thanks,
Tom
26 Feb
26 Feb
1:42 a.m.
Tom Leffingwell wrote:
I have a DSD440 dual RX02 unit, and the first
drive abruptly quit.
The disk access LED just flashes. When you first power it on, you can
hear a slight clicking noise when it first starts spinning. It does spin
though. Right after this noise, the LED starts flashing, and the drive
won't function. Does anyone have any ideas on what might be wrong and
what I can do about it?
Jerome Fine replies:
Try checking the belt - the most common problem with an RX02 is
the belt falling off. I did use a DSD 440 over 20 years ago, but I
can't remember if it also used a drive belt. However, from your
description that the drive spins, my suggestion probably will not help.
Other than that, Tony might know something.
Also, from the ODT prompt on my 11/23, how do
you get it to boot
from the other (DY1) RX02?
Thanks,
Tom
You are probably entering the code by hand. Somewhere, you have specified
a default unit of ZERO. If you have the code, one of us could probably change
the unit number for you.
3:12 a.m.
Tom,
if there is no apparent software solution (what kind of ROM is in use,
is the bootstrap entered manually in octal, ...?), a quick fix might be
to swap the drives or their cables inside the box so that the working
unit becomes number zero.
But, Jerome is right, once we know what kind of bootstrap you use, it
should be possible to replace the unit number. - If it's not such a
tricky beast as the four-word-bootstrap for RL02 disks that uses some
power-up defaults to initialize the controller registers. In such a
case, the sequence would become longer, but it should still be possible.
So, what does your dialog normally look like when you boot your /23?
Regards,
Andreas
Jerome Fine wrote:
--
Andreas Freiherr
Vishay Semiconductor GmbH, Heilbronn, Germany
http://www.vishay.com
>Tom Leffingwell wrote:
...
Also,
from the ODT prompt on my 11/23, how do you get it to boot
from the other (DY1) RX02?
Thanks,
Tom
You are probably entering the code by hand. Somewhere, you have specified
a default unit of ZERO. If you have the code, one of us could probably change
the unit number for you.
4:41 a.m.
The first message I normally see is RT-11 loading...I never have to do
anything, it just boots from DY0:. If I don't have a disk in, or its
messed up for whatever reason, I get:
173326
@
I looked a little closer inside the unit, and it doesn't seem to have a
belt. There's a worm actuator that shakes when on power up when it tries
to move back to its starting place. It can be moved by hand easily, so
it doesn't seem to be binding.
So, what does your dialog normally look like when you
boot your /23?
Regards,
Andreas
Jerome Fine wrote:
--
Andreas Freiherr
Vishay Semiconductor GmbH, Heilbronn, Germany
http://www.vishay.com
>Tom Leffingwell wrote:
...
Also, from the ODT prompt on my 11/23, how do you get it to boot
from the other (DY1) RX02?
Thanks,
Tom
You are probably entering the code by hand. Somewhere, you have specified
a default unit of ZERO. If you have the code, one of us could probably change
the unit number for you.
5:24 a.m.
Tom Leffingwell wrote:
The first message I normally see is RT-11 loading...I
never have to do
anything, it just boots from DY0:. If I don't have a disk in, or its
messed up for whatever reason, I get:
Jerome Fine replies:
A while back, someone mentioned a DSD 440 with a neat trick -
on power up, it reads block zero of drive zero into memory
at address zero - when a time-out occurs - long enough to
read the block, it transfers control to address zero.
Probably I don't remember the details exactly, but perhaps
it is close. If so, you don't have to enter the program
by hand. Which is why it seems to boot automatically.
But there are boot programs for the RX02
which would allow the unit number to be entered at
the time. Since you have access to ODT, that would
allow you to enter the program.
Right now, I don't have access to all that old stuff, but
if I come across it, I will let you know more.
6:58 a.m.
Tom Leffingwell wrote:
The first message I normally see is RT-11 loading...I never have to do
anything, it just boots from DY0:. If I don't have a disk in, or its
messed up for whatever reason, I get:
173326
@
Sounds much like some kind of "turnkey" system that automatically
executes a ROM-based boot routine upon power-up. Depending on what ROM
that is (there are several options), you may be able to change the unit
number by moving some jumper or DIP switch on the CPU board (which
module no, in terms of Mnnnn?) or a separate board (ISTR there is
something like a bootstrap/terminator module)? - Maybe I can find out by
the halt address (173326).
Anyway, I'll go home now (currently sitting in my office) and try to
find a procedure that will at least get you up and running again. Maybe
life gets easier if we can identify the modules that take care of your
boot.
I think, however, RX02 bootstraps were pretty long to type in...
Regards,
Andreas
--
Andreas Freiherr
Vishay Semiconductor GmbH, Heilbronn, Germany
http://www.vishay.com
8:53 a.m.
Thanks...the module its attached to doesn't have an M number (its not a
DEC module) or any number for that matter. It does say Data Systems on
it, so it seems to be a matched pair with the DSD440 unit. I don't have a
BDV11 or anything like that installed, so I presume that the DSD module is
doing it itself.
On Tue, 26 Feb 2002, Andreas Freiherr wrote:
Sounds much like some kind of "turnkey"
system that automatically
executes a ROM-based boot routine upon power-up. Depending on what ROM
that is (there are several options), you may be able to change the unit
number by moving some jumper or DIP switch on the CPU board (which
module no, in terms of Mnnnn?) or a separate board (ISTR there is
something like a bootstrap/terminator module)? - Maybe I can find out by
the halt address (173326).
Anyway, I'll go home now (currently sitting in my office) and try to
find a procedure that will at least get you up and running again. Maybe
life gets easier if we can identify the modules that take care of your
boot.
I think, however, RX02 bootstraps were pretty long to type in...
Regards,
Andreas
--
Andreas Freiherr
Vishay Semiconductor GmbH, Heilbronn, Germany
http://www.vishay.com
10 a.m.
Andreas Freiherr wrote:
Sounds much like some kind of "turnkey"
system that automatically
executes a ROM-based boot routine upon power-up. Depending on what ROM
that is (there are several options), you may be able to change the unit
number by moving some jumper or DIP switch on the CPU board (which
module no, in terms of Mnnnn?) or a separate board (ISTR there is
something like a bootstrap/terminator module)? - Maybe I can find out by
the halt address (173326).
Jerome Fine replies:
I seem to remember someone saying that the DSD 440 reads block zero
of drive zero on power on, then times out and jumps to location zero
to start.
Anyway, I'll go home now (currently sitting in my
office) and try to
find a procedure that will at least get you up and running again. Maybe
life gets easier if we can identify the modules that take care of your
boot.
Might help - and I also usually ask as well, but in this case probably
will not.
I think, however, RX02 bootstraps were pretty long to
type in...
Regards,
Andreas Freiherr
Vishay Semiconductor GmbH, Heilbronn, Germany
http://www.vishay.com
Try typing in the MSCP bootstrap.
The standard RX02 bootstrap uses UNIT=0, so it may need to
be modified to use UNIT=1 for DY1:, however that stuff is not
accessible to me right now. Does anyone have the release
notes for RT-11? It may be there.
Sincerely yours,
Jerome Fine
27 Feb
27 Feb
6:33 a.m.
Tom Leffingwell wrote:
I have a DSD440 dual RX02 unit, and the first drive abruptly quit.
The disk access LED just flashes. When you first power it on, you can
hear a slight clicking noise when it first starts spinning. It does spin
though. Right after this noise, the LED starts flashing, and the drive
won't function. Does anyone have any ideas on what might be wrong and
what I can do about it?
Also, from the ODT prompt on my 11/23, how do you get it to boot
from the other (DY1) RX02?
Thanks,
Tom
Tom,
you do realize that a DSD440 has exhaustive self-tests built-in ?
If you remove the lid and power up the system you should see the error-code in
the row of 8 leds on the controller board.
( the one in the DSD440, not the QBUS controller )
If you do not have the documentation I could help you out with a copy of the
led decoding chart.
Anybody has a spare omnibus controller for the DSD440 ?.....
Jos Dreesen
11:58 a.m.
I noticed the LEDs, but I don't have the documentation. I'm not near the
unit right now, but I'll make a note to check it.
Do you have (or anyone else) any docs regarding the settings of the
jumpers on each of the drives themselves? I took them out and noticed
that most of the jumpers were quite different between the two of them. I
changed the second drive from jumper "DS2" to "DS1", and it booted off
the
second drive, although still don't understand why all the other settings
are different.
-Tom
On Wed, 27 Feb 2002, Jos Dreesen wrote:
Tom,
you do realize that a DSD440 has exhaustive self-tests built-in ?
If you remove the lid and power up the system you should see the
error-code in the row of 8 leds on the controller board. ( the one in
the DSD440, not the QBUS controller )
If you do not have the documentation I could help you out with a copy
of the led decoding chart.
Anybody has a spare omnibus controller for the DSD440 ?.....
Jos Dreesen
28 Feb
28 Feb
5:31 a.m.
New subject: RX02 / DSD440 problemt
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 727 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020228/ade8c071/at…
4 Mar
4 Mar
6:45 a.m.
I've narrowed it down to the either the board in the DSD-440
chassis itself, or the board that's part of the individual drive. It
seems that besides the ID number of the drive, there is some sort of bus
address that is set differently on each RX02 in the unit. Jumpers T1, T3,
T4, T5, T6, and D are set differently between the two drives, besides the
DS1 and DS2 setting.
Do you have docs for the jumper settings?
Thanks,
Tom
On Wed, 27 Feb 2002, Jos Dreesen wrote:
If you do not have the documentation I could help you
out with a copy of the
led decoding chart.
9:35 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1116 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020304/b5a4eb16/at…
12:35 p.m.
On Mon, 4 Mar 2002, Tony Duell wrote:
This sounds suspiciously like a Shugart SA801 drive.
Is there a label on
the drive chassis itself? If it is an SA801, what is the part number of
the PCB?
Thanks! It is a Shugart 801. I thought I mentioned it in my email, but
appartently, I didn't. I don't have the drive in front of me right now,
but I'll grab the number off of it later.
D on 1 of the drives (the first one) is in. One the second drive, the pin
furtherest from the edge of the board is wire wrapped to the center pin of
of the 3-pin DC. Does that make sense?
So basically, there isn't any difference between the two drives in terms
of addressing besides DS1 and DS2 jumpers. That means my problem is on
the board on the bottom of the drive itself. And something tells me it
will end up being an IC that I can't get a replacement for. The actuator
that moves the head doesn't want to go back to track zero on power up.
It vibrates back and forth about a couple of steps and then stops. I
checked the stepper with an ohm meter, and board-swapped between the two
drives.
Still assuming it's an SA801, then
T1 = terminate head load line
(T2 = terminate drive select -- presumably fitted on both drives)
T3 = terminate step direction
T4 = terminate step pulse
T5 = terminate write data
T6 = terminate write gate
D = connect pin 16 to in-use LED input
The termination jumpers should be fitted in the last drive on the cable
only (no matter what it's ID is). I can't see why D would be set
differently on the 2 drives, though.
5 Mar
5 Mar
10:14 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 2359 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020306/1331fc23/at…
12:32 p.m.
New subject: RX02 / DSD440 / SA801 problem
On Wed, 6 Mar 2002, Tony Duell wrote:
Good point...I added it to the subject already. :)
If you're asking about a 'generic' module
(drive, PSU, etc) that may
well have been used in other machines, then post any identification
that's on that module.
I forgot to get the version off the board last night, and unfortunately I
left it the lab today, so I'll have to get it later. Anyway, it does have
a socketed 40-pin ASIC on it right in the center. The only standard
devices I noticed where some 74xx ones, and an LM339 quad comparator. I
didn't notice any transistors, and I was sort of wondering how the stepper
was driven. I would think that if whatever is driving it is blown, then
it wouldn't move at all.
I had considered swapping the socketed 40-pin IC between the working and
non-working unit, but I don't have proper IC extraction tools, and if its
got really flimsy pins on it, I might screw up my only working unit.
I have schematics for 4 versions of the logic
board...
The later ones have a 40 pin ASIC on them. The first worry is that the
stepper control circuit is mostly inside this ASIC, but the actual motor
drivers aren't (they're sections of a ULN2047 chip). The older versions
of the board have the stepper motor circuit built from TTL (which you can
still get) with discrete power transistors for the motor drivers.
Let me know which version board (or at least if theres a 40 pin chip on
it, or if there are 3 TO66-can power transistors near one edge) and I
will see what I can do to help you debug it.
-tony
6 Mar
6 Mar
6:16 a.m.
New subject: RX02 / DSD440 / SA801 problem
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 3704 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020306/a17b0617/at…
7 Mar
7 Mar
3:20 a.m.
New subject: RX02 / DSD440 / SA801 problem
On Wed, 6 Mar 2002, Tony Duell wrote:
There are 2 board versions that use the 40 pin ASIC
shown in my manual --
25136 and 25229. They are pretty similar, so I can give some more hints
withoug further information
I have the 25229.
OK, what I'd do is use a logic probe to look at
pins 25,24,23 of the ASIC
while the head is supposed to be moving. You should see square waves on
all of them. Of course if you have a 3 channel (or more) logic analyser
you can make sure they actually look like stepper motor drive waveforms.
All I have is an HP 54615B 2-channel scope, so I can't look at all three
at the same time, but they all had square waves while it was attempting to
move.
Otherwise, look at the outputs of the appropriate
sections of 1B. If one
of those isn't switching, change the 7404 (this is not a likely fault,
though).
I got square waves out of here. I didn't compare the inputs and outputs,
I just checked to see that a square wave is coming out, instead of seeing
if it was inverted. I've never seen one fail like that though, but I
could go back and check it.
Finally look at the collectors of the appropiate
sections of 1C. Most
likely one is never going low, in which case change the ULN2074.
I checked pins 8, 9, and 16; if I read your diagram right, those should be
the collectors. All them produced the square wave as well.
You mean you don't have a small screwdriver?
I've removed dozens of ICs
with just a small screwdriver. The trick is to work from both ends of the
socket and keep the chip as level as possible. That way you won't bend
the pins.
I do...the last time I used it was on a set of RDC19202-303's (resolver to
digital converters). They cost about $1,800. I didn't mess anything up,
but came close. At that point I decided I'd quit using the screw driver
method on things that aren't easily (or cheaply) replaceable.
However, I think it's time to stop swapping parts
around and to actually
use some test equipment.
I agree...
Any idea on what I should try next?
Thanks,
Tom
9:01 a.m.
New subject: RX02 / DSD440 / SA801 problem
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 2565 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020307/305e555f/at…
10:57 a.m.
New subject: RX02 / DSD440 / SA801 problem
On Thu, 7 Mar 2002, Tony Duell wrote:
OK. It sounds like the ASIC is working correctly. Do
this look like
stepper drive waveforms (3 similar signals with a phase offset between
them)?
As far as I can tell, although I've never worked with steppers before.
Is this with the motor connected?
Yes
If you're getting square waves at the driver
collectors with the motor
connected then either the motor is faulty (but you've board-swapped
between the 2 drives so you know it's OK), or there's an open
connection somewhere.
I checked the connections. Maybe there's a bad diode in there somewhere?
Any other ideas?
Thanks,
Tom
8 Mar
8 Mar
6:21 a.m.
New subject: RX02 / DSD440 / SA801 problem
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1381 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020308/9fe0318f/at…
12:10 p.m.
New subject: RX02 / DSD440 / SA801 problem
On Fri, 8 Mar 2002, Tony Duell wrote:
What you are saying is that you have a good stepper
motor connected to a
good head positioner mechanism. You know it's good because the other
logic board will operate it.
Correct.
You have a logic board that is outputting the right
signals to the motor
Well, its outputing signals...I wouldn't go so far as to say they are
correct.
And yet the motor doesn't run properly.
Nope...
Go back to the good drive. While the head is moving,
look at the stepper
drive singals at the pins of the ASIC (23, 24, 25 IIRC) and at the
collectors of the ULN2074. In particular look at the mark-space ratio and
freqeuncy.
Now go to the faulty drive and look at the same signals. What (if any)
differences do you see?
I'll check it out. What if the relationship isn't correct, like two of
the pulses are on top of each other? Is all the logic for generating the
pulses contained in the ASIC, or is it in turn driven by something
else? The ASIC from the bad system has no problem running the good drive,
so that's not a problem.
6:21 a.m.
New subject: RX02 / DSD440 / SA801 problem
Tom Leffingwell schrieb:
I'll check it out. What if the relationship
isn't correct, like two of
the pulses are on top of each other? Is all the logic for generating the
pulses contained in the ASIC, or is it in turn driven by something
else? The ASIC from the bad system has no problem running the good drive,
so that's not a problem.
Time to look at the DSD440 controller board...
The green leds ( 1 and 9 ) should be on.
Leds 2-3-4 are status leds :
Led 2 : Controller waiting for interface.
Led 3 : controller writing
Led 4 : controller reading
Leds 5 to 8 should be off : they represent a binary errorcode.
I can look them up for you if required...
Jos Dreesen
5:43 a.m.
New subject: RX02 / DSD440 / SA801 problem
Sometimes its just 6 and 8 that come on, other times 5 comes on as well.
Thanks,
Tom
On Fri, 8 Mar 2002, Jos Dreesen wrote:
Time to look at the DSD440 controller board...
The green leds ( 1 and 9 ) should be on.
Leds 2-3-4 are status leds :
Led 2 : Controller waiting for interface.
Led 3 : controller writing
Led 4 : controller reading
Leds 5 to 8 should be off : they represent a binary errorcode.
I can look them up for you if required...
Jos Dreesen
11 Mar
11 Mar
7:05 a.m.
New subject: RX02 / DSD440 / SA801 problem
Tom Leffingwell schrieb:
Sometimes its just 6 and 8 that come on, other times 5 comes on as well.
Thanks,
Tom
6 and 8 on : "indeterminate density" .
Maybe the floppydisc itself is marginal ?
5, 6 and 8 : "invalid drive status during seek"
Give me a fax number off-list and I'll fax you the complete LED-decoding
list.
Does anyone have a copy of the DSD440 diagnostics disc ?
Rgds, Jos
12 Mar
12 Mar
11:35 a.m.
New subject: RX02 / DSD440 / SA801 problem
On Mon, 11 Mar 2002, Jos Dreesen wrote:
6 and 8 on : "indeterminate density" .
Maybe the floppydisc itself is marginal ?
There's nothing in the drive when this is happening. Maybe when it tries
to move, its checking for density?
5, 6 and 8 : "invalid drive status during
seek"
Give me a fax number off-list and I'll fax you the complete LED-decoding
list.
Cool, thanks!
-Tom
9 Mar
9 Mar
7:48 a.m.
New subject: RX02 / DSD440 / SA801 problem
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 2187 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20020309/e497afb8/at…
12 Mar
12 Mar
5:50 a.m.
New subject: RX02 / DSD440 / SA801 problem
On Sat, 9 Mar 2002, Tony Duell wrote:
It's virtually all in th ASIC. There are some
relevant inputs :
Pin 33 (Step pulse input)
Pin 29 (Step direction input)
Pin 34 (Step Enable /)
Pin 27 (Output Enable) ?
You might want to compare the states of the 2 enable pins between the 2
drives. Just in case there are problems there.
They appeared to be the same on both drives, except for a slightly lower
voltage on pin 34 on the good unit (3.7 on the good one, 4.9 on the bad),
while trying to move.
Also, it appears I have an intermittent problem. After awhile, I noticed
the drive was actually positioning correctly. I went to connect it back
to the PDP, and it worked for about 10 power cycles. Then it got to where
it wouldn't work when connected to the PDP, and would work without
it. After a few more times, it got back to where we are now--where it
doesn't work at all. All the connections seem pretty good. The unit was
physically moved during the above events.
2 other areas to investigate. Firstly the clock for
the ASIC. This is the
Xtal (Y1) and associated bits connected to pins 19 and 20. Look for a
5MHz oscillation on at least one of those pins with a 'scope.
I have 5Mhz on both pins.
And POR (Power On Reset), on pin 14. This should be
low briefly after
power-on and then go high. Check the capacitor connected to this pin ( on
the 25229 board, this is C21, .47uF) and the power fail comparator
circuit (on the 25229 board, it's pins 10, 11, 13 of the LM339).
Pin 14 does as you indicate. C21 seems okay. The LM339 is fine, although
I only checked the section you mentioned.
-Tom
8 Mar
8 Mar
6:57 a.m.
New subject: RX02 / DSD440 / SA801 problem
Lost for other ideas, I decided to swap the 40-pin ASIC, and it
checked out okay. What tells the ASIC to start outputting pulses? It
seems I have a problem upstream.
Thanks again,
Tom
8103
days inactive
8117
days old
29 comments
5 participants
participants (5)
-
Andreas.Freiherr＠Vishay.com -
ard＠p850ug1.demon.co.uk -
jhfine＠idirect.com -
jos.mar＠bluewin.ch -
tom＠sba.miami.edu