22 Jan
2017
22 Jan
'17
11:27 a.m.
Hi friends.
I am 100% new to my pdp8e and I am troubleshooting a problem.
The problem is that whenever address bit 7 I asserted, I also see MD bit 4 asserted.
I am hoping there is a simple short somehow between these lines somewhere. I should
mention that my setup has known working boards with the exception of M8310 does not work
(all of my boards were tested in another machine one by one). So I am hoping the problem
noted above might be occurring on the M8310 board itself.
I found a document that describes the signals on the bus located here:
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
I noticed that B1J (MA7) is directly adjacent to B1K (MD4). Could this be related?
So I decided to pull my 8310 and check the resistance between these signals on the edge of
the board. However, I cannot seem to square the signals named in the PDF and what I see on
the card edge connector on the board.
I am wondering if I am not understanding the PDF correctly. I have the board sitting on my
table with the components facing up and the omnibus card edge is at the bottom of the
card. Going from left to right, do I have connectors ABCD or is it something else? Maybe
DCBA?
Other than the reference PDF, is there another PDF that has a more detailed description of
the bus and the signals?
Thanks
Eugene
22 Jan
22 Jan
3:17 p.m.
On Sun, Jan 22, 2017 at 4:27 PM, W2HX <w2hx at w2hx.com> wrote:
Hi friends.
I am 100% new to my pdp8e and I am troubleshooting a problem.
The problem is that whenever address bit 7 I asserted, I also see MD bit 4 asserted.
I am hoping there is a simple short somehow between these lines somewhere. I
should mention that my setup has known working boards with the exception of
M8310 does not work (all of my boards were tested in another machine one by one).
So I am hoping the problem noted above might be occurring on the M8310 board itself.
Could it be a short on the backplane? Or on the jumpers between the
backplanes if you
have 2 backplanes in the box (The standard 8/e box is designed to take
2 backplanes,
I believe the rear one was optional but almost always fitted).
I found a document that describes the signals on the bus located here:
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
I noticed that B1J (MA7) is directly adjacent to B1K (MD4). Could this be related?
So I decided to pull my 8310 and check the resistance between these signals on the
edge of the board. However, I cannot seem to square the signals named in the PDF
and what I see on the card edge connector on the board.
I am wondering if I am not understanding the PDF correctly. I have the board sitting
on my table with the components facing up and the omnibus card edge is at the
bottom of the card. Going from left to right, do I have connectors ABCD or is it
something else? Maybe DCBA?
With the board like that, Connector A is on the right (so the order is
DCBA from left
to right, pin A is on the right of each connector, and side 1 is the
solder side.
Other than the reference PDF, is there another PDF that has a more detailed
description of the bus and the signals?
Do you have the Engineering Drawing/Printset (same thing, just 2
different names)
from bitsavers? There is also an excellent 3-volume
PDP8/e (and /f,
/m) maintenance
manual on bitsavers (I have it on paper too). Volume 1 covers the CPU (which is
what you need now), Volume 2 is 'internal options', Volume 3 is
'external options. In
time you will need all of them.
-tony
3:35 p.m.
On 1/22/2017 12:17 PM, Tony Duell wrote:
On Sun, Jan 22, 2017 at 4:27 PM, W2HX <w2hx at
w2hx.com> wrote:
My friend Sherman has a number of 8/E systems he is going to get going,
and has a setup consisting of a frame, PS, and backplane. He has a
couple of single slot fixtures with switches at the moment which he used
to ring out a potentially working front panel, and found a few failures
there.
Of course after checking the backplane for shorts, and the PS.
He went to the M8330 (Timing board) and has fixed some things on that
board. he will cycle all board sets thru this system setup once he gets
all the way to the processor.
As tony says, I'd start very basic with the backplane. Especially if
there is proximity of the signals there for your specific problem.
One system will be all scrap boards including a front panel which he
retrieved just a tad too late from a scrap operation that had clipped
the flip chip connectors. A strap on Douglas board and rewiring will
fix that one.
There are a lot of places in a system with most of the boards present
that you can have failures, so hard to give a lot of suggestions.
I also hope the other main thing to test, the power supply was checked
out and isn't playing games with you, though it doesn't seem likely it
would make the short you suggest.
thanks
Jim
Hi friends.
I am 100% new to my pdp8e and I am troubleshooting a problem.
The problem is that whenever address bit 7 I asserted, I also see MD bit 4 asserted.
I am hoping there is a simple short somehow between these lines somewhere. I
should mention that my setup has known working boards with the exception of
M8310 does not work (all of my boards were tested in another machine one by one).
So I am hoping the problem noted above might be occurring on the M8310 board itself.
Could it be a short on the backplane? Or on the jumpers between the
backplanes if you
have 2 backplanes in the box (The standard 8/e box is designed to take
2 backplanes,
I believe the rear one was optional but almost always fitted).
I found a
document that describes the signals on the bus located here:
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
I noticed that B1J (MA7) is directly adjacent to B1K (MD4). Could this be related?
So I decided to pull my 8310 and check the resistance between these signals on the
edge of the board. However, I cannot seem to square the signals named in the PDF
and what I see on the card edge connector on the board.
I am wondering if I am not understanding the PDF correctly. I have the board sitting
on my table with the components facing up and the omnibus card edge is at the
bottom of the card. Going from left to right, do I have connectors ABCD or is it
something else? Maybe DCBA?
With the board like that, Connector A is on the right
(so the order is
DCBA from left
to right, pin A is on the right of each connector, and side 1 is the
solder side.
Other than the reference PDF, is there another
PDF that has a more detailed
description of the bus and the signals?
Do you have the Engineering
Drawing/Printset (same thing, just 2
different names)
from bitsavers? There is also an excellent 3-volume PDP8/e (and /f,
/m) maintenance
manual on bitsavers (I have it on paper too). Volume 1 covers the CPU (which is
what you need now), Volume 2 is 'internal options', Volume 3 is
'external options. In
time you will need all of them.
-tony
4:05 p.m.
Hi Eugene,
You might want to test the over the top connectors. I don't remember the
part #, but they do go bad. They can be labeled and swapped with a spare if
have one. Just look to see if anything has changed after every change.
Dirty contacts and bad connections are always a problem.
I would remove the backplane. The foam under it is probably goo. I like to
lay the backplane upside on white paper, and lift one side up about an inch
and drop it.
This loosens up a lot of junk that air and other things don't get out.It's
amazing what you might find on the paper.
clean and inspect before replacing.
Good Luck!!
On Sun, Jan 22, 2017 at 10:27 AM, W2HX <w2hx at w2hx.com> wrote:
Hi friends.
I am 100% new to my pdp8e and I am troubleshooting a problem.
The problem is that whenever address bit 7 I asserted, I also see MD bit 4
asserted.
I am hoping there is a simple short somehow between these lines somewhere.
I should mention that my setup has known working boards with the exception
of M8310 does not work (all of my boards were tested in another machine one
by one). So I am hoping the problem noted above might be occurring on the
M8310 board itself.
I found a document that describes the signals on the bus located here:
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/
pdp8e/Omnibus_legenda.pdf
I noticed that B1J (MA7) is directly adjacent to B1K (MD4). Could this be
related?
So I decided to pull my 8310 and check the resistance between these
signals on the edge of the board. However, I cannot seem to square the
signals named in the PDF and what I see on the card edge connector on the
board.
I am wondering if I am not understanding the PDF correctly. I have the
board sitting on my table with the components facing up and the omnibus
card edge is at the bottom of the card. Going from left to right, do I
have connectors ABCD or is it something else? Maybe DCBA?
Other than the reference PDF, is there another PDF that has a more
detailed description of the bus and the signals?
Thanks
Eugene
5:05 p.m.
Thanks to several who have made suggestions. Noel, Tony, Jim, and Paul thank you. The
system has been very thoroughly cleaned and the PS has been load tested. The front panel
has had some bulbs replaced as well.
I took all my boards (except the front panel) to a friend's house with a working 8e
and tested each board in succession. I found that most boards worked but M8310 was not.
And 8K of 32K was working fine. Above 8K there was a stuck bit in the core. But with 8K
working we were able to boot OS8 with all of my boards except M8310 (oh, and I borrowed
his serial tty card).
I did know about these MA7-MD4 bits that seem stuck together in my machine. But we did not
see this problem in my friend's machine. So I am concluding the problem could be in
one of 3 possible places. The 8310 card, the front panel, the bus itself. Because these
are the three components that were not part of the test in my friend's machine.
Next step was to see if these two bus lines are shorted somehow. After wasting time
looking at a document that was wrong, I checked the M8310 card for shorts between MA7 and
MD4 and none existed.
I am waiting on a douglas electronics card extender in the mail to check the bus itself.
This issue could still be in the front panel or in some logic further upstream. This will
be interesting to troubleshoot.
FYI
That omnibus_legenda.pdf file on bitsavers is CLEARLY erroneous based on what you guys are
telling me and I found another document that indeed indicates that file is wrong. Too bad
that this file is stored on bitsavers when it is incorrect.
I hope this email can be found by the next guy searching for the omnibus details so they
can avoid wasting time.
THIS FILE IS WRONG
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
This file shows the Omnibus signal locations CORRECTLY (extracted from a larger document)
http://w2hx.com/x/VintageComp/PDP-8e/Docs/Omnibus_Card_Edge_Designations.pdf
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of W2HX
Sent: Sunday, January 22, 2017 11:28 AM
To: General Discussion: On-Topic and Off-Topic Posts (cctalk at classiccmp.org)
Subject: quick omnibus question
Hi friends.
I am 100% new to my pdp8e and I am troubleshooting a problem.
The problem is that whenever address bit 7 I asserted, I also see MD bit 4 asserted.
I am hoping there is a simple short somehow between these lines somewhere. I should
mention that my setup has known working boards with the exception of M8310 does not work
(all of my boards were tested in another machine one by one). So I am hoping the problem
noted above might be occurring on the M8310 board itself.
I found a document that describes the signals on the bus located here:
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
I noticed that B1J (MA7) is directly adjacent to B1K (MD4). Could this be related?
So I decided to pull my 8310 and check the resistance between these signals on the edge of
the board. However, I cannot seem to square the signals named in the PDF and what I see on
the card edge connector on the board.
I am wondering if I am not understanding the PDF correctly. I have the board sitting on my
table with the components facing up and the omnibus card edge is at the bottom of the
card. Going from left to right, do I have connectors ABCD or is it something else? Maybe
DCBA?
Other than the reference PDF, is there another PDF that has a more detailed description of
the bus and the signals?
Thanks
Eugene
5:41 p.m.
I would also do a cleaning of your empty backplane with a good vacuum cleaner
nozzle. It is not uncommon for bits of conductive detritus to fall into the
backplane over time. Given that the pins are adjacent on the connector it might
be a possible cause.
Vacuuming out any stuff that has accumulated over time will be beneficial in any
event. I have had mysterious 'opens' occur on a card slot when an old sticky
label that lost is stick fell into the slot, and got wedged between a card pin
and the connector.
Don
On 1/22/2017 2:05 PM, W2HX wrote:
Thanks to several who have made suggestions. Noel,
Tony, Jim, and Paul thank you. The system has been very thoroughly cleaned and the PS has
been load tested. The front panel has had some bulbs replaced as well.
I took all my boards (except the front panel) to a friend's house with a working 8e
and tested each board in succession. I found that most boards worked but M8310 was not.
And 8K of 32K was working fine. Above 8K there was a stuck bit in the core. But with 8K
working we were able to boot OS8 with all of my boards except M8310 (oh, and I borrowed
his serial tty card).
I did know about these MA7-MD4 bits that seem stuck together in my machine. But we did
not see this problem in my friend's machine. So I am concluding the problem could be
in one of 3 possible places. The 8310 card, the front panel, the bus itself. Because
these are the three components that were not part of the test in my friend's machine.
Next step was to see if these two bus lines are shorted somehow. After wasting time
looking at a document that was wrong, I checked the M8310 card for shorts between MA7 and
MD4 and none existed.
I am waiting on a douglas electronics card extender in the mail to check the bus itself.
This issue could still be in the front panel or in some logic further upstream. This will
be interesting to troubleshoot.
FYI
That omnibus_legenda.pdf file on bitsavers is CLEARLY erroneous based on what you guys
are telling me and I found another document that indeed indicates that file is wrong. Too
bad that this file is stored on bitsavers when it is incorrect.
I hope this email can be found by the next guy searching for the omnibus details so they
can avoid wasting time.
THIS FILE IS WRONG
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
This file shows the Omnibus signal locations CORRECTLY (extracted from a larger
document)
http://w2hx.com/x/VintageComp/PDP-8e/Docs/Omnibus_Card_Edge_Designations.pdf
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of W2HX
Sent: Sunday, January 22, 2017 11:28 AM
To: General Discussion: On-Topic and Off-Topic Posts (cctalk at classiccmp.org)
Subject: quick omnibus question
Hi friends.
I am 100% new to my pdp8e and I am troubleshooting a problem.
The problem is that whenever address bit 7 I asserted, I also see MD bit 4 asserted.
I am hoping there is a simple short somehow between these lines somewhere. I should
mention that my setup has known working boards with the exception of M8310 does not work
(all of my boards were tested in another machine one by one). So I am hoping the problem
noted above might be occurring on the M8310 board itself.
I found a document that describes the signals on the bus located here:
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
I noticed that B1J (MA7) is directly adjacent to B1K (MD4). Could this be related?
So I decided to pull my 8310 and check the resistance between these signals on the edge
of the board. However, I cannot seem to square the signals named in the PDF and what I see
on the card edge connector on the board.
I am wondering if I am not understanding the PDF correctly. I have the board sitting on
my table with the components facing up and the omnibus card edge is at the bottom of the
card. Going from left to right, do I have connectors ABCD or is it something else? Maybe
DCBA?
Other than the reference PDF, is there another PDF that has a more detailed description
of the bus and the signals?
Thanks
Eugene
5:59 p.m.
On 1/22/17 2:05 PM, W2HX wrote:
I hope this email can be found by the next guy
searching for the omnibus details so they can avoid wasting time.
THIS FILE IS WRONG
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibus_leg…
This file shows the Omnibus signal locations CORRECTLY (extracted from a larger
document)
http://w2hx.com/x/VintageComp/PDP-8e/Docs/Omnibus_Card_Edge_Designations.pdf
i will delete the file
here is the original source
http://www.classiccmp.org/pipermail/cctalk/2015-June/007475.html
6:15 p.m.
Before deleting this, someone much smarter than me should double check to make sure I am
right that the file is incorrect. It is a shame because it is very nicely done. It even
shows the keyways in the card. But it seems to be backward to me.
I tried to find contact information for that hack42.nl site to suggest they redo the
otherwise useful file, but was not successful.
Thank you
Eugene
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Al Kossow
Sent: Sunday, January 22, 2017 5:59 PM
To: cctalk at classiccmp.org
Subject: Re: quick omnibus question
On 1/22/17 2:05 PM, W2HX wrote:
I hope this email can be found by the next guy
searching for the omnibus details so they can avoid wasting time.
THIS FILE IS WRONG
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibu
s_legenda.pdf
This file shows the Omnibus signal locations CORRECTLY (extracted from
a larger document)
http://w2hx.com/x/VintageComp/PDP-8e/Docs/Omnibus_Card_Edge_Designatio
ns.pdf
i will delete the file
here is the original source
http://www.classiccmp.org/pipermail/cctalk/2015-June/007475.html
7:14 p.m.
The two presentations look to be consistent to me, right down to the keyways.
The bitsavers/hack42 view is oriented is as if you held a board vertically with the
connectors edge pointing towards you and the components on your right, the two faces are
then folded out towards you.
The connectors go up A to D.
This view would also be a view from the bottom of the backplane.
The 8e/Docs 2-page extract is 180 degrees rotated from this.
So the connectors now go down A to D.
On 2017-Jan-22, at 3:15 PM, W2HX wrote:
Before deleting this, someone much smarter than me
should double check to make sure I am right that the file is incorrect. It is a shame
because it is very nicely done. It even shows the keyways in the card. But it seems to be
backward to me.
I tried to find contact information for that hack42.nl site to suggest they redo the
otherwise useful file, but was not successful.
Thank you
Eugene
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Al Kossow
Sent: Sunday, January 22, 2017 5:59 PM
To: cctalk at classiccmp.org
Subject: Re: quick omnibus question
On 1/22/17 2:05 PM, W2HX wrote:
I hope this email can be found by the next guy
searching for the omnibus details so they can avoid wasting time.
THIS FILE IS WRONG
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp8/pdp8e/Omnibu
s_legenda.pdf
This file shows the Omnibus signal locations CORRECTLY (extracted from
a larger document)
http://w2hx.com/x/VintageComp/PDP-8e/Docs/Omnibus_Card_Edge_Designatio
ns.pdf
i will delete the file
here is the original source
http://www.classiccmp.org/pipermail/cctalk/2015-June/007475.html
2771
days inactive
2772
days old
8 comments
7 participants
participants (7)
-
aek＠bitsavers.org -
ard.p850ug1＠gmail.com -
bhilpert＠shaw.ca -
jwsmail＠jwsss.com -
north＠alum.mit.edu -
useddec＠gmail.com -
w2hx＠w2hx.com