[TUHS] PDP 11/34a Serial Communications Problem

[TUHS] PDP 11/34a Serial Communications Problem

[Mark Hare]

Hello all,

This is my first time emailing the list, so please let me know if this
doesn't belong here of if I'm breaking any rules.

A few months ago, I rescued a PDP 11/34a with 2 RL01 drives from the scrap
heap. The unit appears to work fine based on my limited front-panel
testing. I haven't gotten the drives running yet since someone cut the
power cords when the cabinet was being removed.

There is a DL11-W serial line unit/realtime clock (M7856) installed in the
11/34 that I want to use for serial input/output. I have configured the
card for 9600 baud, 8N1. Using some jumper wires, I carefully connected the
card to a serial cable and a computer running a terminal and I was able to
send some characters back and forth successfully.

For a more permanent solution, I designed a simple adapter board that
connects to the BERG 40 connector on the DL11-W and converts it to a DB9
serial port (In restrospect, this product was already available at
https://oshpark.com/shared_projects/uTMf3v08 but I didn't know about that
at the time). I also ordered a 40-pin (non-IDE) ribbon cable to connect the
DL11-W to my adapter.

When I connected everything, the 11/34 would start but no lights would
appear on the front panel. I tried disconnecting the adapter but leaving
the ribbon cable plugged into the BERG connector, but the problem
persisted. When I removed the ribbon cable entirely, the unit powered on
with no problems.

Since this is a straight-through ribbon cable, I don't see what could be
causing this problem. I have checked the continuity of each wire in the
cable, and there doesn't appear to be a problem. I'd appreciate any advice
that anyone has to offer.

Yours,

Mark D. Hare

markhare at buffalo.edu
University at Buffalo
B. S. Civil Engineering '16
M. S. Structural/Earthquake Engineering Student
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20170205/7877de49/attachment.html>

[TUHS] PDP 11/34a Serial Communications Problem

[Michael Kjörling]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1312 bytes --]

On 5 Feb 2017 15:03 -0500, from markhare at buffalo.edu (Mark Hare):
> Since this is a straight-through ribbon cable, I don't see what could be
> causing this problem. I have checked the continuity of each wire in the
> cable, and there doesn't appear to be a problem. I'd appreciate any advice
> that anyone has to offer.

Have you checked to make sure that there aren't any connections
between the individual wires? I'm not particularly familiar with the
hardware of the PDP-11, but a signal going out on one wire and showing
up also on some should-be unrelated wire could conceivably cause all
kinds of weird behavior depending on exactly which wires are involved
and what they are used for.

The fact that merely leaving the ribbon cable connected causes the
problem to appear definitely points to the PDP-11 not agreeing with
the cable (or connections) for some reason. My first suspect in that
case would be an electrical problem with the cable itself _versus what
the PDP-11 wants it like_. Maybe the cable is operating to
specifications, but the '11 wants something different.

-- 
Michael Kjörling • https://michael.kjorling.se • michael at kjorling.se
                 “People who think they know everything really annoy
                 those of us who know we don’t.” (Bjarne Stroustrup)

[TUHS] PDP 11/34a Serial Communications Problem

[Noel Chiappa]

    > From: Mark Hare <markhare at buffalo.edu>

    > I believe I have found the problem.

Excellent!

    > Looking at the DL11-W, I saw that there is an uncovered trace on the
    > circuit board just below the pins of the BERG connector.  It appears
    > that the exposed conductor ends of the cable were making contact with
    > this trace, which shorted the entire cable together.

Shorting any set of DL11-W pins together still shouldn't have caused that
symptom, AFAICS.

I'm too lazy to pull out a DL11-W, and prints thereof, to check, but I suspect
that what actually happened is that trace carries some 'interesting' signal,
and it got grounded 'or something' by the exposed ends of the flat cable.

    > I'm just glad that there doesn't appear to be any lasting damage.

Indeed! I was worried about that...

	 Noel

[TUHS] PDP 11/34a Serial Communications Problem

[Mark Hare]

I believe I have found the problem. Upon closer inspection, I noticed
that the ribbon cable that I am using is terminated in such a way that
the conductors are exposed on the outside of the top of the female
connector. Looking at the DL11-W, I saw that there is an uncovered
trace on the circuit board just below the pins of the BERG connector.
It appears that the exposed conductor ends of the cable were making
contact with this trace, which shorted the entire cable together. I
covered the exposed ends with a piece of electrical tape and the whole
setup now works precisely as intended.

Frankly, I'm just glad that there doesn't appear to be any lasting damage.

Thank you Michael and Noel for your help; I'm sure I'll be back with
more questions as I make progress with this machine.

Yours,

Mark D. Hare

markhare at buffalo.edu
University at Buffalo
B. S. Civil Engineering '16
M. S. Structural/Earthquake Engineering Student


On Sun, Feb 5, 2017 at 3:38 PM, Noel Chiappa <jnc at mercury.lcs.mit.edu> wrote:
>     > From: Mark Hare
>
>     > For a more permanent solution, I designed a simple adapter board that
>     > connects to the BERG 40 connector on the DL11-W and converts it to a DB9
>     > serial port ...  I also ordered a 40-pin (non-IDE) ribbon cable to
>     > connect the DL11-W to my adapter.
>
>     > When I connected everything, the 11/34 would start but no lights would
>     > appear on the front panel. I tried disconnecting the adapter but leaving
>     > the ribbon cable plugged into the BERG connector, but the problem
>     > persisted. When I removed the ribbon cable entirely, the unit powered on
>     > with no problems.
>
> That's extremely odd. There isn't a pin on the DL11-W Berg connector which
> should be able to cause anything like that kind of behaviour. The only
> _possible_ thing I can think of, looking at the list of signals on the Berg,
> is that you are grounding +5 (TT). Either that, or your DL11-W has some
> serious issue?
>
>     > Since this is a straight-through ribbon cable, I don't see what could be
>     > causing this problem.
>
> Me either. But clearly it's not just a straight-through ribbon cable....
>
> I myself wouldn't have gone that route; one can obtain 40-pin IDE/DuPont (they
> are all .1" spacing pins, and basically interchangeable, module keying)
> connector shells, and female pins for same; I would have made a custom cable
> to plug into the Berg with that, to a DE-9S or DB-9P connector (depending on
> whether one wanted one wired as a DCE or DTE). (I myself make such cables, but
> to a DB-25S connector, and then use a commercial DB-25P to DE-9S adapter when
> needed.) Oh well...
>
> Does the DL11-W still work, using the jumper cables kludge?
>
>     Noel

[TUHS] PDP 11/34a Serial Communications Problem

[Noel Chiappa]

    > From: Mark Hare

    > For a more permanent solution, I designed a simple adapter board that
    > connects to the BERG 40 connector on the DL11-W and converts it to a DB9
    > serial port ...  I also ordered a 40-pin (non-IDE) ribbon cable to
    > connect the DL11-W to my adapter.

    > When I connected everything, the 11/34 would start but no lights would
    > appear on the front panel. I tried disconnecting the adapter but leaving
    > the ribbon cable plugged into the BERG connector, but the problem
    > persisted. When I removed the ribbon cable entirely, the unit powered on
    > with no problems.

That's extremely odd. There isn't a pin on the DL11-W Berg connector which
should be able to cause anything like that kind of behaviour. The only
_possible_ thing I can think of, looking at the list of signals on the Berg,
is that you are grounding +5 (TT). Either that, or your DL11-W has some
serious issue?

    > Since this is a straight-through ribbon cable, I don't see what could be
    > causing this problem.

Me either. But clearly it's not just a straight-through ribbon cable....

I myself wouldn't have gone that route; one can obtain 40-pin IDE/DuPont (they
are all .1" spacing pins, and basically interchangeable, module keying)
connector shells, and female pins for same; I would have made a custom cable
to plug into the Berg with that, to a DE-9S or DB-9P connector (depending on
whether one wanted one wired as a DCE or DTE). (I myself make such cables, but
to a DB-25S connector, and then use a commercial DB-25P to DE-9S adapter when
needed.) Oh well...

Does the DL11-W still work, using the jumper cables kludge? 

    Noel