• Please review our updated Terms and Rules here

Adding a Programmer's Console to a PDP-8/a

m_thompson

Veteran Member
Joined
Jul 8, 2014
Messages
1,985
Location
Rhode Island, USA
I have two misbehaving KC8-A PDP-8/a Programmers Consoles that I am trying to get working. The first problem that I need to fix is missing mounting hardware. There are supposed to be two Cabinet Brackets, P/N 7419709-0 and 7419709-1, mounted to the front of the chassis. Screws in the console go into the these brackets to hold the console in place. Part of the drawing from the Engineering Drawings showing the brackets is attached.

Any idea where I could get two sets of these brackets?

Programmer's Console Cabinet Bracket 7419709-0.jpg
 
Don't remember that I have seen those anywhere in mine. But they might be missing. Can't really understand how and where they fit. Do you have the full picture?

My understanding is that the programmers console is snapped on to those always breaking off knobs. Whatever the name is.
 
My programmer's consoles have screws that go upwards and towards the back from the bottom of the panel and should screw into the metal brackets.
Programmer's Console Mounting.jpg
 
My programmer's consoles have screws that go upwards and towards the back from the bottom of the panel and should screw into the metal brackets.

That sounds like a good mod to the original design which as Mattis says were only supported on 4 ‘knobs’. (Mine is like that). When I used them at work back in the 70’s we often needed to remove the panel to access the various interfaces to our test gear and the nylon knobs were always getting snapped off. Much worse now they’ve got so brittle with age!
 
The KC8-A-0 drawings that are revision J and dated 23 January 1975 show the mounting screws.

Interesting. I don’t have those drawings. Just had another look and my panel doesn’t have any screw-holes or earthing cables as your drawing shows. I wonder if there were 2 varieties or if it was a mod to address the fixing knob breakages - it IS quite heavy just to hang on the nylon supports and as I said earlier they tend to snap off if you are taking it off and on a lot.
 
Now I need to actually fix the front panel. I can set and read the SR, so lots of the logic is working.
If I set the MA to 0000 it actually sets to 1400.
If I set the memory contents to 0000 it sets to 1400.

So it looks like there are some stuck bits in the bus interface logic on the front panel.
I already swapped the Option-I board and the behavior was the same.
Time to look at the 8235 chips.
 
Wiggling the jumper cables between the PCBs and squeezing the PCBs makes the 0000->1400 problem change to a 0000->1000 problem. I disassembled the two PCBs and will check for bad solder joints. I will also put some deoxit on the cables that interconnect the two PCBs.
 
Make sure that the cables between DKC8A and the programmers console are good and fitted properly. Those have fooled me more than once-
.
 
Me too! I’ve broken the latch off one side of one of the sockets on the programmer’s panel and it doesn’t take much of a jiggle to get an odd fault turning up. Pressing the connector firmly is my ‘goto’ any time something funny happens.
 
+1 for that. I had a broken latch on the M8316 card. The connectors wasn’t seated properly and I got very far disassembling the front panel until I recognized this. When seated properly everything worked perfectly.
 
I disassembled the two PDP-8/a front panels so I could swap parts between them to see if I could get one front panel working. Now I can get four different bad behaviors with the mixed parts.

I swapped the M8216 board with three spares. That doesn't make a difference. I will swap the CPU board to see of that makes any difference.

Tomorrow I will borrow a PDP-8/a front panel from a machine in the RICM warehouse to see if that one works. If so, then I can use it to debug the other two.
 
I borrowed another programmer's console and M8316 board from an 8/a that John Wilson donated to the RICM.
This one is the early design that attaches with the plastic balls, but otherwise looks the same as the other two later designs.
Now I can make lots more combinations of boards that do different things.

This one has lots of unusual behaviors, the 0, 1, 2 & 3 buttons work OK, but the 4, 5, 6, & 7 buttons result in a 0.
The LA doesn't work right and also does an LXA at the same time.
Entering 2222 shows 0022 in the display, so maybe there is one or two bad shift registers.
 
Too many changing variables. Each time you throw in a new panel you get a new set of problems so
it does not help you close in on the real problem. I know this is not overly helpful, and I know
you'll figure it out.
 
Too many changing variables. Each time you throw in a new panel you get a new set of problems so
it does not help you close in on the real problem.

I was hoping that one combination would actually work, and then I could use that set as a baseline to debug the other boards.
 
Back
Top