PDP11/20 restoring

[Roland Huisman]

Hi,

I just started with tiny baby steps to restore a PDP11/20. It is the somewhat newer model with the black case. Once meant for parts, now this Dutch idiot is trying to make it complete and working again. Probably a long way to go... :neutral: For what I know now these things are missing.

- top and bottom covers
- lock switch and wiring
- power cable harness
- fan panel and wiring

I hope some people can help me with some pictures from or information about missing parts.

The top and bottom covers are not really a problem to make, but I'm missing details. I would like to have some info about how the end of the metal sheets are made at the front side.


I think the top cover is bended at 90 degrees. Can anyone confirm this? Of have a picture?


And I think the bottom cover is folded back. Can anyone confirm this? Of have a picture?


The power connectors from the cable harness are G772 boards. I only have a G772B, but this board seems to be a bit higher and has a bit different layout.

This is my G772B


G772 board which I found on the web.


Can anyone give a good picture from the front and the back and the size of the G772 board? Then I can make a few replica's.

Another thing which got my attention are these capacitor boards. Does anyone have pictures and/or a model name of them? They don't seem to be in all slots and also not in all machines. To me it seems that these are for the 5V only, because that is the only connected power in the not connected slot.



Thanks in advance!
Roland

 

[1944GPW]

I have a FOX 2 aka rebadged PDP-11/15. The box, power supply and most of the KC11 CPU is similar to the 11/20's KA11.
I can measure up the box lid and draw it up in CAD for you. You are correct the edge is folded 90 degrees. I can remove the bottom panel and take a look at it too.
I think also the power paddle boards are the ones you are after so I can take a few pics.
Here is the layout of mine. I don't think there are any electrolytic caps like in yours but I'll take a look when it's both daylight (AND stopped raining, which might be a while).

 

[Roland Huisman]

I have a FOX 2 aka rebadged PDP-11/15. The box, power supply and most of the KC11 CPU is similar to the 11/20's KA11. I can measure up the box lid and draw it up in CAD for you. You are correct the edge is folded 90 degrees. I can remove the bottom panel and take a look at it too.

That would be great! I've put a PDP8/F cover on top of it and I think it is a good reference for the material thickness and size of the edge around it. But to make a real good replica I would appreciate the details. Especially the part behind the front. I can't find any pictures or details from these covers. CAD would be great, but even just pictures would help a lot.

I think also the power paddle boards are the ones you are after so I can take a few pics.

Thanks, I'm very curious.

Here is the layout of mine. I don't think there are any electrolytic caps like in yours but I'll take a look when it's both daylight (AND stopped raining, which might be a while).

Well that is the thing, I have been looking to many pictures on the web. So these capacitor boards just got my attention. But I don't have them either. It just made me wonder if it was a later modification/add on to reduce noise on the 5V power line...

Regards, Roland

 

[nadaveiron]

I've never had an 11/20, but looking at the 8/f cover it looks like an 8/f cover on an 8/e. Did you try an 8/e cover? I wonder if they're the same.

 

[Roland Huisman]

I've never had an 11/20, but looking at the 8/f cover it looks like an 8/f cover on an 8/e. Did you try an 8/e cover? I wonder if they're the same.

Unfortunately the 8/E cover is very different... The 8/E cover is an L-Shape.

 

[pbarton]

Regarding "Another thing which got my attention are these capacitor boards. Does anyone have pictures and/or a model name of them?"
That is a G8000 +8V Filter Network.
They are used in the DP11-A Synchronous Line Interface option. Which is a four slot system unit.
More information here http://www.bitsavers.org/www.comput... DP11-A Synchronous Line Interface Manual.pdf
The DP11-A Synchronous Line Interface option requires +ve and -ve voltages for the modem EIA interface and the PDP11/20 power supply does not provide sufficient +ve voltage for the EIA drivers.
I believe the G8000 module generates +8V from the line time clock signal!
The DP11 is the only Unibus option that I am aware of, that requires the G8000 module.
There should be one tiny diode on it, as well as the caps.

 

[Roland Huisman]

The DP11 is the only Unibus option that I am aware of, that requires the G8000 module.

Thanks, that is interesting to know! This board seem to be exactly
in the position you mentioned. So probably the G8000


But the other is in the CPU part, and this board seems to be different. There is no board
mentioned at this place in the schematics. Maybe this was an update/modification?


I assumed that these capacitor boards were the same. But the capacitors are also
placed in a different way on the boards. (like horizontal or vertical)
So then I'm still wondering about the capacitor board in the CPU part.

Regards, Roland

(Note that both pictures are not may machines but I just found them online in the past)

 

[pbarton]

The 4 slot backplane with the G8000 card will be a DP11-A backplane. It can’t be used for any other purpose.
Did you remove the G8000 card and see the single small diode?

I don’t know what the small card is, in slot A12 of the KA11 processor backplane.
The orientation of the capacitors appears to be different to the G8000 card.
The KA11 schematics suggest that that slot is not used?
Can you check the wire wrap pins for slot A12 and tell me if there are any wire wraps there?
Any power in slot A12? Black = Ground, Red = +5V, Blue = -15V, Orange = +8V, Green = -20V
If there are no wire wraps then the module has no function. It’s just being parked there?
Can you remove the card and attach a photo?
Is there any identification marks on the small card?

Your slot B12 is empty, the M787 is missing. So your 11/20 is without a line clock.

 

[Roland Huisman]

Your slot B12 is empty, the M787 is missing. So your 11/20 is without a line clock.

Well maybe you missed it in the story, but these pictures above are not from my machine.
Last Friday I got this stack of boards which belong to my machine.



This is how I got the machine... In parts, not complete... And I'm very happy with it!


I really like them this way. But like I said, a long way to go...
In this way you are getting to know the machine very well...

The cpu board set seems to be complete... I also got a DD11 from Ed here on the forum.
That seems to be the right expansion panel for the PDP11/20 powered trough a G772 card.

So I'm looking to all of the pictures online so compare them with what I have
and what I'm missing. And what parts I have to clone. This will become a long story... :D

Regards, Roland

 

[pbarton]

Front and the back of the G772 board photo's attached.

 

[1944GPW]

I measured a G772B in the FOX 2. I tried to take out the one going into the memory but it won't come out without removing the thumbscrews and lifting that module up a bit. It's a bit awkward to do that at the moment but I'll get to it sometime.
The finger width is what I calculated after subtracting two thicknesses of the black insulation tape (0.17mm thick) but it probably won't matter much.


I have the chassis top to measure up next.

 

[Roland Huisman]

Thanks, so it is the G772B version which is used in the 11/20. I've drawn that PCB already but at a later moment I thought the B version was wrong model. I thought the original 11/20 was a G772 and in other/later machines a G772B was used. Good to know, thanks! So I can order these boards anyway :D



Regards, Roland

 

[1944GPW]

Here is the top cover panel for the BA11-C (mine is the S subvariant) for the PDP-11/15 and PDP-11/20. The overall panel, then closeup of the front edge measurements and rear edge measurements.
All folds including the front short edge are 7/64" radius. The panel is 16 Gauge (1.6mm) steel and like all things DEC is quite weighty. I reckon it would probably be ok folded up from aluminium sheet too.

 

[1944GPW]

Added photos of top panel

 

[Roland Huisman]

Great, thanks a lot! I've sent you an email too..

 

[Roland Huisman]

I'm working on the H720F power supply at the moment. I was quite surprised to see two 1 Amp 1N4001 rectifier diodes are fused with a 5 Amp fuse. So the load on the 8V could go up to 5 Amp without tripping the fuse. In the meanwhile the diodes will short circuit and there is no fuse in front of these diodes. This will fry a few PCB tracks.


And again fried connectors...


I've seen this in several dec machines. Pulling the full 5V current on just one tiny contact... And if there is a short circuit, traces will be blown away instead of a proper fuse... Since I have to make a new power harness I will add some wires to bypass the connectors. But I'm surprised to see these kind of problems over and over again...

Guess the +5V in my 8/M for example... :???: Made a bypass on that one too...


Regards, Roland

 

[ftcnet]

.. I've seen this in several dec machines. Pulling the full 5V current on just one tiny contact... And if there is a short circuit, traces will be blown away instead of a proper fuse... Since I have to make a new power harness I will add some wires to bypass the connectors. But I'm surprised to see these kind of problems over and over again...

Us DEC field service techs have seen dumpsters full of burnt DEC switching power supplies. Not just the connectors, but good sized charred black holes where PCB material used to be. ex: H744's had a number of ECO's in ongoing attempts to improve the design and reduce the number of scrapped power supplies.

 

[MattisLind]

Us DEC field service techs have seen dumpsters full of burnt DEC switching power supplies. Not just the connectors, but good sized charred black holes where PCB material used to be. ex: H744's had a number of ECO's in ongoing attempts to improve the design and reduce the number of scrapped power supplies.

I spent some time repairing H744, H7440 and H7441 supplies. The four H744 had their bridge rectifier blown. One of them in very spectacular way in that the metal top with the heatsink was gone (wonder where it ended up?). All the H744 had one burnt off AC inlet connector-termninal. It acted as a fuse... The H7440 and H7441 had switch transistor casualties. In the H7441 even most components in the driver stage and the free-wheel rectifier were fried while the H7440 only had a bad output cap aside from the main switch transistor.

 

[Roland Huisman]

So I'm clearly not the only one with DEC power issues :sigh:

Regarding "Another thing which got my attention are these capacitor boards. Does anyone have pictures and/or a model name of them?"
That is a G8000 +8V Filter Network.

That makes sense! while studying other documentations I found that G8000 more often. It seems that I need
one if I want to use a M7800 serial card on a DD11-A SPC backplane in combination with a 720 power supply.



In the meanwhile I found the little schematic of the G8000 and combined with the little picture
I think it should look a bit like this sketch...



Does anyone have a picture of the G8000 front and back so I can copy the texts as well?
Thanks in advance!

 

[vrs42]

Does anyone have a picture of the G8000 front and back so I can copy the texts as well?
Thanks in advance!

I have photos of the G8000A (lacks the diode, I think) here:
http://svn.so-much-stuff.com/svn/trunk/Eagle/projects/DEC/Gxxx/G8000/

Vince