PDA

View Full Version : First PDP-8! 8/a in a desk



Crawford
November 23rd, 2009, 02:43 AM
Folks,

I just procured my first PDP-8, an 8/a in a desk, aka DECSystem 310 courtesy of Tim / Tradde. It used to run, before a change of memory (core to MOS).

I'll start the process of checking it out soon (this week).

Comments/suggestions as how to start are welcome.

-Crawford

pontus
November 23rd, 2009, 03:13 AM
Hooray! Congratulations.

Is it complete with desk and terminal?

I think the 8/a has a switched power supply and thus have smaller capacitors, but I would start by checking those. And remove the decfoam of course.

Does it have the 8/a console?

http://technology.niagarac.on.ca/people/mcsele/images/PDP8A-Console.jpg

Crawford
November 23rd, 2009, 09:59 AM
Yes, it is built into a desk. It has a programmer's console, but not attached just yet. A VT-50 came with it (wired for current loop), but unknown whether it works.

I think I'll try to get RS-232 out of it, and use Mark Csele's check-out routine from here : http://technology.niagarac.on.ca/people/mcsele/pdp8.htm

Cheers,
Crawford

tradde
November 23rd, 2009, 10:46 AM
The programmer panel was not normally part of this type of 8a. Remember it's in a desk
for use in the office. There was even a back panel to cover anything in the back. Don't
want anyone to see a "computer" in there. :) There are switches on the front to power
on or reset only. Reset would boot from the specified device configured on one of the
boards. Usually a floppy or possibly an RK05. The CMOS RAM came from Philipp in Germany as he wanted/needed core for his systems so I got 32k for 16k of core. I'd think it should
be in working order but never got to trying it. The VT-50 I could not check as I didn't
have a current loop interface on my other machine. Hopefully that works too. The last
time I tried the 8a (when it had core) it seemed to work fine with reading/writing memory
manually and loading the small AC count program. I never really used it much as I happened on an 8e and 8i later which I prefer much more. But it's still a pdp-8 so should
be quite usable. Just think you might be able to play Advent since you will have the
required 32k of memory. I provided a few blank single sided single density floppies also.
I also said I would do what I could to help if he finds an issue with it. I'd like to see it
run again and for him to have some fun with it.

Lou - N2MIY
November 23rd, 2009, 04:20 PM
Crawford,

What TTY interface is in the 8/a? Is it a KL8-JA M8655 or KL8E M8650? If so, I suggest converting the VT50 to use RS232. If it's like a VT52, it should be simple. The pictures below show the RS232 adapter that's plugged into the bottom of one of my VT52s. The board has a 1488 and 1489 on it. The schematic must be in the VT52 printset.

http://www.vintage-computer.com/vcforum/album.php?albumid=51

Lou

tradde
November 23rd, 2009, 04:24 PM
Crawford,

What TTY interface is in the 8/a? Is it a KL8-JA M8655 or KL8E M8650? If so, I suggest converting the VT50 to use RS232. If it's like a VT52, it should be simple. The pictures below show the RS232 adapter that's plugged into the bottom of one of my VT52s. The board has a 1488 and 1489 on it. The schematic must be in the VT52 printset.

http://www.vintage-computer.com/vcforum/album.php?albumid=51

Lou

The serial interface is on one of the 3 8a logic boards. There is no M8655 in it. One
could be added I suppose. I believe the VT-50 that he got is the one that was used
with this system. But I don't remember for sure. I gave him the cable to connect
the VT-50 to the 8a.

Lou - N2MIY
November 23rd, 2009, 04:36 PM
Tim,

Well, it sounds like he has it all then, so no need to convert anything. This will be a beautiful setup when it's all running. I dug out the 76-77 minicomputer handbook to read about the 8/a SLU and on the cover is the color picture of the VT5? sitting on the 8/a desk configuration.

Lou

Crawford
November 25th, 2009, 01:22 PM
Well, checkout of the PDP-8/a is proceeding, with some success.

First, I checked out the VT-50 which seems fine. I get characters on the screen, and I set the baud rate.

Next I took the M8316 board and got the switches set for baud rate. I then got the M8316 installed in the backplane with the
Programmers panel attached.

After a quick check of connections I powered the whole thing up. The floppy disk drives and the computer powered on (with a
whole bunch of cans). When the CPU came on, the disks made a hopeful 'pocka pocka' sound like it was trying to boot.

Then I tried some simple programs via the console. No dice. I seem to be able to change and view registers, but memory deposit and
examine do nothing. 07777 as the address and 7777 as the data seems to be the normal output no matter what I try to poke into memory.
I even tried the key by key input test in the PDP8/a User Manual.

Any Ideas?

- Crawford

tradde
November 25th, 2009, 01:53 PM
Well, checkout of the PDP-8/a is proceeding, with some success.

First, I checked out the VT-50 which seems fine. I get characters on the screen, and I set the baud rate.

Next I took the M8316 board and got the switches set for baud rate. I then got the M8316 installed in the backplane with the
Programmers panel attached.

After a quick check of connections I powered the whole thing up. The floppy disk drives and the computer powered on (with a
whole bunch of cans). When the CPU came on, the disks made a hopeful 'pocka pocka' sound like it was trying to boot.

Then I tried some simple programs via the console. No dice. I seem to be able to change and view registers, but memory deposit and
examine do nothing. 07777 as the address and 7777 as the data seems to be the normal output no matter what I try to poke into memory.
I even tried the key by key input test in the PDP8/a User Manual.

Any Ideas?

- Crawford

Are there any switches on the MOS memory board? I know at some point the 8a could
access more than 32k, up to 128k. Don't know how exactly, but I can guess. Maybe
the memory board is strapped for something other than zero to 32k? My next guess
is there is a problem with the MOS memory board. I know I was able to deposit and
read back data before with the core. I even ran the AC count program. I can't say
I like the programmers panel and it's funky with Deposit and Deposit Next, and Examine
and Examine Next. Plus the numeric pad layout is weird. I hope someone has other
suggestions. There is a 8a core board on eBay right now, or there was. There were
no bids the last time I checked.

Crawford
November 25th, 2009, 03:53 PM
Tim,

Yes, I pulled the documents for the MOS memory board, and the switches seem to be correct for setting it to the first 32K (first 2 switches off and the rest on IIRC).

Dave Gesswein has a couple of 8/a's and is local to me so maybe I can ask his Help.

-Crawford

Crawford
November 28th, 2009, 03:39 PM
Bwaa haa haa (mine is an evil laugh!)

I just wanted to let you know I got my PDP-8/A up! (at least to to partial functionality.)

Dave Gesswein confirmed that a G8018 Power supply (wired for core memory) can indeed power MOS memory (contrary to the DEC manuals).

Anyway, about the PDP-8/a, of course it was something stupid. In an effort to get it running, I completely disassembled my front panel, then swapped parts from another front panel board I had. I still still got inconsistent results. So, I pulled out the M8316 board that the front panel ribbon cables connect to... and they were not fully seated into the board connectors. Reseated the connectors, and the front panel started working! (DOH!)

Here's the rule: 99% of computer problems are loose connections.

So the counting program from the Small Computer Handbook works, and then I tried the character output routine from Mark Csele's site. So, that routine works with the VT-50 (using TTY 20mA current loop). Oddly, the VT-50 had an additional relay wired in to 'chatter' each time a character is received (after about 10 minutes I disconnected that).

During this work I pulled the system out of the desk it was installed in, to bring it in from the garage.


-Crawford

tradde
November 28th, 2009, 05:45 PM
Bwaa haa haa (mine is an evil laugh!)

I just wanted to let you know I got my PDP-8/A up! (at least to to partial functionality.)

Dave Gesswein confirmed that a G8018 Power supply (wired for core memory) can indeed power MOS memory (contrary to the DEC manuals).

Anyway, about the PDP-8/a, of course it was something stupid. In an effort to get it running, I completely disassembled my front panel, then swapped parts from another front panel board I had. I still still got inconsistent results. So, I pulled out the M8316 board that the front panel ribbon cables connect to... and they were not fully seated into the board connectors. Reseated the connectors, and the front panel started working! (DOH!)

Here's the rule: 99% of computer problems are loose connections.

So the counting program from the Small Computer Handbook works, and then I tried the character output routine from Mark Csele's site. So, that routine works with the VT-50 (using TTY 20mA current loop). Oddly, the VT-50 had an additional relay wired in to 'chatter' each time a character is received (after about 10 minutes I disconnected that).

During this work I pulled the system out of the desk it was installed in, to bring it in from the garage.


-Crawford

Congratulations then. This is great news. Loose connections can be a pain as they
are so easy to overlook.