• Please review our updated Terms and Rules here

How do I wire wrap used back planes to build a pdp8 straight 8?

Here are some card slot pics from the MARCH pdp 8. At least you'd know the card order for this machine. It's from the R.E.S.I.S.T.O.R.S club of the 60's. Claud Kagan is in some of the photos, who donated the machine to our club in 2008. These are pics of the day he donated the pdp8 to us. As you can see, the computer artifacts were stored in his barn. You may have heard stories - a few years later the barn burned down taking a Burroughs 205 with it.
http://vintagecomputer.net/tcf_2008

thm_digital_pdp8_sn1158_front_panel-c.JPG
 
billdeg said:
Here are some card slot pics from the MARCH pdp 8. At least you'd know the card order for this machine. It's from the R.E.S.I.S.T.O.R.S club of the 60's. Claud Kagan is in some of the photos, who donated the machine to our club in 2008. These are pics of the day he donated the pdp8 to us. As you can see, the computer artifacts were stored in his barn. You may have heard stories - a few years later the barn burned down taking a Burroughs 205 with it.
http://vintagecomputer.net/tcf_2008

thm_digital_pdp8_sn1158_front_panel-c.JPG
Click to expand...
I know a bit about the R.E.S.I.S.T.O.R.S pdp8 and that it was in a barn for a long time, but I did not know about the Burroughs 205, its sad it went with the barn.
 
Sorry, mistake, 25swg is 0.5mm, the 0.9mm or 21swg was used for the pulse distribution.

Core store, yes, a thread on here had someone in Bulgaria (?) selling tubs of ferrite cores, I bought one tub of each size. The Roadrunner wiring is 0.25mm diameter I think, but enamel insulation rather than plastic as with wire wrap wire. It is small enough for the larger cores but not the smaller. It is more than adequate for signals, both on a board and a backplane. No use for power on a backplane though.

As a matter of interest if a wire wrapped backplane is needed then look out for the old CNC controls from the 70s. At Cincinnati Milacron a typical control would be the A5, A8 or A10. A machine tool dealer might still have one. I was half interested in getting one of their computers used in the Big Blue controls, but might be a job too far.

Bob
 
woodchips said:
Sorry, mistake, 25swg is 0.5mm, the 0.9mm or 21swg was used for the pulse distribution.

Core store, yes, a thread on here had someone in Bulgaria (?) selling tubs of ferrite cores, I bought one tub of each size. The Roadrunner wiring is 0.25mm diameter I think, but enamel insulation rather than plastic as with wire wrap wire. It is small enough for the larger cores but not the smaller. It is more than adequate for signals, both on a board and a backplane. No use for power on a backplane though.

As a matter of interest if a wire wrapped backplane is needed then look out for the old CNC controls from the 70s. At Cincinnati Milacron a typical control would be the A5, A8 or A10. A machine tool dealer might still have one. I was half interested in getting one of their computers used in the Big Blue controls, but might be a job too far.

Bob
Click to expand...
can you give me some more info on the A5, A8, A10? the packplane would only be useful if the connecters can take dec flip chips?
 
This is 30 years ago, but. I remember the PCBs having two edge connector fingers, I think they were 0.156" spacing and about 30 fingers, double sided. The chances of the DEC board exactly fitting are zero, but if you were making new boards then it is a source of cheap edge connectors. Bob
 
woodchips said:
This is 30 years ago, but. I remember the PCBs having two edge connector fingers, I think they were 0.156" spacing and about 30 fingers, double sided. The chances of the DEC board exactly fitting are zero, but if you were making new boards then it is a source of cheap edge connectors. Bob
Click to expand...
I am buying some DEC flip chips, and will make any I cant buy, the ones from the 1960's and still cheaper than making my own up? so I would need connecters that would take dec flip chip's
 
I have the wire wrapping gun! and 4 used pdp12 backplains, that is space for 320 flip chips or 50% of the basic pdp8 build!
the pdp12 was scraped a log time ago, the backplains need unwrapping and the stration the pins carefully.
do I need to clean the pins?
 

Attachments

  • IMG_20130621_195203.jpg
    IMG_20130621_195203.jpg
    91.6 KB · Views: 1
Last edited:
I have found some great info, it looks like the PDP11 and posibably others have manufacturing failures problem that can show up with use and age, as oxidized or contaminated pins where used on the backplane, so the solder did not stick to all of them, so I will inspect and test my backplane parts when I get the old wire wrap off.
It would be much better to find the problem now before I put the new wire wrap on!

Trying to fix a DEC PDP-11/34 backplane - diagnosis
http://retrocmp.com/articles/trying...ing-to-fix-a-dec-pdp-1134-backplane-diagnosis

pdp1134_good_and_bad_pins.jpg
>The pin on the lower right and the one in the middle of the photo are probably not coated at all (anymore).
>On the upper right of the photo you can see a pin which is only coated on one side.
>Other soldering joints on the photo are looking good.
 
While contamination is never a good thing - Wire Wrap gets the integrity of it's connection by the sharp corners of the posts being tightly wrapped by the wire. The photos show cold solder joints, which would concern me more. Perhaps you could do some wraps to un-cleaned pins and test the result before deciding how aggressive a cleaning you might require?

When you do clean - Take particular care that the sharpness of the corners is not effected by your work.

Perhaps others will have different opinions and experiences. I've seen a lot of horrible looking backplanes who's only real problems were mechanical or electrical damage, corrosion of the connector contacts themselves, and acid bearing bio-fluids in the connector troughs and on the PCB to which they are attached.
 
I was thinking of using q-tips/cotton wool buds and isopropyl alcohol, and I am also a bit worried about the pins being not as sharp as they have been used before, but I will inspect for cold/dry solder joints and will make some test wire wrap connections, I have saved the old wire and some is quite long.
 
Hi All;
Rory, Just a tip, since I save wire that I have unwrapped.. Cut the ends off of the wire before You use it again.. The old wire that has been wrapped is weakened and so it tends to break, also, when storing in a box it won't tangle as much as the ends tend to catch each other when left as partially wrapped ends..
THANK YOU Marty
 
yes I was going to cut the ends off before reusing the wire, and the old silver plating is black with age! and I am not shore how much I can use in the pdp8 build as the longest bits are about 12 inches long
 
After stripping all the wires off two Honeywell 200 backplanes my wife and I spent a long time clipping off the used ends, straightening the wires and sorting them into bundles of roughly equal lengths. Where a signal is connected to a whole row of pins on different logic boards by wire-wrapping many short wires have to be used, so we've kept all wires down to three inches long, which is enough to join two adjacent pins. If you have large numbers of wires then sort them by length to start with to get the best use out of them.

I spent some time soldering new power bus wires across one of the H200 backplanes for the five different voltages needed by the core memory unit and I can't say that I saw a clear meniscus form on all of the hundreds of connections that I made even though they are currently viable electrically. However, failure of any connection would only affect the power to one logic board, which would be an easily identifiable fault, and the remedy would be to add a wire-wrapped connection to adjacent power pins as a back-up. So far in my test circuits I've had no problems with any wire-wrapped connections failing, so the suggestion that they overcome surface oxidation at the corners of the pins sounds reasonable. On the H200 backplanes virtually all signal connections are made by wire-wrapping with only the heavier power buses soldered.
 
I dont think there would be a problem using 24-gauge wire on the PDP-11/12 backplane? and it may be a good idea to use the thicker gauge wire 24 vs 30 gauge on the pdp8 to avoid voltage drop problems, but I think doubling up or using thicker gauge wire on the supply and ground would be a good idea. would there be a problem using 30 gauge wire on the pdp8, or reusing some 30 gauge in it?

and I have been looking at rope memory, it looks like it could be useful as a boot ROM, I am still looking into the PDP8 boot process and the needed ROM size, looks like it could be made with 12 or 13 ferrite rings a lot of wire per ROM, it would also need a bit of custom logic, and of course a custom front panel with all the control and test switches and status lights possible! I love those blinking lights!
 
Last edited:
Isn't the straight-8 all core-memory? That makes it one big boot "ROM". Hand-toggling the bootstrap paper tape loader was only necessary if for some reason the core memory became corrupted in that area (e.g., over-run by a wayward -- or possibly intentional -- program). At least that's how I remember it working on the 8/L (many moons ago).

That said, it seems to me that you'd need to have it automatically switch out so that "regular memory' would be available as soon as the paper tape loader load was complete and it started operation.

Were you instead thinking of putting the paper-tape loader itself into your "ROM" (thus placing it in a "main" memory area)?
 
Yes the straight 8 uses core memory (RAM) I am still doing research, but I was thinking of a rope memory ROM system that will copy the boot code to the main core memory, so no address space is used for ROMs.
I think there is a good chance I will over right the boot code when using the PDP8 so a rope memory boot ROM could be useful.
I may only have the paper tape loader bootstrap in ROM? I am still looking to find how many instructions it has in it?
 
Last edited:
Hi All;
Rory, since You are Building Your own.. You can do anything You want, like in the "Hackers" Book, and Midnight Wiring Society, they added an instruction to IF I remember correctly it was a PDP 06.. It doesn't have to be DEC sanctioned..
THANK YOU Marty
 
Marty said:
Hi All;
Rory, since You are Building Your own.. You can do anything You want, like in the "Hackers" Book, and Midnight Wiring Society, they added an instruction to IF I remember correctly it was a PDP 06.. It doesn't have to be DEC sanctioned..
THANK YOU Marty
Click to expand...

yes I read about it, the extra instruction stopped one old DEC tape from running! lol
I am thinking of keeping the base pdp8 close to 1968~ stock speck, but have some bolt on extras, like a rope memory ROM, and maybe the option to overclock as my raw cores could have been made in the 1980's and could probably be run a lot faster then the 1960's cores!
I have a few other extras that would be fun to add!

but I want it to be able to run straight 8 software
 
13 instructions in rope memory ROM sounds easy!

>Leo: It says Gizmo SBC6120.
>Steve: SBC6120. Yup. So that is a working control panel that emulates the PDP-8, just
>like what I had when I was 15, and what I may have...
>Leo: Now, do you flip those switches to program it?
>Steve: Yes. Yes. I mean, for example, when the machine is empty, and you want to read
>in a paper tape, you have to key in the so-called "boot loader." So you flip the switches...
>Leo: How many switch flips is that?
>Steve: Oh, it's not that many instructions. I think it's maybe like 12 or 13 instructions.
>It's a very simple loader that is simply - it simply says read the next chunk of - read the
>next data from the paper tape. Store it here. Read the next one. Store it there.
>Leo: But you're flipping those switches essentially with binary code for the Assembly
>instructions.
page 4
https://www.grc.com/sn/sn-177.pdf
 
You must log in or register to reply here.
Back
Top