• Please review our updated Terms and Rules here

What did I do to my PDP-8 today.

I have spent some time with my lab pdp-8/a lately.

6X4sJqJl.jpg


By using switching PSUs it is much lighter to carry around in the lab which is nice. But since it is an 8A100 chassis it lacks the E-connector to support hex wide core memories. I did put a 20V PSU and a -5V PSU from the very beginning so it has now become time to create that little board that should be sitting where there are two holes to the right in the chassis.

I spent some time yesterday working on the CAD layout for the board that would hold five ECS connectors. I also included the four bank select lines to allow for 128 k memory.
 
Yeahh the weight of that DEC stuff... You'd better collect laptops :D I always use a table or chair with stuff stacked on it to get the right height. You don't want to get such a drive into the cabinet by yourself. You can't see the slides while holding a drive or pdp.

When I got my RK05 I thought that it would be a nice place on top of the PDP8/E. And this was the easiest way to put it in the cabinet. But when I got it all in the cables of the PC04 were too short to reach the controller again. And a while later got an eye on another RK05. That one is still waiting for me to pick up. I thought that it would be nice to have them on top of each other. That meant that I had two reasons to swap the PDP8/E and the RK05... :sleepy:

1.jpg 2.jpg 3.jpg 4.jpg

But the future RK05 needs all the rack space. So the power controller and 230-115V transformer had to be moved too. So these are placed in the top of the cabinet.

5.jpg 6.jpg

Now I have the space for the RK05 and below that one goes the RX02 floppy drive... While placing that drive I found out that I had the wrong cabinet slides. But there is a set waiting for me :)

7.jpg

If you like to read the RK05 restoration then read this topic
And one of my RX02 drives is here

Regards, Roland
 
Roland, I had similar problems with device locations for my 8/e. I have a DECtape rack at the top, then the TU56, the RX02, PDP-8/e, and last is the RK05. The power controller interferes with the RK05 so I will need to move it above the 8/e. I think that I have enough room for a second RK05 at the very bottom of the rack. I would like to replace the DECtape rack with a PC04, but I have never found an M840 PC8E at a price that I can afford.
 
We could make one, and add some other things like serial ports.

I have an M840 which Warren had collected along with a reader/punch. The reader punch is not in the best condition if I remember correctly.

It is not a complicated board, it is just that they seem to be pretty rare making them apparently valuable. My suggestion would be to see if this is one of the boards that Vince has drawn up and if so have a few boards made from his work. And it doesn't look like Vince has done one of these.

Mattis, That is a clean chassis. I do like the replacement power supplies. That must make a huge difference in weight as well as in generated heat. Are you planning on putting 128k in that machine? I think that is the only reason to populate the extra slots. I have one of the 128k boards but it does not work. The onboard power supply gets hot and does not generate the necessary voltages. There is probably a bad memory chip dragging it down and that could be a devil to find. I can always put an external power supply on it and force it to power up and see what gets hot.
 
Are you planning on putting 128k in that machine? I think that is the only reason to populate the extra slots. I have one of the 128k boards but it does not work. The onboard power supply gets hot and does not generate the necessary voltages. There is probably a bad memory chip dragging it down and that could be a devil to find. I can always put an external power supply on it and force it to power up and see what gets hot.

Making a modern 128k memory board based on Vincent's current 32k design would be pretty easy. Finding the KT8-a memory management board for 128k will be challenging.
 
My suggestion would be to see if this is one of the boards that Vince has drawn up and if so have a few boards made from his work. And it doesn't look like Vince has done one of these.

I have taken a hint and created an Eagle CAD project for the M840 :). I scanned my M840K (took me a while to figure out how to get my scanner powered up!) and loaded the photos into source control.

Who knows when I'll find the time to draw the thing, though.

Vince
 
Making a modern 128k memory board based on Vincent's current 32k design would be pretty easy. Finding the KT8-a memory management board for 128k will be challenging.

I also thought a 128K board would be cool, and actually did a bunch of work to try and clone the KT8A. I ran into problems as there is quite a bit of logic that is actually hidden inside programmable parts. (I have no idea if I've reconstructed/second-guessed the contents of those parts correctly.)

Vince
 
Mattis, That is a clean chassis. I do like the replacement power supplies. That must make a huge difference in weight as well as in generated heat. Are you planning on putting 128k in that machine? I think that is the only reason to populate the extra slots. I have one of the 128k boards but it does not work. The onboard power supply gets hot and does not generate the necessary voltages. There is probably a bad memory chip dragging it down and that could be a devil to find. I can always put an external power supply on it and force it to power up and see what gets hot.

I want to have a small light omnibus machine which can be used when fixing broken boards on the lab bench. Then this chassi is ideal. Horisontal mounting of boards is nice when taking measurements.

The 8/a HEX memory boards are using -5V and +20V which are fed in on the E-slot. That is my primary reason for adding it. But why not make it identical to a real E-slot by adding the bank select signals and that strange jumper. So 128 k is not what I am aiming at.

I want to be able to work with those core more in this lab machine.
 
Hi Vince,

I have taken a hint and created an Eagle CAD project for the M840

Then it might be a good idea to take the same approach which I did with several cloned controllers.
Use 'modern' replacements for several chips. I've checked the M840 schematic... A few suggestions:

8271 = 74179
8251 > 7442
8881 > 7438
380 > 74HCT02
384 > 74HCT32
NTE9601 > 74122

For the 314 I used different approaches like a 74HCT05 (connect the open collector or open drain outputs
together and pull them up) or a 74HCT32 in combination with a 7427. (Check the RX8E clone that I made)
Very often there are some not used ports which can be very useful to solve a problem like the nasty 314.

Good for you that the 8235 is not on the board. That is the hardest chip I had to replace on the clones.
For several boards I used different solutions, just depending on other free ports. But fun to do!
And I think it is very nice that there are more clone controllers made. This makes it affordable and
easy to expand your PDP8.

But I'm just still wondering what the prices are for these original M840 boards... The standard boards
on Epay are sometimes sold for ridiculous prices. But controllers are just not offered very often...

Besides that, the most boards are offered in the US. So shipping and customs make them
extra expensive for me to buy and get them to Europe... So I'm very happy with my TTL clones :sigh:

Regards, Roland
 
Last edited:
Then it might be a good idea to take the same approach which I did with several cloned controllers. Use 'modern' replacements for several chips. I've checked the M840 schematic... A few suggestions:

Cool suggestions, thanks! I'll start by drawing the DEC board, then think about tweaking it for more modern times.

Speaking of which, it appears that C65 (39uF, installed horizontally up top) on the DEC boards is installed backwards! The marking ("+") matches the way it is installed, but the "+" end is at GND, and the "-" end is at +3V. (Maybe that's why the pull-up for the 3V supply is only 47 ohms, which is much lower than on other boards?)

Vince
 
Speaking of which, it appears that C65 (39uF, installed horizontally up top) on the DEC boards is installed backwards! The marking ("+") matches the way it is installed, but the "+" end is at GND, and the "-" end is at +3V. (Maybe that's why the pull-up for the 3V supply is only 47 ohms, which is much lower than on other boards?)

Hi Vince,

I checked both of my M840 boards and you are right. That the capacitor is placed in the opposite
way as the PCB marking shows. But the + marking is placed wrong...

Left is - and right is + So the capacitors are placed in the right way...

DSC05129.JPG

Regards, Roland
 
Last edited:
I checked both of my M840 boards and you are right. That the capacitor is placed in the opposite
way as the PCB marking shows. But the + marking is placed wrong...

Left is - and right is + So the capacitors are placed in the right way...

A look at
http://svn.so-much-stuff.com/svn/trunk/Eagle/projects/DEC/Mxxx/M840/M840Kfront.jpg
will show that on my M840K, the capacitor is installed to match the markings (which as you say are wrong).

I also don't have that capacitor from 3V to the collector of Q3 (installed as rework on yours). What revision is that? (Mine is stamped "X S" in the plastic handles.)

Thanks,

Vince
 
Hi Vince,

It looks like you have that rework capacitor as well but integrated in your PCB already...
That rework cap seems to be connected to the -15V.
compare.jpg

The other capacitor C65 seems to be 90 degrees rotated on your board. Quite confusing I have to say :D

Regards, Roland
 
I located my M840 and the cap appears to be installed + to + as Vince indicates his is. I ohmed the + side of the cap does appear to go to the omnibus ground. Whats the other side of the cap go to? In circuit my Fluke tells me that cap is 95.7 uf which is greater than the 39 uf listed on the cap. Must be paralleled with some other caps. It is a wonder that this cap has not blown up if connected reversed.

Attached are photos of my board. The smoke from the fires in California, Oregon, and Washington has finally reached South Dakota and we don't have bright sunlight out or I would take the photos outside to get better images. What is the deal with the forum only accepting 256k JPEG files?

My board has some high quality sockets on some of the chips in the lower left corner area. These are clearly factory as the sockets are the original wave soldered parts. One of the sockets is empty. What is supposed to go there or is that unpopulated on your boards?
 

Attachments

  • IMG_2165.JPG
    IMG_2165.JPG
    211.8 KB · Views: 1
  • IMG_2166.JPG
    IMG_2166.JPG
    220.5 KB · Views: 1
  • IMG_2167.JPG
    IMG_2167.JPG
    144.3 KB · Views: 1
  • IMG_2168.JPG
    IMG_2168.JPG
    141.1 KB · Views: 1
Hi Doug,

I located my M840 and the cap appears to be installed + to + as Vince indicates his is. I ohmed the + side of the cap does appear to go to the omnibus ground. Whats the other side of the cap go to?

Probably to the bus on AB2 = -15V. Then the +cap to GND and the -cap to the -15V is completely normal.
Please check my earlier posting. I think we are a bit confused by board revisions. Vince has the
M840K and I have the M840F. That C65 has been moved over the different board revisions...

The smoke from the fires in California, Oregon, and Washington has finally reached South Dakota and we don't have bright sunlight out or I would take the photos outside to get better images.

Must be horrible! Any protections available to not breath in that smoke?

What is the deal with the forum only accepting 256k JPEG files?

My board has some high quality sockets on some of the chips in the lower left corner area. These are clearly factory as the sockets are the original wave soldered parts. One of the sockets is empty. What is supposed to go there or is that unpopulated on your boards?


My board has some high quality sockets on some of the chips in the lower left corner area. These are clearly factory as the sockets are the original wave soldered parts.

Are you sure? These look like they have been soldered by hand to me. But hey we have 256K pictures so maybe I'm wrong ;)

One of the sockets is empty. What is supposed to go there or is that unpopulated on your boards?

On my board is this a DEC97401 (selected 7401) but on Vince his board this is the 8881. They are pin compatible but the 8881 has
a much higher current capability.

Regards, Roland
 
My missing chip is E33 (DEC 8881) which is the Omnibus buffer for the upper 4 bits of the reader port. Have to find a 7438 to replace it with. I don't know how close a 7438 is to the 8881 but it should probably be close enough for one chassis setups.

The cap appears to be C65 and is the bulk filter cap for a 2.5 to 3 volt supply they made on the board. Regulation is 4 diodes in series and so the drop is between 0.6 and 0.75 volts each. I guess 2.5 volts is the reason the cap didn't explode.
 
I located my M840 and the cap appears to be installed + to + as Vince indicates his is. I ohmed the + side of the cap does appear to go to the omnibus ground. Whats the other side of the cap go to? In circuit my Fluke tells me that cap is 95.7 uf which is greater than the 39 uf listed on the cap. Must be paralleled with some other caps. It is a wonder that this cap has not blown up if connected reversed.

Your board looks a lot like mine, except for the sockets and the missing 8881. I see the blue wire near E29; do you also have the cut near BV2?

Attached are photos of my board. The smoke from the fires in California, Oregon, and Washington has finally reached South Dakota and we don't have bright sunlight out or I would take the photos outside to get better images.

It is *still* murky here in Oregon. The smoke seems to be trapped low to the ground by an inversion layer. Even more restrictive than the Covid stuff, as no-one wants to go out in it. (Trash pick-up was missed yesterday.) Was able to at least get to the grocery the other day, so at least that's open.

Vince
 
Probably to the bus on AB2 = -15V. Then the +cap to GND and the -cap to the -15V is completely normal.
Please check my earlier posting. I think we are a bit confused by board revisions. Vince has the
M840K and I have the M840F. That C65 has been moved over the different board revisions...

I have the M840K revision also. And that cap is C65 on the diagram which goes across the 4 diodes in series. It is definitely installed backwards.

Must be horrible! Any protections available to not breath in that smoke?

It is not what I would call horrible. What we have outside right now is more like a really clear day in Los Angeles in the 1970's. I can see the sun now.
You can't even smell any smoke (yet). I was in Oregon for part of the fires and that is pretty horrible. Has it cleared up any Vince?

Are you sure? These look like they have been soldered by hand to me. But hey we have 256K pictures so maybe I'm wrong ;)

Grrr, Grumble. You are not wrong. Looked at it without my readers before I posted. Under the magnifier I can see that they were replaced. It is a pretty good job. They cleaned up all the flux and didn't overflow the solder.

On my board is this a DEC97401 (selected 7401) but on Vince his board this is the 8881. They are pin compatible but the 8881 has a much higher current capability.

Looks like the 7438 can sink 48 ma. That is quite a lot actually. How much could the 8881 do? I guess if you needed more drive you could sub a 74S38 since it can drive 60 ma. It doesn't look like the higher speed should be an issue on the application on this board.
 
Back
Top