• Please review our updated Terms and Rules here

PDP 11/23 reassembly and restoration

ThisPostContainsComputers

Experienced Member
Joined
Oct 31, 2015
Messages
119
Location
Minnesota
Well, today I have finally managed to get my hands on some "larger" PDP-11 hardware. The ad I picked it up from was originally was originally only for the drives, but a system unit, rack, media, and cabling ended up being included to. All in all I ended up driving home with:
  • A 3rd party chassis packed full with an M8186 CPU, assorted DEC cards, and a number of 3rd party cards.
  • 1 RX02
  • 1 RL01
  • 1 RL02
  • Assorted 8" floppy disks, 2 RL02 disks, 1 RL01 disk
  • DEC branded rack
  • Most (all?) of the needed cabling
Everything looks to be in decent condition apart from being a tad bit dirty. I have yet to actually unload any of this (Besides the CPU) out of the car.

wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==

wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==

wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==


Anyway, moving on to the main chassis. On the front panel it states that it is a "ADAC System 1000", with a number of ADAC branded cards inside. On the topic of cards, this machine is packed full of them. Most of the cards have been loosely thrown into the backplane, and are not actually plugged in. Most of the cards are fairly straight forward, thought I think some have them may not be front the original set up of this machine, and instead just thrown in there when the machine was sold as surplus. There are multiple M8047 cards, plus multiple RAM cards (3rd party 128K, and a M8044). I am assuming the ADAC cards are just Digital<->Analog cards. Also included is a MLSI-512 graphics card.

wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==

wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==

wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==



The guy I got it from says he bought it as surplus from the University of Minnesota. Pretty sure it was for data collection and logging, like many systems of this configuration are. I will probably have more questions once I start unloading the rest of the hardware tomorrow, but for now I do have a couple. First off, how do you actually open the RL0X drives. I was going to open them up in order to lock the heads before moving them, but wasn't quite able to figure it out on site. There appears to be what I think is a latch on the top, but I am assuming they are electromechanically locked. I am also curious on to what the order of operation for getting these RL0X drives fully checked out before trying to use them is? Wouldn't want to wreak my limited supply of disks now would I?

Thank you, Gavin
 
The latch on the top of RL drives are indeed how you open them. But there is an electrical interlock on them. If you don't have power on them, newer revisions have a small service lid on the right side, which you can open with a screwdriver. Inside you can then access the interlock pin - pull it down and the latch on the top can be operated.
Older revisions don't have this, and you will instead be required to remove the rear part of the top cover (four screws) so you can get it all opened up.
Once this far, the transport safety for the heads should be easy to do. You need a smaller straight screwdriver for that part.
 
I have a video on YouTube of my PDP-11/23 with its RL drive, also includes a tape drive but just ignore all that and watch the RL drive part. You have to have the drive powered on and connected to the controller before you can open the drive housing. Would assume that unless the previous owner were morons they would have parked the drive and removed the disk pack at that time. The head has a small enclosure that it lives in when the head is parked, there is also a small metal cover that can be held in place to keep the heads from coming out when no pack is in the drive but have never see anyone use that additional step but the heads tend to stay parked in the drives during normal movement.
About the worst thing you can do to that drive is shut it down without parking the heads first! But beyond that the drive is so primitive and so big its almost indestructible and unlike most drives built after its easily repaired when it has an issue.
The RX drives are also fairly bulletproof. but they do have an annoying habit of the motors running the entire time power is applied.
Link to the video of my 11/23:

https://www.youtube.com/watch?v=oWnjuLS6jxI&list=UUhs8dBmjyQgd2k6mvm3Sizw& index=17
 
The "parking" of the heads always happen. Even if you disconnect power on a running drive. There is enough charge in internal capacitors to retract the heads at all times. This is essential, as the same thing could happen during a normal power outage. Not everyone had/have UPS for these kind of systems...

The small metal plate in front of the heads should always be moved to block the heads when you transport the drives. It's called a transport safety for a reason.
And whenever you move drives, you should *never* have disk packs in there.

If your drive has some issue and the heads are left out at power down, you do risk a head crash, which is not a nice thing. Same thing if transporting without locking the heads down. If the carriage moves out, you then have the heads hitting the other heads. If the heads are damaged, the drive is pretty much a write-off. There are no spare heads available anymore, and haven't been for a long time now.
 
Man, I would not trust that! I always park the heads before powering down the system and wait for the ready light before loading them.
 
Man, I would not trust that! I always park the heads before powering down the system and wait for the ready light before loading them.

Of course you shouldn't "trust" it in the sense that this is how you would normally operate. I am merely pointing out that the devices were designed for this, as involuntary power loss sometimes does happen, and not designing for it would have been an extremely serious issue. Lots of companies depended on these kind of systems, and if a powerloss would have been all it took to bring down the whole company (potentially), it would have been unacceptable, obviously.

Just because safety measures are in place however does not mean that you shouldn't follow normal procedure when operating. Just as you shouldn't stop your car by driving into the one in front, even though you have seat belts and airbags, and possibly even collision detection and automatic breaking. You should still brake in good order.

As for the loading, there is no reason to wait for the ready light before pressing the load button. The load cycle will only happen once the drive is ready. The load button don't affect anything until that point. So by waiting you just add some extra time. Not that it hurts, but it definitely don't add anything either.

But I was pointing out that your implications about the "badness" of having the drive powered off before spun down is largely unfounded. It is a bad practice, but it should never have any bad actual effects.

I would say your habit of not using the transport safety when transporting the drives is a much worse habit. There is no reason not to put on the transport safety before moving the drives, just as there is no reason not to spin down the drives before powering off. But while the engineers had to explicitly design the hardware to handle a power loss before spin down, there was no reason to design a safety backup for the transport safety.

And as I mentioned, you can open the drive without power. Almost all drives have the service hatch on the right side, but some old revisions do not. It's all documented in the manuals...

By the way, the terminology of parking a drive don't really apply to RL drives (or any other drive with removable packs). It's something that came with later enclosed drives that didn't have a proper retraction of the heads. Instead they are designed with a "parking area" for the heads, where they are allowed to land on the disk surface and rest there. That is never the case with the RL drives. The heads are retracted, and mechanically separated from the disk surface. So what you usually said (say) is that the disk is spun down or spun up, and the heads are loaded or retracted (or unloaded).
Hence also the "LOAD" button and ready lamp.
 
Does anybody know where I can find the documentation for the ADAC System 1000? More specifically, the backplane and the chassis power supply. I am unsure if the backplane is Q/Q or Q/CD, and slots 3-11 apparently have something funky going on with +-15V. Having a schematic for the power supply would also be nice when I check out all of the voltages. I would like to have a full grasp on how the system is put together before I risk frying any of the 14 QBUS cards that were haphazardly thrown into the backplane.
 
Small update: After tracing out a decent amount of the pins, I am 90% sure that this is a Q/Q backplane. All of the data lines that I tested in the A/B column were continuous with the ones in the C/D column. Interestingly enough, the -/+15 volt rails that are stated on the case to go from slots 3 to 11 actually go from 5 to 18. On a more frustrating note, my backplane appears to have a short somewhere between +5V and ground. Finding said short has proven to be a "fun" challenge that I have yet to complete...

(Edit: Looks like the continuity "issues" was resulting for a resistor block putting 20ohm between GND and +5, causing my multimeter to read a short. No idea why they have that like they do besides having a dummy load for the power supply)
 
Last edited:
We used to use ADAC I/O on our power station control systems. Unfortunately, we have shut them all down now though!

I still have the software (and manuals for the software) though...

I could use my contacts to see if I could locate some hardware manuals for you, but I wouldn’t hold my breath!

Anyhow, the cards we supported were:

1616OII Digital Inputs.

1616HCO Digital Outputs.

1601GPT Watchdog.

If you can post a list of card numbers I’ll see what I can do for you.

Dave
 
(Edit: Looks like the continuity "issues" was resulting for a resistor block putting 20ohm between GND and +5, causing my multimeter to read a short. No idea why they have that like they do besides having a dummy load for the power supply)

20 ohm? That's not far from a short. As far as loads goes, that's a lot of load. 1.25W unless I'm confused. I hope you have some beefier resistor than the standard 1/4W ones...
 
20 ohm? That's not far from a short. As far as loads goes, that's a lot of load. 1.25W unless I'm confused. I hope you have some beefier resistor than the standard 1/4W ones...

PXL-20210517-050214922.jpg

The resistance between GND and +5V is done using 3 of these resistor banks. Each one of them does 72ohm each, and they are in parallel making the entire system be about 24 ohms.

We used to use ADAC I/O on our power station control systems. Unfortunately, we have shut them all down now though!

I still have the software (and manuals for the software) though...

I could use my contacts to see if I could locate some hardware manuals for you, but I wouldn’t hold my breath!

Anyhow, the cards we supported were:

1616OII Digital Inputs.

1616HCO Digital Outputs.

1601GPT Watchdog.

If you can post a list of card numbers I’ll see what I can do for you.

Dave

This system also came with three ADAC cards, they are:

Model 1023AD (Has a ADAM 100 Module on it)
Model 1023EX (A large amount of the sockets are unpopulated, assuming this is normal)
Model 1620 DMA

Am I the only one who does not see the pictures in the opening post? I'm very curious :D

Here are some direct links to all of the images I have posted

https://i.ibb.co/nCtmZsF/PXL-20210513-015950678.jpg
https://i.ibb.co/s3ZM9MH/PXL-20210513-015907032.jpg
https://i.ibb.co/c3h3H64/PXL-20210513-062623222.jpg
https://i.ibb.co/SQvmDcH/PXL-20210513-015005035.jpg
https://i.ibb.co/NtzmWMy/PXL-20210513-014945447.jpg
https://i.ibb.co/W6bR8qT/PXL-20210513-014935816.jpg
https://i.ibb.co/yVm2kHK/PXL-20210517-180913834.jpg
https://i.ibb.co/VCBfkby/PXL-20210517-050214922.jpg
https://i.ibb.co/28TmhdP/PXL-20210513-062800349.jpg
https://i.ibb.co/HFvFcMb/PXL-20210513-062739274.jpg
 
Capture2.png


Good news! The CPU appears to work. After a minor explosion in the power supply and serial configuration issues were cleaned up, the ODT came up without a hassle. After taking inventory of the system, it appears that it comes with the full 256k of RAM (spread between two M8047 cards, a M8044 card, and a third party 128k card).

I am still searching for the schematics for the power supply of my ADAC System 1000 chassis. I would preferably like to recap all of the tantalum caps to avoid any explosions in the future. I did figure out that the backplane is a Q/Q backplane, with the CPU slot being at row 1 column AB. The interrupt priority chain runs in a serpentine fashion down the backplane.
 
Well done!

If you now have ODT running, you can enter a simple terminal test program... I feel a Hello World! program coming on here...

Dave
 
Way more fun, loop that increments by one and displays the results.

1000 005000 CLR R0
1002 12701 MOV #177564 to R1
1004 177564
1006 105711 Loop TSTB R1
1010 100376 BPL Loop
1012 110061 MOVB R0, 2(R1)
1014 2
1016 005200 INC R0
1020 000137 JMP #1006
1022 001006

Over ten years ago I did a video of my 11/23 running this loop, at:

https://www.youtube.com/watch?v=HuX9Jm727Tw

That’s back before I got my RX or RL drives up and running.
 
Way more fun, loop that increments by one and displays the results.

1000 005000 CLR R0
1002 12701 MOV #177564 to R1
1004 177564
1006 105711 Loop TSTB R1
1010 100376 BPL Loop
1012 110061 MOVB R0, 2(R1)
1014 2
1016 005200 INC R0
1020 000137 JMP #1006
1022 001006

Over ten years ago I did a video of my 11/23 running this loop, at:

https://www.youtube.com/watch?v=HuX9Jm727Tw

That’s back before I got my RX or RL drives up and running.

Nice and easy. A couple of comments:
The CLR R0 at the start isn't really needed.
The instruction at 1006 is actually TSTB (R1)
1020 could be a BR LOOP (772) - one word less, and makes it PIC
 
Back
Top