• Please review our updated Terms and Rules here

Need a bit of help resurrecting my PDP-11/73

GreyHairedDECfan

Experienced Member
Joined
Dec 5, 2021
Messages
127
Location
Northern Maryland
This is a machine that I set up years ago at a customer site and acquired after it was replaced with something newer. Last time I had it running was about 15 years ago, it's been in storage since.

I took it out and set it up in one of my spare rooms. Carefully vacuumed everything out and cleaned the case with cleaning wipes.

Plugged it in and about an hour later smelled a little burnt odor. Note that this was without powering the machine up, just plugging in the power cord. But hoping that could just be an errant spider. No evidence of animals or chewed wiring at all inside, that I can see.

I hooked it up to a laptop running Putty and can communicate with it just fine, but (uh oh) it reports that DU0: is not ready and won't boot. I did pull the back panel and reseated the boards and ribbon cables. Did not attempt to pull the drive (RD54 I think, but could be wrong) to reseat that end of the cable.

Where should I go from here? Appreciate any help. Some pics attached. BTW does the neon lamp in the power switch indicate something (the button cracked unfortunately. It lights up solid orange at power-up, but then changes to a flicker and eventually nothing).

I've been wanting to get this baby up and running again for a while, and was pleasantly surprised to come across this forum!
 

Attachments

  • PXL_20211205_213339391.jpg
    PXL_20211205_213339391.jpg
    105.8 KB · Views: 15
  • PXL_20211205_213320604.jpg
    PXL_20211205_213320604.jpg
    123 KB · Views: 17
  • PXL_20211205_212815895.NIGHT.jpg
    PXL_20211205_212815895.NIGHT.jpg
    82.6 KB · Views: 15
Welcome...

The power switch and inbuilt neon lamp just need replacing at some point in the future - that’s all.

The indicators on the front look OK.

Can I suggest performing a Map at the boot-up prompt to see what I/O cards it thinks it has?

Can I also suggest putting your location in your profile? You never know, you may have someone knowledgeable close by you able to help?

That would be my “starter for ten”.

Dave
 
Possible "sticktion" on the RD54. It may not be the best idea but a sharp tap with your hand on the RD54 may release the heads if they are stuck to the platter.
 
So some more testing today. There is a M7546 TK50 tape controller, and a M7555 RQDX3 hard drive controller installed. I pulled both and reseated them in the backplane. I also took both drives out and reseated the power and data cables. When the unit is on, the tape drive power button lights and the drive gear spins. I don't see any LEDs lighting on the hard drive, but I can hear it spin up (or at least it sounds as if it spins up, starting when it tries to auto boot (see attached terminal log).

I did notice that with the back cover off (laying down), there are no LEDs lighting on either the TK50 or RQDX3 drive controllers. Are there supposed to be? The TK50 controller has 2 red LEDs within clear view, but neither of them lights up. Can't see any lights on the hard drive controller.

If I move either of the controllers to another backplane slot, will that work (if the current slot is bad somehow)? I seem to recall something about needing a grant card or something like that to pass data flow across unused slots, but I could have that all wrong - been quite a while
 

Attachments

  • putty.txt
    6 KB · Views: 2
Is the "46" indicated on the 7-segment display there permanently - or was it just there when you took the photograph of the boot-up process?

Dave
 
Is the "46" indicated on the 7-segment display there permanently - or was it just there when you took the photograph of the boot-up process?

Dave

It's sitting on '4' after it attempts to boot. Does that correlate with the "Message 4 Entering dialog mode" that shows up in the log after it fails to boot?

Also after pulling the hard drive and tape drive cards and reseating everything again, now during boot MU0 shows up as a non-existent drive (whereas before it was "Not ready"). Do I simply have a dirty connection between the card and the backplane?
 
Yes, if the 7-segment LED display ends up as a '4' this is indicating that the boot ended up in 'dialogue mode'.

It is possible that you have a bad contact - of course.

Are you able to perform a dialogue Map command and see if the controllers exist?

If they do, the next step may be to load the XXDP diagnostics and run them on the CPU, memory and the controllers. One option is serially from a TU58 emulator...

Dave
 
Here's the map results:

Commands are Help, Boot, List, Setup, Map and Test.
Type a command then press the RETURN key: M


15.206 MHz

Memory Map
Starting Ending Size in CSR CSR Bus
Address address K Bytes address type type

00000000 - 17757776 4088 17772100 Parity Qbus

Press the RETURN key when ready to continue

I/O page Map
Starting Ending
Address address

17760020 - 17760036
17760140 - 17760146
17765000 - 17765776 CPU ROM or EEPROM
17772100 Memory CSR
17772150 - 17772152
17772200 - 17772276 Supervisor I and D PDR/PAR's
17772300 - 17772376 Kernel I and D PDR/PAR's
17772516 MMR3
17773000 - 17773776 CPU ROM
17774500 - 17774502
17777520 - 17777524 BCSR, PCR, BCR/BDR
17777546 Clock CSR
17777560 - 17777566 Console SLU
17777572 - 17777576 MMR0,1,2
17777600 - 17777676 User I and D PDR/PAR's
17777744 - 17777752 MSER, CCR, MREG, Hit/Miss
17777766 CPU Error

Press the RETURN key when ready to continue

I/O page Map
Starting Ending
Address address

17777772 PIRQ
17777776 PSW


Pretty sure I have a TK50 tape with XXDP on it, but of course if the tape drive won't work that isn't going to help.
 
I was able to remove and reseat the TK50 controller (yet again), and now I have a red LED on the TK50 controller. Also the Map command shows the tape drive as Not ready, now, instead of Non-existent. So some progress.
 
Got a picture of the cards inside? It's weird that it didn't see the MS: controller at first, that's odd and I wonder if there is a break in the interrupt chain before the RQDX3. Also are you sure it ran before it was parked, the RD54 requires different jumper settings on the RQDX3. Also make sure the drive is plugged into the Drive o sockets on the BA23 frontplane.

Also possible the drive is just sad. Can you pull it out and power it up with an external power supply?

The big problem with TK50's and TK70's is that the tape capstains have bearings that dry out over the decades. Removing the capstain (count the number of turns on the bolts, then put them back with the same # of turns) and putting a drop or two of watch oil with an oiler will restore the bearings. Without it the capstain drags on the tape and it faults unhappily. With and they run great.

(Also grab a TK70 controller from Ebay if cheap, it's a lot faster than the TK50 controller even with a TK50 drive)
 
If the RIFA capacitors have gone - they generally go one way or another. By this I mean that the machine doesn’t work at all (no power) or the machine works but the RIFA capacitor doesn’t do its job because it has died. Some people just remove and jumper out the RIFA capacitor in this case...

I didn’t spot the message about the memory test being “stepped”. I would enter dialogue mode and re-enable all of the POST checks. Someone has removed it to cause the boot to work quicker. At this point in time, you want all of the checks enabled.

That’s a good call about the bus continuity (especially with the MS controller at 17772522 being missing). If it was in the bootlist I would expect it to exist. Is it physically there or not? If it was, has the tape drive and/or controller been removed at some point? The next step is to document all of the cards within the QBUS chassis and identify where they are in the bus slots. If there is a gap somewhere, that may affect the operation.

Incidentally, I have plenty of TK70 controllers here to give away. In fact, boxes of QBUS cards free to a good home. You will just need to pay the shipping charges. I will be doing a new inventory shortly of what is left. I know for a fact I have half a box of DLV11 serial cards!

Dave
 
Here is how the cards are installed in the backplane. This is exactly how it was since the late 1980's when I set this machine up originally.

Slot 0 (far right when looking directly into it from the back opening):
..Full-height M8190 (pretty sure this is the CPU as it has the console cable on it)

Slot 1 (next one to the left):
..Full-height M7551 (think this is the memory card)

Slot 2 Upper:
..Half-height M3106 (DZV11 used for serial communications with control equipment)

Slot 2 Lower:
.. Empty

Slot 3:
.. Full-height M3104 (DHV11 used for terminal communications)

Slot 4 Upper:
..Half-height M7555 RQDX3 controller

Slot 4 Lower:
.. Half-height M7546 TK50 controller

Slots 5, 6, 7:
.. Empty

The machine last ran (fine) back in about 2005 or so.

It's pretty dusty inside on some of the boards and air passages. I'm wondering if I should pull the internal case out of the outer case/stand and open up the metal internal case, pull everything and clean it well. Or am I likely to make things worse (envisioning dried-out plastics breaking, static electricity damage, etc.).
 
The slot allocation looks OK. So the machine never had an MS device actually installed?

Yes, take the cards out and give them all a bit of care and attention. Use sensible antistatic precautions if you don’t have an antistatic mat and strap.

Make sure the cards go back in the same place.

Blow the dust (and dead bugs) off the cards. Do you have any edge connector cleaner?

Off on a business trip now for a few days. Good luck...

Dave
 
That's an RD53. They are notorious for not coming up properly due to stuck heads. Not to the platter, but to the head stop in the drive. I've found that whacking it on the side with a rubber hammer while powering up will sometimes free the head, other times you have to take the top off and move the head assembly.

Note that while the historical wisdom was that taking off the top is instant doom, these drives did have filters in them to filter out any particles anyway so they should probably be ok for the short term. Still, I would back up the data as soon as possible.

Note2: If you get an RD54 you do have to change jumpers on the RQDX3.
 
That's an RD53. They are notorious for not coming up properly due to stuck heads. Not to the platter, but to the head stop in the drive. I've found that whacking it on the side with a rubber hammer while powering up will sometimes free the head, other times you have to take the top off and move the head assembly.

Note that while the historical wisdom was that taking off the top is instant doom, these drives did have filters in them to filter out any particles anyway so they should probably be ok for the short term. Still, I would back up the data as soon as possible.

Note2: If you get an RD54 you do have to change jumpers on the RQDX3.

How hard do you whack them... tap, gentle hit, or really smack it? I can give it a go.

Is the controller able to detect that it's not working and that's why it's reporting Not ready?

Update: I tried the whack, no luck. Took the cover off the drive. I see what you meant by the air filter. Anyhow, I could not get the head arm to move at all, using careful gentle persuasion. Should I have been able to?

I really wish there was a way to get some indication that the RD53 is even getting signals. I reseated all the cables on the front plane, no difference.

I may see if I can boot off of an XXDP tape cartridge.
 
Last edited:
Can't boot off of the MU: drive. When I try, it reports Non-existent drive. Also the TK50 has something stuck, it won't let me eject a tape. Had to remove the drive and flip up the solenoid-driven latch to allow the cartridge to be removed. Fortunately there wasn't tape all over the place.
 
Back
Top