DSD 880 Issues

[gatos2locos]

All,

I just purchased a DSD 880 (fixed drive + 8" floppy drive) together with a DEC PDP 11/23. All very clean and was working before shipping (prev owner used hard drive lots but did not know if floppy drive worked; it booted from fixed drive). Was professionally shipped (by UPS) and seemed to be well-done and undamaged at arrival. However, I am having drive issues. I might be something simple.

Both head and spindle were locked. I unlocked both. At power-on, after fan is on and some time I get an error code "06" which in the manual is written as basically a seek timeout. I used the hyperdiagnostic controls and seem to be able to get floppy drive to operate but not sure if everything works yet.

Some suspicious things:

1. when unit arrived a cable inside, from a board to the front display panel, was unplugged, as if the connector got disconnected during shipping. I reconnected it -- without it, the front panel was simply fully off.
2. after powering on the unit and some seconds past, I think I hear the floppy drive as being tried to be used and then immediately I get the "error code 06" and a LED on indicating problem with the Winchester (fixed drive). Seems a bit odd. Is it that it tries to move floppy disk, then immediately goes to fixed drive, which immediately fails and gives error code 06 (seek time out). I am not sure if I am hearing the drive spin.

I read above about the some dip switches that can alter the boot to use floppy? Could they have gotten moved by cables during shipping? Why does it try to move the floppy? Is this typical? I think I can see the "axis" of the motor that spins the fixed drive and I believe it is not spinning. Can the power to it be unplugged somewhere?

Suggestions?

 

[RSX11M+]

An internally disconnected cable is signs of severe and potentially damaging G forces encountered in shipment. If it was insured, you should contact the shipper.

Meanwhile, I suggest you analyze the internal cable that got unplugged, and look for others that could have been similarly effected as well as tell-tale signs of strain induced damage elsewhere. [bent mounts, PCB cracks, chips out of sockets, etc...]

Stress this severe might move a slide switch, but PCB mounted rockers are usually "G-force neutral" and are less likely to change unless directly impacted. [I.E. - by a loose cable end?]

Is this behavior you're describing with the unit StandAlone, or is it connected to the 11/23 chassis and QBUS controller?

You can download the manuals for that from BitSavers if you haven't already. Troubleshooting sections should be helpful. I'll look am mine here as soon as I finish this post to see if there's anything to add.


Notes about 18 vs 22 bit QBUS controller: [nothing to do with your issues - yet]

• The manual on Bitsavers is vintage 81. I don't think it has any reference to the Q-22 version of the controller, however they were similar enough that the documentation should still work. [my recollection]

• 11/73 users should be aware that the controller may have timeout issues when used with the DUAL wide 11/73. I think this was not the case for the QUAD 11/73, but I'm not certain. [maybe I can find my notes here someplace]

• Since we never used the 22-bit 11/23's with the DSDs, I don't know if there is a similar issue.

• After a conversation with the manufacturer back in the day, I was given the impression that they were not going to revise the board to address the timeout issue, probably because the unit was about to be discontinued.

 

[RSX11M+]

Referring to the manual:

Please perform the "Initial checkout and Acceptance Tests" in section 3.7 steps 1-7 [ those to be performed without being connected to the 11/23 host system and controller ]

That should give us a starting point.

• Afterthought: Please omit tests that require the Winchester to be Write-Enabled until you are able to perform an offline backup of it's contents.

 

[gatos2locos]

Thanks for the suggestions! The DSD unit is self-contained and in theory I can run tests without the main CPU. I am not doing the write-enabled tests. The packaging was styrofoam and bubble wrap quite well done. Some metal tabs are bent (in this disk unit and in the main CPU unit which came in a separate box). The metal is quite thick. I asked the seller if it was bent "he said no" but I am unsure of the certainity of that. For the metal tabs on the disk unit to have bent the fall would have had to be tremendous and some magic would have had to occur for it to be damaged without affecting the styrofoam and the cardboard box. Thus, while a major physical fall is possible, it would seem unlikely.

The cable that was unplugged is odd. I have no explanation, except maybe it was "loose" and it just needed a soft push to come off.

I have not had much time recently to continue debugging. However, I did notice that if I pull the "fuse" for the Winchester drive, symptoms are the same (and fuse measures as just fine). I need to remove more boards to get deeper into the machine -- maybe some other cable slipped off...

??

I have all manuals in actual printed/book form...

Referring to the manual:

Please perform the "Initial checkout and Acceptance Tests" in section 3.7 steps 1-7 [ those to be performed without being connected to the 11/23 host system and controller ]

That should give us a starting point.

• Afterthought: Please omit tests that require the Winchester to be Write-Enabled until you are able to perform an offline backup of it's contents.