Hello.
I have a partially defective HP 82901 drive.
I don't know this drive by model number, but IIRC it's a daul 5.25"
drive with an HPIB interface. I have an HP drive that fits that
description, unfortunately I can't get to it quickly, so I can't verify
the model number, but I do remember some things about it.
I found the problem on the controller card.
I found a service manual on HpMuseum site, named
"82901-82902_CEServiceHandbook-Partial_17pages_1982.pdf". Following it, I
made all self tests and the problem is on the read circuit. I'm sure about
that because the two Tandon drives work fine with another controller card.
In this case, the service manual suggest to verify read oscillator
frequency. Unfortunately the section VI of the manual describing the
procedure is completely missing from the pdf file.
I would strongly recoemdn that you don't _adjust_ anything at this point.
It's fine to check the freqeucny, but don't tweak anything until you are
sure the fault is just misadjustment.
Anyway, from what I remeber, the controller card is based round a 6802
processor [1]. The actual disk contorller chip, anohter 40 pin device, is
made by Fujitsu, and has the same pinout as the 1793 _apart from being
+5V supply only_.
[1] If you've got chips marked with the HP house codes (1820-xxxx
numbers, maybe soem 1818-xxxx ones), let me know what they are and I can
see if I can find the standard equivalents.
That mioght be enough to get you started. Grab a 1793 data sheet and
check the clocks at the pins of the disk controller. There is probably a
proper procedure for adjusting the free-running requency of the data
separate VCO, but at this point, what you want to do is see (a) is there
a read clock and (b) is it possibly a sane frequency.
I can't remember much abotu the PSU in this unit. IIRC there is a
seaprate PSU PCB (and presumably you can check the outputs of that quite
easily). It's quite possible the VCO circuit needs a -ve supply line.
This might come from the main PSU, but on at least one HP controller, it
was producaed by a charge pump circuit clocked from a divided-down
version of a master clock. That might be worth looking for and checking.
If you're really stuck, I'll excavate my unit, pull the covers, and see
what I can quickly deduce.
-tony