[time-nuts] OT: Help requested with Solartron 7081 DVM
David C. Partridge
david.partridge at dsl.pipex.com
Thu Apr 16 15:12:14 UTC 2009
Hi folks, apologies for yet another OT post.
Is there anyone near (within say 30-40 miles) to Kenilworth, Warwickshire,
England that can assist with resolving a fault on a Solartron 7081.
Bill Ezell (BIG thank you) in the USA has been a providing a massive amount
of off-line assistance by email but we don't seem to be getting to finding
the "smoking gun", so I think we *may* now be at the stage of needing
someone with stocks of RAM that will fit this(*), a ROM reader/burner(**),
and probably a logic analyser.
The obvious fault is that the display and LED annunciators are completely
blank, though sometimes it will flash most/all of the LED annunciators
simultaneously at about a 1 second interval. The display does work as it
lights up if I disable the power down logic (which disables the processor
board). As far as I can tell there is no DMA activity (display and I/O to
floating logic board are driven using DMA). The really annoying thing is
that it fired up ONCE while I was debugging the low +5V supply on the earthy
I/O board but has never spoken to me since then.
Loking at the CPU (68B09) pins, the IRQ line appears to be held low, and
typically it is the GPIB chip (Intel P8291A) asserting the interrupt, the
interrupt isn't getting cleared by the CPU reading the status from the GPIB
chip. Removal of the GPIB chip moves the asserted interrupt to the next
lower priority device which is the RS-232 interface, but otherwise things
stay the same.
Power supplies are good, I've reflowed most pins on the CPU board and
checked connectivity especially where previous re-soldering work had been
done, and the address select lines are being polled by the processor. There
are no obvious shorts between or to ground on the data and address lines.
(*) RAM is 6 off 5516/6117, 6 off 6116, or 2 off 6264 (currently fitted RAM
(**) ROMs are TMS2564 or TMS2764 EPROMS (or equivalents)
The Service Manual suggests this is one of ROM/RAM/CPU/DMA chips or the
display related logic on the I/O board (I've checked most of this and it has
outputs that match the pretty static inputs).
More information about the time-nuts