11/45-1, Jay-0

From: Tony Duell <ard_at_p850ug1.demon.co.uk>
Date: Fri Feb 18 17:55:55 2005

>
> Checked out my supplies and powered up. Sometimes the processor comes up
> "pause", sometimes no status light. I checked the light test and many lights
> are burned out. It doesn't seem to retain any values I store in memory, and

First thing. Replace those burnt-out bulbs. You will go insane trying to
sort out an 11/45 without a good panel. Period.

> "addresses error" was a frequent visitor. It's very possible I'm not
> operating the panel correctly, it's a bit different than what I'm used to.
>
> Just out of curiosity, I'm wondering about checking the voltages of the
> regulators at the backplane. I found the charts to decipher the pin
> references, I believe the first pin called out to check for +5 was A02A2. I


There are +5V on pin A2 of most of the connectors (maybe not the memory
clots), ground is pin C2 and T1. -15V on B2. You need both those voltages
for the CPU to run.

Incidentally, the metal chassis is connected to the 0V (ground) rail too.
It's OK to use that for PSU voltage tests, but I wouldn't use it for any
singal tests (too much impedance, possibly).

Can you tell me what boards (M numbers) you have in the CPU section of
the backplane. Do you have the FPU and/or the MMU?

> can find that pin, but where exactly is the ground reference for that? I
> checked the voltages at the regulators instead, but was just curious for
> future reference. All the grounds appear to go to the backplane traces from
> the mate & locks so there's no obvious lead going to a pin on the backplane.
>
> I'm not completely sure where to head next - I feel completely uneducated
> enough to know what to look at, so I'm going to just plow through more
> manuals. My initial suspicion is something with the core setup, I wonder if
> the MF11-L assigns address space based on slot or on jumpers. I tried

Jumpers on one of the boards, I am almost _certain_ of that.

> swapping boards (sorry Tony!) in the core section and the problem didn't
> seem to change at all. It's sucessive presses of DEP supposed to
> automatically increment the address? Not sure it's doing that.

Yes, it should, but only if you don't get a bus error. If you get an
error, the address doesn't change.

Bus errors are normally due to a device/memory not existing at the given
address (I assume you've realised by now that all I/O devices are memory
mapped). Basically, the current bus master (it'll be the CPU for you at
the moment) outputs an address, control lines, and maybe data (if a write
cycle), then assertes MSYN. The selected device asserts SSYN to say it
exists, If it doesn't do so within a certain time, you get a time-out,
and a bus error.

I assume you've set the rotary switches sanely. I think you want the
address one in one of the 'physical' positiions, the data one in the
'data path' position.

OK, what to do next. I asusme you don't have an M792 boot ROM (diode
matrix card). If you do, stick it in (any SPC slot) and try reading it from
the panel. This was a big help for me, being able to relate the light
pattern to the diodes on the card.

Certainly check the PSU voltages. You should have +5V, +15V and -15V on
the CPU backplane. I think your core needs +20V or something strange.

Check the ACLO and DCLO pins on the backplane too (thin yellow and puple
wires on the power harness, I forget where they actually go). If one of
those is low (0V), the CPU will be trying to do a power fail shutdown,
and will not repsond to the panel in a sane way.

Do you have a KM11 maintenance PCB? In fact just what do you have?

-tony
Received on Fri Feb 18 2005 - 17:55:55 GMT