21 Feb
2009
21 Feb
'09
4:26 p.m.
I have just collected a MicroVAX II which has been in storage and has not
been powered on for 20 years. This is my first machine of such an age,
unfortunately I am not particularly knowledgeable at the electronics level
(I studied circuits academically 25+ years ago and can solder a bit, but
that is as far as it goes). I know I will need to treat it carefully in
order to get it working again. I plan to open it up and make sure I clear
out any debris etc, but beyond that I need advice from those with the
experience and knowledge that I lack on how to go about powering it up
carefully.
IMHO the most important thing to do after checking for obvious problems
-- loose parts, evidence of overheating/burning, etc, is to test the
power supply (PSU) on a dummy load. A defective PSU could wipe out every
chip in the machine!.
Since you're happy to dismantle the machine anyway, this should be quite
easy. With all the PCBs out of the backplane (make notes/diagrams as to
what goes where), and with the power cables unplugged from the drives,
connect a 6V car bulb between the +5V line and the 0V line (you can find
these by tracing back from the known Q-bus pinouts if you can't find them
any other way) and power up. Mearue the 5V line with a voltmeter, and if
possible the other power lines (if you can find them). Only when you know
they're correct should you plug the logic boards in.
Oh, I would recomend opening up the PSU (mains disconnected, of course),
and looking for signs of trouble in there. And check all the fans are
rotatig when you turn it all on
-tony
21 Feb
21 Feb
5:04 p.m.
cctalk-bounces at classiccmp.org wrote:
IMHO the most important thing to do after checking
for
obvious problems
-- loose parts, evidence of overheating/burning, etc, is to test the
power supply (PSU) on a dummy load. A defective PSU could
wipe out every
chip in the machine!.
In addition, pay particular attention to the PSU wiring harness.
The early rev of harness on (iirc) the BA23 had a habit of
catching light. I think it was slightly underrated for the
maximum power load and so would deteriorate over time. It
would eventually reach the point where one of the N conductors
would fail, leaving N-1 carrying way too much load and smoke
ensued.
Field service were (again, iirc) supposed to swap these out if
they came across them. But if your machine has truly not been used
for twenty years, it ended up in storage relatively eraly in its
life and so may not have had a chance for some FS TLC.
A little bit of googling should find you the relevant details
(or you could wait for the next post which will no doubt be
from someone with much better memory than me :-)).
The only other thing I would add to Tony's post is that once
you think it isn't going to fry your boards (or burst into flames)
then you only need the CPU and one memory board in. Then you hook
up a terminal to the console (9600-8-N), set the switch properly
and check that you get some kind of life on the console.
At that point you can start to add the remaining boards (checking
that Qbus grant continuity is OK as you go) and build your system
back up again.
There are plenty of docs you can track down on Manx
(http:://vt100.net/manx).
You can google for the various Micronotes too, some of them are relevant
to the MicroVAX, others to the Qbus, and all of them are interesting
background reading.
Antonio
arcarlini at iee.org
8:41 p.m.
At 05:04 PM 2/21/2009, arcarlini at iee.org wrote:
cctalk-bounces at classiccmp.org wrote:
If the cables between your power supply and the backplane are ribbon
cables with IDC connectors, they can overheat (often you'll find them
showing discoloration.) If they're separate wires with white connector
blocks, you have the updated harnesses.
I haven't seen this problem with BA23 boxes, but have seen several
BA123 boxes with signs of overheating.
-Rick
IMHO the most important thing to do after
checking for
obvious problems
-- loose parts, evidence of overheating/burning, etc, is to test the
power supply (PSU) on a dummy load. A defective PSU could
wipe out every
chip in the machine!.
In addition, pay particular attention to the PSU wiring harness.
The early rev of harness on (iirc) the BA23 had a habit of
catching light. I think it was slightly underrated for the
maximum power load and so would deteriorate over time. It
would eventually reach the point where one of the N conductors
would fail, leaving N-1 carrying way too much load and smoke
ensued.
Field service were (again, iirc) supposed to swap these out if
they came across them. But if your machine has truly not been used
for twenty years, it ended up in storage relatively eraly in its
life and so may not have had a chance for some FS TLC.
A little bit of googling should find you the relevant details
(or you could wait for the next post which will no doubt be
from someone with much better memory than me :-)). 
10:37 p.m.
On Sat, Feb 21, 2009 at 7:41 PM, Rick Murphy <rick at rickmurphy.net> wrote:
If the cables between your power supply and the
backplane are ribbon cables
with IDC connectors, they can overheat (often you'll find them showing
discoloration.) If they're separate wires with white connector blocks, you
have the updated harnesses.
Confirmed.
I haven't seen this problem with BA23 boxes, but
have seen several BA123
boxes with signs of overheating.
We had a MicroVAX I that was FS-converted to a MicroVAX II ($20K,
IIRC, for boards, disk and license upgrades). When we finally traced
the acrid smell in the testing area, we found that the middle 25% of
the contacts on the IDC cable were charred, and, I think, the
connector housing itself broke apart when the cable was removed.
It may have helped that we had a totally-loaded BA23 since we needed
9MB of memory, serial, tape, *plus* room for our own 3rd party board.
There wasn't an empty slot in the machine.
The good news is that I still have that same machine and it still
works just fine 20+ years later, with the replacement DC power
harness.
-ethan
5606
days inactive
5607
days old
3 comments
4 participants
participants (4)
-
arcarlini＠iee.org -
ard＠p850ug1.demon.co.uk -
ethan.dicks＠gmail.com -
rick＠rickmurphy.net