Subject: VAXBI trouble -- I just killed a bunch of boards -- haunted VAX in my basement
To: None <port-vax@netbsd.org, cctalk@classiccmp.org>
From: Gunther Schadow <gunther@aurora.regenstrief.org>
List: port-vax
Date: 05/27/2002 22:43:41
Hi, my pulse is still at 100.

Today I brought my second VAX 6000-400 into the basement. That thing
is haunted! Bringing into the basement entails screwing everything
off and then back on. Everything went so beautifully. Having it done
the second time I noticed that the cables almost find their way
by themselves, just listen to them and they show you how they want
to be plugged in :-).

Anyway, it didn't work very well. The XMI side alone could be made
to work fine, but the VAXBI side doesn't want to work. I checked
the wiring once, twice, there is really not much to it, just one
big plug that goes into the power regulator and the two HUGE contacts
for the actual power. Nothing difficult.

I think I burned a couple of XBIA and XBIB boards today. Nothing
went off in smoke or flames but one spare after the other would
stop working right. I thought I'd do a little board-testing in my
working VAX, but even there the things behaved strangely, even to
the extent where I thought that may be one XBIA board destroyed
the attached XBIB board and vica versa.

Although I have accumulated a load of XBIA/XBIB boards my reserve
went preciously low. I shot with almost all I had and there were
quite a few casualties in the battle. But all to no avail.

I'm lucky my working machine is back working again. But I should
find a more efficient and less dangerous way to debug that other
machine. There is just too much combinatorics involved in the
trial and error method. There are

1 XMI bus
2 positions for XBIA
2 VAXBI busses, that are interconnected

everything seems dependent on everything else here. How can this be
decoupled so that I can test one thing at a time? The next I'll try
is actually replace the VAXBI backplane cage. But what would DEC
field service have done? Is there a technician's handbook that
would give some ideas?

thanks,
-Gunther


-- 
Gunther Schadow, M.D., Ph.D.                    gschadow@regenstrief.org
Medical Information Scientist      Regenstrief Institute for Health Care
Adjunct Assistant Professor        Indiana University School of Medicine
tel:1(317)630-7960                         http://aurora.regenstrief.org