[unix-pc.general] 3b1 death

brown@cs.rochester.edu (Chris Brown) (01/23/88)

I lost power at home, and my Global surge protector kept power off to
my 3b1 (2MB, 67M disk).  5 minutes after repowering, machine dies:
no diodes, no cap.lock light, screen frozen, no reset button.  However,
fan, disk still working, screen lit if frozen.  Power down and wait,
power up, a few minutes go by same thing happens.  Power down and wait,
insert diagnostic floppy, powerup and part way thru parity checking,
start getting errors and wham same lockup happens.

Open it up in the shop (not so hard...put it on its back to get started),
crank open the motherboard cover, and it runs for 3 hours no problems
except the cooling-fin gizmo in the power supply is killingly hot and the
68010 chip is possibly a bit overheated.  I had thought hot chip on motherboard,
or dying piece of powersupply (not all of it, of course).

Now I'm stuck.  There were no papers over vents, the machine was on an
uncluttered desk, the fan was making a new speed oscillation effect
(wurra wurra wurra) but the airstream was strong and seemingly cool. Room
temp was 69 or less.  Machine is 5 months old, ran in hotter temps no problem.

I don't see how power loss could hurt. My technician here thinks the cooling
is marginal for the stuff in the cabinet and wonders if another fan outside
to increase the partial vacuum would help suck more air thru...I dunno
about that.

So I guess my question is whether this sounds like a heat problem, have
others beefed up their cooling or felt like they needed to, does it make
sense to drop a new 68010 chip in with no further evidence, should I go
to a heatgun (ugh), or what.  I am working on AT&T service but not making
contact yet, and I don't see what they would do that my local technician
hasnt
done yet.  Any suggestions sincerely appreciated.  And thanks to you with
advice about the fast clock.    

Chris (Life sucks when your 3b1 is sick) Brown

cts@ixhte.ATT.COM (John "C". Sucilla - AT&T Network Software Center) (01/28/88)

In article <6126@sol.ARPA> brown@cs.rochester.edu (Chris Brown) writes:
>I don't see how power loss could hurt. My technician here thinks the cooling
>is marginal for the stuff in the cabinet and wonders if another fan outside
>to increase the partial vacuum would help suck more air thru...I dunno
>about that.

I was getting a lot of grief from my fan too, the thing would seize up
every once in a while when it was running in its slow mode, the machine
would over heat and crash. So I called randy suess over on chinet and
he told me how to bypass the speed control for the fan. I did that and
it helped a little but the machine would still just hang for no apparent
reason. Finally I realized that the fan was probably sucking too much
current out of the 12v section of the power supply (remember, this fan
seizes up at slow speeds) and I replaced it with a 110v version. Its
been up without a single crash for almost a week now. Hope this helps...

mark@cbterra.ATT.COM (Mark Horton) (01/29/88)

I had a similar problem.  I'm not sure this will help, but I'll
relate what happened.

My 3B1 (40MB/1MB) was running happily in my basement when my wife
got a court order to throw me out of the house.  (Long story, suffice
it to say she abused the laws and I'm 98% sure I'll get my kids and
house back in the final showdown.)  She then unplugged the 3B1 without
so much as turning off the power switch.

It sat there for 2 weeks while my lawyer tried to get her to release
the machine.  Finally they reach agreement, and she brings it up into
the garage and lets it sit there for hours.  I pick it up, carry it
to its new home, set it down, and plug it in.  Friends are warning me
about condensation, but it's already coming up.  It boots fine and
goes into its fsck ritual.  I'm still carrying stuff in, and 30 minutes
later I check on it, it's tripped the circuit breaker.  Repeated attempts
to boot it show it comes up fine, but after 4-20 minutes, the breaker
trips suddenly, without warning, and it shuts off.

Next day I open it up to check that the fan cable is still connected,
which it is.  I run it with the cover off, it runs fine.  I put a fan
blowing on it, it's happy for days.  Then I put the cover back on, and
it's still fine.  I never did put the screws back in, but the cover's
on tight, so I doubt that matters.

I do notice the fan is a strange beast that likes to change speeds if
I look cross-eyed at the box.  Moving it especially changes speeds.
I suspect that the combination of moving, temperature change, humidity
change, and phase of the moon were all involved.

	Mark

mikes@ncoast.UUCP (Mike Squires) (01/29/88)

In article <6126@sol.ARPA> brown@cs.rochester.edu (Chris Brown) writes:
>I lost power at home, and my Global surge protector kept power off to
>my 3b1 (2MB, 67M disk).  5 minutes after repowering, machine dies:
>temp was 69 or less.  Machine is 5 months old, ran in hotter temps no problem.
.....stuff deleted......
>I don't see how power loss could hurt. My technician here thinks the cooling

(I worked for a while as an analog (high end audio) technician)
Sounds like the PS emitted a spike when the line voltage went down.  This
would make something go marginal.  First choice for testing would be the PS
(scope 5V line with covers on); check total current with covers off (may be
too high due to bad part); try to cool hot parts with additional heat-
sinking with covers on.

Mike Squires  Allegheny College Meadville, PA 16335 814 724 3360
If it doesn't use output transformers I'm just not interested........