[comp.sys.sun] Additional Sun upgrade please

fitz@cive.ri.cmu.edu (Kerien Fitzpatrick) (02/09/89)

While the commentary about upgrades is going around - how about
considering this one.

I'm not too sure how many others out there are stuck with 3/75M-8s, but
30% of our clients are 3/75s and they fail (power supply mostly or the
motherboard goes from overheating) a couple of times are year.  This was
compounded by Sun restricting the repair timeframe to only making the 30
day repair available.  It seems that the problem with these units is a
weak (and poorly designed)  power supply and poor ventilation.

It sure would be nice if Sun would be magnanimous and say "hey, we made a
mistake . . . " and give 3/75 owners a good deal on upgrading to a 3/160
housing.  Others at this university have taken the CPUs and memory boards
from 3/75s and put them into housings for 3/160s (the 3/160 boards where
put into cages on mobile robots,etc).  Our 3/75s spend almost 40% of their
year being repaired (unless you want to pay around $1000 for a new supply
or $600-$800 for a quick turn-around repair).

When I queried the people at the (800)USA-4SUN number they told me I
should rid myself of the 3/75s and purchase 3/60s.  There isn't anything
wrong with the CPU, memory board, or monitor - it is just a @%!$@$#$
enclosure.  If you are going to upgrade the donkey 3/50 why not save us
3/75 owners?

Kerien Fitzpatrick
Field Robotics Center
The Robotics Institute
Carnegie Mellon University
Pittsburgh, PA 15213
arpanet: fitz@rodan.frc.ri.cmu.edu
(412)268-6564

paula@june.cs.washington.edu (Paul Allen) (02/23/89)

fitz@cive.ri.cmu.edu (Kerien Fitzpatrick) writes:
>[describes an awful problem with 3/75's losing power supplies and
backplanes, compounded by the lack of a full maintenance contract.]

I don't remember seeing this discussed in Sun-Spots before now, so here's
a description of our experience with the 3/75.

We bought our 3/75's when they were first announced and currently have
about 10 of them.  We've replaced power supplies, and sometimes the whole
enclosure, on all of them.  The problem is the connector between the power
supply and the backplane.  The contacts oxidize over time, resulting in
heat generation and ultimately a burned connector.  The worst one I saw
was so badly 'carbonized' that the connector crumbled when we tried to
unplug it.  In most cases, Sun field service simply replaced the power
supply after carefully cleaning off the posts that the connector fits
over.  We had a couple cases in which things were so far gone that the
engineer judged it necessary to replace the whole base.  I can only think
of one machine that has had this fix applied more than once.  We haven't
had a 75 fail in a long time.

I would think that it should be possible to fix a failed 3/75 by simply
picking up a new connector at your local electronics supply house and
replacing the burned one.  This assumes that the situation hasn't gone so
far that the backplane has suffered damage.  Kerien described having to
wait 30 days for Sun to fix a machine.  The investment of a couple hours
in replacing the connector would seem to be worthwhile!  The critical
point is that the backplane must be undamaged, and the posts that the
connector fits over must be scraped completely free of oxidation.

>It sure would be nice if Sun would be magnanimous and say "hey, we made a
>mistake . . . " and give 3/75 owners a good deal on upgrading to a 3/160
>housing. ... 

I agree that Sun should fix these problems, since it's clearly caused by a
poor quality connector.  The presence or absence of a maintenance contract
should be irrelevant.  Computers that break because of shoddy materials
are bad for business!  

Paul Allen

Paul L. Allen                       | pallen@atc.boeing.com
Boeing Advanced Technology Center   | ...!uw-beaver!ssc-vax!bcsaic!pallen