On Thu, 2007-01-25 at 23:10 +0000, Tony Duell wrote:
1) 90-odd% of DZ11 failures are defective RS232
driver/receiver chips,
probably damaged by spickes induced on the RS232 cabling
Back in the ancient days, I worked repairing computer boards for a
company that made warehousing robots. The robots communicated with the
central computer via the ac mains, buffered by optical isolators.
Boards were not technically repaired, but replaced. A customer would
send in a failed board (and they had GREAT diagnostics to show when they
failed) and would be sent a functional board in return. They kept a
couple spares of each kind on site, usually, as well. I took the failed
boards, and repaired them, and put them in the shippable stock to be
sent out in exchanges. (New sales were all freshly manufactured.)
I worked out a system for repair... I simply removed the optical
isolators, put in new ones, and tested the boards. That simple ploy had
about a 90% success rate. I then asked to make 25% of repair fees in
lieu of salary. I should have done it in the OTHER order... Board
swaps cost about $400, and I did about four a day using their system,
and about ten a day using my system. Of course, once I was bringing in
$4,000 a day, there was no way I was going on commission...
A bit of digression, but the point is that, in some cases, it is
only certain particular components which are subject to frequent
failure, and in these cases, if no others, board repair is cost
effective. (More cost effective if they were to socket those parts...)
3) The chips are cheap and easily availble.
They're pin-through-hole
DIPs. Dike out the old one and solder in a new one.
Likewise in my example -- the boards each had two Optical Isolators,
which were significantly less than a dollar the set.
Peace,
Warren E. Wolfe
wizard at
voyager.net