22 Mar
2014
22 Mar
'14
1:45 p.m.
On Mar 21, 2014 3:48 PM, "Tony Duell" <ard at p850ug1.demon.co.uk>
wrote:
It's not even that conssitent between capacoitors of the same nominal age
and manufacture. I am sure many of us have had one original capacitor in
a device fail (oxide disoiving -> short circuit) while the others, same
manufacturr, go on for many more years.
can somebody please expalin why I rarely see
capacitor failure, for all I
work on machines a lot older than this Osborne?
It depends a lot on the manufacture of the capacitors, and seemingly not
necessarily in ways that strongly correlate with other specific measures of
"quality". I definitely don't understand it. I would expect the rate of
oxide layer breakdown for non-use to be consistent across all aluminum
electrolytic capacitors, but it isn't. My best guess is that the
electrolyte chemistry somehow stabilizes the oxide. Perhaps they use
additives with that specific purpose. Certainly the electrolyte formulation
varies considerably, and often is a trade secret.
After a decade of non-use, some (but not all) of the 40-year-old capacitors
in the PDP-1 had to be reformed, and a few of those could not be reformed.
It is entirely possible that capacitors made more recently may have a
slower rate of non-use oxide breakdown.
The cynic in me would suggest that the more modern ones are more likely
to fail :-)
More seriosuly, of course I have had electrolytic capacitors fail. But
I've certainly not had to repalce all the ones in all my classic
ocmptuers -- in fact many machines, 40+ years old, are still runnign with
all their original capacitors.
When a machine stops working, IMHO, you should not blindly re-cap it.
Suppose it still doesn';t work after you've done this. What do you then
conclude? That the original fault is still there, it wasn't a bad
capacitor? That one of the replacement capacitors was faulty? That you
made a mistake fitting them?
If you have a fualt, like nist spikes on a power line, that could be due
to a faulty capcitor, then of course you _check_ the capacitors. And
replace (or reform) them if necessary. Capacitors can fail. But changing
parts for no logical reason will get you nowhere fast.
-tony