Re: Electrolytic caps and Apple analog circuits (semi on topic)

["Matt J. McCullar" <mccullar@flash.net>]

I worked on jillions of Apple Computer switch-mode power supplies.  I've
worked on far more various types of video monitors of all kinds, too.

The biggest enemy of any electronic component is heat.  That's what degrades
electrolytic capacitors, and why some seem to last forever, whereas others
have taillight warranties.  Some caps are exposed to high-temperature areas,
while others are not.  There's plenty of heat inside a video monitor, and
dust tends to collect inside.  That traps even more heat.

The older a video monitor is, the more likely that it's got at least one bad
electrolytic capacitor -- perhaps several.  It's been my experience that
it's not a change in the capacitance value itself, so much as a big increase
in a capacitor's Equivalent Series Resistance -- or ESR.  It's like a cap
has been bridged with a resistor of just a few hundred ohms.  That causes
all kinds of weird things to happen.  True, some caps do just flat-out open
up.  And a cap can change its capacitance value dramatically without
affecting its ESR.  But in my experience, the increase in ESR is the most
common failure.  It's best to just replace them all if you can, at the same
time.  Reflow the solder joints while you're at it.

When I worked on Apple II power supplies, the most common failure was the
same electrolytic capacitor.  It wasn't on the DC output sections, it was in
the feedback loop for the main switching transformer.  That caused the
failure of all output voltages.  At the time I didn't have access to the
gosh-wow cap testers I have now, so I don't know offhand if it was higher
ESR (which I now consider likely) or capacitance change, or what.  I do know
that replacing that capacitor cured the problem.  Soon my job consisted
mainly of knocking out rivets or removing screws from those power supply
lids, replacing that cap, and Voila!

As for the video image shift in a monitor, that's more likely due to some
sort of timing issue.  There might be a potentiometer adjustment inside the
monitor for that.  A phase shift in the horizontal sync signal (or even a
lengthening or shortening of the sync pulse) could cause that, too.
Critical timing circuits tend to use high-precision capacitors such as
tantalum, not electrolytic.  Look for a horizontal sync adjustment pot; it
may be a horizontal phase adjustment pot, too.  Look for a pot marked HSYNC
or HPHASE.  Be sure to use a magic marker to note exactly how the pots were
set before adjusting them.