[AMRadio] Condx. caps

Donald Chester k4kyv at charter.net
Wed Oct 19 13:08:18 EDT 2016

> So, after a week of
> incremental voltage increase, rest time and cycle it again, TLC'ing it
back to
> life, it looked like it might make it OK, only to quit in glorious
fashion.  Glad I
> saved my prepared Newark order, now I just have to click 'buy'.

I have an old Heathkit capacitor checker (dumpster find) that works pretty
well for checking  leakage, but the capacitance bridge is so inaccurate that
it's useless.  It has 0-600 volts DC output, using a magic eye tube to
indicate leakage current.  It can be used to  recondition electrolytics by
attaching the capacitor, turning up the voltage until it shows leakage.  Let
it sit, and the leakage current gradually drops down until it is close to
normal. Then I turn the voltage up until it shows excessive leakage current
again, and repeat the process.  Keep going until it reads normal at full
rated test voltage.  If it takes more than an hour or so for the leakage
current to approach normal, I discard the capacitor. After an apparently
successful re-condition,  I discharge the capacitor and let it sit overnight
or maybe for a day or two. Then, if it still checks good with no more than
maybe a slight rebound of leakage current, I'll re-use the capacitor.  If
significant leakage has returned, I discard it, since with any piece of
equipment it may sit idle for weeks or even months without being turned on.
I have found that even brand new electrolytics tend to show a slight leakage
before the rated voltage is reached but they re-form easily, so I run the
procedure on every capacitor before I use it.  I have caught a couple of
brand new but bad capacitors that way.  I rarely use an ancient electrolytic
for a project even if it appears to recondition OK, unless that's the only
one I have and don't want to wait for a new order to arrive, but I'll make
note of the capacitor and make sure it gets wired into the circuit in a
fashion that would make it easy to replace.

With non-electrolytics, even if they check prefect on the checker, once I
know for sure I can apply full voltage without the capacitor  drawing
excessive current or shorting and destroying the meter, I then measure the
actual leakage with full test voltage applied using my DVM set to the
microamp  range.  I suppose a capacitor could short unexpectedly just the
instant I was taking the measurement; that's just a crap shoot that I have
taken for years and so far never damaged the meter, although I seem to be
prone to just exactly that kind of Murphy's Law event. The Heathkit checker
has a series resistor that would limit the current in the event of a  short.
I don't know how much current it would take to damage a DVM set to
microamps, but I know it wouldn't take much to destroy a regular moving coil

A have a junk box where I save old wax capacitors, just in case I might ever
decide to re-stuff one for a vintage restoration or replica, something I
have only actually done once or twice.

Don k4kyv

More information about the AMRadio mailing list

This page last updated 18 Feb 2018.