[AMRadio] Power factor

Bill Smith billsmith at ispwest.com
Thu Apr 4 14:48:00 EST 2002

Hi Don

George and Donald have given dissertations well beyond my capacity, and have
described the culprit: resistance.  An ideal capacitor is comprised of two
surfaces which are in proximity to each other and separated by a dielectric.
The surface can accumulate electrons given applied voltage.  Of course all
capacitors are made of real materials and have limitations, and/or may be
optimized for size or cost.


George identified two failure modes, one of leakage, another of series
resistance, discussed below.  A paper capacitor is made of thin sheets of
aluminum foil separated by sheets of paper, all rolled into a tube.  You may
find it  interesting to take an old wax paper capacitor apart.  Apparently,
(I really don't know for a fact), aluminum from the foil migrates over time
into the paper insulation and creates an undesired dc path.  At any rate,
old capacitors are found to develop "leakage" resistance in parallel with
the capacitor, or across the capacitor's wire terminals.

A quick go/no-go paper capacitor test can be accomplished with a analog VOM
(Simpson 260 et all).  Simply switch to the 10,000 ohms scale and connect
the test probes to the capacitor's terminals (out of circuit).  You should
observe the meter "kick" as the capacitor charges, but the meter should
quickly fall back to infinite resistance.  If it measures any residual
resistance, if the meter doesn't fall all the way back, the capacitor is

You won't see much, or any kick on smaller caps (1000 pf or 0.001 mfd).  On
larger caps (0.05 - .1 mfd range), temporarily remove a probe, wait a second
or two and reattach.   If the capacitor can hold a charge for this period of
time, you won't see a kick when you reattach.  If the capacitor is leaky, it
will discharge the meter test voltage and the meter will kick again, and the
part is bad.

A VTVM will work also, but typically uses 1.5 volts.  The Simpson (at least
the newer meters) uses a 9 volt battery on the highest ohms scale, which is
a better test.  Realize that these tests are made at a fraction of the
hundreds of volts which are typical operating voltages found in tube
equipment, so any leakage at these lower voltages is a sure indication of
component failure.  Leakage may not be a significant factor depending upon
where in a circuit the component is used, but I don't make this distinction,
I replace the part.

Capacitor testers, which are readily available at swap meets for $10-25 have
a leakage test function.  They test leakage at the rated voltage of the
component.  If you are just starting out, they are a very useful addition to
the bench.


I have used power factor very loosely, but what I have been talking about is
resistance in series with our ideal capacitor.  Again, using the wax paper
capacitor as the guinea-pig (you can take one apart without too much risk of
PCB exposure as far as I know), the wire terminal ends of the capacitor are
attached to the coiled aluminum by what looks like a wad of solder!  I don't
think there is much more of a bond from a terminal of an electrolytic to the
aluminum foil (inside those devices) either, perhaps there is an
electro-weld, but in both cases the tendency for a connection to open up is
common.  While there are other failure modes such as electrolyte drying,
component attachment gives a simple concept of resistance developing in
series with the capacitor.

Again, George well described the differences between various types of
components and why some are selected for certain applications over others.
In our equipment, one can again reach to the capacitor tester to verify
whether a part is still functional.  If a small capacitor is open, it will
not test well in a capacitor bridge.  Most bridges have a "green-eyed
monster" indicator tube as the indicating device.  During capacitor
measurement, a pie-shaped shadow inside the tube should close fully when the
tester is adjusted to indicate the correct value of the capacitor.   Poor
capacitors (high internal resistance) will not allow a full closure during
capacitor measurement.  Very open capacitors will not show any closure of
the shadow.

A special test is provided on capacitor testers for larger values of
capacitance.  It is often labeled "power factor", which is the source of the
reference I made previously.  The power factor test is a deliberate
resistance inserted in the internal bridge circuit of the tester to balance
(mirror) any series resistance inside the component under test.  Russ
mentioned good parameters.  Many modern electrolytics are constructed with
deliberate resistance built-in (so that they don't blow up when connected
directly to a solid-state diode in a high-voltage supply), but as old
capacitors fade, their power factor or series resistance increases.  This is
an undesirable effect, because it reduces the ability of the component to,
for example, smooth the DC voltage in a power supply.

Tests with a capacitor tester should be made with the component out of
circuit.  An oscilloscope which has good response at the operating frequency
of the equipment can be an invaluable aid, too.  I once found an open
ceramic capacitor in a broadcast transmitter operating on 80 meters with a
'scope.  I attached the scope to the screen of a driver tube, and found all
sorts of RF present.  It was well bypassed with an 0.01 mfd ceramic cap,
which left me scratching my head.  I replaced the part, and the RF went
away, then tested the capacitor on a capacitor tester.  The part tested
perfectly ok!  Who knows how many similar components are quietly giving fits
in other equipment.  But generally, a capacitor tester provides a good test
to identify failed components in our boatanchors.

Mica capacitors are normally considered to be bullet proof, but they can
also fail. The square, phenolic-encased mica capacitors generally open when
they fail, but can exhibit leakage also.  I have found several leaky, open,
bad mica caps in an R-390.  Interestingly, the caps I would have suspected
to fail were the tubular caps, but the Vitamin-Q type have an excellent
reputation and are apparently good in this receiver.

73 de Bill, AB6MT
billsmith at ispwest.com

----- Original Message -----
From: "Donald Chester" <k4kyv at hotmail.com>
To: <billsmith at ispwest.com>
Cc: <amradio at mailman.qth.net>
Sent: Thursday, April 04, 2002 8:24 AM
Subject: [AMRadio] Power factor

> >From: "Bill Smith" <billsmith at ispwest.com>
> >If the cap shows high leakage, or a high power factor, it is probably
> > >spent, but if it acts as a new capacitor you have saved a good part.
> What exactly do you mean by a capacitor's power factor?  How do you
> it and exactly what are you measuring?  I am familiar with the concept
> talking about transformers, but what about capacitors?  I would assume the
> power factor of a good capacitor standing alone would be zero, since the
> (ac) current would lead the voltage by 90 degrees.
> Don K4KYV
> _________________________________________________________________
> Join the world's largest e-mail service with MSN Hotmail.
> http://www.hotmail.com
> _______________________________________________
> AMRadio mailing list
> AMRadio at mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/amradio

More information about the AMRadio mailing list

This page last updated 16 Dec 2017.