[Collins] Caps. Paper or Plastic which is better?

[Dr. Gerald N. Johnson, electrical engineer]

"Good 'ol" ceramic disks do not have a stable dielectric. The high
dielectric constant ceramic necessary to make .1 and .01 cap with
adequate voltage ratings has a wild temperature coefficient. They can
loose 80% of their C going from room temperature to the limit of their
operating temperature range. They can loose 30% just being nestled
against a tube socket. I learned that the hard way, long ago.

That dielectric is usually barium titanate, which is piezo electric.
That contributes to receiver distortion because that makes the
capacitance voltage sensitive and if across the audio output transformer
can cause the capacitor to act like a tiny speaker. You can check for
piezo electric effect by connecting a capacitor (I recall using a .01
for this experiment 40 or so years ago) to a VTVM on the 1.5 volt DC
scale, then lay the capacitor on a small vice and tap with a small
hammer. You'll see significant fractions of a volt. You get more the
harder you hit, until you smash the capacitor, which you surely will in
this fascinating experiment.

I like orange drops because I have abused them and found them perfect
after the abuse, like a wash in my pants pocket through the manual
maytag, including a pass through the wringer while wet, then dried in a
high speed electric clothes dryer with no change in leakage. There are
other makes that I've not personally tested so rigorously that are
coated with deep maroon or sea foam green.

Replace all the oil paper capacitors. Don't bother testing them, unless
they have a second insulation of mylar (polycarbonate) they have been
leaking for 20 years. And reducing receiver gain by lowering screen
voltages, upsetting AVC by loading the AVC line, and shortening output
tube life by applying positive bias to the control grids. Testing
requires removing them from the circuit. Its easier to skip the test and
install new with longer leads than to try to work the olds back in, when
the occasional capacitor passes the leakage test.

73, Jerry, K0CQ, technical advisor to the CRA.
-- 
Entire content copyright Dr. Gerald N. Johnson, electrical engineer.
Reproduction by permission only.