[R-390] Paper Capacitor Replacement II

[Charles P. Steinmetz]

>So, an all-polystyrene "array" would be best?

Yes, aside from temperature compensation and any need for operation 
at VHF or above (not likely with a 0.5 uF cap, eh?).

>I found some 0.047uF @ 1000V.  Ten of those in parallel and then 
>some smaller value polystyrenes to get it to 0.5uF within as close 
>as I want to get it.
>
>The idea of applying a constant correction factor isn't very 
>appealing and getting this new capacitor to under 0.05% or better 
>should not be too hard to do (as long as the met lab doesn't turn me away).
>
>Any problem/caveats with this approach?

Only temperature compensation, which it would be relatively difficult 
for an amateur in a home lab to organize without a substantially 
better bridge or C-meter than the one being repaired -- which I 
gather is not your situation.  Polystyrene capacitors typically have 
tempcos of -100 to -300 ppm/degree C, so you should be able to hold 
+/- 0.1% over at least +/- 3 degrees C above and below the 
temperature at which they were measured -- acceptable for a lab 
instrument.  You may want to let the bridge warm up fully and measure 
the temperature inside, and ask the lab if they can measure your cap 
at that temperature.

BTW, correction factors are a fact of life in the calibration 
world.  Don't be scared of them.  Just keep a pocket calculator handy.

I'm curious where you found your 1000v polystyrenes, and what brand 
they are.  (I like the Eurofarad polystyrene caps, but I don't know 
if they make anything above 500v.  ACC does, but only to order.)

Best regards,

Charles