[ARC5] SCR-522 Project: Finding Bad Bypasses in 20 Minutes.

[Richard Knoppow]

     To repeat something that has been said many times, if the 
bypass and coupling capacitors are paper and are old they _will_ 
be bad.  How bad depends on age and the amount of DC voltage on 
them as a ratio of the rated voltage. Some are better than others 
but all degrade with time. Since moisture in the element is a 
contributor various schemes have been used to avoid moisture. The 
best seem to have been the oil filled variety but those don't 
last forever either. The problem is the degradation of the paper. 
Its an effect similar to the browning of newsprint.
     Old text books sometimes had some indication of the life 
expectancy of paper caps as did some manufacturer's literature.
     Modern plastic film caps have much better characteristics 
than the paper caps did when new and, in general, do not age or 
become degraded with age. Probably the best are polypropylene but 
other types of dielectrics are also good.
    Leakage, that is, parallel resistance, allows some DC to 
flow. This is a problem with bypass caps but can be very serious 
with coupling capacitors where it can change the bias values. 
Leaky coupling caps cause distortion in audio amplifiers and can 
cause bias to become positive resulting in the tube drawing far 
too much current and damaging the plates. In solid state circuits 
related problems can happen.
     In much military equipment the capacitors are in metal cans. 
One can replace them by using standard lead capacitors mounted on 
terminal strips or by restuffing the cans. I have never had much 
success with restuffing probably due to lake of patience because 
many others do fine with it.
     Suspect caps can be tested with a capacitor checker or by 
measuring the DC current they pass. Capacitance bridges that 
measure only value and ESR (or dissipation factor) will not spot 
high leakage caps since they do not measure _parallel_ 
capacitance, only series capacitance.  High ESR can affect the 
performance of a cap but will not cause the problems leakage 
does. Both must be checked on.
     Electrolytic caps also have a limited life and should be 
tested. In general, the familiar aluminum type electrolytic, has 
higher leakage and ESR than paper, plastic, mica, ceramic types 
but there is a limit on acceptable amounts. In general, they do 
not make good coupling caps because of the leakage but are often 
found in solid state circuits where the impedances are so low 
that very high values of capacitance are necessary.
     There is a reasonable amount of guidance in web literature 
on acceptable values of leakage for various kinds of caps. Both 
ESR and leakage should be as low as possible, i.e., low ESR, high 
parallel resistance.
     Failing paper caps can be insidious; the leakage voltage can 
go up slowly, eventually causing tubes to run hot and causing 
distortion. It can happen gradually enough to be easy to miss 
until it gets very bad. This is one reason for justifying the 
wholesale replacement of capacitors.

On 6/29/2018 11:45 AM, Jim Haynes wrote:
> Another approach is just to assume all the bypasses are leaky, or 
> soon
> will be, and replace them all.


-- 
Richard Knoppow
1oldlens1 at ix.netcom.com
WB6KBL