[Prev][Next][Index][Thread]

Re: Partial Discharges.....A good reason for cap failure?



"Very good description of cap failure over time. This is interesting and
does sound like a valid reason why caps fail over time. Even
professional
caps have discharge life expectancies. We should learn from the
manufacturers here that reducing the voltage across the cap (series
arrangements) promotes the best chance of longevity, especially since if
we will never be able to match a cap manufacturers processes,
cleanliness,
and experience. However, we can series up what we have and increase the
life.

Bart"

	When I worked at Hughes Aircraft I was quite involved in the design of
high-power microwave transmitters, and was a customer for some of the
pulse-forming networks our components group built.  The capacitors were,
of course, a key element.  Our "capacitor guys" maintained that almost
all premature failure (within rated voltage) was caused by corona in
trapped gasses, which resulted in ozone, which eventually destroyed the
insulation and started a breakthrough path.  One of their tests for good
capacitors was to use a radio receiver connected across a resistor in
series with the capacitor.  The "high" side of the capacitor was fed
from a variable voltage, 60 Hz transformer.  As the AC voltage was
increased, a value would be reached at which the receiver would start
picking up a 60 Hz "sizzle".  The voltage at onset and the level of
noise were interpreted in terms of the potential for failure.

	Point was that, even with vacuum impregnation, there were always voids
in the capacitors which produced corona when the applied AC voltage was
high enough, and that the corona induced insulation failure.  I would
think that similar tests would be easy to perform on the home built
capacitors often discussed here.

Ed