> // During a glitch, the ceramic case tends to explode and throw
> schrapnel around one's abode.
Wow! Is that ever a true statement.
However, this result is from selecting a resistor not robust enough
for the job in the first place.
Most commercial (MRI, etc.) amps I have seen use a 100 watt power
resistor mounted on the power supply chassis, as Ian suggests. It is
always mounted on insulators. The ones with the slick brown shiny
enamel are best. The 100 watts is not for power handling. It is to provide
a lossy path for HV arcs between turns. This is a poor man's HV resistor.
In applications where the power supply is on the RF deck chassis, and
space becomes a problem, one must bite the bullet and spring for a
real HV resistor. The value of the resistor depends upon several things.
It must limit current until the primary fuses/breakers do their thing and not
self destruct in the process. Usually 10-50 ohms is what one ends up with.
Eimac has a tech sheet with complete details on how to design and test
such a device.
FAQ on WWW: http://www.contesting.com/FAQ/amps
Administrative requests: amps-REQUEST@contesting.com