Topband: 200pF HV-capacitor for shunt feeding

[Tom Rauch]

> If the coax has about  28 pF per foot (memory again), then
you should  be
> able to get up to a bit over 600 pF with a single coax.
If you use the two
> paralleled coaxes, then this will still allow you to get
up to about 300 pF,
> so again it should be satisfactory.


Coax is a transmission line, it has both series inductance
and shunt capacitance and it has standing waves.
Ignoring that effect is the same thing that causes trap
calculators to have error.

A length of coax used as a capacitor is actually only an
open stub. The voltage across the open end is higher than
voltage across the end where power is applied, the
"capacitance" is higher than the pF/ft we calculate using
only length times capacitance for unit length. Think of it
this way, if the line is 1/4 wl  long or longer capacitance
goes to zero and then negative (inductive).

One way increase Q, reduce loss resistance, and decrease
peak voltage is to fold the transmission line back around
and connect the shield at the what would have been the open
end and the center conductor at would have been the open end
in parallel with their respective points that are being
normally fed. This does reduce the effective capacitance.

A large error in calculated C occurs unless the line is very
short in terms of wavelength. In other words, if you don't
provide room for those standing waves that increase loss,
voltages, and modify capacitance to stand...the capacitor
will look more like a capacitor and less like a transmission
line with open circuit load.

As for PC boards, they generally make *very* poor
capacitors. So does glass and aluminum foil and many other
things being used.

73 Tom