Rick wrote:
On Wed, 04 Apr 2007 21:48:08 -0700, Bryan wrote:


A stub of coaxial cable will work... approximately 30pF/foot for 50 ohm
types... approximately 20pF/foot for 75 ohm types.


Good evening, Bryan.

Thanks, but there are a couple of problems... the antenna will be for 160
through 40 and so two of the traps will be for 80 meters... I forget what
the capacitance will be but I think it was somewhere in the 200 pf range.
Also, the antenna is meant to be portable, and so the traps need to be as
physically small as practical. Also the power will usually be low... as
low as 5 watts, usually no more than 50 and never more than 100 watts...
so I guess I need to make the antenna as efficient as I can... kind of a
contradiction in terms for trapped antennas especially where the traps
need to be physically small.

If I could find the doorknob capacitors for a reasonable price that would
be ideal, but $15 each is way too much.

On the other hand, the ones from RF Parts are all at least 4 KV, and up
from there (40-50 KV). I don't know how to accurately calculate the
voltage rating I'll need (I'll post that as another question under another
topic), but my guess is that for 100 watts max, I don't need 5 KV, and may
not need 1 KV.

Who else besides RF Parts has these things and has a decent website?

I've bought a lot of HV capacitors for various applications over the
years, both surplus and new.

RF parts has fairly reasonable prices for new components. The actual
manufacturer Ceramite, is a division of Vishay. TDK also makes this
kind of capacitor. And, for surface mount, there are some high
current/high voltage RF capacitor manufacturers (ATC, Murata Erie,
Dielectric Labs). These things aren't made in huge volumes and they're
fairly expensive, brand new.

You might find surplus or used parts much cheaper (e.g. eBay)..
Something to watch out for.. there are two very different series of
"doorknob" style capacitors. One is used for DC kinds of applications
and can't handle much RF current (you used to see these as filter
capacitors in flyback type power supplies at 15.75 kHz). The other is
designed for RF current. The ones that can handle current cost a LOT more.

And, if you're buying used... The failure/wearout mode with overvoltage
is that the metalization on the ceramic gradually goes away, reducing
the capacitance. The failure/wearout mode with overcurrent is
microcracks in the ceramic dielectric from thermal stresses. The
capacitance tends to be fairly constant (these aren't high precision
devices anyway.. 20% is a close tolerance part), but they'll have a C
that varies with voltage and temperature, and the breakdown voltage will
decrease ("treeing in the dielectric").

Even the RF devices aren't intended for use in a tuned circuit.. they're
used for DC blocks and RF bypassing, where the capacitance isn't
critical. They often have a fairly large temperature coefficient, as well.

What you might want to look into is capacitors with very low ESR made
for power converters. The Cornell Dubilier 942P or Wima FKP1 series,
for instance. They use polypropylene dielectric with extended foil
construction and 0.15 uF 2kV parts run a couple bucks. They're widely
used in 1MHz power applications (and can be run at 2 or 3x their
voltage ratings, if you want to be bold and thrifty). The key specs to
look at are RMS rf current ratings (which are thermally limited, by the way)

If they aren't low loss enough (or you can't get them in small enough
capacitance), you might look at paralleling or stacking SMT ceramic caps
from the vendors above. 1kV parts are readily available. I have a 1kW
13 MHz power amp here that uses 11 pieces of a 56 pF cap in parallel in
series with the RF and a half dozen 100pF 2.5kV units in shunt in the
output network.


Jim, W6RMK