tom wrote in
t:

On 5/4/2010 3:58 PM, Owen Duffy wrote:
The technique presumably is to insert a (lossy) capacitive reactance
in series with the feedpoint, and such that the reactance decreases
with freqeuency, thus exacerbating the natural feedpoint impedance
change.

Owen

I am curious about your statement. You say "insert a (lossy)
capacitive reactance in series". Why would a braidless piece of coax
inserted in a tube have significantly different loss than the intact
coax of that length?

I've made matches made that way for decades that ran at full or near
full legal limit on 6m. I'm pretty sure that any significant loss
would have shown up as dripping plastic. The matches when taken apart
after years of use show no sign of heating.

Tom,

My use of "lossy" was to remind readers that capacitive reactance
obtained by using such a transmission line element is a relatively lossy
'capacitor'. For example. an o/c stub of RG213 for a reactance of -10
ohms at 144MHz has a resistance of about 0.1 ohms, or a Q of about 100.
That is not a huge loss, but quality capacitors achieve much higher Q
than that.

So, I don't know why one might use such a thing in a driven element,
introducing say 0.2 ohms of resistance which consumes about 0.4% of the
power if it was a R=50 feedpoint, when a similar reactance could be
obtained by a slight shortening. The purpose is probably not for
frequency compensation, it works the wrong way.

Is the loss significant, not really in this case, and it won't melt the
PE, but TL derived capacitors are relatively lower Q.

Owen