Dear Andy, LF Group,
The idea that any short feeder of only a few electrical degrees has
virtually no effect on impedance seems to be popular, but it just isn't true
as you show here. A more realistic statement would be: If the load impedance
is of a similar magnitude to the feeder Zo, the short feeder will have
little effect on the impedance. If the load impedance is much higher than
Zo, the feeder will act as a shunt capacitor (100pF/m for Zo 50ohm), and if
the load is much lower impedance, it will behave as a series inductance
(0.25uH/m for Zo = 50ohm).
Since the impedance at the point the feeder connects to the antenna tuner is
at a fairly high impedance in this case, the 12m of 50ohm feeder behaves
roughly as a 1200pF shunt capacitor, giving the parallel resonance peak.
Moving the variometer to the antenna end of the feeder gets rid of a third,
higher, resonant frequency, approximately where the variometer inductance is
series resonant with the cable capacitance. But it does nothing to get rid
of the parallel resonance in your plot, just shifts it a bit in frequency -
all you are doing is adding to the loading coil inductance by a small,
variable amount. I think with a high-Q antenna, these multiple resonances
The variometer-in-shack scheme becomes more manageable if the feeder has
less capacitance. This can be done by having a high Zo feeder - probably the
best feeder is just 2 wires, separated by a metre or more. This has a high
Zo, more similar to the off-resonance impedance of the antenna, and a
capacitance of probably < 10pF/m - the parallel resonance would still be
there, but would be much further from the wanted series resonant frequency.
I have used such an arrangement a few times, and it worked rather well - the
wire between the variometer and loading coil needs to be reasonably well
insulated and spaced from conducting objects. The wire from TX earth to
antenna ground can just be laid on the ground.
Cheers, Jim Moritz
73 de M0BMU