Hello,

Compensating an off-resonance antenna with increased TX power
seems like a brute force method to me. Why not build a simple phase
measuring thing that outputs a +/- voltage that is proportional to the phase
relationship between voltage and current in the feeder? After all, the
theoretical definition of resonance is that the voltage and current are in
phase with each other.

Assuming zero output at 0 degrees, the output from the "phase
sensor" can be used to control a servo motor on a variometer which
brings the system into perfect resonance whatever the wheather...
An op-amp in the servo loop, surrounded by suitable R's and C's, will
make the system slow and stable (dominant pole). It doesn't have to
be faster than the changes in wheather ;-)

One of my ideas for a "phase sensor" is:

The feeder voltage and current can be sampled with a resistive (or
capacitive, or both) divider and a current transformer, respectively.
These two signals are then amplified and clipped to square waves.
If these signals are fed into an XOR gate (74HC86 etc), the output
from the gate will be a PWM signal that is proportional to the absolute
value of the phase error. The PWM signal can be converted to a DC
error signal by a simple RC lowpass filter (136 kHz >> the servo loop
bandwidth). The sign of the error signal can be found by connecting
these two signals to a D-type flip-flop (74HC74, 4013 etc).
The current and voltage squarewave signals are connected to the
clock input and the D inputs respectively. The sign bit will appear on
the flip-flop Q output. This sign bit controls the direction of servo motor
travel and the filtered PWM from the XOR gate is the magnitude of the
error.

Depending on actual implementation, some kind of "gating" may be
needed to prevent the servo loop from "run-away" during key-up periods..


Just an idea...

73 de Johan SM6LKM