>The question however, is this. Must the two pieces of Heliax be exactly equal
in length? Or, can they a multiple of a wavelength longer as measured by my
MFJ 259? If so, do they have to be a full wavelength longer or can it be
just a multiple of a half wavelength? My gut feeling tells me that
everything would be fine if added an extra full electrical wavelength. I
think just using multiples half an electrical wavelength would put them into
out of phase.
Your gut feel is correct. A feedline 360 electrical degrees
more than the others will work OK. 180 electrical degrees (0.5 wl)
will put the antennas out-of-phase which is NOT what you want.
I very precisely matched the feedlines on my 3-stack using the
MFJ-259. Later, I once was considering replacing my top KLM-610
monobander with a KT-34XA, so I modeled everything to try to understand
what the effect would be due to the booms not being the same length and
the feedpoint not being physically in the same place. I discovered
phase match is NOT very critical as long as it is within about 10 degrees
(I think the number may be closer to 30 but I didn't keep my data).
If I were you, I would first carefully measure the velocity
factor of your hardline (each one separately if they are from different
material). Then simply calculate the additional length of feedline
for 360 electrical degrees using [(983.56/f) * Vf] where f = frequency
in MHz and Vf = velocity factor. For example, assuming 28.4 MHz and
Vf of .87, you would need to make the longer feedline 30.13' longer
than the short feedline. Assuming you measure correctly, you could
have an error of +/- 30.13/36 or ~10" before you approached a 10 degree
error. If you are a purist, you can use your MFJ to match exactly,
but it is not nearly as critical as I once believed.
73, Bill W4ZV
P.S. For example, assuming phase shifts of 0, +10, and +20 degrees
for the 3 antennas going up my tower, gain only changes by a negligible
-0.08 dB and vertical beamwidth decresases by a negligible 0.1 degrees.