I thought my experience might be of some interest, even as a negative example.
Earl, K6SE was nice enough to model my one-tower setup, using techniques
that he has proven to be quite accurate in predicting shunt feed
characteristics. It turned out that the coax feedlines for my 80-meter
lazy vee dipole array were a big problem. The way it is now switched, one
feedline is always connected, and because the feedlines run down inside the
tower leg, I think that shield is effectively tapped to the tower at the
point where the feedline enters the tower. As a result, there are whopping
currents induced on the feedline, and it totally screws up both the
radiation pattern and the nice normal procession of increasing R values as
you go up the tower.
Wanting to see if the model was telling the truth, I tapped the tower at 50
feet, where a model without the 80m feedline said I should find 50 ohms.
What I found instead was 24 ohms plus a lot of inductive reactance. Good,
I thought. Wrong! Attempts to tune out the reactance with a series cap
failed miserably, and I also discovered that the R component changed not at
all over an octave in frequency. Clearly, there is not a viable tap at
that level, and while I wonder where all that R comes from, I'm going to
abandon the shunt feed for now and start looking at trees for an inverted L.
This seems to be confirmation of the importance of having no feedlines
entering or leaving a shunt fed tower above ground. Or am I
misunderstanding?
73, Pete N4ZR
|