[TowerTalk] tower mounted vertical

[Terry Conboy]

At 05:03 PM 2005-06-11, Tim Kass wrote:
>If you slope the radials, that will lower the impedance a bit, for a full
>size 1/4 wave ground plane it will match 50 ohms...either way it shouldnt
>change more than about 20 ohms and should still present a decent match to 50
>ohm coax.

I'm sure Tim meant to say that the impedance will go up with sloped 
radials.  I modeled quarter-wavelength verticals on a 50 foot tower, with 
four 45 degree sloping radials.  The feed impedance at resonance varied 
between 33 and 50 ohms, not bad.  (Many people don't realize that a ground 
plane with four horizontal radials in free space is about 22 ohms instead 
of the 36 ohms plus ground losses seen with the radials on the dirt.)  Of 
course, with a trap antenna, things will likely be somewhat different.

>However another point, moving the vertical that high may help
>with receive signals, but a lot of your transmitted signals will go straight
>up or nearly so, fine if you want short DX.

There is some increase in high angle energy, but there will still be fairly 
good signals at low angles.  The worst case I saw in modeling was on 30m, 
where a 1/4 wavelength vertical on a 50 foot tower had a lobe of +4.2 dBi 
at a 46 degree takeoff angle and +0.8 dBi at 15 degrees.  On 20, 15, and 
10m, the higher angle lobes were roughly +/- 1 dB with respect to the low 
angle lobes.  On 40m, the high angle lobe was down -2 dB.  Don't forget 
that raising a vertical above average-conductivity ground increases the low 
angle energy quite a bit.  Not like sitting on salt water, but a nice 
improvement.  (I have the modeled impedances and gains vs. angles if you're 
interested.)

Part of the situation, as W8JI often reminds us, is that a ground plane 
antenna doesn't decouple the feedline (or in this case, the 50 foot tower) 
very well, if at all.  When the tower is near a multiple of 1/2 wavelength, 
there can be lots of current flowing down the tower (or unchoked 
feedline).  This is the case on 30m, 15m, and 10m.  The modeling I did 
assumed the radials are connected to the top of the tower.

73, Terry N6RY