[TowerTalk] Broadbanding 80 meter dipole

[Jim Brown]

On 8/18/2017 7:34 AM, Grant Saviers wrote:
> One not in the current handbook is the Witt/Leeson tuned coaxial 
> resonator feed method,  see 
> http://www.robkalmeijer.nl/techniek/electronica/radiotechniek/hambladen/qst/1993/09/page27/index.html 
>

Garry, NI6T, turned me on to this several years ago, crediting it to 
W6NL. Dave, who teaches at Stanford, has long included it in his 
classes, but denied credit for it, saying that it was far older than 
him. And like me, he's no pup! I included the technique in a talk I did 
some years ago on using SimSmith to design antenna matching networks.

http://k9yc.com/PacificonSmithChart.pdf

I showed that the technique is not limited to a one-wavelength matching 
section -- indeed, a SimSmith model shows that it works fine with any 
whole number of half-wavelengths up to about 4.

This technique does not, however, work for a multiband dipole to cover 
80/75/40. I have such a dipole, so I worked long and hard at a model and 
never arrived at anything that looked good on both bands. The problem 
occur on 40M.  Instead, I switch in stub matching networks in the shack. 
To design them, I measure feedline Z with my DG8SAQ VNWA 3e, export the 
data to SimSmith, and do the design work there. This is also included in 
the SimSmith talk. Among other things, if line length and coax type are 
known, SimSmith can show plots of feedline loss vs frequency. SimSmith 
has data for many popular feedline types, and it's easy to measure the 
electrical length of a feedline with the VNWA 3e.

http://sdr-kits.net/VNWA3_Description.html

Another point. Even if the tower is not resonant, it is still an active 
element of any vertical antenna in the vicinity. N6LF alerted me to this 
several years ago in an excellent study that he posted to his website.  
I verified this by building a model of my 160M Tee vertical with a 
grounded tower that wasn't even close to resonance and about 80 ft from 
it,  and saw a bump of several dB in the horizontal pattern.

The strength of the effect also depends on the impedance to earth of the 
ground for the tower at the operating frequency. That impedance (mostly 
R) shows up as a resistor in series with the base. I have two 
quarter-wave wires sloping in opposite directions off the top of my 120 
ft tower to a feedpoint about 60 ft from the tower. They are insulated 
from the tower by a 10 ft length of 4-in PVC conduit. Each wire has four 
radials elevated about 20 ft (this height recommended by N6BT), and I've 
added 8 on-ground radials to the base of the tower to lower the Z of the 
earth connection (because the tower is a passive element).  NEC predicts 
about 2 dB gain from this antenna, and on-air switching between the two 
sloping verticals confirms the directivity (although not the gain, which 
is FAR more difficult to measure).

73, Jim K9YC