Topband: Re: Increasing bandwidth 160m L

[Donald Chester]

Don't forget that sometimes increased bandwidth indicates losses within the 
system.  For example, a vertical with an inadequate ground system may 
exhibit a large bandwidth, but as you install more radials the bandwidth 
decreases as the efficiency increases.

The ultimate example of the bandwidth vs efficiency tradeoff would be the 
infamous Maxcom antenna matcher, which was essentially a dummy load with 
nonresonant antenna wires attached. The wires radiated enough rf to produce 
a detectable signal at some distance, while the SWR remained close to 1:1 
from below the broadcast band to vhf.  I have seen other "broadband" antenna 
schemes that included resistive elements to broaden out the resonance 
characteristic.  I believe B&W sold one sometime ago, that was reasonably 
(>50%) efficient.

If you put  up a new antenna and get unexpected bandwidth, treat in with 
suspicion until you have verified that there are no losses in the circuit 
that effectively add resistance.

I built my quarter-wave vertical about 20 years ago.  It consists of 127' of 
tower on a base insulator, with 120 buried quarterwave radials.  Everything 
in the earthing system is brazed together with silver alloy solder.  At 119' 
( top guying level), I have the middle of a 135' long dipole attached, with 
open wire feeders going down to the base through the interior of the tower, 
using plexiglas spacers every 10'.  When the tower is used as a vertical, 
the dipole feeders are disconnected at the tower base using a ceramic knife 
switch, and left floating.  The base of the tower is matched to RG-213 using 
a simple L-network tuned for 1:1 SWR at 1900 khz.

When I first fired it up, I had expected a bandwidth of about 60 kHz, based 
on previous experience with inverted-L's and shortened verticals.  I was 
surprised that the bandwidth was broad enough that I could work the entire 
band, 1.8 - 2.0 mHz, with SWR not exceeding 2.5:1.  The first thing I did 
was to check all connections for good electrical contact and make sure the 
ground radials and the rest of the earthing system had good integrity.  
Everything checked out OK, and when I started using the antenna I got the 
expected good signal reports throughout the band, all over N America.  I 
just left the L-network permanently tuned to 1900 kHz and adjust the 
matching network between the transmitter and coax feedline when changing 
frequency.  I have never tried to measure the field strength at a specific 
distance away from the tower to accurately determine its radiating 
efficiency.

Base impedance measurements gave  interesting results.  Normally a quarter 
wave vertical can be expected to measure somewhere around 40 ohms.  Mine 
measured as follows, using a General Radio 916-AL impedance bridge:

(with the dipole feeders disconnected at the base and left floating)

1800 kHz  117 + j233
1900 kHz  185 + j316
2000 kHz  320 + j425

Thermocouple rf ammeter readings at known rf power levels verified the above 
measurements.

Obviously, the dipole adds considerable top loading to the vertical, making 
it closer to the classic "vertical tee" than a true quarter wave vertical.  
The dipole is not electrically connected to the tower at any point, but the 
close proximity of the concentric feedline - tower combination provides the 
coupling.

Interestingly, electrically connecting the ends of the disconnected open 
wire feeders to the tower base yielded the following results:

1800 kHz  28.8 - j10.5
1900 kHz  45.5 + j30.5
2000 kHz  70  + j99

I always use the vertical with the feedline floating at the base.  I have 
never attempted to analyse the antenna, since the capacitive coupling of the 
tower to flat top through the feedline over the full length of the tower and 
feedline would be a very complex equation to calculate, much like the 
recently discussed topic of using a significant fraction of wavelength of 
coax as a capacitor.  I happened upon an efficient wideband antenna system 
and have been taking advantage of it for over 20 years.

-k4kyv