[Towertalk] grounding an elevated vertical

[Tom Rauch]

> There was significant quantitative work done on 1/4 wave verticals
> with elevated radials, in the late 70's and early 80's.  This was
> intended for AM broadcast use, but the experimental work was carried
> out at 1.8 MHz by amateurs who are also consulting engineers.

Actually there was no quantitative work done at all, other than 
models. The only data from measurements used standard FCC 
slope-of-ground-conductivity measurements, and they can easily 
be all over the place for accuracy.

When i looked at data that was taken from these measurements, 
they didn't even use the mean FS value where the data had 
"flyers". (Flyers are where the FS goes through abnormal changes 
in level with distance.)

The only accurate way to measure an efficiency change would be 
to only change the radials, and measure at the same points with 
ONLY a radial change. To the best of my knowledge that was only 
done a few times, and the change in FS was significant in favor of 
the conventional system.    

> To be sure, copper plating earth will improve the performance of a
> vertical.  It's a question of marginal improvement in efficiency
> versus expense.   120 buried radials are the broadcast norm.  But 16
> radials will get you to 90% efficiency.  And 3 or 4 elevated radials
> will do the same.  

16 radials does not normally get you to 90% efficiency, and neither 
does 3 or 4 elevated radials unless they are 1/4 wl high or higher 
above earth.
 
> Empirical data presented to the FCC showed that 4 radials, elevated @
> 10' were the equivalent of 120 buried radials, in terms of field
> strength. If I recall the experiment, though, there were REMAINS of
> those 120 radials still in the earth, and the ground conductivity was
> exceptionally good.  

The only system like that I recall was in the Carolinas, were soil is 
not particularly good. That paper had almost zero technical merit 
because:

1.) The measurement was based on using slope of attenuation with 
distance to calculate expected FS, and relatively little data was 
taken. You can get almost any result you like within ~3dB using 
that system, or more if the engineers aren't very careful or totally 
unbiased.

2.) An unknown amount of original radials remained.

3.) The station where the measurements were made reported poor 
coverage to me on the telephone, but they said they really didn't 
care because they only needed to cover a few miles. They claimed 
a proof was never done, and never needed, because they were a 
250 watt non-directional (running at reduced power).

Elevated radials have largely fallen from use now at BC stations, 
just as they did in the 1930's when the original RCA study was 
made.

If we think back, virtually all original ground systems were small 
resonant counterpoise systems. Then we got smart, did tests, and 
started using the better system. Later, we tried going back to the 
system that was abandoned in the 30's!!!

Two steps forward, one step back?

Bottom line is use as many radials as you can as long as you can, 
and if you can only use a few radials by all means elevate them. If 
you have 8 radials at 1/4 wl height, it is indistinguishable from 60 
radials or 120 radials on the earth (based on IEEE papers detailing 
measurements comparing one system both ways).

If you have 4 radials, try to keep them 1/2 wl above earth.

In all cases with small elevated radial systems you should use a 
feedline choke at the antenna, and keep the radials isolated from 
earth paths at the operating frequency.

Even a 1/4 wl groundplane with four radials has considerable 
feedline "earth currents" unless it is choked at the feedpoint.  
73, Tom W8JI
W8JI@contesting.com