[Top] [All Lists]

Re: Topband: Elevated Radial Efficiency - an inordinately long post

To: "William Culpepper" <>,<>
Subject: Re: Topband: Elevated Radial Efficiency - an inordinately long post
From: "Tom Rauch" <>
Reply-to: Tom Rauch <>
Date: Thu, 29 Nov 2007 12:15:15 -0500
List-post: <>
Hi Bill,

I strongly disagree that the early tests and articles meant 
very much. First, they deluded Hams into thinking 120 
radials are needed to achieve maximum efficiency in an 
array. The truth is the efficiency curve of eff vs. number 
of radials flattens out in most cases with as few as 40-50 

Second, it wasn't a good test for measuring a variation of a 
few dB.

>I am ONE of the people who claim that four elevated radials 
>can have
> approximately the same efficiency as 120 buried quarter 
> wavelength radials.

The only possible way to make that test with any reasonable 
accuracy would be to install four elevated radials over 
virgin soil and then remove them and replace them with 60 or 
120 radials on the ground while measuring FS at the very 
same points. I'll explain why below.

> I have installed such systems at three Standard Broadcast 
> stations in the
> United States, and made field strength measurements that, 
> when analyzed in
> accordance with FCC procedure, showed that the 
> unattenuated field strength
> at one kilometer was essentially the same as the FCC 
> criteria for broadcast
> antennas with 120 buried  90 degree radials (Figure 8 of 
> Part 73 of the FCC
> Rules).

The FCC test works like this:

1.) A person measures the slope of the FS changealong 
straight lines radiating out from the antenna for a 
considerable distance.

2.) From that slope a person finds a "best match" to a 
graph. This graph then estimates the average conductivity 
along that path.

3.) From that estimated conductivity a person then estimates 
the expected FS and expresses it normalized to some 
standard, like a kilowatt at a kilometer.

The problem is the variables and the human element. There 
are many wobbles and variations in the slope of FS curves. 
Sometimes even moving just a few feet, a person can make an 
array have any efficiency he wants within reason! It's easy 
to have 10dB of variation with radial distance with just a 
few feet of movement!! So what does the person use?

As an example if we look at the FS measurements Belrose used 
in an elevated radial article, the slope of the FS could be 
matched to two or three different curves. Not only that, he 
picked the "best" radial direction for estimating a high 
efficiency. The "wobble" or deviation in slope was 2-3dB or 
more over very small distances in points along a radial 
direction. It just happened that picking the worse sloped 
curve made the earth look lossy, and that made the array 
look better because it makes the EXPECTED FS look much 
lower. This means the array efficiency looked better. If the 
mean value of slope was used the earth loss would have 
looked better, and the array efficiency looked worse. Even 
by picking another direction the array would have looked 

The FCC estimate is also an average of the entire path, not 
the localized conductivity. It treats the earth as a 
homogeneous media with constant conductivity. That isn't the 
real earth.

Making the tests even less reliable, one of the early 
elevated radials tests was made over soil that contained 
unknown remaining radials from a earlier ground. Who can say 
how much effect that added to the mix??

So how would we do a good test with minimal chance for 

We could do a good test by changing nothing except the 
radials, and measuring at exactly the same points. Then we 
don't need to match dozens and dozens of readings into a 
sloped curve, and then transform that result into another 
estimate of what we should have at a certain distance, and 
then find a spot at that distance that proves we get what we 
guessed we should get.

We simply go back to the very same point and record the 
change with only one thing changed, the radials. The result 
is the exact deviation caused by the ground system, within 
limits of human error and instrument calibration. The result 
is not a long process of matching curves to estimate the 
soil conductivity, and then extrapolating that conductivity 
data into an estimate of efficiency based on the estimated 
soil conditions along the path and expected FS at a certain 

We actually did several tests where FS was measured with a 
small elevated system and with no changes at all other than 
the radials FS was remeasured. In those tests four elevated 
radials that were ground isolated were compared to a 
conventional ground of 60 radials with no change except the 
radials themselves. The results showed the small elevated 
system was a few dB to 6 dB down from 60 radials. This test 
repeated with similar results at three locations, one of 
which was WVNJ radio. WVNJ was a good test because they 
never had a conventional ground, they started with four 
elevated radials. It was easy to watch the FS at measurement 
points increase 2-5dB as the ground system was changed by 
adding more and more radials.

Roy Lewallen was out here last year, and we found four 
elevated radials about the same as 10-15 radials in the 
dirt. That was on 40 meters. Why would I put up with all the 
headaches of suspending an antenna and radials up several 
feet and NOT having good lightning protection when I could 
just tack 10-15 radials down on the ground and have a normal 
system that works exactly the same?

73 Tom 

Topband mailing list

<Prev in Thread] Current Thread [Next in Thread>