[TenTec] Radials

[David Kelley]

>And once again, the question of 30x 1/4 vs 60x 1/8 wavelength radials
>remains unanswered.
>I find that to be a very interesting and practical question.
>In all of my reading in the past, including W6SAI's entire book on vertical
>antennas, I don't recall that ever being addressed.
>
>73
>Rick

I think this question has been answered.  Tony W4FOA cited the following
article on the NCJ web site by Al Christman K3LC, in which ground-
mounted verticals with buried radials were modeled using EZNEC 4:

http://www.ncjweb.com/k3lcmaxgainradials.pdf

The basic lesson I get from this article is that you should bury as much
*total* wire as possible.  That is, a total of 4000 feet of wire is going to
outperform a total of 2000 feet of wire.  For a given total length, there is an
optimum length for the individual radials that will maximize the gain.
However, based on the data presented in the article, doubling the total
amount of wire yields less than 0.5 dB of additional gain (for average
ground), and the gain penalty for not optimizing the individual radial lengths
is no more than approximately 0.3-0.4 dB.

I have forgotten which band the original question referred to, but let's say it
was 160 meters.  A free-space quarter wavelength on that band is approx.
130 feet; 1/8 wavelength is 65 feet.  Thirty 1/4-wave radials amount to a
total wire length of 30 x 130 = 3900 feet - let's call it 4000 feet.  Sixty
1/8-wave radials yield the same total length.  Referring to Christman's
Fig. 3 (assuming average ground and a 1/4-wave vertical radiator), we see
that 4000 feet total will yield a gain between 0.6 and 0.9 dBi.  The optimum
number of radials, which gives 0.9 dBi of gain, is 30-40.  Let's use 36 (one
radial every 10 degrees).  Since a total of 4000 feet of wire is available, the
individual radial length should be 4000 / 36 = 111 feet.  Of course, a few
feet longer or shorter will not make a noticeable difference, so excessive
effort should not be expended to get the lengths "just right."

As K3LC states, these results are valid only for 1/4-wave radiators.  He
doesn't state in his article how the results would change if the radiator
were shorter than 1/4 wave.  However, since the radiation resistance
decreases as radiator height decreases, a larger amount of current will
flow in the radiator and in the ground for a given amount of input power.
The quality of the ground would have to be even greater to achieve a
given gain target for a shorter radiator.

73,
Dave NB4J


--
David F. Kelley, Ph.D.
Assistant Professor
Electrical Engineering Dept.
Bucknell University
Lewisburg, PA 17837
(570) 577-1313
dkelley at bucknell.edu