Richard Clark wrote:
On Thu, 2 Sep 2004 14:27:37 +0000 (UTC), "Reg Edwards"
wrote:


Incidentally, ground loss is not only smaller in sea water, it is also
smaller with soil resistivities of several thousand ohms and greater.
There's a maximum somewhere in between.


Hi George,

The statement above falls into the category of "Old Wives' Tales."
Given the choice for conductors, Sea Water ranks 6 or 7 orders of
magnitude in worse conductivity than any metal (or even carbon) you
would care to pick. By this logic, you should do everything in your
power to operate in an open pit coal mine. ;-)
. . .

It's not an "old wives' tale" at all.

We have to remember that there are two loss mechanisms involved with
antennas. One is the resistance encountered by ground current flowing to
one of the feedline terminals in a ground-mounted antenna. This can be
reduced to an arbitrarily small value by installing radials. The other
is loss incurred when the field strikes the ground and reflects. This
second loss is generally negligible for horizontally polarized antennas
except at high radiation angles, but is very significant for verticals
at low angles. This loss occurs mainly beyond the far edge of most
radial fields, so there's usually nothing you can do to reduce the
reflection loss except to move the antenna.

Perfectly conducting ground has no loss, and perfectly insulating ground
has no loss. When considering ground system loss (resistance encountered
by local ground currents), the loss is maximum somewhere in between.
Doing some experimental modeling with EZNEC/4 (NEC-4 based), I've found
that the maximum loss for a radial ground system in the HF range
unfortunately hits somewhere around average soil conductivity. It's not
an abrupt maximum -- the loss varies fairly gently with conductivity.

Reflection loss has to be considered a little differently. While the
same statement about perfectly conducting and perfectly insulating
ground is still true, if you had perfectly insulating ground, radiated
power would be lost to useful radiation by penetrating the ground,
whether or not it got dissipated as heat. (In reality, it would of
course eventually get dissipated, since no ground is perfectly
insulating.) Looking just at the amount of radiation that ends up above
the horizon, and neglecting conductive ground current loss, the
reflection "loss" does indeed seem to increase monotonically as the
ground conductivity decreases.

The ground reflection loss can easily be evaluated with any version of
EZNEC, including the demo. Choose a vertical antenna such as example
file Vert1.ez, set the ground type to Real/MININEC, wire loss to zero,
and the plot type to 3D. There should be no resistive loads in the
model. Then look at the Average Gain shown at the bottom of the main
window after doing a plot calculation. The deviation from a value of 1
(0 dB) represents the fraction of the applied power that's lost in the
ground reflection.

Roy Lewallen, W7EL