Just a couple of thoughts from an old broadcaster's perspective, if I may.
The use of radials with an MF or LF vertical antenna has, for several
decades, been more for the purpose of enhancing an earth connection than
replacing it.
A collection of (usually quarter-wavelength) radials can, of course, replace
the earth connection entirely if it is sufficiently remote from earth,
serving then as a ground plane. In this instance, it is also possible to
model such an antenna as a dipole with one half consisting entirely of end
loading, though I don't know if there is any merit in doing so.
As one gets closer to earth, however, the current in each wire of the ground
plane induces counterparts in the earth, and the inductive coupling between
them means earth losses will show up in antenna losses even with no direct
electrical connection. At mediumwave broadcast frequencies, general practice
has been to not try to separate the ground plane currents and induced earth
currents, but to reduce losses as much as posible by placing the copper of
our radials in amongst the earth currents.
To counter a misimpression that sometimes arises when discussing broadcast
antennas, let me note that that approach was _not_ chosen because it results
in the lowest losses; in practice, that seldom turns out to be the case.
However, from an engineering approach, it is more practical to implement than
any of the alternatives, and results in a stable antenna system of
substantially predictable characteristics.
Alternatives that have been used in MF broadcasting include elevated radial
systems (an approach developed using NEC) and counterpoises. The distinction
between them is that an elevated radial system does not attempt to isolate
the radiator from earth entirely, but to reduce the coupling between the
radials and the physical earth enough to minimize earth losses; whereas a
counterpoise system attempts to "capture" all of the displacement current of
the antenna by virtue of being maintained 180 electrical degrees apart from
both the earth and the vertical element simultaneously.
The practical drawback of elevated radials is that both they and the base of
the radiator have to be elevated 3 to 5 meters or more above the earth at
mediumwave broadcast frequencies to show any net benefit. I know of one
elevated radial system here in the southeastern US (a region of terribly low
soil conductivity) which reduced loss resistance in the antenna by one third.
It might have done better but for the difficulty of elevating all parts of
the system any higher.
The practical drawback with counterpoises is the difficulty of establishing
and maintaining the electrical balance between the counterpoise, antenna, and
earth, necessary to minimize the currents flowing in the earth from the
antenna itself. In ideal circumstances, the earth currents are matched by
currents of their own within the counterpoise, not related to the current in
the antenna. If the requisite balance is not achieved, though, the adverse
interactions can cause losses to be worse than for a conventional ground
system.
As a consequence of physical limitations most of us endure with our antenna
systems, efforts to control losses generally follow the pattern established
for MF work: more metal in or above the ground. Results, however, often
differ from the expected. Sometimes more radials enhance performance, but
sometimes they don't. Sometimes connections to water or gas mains help, and
sometimes they hinder. Why?
With the greater depth of skin effect at LF than MF, it appears the
interaction of ground plane currents and earth currents is harder to judge by
rule of thumb. The earth connection most of us LowFERs achieve with radials
at or near the surface is not nearly as close to the effective "virtual
height" of ground as it would be at MF. Thus, even with wires planted firmly
in the soil, we have a system of radials hovering above the effective ground,
with all the attendant losses via the induced currents below, yet
simultaneously connected to that virtual ground by way of an infinite number
of resistors due to the direct electrical contact with the surface.
Looked at in this way, our conventional ground systems aren't ground planes
so much as counterpoise systems, matched (or mismatched) to their environment
at random. That might account for why one person adds ground rods to a
system and improves his signal, whereas someone else does the same thing but
increases his losses. Same laws of physics at work...but different values
for the variables, and hence a different way of having to look at them to
achieve the desired result.
73,
John KD4IDY
|
