Re: NEC-LIST: Near field (over ground)

In Stan Tarnavskii's data:

  GN2,0,0,0,14.,.0001,
  FR0,1,0,0,10.,

the complex relative permittivity is 14 - j3(10^-3). With this small
a loss tangent the lateral wave is very significant. This is the wave
that goes from the antenna into the ground, travels through the ground
and then emerges at the evaluation point. The oscillations in
amplitude are the interference between the lateral wave and the direct
wave. The Norton approximation, used beyond one wavelength, does not
include the lateral wave, so the oscillations disappear. In most
antenna applications the lateral wave is completely attenuated and not
of interest. It would be of interest if you are trying to use it to
determine the ground parameters, as in some of King and Smith's work,
and then NEC would not give useable results beyond one wavelength.
NEC-4 has the same limitation. The wave in the ground is included as
a first-order asymptotic approximation, and it goes to zero when the
ray grazes the interface. The wave above ground includes surface wave
and higher order asymptotic terms, so it gives a valid result along
the interface.

The real limitation on distance for the near field is when the phase
information is lost in the total distance. That would depend on the
computer precision, but should be greater than 1.e5 wavelengths. It
can be checked by computing near fields at distances increasing by
factors such as 2 or 10, and seeing when the field stops falling off
as 1/R. The evaluation may fail sooner for field over ground due to
the increased complexity of the expressions. In fact, in example 3 in
the NEC-2 and 4 manuals, the ground wave calculation at 100000 m is
near the limit, so results may vary from one computer to another.

Jerry Burke
LLNL
Received on Fri Jun 23 2000 - 21:58:58 EDT