Re: NEC-LIST: Defining Real Ground Using MININEC

Hi:

I've been having trouble with the MININEC software package. I defined
a very simple case of one wire in meters from (0, -50, 10) to
(0,50,10). The parameters I used are frequency=10Mhz, real ground of
dielectric 13, conductivity 0.002, distance 1E38 and height 0. I
defined a center fed voltage source of 1V and looked at the near
fields in a plane perpendicular to the wire, centered at the source
location. However, when I run this model while spliting the wire into
30, 50, 80, 100 and 500 segments, I found that the near electric field
values don't converge but are in fact diverging. What's more puzzling
is that the phase of the field components at the same coordinate vary
wildly between the different runs. Plus, after running the model with
a perfect ground and running it with a real ground but using
conductivity 3E8, which is almost like a perfect ground, I found that
the electric field results are wildly different, off by a factor of
100 to 1000. I also noticed that these differences were much greater
for near field values at coordinates with the same height as the wire.

Has anyone else encountered a similar problem or might know the solution?
Please help.

Thanks.

Wai-Yin Shum
 
Received on Mon Aug 10 1998 - 10:56:01 EDT