Re: NEC-LIST: Image Plane Log Periodic Array

Davor Virkes wrote:
>...If anything seems odd, check for currents in wire mesh. Any
>irregularity counts for mistakes in segmentation, mutual connection
>of segments, or mesh size....

One should always check the current distribution computed for any
antenna model for physical "reasonableness." This check is more
important when the model has more segments and/or more wires. It is
most important for a mesh, because NEC-2 is known to have problems
with small loops. (NEC-4 is better.)

Probably the quickest and easiest possible check is to use the
auxiliary program NECPLT (included in the NEC-2 distribution) to draw
the wire structure with arrowheads showing the magnitudes and
directions of the real and imaginary parts of the current in the
middles of all segments. For very large models I have also used
Microsoft Excel to make a variety of graphs of the current
distribution. [Simply copy the relevant part(s) of the NEC output
file, let Excel parse it into columns, and do a little editing. With
a little practice, this is quick and easy. If you're up to speed
with Excel macros, it's even easier.] An appropriate color, "3-D",
ruled-surface or contour plot is very informative.

A good plot does more than to reveal modeling geometry and/or
numerical errors; it provides physical insight. I'd never thought
much about the current distribution on a ground-plane before seeing a
plot of the segment currents in my 4.5-wl-wide (45 segments wide)
hexagonal ground-plane model (with a turnstile at its center). Then
I realized what should have been obvious, that there is a
standing-wave pattern on the finite ground plane.

-Chuck.

-- 
The NEC-List mailing list <nec-list_at_gweep.ca>
http://www.gweep.ca/mailman/listinfo.cgi/nec-list