NEC-LIST: Follow-on 1-m cylinder with wire to gnd.

I apologize if there are long lines.

A number of respondents suggested using a wire mesh instead of patches to model the cylinder.
It worked, so I investigated in-between versions. With all wires the resonance peak is at 16.9 MHz,
which seems reasonable as the LC frequency, and is independent of the number of segments
 in the connecting wire. Modeling the side and top of the cylinder with patches
changes the resonance slightly to 18.9 MHz. Adding patches to model the bottom outer 80% of its radius changes the
resonance to 31.7 MHz. Modeling the entire bottom with patches moves the resonance to 47.1 MHz. These
results are all with the 1-m long, .01-m radius connecting wire in 11 segments. Fewer wire segments with the bottom
represented by patches lowers the resonance frequency toward what I think is the correct value of about 17 MHz.

Assuming I haven't done something stupid, the problem seems to be in the connection between the axial wire
and the bottom surface of the cylinder. I am especially surprised that modeling the inner 0.1 m of the bottom with wires
and the outer 0.4 m radius by patches creates a big error, i.e., almost doubling the resonance frequency. It would
be great of someone who understands the mechanics of NEC could interpret this behavior so that we could
know when patches can and cannot be used.

-- 
The NEC-List mailing list
NEC-List_at_robomod.net
http://www.robomod.net/mailman/listinfo/nec-list