NEC-LIST: Simulation of surface structures in NEC2. Microstrip model.

From: Andriy Gordiyenko <agordiyenko_at_email.domain.hidden>
Date: Mon, 05 Mar 2007 20:15:32 -0500

Hello,

 

I'm trying to simulate microstrip structure using NEC2. I tried to use
patches for line and exating with V-source wire for excitation of this line.
But, results show just small influence of patches, presenting line, on wire
currents, which are actually results. Are patches not suitable for such
simulations? I believe that microstrip line simulation is not much different
from antenna simulation, but one need get other parameters from simulation
out. And, if this simulation doesn't show good results, how can one trust to
results of antenna simulation. Another thing, I tried, is simulation of
patch-system using current source at the place of of one of patches ( code 4
in excitation card ). This approach works better, I do could see big
influence of line-shape onto currents in patches and also in wires, but at
some number of patches program starts to give out some words instead of
currents values, may be some calculation mistake. As patches should create
closed volume, I duplicated them and closed the volume too, but this had the
same effect.

Question: Are there some limitations of number of segments in NEC2? Does
someone know a way to model microstrip line with patches which gives out
relevant results?

Another model which does give some results out is wire-model. There is
another problem here with this model. Results depend very much on
presentation of line with wires (wire-mesh). And this is doesn't mean that
if model has more wires, then results are better. It looks like wires should
be at the place of currents in structure (actually it's also one of
simulation purposes, who need such simulation then:-)). One more issue -if
this simulation has so high potential error for simple model, what happen in
complicate structure?

Is there some approach for simulation of surfaces with wiremodels, which
gives out good results? As results I mean S-parameters, mostly.

 

I appreciate any help. Thank you very much.

 

 

Andrey.

-- 
The NEC-List mailing list
NEC-List_at_robomod.net
http://www.robomod.net/mailman/listinfo/nec-list
Received on Tue Mar 06 2007 - 01:15:45 EST

This archive was generated by hypermail 2.2.0 : Sat Oct 02 2010 - 00:10:46 EDT