RE: NEC-LIST: Modelling the RF absorber

Andre, and others,

That thought crossed my mind too... I'll play around with it. I'll let you
know if I find anything interesting or useful. :)

I'm fairly new to computational electromagnetics... so I don't yet have a
good feeling for the real limitations of MoM (but I'm learning, and I'll be
ordering books shortly :-). From what I've read so far on this list, I'm
starting to think that modeling a GTEM with NEC is futile, but it's all a
learning experience.

Is there a theoretical reason why NEC does not do well with entirely
enclosed simulations (like TEM cells)? And by "does not do well", should I
expect entirely bogus results, or do I just need to be very careful about
how I set up the model?

Also, because a GTEM isn't symmetrical, I would assume that the currents on
the top side of the septum are different than those on the bottom side of
the septum. The NEC documentation mentions that the formulation used for
surface patches does not work well for thin bodies. So a wire mesh might be
more appropriate - but NEC assumes uniform current on a wire... so a single
wire mesh won't work either. Perhaps two wire meshes?

-Daryl.

> -----Original Message-----
> From: nec-list-admin_at_gweep.ca [mailto:nec-list-admin_at_gweep.ca]On Behalf
> Of fourie_at_poynting.co.za
> Sent: Tuesday, April 02, 2002 12:28 PM
> To: nec-list_at_gweep.ca
> Subject: RE: NEC-LIST: Modelling the RF absorber
>
>
> Daryl van Vorst asked: "Is there a way to approximate an RF
> absorber in an NEC2
> simulation?"
>
> Daryl,
>
> I have modelled planer lossy surfaces with NEC before using a
> wire grid with each of
> the grid wires resistively loaded. We modelled "teledeltos paper"
> or "space paper"
> which has a few 100 Ohm per square resistance (not actually
> 377/sq like I thought).
> Was very successful. I believe if you extend it to a 3D sort of
> wire mesh representing
> the absorber it should be possible to do using wire MoM. Some of
> the FEM and TLM
> methods are certainly more suitable, but if you are used to NEC
> models and have it
> availabel it may work fine and be good fun. Please let me know if
> you try it and it
> works (or does not). We are still considering the same technique
> to model dielectrics
> (using capacitors in the wire grid or mesh) - have not really
> tried this but should be
> quite useful for modelling small dielectric components (radomes?)
> if it works.
>
> The reference below describes the study and measurements we have done if
> anybody is interested:
>
> A R Clark, A P C Fourie, and J Dresel. Modelling lossy surfaces
> using a wire
> Method-of-Moments code. IEE Proceedings-Microwaves, Antennas and
> Propagation, 147(3):247-248, June 2000.
> Regards
> Andre Fourie
>
>
>
> Tel: Intl + 27 11 262 5155
> Fax: Intl + 27 11 262 5156
> Website: http://www.poynting.co.za
> email: andre.fourie_at_poynting.co.za
> Papermail: Prof. APC Fourie, PO Box 76579, Wendywood, 2144, South Africa
> --
> The NEC-List mailing list <nec-list_at_gweep.ca>
> http://www.gweep.ca/mailman/listinfo.cgi/nec-list
>

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