Re: NEC-LIST:MSAs in NEC

I have been toying with the idea of modeling a dielectric object with a mesh
of wires with imaginary conductance.

I figure that the conductance needed to model a particular dielectric constant
and loss factor will depend on the meshing density and the frequency of
operation but it seems like it should work

About the only thing that might keep you from this technique would be the
computational expense (lots of segments for the dielectric volume)

Any comments on this idea?... or maybe a reference

Wayne S

----- Original Message -----
From: "McCabe, Brian L" <BMccabe_at_SIKORSKY.COM>
To: <nec-list_at_gweep.ca>
Sent: Monday, April 22, 2002 3:36 PM
Subject: RE: NEC-LIST:MSAs in NEC

> The short answer is that microstrip antennas can't be simulated in NEC or
> its variants. There's no way to model the dielectric slab.
>
> NEC is based on the free-space Green's function, which means that the
> physical problem must be expressed as an unknown surface current
> distribution radiating in free space. The user models all 'matter' in the
> problem as discrete wire segments.
>
> NEC allows you to add a dielectric coating to wires (I haven't used that
> feature) but that can't be used to model a chunk of dielectric material.
>
> I'm not familiar with Ouray, but there are a number of commercial moment
> method codes for microstrip antennas. They're based on "spectral domain
> Green's functions" for multilayer environments, which are derived by a
> process similar to transmission-line analysis. These moment method codes
> are more accurate than the cavity approximation.
>
> Methods like finite difference time domain and finite elements can also be
> applied to microstrip geometries.
>
> Depending on your needs and expertise, you could even write your own moment
> method code for a simple microstrip dipole or patch. There are many papers
> by D. Pozar and his group at UMass (N. K. Das and D. M. Pozar, "A
> Generalized Spectral Domain Green's Function for Multilayer Dielectric
> Substrates with Application to Multilayer Transmission Lines," IEEE Trans.
> on Microwave Theory and Techniques, Vol. MTT-35, pp. 326-335, March 1987 is
> a reasonable place to start.)
>
> Hope this helps,
>
> Brian
>
> -----Original Message-----
> From: Graham Kinch [mailto:graham_kinch_at_hotmail.com]
> Sent: Friday, April 12, 2002 7:17 AM
> To: nec-list_at_gweep.ca
> Subject: NEC-LIST:MSAs in NEC
>
>
> I know I'm over looking the obvious here, but if someone could point out
> the obvious to me it would be much appreciated!I have simulated a 435MHz
> microstrip antenna in 'Ouray Microstrip Antenna Designer' software. This
> software is based on the Cavity Model. When I try and simulate the
> dimensions suggested in Ouray in NEC software it simply doesn't work, I
> get very strange results such as positive return loss! The software I'm
> using is Win NEC Pro and Super NEC, both NEC2 based programs. I'm also
> trying to simulate a plate antenna in both and getting very similar
> results. The solid plates in both the microstrip and the plate antennas
> are simulated in both pieces of software as wire meshes, with the wire
> spacing of the mesh less than 1/10th of a wavelength. To the software
> this appears to be solid. The problem is that in both antennas the
> distance between my ground plane and the plate is much less than 1/10th
> of a wavelength, approximately 10mm. Could the problem be that the NEC
> software assumes that the ground plane and top plate are one solid
> structure with no gap between them? Can microstrip antennas even be
> simulated in NEC?Any suggestions would be much appreciated. Graham
> KinchDublin
> Institute of Technology.
>
> ------------------------------------------------------------------------
>
> Chat with friends online, try MSN Messenger: Click Here
> --
> 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
>

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