Re: NEC-LIST: Q: cylindrcial structure

> Date: Mon, 5 Jan 1998 09:37:04 -0500
> To: nec-list_at_ee.ubc.ca
> From: Chuck Counselman <ccc_at_space.mit.edu>
> Subject: Re: NEC-LIST: Q: cylindrcial structure
> Reply-to: nec-list_at_ee.ubc.ca

> Boon Phing wrote:
> >Hi, I am using NEC2 to generate a cylindrical structure (open ended
> >cylindrical waveguide). I am using wire grid modelling for the base
> >and the surface patch model(SM) for the wall. When the connecting
> >portion ( the base with the wall), do I have to satisfy any condition
> >like the wire must be at the centre of the patch or one of the corner
> >of the patch.

Chuck Counselman responded:
> Having never used surface patches, I can't answer your explicit
> question, but you raise another question, implicitly: Can patches be
> used to model an open structure? I thought that NEC-2 was able to
> model only a closed volume (within which the fields are zero) with
> patches.

Chuck is correct. The surface patches in NEC2 can only be used to
model closed surfaces. Furthermore, my experience has been that the
surface patches in NEC2 do a fine job of modeling scattering problems,
but they do a very poor job of modeling problems with a local source
(like antennas). The point matching technique employed by the patch
portion of the code does not enforce continuity of current between
patches. Wire grid models are generally much more accurate although
choosing the correct grid geometry and wire radius is somewhat of an
art. Neither approach (wire grid or patches) will give you an
accurate representation of the fields near the surface.

---------------------------------------------------------------------------
Dr. Todd Hubing email: thubing_at_ee.umr.edu
Electromagnetic Compatibility Laboratory Phone: (573) 341-6069
University of Missouri-Rolla FAX: (573) 341-4169
Rolla, MO 65401 web: www.emclab.umr.edu
---------------------------------------------------------------------------
Received on Sat Jan 10 1998 - 13:03:09 EST