Re: NEC-LIST: NEC 4.1 Anomaly

At 03:15 PM 3/21/2006, Davor Virkes wrote:
>Dear Frank,
>at first I ignored your post as mere NEC4 bragging, just like "look Ma, no
>hands" thing, but on the other look, there are some details I noted on
>NEC2 before. (please note that I have no access to NEC4 so far) I've noted
>the same anomaly when rotating meshes using MSexcel. Too often some
>segments get detached due to rounding, and mutual distances seem
>negligible. Segment currents are the only method of verification.
>There is another one I found funny, and often useful: zigzagging wires
>with odd numbers of segments, in a way that there is no nod in a middle
>acts as the wires are mutually insulated - great tool for simplifying
>crossed loops.

I forget the exact number, but both NECs actually determine the
connectivity by comparing the X,Y,Z positions of the segment endpoints, and
if it's small enough, they're deemed connected.

One needs to be careful when using some form of preprocessing to translate
model files (e.g. Excel) for just this reason. An arbitrary list of
wires/segments doesn't carry any information about connectivity. On the
other hand, if you regenerate the NEC model from some abstract model on the
fly, you can make sure that segments that are supposed to be connected stay
that way. The most problem is when you are doing something that you've
gridded, and hoping that the segment ends line up for both directions of
the grid.

I've heard it suggested that one can explicitly generate all the segments
(that is, all GW have just one segment). Makes for a huge model file, and
slows down the read in time, but since the underlying model is identical,
the modeling time should be identical.

>I've lost some edge with NEC as I did not run a single simulation for over
>a year now, but I am still waiting for NEC4 to become available. Any news?

I would imagine it will stay export controlled for a while. Have you
checked the Lawrence Livermore site?
http://www.llnl.gov/IPandC/technology/software/softwaretitles/nec.php

The screening form for non-US customers seems pretty straightforward, and
unless you're in the N,B,C weapons business, you probably qualify to get a
copy.

>Davo

James Lux, P.E.
Spacecraft Radio Frequency Subsystems Group
Flight Communications Systems Section
Jet Propulsion Laboratory, Mail Stop 161-213
4800 Oak Grove Drive
Pasadena CA 91109
tel: (818)354-2075
fax: (818)393-6875

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