Re: NEC-LIST: Average gain test and diagnosing the source of the problem

Wire grid modelling is certainly tricky.
Really the #1 and #2 error sources are just the modelling rules--segment
length to diameter, etc.
Watch the bends. A big gotcha is a bend with short segments where the
centers of the segments lie in the coincident (overlapping) bend area.
The equal-area rule works ok; my experience has been that box-y models work
ok at about 1/2 the rule. Can't comment on curved surfaces.
I triangulated a working model once and found it to be more effort than the
results justified.
Might I suggest a rules-checker...the ancient "NecDraw" by Bob Just
(available on the NEC archive, or ask me) does a very good job at
checking. I implemented the rules in a Visual Basic implementation of
NecDraw that handles unlimited segmentation.
Remember too that the average gain values are based on the input impedance
of the "antenna"...so the feed area needs to be modelled even more
carefully (and in more detail, if possible) than the rest of the structure.

>"wilfam" <wilfam_at_technet2000.com.au> writes:
>
>problem
>
>
>Hi,
>
>If the average gain test shows that your model has a problem then what is
>the best way to identify the cause of the problem ?
>
>There is some great information available about this test but not many
>examples of "issues" and how they were identified and resolved.
>
>I'm keen to learn.
>
>My particular interest is antennas on wire grid models. I have looked at a
>number of great papers that describe wire grid guidelines.
>
>I have a model (source unknown) that had its square wire-grid
>"triangulated". In an effort to make the model fit the guidelines I set
>about progressively removing the "triangulating wires" in an attempt to be
>more compliant with the guidelines. The only trouble was that the average
>gain test results got progressively worse.
>
>I have a colleague that is building a different model from scratch. As he
>is building the model his gain test results are also getting worse...but he
>is unsure why.
>
>We appear to be missing a piece of the puzzle.
>
>I'd appreciate your help.
>
>Thanks,
>
>Michael.
>
>--
>The NEC-List mailing list
>NEC-List_at_robomod.net
>http://www.robomod.net/mailman/listinfo/nec-list
>
>--
>The NEC-List mailing list
>NEC-List_at_robomod.net
>http://www.robomod.net/mailman/listinfo/nec-list

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