As I just posted in another followup, MultiNEC (add-on for EZNec and
other NEC programs) takes care of just this sort of thing for you, and
takes much of the tedium out of the process. As an Excel spreadsheet,
it does require that you have Excel on the computer you're using.

Cheers,
Tom

Reg Edwards wrote:
"JC" wrote
Thanks for help, I think I wrongly explained my problem, here is the
question:
1/ I design an antenna, for instance a 3 el 20m beam.
2/ I enter into EZNEC wires dimensions, spacing, height, source.....
3/ EZNEC calculates gain, F/B, SWR....and results are acceptable.
4/ Now let's suppose my objectives are max F/B as I have a QRM
source
opposed to my favourite transmitting direction and SWR 1.5 on a
given
frequency range as my transceiver is very SWR sensitive and I can't
use an
antenna tuner.
I accept changing wire lengths and spacing but not boom length.
Is there a way to have EZNEC, or another software, doing automatic
iterations until it reaches the best F/B-SWR compromise ?

----------------------------------------------------------------------
--------------

Let us see what would be involved if you had EZNEC and had to do
everything else yourself the hard way. You already have a crude,
satisfactory design for 3 elements, wire lengths, wire diameters,
spacing, height, etc.

Only the boom length and presumably wire diameters and height are
fixed and you wish to optimise everything else for maximum F/B ratio
and minimum SWR.

Everything else comprises : 3 lengths and 1 spacing. This makes a
total of 4 independent variables.

You now vary the first variable over a range of say 4 increments,
keeping all the other variables constant and keeping a record of the 4
results of F/B ratio and SWR

You then vary the second variable over a range of 4 increments,
keeping all the other variables constant and keeping a record of the
results.

You continue to do this until you have done all possible combinations
of the 4 variables. You will have a 4-dimension array of results of
F/B ratio and SWR, making a total of 512 observations.

Now search the observations until you can find the maximum of F/B
ratio combined with minimum of SWR

If it looks as though the minimum SWR or the maximum F/B ratio lies
outside the 4-dimensional array then shift the variables in an
appropriate direction and repeat the whole procedure until a max and
min are found.

The trouble with modelling programs is you have to enter element
lengths and spacing via the keyboard. It would be nice to have a
program to do it for you. I am unfamiliar with the situation. Such a
program might exist - one which outputs F/B ratio and SWR.
----
Reg.