[TowerTalk] Stacking Antennas

[jimlux]

Brian Machesney wrote:
> Mention was made of simulating with NEC to see what one would get from a
> particular stack. Assuming the results are not what one wants right out of
> the gate, how can one optimize the antennas to get the "best" or "desired"
> results? I can't find a replacement for Yagi Optimizer which, I understand,
> was used by WA3FET to optimize his OWA designs for stacking.
> 
> How did YO work? Did one simply parameterize the antenna design and then let
> YO plot patterns for many different combinations of parameters? If so, I can
> do that with MultiNEC (also, sadly, no longer available). 4nec2 says it
> includes,"AO-style gradient/hill-climbing optimizer" and "Genetic Algorithms
> based optimizer." Does anyone have experience with this aspect of 4nec2?
> 

In 4nec2, one could set the model up with the distance between the 
antennas or the feedline lengths as variables, and then run the 
optimizer.  The challenge with all optimizers is setting up the goal 
function.  Are you interested in gain at a particular elevation angle? 
F/B ratio, etc.  4nec2 has a fair number of options here, but they might 
not include what you want.

It might come down to manually iterating over the range of some 
parameter and evaluating the pattern plots by eye.