[TowerTalk] Funniest thing I've seen in weeks

[Jim Lux]

At 04:08 PM 6/30/2004 -0400, Joe Subich, K4IK wrote:

> > From: Jim Lux
> >
> > You're right, though, they all reradiate because of mutual
> > coupling, and there's no particular reason why you couldn't
> > for instance, make an antenna  with 3 reflectors and a driven
> > element.
>
>Other than the empirical (and some computer) studies that seem
>to show that there just isn't any return on the investment in
>additional reflectors.  For some reason (and I'm not an expert
>in EM) there is so little coupling from additional reflectors
>after the first that they do not provide a real advantage.
Have you any references on this phenomenon that I could look for?  From an 
EM standpoint, the coupling from one element to another is essentially 
independent of whether or not there's another element in between (otherwise 
there would be no advantage for multiple directors).  I have a hard time 
believing there's any inherent electromagnetic reason for preferring 
directors over reflectors.  Maybe it has to do with element losses or 
something?


I just ran a couple quick models with a 3 element 20m yagi, putting the 
feed point on each of the three elements (keeping element lengths and 
spacings, hence the mutual coupling,the same).  The general shape of the 
pattern was identical, with the gain changing slightly (0.5 dB out of 
12dBi), probably attributable to the less ideal current distribution when 
feeding the front or back element as opposed to the middle.

I should think that if you optimized the antenna, there wouldn't be a 
significant difference between feeding at the back or the front.  Feeding 
in the middle is always going to be better.  In fact, doing a quick 
optimization, I get F/B ratios within 0.5 dB and gains within 0.5dB for 
feeding either the front or the back element

One aspect is that the feedpoint impedance may be "nicer" with one 
reflector and many directors than with one director and many reflectors, 
although I have no basis for the surmise.

>The fall-off in incremental performance is even more (expo-
>nentially) after a second reflector.  On the other hand, the
>performance increase per additional director seems to be fairly
>linear with the percentage increase in boom length.

This would be true for ALL arrays.. the performance is proportional to the 
aperture.  Yagi-Udas just happen to be end fire arrays where the aperture 
is in the same direction as the preferred propagation direction.  There are 
some subtleties in superdirective arrays, but, as you say, directivity/gain 
is proportional to length to the first order.


>73,
>
>    ... Joe, K4IK
>