[BC] Engineering Advice

[Richard Fry]

Peter Moncure wrote:
>Binomial feeds are somewhat less efficient for the vertical aperture
>they use, but that cost is more than repaid by the elimination of
>locally induced multipath, as Glen suggests ... They can, for reasons
>not entirely clear to me, also exhibit less variation in axial ratio when
>side-mounted.  (anyone care to comment on that, RF?)
_________________

The reduction in multipath from a binomial array at receive locations near 
the tower results from the lack of nulls in its elevation pattern, at least 
theoretically -- but probably not completely so in a practical installation. 
Conventional power division with multibay arrays produces several elevation 
pattern minima, which are subject to multipath interference from the higher 
fields radiated just outside the minima.  The affect of the tower adds a lot 
of unpredictability to all this.

The very broad main lobe of a binomial array tends to put a lot more r-f 
on/near the ground fairly close to the tower site, which can worsen the 
potential for blanket interference to receivers there.

Although it has been claimed by some FM antenna OEMs, AFAIK it hasn't 
conclusively been proven that reflections of elevation pattern sidelobes 
from surfaces/conductors near the ground around the tower site will cause 
multipath interference dozens of miles from the tower site.  The direct ray 
from the main lobe is much stronger out there than any reflection of its 
sidelobes, because (1) the power radiated in the sidelobes is much lower, 
(2) reflection coefficients usually are not very high, (3) of the improbable 
angles needed for the reflection to reach distant points, and (4) the 
propagation paths for such reflections generally have very poor clearance 
and high loss to distant locations.

Binomial arrays are 1/2-wave spaced, so that may improve the equality of 
h-pol with v-pol, but not necessarily improve the true axial ratio, which is 
the parity of fields at all rotation angles.

RF