As a first order approach... use the phased array concept of pattern
multiplication ... Figure the pattern for two isotropic antennas fed in the way
you want, then multiply that pattern by the pattern of one element. This
assumes two identical antennas, of course.. Given that one is probably at the
top of the tower and one is "next to" the tower, this might not be a valid
assumption. You might try a simple model with just a pair of dipoles (so that
you get the polarization and images taken care of).
The other issue to think about is how are you going to do the phasing? You
have to know the mutual impedances of the antenna, and then figure out the
networks (or lengths of transmission line..same thing essentially) to get the
right drive currents. I also suspect that the Z of a reasonably high gain yagi
changes quite a bit over a fairly small frequency range, so the phasing scheme
may not be particularly broad band.
On the other hand, you're probably not looking to steer the beam with
milliradian precision, nor to keep all sidelobes 50 dB down, so a pretty crude
phasing scheme might work fairly well in practice.
One of the big advantages of phased arrays (especially with limited numbers of
elements) is not that they have particularly high gain or narrow beamwidth
(compared to the component elements) but that they can put a null on something
you don't want. You could, for instance, reject signals coming from a higher
elevation angle (i.e. stations closer to you) to favor the DX signal.
For this kind of thing, a simple LC phasing box, a coax "T", and a couple of
DPDT reversing switches might work pretty well. You tune the phasing to
enhance the interfering signal, then flip the phase of one element, which puts
the null on the interferer. (There was something in one of the Antenna
Compendia about this, using a pair of verticals). Sure, the current balance
isn't ideal, etc., but, if it works, who cares whether it matches the
theoretical model.
jljarvis wrote:
>
> Folks,
>
> Anybody got any feedback they'd care to offer on the topic of
> feeding a 2 yagi stack OUT of phase?
>
> Anybody experimented or modelled with less than 180 degree
> phase reversals? i.e. 90, 60, 30...and what happens with
> vertical angle lobe steering?
>
> I'm working up to an experiment preceeded by a model,
> and possibly only a model, if it doesn't look good...
> but thought I'd ask for inputs first.
>
> Tnx.
>
> Jim n2ea
>
> _______________________________________________
> Towertalk mailing list
> Towertalk@contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
|