I guess I'll have to break down, buy a copy of EZ-NEC and play with it:-)

By the way, are you suggesting that 90 degree rotation is all I'd need,
simply because the dipole radiation pattern is so wide, broadside to the
radiator? If that turns out to be true, then the problem becomes vastly
easier to deal with, and could even be done mechanically by simply
switching around the feed line to one dipole (easier to do with ladder
line - the preferred feed line to a dipole - than coax, of course)
although that might not be the BEST way to do it.

Harry

Roy Lewallen wrote:

The two dipoles don't couple to each other, so you have a lot of options
for feed systems. Transmission delay lines are one. Another is lumped
circuit networks -- an L network can be designed to effect a range of
phase shifts.

I think you'll find, though, that you'll hardly be able to tell when you
rotate the array in any increment smaller than 90 degrees, and unless a
signal is pretty close to a null, even that rotation won't make a
striking difference. It's a problem that's easily modeled, so I'd start
by seeing just what I might gain from it before going to the trouble.

Roy Lewallen, W7EL

Harry Gross wrote:

Hi all,

I'm toying with the idea of an 'electronically rotatable' dipole setup.

Consider: a dipole oriented N/S
another dipole oriented E/W
They cross each other at the mid-point of each, forming a big
plus sign, like so:

|
|
----+----
|
|

If you feed the N/S dipole, you get an E/W radiation pattern.
Similarly, if you feed the E/W dipole, you get a N/S radiation patter.

If you feed the N/S and E/W dipoles simultaneously, you get a NW/SE
radiation pattern (for example), and if you flip the feed of either
one of them 180 degrees, then you get a NE/SW radiation pattern.

So far, so good. Now for the fun part. If you feed one of the
dipoles with a signal that is only 90 degrees out of phase (or 270
degrees), you can get radiation patterns that take of in the NNW/SSE
or WNW/ESE directions, and if you feed the other dipole 90 or 270
degrees out of phase, you get NNE/SSW and ENE/WSW directions.

What I am looking for is a circuit that can perform the necessary
phase delays, without having to use miles of coax coiled up in the
shack. Ideally, the circuit would be able to perform the phase shift
on all (or at least several) ham bands, so that it could feed into a
pair of fan-dipoles for operation on multiple bands.

Anyone have any suggestions on just how to accomplish this? In the
extreme case, you could have a continuously adjustable system, that
would permit you to point the 'virtual dipole' in any direction at
all, which might be useful when attempting to null out an interfering
station off to the side of the desired signal. However, in practice,
I suspect that the eight positions mentioned above would be sufficient
to do the same thing (albeit with the possibility of somewhat more
signal loss due to fixed aiming directions).

I seem to recall seeing a quote in the ARRL Handbook (or the ARRL
Antenna Book) from Roy Lewallen, indicating that there is no simple
way to do this. Am I mis-remembering (or mis-interpreting) what I read?

I know that the military does this sort of thing with radar systems,
and has for some time. However, their budget is quite a bit bigger
than mine, so perhaps their solution isn't 'simple':-)

Any thoughts on this, anyone?

Harry Gross