Something to keep in mind is that phased arrays with a large number of
elements are simpler in one respect than ones with just a few elements.
In the first case, you can make the assumption with reasonable accuracy
that the feedpoint impedances of all elements are the same, since all
are in essentially the same environment with respect to the other
elements. For most simple phased arrays, you can't make this assumption.
So while pattern analysis of simple phased arrays is easy once you
assume equal element currents, actually getting those currents can be
more difficult than you might assume. If you're interested only in the
patterns and not how you'd actually get the currents you need, simple
trig is enough. But that's not enough to enable you to actually design
and build one that'll work as planned. Chapter 8 of the _ARRL Antenna
Book_ describes what has to be done to make simple arrays work properly,
as well as describing a few common simple arrays. There are also a few
examples of phased arrays with the free EZNEC demo (http://eznec.com),
where you can save yourself the math and immediately see the effect of
changing element currents.

Roy Lewallen, W7EL

David Harper wrote:

Thanks! You know any simple phased array configurations that are easy
to mathmatically model? I've played with a 4-element array all
located on the same axis just for fun to see what resulted (and to
make sure my equations were right), but I'd like to find some
real-world systems (hopefully without too many elements) to play with.
Any digrams exist on the net? (dimensional relationship between
elements, etc).

Thanks again!
Dave

Below

One method of eliminating* sidelobes from an array is to use 1/2 lambda
element spacing and in-phase, binomial power distribution to the elements.

This has been done in FM broadcast transmit arrays to reduce radiation
levels on the ground immediately adjacent to the tower. Paper 5 at
http://rfry.org shows some of the considerations for its use. Paper 10 at
the same site is a slide show wherein slides 20 and 23 also deal with this.

R. Fry

* in free-space theory, anyway
____________

"David Harper" wrote
I believe you didn't fully understand my question: How do the other side
lobes get reduced and/or eliminated? Theoretically, in some phased
array antenna configurations, some lobes have an undesirably high
a gain. I was wondering what filters / engineering work arounds were
used to mitigate this. Other posts have adequately answered this
question, however. Thanks anyway.

(David Harper) wrote in message . com...
I had a simple question in regards to phased array antenna patterns.
If a phased array is trying to send a narrow beam in a specific
direction, how do the other side lobes get reduced and/or eliminated?
Are the individual antenna transmitters/elements not omnidirectional
themselves? If not, what are the characteristics of their patterns?

I ask this because I'm trying to understand how tracking radars can
send narrow beams in the desired direction without significant
secondary lobes to interfere with returns from the desired lobe.

Thanks in advance for any insight!
Dave

Thanks for all the responses people. I appreciate it!

Dave

On 14 Jun 2004 11:53:37 -0700, (David Harper) wrote:

Thanks! You know any simple phased array configurations
that are easy to mathmatically model?...

In a follow-up to your post Roy, W7EL, suggested modeling
an array with his **FREE** EZNEC demo. I'll second that
motion. EZNEC is well worth learning to use if one is doing
any kind of antenna analysis, even back-of-the-envelope
guesstimates.

Here's what I would do to get a high-gain array with EZNEC:

Place a large number of 1/10-wavelength wires in a row,
end to end, and separated from each other by 1/2 wavelength.
Place a current source in the middle of each, up to the
maximum number allowed by the EZNEC demo version. Set each
current source to 1 amp at zero degrees. Force each wire
to consist of only one segment - this makes the current in
it uniform. Have EZNEC evaluate the pattern in the same
plane as the wires. This is a one-dimensional phased-array.

You can vary the amplitude and phase of the individual
current sources to experiment with beam steering. And you
can change the spacing for the individual elements out to
1 wavelength and beyond to see what that does. It's a lot
of fun to watch what happens.

This should be a relatively easy configuration to write a
far-field expression for if you are so inclined. Math never
was my strong suite, so I'm out of my element and can't give
you much help. (Hey, we're just a bunch of ham-radio operators
here).

Have fun!

Jim, K7JEB