"Jim Lux" wrote in message
...


The patch I'm building uses an air dielectric so I'm using about .47
Lambda on a side.
This site is my guide; http://www.rc-cam.com/gp_patch.htm
I'm feeding with coax thru the backplate with the center conductor to the
Patch. (probe feed)
Since the Patches are 0.47 Lambda, 1.0 Lambda spacing would leave 0.56
Lambda between Patches.
Probably to much. At 0.5 Lambda the patches would almost touch. Probably
to close.
So it's between 0.5 and 1.0 Lamda.
What criteria do I use to choose the spacing?

Are you concerned about grating lobes?

Just looking for the main lobe to have max gain.

How much work do you want to go through to calculate or measure mutual Z,
so your feed network works?

I hope my 50 ohm patches stays 50 ohms, so I guess I they need enough space
so
they don't have much mutual coupling. (I want my cake and eat it to) :-)


Far apart reduces mutual coupling, so you can use a simple feed network
(i.e. power divider and equal length lines), but gives you grating lobes.

You have to calculate the pattern of your array (you could start with the
pattern of an array of isotropic elements, and then just multiply by the
pattern of one patch, which is typically some sort of cos(theta)^n looking
thing..

If far apart gives you an ok pattern, then go with it. It's easier,
although bulkier.


I'm quite confused at this point, I'm going to build one using the spacing
at
http://www.darc.de/distrikte/g/T_ATV...d-Array-GB.pdf
But instead of the Biquad antennas, I'll use the Patches.
I'll build it and see if the come, (the RF waves that is)
Do you think 0.45 Lambda is far enough so that mutual couple won't affect
the input impedance?

Mike