Re: NEC-LIST: Modeling a planar array

If you're looking to point the beam normal to the plane of the array, then
the phases are all the same.
If you're pointing off to the side, then the easiest thing to start with is
to use a phase that is proportional to the displacement in the look
direction:
If x[] is a vector representing the position of the phase center of the
element
and l[] is a unit vector representing the look direction

then
phase shift = -2 * pi/lambda * dotproduct (x[], l[])

You can use other phase shifts (this is what superdirective arrays, do, for
instance)

Something to be aware of though... with elements close together, there will
be significant mutual interaction among the elements, so you'll need to
drive them with current sources to get the pattern to come out right. If you
use voltage sources (or 50 ohm sources), the mutual Z between elements will
result in changes. There will also be certain scan angles/look directions
where the gain drops off.

<BMW9009_at_aol.com> writes:
> Hi all,
> (please bear with me, I am new to NEC and modeling)
>
> I am trying to model a MxN planar array (using Chebyshev and Taylor series
> approx. for each element weighting)
> Lets say that I have the following parameters:
> 10x10 array
> f = 10GHz
> element spacing in x dir = 1/3 wavelength
> element spacing in y dir = 1/2 wavelength.
>
> I have been using a matlab program to calculate the x-dir linear weights
and
> y-dir linear weights (same parameters used except element spacing). Then i
> multiply these weights together so that I get the magnitude of each
element in
> the planar array.
>
> I was wondering how to get the phases of each element though.
> Can someone please help me?
>
> thanks
> -JW
>
> --
> The NEC-List mailing list
> NEC-List_at_robomod.net
> http://www.robomod.net/mailman/listinfo/nec-list

-- 
The NEC-List mailing list
NEC-List_at_robomod.net
http://www.robomod.net/mailman/listinfo/nec-list