[ Reply | Next | Previous | Up ]
From: Dan Handelsman, N2DT
Date: 16 Apr 2002
Time: 18:36:44
Remote Name: 168.100.201.177
David,
LP pointed me to a reference that may be easily found by amateurs; Bob Haviland's article on Patch antennas in the Antenna Compendium Vol. 5
Haviland has these points to make:
1 - the usual patch is about .5 wl wide and 1 wl long. This is a conducting sheet set over a dielectric material which is interposed between the patch and a much larger groundplane.
2 - The spacing between the metallic planes - the patch and the groundplane - is a function of the dielectric constant of the material between them and is about .01 wl.
3 - of critical importance is where on the perimeter of the patch surface it is fed. If the patch is .5 wl wide and the feedpoint is just brought in from the edge near the middle of a 1 wl side, it sets about the following relationship between the currents along the patch:
4 - the field fringes at the edges of the patch - near the feedpoint and at the opposite edge which is about .5 wl away - and the edges of the plate radiate as a gap or slot antenna. They are in phase and separated by .5 wl. They reinforce each other in the diretion orthogonal to the surface of the plane.
5 - the offset of the feedpoint from the edge and the size of teh perimeter affect the Zin.
6 - bandwidth is very narrow.
So the fringe currents at the edges - at the feedpoint and contralateral to it - result in two radiators in phase.
This would be impossible to model with NEC-2 and difficult for NEC-4 due to the a) narrow separations between plates and b) the dielectric effect if anything other than air is contemplated.