On Mon, 10 Aug 2009 11:44:52 -0700, Roy Lewallen
wrote:

I also did some experiments in the early '70s to see if CP would reduce
fading. I built a couple of types of omnidirectional CP antennas -- a
"skew planar", and a copy of a commercial FM BC antenna, for mobile use
with the local 450 MHz repeater.

A "halo" type of antenna? Some of the commercial broadcast FM
antennas are eliptical polarized. Most of the signal is horizontally
polarized, but there is a small vertical component in order to improve
performance in vehicles.

We tried several antennas at the repeater end. Something like this
one seemed to work best:
http://iris.nyit.edu/~sblank/VPFMfig5.gif
We had 4 elements with a coax cable phasing mess.

I soon discovered that as soon as I
placed the antenna over the top of the car, the polarization became
nearly linear. I've since learned that it's because of the nature of the
reflections from the ground plane, and it's easily seen with EZNEC+.
When I put the antenna far enough away from the car to minimize
reflections, the lowered gain offset any possible advantage. Overall,
they worked out worse than a conventional vertically polarized antenna.

I tried to use CP on both ends and eventually gave up. Thanks for the
explanation, but I have a different theory. The polarization changes
sense (direction) every times it's reflected. We standardized on RH
CP. When the RH CP signal hits the car, it is reflected as LH CP. If
the LH CP signal arrives at the repeater antenna, which is RH
polarized, they cancel. If it became linear, it would theoretically
only present a -3dB polarization loss, which is not huge.

It might have been interesting to try CP at the repeater, but that was
never done.

I can testify that it worked quite well for solving the specific
problem. We were trying to eliminate picket fencing (frequency
selective fading or Rayleigh fading). While there were some half
hearted experiments with various CP mobile antennas, the major effort
was at the repeater end. This was about 1971 so the technology used
was rather crude. One student was doing his senior project (reqd for
graduation) around this test. Several of us were enlisted to help.
When was in the land mobile radio biz many years later, I repeated the
tests with similar results.

We hung a thermal chart recorder onto the first limiter testpoint (on
a Motorola Sensicon T43 receiver) and plotted signal strength versus
time as a mobile drove through the problem area with the xmitter keyed
continuously. The test was repeated with various tower mounted
antennas. The linear antenna had more signal (gain) than CP, but also
had many more fades, what were far more pronounced. In all, I would
call it an improvement in quality, but not in range.

The problem with ground reflection ruining the circularity makes it very
difficult to achieve circular polarization for HF skip communication.

Well, I supplied several examples of commercial HF antennas that are
circularly polarized. I'm tempted to try building one, just to see
what works or breaks.

A second problem is that the majority of CP antennas, such as the
quadrature fed crossed dipole "turnstile", are circular only directly
broadside, and increasingly elliptical as you move away from that direction.

That's why high accuracy GPS antennas use choke rings at the antenna.
It widens the pattern so that it picks up more of the sky, but also
maintains some semblence of CP at the horizon.

Any interest in me scanning and posting the chapter on circular
polarization repeater antennas from the TAB book? 13 pages with some
low quality pictures.

Roy Lewallen, W7EL

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558