Jerry wrote:

Hi Chris

The Lindenblad antenna is fed to produce a null toward zenith. The
Lindenblad antenna as defined by Brown and Woodward in the mid 1940s for TV
transmission, has an omniazimuth radiation pattern.
The DCA has no zenith null.

If you consider an antenna with an overhead null to be the same as an
antenna with no null to be the same, I have no expectation that you and I
will agree.

The DCA offers little advantage over a Quad Helix when radiation pattern
is considered.
The DCA is slightly more sensitive toward the horizon than the Quad Helix.
.
The bandwidth of a DCA is far wider than a Quad helix.
The DCA is very insensitive to dimensional errors when built by an
amateur. The Quad Helix is extreemely demanding of prescission of
construction.

The original subject of this thread was related to building an antenna for
reception of Low Earth Orbiting satellites. I figured the OP could
appreciate knowing that a DCA will perform better than a Lindenblad and
needs no series matching transformors.

Jerry KD6JDJ



Just how good does this antenna have to be. It's not like it's being
used as a probe to measure randomly polarized signals, where AR=1 is
really important.

Quad helix antennas have a reputation for being demanding, but that's
where the performance requirements are demanding. Considering that quad
helix antennas are made by the millions for GPS and by the thousands for
WxSat use on boats, they aren't all that picky, because conventional
mass production tolerances are "good enough".

Relax the performance requirements and the helix is no more or less
difficult than a turnstile or Lindenblad or CP patch. Before the advent
of modern modeling tools, *designing* a quad helix was a huge chore,
especially if you didn't want to use a quad hybrid in the feed network,
but wanted to do the "one a bit long, one a bit short" to get the 90
degrees.

But, returning to the original question, why not a turnstile (crossed
dipoles fed 90 degrees out of phase)? For LEO satellites, you don't
really want a hemispherical pattern anyway. You want something with more
gain at the horizon where the slant range is much greater (thousands of
km) than at zenith (where the range is hundreds of km).

And, for that matter CP is probably not worth worrying about. The loss
from a perfect CP to a perfect linear is 3dB. If you're in a situation
where 3dB is going to kill you, you've got other problems to worry about.

Where CP is really, really nice is when you want to kill multipath from
close by reflectors. Or in deep space applications, where you don't
know the linear orientation of the transmitter/receiver (and you also
ARE worried about eke'ing out the last tenth or hundredth of a dB of
performance)


Jim, w6rmk