On Thu, 12 May 2011 07:05:36 -0700 (PDT), Michael
wrote:

I suspect the underling problem may be
too many arm chair antenna experts plugging their designs in to
computer antenna modeling software, and not enough people actually
going out in the backyard and building the antenna and comparing it's
performance to a real world dipole for the same frequency.

Hi Michael,

All very true, but be careful of the conclusions you thinks this leads
you to. For instance:

There are
many web pages touting the supposed gain of the Lazy H with EZNEC
plots posted as supporting evidence, but real world performance has
shown that it is consistently out performed by a plain old dipole.

The Lazy H, in professional installations known as Sterba Curtains,
are THE antenna of the big gun Shortwave Broadcasters. Engineers,
stations, and their backers looking at field data do not erect costly
antennas without good judgment. In the past, we have had contributors
here who have designed for these big guns.

Now, as to what you see as "evidence" on the Web in the form of EZNEC
software reports (something I've been using for nigh-on 20 years) has
to be looked at in the details, and the details that are significant:
1. The presence of ground
2. The quality of ground
3. The antenna elevation above ground
4. The presence of conductor loss
5. Transmission line specification
6. Transmission line drive
7. The load Z
8. The system gain (or loss)
at a minimum. That is a lot of detail to hold in the mind while
trying to compare two antennas, much less 3, 4, or 5.

One of the glaring mistakes between the outcome to this and "reality"
(and this undoubtedly conforms to your experience) is that the
placement of the source (transmitter) in software is vastly different
than the placement of the transmitter (source) in physical reality.

For instance, the twist in the line between elements is to enforce a
necessary phase relationship so that you can fee it at the bottom.
Otherwise you can go with a non-twist if you feed it in the middle.
The design varies in particulars there. So what? You can feed either
a twisted connection, or a non-twisted section in the middle; BUT for
the second version, the line back to the transmitter MUST BE dressed
AWAY from the middle of the interconnecting line. Proximity will
confound everything. Proximity for this second choice is almost
guaranteed in a casual installation where an EZNEC "schematic" of the
system offers only a blob called "source" in a spot of convenience
that does not exist in reality.

So, the long and short of it is:
Do not feed in the middle of the interconnecting line.
Observe proper phasing requirements.

You can perform your own EZNEC analysis using the free version. The
benefit of this is you can control all the variables, adjust items 1
through 6 in the list above, and you can replicate your experience.
Doing this will also reveal what you need to do to optimize your
antenna.

73's
Richard Clark, KB7QHC