On 5/31/2011 3:40 PM, John S wrote:
On 5/31/2011 5:04 PM, K7ITM wrote:
On May 31, 1:35 pm, John wrote:
On 5/31/2011 1:52 PM, Jeff Liebermann wrote:



On Sat, 28 May 2011 14:25:47 -0700 (PDT),
wrote:

How many amateur radio operators use this kind of academic preening
when they are putting up a dipole.

Me, me, me. Even the simplest antenna is influenced by nearby
structures, towers, poles, elevation, guy wires, position of coax
feed, chain link fences, and grounding system. That makes a simple
dipole not very simple. I've helped a few local hams model their
houses allowing prediction of takeoff angles, mysterious nulls,
optimum height, and cut length. While modeling (I use 4NEC2) does
take some learning and understanding, it does offer an improvment over
the tradition ham radio cut-n-try.

Too much hand-waving here to be useful to most folks.

Speak for yourself please. I like postings that are over my knowledge
level so that I learn something new. It's also nice to know *WHY*
things work, or don't. Learn by Destroying(tm).

I agree, Jeff.

I like antennas that are naturally short-circuited by design and can be
grounded, making the feed point essentially grounded for DC and lower
frequencies. One such antenna is the folded unipole. Its only problem is
that the feedpoint resistance is about 120 or so ohms.

So, I had this idea. The usual monopole (or ground plane) has about
30-35 ohms resistance. To get 50 ohms it is common practice to droop the
radials about 45 degrees. Since that raises the feedpoint resistance,
would raising the radials lower the feedpoint resistance of the folded
unipole and, if so, what effect would it have on the pattern?

EZNEC said to raise the radials of the folded unipole about 23 or so
degrees to get 50 ohms and the pattern would not be affected.

So I built one and it works swimmingly. I had to make some minor
adjustments in element lengths but that was fairly easy with the vector
voltmeter. Hooray for modeling.

Cheers,
John - KD5YI

Hooray also for using your head, John, and realizing that raising the
radials would _probably_ have that effect -- then having that verified
by a model, and then by an antenna that works well for you in
practice.

As a matter of fact, I can see where just jumping into it without the
benefit of modeling would probably have resulted in giving up on it. I
had to adjust many things (such as radial tilt) before I learned about
how things were going to be affected. Modeling is like having an antenna
breadboard but a whole lot less work.

Another way that should work: make the two parallel conductors
different diameters, with the correct spacing. You might also try
making a self-supporting grounded quarter wave, resonant with its
radials, and fed with a parallel conductor that doesn't go all the way
to the top of the quarter wave...

I looked at those (EZNEC) years a go and was never satisfied. I have
learned much more now, so I might benefit from another look.

So there are three different arrangements, perhaps with pretty similar
electrical characteristics, and you can then pick among them for the
one that suits your construction practices the best. Perhaps there
are some more "grounded" monopole designs you throw into the mix.

Yes, if you're "just throwing up a dipole," maybe you don't worry
about things like this, but there are those of us who like to think a
bit deeper about things. I can only hope I remain infinitely tolerant
of those who like to think much deeper than I about many things.

Cheers,
Tom

C'mon, Tom. You're a very knowledgeable person and I value your input.
That's plenty deep.

By the way, I used Walt's inherent balun to make a diamond-shaped
antenna which needs no additional balun. It is 50 ohms at the feedpoint.
It, too is inherently short-circuited and, with the inherent balun,
probably groundable.

It was an interesting exercise that went like this:

1. I like a loop for the inherent short-circuit.
2. It has about 100-120 ohms terminal resistance. I want 50 ohms.
3. It needs a balun. But, I don't really want one.

A folded dipole is about 300 ohms. A half-wave shorted transmission line
is about 0 ohms. So, if you take a shorted half-wave transmission line
and spread the wires apart at the 1/4W point all the way to where it
becomes a folded dipole, it seems to me that the terminal resistance
will go from zero to 300 ohms and 50 ohms is in there somewhere.

I tried it in EZNEC and found that to be the case. I found that, if the
acute angle of the rhombus is about 51.5 degrees, then the terminal
resistance is about 50 ohms (adjust perimeter along with angle to get
50+j0).

Ok, fine. That takes care of everything but the balun. In Walt's
Reflections III, he discusses the half-turn bifilar loop (page 22-10).
But what was intriguing was the inherent balun. Aha!

So, after modeling as well as I knew how, I constructed a rhombus
(diamond-shaped) antenna with the right half of the diamond being coax
(inherent balun) and the left half of the diamond being 14 ga wire.




Are you feeding this with coax? Why not use a big choke at the
feedpoint on the outside of the coax (i.e. a bunch of suitable ferrite
toroids)?

It's not like antenna itself cares which side is connected to shield or
center conductor (as long as no current is being carried on the outside
of the shield?)

It should work basically identically as a scheme with a coax balun, so
you could choose which ever is mechanically or cost-wise more
convenient. (I happen to have a box full of 31 mix toroids for such
things, you might happen to have extra coax or a clever way to support it..)