Though I haven't actually done this, the way I'd approach it is to
determine the desired resonant frequency of that element, standing
alone. It's not so much the length of the element as the element's
reactance at the operating frequency that's important, and if you make
an element the same diameter and length as in the simulation, and trim
the coil so that your physical element resonates at the same frequency
as the one in the model, you should be OK. That's my reasoned
approach, but for sure listen to those with experience before taking
my advice on it. You should be able to determine the resonant
frequency by modelling just that element separately, with a source
somewhere near the center.

When you posted earlier about your shrunken design, I had the mental
image of a loading coil in each side of the element, and the other
poster's comment about using a shorted stub shorter than 1/4 wave to
get the same inductive reactance gave me the idea for what should be a
pretty rugged, and easy to build, way of doing it...the center of the
element would be tubing, and the ends would be rod. The rod would go
into the tubing, forming the coax stub on each side. They could screw
in, so you could store the antenna in a very small space. I don't
recall ever seeing such a design, but it should work, and avoids the
trouble of coming up with a good mechanical arrangement for the
loading coils. Same thing could be done for shortened-element HF
beams, too.

Cheers,
Tom

"George" wrote in message hlink.net...
I have a question that's probably easy for anyone with practical experience
building UHF antennas. I need to make a shortened dipole (from tubing) at
450 MHz without sacrificing too much gain compared to full-size.

I have a design based on EZNEC. Total shortened "half-wave" dipole length
is 0.32 lambda. Half of that of course is in each of the two elements.
There is a loading coil in the middle of each element. EZNEC assumes zero
length, non-radiating coils. In reality, the coil length is significant
compared to the element length.

So, how do I build this thing considering the real coil length? Can the
coil be inserted in the middle of each element without reducing the element
length by the length of the coil? In other words, would final element
length be 0.32 lambda plus coil length? Or, should the elements be
shortened to maintain overall 0.32 lambda end-to-end length?

Thanks.

George K6GW