(no subject)

I've only been using NEC-2 for about a week. I got it with a slick little
windows shell that has helped me get up to speed (Nec-Win by Paragon).

I've been experimenting with bowties and bi-cones in an effort to get
broadband with simple structures. My question is about the proper use of
transmission lines model.

In one case, I took two equal length wires and made them intersect in the
middle to form a simple bowtie antenna. I broke each wire into 15 segments.
The wires intersect at segment 8. I connected the two wires together at
segment 8 with a zero length transmission line (with Z0 set to 50, although Z0
shouldn't matter). I connected a voltage source across the middle segment of
one of the wires. Ran simulation, patterns and impedance look good according
to my past experience.

Now to fiddle with it. I think a lop-sided bowtie will outperform a
symmetrical one, so I change the length of the two wires so they are somewhat
different from each other. After re-run simulation, I see antenna does perform
better.

Here's the catch. If I move the voltage source over to the other wire, I get
different impedance values. While the real part was within about 5%, the
reactance can vary wildly depending on frequency. I would think that a zero
length T-line can't do any transformation. Or is it more to do with how the
middle segment is excited? If so, would I need to break up into more segments?

I also want to simulate my bowtie with the ends closed (like a UHF antenna for
a television set). I don't know if I should use T-lines to connect the closing
wires or not. The results would be different.

If any of you can help me or point me to some good reference material,
I'll appreciated it a lot.

RC Looke
Received on Fri Nov 24 1995 - 21:32:00 EST