Re: NEC-LIST: Modelling thick wires in NEC-2

Roland Mueller <rmueller_at_rami.com> wrote:

>I am trying to model a octave bandwidth cylindrical dipole with two
>wires, 12 segments each, that have a segment length to diameter ratio
>(l/d) of 0.6. I have read that for accurate calculations l/d should
>be > 4. What is normally done to get around this problem?
>
>I am thinking about possibly using several parallel thin wires
>positioned on the surface of cylinder to model the wires.

I'm thinking that your thinking is OK. Use the so-called equal-area
rule to set the radius of an individual wire. I.e., set the wire
radius, r, such that 2*pi*r equals the center-to-center spacing
between parallel wires.

I think you'll need to cap the ends of your hollow wire-cage cylinder,
too. I'd do this with a radial array of wires connected to the
cylinder-wall wires at the circumference, and to a common node at the
center. I'd taper the radial wires.

Finally, I'd run a (polygonal) "wire" around the circumference at the
end(s) of the cylinder, because I'd expect charge to concentrate
there.
Received on Wed Aug 19 1998 - 10:28:55 EDT