Anyone know how to use square rod in modelling programs?

I'm sure it's a small effect, but I'm interested in the *proper* way to
account for it vs. round rod.

I've used the mean radius of the inscribed and circumscribed circles
and that turned out fine,
but I don't have much ability to make good gain and F/R measurements.

It seems that the corners could affect element coupling...

Should I use the circle that has the same area? Maybe the geometric
mean of the radii of the inscribed and circumscribed circles?

I've got a bunch of 1/4" square rod that I want to make 435MHz yagis
out of, and pretty much all web searches I could think of on square
antenna elements either get zero hits for being too specific or talk
about quads a lot.

The square rod is great mechanically; I can slap a piece of PVC pipe on
a milling machine, mill 1/4" slots through very accurately spaced, and
hold the elements in with cable ties.

Anyway, I'm just going to try the geometric mean and build another
antenna, but I'm interested in any comments on this.

73,
Dan
N3OX
www.n3ox.net