N0GW wrote:
. . .
I wonder if Roy has a simple way to include the coil Q or losses in the
EZNEC model. I'm sure there is a logical way to work it out but I
haven't had time to think it through yet.

Sure. If modeling it as a helix, simply specify wire loss for the type
of wire or plating on the wire used for the coil. (The loss in the coil
will be much greater than the loss in the straight wire, so it's no big
deal if the straight wire is made from some other material.) If modeling
it as lumped loads (not as accurate), include the appropriate value of R
in the loads. You can quickly and easily see how much loss the coil is
causing by comparing gain with the loss present and absent.

A few experiments showed a pretty close agreement between EZNEC's helix
model with wire loss and Reg's inductor program with regard to L,
resonant frequency, and Q. I didn't, however, run tests with a wide
range of coil geometries and wire sizes.

The EZNEC helix model will underreport loss if wire spacing is less than
one or two wire diameters, since it doesn't account for proximity effect
(uneven current density around the wire).

Although you can manually (and tediously) build a helix model with EZNEC
v. 3.0 or EZNEC-ARRL which is v. 3.0, or create one with an external
program and import it, EZNEC v. 4.0 has a built-in helix creation
feature that generates a helical coil with a few keystrokes.

Roy Lewallen, W7EL