It's not practical, and often not possible (due to inability to model
proximity effect), to model a loading coil on a turn-by-turn basis.
Where a loading coil is physically long enough to have a current
gradient across it, I've always believed it's adequate to model it as a
wire whose outer diameter is the same as the outer diameter of the coil.
Usually, a single lumped inductive/resistive "load" placed in the center
of that wire is adequate, although if the coil is particularly long, you
might want to split the load into several equal pieces distributed along
the fat wire.

This approach assumes that the current gradient across a coil is the
same as for a wire of the same outside diameter as the coil. I don't,
however, have any measured data to confirm this. It would be a fairly
simple matter to put toroidal current transformers just above and below
a coil and compare their values to what EZNEC or other programs report.
If anyone has done this or is interested in doing it, I'd love to see
the result along with a detailed description of the methodology.

Roy Lewallen, W7EL

Yuri Blanarovich wrote:
Question to Roy,
can EZNEC model loading coils in loaded antennas, like mobile whips including
the physical properties of the coil?
I just had some exchange with W8JI on eHam.net
http://www.eham.net/articles/5998
where he is ignoring the fact that loading coil is part of the radiator and its
physical properties, where length to diameter ratio have impact on the current
distribution in the coil.
He is claiming that modeling programs confirm his statement that the current is
the same at the both ends of the loading coil (close in short coils) . I am
having hard time to see how the modeling program is capturing physical
properties of the coil besides inductance, Q etc.
W9UCW has done considerable amount of work on the subject and has measured that
the current diminishes in the loading coil away from the feedpoint.

Thanks!

Yuri, K3BU