Inductance calculations

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Antonio Carlos M. de Queiroz by way of Terry Fritz <teslalist-at-qwest-dot-net>" <acmq-at-compuland-dot-com.br>

Hi:

I have implemented in my Inca program some routines for inductance
and mutual inductance calculations using elliptic functions in the
solution. The method is essentially the same outlined in Grover's
book, and exactly as described by Maxwell.
I left all the approximate formulas and numerical solutions using
true spirals available too for comparison.
The approximations that are still incorporated in the calculations
are that the turns are closed circles, the radius of curvature of
the circles is much larger than the radiuses of the wires, and that
the current flows uniformly distributed in the wires. The "exact"
formulas assume integer number of turns (for awhile).

The results for mutual inductances seem perfect, and the results
for inductances of long coils too. For some coils, however, the
inductances differ considerably from the results obtained with
Acmi or Fantc (or even Wheeler). A typical case (the worst that
I have found) is the example of two helical coils in Fantc.
The primary has:
turns= 5;
height= 0.20 m;
radius= 0.50 m;
wire diameter= 0.002 m
Inca obtains 46 uH, while Wheller, Acmi, and Fantc give 38-39 uH.

Inductance calculations by Maxwell's method are quite badly
conditioned numerically, but I don't so great discrepancy with
other geometries.

Could someone verify?
The program is in http://www.coe.ufrj.br/~acmq/programs

Antonio Carlos M. de Queiroz