I use FEMM to model loudspeaker motors and coils.
Until now, I modeled coils as a single, non-conductive
cross-section with a current numerically equal to the
number of turns (I model them as non-conductive to
eliminate Hall effects that wouldn't be present in
real coils). In these cases,I believe the Lorentz
force is numerically equal to the ubiquitous speaker
parameter "BL".
To investigate this simplification I created two
models: one with the single cross-section I described
above, and one with the cross-section divided into 50
smaller sections/blocks (to simulate 50 turns). In
the first case there is one material property marker
assigned to a circuit with 5 Amps; in the second case
there are 50 separate property markers, each assigned
to a circuit with 0.1 Amp (5 total Amp-turns in both
cases).
My confusion: I expected similar results, but found
them very different. Moreover, I found that I could
get identical results if I use equal circuit currents
in both models, regardless of how many sections in
which the coil was divided. Is there an error in my
methodology for modeling turns explicitly/implicitly?
I didn't find anything enlightening by searching prior
messages....
Thanks,
Scott