Since others have observed that these pairs of coils work fine with
on off commutation, you have an assumption weighing you down, I think.
The carefully created sine wave might do more movement work if it
was a sawtooth that goes high right after the magnetic field peaks,
then trails off as the magnetic field
moves away from it.  That kind of wave shape would give you a watermelon
seed
squirting leverage for making more motion with same juice.  The duty
cycle as used is probably much less than with an all positive level shifted
sine wave like I think you describe...plus they will overlap and drag
against each other
if active (nonzero) more than 120 degrees.

To think about it:
Break this motor down into three overlapping one coil motors that have
two segment commutators that kick in just after the coils pass the pole
pieces,
and the commutator segments break at 90 degrees, then the motor freewheels
to the other
polarity of the two segments and the cycle repeats.  With three of them,
you will have six driving times to generate a current in one direction or
the other, and to control speed, modulate the cutoff slope and/or square
wave end time.

John G.
Austex

> -----Original Message-----
> Indeed it does but the whole thing (my totem-pole drivers with
the 2N3904 - 2N3906 combo) generates far too much heat. The resistance
through each coil is 4.6 ohms and I am having trouble driving it. My
question is more specific than the theory of operation of the device which I
believe (until convinced otherwise) I understand. I need some information on
the nature of the drive circuit that feeds the coils.