The SOHC4 starter has a field coil and an armature. They are normally wired in series, and both make magnetic fields. The field pole pieces are stationary, and the armature poles reverse polarity as different commutator sections pass under the brushes. The force from N-S attraction and like poles repelling is what induces armature rotation and creates torque.
Yes, the starter can make power, by breaking the series connection internally between field and armature, and using a separate output connection to each brush holder. Then placing a voltage on the field coil(s) to create a magnetic field. Spinning the armature in this magnetic field will induce a voltage in the armature that can be tapped off from the brushes.
I don't know how efficient this would be. There is a brush to commutator phase relationship with the field coil poles pieces that I haven't analyzed. I don't have an unattached starter motor to examine and measure.
Anyway the output would be DC so no rectifier would be needed.
Practical considerations:
The starter drive is normally through a one way clutch. It keeps the starter motor armature from spinning unless power is applied to it in starter configuration. It also keeps the engine from driving the motor armature to speeds where centrifugal force would make the armature winding fly out of capture. With a starter loaded, it spins about 3200 RPM, It drives the crank at a 22 to 1 ratio. If this ratio is maintained by direct coupling, the starter motor would turn 22 times faster than a 10,000RPM crank speed, (220,000). The starter no load speed rating is 11,000 to 20,000 RPM. I doubt it would survive 220,000 RPM.
Brush life was chosen for intermittent duty starting applications. I expect constant armature spin operation would require frequent brush replacement.
Full field strength during starter operation saw current in the 120 AMP range at 12V, so the field wires in the stater are likely to be a pretty heavy gauge.
Theoretically, the armature could deliver 120 AMPs of power with the field at full strength. I don't know what the field resistance is for the starter motor. You would have to select a field coil regulator that could handle the currents demanded by the starter field coil.
I learned from my bike mechanic buddy that the CBX uses a slipper clutch on it's alternator drive to keep from over stressing/over-winding the alternator.
I think I stayed up too late for this post.... Might have to edit it tomorrow...but now...must sleep....
