I guess the circuit cannot drive a top that spins faster than 5 x 2 = 10 ms per rotation, or 6,000 RPM.
I think you're right, assuming the base drive generated magnetic field is much greater than the top's magnetic field.
I did a pendulum setup and got two drive pulses, one long (about 20 ms) and one short (about 5 ms).
I started thinking that the regenerative feedback was going into oscillation due to the slowly changing magnetic field of the pendulum. Now I'm not so sure this is correct. Sorry if I'm getting to far off topic.