At least in part I think I may know how a tiny piece of the TPU works.
I have known for a long time that with the help of the aether it is possible to pull in fields or project them, although there could be many slightly different takes on how to happens but I can tell you that much is possible and it seemingly generates energy.
What is quite new though is that I just looked at the Joseph Newman device that JLN and Stefan Hartman did work on about a decade ago.
http://jnaudin.free.fr/html/NMac0626.htm
They were able to create a negative current that was against the applied voltage (or the 'collapsing emf') by way of making a noisy commutator.
It seems that the coil is pulling in the field from the rotor more strongly as the coils connection breaks.
Such a make and break is of course the kicks SM talks about, they are creating a momentary increase of induction from an external EMF.
A few things are worth noting, it was found that very very very little current flowed in the Newman coil despite having a very high applied voltage, it fact there were no sparks observed at the commutator. So you only need flea power applied.
Another likelihood is that you can get the same effect by having not the coil it's self with the noisy contact, you can probably have the noisy aetherically active coil wound over at a control coil
So a likely experiment would be to create a noisy coil with a breaking contact around a collector (collector could likely be multistrand) and monitor the current out of the collector, see how it picks up a field from a nearby AC coil, and then see how it reacts when the coil wound over it is powered with the noisy contact.
Another note would be that you should ensure that you wind the coil over the collector as in 2 back and forth layers so it has no magnetic inductive effect on the collector coil.
It is likely that many here could try this very easily with TPU's they have already built.
Personally I am probably going to test the effect in a form closer to the Newman motor form first.
Hi aether22
A while ago I considered that SM may have had a small DC brush motor embedded in the TPU as a spike generator. It would also explain the gyro effect and slight vibration. So I did some experiments decoupling the motor from the source voltage with some small RF chokes, then differentiating the pulses with a small cap feeding a nominal resistor load.
The spikes are very fast, sharp, bidirectional and fairly random, although there is some repetition as expected. Fed into a transistor switch, these pulses could create some interesting drive for a TPU.
Possibly, in a TPU, with some positive feedback, the motor would windup and perhaps find the "sweet spot" like a VCO (voltage controlled oscillator)
....V