Shorting coil gives back more power

[MasterPlaster]

The Kunel stuff tends to use the control coil to neutralize (hide) a magnet from the pick up coil(s) and let go. In this case I think it's really just shifting the bloch wall, in a very powerful way.

No it that is not the right thinking. When you bring two similar poles of magnets
you feel the magnetic force. A bit like pushing two baloons together. This is how I see the magnetic fields.

Also, think of a bow and arrow, you pull the bow back to the right amount. too little or too much is no good. The magnetic field is the pressure behind the bow.

So you fire the magnetic field and you let go!

I have a feeling that on collapse of the holding force, the earths magnetic field over compensates and that is the bounty.

[romerouk]

@popolibero
The magnet with the gap must be in repulsion mode. I have tried all Magnacoasters setups but without gain at the level Richard is claiming. I am sure he's got another secret there, maybe coil lenght in relation with the magnets, the coil itself,winding,...
Recently I have tried to short the driving coil and the spike I got burned all components, transistor, hall,... The spike was about 1600v.
Richard has the same problem for long time now, the spike burns the components and is difficult to control. He is using a chopper transformer but even there he has some problems too. Dumping the charge in

Shorting the BEMF is a better approach, timing is very important.
I have redone my circuits and now I am using better components.
One thing to mention, I am not shorting the coil directly, I am doing it with a capacitor, in fact the shorting is adding a capacitor in parallel with the coil.The capacitor value in my case is 0.33uf.
The coil I use is not the old MOT fan anymore, I have made a coil with 500 turns, 0.8mm wire.

[Feynman]

Okay , it sounds like this is promising ,but we really need to put this together as an open-source PDF for this to be viable.  It's hard when the information is scattered around in 10 different places.

Anyway, I'll try to dig more into this, but the basic idea is what.  Solenoid in parallel with 3-4 magnets driven with 20-50khz ... okay, cool... what waveform drives the solenoid?  How does the inverter fit in to the schematic?

To be honest, I feel more confident replicating the Boyce TPU than the Magnacoaster thing, but maybe thats cause I've done more research on it, and programming microcontroller pulses seems to be easier to me somehow.

[giantkiller]

'nuff sed...
http://www.youtube.com/watch?v=_lU2Ijs_8TY

[Feynman]

'nuff sed...
http://www.youtube.com/watch?v=_lU2Ijs_8TY

Thanks GK, that's really interesting.  Putting the magnets and toroid against the television to measure the flux was a good idea.   Still I'm not clear on how to get from this to a practical system...

In the Kunel patents, the systems still use battery banks.   As far as I'm concerned, I want 'solid-state' or bust... no Bedini battery banks.  Just capacitors and invertors etc.

But what you've posted is definitely promising, hopefully you're able to use that information to build a new circuit.