A truly overunity Transformer / Meg

[gyulasun]

Hi Rossen,

Thanks for the answers.  In the meantime I realised I gave a misunderstandable answer for your question 1 but by then it was not possible to edit my reply for the second time.  So on the appearing flyback pulse I meant it created from the already induced output voltage under load on the output coil and I did not mean a flyback pulse created in the input coil from the input pulse. I should have written it more clearly.  But it turns out from my reply# 75 this morning that I am aware of the good input-output coil isolation from the mutual induction point of view, simply coming from the arrangements of the cores. 

By the way, am I correct in assuming that the flux from the permanent magnets shifts the magnetic operating point from the zero point of the B/H curve in the input core to a (chosen) direction and this way the input power needed for the input core saturation may get reduced by a certain amount because the distance on the curve is less in the PM biased case towards the saturation state than moving from the zero point?  Logic says so for me...  

Cheers,  Gyula

PS: I started to collect some ferrit E and C cores from some junk boxes...

[gyulasun]

One thing I never found a real solution for (yet) is this:
Once you switch-on the input coil so it switches away the field of the magnet to the output, it will return once you turn-off the input coil. Now... it should be possible to somehow recover this returning field of the magnet. Not sure what the best method is. But the returning field should hold considerable potential power. And due to the airgap it is returning on its own, so it is 'free' for the taking. Only thing is how do we take it.

Hi Steven,

If I may chime in with my thought on your question:  in case of the Bulgarian MEG I found the input core was assigned to be always a closed core either a toroidal shape or two rectangular E facing each other, through which the flux equally divided on the two or three legs.   
So whenever the flux goes through fully between (say) the top and bottom parts of the core, induction does occur in the coils wound on the legs but these voltages are with the same polarities in the coils and so their difference is very low. In other words this a needed property we expect from the input coil as isolation (I mean separation) to the rest of the magnetic circuit, actually this deliberate construction greatly helps in separating input from output. 
However, if we were to use further coils on the input core that are not used for saturating the input core at all AND would connect them in the correct phase either parallel or in series then the returning flux of the PM surely could be utilized.  However we also have to think about the separation between these extra coils and the output coils: they may have a mutual effect on each other...  PERHAPS making the extra input coils also resonant by capacitors can help here too. And yes, air gap is also useful here.

rgds,  Gyula

Edit: I am aware of the above assumption of more or less equal induction in case of the E cores legs is a simplification from my part because the central leg usually has got a bigger cross section then any of outer legs but I think this should be taken into consideration at the number of input coil turns on the legs for the correct saturation to happen.

[Getca]

Hello, Gyula...

It seems to me You begin to get the right answers... There are two important things to get - the output circuit and shunt principle of work.
The shunt...We have to get its core saturation and maximum magnetic resistance decreasing using a minimum energy. Think about its working area over B-H curve and the PM flux dependence. Also there is an acumulated energy in the shunt core as the saturation result, but how can we use it?
The output circuit...Think about the flux changes and the voltage induction dependence. Please see the GIF animation...The control core saturates, the flux goes in the output coil, the control pulse is switched off and... You could continue... So, here is the key to minimize the LENZ reaction. Think how does LC-circuit work, what's its voltage and current in the circuit and the phase difference...The resonsnt circuits are used in the switching PS too, but there is an electrical control. However what happens in the MEG?..We have a magnetic control of this LC circuit, because there is only a magnetic flux switching.
I hope my brief thinking would be usefull for You...

Regards,
Rossen

[Kator01]

Hi all,

good question Steven. I have to ponder on this on.

Now please find attached a diagramm of the permeability of pure iron in dependance of flux-density ( Aw/cm = Ampere-Windings / centimeter ). This is an old physics-book ( unfortunately in german ) dated 1940.

The topic :

Permeability in Fe ( pure iron ) is not constant.

@getca :  if you read this diagramm you will find the answer to your output-voltage-scopeshot. The horizontal part of the voltage-curve is the area in the attached diagram where permeabilty-change is strongest ( the steep rise on the left ) Before this steep rise flux-change is quick ( an so is voltage-change at output ) but once you enter the steep part of the  permeabilty-curve  most of the energy ( dB/dt ) is first  absorbed by the alignement of the tiny elemtary-magnets in the Fe-metal-grid. Now at the peak of the curve ( most of the tiny-magnets are aligned) any flux-change again reflect itself in steep rise of output-voltage.

Best Regards

Kator01

[khabe]

Found interesting doc. - have not read it myself yet - seems like critique  ... poor Naudin 
http://www.nuscam.com/pdf/garbage_physics.pdf
regards,
khabe