Kapanadze Cousin - DALLY FREE ENERGY

[NickZ]

  But, they do.  So, it may be that the mosfets don't have to provide for such a high output, and the results shown are caused by the heterodyning of the frequencies, instead, and not from hidden sources as you suggest. So, maybe T-1000 is correct, and that is why he suggested the protection diode between the yoke coils, and the mosfets. However, this mentioned diode is not included in any of the Akula/Ruslan schematics, if it were really needed, especially at the 4000w output loads, it should be shown on the diagrams. Or not?
  I don't see Akula or Ruslan needing to use large car batteries for their tests, or to kick start their devices.

  Hoppy, the batteries, big or little don't have to run the device loads, as they are meant only for short term (seconds, to a minute or two) testing purposes, not to try to run these devices themselves. But, as Gyula has mentioned we should not try to run these tests, at higher than the recommended mosfet wattage ratings.  And so for now I will keep the load at less than 200w. And tune the system to that amount of draw, and also, further tune for the "effect", instead of maximum bulb brightness, or output levels. As Nelson has also suggested doing, as well.

[Hoppy]

  But, they do.  So, it may be that the mosfets don't have to provide for such a high output, and the results shown are caused by the heterodyning of the frequencies, instead, and not from hidden sources as you suggest. So, maybe T-1000 is correct, and that is why he suggested the protection diode between the yoke coils, and the mosfets. However, this mentioned diode is not included in any of the Akula/Ruslan schematics, if it were really needed, especially at the 4000w output loads, it should be shown on the diagrams.
Or not?

Nick,

T-1000 diode suggestion is not going to help the mosfets much to drive a lamp load at 2000W - 4000W, so its fair to assume that if this power level is genuine, then it has little to do with output drive capability of Ruslan's PWM. As to whether Ruslan actually used these additional diodes, the circuit diagrams are of little help, as we don't have any idea how accurate they are.

What 'effect' do you intend to tune for?

[TinselKoala]

I still think there is something wrong with Nick's Mazilli. Either it is not switching at the zero-crossings as it should, or there is too much loading of the body diodes, or both. The scope should reveal whether it is switching properly. And you could try supplementing the body diode with an external diode capable of high currents, like a six amp fast diode.

The reason I am saying this is because I run the Mazilli/Royer driver of my TKoil VII regularly on 24-26 volts, 7-10 amps input, from two SLAs in series, supplemented with a brute-force DC PSU in parallel, and its mosfets (IRFP260n) do not overheat, they are on small heatsinks as I have shown before. I am running at a slightly lower frequency though. My crossover diodes are UF4007s.

Layout, layout, layout! Symmetry, and heavy Drain-Source current paths.

[Dog-One]

I still think there is something wrong with Nick's Mazilli. Either it is not switching at the zero-crossings as it should, or there is too much loading of the body diodes, or both.

Yes, I would have to guess there is overlap--both transistors are on at the same time for some duration.  If it were me, I would try changing the bias.  Work your way up until neither transistor fires, then gently bias back down.  The scope will definitely tell the story.

[lost_bro]

I still think there is something wrong with Nick's Mazilli. Either it is not switching at the zero-crossings as it should, or there is too much loading of the body diodes, or both. The scope should reveal whether it is switching properly. And you could try supplementing the body diode with an external diode capable of high currents, like a six amp fast diode.

Layout, layout, layout! Symmetry, and heavy Drain-Source current paths.

Good day NickZ

I have attached a drawing of one possible solution to the MOSFET overheating issue........ A Schottky diode is best used for the reversed-series blocking diode.
Because it only has to function (blocking capacity) for a very brief period each cycle, even a smaller ie; 40volt Schottky will function in this design when switching over 200volts thru the FETs.
I personally use this solution also in my SSTC H-bridge.  Works very well!!!

Layout, layout, layout!    A Sh*tty layout / design will bring out the worst even from the best components.........

Forgot to mention that the MUR diode parallel to the MOSFET will *need* to be on a heatsink.  You will be surprised how much heat/free wheeling current passes thru this diode when the body diode is effectively blocked and out of service.

take care, peace
lost_bro