Akula0083 30 Watt Self Running Generator.

[TinselKoala]

Once the 494 is oscillating and driving the mosfet, there will be spikes generated whenever the L1 coil is turned on and off. The two MBR3545 diodes charge up the capacitor C3 and the second one helps keep the charge on C11 (I think). The LEDs and the inductors, the mosfet and capacitors C3 and C11 then keep the LED bank on while the circuit is operating with "start" button depressed, and after the "start" button is released they can keep the LEDs lit for a while, I think. The output LED side of the circuit, including the inductors, looks quite a lot like a JT with a wireless receiver (essentially the second MBR diode) attached.

I think. Maybe.

[TinselKoala]

Dammit. I actually have all components necessary to build the silly thing on hand, except my 494s are the CN type. Now see what you have done? I don't have room for all this (possible)  HV stuff in my lab!!

Well, almost. The high current Schottky diodes.... why are these needed? There certainly aren't going to be sustained currents of the magnitude that those can handle. 35 Amps!  Why would a 45 v, 3 amp Schottky not work?

[TinselKoala]

Ah. I see that GL's excellent PCB board uses the smaller 15SQ45 diode, a 15 amp unit. Some other component subs also. The "fact" that the 3205 mosfet needs a large heatsink is telling you something significant: It is dealing with HV spikes that are avalanching it continuously, and/or it is being underdriven or too fast for its rating. This mosfet has a very low intrinsic Rdss and runs cool in every application I have used it for.

[elementSix]

 

[elementSix]

I would Like to point out that we must be very careful about the version of this Circuit that we work from.  I believe that the one I posted earlier is the original.  The one that is on the first page is XNAME41's altered version which he says is the one he used to do his replication.  Which is likely that Akula changed one little part to make us work for it, instead of getting a hand out.  But the ones I have posted below are both in very poor quality, which means that they have been changed and altered to make them hard to read and also they changed most values on resistors and capacitors so that we could not replicate this setup.  It could be a precaution in case it is a real working circuit or that they know for a fact that it works.  Anyways I have 5 different versions on my cpu.  I am posting the 2 I believe are total fakes, just because of their quality and the windings on the L1 are wound the same direction, unlike the picture that Akula put on them.  Which shows the Cw and CCW windings.  Its hard to read them, but you can see that they are totally different.  So we should probably work from Grumages, post/ XNAME41's version or the one I posted earlier with the questions.  I think that is the original.  So check the Russian page to see which one Akula posted.

Good Luck to us all.