Almost forgot.
A schematics for the latest Akula's circuit has just surfaced on a Russian forum.
Here it is.
~A

One more thing. Both chips are TL494.

~A

Whatever, dude.

Oops, I forgot to invert the lower channel. "Correct" now:

 

3.0 V,  50 milliAmps input power, brilliant blue and white LEDs.

 
 

its inductance in cups should be 115 uH, cups on a copper foil glued to it connect oscilloscope .

A question, if I may...

Why 115uH?
Also, when you glue a copper strip to the pot do you, actually glue 2 strips for the lower and the upper part of the pot or just one to wrap both?

~A

my English is bad but I'll try to explain that you have no chance to tune Akula's  device if you don't know how it works. I hope Wesley  will  translate latest Akula's conversation on skype. but in short, Akula's devices is working on core resonance. It's different cores - different resonances. for example one Akula's core have 3,5 Ghz resonance. transistors switching time need to be shorter than this frequency. the high and low frequency generators need to be harmonics oscillators on  cores resonance frequency.
from start you need to find  core resonance that you have.
the transistors switches need to be switch on-off in one time
and there is more what to be explained but I hope somebody do better than me

Hello All:
I believe this to be relevant concerning ferroresonance and dimensional resonance:
and in relation to AC power systems:

"Ferro-resonance can generate overvoltages of upto 12 times line to ground source voltage upon opening a single phase device or poorly synchronised 3 phase device. Voilent failure can occur exposing personnel to the high voltage failure and accompanying conditions. Ferro-resonance conditions can result in damage to lightning arresters, switching devices, buried cable, transformers and associated equipment.
Ferro-resonance can be initiated when all of the following elements are present and the switching means at the dip point or takes off when a single phase device or an unsynchronised 3 phase device that does not operate all phases within 1/2 cycle.
1.System grounded at the source, but with no ground at the transformer bank
2. Shielded cable length or overhead conductor length sufficient to create the capacitance necessary
3. Transformer size that permits saturation of the iron core at the operating voltage
4. Transformer unloaded or lightly loaded.
Prevention or control of ferro-resonance can be by:
1. Using wye-wye transformer withboth neutrals grounded and tied to systems neutral
2. Using only phase-to-neutral (not phase-to-phase) transformers connections for single pahse transformers.
3. Limiting length of underground cable
4. For poorly synchronised switching devices, extra resistive load of 2% of the transformer capacity on underground cable. "
Ref:www.tpub.com/doeelecsafety/electri... safety

Food for thought, and it does correlate with the Russian stories of the cores self-destructing after a short runtime:

take care, peace
lost_bro

Well done TK, so what about the Secondary coil ?
What voltage and current can you get from it ?
Do you feed this spikey signal to the primary of the flyback transformer
and is the pure sine wave then at the secondary ?
Can you bridge rectify the secondary and feed it back to the input then ?

Many thanks.

Regards, Stefan.

The circuit is wired exactly as Akula's schematic below, except that I added a 2.2 pF ceramic capacitor across the 3.3K resistor.  (And my big input reservoir cap is 3300 uF not 2200 uF.) I have the smaller of my inductances (about 6 milliH) connected to the 34063 side and the larger (about 20 milliH) connected to the 4069 side. I am using BUZ11A mosfet, CD4069CN from National Semiconductor, MC34063API from ON Semi. One blue and one white LED.

The system is very sensitive to input voltage, potentiometer setting and coil ferrite gap (thanks for suggesting I look at varying this gap). Also coil phasing, we are all used to swapping connections of one coil if it doesn't "work" the first time, right?

The spike voltage on the top trace is set by the pot and the DC_DC chip's other components, but the range of "tuning" to see this waveform is so narrow that the spike voltage does not vary in that range. It maxes out at about 2.5 times input voltage. Akula's "80 v p-p" here must be from setting his probe/scope attenuation incorrectly, as I am seeing exactly 1/10 that value at maximum.

The sinus waveform can be gotten to make a little more voltage than is shown on the scope but then the top trace changes as well, by reducing the length of the bursts of oscillation, and if the pot is turned too far in either direction the sinus flattens out and no bursts happen.
I don't think it can be connected differently or looped back in the way you are thinking.

I think that the inductances I'm using are too large, which accounts for the slower main frequency I'm getting. I think this, because when I separate the core halves, the frequency of the sinusoid goes up and up, until the cores are too far apart and the thing stops working.

The ferrite I am using is from a TV flyback transformer. It is conductive, measuring almost exactly 50 ohms per centimeter on the surface, contacted with point probes. The coil bobbin inner tube is a brass tube and it is in contact with the ferrite and connected to the inner winding as well.

But it still goes straight to DC and fades out in a couple seconds when I pull the input power  plug.

 

Я хочу повторить то, что вы сделали, и не против, как это работает, но если вы думаете, что мы не должны быть в состоянии простого копирования, то я понимаю, но зачем создавать головоломки?
Пожалуйста отправлено феррита чашку Magpwr, он был одним из первых, кто хотел повторить из вашего видео.

спасибо

I wish to replicate what you did and don't mind how it works but if you think we should not be able to simple copy then i understand but why create puzzle?
Please sent ferrite cup to Magpwr, he was one of the first who wanted to replicate from your video.

Thanks

hi scratchrobot,

I might be the first person to be interested in the device but i am definitely a lousy presenter through youtube video.Hence i am not the right candidate maybe TinselKoala would do.

Do reveal to us if your replication is a success since you have the exact same Russian made ferrite pot core.

I have read this somewhere in internet it looks some of the old soviet made ferrite core are irradiated as part of the manufacturing process."Sorry i am unable to validate if this is true".

I am just thinking if this irradiation process do play a important "OU" role from the particular Russian Yoke to this recent ferrite pot core.

I have recently received the MC34063 but not the ferrite core from ebay yet.

Just for fun here's another scopeshot, at the same parameters as above.

The shot below is as described above, with a 1.7 mm thick wooden spacer between the two halves of the flyback core outer leg. Inner leg is probably still touching inside the bobbin. 3.0 V and 50 mA input.

There is something wrong here somewhere.   

The sinus I am seeing seems to be at the local powerline frequency. I know it is coming from the coil and mosfet switching though, because I can feel and hear the coil vibrating.

I set up the measurements to read the p-p voltage and frequency of the sinus. The two shots below are first, with the core and spacer and spring clamp installed, and second, with the core unclamped and no spacer, chattering away, and retuned for a stable waveform set. Current draw went up to over 300 mA.

OK, I'm done with this coil I think, and I'll be winding another one shortly. Akula couldn't be picking up a harmonic of his powerline frequency, could he?