Akula0083 30 Watt Self Running Generator.

[4Tesla]

Yes, I've been thinking about diode effect of leds. So here is updated skeleton schematic so we can see what's going on in the circuit.

Good idea.. I like how you are working on simplifying the schematic so we can see how it works!

[forest]

Unfortunately I think it maybe related to resonance. In such case transformer inductance is also important in relation to circuit operating frequency. If not matched then no ability to self-run. I have such feeling after watching this video https://www.youtube.com/watch?v=afPQtDbn_3s&list=UU_zn0cdzv_qfsxQUm4e7R9w

[MileHigh]

I see that some of you will do a PCB project for this one.  Energia9 said "down to earth measurements" and that is the key.  I will assume most replicators will have oscilloscopes.

You notice that this circuit is a variation on familiar technology.  I am intentionally being very generalized:  There is some kind of oscillator circuit that creates a pulse waveform.  There is a MOSFET or transistor switch pulsing on to energize an inductor, or possibly a transformer core.  The output signal may go through other inductors or decoupling capacitor sections until it drives a load.  The load may be a string of LEDs or something else.

Once you have built it and get over the excitement of making initial input and output measurements and trying different loads, etc., then the real work begins.

You need to discuss how to make a serious power input measurement from the battery source.  They way you normally do it might be good, but it's important to discuss it and have everyone agree that it's the right way to do it.  Then if you have five replicators, and they all agree on how to make the measurement, and they share their data, you have a very good start.  Then you can do the same thing for the output.  You all discuss among each other and agree on how to make the output measurement and all agree that it is a valid measurement.  Since you are all using a real PCB, I am assuming that your setups will be nearly identical, so it should be pretty easy to do.

That leads up to the most challenging part.  It's to figure out how the circuit actually works.  What you want to do is follow the signal path through the schematic with your scope probes.  The key thing to do is simply leave Channel A on the oscillator signal, a.k.a., the "master clock."   I just looked at the Russian schematic and it's pin 10 of the TL494 chip.  Trigger your scope on Channel A and then use Channel B to look at any other signal on the schematic.  Then if you are serious, you will get a pencil and graph paper and you will measure the different voltages at various test points and draw them on your graph paper all synchronized to the master clock.  Or do the same thing digitally.  In theory you can just stitch together scope screen captures and stack them one on top of each other with an image editing program and then save that as a new image.

You can look at the waveform before the pair of transistors, and then at the gate input to the MOSFET.  You can see how the transistors are there to fill/empty the gate capacitance as fast as possible to make the MOSFET switch as fast as possible.  Then you look at the voltages on both sides of each transformer coil and draw them out on your graph paper.  Four trace captures will give you a very good idea of what the transformer is doing, no wild guessing.  You look at the voltage at the top of the series of LEDs and draw that trace on your graph paper.  Do it for as many signals as you want.  You can take a picture of your "14-channel scope capture" with each of the 14-channels properly identified, and with a proper scale shown for the Y axis for each trace.

Then you look at the scope captures and your schematic and figure out how the circuit works.  Is it when the MOSFET switches on that power is pumped into the LEDs?  Or is it when the MOSFET switches off and the magnetic energy stored in the transformer core gets pumped into the LEDs?

What if you change a component value or the number of turns of one of the sides of the transformer?  If you have your initial timing diagram as a reference, after you change something you check out the before/after signals.

So those are some big pointers for you.  From reading the thread I know that many of you will benefit tremendously and learn a lot from this experience.

When you look at the timing diagram, you start to see how action A, results in action B, which leads to C and D, etc., until the LEDs light up.  If you can collectively figure that out, so much the better for the group of replicators and followers.

MileHigh

[a.king21]

The L2 coil is also key.
The L2 coil is OVER the gap in the core.
We know that a core gap creates a DOUBLING of the frequency and also creates
ELECTROSTATIC pulses.
We know that the Cover over that trafo should NOT be short circuited, therefore it does not act primarily as
an electromagnetic shield, but as a electrostatic pickup.
The electrostatic energy is channeled to negative ie a pseudo earth.
Electrostatic induction is COP2 minus system losses.
All within the laws of physics.

[a.king21]

RE RESONANCE AND TUNING.


We know that the gap in the core acts as an additional TUNING mechanism.
Think of it as two plates of a capacitor.
This device therefore can be TUNED TO RESONANCE by manipulating the CORE GAP.