3v OU Flashlight

[TinselKoala]

Yes, I can agree with that, the bursts load the supply causing the ripple sine. Either that or the sine is superimposed on C5 by the output of W2.

  But lets get to the meat of the question from the first line i.e. the rationale for using a scratched out portion of a schematic from the video. Also, what is the proposed utility of the FET and load resistor R3?

I don't think I'm using the "scratched out" or rather greyed out portion of a schematic. Please see below.
The FET and the R3 appear in the schematic that is purportedly the schematic used to make the scopeshot and the "perpetual" shining LEDs. Beyond that you will have to ask Akula what they might be for.

Note also that in the diagram in question C5 effectively shorts the output of winding #2 at high frequencies, creating quite a stress on the driver transistor in the chip and further reducing efficiency. It is a good thing there is current limiting in that 34063 chip.

Yes I see that. That's one reason why you can't use very small inductances like the microHenry range. The input current goes way up and the chip heats significantly, even on the 3.0 volt supply.

Would anyone agree that most of the other Akula designs utilize the mixing of two frequencies in the transformer, or at least a variable bias to the core in one of the windings?

Of course I understand that there is no obligation for you or anyone to answer, I am presenting these questions respectfully as they have troubled me.

Yes, that's what it looks like to me. That is the Red Herring in these designs. They all (when corrected properly) "work" to produce the waveforms and measurements (under power) just fine, and the "theory" of ferroresonance, beat notes, phase reinforcement, resonant pumping, all of that is the saturated color of the Red Herring. You chase after the Red Herring, thinking it will lead you to the Bright Shiny Objects.... but in reality you are just another fish and if you aren't careful you'll get eaten by the sharks.

Even in this design, as you can see in the original Vadik Guk version above, there are two oscillators meant to be beating against each other. (Some stages of the inverter are used as oscillator, and in the "scratched out" portion of the diagram below, the same is happening.)

[Vortex1]

Thank you for your thoughts on this. Attached below are the two schematics I have on file, one by Vadik Guk, another variation by verpies.

In the Wesley video schematic, around the scratched out 4069, I see two oscillators, but these seem to be stand alone LED pulsers, thus not very influential to the core except for loading effects of the supply. The remaining inverter (pins 12,13) appears to be sensing voltage on W2 and acting as a driver for the FET. Again this FET and resistor seem to be directly shunting the supply and does not seem to have a useful purpose, unless it's purpose is to provide a variable loading of the rails based on W2 output.

[verpies]

This schematic was an earlier work in progress based on magpwr's reverse-engineering work done directly from Vadik Guk's video.  This schematic is not accurate.

The other schematic is a more recent one drawn by me after Semenihin-77 posted his schematic of the same circuit here.  This schematic is accurate if Semenihin-77 posted a good info.

[avalon]

Ok,  same circuit (TKLABS V1.4) with same components, now with much higher inductance on the E-core transformer.
I have one side 600mH, other side 100mH 

After some testing i found that the higher inductance coil should go to the inverter (W2)
This way the 1:6 ratio ensures an adequate input pulse for this inverter to produce a good output (gate) signal.

I get the following screenshot (to get the MOSFET activated i needed to raise the supply voltage to 4V (@ 100mA).

All signals compared to ground

CH1 (yellow) top of D1 diode
CH2 (blue)   pin 13
CH3 (purple) gate MOSFET (pin 12)
CH4 (green)  Drain MOSFET

Regards Itsu

This looks just spot on.
As a matter of interest, have you checked the native resonance frequency for the core?

~A

[itsu]

This looks just spot on.
As a matter of interest, have you checked the native resonance frequency for the core?

~A

no, i have not checked that, you mean like this E-core picture?


@All,

I also was wondering what this MOSFET/transistors function was in this circuit, anyway  i have swapped the MOSFET for a transistor (MJE13009) and now i can get a similar screenshot as the earlier one, but now with a supply voltage of 2.3V (@50mA) see screenshot below.

Still with this high inductance (100mH/600mH) in the transformer.

Same setup as earlier screenshot, but purple is now on the base, green the collector.

Regards Itsu