3v OU Flashlight

[tysb3]

@ Pirate88179The problem is that the claimant don"t want to disclose all essential important details. and if you don't believe him you are wasting the time with no reason.

It's better reason to show, how you can fake this device.

[TinselKoala]

Here are some more facts I've gathered.

1. The TIP3055 works, whereas the mosfet I tried did not. That is, the bipolar transistor is switching fully when it gets the appropriate output voltage from the inverter chip, and it pulls down the positive rail when it does, and I get a good signal, although not a pure sinus or at 270 Hz, from the testpoint, at certain very precise settings of the trimpot.

2. The signal from the anode end of the Schottky diode I am seeing is very similar to the Akula signal, but at about 1/10 the claimed amplitude, and still without the regular response to the amplitude of the sinus. I get 7-8 volts maximum amplitude from the chip here. Have we determined for certain that Akula's probe for this channel is incorrectly set?

3. I tried a 34063 chip from another manufacturer and got somewhat different waveform results. The first chip is marked "sum" which is a mark I don't recognize (the chip from the dollar store USB adapter.) The chips I ordered from Ebay arrived today, and they are marked "ON" which is ON Semiconductor and corresponds to the data sheet attached below, for 34063AP1. I think I like the ON Semi chip best.

4. The system is also very sensitive to supply voltage and current. The first PSU I used was my small HP721A which has 225 mA current limiting. But I've found that at 3 volts supply the circuit wants to draw as much as 270 mA, so some of my early results are contaminated by the PSU current limiting function rolling back the voltage. With a stronger regulated supply I get, again, somewhat different performance at 3 volts input.

5. No matter what I try so far, in terms of coils, phasing, pot settings, PSU settings, whatever... as soon as I unplug the power the system reverts to straight DC to the LEDs from the capacitors, and it fades out over a few seconds.

[tysb3]

 @Tk
I think not only in Russia secret military lab but in USA secret military lab knows  more than Akula.
Akula just "wasting"  their secrets.

[TinselKoala]

Yes, you are probably right.

ETA: I forgot to mention that the 270 mA is the maximum draw. The current draw goes way down when the trimpot is adjusted right, and for the alt.snakeoil trial run with hidden battery I had the current draw down to around 25 or 30 mA by adjusting the pot to the point where the LEDs just started to dim.

[TinselKoala]

OK, here is an example scopeshot from my build.

I am inputting 3.0 volts and the system is drawing about 80 mA. I have the TIP3055 transistor instead of a mosfet in place. I have the larger inductance of the flyback ferrite transformator hooked to the 34063 side and the smaller inductance to the inverter side. Phasing makes a difference. Scope probe positions as on the schematic. The LEDs are blindingly brilliant, much brighter than in the alt.snakeoil video. Potentiometer adjustment is critical.

The rest of the needed info is on the image. The top trace portion that is outlined in yellow, and the corresponding areas on the other cycles, is filled with the spiky output of the 34063 but at very high frequency; the scope can't show them at this slow timebase. They appear to be happening at about 40 kHz but the frequency fluctuates rapidly.

So far, this is the most "sinus" like signal that I have been able to get from the inverter Pin 13 - coil "secondary" connection. Doesn't it look like a great shark fin? I wonder if that has any significance.



I suppose I should hook up the DSO for more detail. It's a lot easier to explore the problem space with the analog scope though, since changing settings is a lot easier on the analog.