Lasersaber strikes again. A joule thief king ?

[conradelektro]

Lasersaber shows the characteristic "ringing bursts" of his JSR Looper in his videos:

http://www.youtube.com/watch?v=5EkXNRWAi1Y (minute 6:50, on the base of the transistor, he calls it "over the load = LEDs")

http://www.youtube.com/watch?v=6B79UJGoNJE (minute 1:00)

http://www.youtube.com/watch?v=Xk35CpCFg1w (I recall that Lasersaber also showed a scope shot over the colector/emitter in this video, but it does not work on YouTube at the moment)

My tests show that one only gets the very low power draw if the circuit shows these "ringing bursts" and I think that these "ringing bursts" are the only true "replication of Lasesaber's JSR Looper".

Please see the attached scope shots.

I do not want to prove anything and in my "replications" the LEDs do not give useful light unless one adjusts the circuit with a 1M resistor (from the base of the transistor to positive rail) to a power draw of about 100 µA. In my replications I could get a power draw below 1 µA but the LED was only glowing very feebly.

May be the "ringing bursts" are of no use, but I think that they are the characteristics of a "JSR Looper".

Lasersaber should make a few more scope shots to clarify that.

Greetings, Conrad

P.S.: I also show the "ringing bursts" with a 2N1304 in this post http://www.overunity.com/14591/lasersaber-strikes-again-a-joule-thief-king/msg403065/#msg403065

And one can produce these "ringing bursts" with any transistor and with almost any coil with a circuit published by Vortex1
http://www.overunity.com/14591/lasersaber-strikes-again-a-joule-thief-king/msg403216/#msg403216

[MarkE]

The ring-out is a result of discontinuous conduction.  It means that you do lose a bit of energy, but generally speaking that will be very small and is determined by the parasitic capacitance in your coil.  For circuits like these that are strained to turn the transistors on and off really hard and fast, discontinuous conduction is likely to deliver much better efficiency than continuous conduction.

[Collapsingfield]

By my experiments and simulations by Ltspice everything depend on the parasitic capacitance of the two coil. Sometimes you need to use additional small (10-300pF) capacitor between the two coil to start ringing. In my opinion the continuous ringing needed to reach enough intensity. I think the main "trick" is the ringing in the coils and between them by the parasitic capacitance feed by pulses and the ringing current on the LED (because of the capacitance in closed state??) so the charges are  "reused". It is more efficient then DC discharging calculated by stored charge in the condenser. For example 10000uF at 9V store 90mColoumb= 90mAs. So you can light a LED with 1mA for 90 second - yes, I now that it is not easy in practice - this calculation is good theoretically for the best case. The SJR is able to light the same LED for much longer. Additional advantage of this solution is the big independence of the voltage in the condenser. It can suck out almost every energy from the condenser.
Regards

[conradelektro]

By my experiments and simulations by Ltspice everything depend on the parasitic capacitance of the two coil. Sometimes you need to use additional small (10-300pF) capacitor between the two coil to start ringing.

@Collapsingfield: Could you please explain how to connect the "additional small (10 - 300pF) cap" in the JSR Looper circuit which I show e.g. here:
http://www.overunity.com/14591/lasersaber-strikes-again-a-joule-thief-king/msg406424/#msg406424

You could download the circuit drawing and the "additional cap" could be added with the programm "Paint" which is on every Windows PC (then attach the altered circuit drawing to your post). Or you could post the circuit you used in your LTSpice simulation.

Greetings, Conrad

[Collapsingfield]

Here is the Ltspice file: