Lasersaber strikes again. A joule thief king ?

[conradelektro]

But meanwhile the Loopstick meets LS Looper video demo is done. The change in brightness in the LEDs doesn't show up so well on the video but I hope you can at least tell a bit of what's going on.

http://www.youtube.com/watch?v=eqBK00tuI8Y

@TinselKoala:

Very interesting, may be the 20 turns (one end free) act as an antenna and you are tuning to the 60Hz of the 110 V mains or some radio station or cell phone tower near your lab?

May be you can check with your scope whether your loop stick circuit is doing the "ringing bursts" which I show here http://www.overunity.com/14591/lasersaber-strikes-again-a-joule-thief-king/msg405416/#msg405416 or here http://www.overunity.com/14591/lasersaber-strikes-again-a-joule-thief-king/msg403109/#msg403109 (this is the characteristic ringing of LaserSaber's circuit).

To get a very low power demand one needs these "ringing bursts" instead of the "normal Joule Thief type oscillation" (measured over collector/emitter of the transistor).

If one uses the MPSA18 almost any coil will cause oscillation, but only with a very "special coil" (high inductance, 2:1 ratio) one gets these "ringing bursts".

These ringing bursts can be produced with any transistor and almost any coil if one uses a circuit published by Vortex1 here http://www.overunity.com/14591/lasersaber-strikes-again-a-joule-thief-king/msg403216/#msg403216 . But the power draw is at least ten times higher than with LaserSabers circuit (and the MPSA18).

A far as I have come:

- use a MPSA18
- then try to get a coil similar to LaserSaber's pot core coil (I think it needs high inductance and a 2:1 ratio)

If you can confirm the "ringing bursts" in your loop stick circuit, you have successfully replaced LaserSaber's coil with your own. And your circuit works with a very low supply Voltage as well (which is nice). Low Voltage (below 1.5 Volt) and low current (less than 1 µA on average) is a neat trick if the LEDs are at least glowing nicely. A magnet shaker would be a good enough power source (one shake and go for several minutes).

Unfortunatelly some chores are keeping me from experimenting at least for an other week.

Greetings, Conrad

[conradelektro]

I wound a strange coil with Litze wire (0.75 mm diameter, thick isolation). It has a 16 turn primary and a 45 turn secondary.

LaserSaber's latest circuit shows the "ringing bursts" with this coil (transformer) but it needs a 20M (or 10M) resistor from the base of the transistor to the positive rail in order to work and it only works if the secondary is steped up.

The LEDs are brighter than I ever could do with LaserSaber's SJR Lopper circuit.

See the attached files for details (circuit, photo with power draw measurements, scope shots show "ringing bursts").

My ideas about getting the "ringing bursts" with LaserSabers SJR Lopper V3.0 circuit http://laserhacker.com/?p=420:

- use the MPSA18 transistor
- coil (transformer) should have a very high inductance (pot core will help, step down is good but I could not replicate that)
- put a 10M to 20M resistor between base of transistor and positive rail (this will help if the coil is not optimal)

To make my replication better (brighter LEDs for less power) I still have to find the right wire for my transformer core.

Greetings, Conrad

[TinselKoala]

What you are calling "ringing bursts" are simply the response of a fairly high-Q tank circuit to a stimulating pulse that is happening at a much lower frequency than the resonant frequency of the tank.

It actually represents an energy loss, I think. The whole of the energy represented by the square-topped part of the pulse just before the ringing, is dissipated in the resistance of the tank components and radiated as RF at the ringing frequency, during each nice exponential decay of the pulse.

Do you have a phototransistor or CdS photoresistor? It would be interesting to use such a device to monitor the light pulses from the LED and compare them to the ringing burst trace to see when the light is actually on during the waveform.

I've got a lot of similar example waveforms but here is the easiest one to look at that I could find. The Arduino inductance meter puts an unknown inductance into a tank circuit with a good quality known capacitor, and "strikes" the tank with a single sharp pulse and then observes the ringdown, determining the frequency. Since the capacitance is known, the inductance may be calculated from the ringing frequency.

http://www.youtube.com/watch?v=Qx3B89379eQ

Also, I see that your schematic says "2V-20V". There is a hundred times more energy in a cap charged to 20 volts than the same cap charged to 2 volts.... no wonder small caps give long runtimes, if you are allowed to charge to "high voltages" like 20 volts!

[d3x0r]

What you are calling "ringing bursts" are simply the response of a fairly high-Q tank circuit to a stimulating pulse that is happening at a much lower frequency than the resonant frequency of the tank.

It actually represents an energy loss, I think. The whole of the energy represented by the square-topped part of the pulse just before the ringing, is dissipated in the resistance of the tank components and radiated as RF at the ringing frequency, during each nice exponential decay of the pulse.

Do you have a phototransistor or CdS photoresistor? It would be interesting to use such a device to monitor the light pulses from the LED and compare them to the ringing burst trace to see when the light is actually on during the waveform.

I've got a lot of similar example waveforms but here is the easiest one to look at that I could find. The Arduino inductance meter puts an unknown inductance into a tank circuit with a good quality known capacitor, and "strikes" the tank with a single sharp pulse and then observes the ringdown, determining the frequency. Since the capacitance is known, the inductance may be calculated from the ringing frequency.

http://www.youtube.com/watch?v=Qx3B89379eQ

Also, I see that your schematic says "2V-20V". There is a hundred times more energy in a cap charged to 20 volts than the same cap charged to 2 volts.... no wonder small caps give long runtimes, if you are allowed to charge to "high voltages" like 20 volts!

I think these are the ringiog bursts he actually means... http://youtu.be/8mJX5qW4ghk?t=2m25s
I have not seen this behavior in any configuration; I only get a continous pulsing...
(conrad)'s don't look anything like this ...

[TinselKoala]

I think these are the ringiog bursts he actually means... http://youtu.be/8mJX5qW4ghk?t=2m25s
I have not seen this behavior in any configuration; I only get a continous pulsing...
(conrad)'s don't look anything like this ...

Oh, sure, I see what you mean. Those are much more like the pulse trains from one of the Akula circuits.

Yes, my Loopstick device does things like that too, depending on the voltage and the setting of the ferrite into the loopstick. It does even weirder things too. I have never charged it to over 3 volts though.

I'll make a short video showing scopetraces later on this evening. It has at least two modes of oscillation. One where it makes "burst packets" like shown in the video, except within the bursts there is a high-frequency nearly pure sinusoid waveform. The other is a train or "segmented train" of ringdown waveforms similar to conrad's above.