Joule Thief

[TinselKoala]

Yes, varying core saturation can produce many interesting effects. The MEG is an electronic system based on this idea and the eOrbo and ssOrbo of Steorn are also dependent on varying core saturation, in a mechanical and an electronic implementation respectively. The "Mag-amp" or magnetic amplifier has been around for a long time.
I've got that long toroidal ferrite "bead" that I used in the toroid winding demo video wound with some toroidal wraps, then I put a circumferential wrapping around the outside. It is very interesting to  monitor this coil on the scope while the other wraps are doing the "JT thing" in a Joule Thief.
I just put into operation my current-limiting PS, and now I'll be able to see what happens to the JT when I energize this circumferential coil which will change the permeabilty/saturation level of the core.

The "thing" is now, at 1645, back up to 1.320 V from an earlier low about two hours ago of 1.318 V. The only thing that has changed is that I've turned the window air conditioner on to reduce the mugginess in the DeepBunker. Could my instruments and/or this circuit be that sensitive to temperature and humidity?

[e2matrix]

Here is a point to consider when looking at very small voltage increases in this circuit.  I'm not sure if this applies to 5mm type LED's but I know that the surface mount type higher wattage LED's like 1 watt and up will create a voltage and act like a solar cell when exposed to a light source - Sun or even room light could create a small charging effect from the LED.  That would not account for the effect Gadgetmall had and especially not the runaway one but it might account for some other small voltage increases in the circuit.  I discovered this when I had an LED flashlight that would not turn on if it's pointed at the Sun (a design booboo by the mfgr.).

[NickZ]

  I do use a permanent magnet next to or under my various Ringer's transformer coil. It does change the circuits frequency, but I don't see much difference in the actual voltage output, or bulb brightness. You can hear how the frequency is raised or lowered in pitch.  I go by the volume level of the ringers' tone, not the higher or lower pitch, but the volume level increase of the ringing. The more noise it makes, the higher the output. There probably are sweet spots, though.

[tysb3]

hi TK,
its maybe help for you:    " Just completed my first experiment. What I found: battery voltage much depends on the temperature."

http://www.overunity.com/8334/second-stage-joule-thief-circuits/msg237123/#msg237123

[TinselKoala]

@tysb3That is excellent work! Real data carefully gathered and properly presented. I am so pleased with you.... that I am not even embarrassed that we seem to be covering old ground again. It looks like there was a lot of good work done in that thread that I have been completely missing.
I also am amazed by the pulsetrain graph right after your temperature plot. That is exactly what I saw when my LSJT first began to oscillate! It would make groups of pulses just like that, same appearance and timing and intervals. This effect didn't last for me, though and now the circuit is making continuous oscillations. I don't understand what I did to make it do the groups like you show.
Your temperature graph seems to reflect what I'm seeing. There is a slow overall decline, with periods of steady voltage and some periods of slight climbing. I haven't been keeping good records but my impression is also that warmer temperatures make the voltage go up a bit.
Again... thank you so much for posting the link to your earlier work. That is the way to do it!