Joule Thief

[TinselKoala]

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

[hartiberlin]

I posted this already in 1999 when I experimented with
a 12 Volts Fluorescent tube driver circuit, which works simular to a Joule Thief.

This also had some good Electron sucking capabilities from the ground,
so I am posting it here again as a reference.


The interesting thing was, that the Fl tube had a full square wave on it this way, having the diode
across the fl tube.

And the fl tube was pretty bright, but the circuit did not consume much more
when the tube was connected or not. So the difference input power between no fl tube connected and
fl tube connected was very low...

Regards, Stefan.

[crowclaw]

These FL tubes consume very little current once they conduct at a  higher enough voltage. I posted some time back a modified 14 watt ES domestic lamp lit up brightly held in my hand. The electronic drive circuit was removed and a modified camera flash circuit plus a single AAA battery was shoe horned into the screw in base.
Probably a similar effect here, the drive circuits alone is take all the current... if it's actually providing drive or just coupling the HV straight through the coupling transformer is debateable. Would be interesting to try removing the drive circuit completely and coupling the HV onto the transformers primary or again just like a modified FL direct onto the tube. Similar to neon gas that glows when a high voltage is applied, so FL tubes glow the same and as such very little current is required to achieve reasonable light.
Hence I'ts always the drive circuits that take the current that's necessary to create a high enough working voltage.
Kind regards Merv

[powercat]

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

Hi TK great videos, eficiency just keeps getting better and better,
10,000 seconds is approximately 2 hours 46 minutes I wonder how much light energy there is from the LED to run a small solar cell and put power back into the capacitor, then how much longer would it run ?
It is such a breath of fresh air to look at a thread where everyone is helping each other and no one is making any extravagant claims they can't prove.

[gadgetmall]

I have confirmed this circuit is >1 with a nicad . I need someone to replicate it . My start voltage was 1.228 yesterday at 11am  . today at 10:12 am it is now 1.248 and lit and continuing to rise . No emf devices are on near it and My Android phone says no emf in that area other than the device itself. ..My heart is Pounding ! you must use exactly the parts including  the germanium diodes and .0001uf disk cap  on your feedback legs . My unit is not grounded and no antenna.Now i am going to dismantle it and shorten all the wires and Box it .

I know i might jump the gun sometimes but i know what I see . I don't have an explanation for it but i like it all the same . This little transformer is an audio transformer with one winding i got out of a radio . it looks like a small E-Core . There was enough space to wrap a bell wire primary . I started with 12 turns and ended with 8 working perfectly . I used TWO diodes .. the Extra diode turned opposite from the first one . The Eureka moment cam about 8am when i placed a cap between the extra diode . the battery was checked yesterday and it was 1.228 standing volts before the circuit was energized . I simple put a cap this morning without shutting it off and it is really charging very good now . it is already 1.251 this second . all i can say is WOW>