Joule Thief

[Pirate88179]

This is the smallest JT I have made to date.  I cheated and used a modified version of the garden light circuit board.  You can see another little board in the photos.  These are the ones with the gray blob over the chip.  All of the chips I have looked at would not fit under that blob so possibly it is an smd version?

This is a very bright little light and will light up down to .6 volts.  It draws less than 8 mA's when fully illuminated.

The case for the light is made from a 35mm film container lid and the base is made from a cut down Airborne (anti-cold herbal remedy) container.  It is the smallest package I could readily make but, there is a lot of room inside so it could be place in an even smaller package.

Bill

[Pirate88179]

https://www.youtube.com/watch?v=D6XVcqOs-Cg

Here is a short video of my Micro Joule Thief Light.

Bill

[Farmhand]

Cool stuff, your right the 52521H IC isn't much use for extended run times or for what I want
which is a more efficient circuit to run from a galvanic battery.

The setup in the picture below gives only 0.5 volts under load with about 1 to 2 mA, but I adjusted
the base resistance so that it runs the JT in burst mode powering two 5 mm LED's, meaning it
oscillates at 60 kHz until the voltage drops enough to pause the oscillations very briefly, that makes
the LED's brighter but flashing, another cell or a bit more current and it would run continuous.

It can run at 0.38 volts and more current. The cell has galvanized steel and
carbon arc gouging electrodes (copper coated with an inch of exposed carbon in the ground).

[Pirate88179]

Cool stuff, your right the 52521H IC isn't much use for extended run times or for what I want
which is a more efficient circuit to run from a galvanic battery.

The setup in the picture below gives only 0.5 volts under load with about 1 to 2 mA, but I adjusted
the base resistance so that it runs the JT in burst mode powering two 5 mm LED's, meaning it
oscillates at 60 kHz until the voltage drops enough to pause the oscillations very briefly, that makes the LED's brighter but flashing another cell or a bit more current and it would run continuous.

It can run at 0.38 volts and more current. The cell has galvanized steel and
carbon arc gouging electrodes (copper coated with an inch of exposed carbon in the ground).

Nice set-up.  Try using magnesium for the - electrodes and you should be able to get about 2.00 volts out of it.  That's about what I get, not counting those high spikes which will charge my 2.7 volt supercap fully.  I have a video on youtube running 200 leds off of my set-up and, I can also run a Bedini motor from it.  Get the largest block of magnesium you can and also, I would strip the copper off of the carbon rods as well.  Place the carbon rods on the north/south meridian to the north, and the magnesium to the south on the same line.  Not magnetic north/south, although, depending upon where you are these can be close.

I also have a cool video showing really high spikes all over the place on my scope.  It is these spikes that I assume are charging up my supercap (650 F) to 2.7 volts when I can only measure about 2 volts with my DMM.  I have been running this system for a long time now...well, actually right now it is not hooked up but maybe I should revisit this and explore it further?

Bill

[TinselKoala]