Joule Thief not working


I would like to experiment a little bit with joule thief circuits.
Unfortunatly I seem to cant get it working.

I build a circuit with help of a schematic I found on the internet, as you can see
in the middle of the image beneath.
On the left on the image I measured the voltage where the LED should go and I get
a voltage level of 0.58V whereas a direct reading of the solar cell gives me a reading
of 2.55V (right picture).
Im using a 2n2222 transistor.

If there would be too less woundings on the ferrit torrid it should be at least the same
as the direct reading right?

Is there any wiring problem? How do I boost my voltage level right? Because there has
to be something wrong with my setup… :~

Thanks very much for help! :slight_smile:

Two things, first I suspect that you can't get enough current from your solar cell to make it work. Try replacing it with a battery just to test.
Second you can't make the circuit on bread board, those circuits are very layout dependant. You normally need a PCB.

Thanks Mike!

I tried with an old 1.5V AA battery with no luck but no difference. Direct
reading 1.2V and in circuit also 1.2V (ok here we have some progress).
But I have to see some sort of oscillation right?

Hm but eg. these "freeform" circuits working as well :frowning:

Is the transistor the problem?

Any other ideas? :~

Thanks in advance!

You need to keep the leads as short as possible for that circuit - blob-soldering it together on a prototyping board would be far better (if not very pretty). Components an inch or two apart are going to experience far far less stray inductance.

Most croc-clip hookup leads I've bought are utterly hopeless BTW, very high resistance (I've had some with several ohms or so from clip-to-clip and with intermittent connection). I buy croc-clips and solder my own leads from proper heavy-guage wire so they are good for decent currents and have decently low resistance. For instance I've got a nice pretty looking lead with matching rubber boots that's about 0.5 ohms, and a similar length one that I made myself that is 0.02 ohms... 25 times better.

Incidentally 0.5ohms in the wrong place in a power-conversion circuit can easily stop it working. Check those leads!

Your toroid looks pretty loose and with too few turns. The wire should be wound tightly onto the core and ideally be a single core conductor. Lacquer covered single core copper wire (like that in motor windings) is commonly used as the thin insulation gets the windings close together. I would go for 15-20 turns each to get a good flux up.

Also notice the dot notation on the toroid. The primary and secondary windings have to be wound in the correct direction relative to each other or the polarity will be wrong. (see Polarity (mutual inductance) - Wikipedia).

I agree with Grumpy Mike regarding the solar cell. These circuits make their "extra" voltage by drawing extra current on the input. Keep it simple to start with and replace the uncertain variable of the solar cell with a battery.

Thanks guys!

I will investigate further on weekend considering your advises! :slight_smile:


I didn't have any trouble getting a Joule-thief circuit working on a breadboard, but I have my doubts about those long jumper leads you're using to connect to the coil.

Step 1: try reversing one of the coil connections to get the "dots" correct.
Step 2: tighter turns, more windings, and shorter leads for the coil.