How does this simple circuit operate?

[walid1]

Hi
this is the circuit

thanks

 

[Hero999]

Google for Joule Thief and if that doesn't answer your question Google for blocking oscillator.

 

[walid1]

Thank you hero999 it helped me

 

[Kevin Weddle]

Hello Walid, I have never seen an LED in parallel with a 2N3904 transistor. Does it have current gain, voltage gain, or both? The power source doesn't have a voltage value.

 

[audioguru2]

Hello Walid, I have never seen an LED in parallel with a 2N3904 transistor. Does it have current gain, voltage gain, or both? The power source doesn't have a voltage value.

It is a simple transformer-coupled oscillator and voltage stepup circuit. Its output voltage is as high as the LED forward voltage which is about 3.5V for a white or blue LED.

It works with a "dead" 1.5V battery that is about 1V. It keeps working until the battery voltage drops to about 0.3V.

The internet has some wrong explanations about how it works. I found a correct explanation and schematic: View attachment 41199

View attachment 41200

 

[steven2]

thats a joule thief circuit  it sucks the last remaining power out of a dead battery and are very usefull  ive biult a number of these and have a bigger one i use to recharge my watch battery with and desulphate my 12 volts slr battery

 

[steven2]

heres the 2 bigger joule thiefs i used to recharge flatt button cell watch batteries and desulphate dead sla batteries and even charge a photoflash capacitor to

View attachment 41201

 

[steven2]

View attachment 41202

 

[Kevin Weddle]

Instead of an LED, could you use a diode and a small light bulb in series? Maybe a single 1.5V battery to power the small bulb? It looks like the component values would be difficult to determine in order to get it to oscillate.

 

[Hero999]

No because it's not powerful enough to power an incandescent lamp.

If you need to run a small incandescent lamp from a single AA battery then buy one which is rated to 1.2V

 

[Kevin Weddle]

Well of course. It's an interesting circuit which Steven has built and works. It is difficult to expect circuits to function unless the component values are exact and the design accounts for component tolerances.

 

[audioguru2]

The Joule Thief is a very simple transformer-coupled blocking oscillator with very few parts.
Almost any amount of turns on the transformer will work and a power transistor can be used (like Steven did) when the base resistor value is reduced.

It is such a simple circuit (using positive feedback which causes oscillation) that component values and tolerences hardly make any difference.

 

[steven2]

heres another one

View attachment 41203

 

[steven2]

heres the one that i biult from  one of the circuits featured on youtube that i used to desulphate my 12 volts sla battery and charge a photoflash cap up enough for its neon to come on. i brought 7 meters of insulated red wire and 7 meters of black  and i wound more than 90 turns , i thread the pair of wires through the inductor ring  about a meter then wound that on clockwise then when it was done i got the longer bit left and wound that over the first winding counter clock wise  so i made mine a bifilar wound  clockwise and counter clockwise wound , its not that i dont know if the way i wound it makes any difference but it was just an experiment

View attachment 41204

 

[Hero999]

heres the one that i biult from  one of the circuits featured on youtube that i used to desulphate my 12 volts sla battery and charge a photoflash cap up enough for its neon to come on. i brought 7 meters of insulated red wire and 7 meters of black  and i wound more than 90 turns , i thread the pair of wires through the inductor ring  about a meter

Well I don't see the point of the neon. If the current is low enough for the neon to light without blowing up, it will have a maximum current rating of something liek 10mA so it can't have a high enough current to charge a SLA. The neon could just as easily be replaced with an LED and it will work fine.

then wound that on clockwise then when it was done i got the longer bit left and wound that over the first winding counter clock wise  so i made mine a bifilar wound  clockwise and counter clockwise wound , its not that i dont know if the way i wound it makes any difference but it was just an experiment

If you wound it clockwise then winding anticlockwise will reduce the inductance because the magnetic fields cancel. For example, a transformer with a 20 turn secondary, with 5 of the turns wound in the opposite direction to the rest, will be equivalent to the same transformer with a 10 turn secondary, 15 - 5 = 10. Winding in the opposite direction is counter productive because it increases the series resistance. You're circuit would be better if you simply removed the turns wound in the opposite direction.

 

[steven2]

and yet it works fine no problems and never had the neon blow out at all  and i charged a photo flash capacitor up to the voltage rateing of the neon and it light up well . heres a picture of it the pink glow comes out on film but not when you look at it directly , and dispite haveing some counter clockwise wounds it still works like the one featured on youtube  and for 12 volts i got much more voltage output than i  put into it  so if my conter clockwise turns do any cancelling out  then why do i still get more than i put into it  as for the neon well i followed the giudelines on youtube and put it in  and it works fine  in fact what you see on the project box pannel was originally a 12 volts dc gl;obe auto globe but i converted it to hold a 90 volts neon , ill tell you exactly how much voltage output i measured when i find my notes in amoungst all the 169 gigabytes hardrive

View attachment 41205

 

[Hero999]

and yet it works fine no problems and never had the neon blow out at all  and i charged a photo flash capacitor up to the voltage rateing of the neon and it light up well . heres a picture of it the pink glow comes out on film but not when you look at it directly ,

The neon is pink because it's ridiculously overdriven. The colour is different at higher currents because the lamp pressure will be higher and other gasses in the mix will be ionised.

It works but for how long? The neon will eventually burn out.

and dispite haveing some counter clockwise wounds it still works like the one featured on youtube  and for 12 volts i got much more voltage output than i  put into it  so if my conter clockwise turns do any cancelling out  then why do i still get more than i put into it  as for the neon well i followed the giudelines on youtube and put it in  and it works fine  in fact what you see on the project box pannel was originally a 12 volts dc gl;obe auto globe but i converted it to hold a 90 volts neon , ill tell you exactly how much voltage output i measured when i find my notes in amoungst all the 169 gigabytes hardrive

You didn't read my post properly, I didn't say it wouldn't work, just that the way you've wound it makes it less efficiency. Try removing the extra turns, it will be more efficient.

The voltage across the neon will also be much less than 90V with the high current you're putting through it.

Did you measure the current with a true sine wave meter or an oscilloscope with a current probe or sense resistor? Don't bother trying to use a meter, the reading won't be accurate.

 

[steven2]

my atempt at posting my note failed its to big i geuss theres not enough room here to upload it all

 

[Hero999]

How large is the file?

I just reread your post and noticed something else:

heres a picture of it the pink glow comes out on film but not when you look at it directly

The fact that it appeard pink on film but not when you're looking at it implies it's emitting invisible radiation, probably UV, try putting some white paper next to it to see if it glows blue.