Joule Thief

[jeanna]

@jesus,
Thank you for your suggestion. Is there anyone else that is seeing that 2 makes a difference?

@xee2, so the cap across the battery which appears to offer nothing, is protecting the circuit and allowing a spike to be there faster than a battery would?

@all,
About the circuit with the uneven turns. I see nothing non-linear here as much as I try. I can run down all the numbers for you if someone wants. I am happy to do that. Here is the recap.

The basic circuit with its
1k base resistor, the
2N3904 transistor,
a VR in place below the base with its cap || to it. but the VR is set to zero...and
5+5 bifilar with 120 turns secondary
the secondary to the leds :

The rectified voltage of the secondary is 34.6V both ways
The amp draw is 16.8mA

As I added turns to the primary collector coil the rectified volts AND the amps both went up.

Any time I turned the variable resistor with its parallel cap to anything above zero all the way to 10.8K the amperage went down, but the rectified voltage did too.

--

So, it is all fine to light leds with low amperage and the high voltage which is oscillating, but I am not yet prepared to think I am getting near to replicating this fuji transformer.

The fuji circuit has a clear metallic capacitor in a place that is not on the base of the transistor, I think. This is not the big-mama cap, but something that seems to point towards the second coil which could be a relay. I don't want to add more than one thing to my jt coil at a time and this base cap has not been beefing up my circuit.


@gadgetmall, I am trying to remember if the joule thief basic is where you have added this 68pF cap? I think it is. Do you have any suggestions here?


@MK1 / Ist,
When you count the rectified volts, are you adding the 2 directions together? If the rectified volts read 39v each way is that considered 39vac? or is it 78vac?

@jim,
How about a have-a-heart trap with the modifications that after they are in the trap they go down a hallway to a hamster running wheel which is hooked up to an induction generator? Maybe they can cook your morning coffee. That way, you both win. 

Thank you,

jeanna

PS I think kubikop is just showing the peltier effect can go both ways. Really, he boiled water for one side, and he used ice cubes for the other and lit a led for 10 minutes. This is interesting but does not show efficiency. IMHO  Now, if a peltier doodad can be run by the high frequency voltage coming off a joule thief, that would be very exciting. In my next order from allelectronics I will get the one they have for $14. It is too high priced for a usual order from them, (especially after those leds) but it is a question I want answered.

Hey Gadgetmall, wanna try it?

[xee2]

@ jeanna

allowing a spike to be there faster than a battery would?

Yes.

Is your secondary  coil split into two separate coils? When you give secondary voltages each way, is that the voltages for the two parts of the secondary coil?

[Pirate88179]

http://www.youtube.com/watch?v=JXo4-gYBczc

My cat is feeling a bit better so I made a quick video of 100 leds running off a single 2.3 volt, 10 F supercap which is powering the modified Fuji circuit.  I am considering ordering a 100 farad cap tonight. Have to check finances first.

Bill

[Mk1]

@jeanna

Remember i got 4 coils 2 pickup coil, i don't even look at the coil voltage both ways , there usually about the same once rectified,

On this picture you see output from pickup coil 1 in red for 67 volts, and 2 in blue for the second 67 volts , all the test are done with the same design of 4 coils , i did it on many different core, and the output is really adjusted by the resistor that goes to the transistor base, on this core 1k is a ok , but not for the other ones. Each different toroid of ferrite or powdered iron will work but at there own speed (adjusted by the resistor) and induction, some need more coil turn some less (on the jt side)for getting enough induction for a good pickup coil output.


This core dark green works at 1k when used with a 2n3904
The yellow with white edge needs 5 ohm with a 2n3904
The green with a blue edge needs a 55 ohm with a 2n3904
The green with a red edge is under testing
All those are the same size toroid

@all

By the way , to me iron rust is iron even if baked in ceramic over , ferrite is mostly iron and powder iron  or not it will work when done right , in french "fer" means iron.

Mark

I got pc issue damn the trojan! I getting harder and harder to be with you friends.

[jeanna]

double post --- sorry