Big Joule Theif

[Tectalabyss]

Tec & all,
Variable resistors are necessary to find the lowest current draw for the circuit to operate as you want it to. Once found, it can be measured and replaced with a fixed one. If the JT is used for different things, many leave it in so it is versatile for different applications.
Using a meter on a JT circuit is almost useless. You can put in a 1ohm resistor off the battery and measure the current draw across the resistor using the volts range. 1v = 1ma current draw. Sometimes the circuit will dim from that. If you are charging a cap, it can be measured, but not much more. The spikes occur to fast for a meter to read. Most of the reading you see posted on the high voltages and such are from scope measurements.
Good point Tec.

@Vaporizer.
Thanks You made it a lot clearer than I ever could.I don't have a scope yet just my bearskin's & stone knife's I just have to backtrack and recheck my finding's a few time's to be a little more sure and then I can't say with 100 % . But I love the learning. Thanks also for the link's.A lot!!! of very useful reading. Tec

[pentiger]

***Edited below

Hi Everyone

I was away for couple days trying to resolve my problems with the big toroid.

Today I returned and was surprised by this much new info from you.

I need to ask Vaporizer, I loved your modified Fuji circuit, I noticed you have a big transistor there, can you tell me what kind it is?

I did try to use the Fuji JT and also burned the original transistor which is very hard to find from what I saw.

I did have luck with my 3inch toroid.

YouTube - Big joule thief finally working

It was a great project. I need to get myself more toroids and make more of these boxes.

pentiger

--------------------------

I thought it worked, I am getting tired of it. It lasted 30 minutes

I thought I did everything perfect this time around.

Back to the drawing board I guess. I'm gonna cry.......

[pentiger]

where do these come from:

Thanks.

[NickZ]

Guys:
I wonder where he got this coil, (IST- newest coil). He has been posting on Overunity.com's Joule Thief thread going by the name -Osiris, now. He was the one that came up with the toroid tuning method mentioned in my last post here. I would not have expected the results that he came up with on those tests. The more windings on the secondary coil, the more juice, but not so on the primary side. So, it is very important to have just the right L1 to L2 turn combination. Or poor results may be the outcome. That might be what is going on with some of our Jt circuits less than brilliant results. I also notice that not all transistors of the same type run the same, some just aren't any good at all, some are very strong but can also use up the run battery's charge up way too fast. Also, if the transistors gets too hot just once, they are never the same, even though they keep working, but not like before.
I'm still working on finding a feed back system that works, and keeps the run batteries voltage level constant, otherwise its useless, as the frequency is always changing, and no circuit will work properly like that. Resonance has to compensate for all the loses...
Big Coil Builders... good luck to you all.

[cikljamas]

cikljamas,
Don't go with cfl because I did. I was very impressed with Lidmotors vids and bought 4 9W on and order from Goldmine. Of course I found out later LED's are better bang for the ma so most say. Now I have to get some better led's. lol
I have some 12,000mcd white 10*, but they aren't very impressive. I'm looking at a way to diffuse them. The one 13w spiral at good brightness, 2.25v draws 465ma. They lasted for abt 2.5hrs before dropping out. Not really suitable IMHO for long outage use. I'm abt done with the fuji transformers. I can use them for 1 aa and only draw abt 130ma so I should get 20hrs from my power outage "safety" lights to sit around the house. Next is back to coils and see what I can do.
Good luck on what you try and keep us posted!

Hi Vaporizer !
Thanks for answer !

I just wound one coil L1 4 turns, L2 8 turns, and L3 120 turns, and after
FWBR it gives 220 V which is in Europe standard but without FWBR it is
about 100 V with 1,5 batterie...But what is total mystery for me is this :
if i get voltage that i need for cfl to work what about amps, i mean could
one 1,5 V battery give enough current to supply such a thing as it is cfl ???
Do you maybe use some special cfls or it is common cfl for AC current from
the wall socket ?

Maybe stupid questions, but i recall how i made a year ago two inverters
12/220 V with four 2N3055 and i know how these transistors become very
hot very soon so i just doubt whether 1,5 battery with a help of one little toroid of 1 inch can light common AC cfl, it seems unlikely to be true, but who knows...There is lot of things in electronics that i still have to learn, and when i do not know something i just ask without too much shame...

If i understood you well you suggested me to give up from cfls and stick
up with leds ? I already have two little JTs that gives very good results
lightening leds 30000 cd, pretty nice stuffs, and maybe it will be something
that i am going to use much more often then cfls, but i have to try that
(cfls) too, just to see how it works...

Thanks for understanding !

[synchro]

Flash mod.

I busted up five or six disposable cameras and tried as many different failures before I tried this Kodak funsaver modification. Turned out to be real easy and worked like a charm. I"m lighting a flouresent off a 1.5 volt AA. Here's the link:

YouTube - Micro Camera Circuit Mod!

[pentiger]

I busted up five or six disposable cameras and tried as many different failures before I tried this Kodak funsaver modification. Turned out to be real easy and worked like a charm. I"m lighting a flouresent off a 1.5 volt AA. Here's the link:

YouTube - Micro Camera Circuit Mod!

Hi, great tutorial. Do you have a photo of it lighting a cfl?

Thanks.

[xee2]

@ synchro

Easy flyback circuit >>> YouTube - 9V battery Jacob ladder

NOTES:
1. correct transistor part number is MJE13007
2. pin numbers will depend on the flyback you use (pin numbers on schematic are for the flyback I used)
3. I got the circuit idea from Kooler
4. circuit will light fluorescent tube from one AA batttery with only 10 mA (decrease resistor value to get brighter tube)
5. WARNING - high voltage can be dangerous, this circuit is only for those who can work safely with high voltages (output voltage is about 6000 volts)

[NickZ]

Guys:
I made a new Jt today using a one inch toroid with a secondary. My first one with a secondary. Connected it to a TIP3055 transistor, and was able to not only light 20 Leds nice and bright, but also a couple Neons, between the C and E rail, at the same time. First time for me lighting Neons. I also added an extra capacitor (0.047mf) between the E and Base rails which added about 15 percent more brightness. I tried connecting all the available leds that I had, onto the Jt, and it light them all (about 40 or so). Probably could have light many more.
The transistor was getting hot, but when I added the extra capacitor and a neon or two, it helped to help keep it running cooler.
If I could get more fit more windings on the secondary, and also use two AAs, I'll bet that it would light a CFL. It won't light it with just one AA. Anyways, I had my first minor shock from that Jtc today, so I must be doing something right.
My analog meter when off the scale... then later would not move at all while the leds and neons are bright. I'm still trying to figure it out. Pretty neat though, can't wait to try those big coils, like the one in the picture on my last post. Looks like a 5 or 6 inch toroid.
NZ

[Vaporizer]

@ pentiger,
After seeing a vid, I guess that one I have is a Kodak. They arent supposed to be as good as a Fuji, so actually I feel better to get what I did out of it. lol
I read somewhere they werent worth using, Fuji was. The drawings for the Fuji show 45ga wire on the output. That is so fine, I can't imagine it lasting long under a big load.
The transistor in my pic is a TIP31C. When I upped it to 2 AA rechargeables @ 2.25v it cooked the original transistor in just over 10 seconds from heat. The TIP31C handles it well. It may not be the best one for the circuit. I have some 2N3055's I need to try and a few others. then order what is best.

The toroids I have in all my pics so far came from ebay. I just searched and found 3" & 2 3/8" dia. Pure luck. I almost have one of the 2 3/8" wound to test out. I saw a vid of someone lighting a 13w spiral very bright, but they said continuous use fried the transformer(Fuji). That doesnt meet my need for power outages. They were using it with 3 AAA rechargeables from a phone pack. Raising the input voltage is an easy way to get more out since its multiplied through the circuit. Making a better circuit, thats tougher for sure. I'd like it to do well on a AA or AAA. In an emergency, I can be flexable. lol
Sorry you lost yours. What failed?

@ Nick,
In reading all that I did, the link you posted was maybe the 1st from him that actually showed anything. Most I saw were posts of great things & claims, but never any details others could use. I thought he was doing this as a business and was tooting his horn, but keeping it proprietary.
The L1 to L2 ratio I havent figured out yet. The ratio to the secondary is a simple step u transformer. The more winds the higher output voltage, but the higher you step it up the less current it has. The BJT started with a 1:1 L1, L2 which I thought odd as people posted scope shots of a higher voltage. The little thought I I have give it so far seems the trigger voltage fires the circuit from part of the coil and the collapse comes back from it and the other portion of the coil. This should be where the increase comes from. Say 3 winds L1 fires it and 10 total (L2 7 winds) collapses back. Without controlling the base or loading it somewhere, E-C, with something it should continue to increase till failure.
I totally agree with you on the transistors. All of the same number will not be the same and once hot to the touch, they are never the same, but still work to some degree. I still occasionally swap one to see if I damaged it.
Adding the cap does help, but it did increase my current draw. A .047 won't pass dc so it has to be getting ac from the E line and shuttling it back to the base for use. That would increase the saturation and increase the brightness.

@ cikljamas,
I wasn't trying to discourage you from cfl's. I like the way they light.....lol
I have been reading that LED's are so much more efficient is all. Don't kill your meter trying to measure the output coil. It may show 220v and have 600v spikes that are to fast to read resulting in meter death. The "straight" flouro tubes are ran just as they are. The compact ones have caps, neons and circuits in the lights base that need removed or they won't work. I dont think current has much to do with lighting florescent lights the way we are. The higher the voltage the brighter they get. All I can think is the high voltage excites the vapor in the bulb so it floreses the phosphor coating. Excite the gas more, brighter light.
The previous post of the 9w "U" compact had a cap & neon in the base I had to remove. I have a 22w round "halo" florescent that the same circuit lights with no mods(it is a normal 120v AC), but it was dim due to my low output vs its size.

@ xee2,
I've read a lot of your work and its been great. Thank you for sharing. Nice ladder on 9v. TV flybacks can normally pack a lot of current and 30,000v under normal operation. 6000v doesnt surprize me at all. Lighting a cfl with 10ma does though and it somewhat confirms my theory on exciting the gas to a higher potential.

[cikljamas]

@ cikljamas,
I wasn't trying to discourage you from cfl's. I like the way they light.....lol
I have been reading that LED's are so much more efficient is all. Don't kill your meter trying to measure the output coil. It may show 220v and have 600v spikes that are to fast to read resulting in meter death.

Hi Vaporizer !
Why did not you tell me that before, i just killed my analog meter !
I tried to repair this stuff, but i just can not figure out what in the meter
failed to be able to change that with healthy one brand new component...
Fuse and diode protection is written on that meter, and fuse is untouched,
diodes are untouched too, but meter is dead, so what is broken in there
after all ???

neons and circuits in the lights base that need removed or they won't work.
.

Do not get that...Some picture would be of much help !

Thanks and cheers !

[Vaporizer]

@ cikljamas,
Sorry to hear abt your meter. So its my fault? Every thread I read in always mentions to be careful using a meter on JT's. Especially big ones. A good reason to read threads from the beginning?

Here is what is removed from a spiral cfl hidden in the base.
There is a video on youtube somewhere.
Gutting Old CLFs | Jim On Light

[pentiger]

@ pentiger,
After seeing a vid, I guess that one I have is a Kodak. They arent supposed to be as good as a Fuji, so actually I feel better to get what I did out of it. lol
I read somewhere they werent worth using, Fuji was. The drawings for the Fuji show 45ga wire on the output. That is so fine, I can't imagine it lasting long under a big load.
The transistor in my pic is a TIP31C. When I upped it to 2 AA rechargeables @ 2.25v it cooked the original transistor in just over 10 seconds from heat. The TIP31C handles it well. It may not be the best one for the circuit. I have some 2N3055's I need to try and a few others. then order what is best.

Thanks. I had a FUJI circuit and the transistor D1960 blew up when i upped the input voltage to 3v. I tried the tip31 and it doesn't work, that's why I am asking. I guess I will have to get myself a kodak camera now

I still struggle with my big JT as the batteries die to quick on me.
I have a 1k ohm variable resistor and rheostat at the battery and it dies so fast that it's not funny.

I wish that there was bulletproof formula for achieving the perfect JT.

pentiger

[Vaporizer]

Hi pent,
I'm as new as anyone else to these things. Mine had a D2470. They were old boards I got 2-3yrs ago already stripped out. I lost 2 of them. 1st one I didn't catch why and the 2nd one I was feeling for heat and I was to slow lol. The Kodak if I remember reading right has abt 259 ohms on the output and the Fuji is abt 375? If the wire were the same, the Fuji has alot more turns and would have a higher output voltage.

The Tip31 should work. Abt any NPN will, its a matter of what you intend to load it with or how high a voltage you are running.

There is no sure cut & dry method. You can get close though then fine tune.
Have you checked your wiring closely? Do you have a huge load? Something is causing all the heat and battery drain. Could your coils be shorted to the core?
Can you post a drawing or pic of your circuit? If you are using just a trigger & primary coil with 1 LED across the E-C, it should not get hot or draw hardly any current at all. If you backed up to that point, you can add a 1 ohm resistor from battery + to the circuit. Then set you meter for volts and read on both sides of the resistor with your probes. Not to ground. Red on one side, black on the other side of the resistor. The reading in volts is now the current being used. 1v=1ma. So, if it's reading .465v using 1 LED, that's 465ma and way to much. It should be just a few ma. If its high, your transistor or coil is bad somehow. That is all with 1 AA or AAA.

Hope that helps you find the problem.

[NickZ]

@Pen:
Maybe try to reverse the polarity off the secondary. See it that helps. It may be some simple little thing that keeps it from lasting longer.
I notice that the leds bulbs will light both ways on a secondary, but they are a bit brighter connected one way than the other.
I've had the same problem with short duty cycles, but sometimes it just goes away and everything work fine again. I still can't figure all that is going on yet.