Big Joule Theif

[Tectalabyss]

Pentiger and All:
I just bought something very similar to those Solar Christmas lights, but without the solar part.
My light set has a mode controler, 50 leds lights, but runs on 120 volts AC power instead of solar. I had to cut the three Led lines all to pieces to get it to work with my newest joule thief circuit. As there were various resistors inside the three lines. Jeanna helped me out on how to go about it, which really did help.
Pentiger- It may not be a Jt inside the lights controler but instead a Led driver circuit.
The 3 inch coil that you have looks great, just keep trying different transistors and resistors, until you get it to work the way you want it to. You really need to get some small pots to help tune it, and tuning may be all you really need to do, to get it right. Good luck with it.
Tec: I use various transistors, like the Tip3055, 2n2222A, 2k2222, or whatever works, although many transitors can function to some degree, there are big differences between each one. And the heat issue is fairly complicate when tryng to get more output. Finding the right components for the job is the trick, and building a Jtc that will Resonate with the particular toroid coil, without heat, or quickly draining the run battery's charge.
Jeanna suggested to try a thousand things...
NZ

Hello NickZ and All.
I've also tried many transistors even some of the older ones I had pulled from old tv's an radio's back in the day when they weren't that old I am still learning the newer electronics My job over the past years have been 3 phase 480 volt industrial Maintenance.I believe the JT or some variation of it could be used for a better power supply sadly I just don't have the skills to be of much help. Tec

[NickZ]

Pentiger and All:
I just bought something very similar to those Solar Christmas lights, but without the solar part.
My light set has a mode controler, 50 leds lights, but runs on 120 volts AC power instead of solar. I had to cut the three Led lines all to pieces to get it to work with my newest joule thief circuit. As there were various resistors inside the three lines. Jeanna helped me out on how to go about it, which really did help.
Pentiger- It may not be a Jt inside the lights controler but instead a Led driver circuit.
The 3 inch coil that you have looks great, just keep trying different transistors and resistors, until you get it to work the way you want it to. You really need to get some small pots to help tune it, and tuning may be all you really need to do, to get it right. Good luck with it.
Tec: I use various transistors, like the Tip3055, 2n2222A, 2k2222, or whatever works, although many transitors can function to some degree, there are big differences between each one. And the heat issue is fairly complicate when tryng to get more output. Finding the right components for the job is the trick, and building a Jtc that will Resonate with the particular toroid coil, without heat, or quickly draining the run battery's charge.
Jeanna suggested to try a thousand things...
NZ

[pentiger]

hello I found something and wanted to share with you guys.

Buy Solar light, outdoor solar light, Solar led light, wholesale Solar Powered 60 Blue LED Christmas String Fair Lights at Aliexpress.com

commercial Joule thief with 60 leds.

pentiger

[Tectalabyss]

Guys:
Although I have not had problems with the Jt transistor overheating, I have notice that if I remove the led that is across the C and E rail, everything gets very, very hot. If I left it like that, the transistor would melt down. But if I use one low voltage 2 volt red Led bulb there (between the C and E rail), there is absolutely no heat. The point being that it may be important to include at least one led between the rails, even at the expence of it lowering the secondary voltage output, by a little bit. No heat may also be related to longer run times or duty cycles, as there is less waisted energy.
NZ

Hello NickZ and all.
What voltage are you running type transistor? I have never had a problem with heat on my transistor unless i tried to run to high a voltage with or with out anything on the e/c do you think maybe that is whats going on ? Tec

[Tectalabyss]

@ Tectalabyss,
I see. That point (E-C) is normally 1 LED or neon to show the oscillator is running, nothing more. Many remove it once a stable secondary load is established to reduce primary current draw from the battery. The paralleled LEDs there are lit by the feedback signal, but are also still on the source battery side using battery current too. Most use a separate winding(pickup) on the toroid to run them. This way all they see is the voltage developed by the collapsing magnetic field of the oscillator, never actually connected to the battery side, and totally isolated.
That should help your time run. Lidmotor has a good example of that with Jennas light vid. He has a 3" toroid running a CFL. The JT is only ringing the toroid, everything else is picked up from the tertiary(3rd) winding. He has room in that vid for many more separate windings too. Its pretty impressive to watch when you think of what its doing. I'll find the link if you need me too.

Edit: Actually, the one LED/neon also protects the transistor from HV spikes till a load is used to do more than just a simple JT circuit. Needed to clarify so ppl didnt start blowing transistors!

@ Vaporizer.
Thanks for the reply,That was with a small 1" toroid and a small secondary I have even tried running a 1" toroid small secondary nothing on the e/c secondary running back to run battery on another 1" JT 1" with a secondary.
(You make things clear and I for one am glad to have you Here on Energetic Forum) Thank you. Tec

[pentiger]

Guys:
Although I have not had problems with the Jt transistor overheating, I have notice that if I remove the led that is across the C and E rail, everything gets very, very hot. If I left it like that, the transistor would melt down. But if I use one low voltage 2 volt red Led bulb there (between the C and E rail), there is absolutely no heat. The point being that it may be important to include at least one led between the rails, even at the expence of it lowering the secondary voltage output, by a little bit. No heat may also be related to longer run times or duty cycles, as there is less waisted energy.
NZ

Wow Nic, great idea. I will have to try that, how about a small resistor instead.

Unfortunately today I will not be able to test anything so maybe tomorrow.

Good luck.

[NickZ]

Guys:
Although I have not had problems with the Jt transistor overheating, I have notice that if I remove the led that is across the C and E rail, everything gets very, very hot. If I left it like that, the transistor would melt down. But if I use one low voltage 2 volt red Led bulb there (between the C and E rail), there is absolutely no heat. The point being that it may be important to include at least one led between the rails, even at the expence of it lowering the secondary voltage output, by a little bit. No heat may also be related to longer run times or duty cycles, as there is less waisted energy.
NZ

[Vaporizer]

@ Tectalabyss,
I see. That point (E-C) is normally 1 LED or neon to show the oscillator is running, nothing more. Many remove it once a stable secondary load is established to reduce primary current draw from the battery. The paralleled LEDs there are lit by the feedback signal, but are also still on the source battery side using battery current too. Most use a separate winding(pickup) on the toroid to run them. This way all they see is the voltage developed by the collapsing magnetic field of the oscillator, never actually connected to the battery side, and totally isolated.
That should help your time run. Lidmotor has a good example of that with Jennas light vid. He has a 3" toroid running a CFL. The JT is only ringing the toroid, everything else is picked up from the tertiary(3rd) winding. He has room in that vid for many more separate windings too. Its pretty impressive to watch when you think of what its doing. I'll find the link if you need me too.

Edit: Actually, the one LED/neon also protects the transistor from HV spikes till a load is used to do more than just a simple JT circuit. Needed to clarify so ppl didnt start blowing transistors!

[Tectalabyss]

Run time's

Hi Tectalabyss and thanks for the welcome. 2 weeks? Are you talking abt the LED thats across the E-C junction? On the feedback, are you trying to extend the run time using it? I've read so much I need to go back through my folders I saved and reread a lot. If I remember right, most say you can't refeed the voltage back into the source battery. It just doesn't work. It needs to feed a separate charging battery. I know the Bedini circuit is this way so I may be confusing them.

I have gotten hung up on using the disposable camera circuits I had , which are basically JT's, to run a 9w CFL. A 1.25v AA I can get abt 10hrs from it at maybe 1/4 brightness. I thought it was 1/2 bright but now I dont think so. I upped it to 3v and saw very little diff other than it blew the transistor...
The coil is tiny. It will set inside the center of a goldmine toroid. I changed the transistor to a TIP series and back to 2.5v. It lit much brighter but was getting pretty warm. I looked and the Base resistor was 470ohm. I removed it and put a 1K variable in it. Interesting results. There are 2 settings that give the brightest light. One the transistor stays cool to touch and the other the transistor is hot. On the "cool" setting its still 2x as bright with 2.5v(compared to a single AA) and I'm getting a min of 10hrs on the 9w CFL using 2 AA 1.25v rechargeables. I tried winding a goildmine toroid with a 10:1 ratio and adding it to "kick" the camera coil up. It just doesnt work. I tried isolating it with diodes, caps...nada. Its like the last coil wants to much. If I would wind it with thinner wire so it had more resistance it may work and a smaller core too. The camera output coil is abt 390 ohms.

I have a pic saved I need to find where someone series 2 of them I think. Might try that, or mount it as is and move on. I have 2 3" new toroids I ordered from ebay and 10 2 3/8" that still haven't arrived.

Nick is right. Unless you gut the CFL it will not work. Mine had what looked like a neon in it with a cap.

pentiger, nice vid and that's a lot of light. I cant say if that is normal or not as I'm not quite to that next step yet. You might try the variable base resistor to see if you are using to much to "trip" the transistor. I did read that most use 350-450 turn "pickup" windings. More than that , little increase was noticed.

Hello Vaporizer and all.
That's with a small secondary running to a feedback circuit running 10 5mm white leds in parallel across the e c junction by the second week they were getting dim so I didn't post anything about it. Tec

[SkyWatcher]

Hi folks, Hi pentiger, Ok thanks for the coil turn specs. So as long as there are no short circuits in any of the coils, the fast voltage dip could be because D cell alkalines cannot carry much amp compared to rechargeables. If you look on the web, you'll find with certain amp draw, D cell alkalines will give a drastic reduction in there capacity. That could be part of the reason. Also the JT circuit can heat up the transistor with 2 cells or more, though the heating can be reduced using only the flyback with diode. What this means to me is that even if one adjusts coil resistance and base values the JT circuit does not drive the transistor properly when voltage increases, that's why I mostly use an oscillator similar to bedinis, it's easier to keep heat out of transistor and gives better output.
peace love light
Tyson

[pentiger]

@vaporizer
Thanks, I wasn't sure if 360 turns is enough. Today I tried changing resistors, ( I need to get myself a rheostat or two) and it's strange - I tested the light with modified CFL and 64 leds and both bulbs need different values. The rheostat would solve the problem of adjusting the circuit for a specific bulb. My goal is to set it up in a box with a regular 110v outlet and a switch so I can connect a night lamp with a modified bulb.

@all
I am using the lidmotor's circuit with 3 and 13 turns on the primary, But I am thinking this my be wrong for this toroid. I will try 5 and 5 and maybe other combinations. If you guys know what influence the number of each turns has on the circuit, please post.

Good luck with you tests.

[Vaporizer]

I apologize for the long posts. I'll try to do better.

I did forget to mention one other strange effect I found. When running on a singe AA 1.25v, I could stand a small medium strength neo magnet on top of the camera coil and I could dim or make it a little brighter than without it by rotating it. My guess is the coil is just below saturation and the magnet, when in the right position, adds a little extra to the cores field that is collapsing with it increasing the output.

This lead me to yet another thought I want to look into. There has to be a formula somewhere for calculating the max saturation mass of a given core size. One that would indicate the gauge wire vs voltage, turns or resistance needed...ect...... I have never seen this posted.

[Vaporizer]

Hi Tectalabyss and thanks for the welcome. 2 weeks? Are you talking abt the LED thats across the E-C junction? On the feedback, are you trying to extend the run time using it? I've read so much I need to go back through my folders I saved and reread a lot. If I remember right, most say you can't refeed the voltage back into the source battery. It just doesn't work. It needs to feed a separate charging battery. I know the Bedini circuit is this way so I may be confusing them.

I have gotten hung up on using the disposable camera circuits I had , which are basically JT's, to run a 9w CFL. A 1.25v AA I can get abt 10hrs from it at maybe 1/4 brightness. I thought it was 1/2 bright but now I dont think so. I upped it to 3v and saw very little diff other than it blew the transistor...
The coil is tiny. It will set inside the center of a goldmine toroid. I changed the transistor to a TIP series and back to 2.5v. It lit much brighter but was getting pretty warm. I looked and the Base resistor was 470ohm. I removed it and put a 1K variable in it. Interesting results. There are 2 settings that give the brightest light. One the transistor stays cool to touch and the other the transistor is hot. On the "cool" setting its still 2x as bright with 2.5v(compared to a single AA) and I'm getting a min of 10hrs on the 9w CFL using 2 AA 1.25v rechargeables. I tried winding a goildmine toroid with a 10:1 ratio and adding it to "kick" the camera coil up. It just doesnt work. I tried isolating it with diodes, caps...nada. Its like the last coil wants to much. If I would wind it with thinner wire so it had more resistance it may work and a smaller core too. The camera output coil is abt 390 ohms.

I have a pic saved I need to find where someone series 2 of them I think. Might try that, or mount it as is and move on. I have 2 3" new toroids I ordered from ebay and 10 2 3/8" that still haven't arrived.

Nick is right. Unless you gut the CFL it will not work. Mine had what looked like a neon in it with a cap.

pentiger, nice vid and that's a lot of light. I cant say if that is normal or not as I'm not quite to that next step yet. You might try the variable base resistor to see if you are using to much to "trip" the transistor. I did read that most use 350-450 turn "pickup" windings. More than that , little increase was noticed.

[SkyWatcher]

Hi folks, Hi pentiger, have you checked amp draw and can you check resistance of your bifilar, it is possible you have a short to core. Other than that, can you post exactly what circuit and parts and specs your using.
peace love light
Tyson

[pentiger]

Hello All.

I have a small question how long are the run time's with a 1.5 volt AA battery JT setup.I have gotten them to work but never have gotten the run times to more than two weeks and that was with a feedback circuit that didn't work that well. Tec

Two weeks? What load? Mine dies after couple hours.

You are lucky