RimStar, to a Hobbyist such as me, you are a RockStar! Your videos are packed with fast paced information and lots of it. I am shocked that your channel does not have a million followers!
Thanks and you're welcome. I taught for a software company for 15 years as well as various other smaller teaching jobs over the years. It's an addiction. :)
Thanks Todd! It was definitely better thanks to your input. That whole analysis bit added to the solar cell part was your suggestion, and that's one of the best bits! The clickable annotations at the end aren't working at the moment, but hopefully when they do you'll get more visitors for your great videos. There's a link in the description too. Thanks again, my friend.
It's the same as for an NPN transistor, the only thing that changes in the circuit is to reverse the battery. I'm also guessing you mean an LED, but I've included enough info in case you mean a regular diode. The short leg of the LED/the negative/the cathode/the end with the stripe goes to the transistor's emitter. The other end goes to the transistor's collector.
Thanks, and thanks for watching. Thinking back, I don't think I've ever seen them in things I've scavenged. I do see those flat, square, black types attached to heatsinks a lot, but I don't know their specs having never worked with them. Being attached to heatsinks, I imagine they have to handle power too. May be worth looking into.
I don't know if silicon solar cells make use of infrared though I suspect not, or not much since that's one of the advantages of multijunction gallium arsenide solar cells. And thanks man! Glad you like the video.
Thanks! I do take suggestions. e.g. This video wasn't even in the plans until I had so many questions and suggestions from my previous video. But unless there's a very large temperature difference and many thermocouples I suspect, there won't be enough power to run this circuit, even with the low power transistor version. Maybe a themocouple powered by concentrated solar sunlight would work though. The Voyage RTG used a hot-cold temperature difference of 700C or 1260F.
Thanks Dan! It's a good thing I'm a regular commenter on your wonderful videos otherwise you'd have forgotten me! :) Thanks for subbing and welcome! Hopefully I'll keep you as interested as your channel has with me.
I was thinking of using little power such as a few batteries to drive high voltage AC appliance (I actually want to make a fume extractor) now you point me to a shortcut. Thanks. It's really surprising to me though, that the "spikey power" can drive CFL, I need to investigate why this will work for CFL, but not some other devices.
Thanks. And I looked for the Ah but they never seem to include them, not even on the packaging. As for the turns, I wrapped them myself so sure. It's the same coil from last week's video. 256 turns of 30 AWG, then 10 turns of 24 on top of that, followed by 5 turns of 24 on top of that (I mistakenly said 18 AWG for the two 24 AWGs in the video.)
Been wating all week for this LOL Still have some unanswered questions but all in all it was a very good video with a lot of info, ideas, and examples.
But it caused interference in the AM band which is 550 kHz plus, though with decreasing strength as it got further up into the shortwave band which is above the normal AM. So it does make sense that by the time it got up to FM, the interference was gone. It also filled the whole AM band, so I'm thinking it wasn't just harmonics.
Nice video! This circuit is quite fascinating! Have you thought about making one that could do useful work from a small voltage source like a thermocouple (or several of thm)? It would be fun to see if you could light an LED or power a microcontroller from just room temperature. It would be a cool way to demonstrate how the RTGs the power things like Voyager or the Curiosity Rover work! Not sure if you take suggestions... just a thought.
I don't know any temperature ranges. I just felt and if it was too hot to touch after a second or so I figured it was getting hot and I probably didn't want it getting hotter.
Great test demo. I think I'm getting better results with pnp's too. Especially when driving a 3rd flyback. I haven't done as many as npn type though, so results are pretty patchy. hour and a half is pretty decent for 2 AA's I think. Any ideas on the Ah of the aa's and the turn specs on that core?
Hi I am not good at English. I tried to follow you but I move the negative to the positive terminal of the power supply and I added a diode , transistor(emitter) .It Very good results.
I use Mosfet same as NPN transister.I have used chargers battery but it surprise to input and output not equal it is input 12V 0.12A. It is out put 100V 0.04A. it mean output 4W more input 1.44W.it may be caused Black EMF. It similar Bedini Motor.
+Electronic Experiments Little modify. Transistor use current but mosfet use voltage. plus.google.com/u/0/102841827576367305980/posts/fT9k9cYKcWr?pid=6206054606062395042&oid=102841827576367305980
Wow, very, very, very good, I don't even know if I could ask for better. Congratulations. Love for this video. I'll copy this video. One question, does the oscilloscope not be damaged with a voltage peak above 480 volts?
The 2N3055 transistor can work at 60 volts and 15 amperes maximum. You may give a try to operate your cirquit at these values by replacing the thin wire of the secondary coil with another one so you can make your CFL brighter.
Good point. At 2:23 in the video the peak voltage across the emitter and collector is 38.8 volts, though I don't know the current. It looks like there's room here to get the voltage higher on the input side and reduce the turns on the output coil. Though I didn't adjust the potentiometer during this measurement.
Question about the heat? You said in the video that the resister and the batteries where hot. 1 how much heat is considered bad? The reason I ask is that most people say something is hot when the actual temperature is around 110 or 120 degrees. Any hotter we will start to be burnt. I had a emergency call to repair an eddy current brake on an overhead bridge crane. (it keeps the load from free falling to the ground) The maintenance person said the eddy brake... (see next post)
nice!thanks for posting...i am still gathering the stuff i need...do you know where i can get those big npn and pnp transistors...i will have to scavengethem out of somewhere..radioshack and other stuff dont work here..
I have a question related to using the solar cell to measure the light output. Having a difficult time accounting for the negative-going pulses at the solar-cell output. Is it possible that there is inductive or capacitive coupling between the lamp (and associated wiring) and the solar-cell causing this, rather than the cell's output itself? It could be tested by increasing the distance between them and/or shielding all but a small aperture through a metal enclosure that a bit of light passes
The joule thief output isn't DC. In this video you can see it's very spikey and the power is during those spikes. So it wouldn't work. If you're already powering your joule thief with a DC power source, why not just connect the DC to AC transformer to that source with no joule thief? Adding the joule thief just adds losses. All a joule thief does is create periodic high voltage spikes for things that need a higher voltage than the original power source and that can work with spikey power.
I doubt it very much. A motor like in a 220v AC fan needs an electrical current that varies according to a sine wave. The waveform from a joule thief is no where near enough to that to make it turn.
Yes, the resistance is at least the parameter you can easily change if you use a potentiometer. You can see it happen a little at 3:14 in the video. Though if you could vary the supply voltage than that would too. I'm guessing the number of turns on the coil would too but that's not easy to adjust dynamically.
DEBASISH DUTTA 92v is very low. Are you sure the insulation to the wires on your coils didn't break down and are conducting through corona? That's what happened to me and caused me to use thicker insulation wherever there would be high voltage between adjacent wires. I talk about this at 2:10 and 3:10 in my video "How to Make Joule Thief Light a CFL - Jeanna's Light". Also, recheck the way you connected the coils to the rest of your circuit. The connections are backward from what you'd expect.
Glad to see that the light works with a 9 volt (with an extra resistor) and that the pulsed DC can easily substitute AC within the toroid system. I might actually take up the project to build the CFL lantern to run off a 9 volt. One question that comes to mind is could I substitute the magnet wire in the toroid with a 1 mm, insulated telephone wire stripped from a telephone cable?
RimstarOrg Okay, a little update on my progress on making this circuit. I acquired a 2N3055 transistor for the circuit; then, I was able to go get a monster toroid (goo.gl/WNuJo2) with 61 mm outer diameter for the 1 mm insulated telephone wire. Now, I am getting a potentiometer and maybe an extra resistor. Do you have any tips on improving the design? Such as, a resistor on the positive in combination with a lower ohm rated potentiometer or wind more with the secondary and use electrician's tape for additional insulation. I do enjoy these videos and design assistance since I will be doing a research project soon with different battery types and brands to see which one might last the longest in terms of battery brand per type without too much compensation in terms of lighting with the CFL Joule Thief Circuit.
Destroyer9747 Nice toriod. With that you should be able to put plenty of turns without the two ends coming close enough together for the need to insulate the two ends from each other.
RimstarOrg Thanks, I am glad that the toroid is sufficient enough for the secondary. I might have 450-550 on the secondary if I can fit that many without too much amp draw for the circuit or overloading the bulb (if possible).
I know what a newman motor is but none of the details, so I can't say whether a joule thief will run one. Keep in mind that the tolal energy put out by a joule thief is less than the total energy in. It isn't overunity.
Thanks. But an LED is cheating! :) There's something just so cool about holding a lit, cordless CFL in the palm of your hand. That thumbnail for my previous video reminds me of the photo of Tesla seated, holding a lit lightbulb (not claiming to be like the great man, of course.) Rechargeables would be nice. You should see the stack of non-rechargeable AAs I have here from making these videos. :(
Will the efficacy / efficiency of the whole system (with the 5W CFL) be anywhere close to comparable to a modern LED setup given the same 2 AA batteries as the power source???
you could hack a hand crank flash light and store the power in rechargeable batteries and you could take a small water tight container and pour mineral oil in it and sink the transistor and keep it cool
No, I didn't. It would be difficult to do anyway since the output of a joule thief circuit is so spikey. It is a fairly inefficient circuit though, so the output watts would probably be much lower than the input watts.
@@RimstarOrg adding appropriate capacitor parallel to output coil will make the output smoother and i believe this will make it easier for measurements. i found online that if three coils are wound on the toroid (sensing coil, induction coil, secondary coil), the secondary coil will generate good current and voltage and won't make joule thief draw more current from source during on load conditions. this way the output might surpass input.
There's no reason why they should think it's a fake. It doesn't put out more power than it uses. And it's an old circuit, also called a blocking oscillator. Joule thief is a misleading name.
Sir i need to ask a question, i have an experiment that combining newman motor to drive 2 mini axial flux generator. The question is can i use joule thief dc to dc booster from second generator coild (which is output 3 volt dc (at 1000 rpm) --> 70 / 100 volt dc) to feed the newman motor?. so the machine keep running it self?. I will be waiting to hear from you. Best Regards.
I've never tried, so I don't know how less efficient it would be due to the induction. You could also stack washers with nonconductive spacers in between them, though you'd still have eddy currents in each washer.
The only path left from the original CFL is the tube but I don't know if the tube will last less time than if it would have if it was still used as part of the original CFL. The joule thief circuit itself is not a very efficient circuit to start with you you'll definitely be using more power for that same amount of light.
+rimstarOrg, I can't reply directly. So if I add a DC to AC transformer to joule thief circuit as a bridge linking to a 220v AC fan, do you think it has a chance to work? I'm very inspired by your video and website, thank you.
I am making a school project and I need your help. What is the output frequency in your normal joule thief(which runs an LED) if you use fresh batteries(3.0 volts). Also give information about output current. Does a normal multimeter show the same accurate results as that of an oscilloscope? Thanks in advance!
I've never tried a normal LED joule thief with fresh batteries. You can see what I have tried here ua-cam.com/video/B61DU7yEsPM/v-deo.html in my video about making one. The frequency as you can see around forty-five seconds into that video is around 21 kHz (around 46 usec per cycle) using a 1 kilohm resistor. I don't know how to measure output current since the output is nowhere near flat DC nor sinusoidal AC. A multimeter isn't as good as an oscilloscope for this since a multimeter is designed for flat DC and sinusoidal AC and doesn't show the waveform. Some multimeters have min and max feature where they can show minimum and maximum values so that helps, otherwise you're just looking at values from random points in time.
RimstarOrg Can you try using a full wave rectifier and high voltage capacitors to see if you get an almost flat DC output. I am making a 'Railgun' for Physics project, therefore I need high voltage power supple (around 300-400 volts) to charge my capacitor bank(around 5000 microFarads). Perhaps those frequent pulses wont be able to charge the capacitors as fast.
I do have a page on my website contibuted by someone else who turned the output of a joule thief into DC for making a clock work rimstar.org/science_electronics_projects/joule_thief_powering_clock_ls.htm but it's the low voltage version. It might be possible for high voltage, but it may also be that when you do the smoothing you'll find that it smooths to too low a voltage. I can't say for sure. I also don't know how long it would take to charge your capacitor bank, so it may or may not be a while between firings. It's too hard to me to say anything for sure with the joule thief's type of output.
I couldn't find any audio transformers in our local stores. But that could just be our stores. Yours might be better. There are probably many sources online but one I know of is www.scitoyscatalog.com/category/R.html I'm not sure if it qualifies as high power.
Than Zamp Oh! The "audio" in front of transistor thru me off. I've never heard them called audio transistors so I've never looked for them. I buy my power transistors at local electronics stores. One store I go to has only NTE 130, which is equivalent to the 2N3055. The other store has 2N3055 and others but I don't remember the part numbers.
RimstarOrg, Thanks for the info about your transistors. I just want to know if any type of high power transistor will work for a joule thief experiment, because I used a lower power transistor taken form a damaged CD-R device, I managed to use it succesfully as amplifier, but in my joule thief experiment with a CCFL transformer *(from a LCD screen)* wasn't working no matter the polarity I used. The strange thing: it is not damaged, because I tested it with my digital multimeter.
Than Zamp I honestly can't say which ones won't work (well, except for very small ones like 2N4401), only the ones I've mention which I know have worked.
double the turns on the primary with thinner awg , and double the turns on the secondary with thicker awg, 1 awg difference is good , this way use less power and still get high voltage lower frequency may happen not sure lmk
I have question to how did you turn the toroidal transformer?? it how be a like a normal transformer turn like primary will be 9 turns and second will be 23 turns to make 23 volt?? I don't get much.
+talastas789 I wound the toroidal transformer by hand. You can see it in the video where I show how it was made ua-cam.com/video/FkLET8MhRbU/v-deo.html.
+RimstarOrg I get it!!! its like a real winding a toroidal one!! but when primary will be much low turn to meet a 9 to 10 input right?? so yeah I did get it !! thank for it!!!
+talastas789 I don't understand your question about the 9 to 10 input. What are you trying to achieve? This will not put out 23 volts DC. A joule thief circuit puts out a voltage that is very spikey, as you can see in some of the oscilloscope clips in the video. If you are looking for how to calculate the number of turns needed for a transformer here's a webpage that seems to cover it www.sayedsaad.com/fundmental/21_TURNS%20AND%20VOLTAGE%20RATIOS%20%20.htm
Maintenance stated they used their hand and had burnt themselves. The normal operating temp of an eddy current brake is 1200 degrees So yeah your going to get burnt bad. I just was wondering how some heat verses a lot of heat effects battery life and current.
sir i have 12V 7.8Ah old desktop UPS Battery with good backup i want to use it for light CFL at least 3 hours continous backup needed using joule thief method . could you please suggest me what kind of material i needed. what rating of register etc .....
sameer khaan I'd go with a power transistor like in this video and as for the resistor, I'd put in a potentiometer (variable resistor) and adjust it, also like I did in this video. That's the only way I know with this circuit to find out what resistance to use.
I made a CFL light that runs off 2 C batteries and is powered by a modified camera flash circuit. Runtime seems to be a lot and it's bright enough to see in the dark. the batteries inside seem to be good still. Here's a video. ua-cam.com/video/GvVi5zd57c4/v-deo.html
Jozsef Ienciu It'll depend on what size wire you use and the dimensions of your toroid core. In my case I got 256 turns. See this video of mine showing details of how I made it, including winding the coils "How to Make Joule Thief Light a CFL - Jeanna's Light".
The spark is probably a little weak with these specific components but this fly swatter (ua-cam.com/video/sQCKNZMNCo0/v-deo.html) is the same basic circuit with some parts added and can be used as a taser.
I'm not sure you'd get a shock, or at least feel the shock, when using the specific components used in this CFL joule thief circuit. But you'd definitely feel the shock from the fly swatter joule thief circuit. It'd be a painful, though harmless, shock (unless you have a heart condition).
Oh no I didn't mean to sound like that hehe. I am not a true expert either , I have read lots and lots about tesla coils though, I think I am 99,9% sure, I think I am. The spike positive spike is the difference. at any rate, My goal is just a continues wave SSTC. Thans
How many watts is the CFL that you started with? The longer the coil of the CFL, the higher the voltage you'll need. Maybe that's your problem. You could also try connecting more batteries in series to give it a higher voltage to start with. And the only other suggestion I can think of is to try a lower resistance resistor.
You'd have to at least wind a new coil for the zapper and possibly replace the resistor. That's assuming your zapper is using the same circuit that was used in a zapper I once took apart.
you helped me lot iam so lucky find you one last thing what about induction module output can you make a vid i think cfl can glow damn brighter then any of these so higher frequency if it douse i will order induction module
RimStar, to a Hobbyist such as me, you are a RockStar! Your videos are packed with fast paced information and lots of it. I am shocked that your channel does not have a million followers!
Thanks Ronald! I'm glad you enjoy my videos. Thanks for watching them!
Thanks and you're welcome. I taught for a software company for 15 years as well as various other smaller teaching jobs over the years. It's an addiction. :)
You're welcome! It's my pleasure making the videos. Thanks for watching and welcome to my channel!
My pleasure. Thanks for watching.
Glad you like it. Thanks for watching.
Thanks Todd! It was definitely better thanks to your input. That whole analysis bit added to the solar cell part was your suggestion, and that's one of the best bits! The clickable annotations at the end aren't working at the moment, but hopefully when they do you'll get more visitors for your great videos. There's a link in the description too. Thanks again, my friend.
Thanks! And thanks for watching.
Glad you liked it! Thanks for watching.
It's the same as for an NPN transistor, the only thing that changes in the circuit is to reverse the battery. I'm also guessing you mean an LED, but I've included enough info in case you mean a regular diode. The short leg of the LED/the negative/the cathode/the end with the stripe goes to the transistor's emitter. The other end goes to the transistor's collector.
Thanks! I'll take awesome any day. Glad you liked it.
Thanks, and thanks for watching. Thinking back, I don't think I've ever seen them in things I've scavenged. I do see those flat, square, black types attached to heatsinks a lot, but I don't know their specs having never worked with them. Being attached to heatsinks, I imagine they have to handle power too. May be worth looking into.
Thanks! If I recall it can handle up to 1000 volts. It also helps that they're spaced out spikes.
Thanks. And I'll take your word for it. I'm no tesla coil expert, though I'd like to make one at some point.
I don't know if silicon solar cells make use of infrared though I suspect not, or not much since that's one of the advantages of multijunction gallium arsenide solar cells.
And thanks man! Glad you like the video.
Thanks! I do take suggestions. e.g. This video wasn't even in the plans until I had so many questions and suggestions from my previous video. But unless there's a very large temperature difference and many thermocouples I suspect, there won't be enough power to run this circuit, even with the low power transistor version. Maybe a themocouple powered by concentrated solar sunlight would work though. The Voyage RTG used a hot-cold temperature difference of 700C or 1260F.
I agree! The look like magic.
igreat video! incredible that the scope wasn't damaged by the 700 volt input
Thanks. Glad to hear is was worth the wait.
Thanks Dan! It's a good thing I'm a regular commenter on your wonderful videos otherwise you'd have forgotten me! :) Thanks for subbing and welcome! Hopefully I'll keep you as interested as your channel has with me.
Thanks. Since you replied to me, simontay1984 may not get notified so I replied to him to see your comment.
That would be neat. It'd still have to use a compact fluorescent light since LED flashlights already exist so it wouldn't look any different.
I was thinking of using little power such as a few batteries to drive high voltage AC appliance (I actually want to make a fume extractor) now you point me to a shortcut. Thanks. It's really surprising to me though, that the "spikey power" can drive CFL, I need to investigate why this will work for CFL, but not some other devices.
Good set of tests, very thorough. A 1W LED mains bulb would seem ideal for this circuit...and rechargeables.
Thanks. And I looked for the Ah but they never seem to include them, not even on the packaging. As for the turns, I wrapped them myself so sure. It's the same coil from last week's video. 256 turns of 30 AWG, then 10 turns of 24 on top of that, followed by 5 turns of 24 on top of that (I mistakenly said 18 AWG for the two 24 AWGs in the video.)
But you excelled at music!
Been wating all week for this LOL
Still have some unanswered questions but all in all it was a very good video with a lot of info, ideas, and examples.
But it caused interference in the AM band which is 550 kHz plus, though with decreasing strength as it got further up into the shortwave band which is above the normal AM. So it does make sense that by the time it got up to FM, the interference was gone. It also filled the whole AM band, so I'm thinking it wasn't just harmonics.
Thanks. I'll give it a try.
This with the magnets is a f****** good idea . I use it now .
Nice video! This circuit is quite fascinating! Have you thought about making one that could do useful work from a small voltage source like a thermocouple (or several of thm)? It would be fun to see if you could light an LED or power a microcontroller from just room temperature. It would be a cool way to demonstrate how the RTGs the power things like Voyager or the Curiosity Rover work! Not sure if you take suggestions... just a thought.
A foil cap? What exactly do you mean? What capacitance? Can you clarify? Why am I speaking in questions? :)
I don't know any temperature ranges. I just felt and if it was too hot to touch after a second or so I figured it was getting hot and I probably didn't want it getting hotter.
Excellent Video and tests Master. Keep up the great work!
those joule thief are awsome.
Interresting as always ,thanks !
Great test demo. I think I'm getting better results with pnp's too. Especially when driving a 3rd flyback. I haven't done as many as npn type though, so results are pretty patchy. hour and a half is pretty decent for 2 AA's I think. Any ideas on the Ah of the aa's and the turn specs on that core?
Hi I am not good at English. I tried to follow you but I move the negative to the positive terminal of the power supply and I added a diode , transistor(emitter) .It Very good results.
+Kasamepisut Susiritumrong I'm glad to hear it worked. It sounds like you have a PNP transistor.
I use Mosfet same as NPN transister.I have used chargers battery but it surprise to input and output not equal it is input 12V 0.12A. It is out put 100V 0.04A. it mean output 4W more input 1.44W.it may be caused Black EMF. It similar Bedini Motor.
+Electronic Experiments Little modify. Transistor use current but mosfet use voltage.
plus.google.com/u/0/102841827576367305980/posts/fT9k9cYKcWr?pid=6206054606062395042&oid=102841827576367305980
The magnets are a neat trick.
Excellent Video, thanks.
Wow, very, very, very good, I don't even know if I could ask for better. Congratulations. Love for this video. I'll copy this video. One question, does the oscilloscope not be damaged with a voltage peak above 480 volts?
Thanks. I'm glad you like it. Thanks for watching. Re your question, my oscilloscope is rated for 1000 volts.
The 2N3055 transistor can work at 60 volts and 15 amperes maximum. You may give a try to operate your cirquit at these values by replacing the thin wire of the secondary coil with another one so you can make your CFL brighter.
Good point. At 2:23 in the video the peak voltage across the emitter and collector is 38.8 volts, though I don't know the current. It looks like there's room here to get the voltage higher on the input side and reduce the turns on the output coil. Though I didn't adjust the potentiometer during this measurement.
Question about the heat? You said in the video that the resister and the batteries where hot. 1 how much heat is considered bad? The reason I ask is that most people say something is hot when the actual temperature is around 110 or 120 degrees. Any hotter we will start to be burnt.
I had a emergency call to repair an eddy current brake on an overhead bridge crane. (it keeps the load from free falling to the ground) The maintenance person said the eddy brake... (see next post)
Thanks.
nice!thanks for posting...i am still gathering the stuff i need...do you know where i can get those big npn and pnp transistors...i will have to scavengethem out of somewhere..radioshack and other stuff dont work here..
I have a question related to using the solar cell to measure the light output. Having a difficult time accounting for the negative-going pulses at the solar-cell output. Is it possible that there is inductive or capacitive coupling between the lamp (and associated wiring) and the solar-cell causing this, rather than the cell's output itself?
It could be tested by increasing the distance between them and/or shielding all but a small aperture through a metal enclosure that a bit of light passes
The joule thief output isn't DC. In this video you can see it's very spikey and the power is during those spikes. So it wouldn't work. If you're already powering your joule thief with a DC power source, why not just connect the DC to AC transformer to that source with no joule thief? Adding the joule thief just adds losses. All a joule thief does is create periodic high voltage spikes for things that need a higher voltage than the original power source and that can work with spikey power.
I doubt it very much. A motor like in a 220v AC fan needs an electrical current that varies according to a sine wave. The waveform from a joule thief is no where near enough to that to make it turn.
which parameter affect the frequency of operation?? Is it the resistor ??
Yes, the resistance is at least the parameter you can easily change if you use a potentiometer. You can see it happen a little at 3:14 in the video. Though if you could vary the supply voltage than that would too. I'm guessing the number of turns on the coil would too but that's not easy to adjust dynamically.
Thanks..I have tried to light the cfl but still not done..
Getting maximum 92v peak to peak at out put terminal..
I have used 275 turns.
DEBASISH DUTTA
92v is very low. Are you sure the insulation to the wires on your coils didn't break down and are conducting through corona? That's what happened to me and caused me to use thicker insulation wherever there would be high voltage between adjacent wires. I talk about this at 2:10 and 3:10 in my video "How to Make Joule Thief Light a CFL - Jeanna's Light". Also, recheck the way you connected the coils to the rest of your circuit. The connections are backward from what you'd expect.
OK let me try again ..
You can try the tiny 2N7000 MOSFET which costs a dime at Mouser or Futurlec, but it requires 3V gate voltage.
Add a foil cap across the leads of the battery. That will improve the performance allot
Glad to see that the light works with a 9 volt (with an extra resistor) and that the pulsed DC can easily substitute AC within the toroid system. I might actually take up the project to build the CFL lantern to run off a 9 volt. One question that comes to mind is could I substitute the magnet wire in the toroid with a 1 mm, insulated telephone wire stripped from a telephone cable?
Destroyer9747 You can substitute 1mm wire but you need to fit a few hundred turns so you'll probably need a bigger toroid.
RimstarOrg Thanks, I'm glad that I can at least use 1mm telephone wire that I salvaged from a telephone cable.
RimstarOrg Okay, a little update on my progress on making this circuit. I acquired a 2N3055 transistor for the circuit; then, I was able to go get a monster toroid (goo.gl/WNuJo2) with 61 mm outer diameter for the 1 mm insulated telephone wire. Now, I am getting a potentiometer and maybe an extra resistor. Do you have any tips on improving the design? Such as, a resistor on the positive in combination with a lower ohm rated potentiometer or wind more with the secondary and use electrician's tape for additional insulation. I do enjoy these videos and design assistance since I will be doing a research project soon with different battery types and brands to see which one might last the longest in terms of battery brand per type without too much compensation in terms of lighting with the CFL Joule Thief Circuit.
Destroyer9747 Nice toriod. With that you should be able to put plenty of turns without the two ends coming close enough together for the need to insulate the two ends from each other.
RimstarOrg Thanks, I am glad that the toroid is sufficient enough for the secondary. I might have 450-550 on the secondary if I can fit that many without too much amp draw for the circuit or overloading the bulb (if possible).
Re MOSFETs, see Acme Fixer's suggestion in his comment below.
Great demo. Thanks
I wonder how long would stay light (6:38) if the solar cell would be aded to harvest back some of the energy from light?
I know what a newman motor is but none of the details, so I can't say whether a joule thief will run one. Keep in mind that the tolal energy put out by a joule thief is less than the total energy in. It isn't overunity.
That I wouldn't know. While I know what they are, I've yet to play around with MOSFETs.
Thanks for the Info on the joule thieves BTW. IT's a really gr8 instructional video. Are you a teacher?
Thanks. But an LED is cheating! :) There's something just so cool about holding a lit, cordless CFL in the palm of your hand. That thumbnail for my previous video reminds me of the photo of Tesla seated, holding a lit lightbulb (not claiming to be like the great man, of course.) Rechargeables would be nice. You should see the stack of non-rechargeable AAs I have here from making these videos. :(
Thank you a lot for knowledge.
Will the efficacy / efficiency of the whole system (with the 5W CFL) be anywhere close to comparable to a modern LED setup given the same 2 AA batteries as the power source???
I doubt it would be as efficient as a modern LED setup.
you could hack a hand crank flash light and store the power in rechargeable batteries and you could take a small water tight container and pour mineral oil in it and sink the transistor and keep it cool
did you also do input vs output watts calculation?
No, I didn't. It would be difficult to do anyway since the output of a joule thief circuit is so spikey. It is a fairly inefficient circuit though, so the output watts would probably be much lower than the input watts.
@@RimstarOrg adding appropriate capacitor parallel to output coil will make the output smoother and i believe this will make it easier for measurements.
i found online that if three coils are wound on the toroid (sensing coil, induction coil, secondary coil), the secondary coil will generate good current and voltage and won't make joule thief draw more current from source during on load conditions. this way the output might surpass input.
Have you done any measurements of efficiency ? Power in power out ? How lossy are these systems ?
Nice. So except for the very high positive first bang of 700v, the wave are damped ac just like out of a tesla coil
There's no reason why they should think it's a fake. It doesn't put out more power than it uses. And it's an old circuit, also called a blocking oscillator. Joule thief is a misleading name.
Just to be clear, the magnets aren't my idea. I've seen quite a few people use that trick. Good trick anyway, though.
GREAT VIDEO! Just added to my feed. Also, I just realized I had not been a subscriber! y!kes. SUBBED fav. great channel.
I just have seen it in your video ,now i don need batteryclips ;)
Sir i need to ask a question, i have an experiment that combining newman motor to drive 2 mini axial flux generator. The question is can i use joule thief dc to dc booster from second generator coild (which is output 3 volt dc (at 1000 rpm) --> 70 / 100 volt dc) to feed the newman motor?. so the machine keep running it self?. I will be waiting to hear from you. Best Regards.
Nice video!
Good video.
Can you use a washer for the winding ring, or does the induction in the washer ruin it
I've never tried, so I don't know how less efficient it would be due to the induction. You could also stack washers with nonconductive spacers in between them, though you'd still have eddy currents in each washer.
Will operating the CFL using this method reduce the CFL's life?
The only path left from the original CFL is the tube but I don't know if the tube will last less time than if it would have if it was still used as part of the original CFL. The joule thief circuit itself is not a very efficient circuit to start with you you'll definitely be using more power for that same amount of light.
Is it capable of driving a piezo?? With source voltage below 1v
+rimstarOrg, I can't reply directly. So if I add a DC to AC transformer to joule thief circuit as a bridge linking to a 220v AC fan, do you think it has a chance to work? I'm very inspired by your video and website, thank you.
I am making a school project and I need your help. What is the output frequency in your normal joule thief(which runs an LED) if you use fresh batteries(3.0 volts). Also give information about output current. Does a normal multimeter show the same accurate results as that of an oscilloscope? Thanks in advance!
I've never tried a normal LED joule thief with fresh batteries. You can see what I have tried here ua-cam.com/video/B61DU7yEsPM/v-deo.html in my video about making one. The frequency as you can see around forty-five seconds into that video is around 21 kHz (around 46 usec per cycle) using a 1 kilohm resistor. I don't know how to measure output current since the output is nowhere near flat DC nor sinusoidal AC. A multimeter isn't as good as an oscilloscope for this since a multimeter is designed for flat DC and sinusoidal AC and doesn't show the waveform. Some multimeters have min and max feature where they can show minimum and maximum values so that helps, otherwise you're just looking at values from random points in time.
RimstarOrg Can you try using a full wave rectifier and high voltage capacitors to see if you get an almost flat DC output. I am making a 'Railgun' for Physics project, therefore I need high voltage power supple (around 300-400 volts) to charge my capacitor bank(around 5000 microFarads). Perhaps those frequent pulses wont be able to charge the capacitors as fast.
I do have a page on my website contibuted by someone else who turned the output of a joule thief into DC for making a clock work
rimstar.org/science_electronics_projects/joule_thief_powering_clock_ls.htm but it's the low voltage version. It might be possible for high voltage, but it may also be that when you do the smoothing you'll find that it smooths to too low a voltage. I can't say for sure. I also don't know how long it would take to charge your capacitor bank, so it may or may not be a while between firings. It's too hard to me to say anything for sure with the joule thief's type of output.
awesome! it's what i can say.
I have a question: can I buy from a local store any type of high power audio transistor in case I can't find yours?
I couldn't find any audio transformers in our local stores. But that could just be our stores. Yours might be better. There are probably many sources online but one I know of is www.scitoyscatalog.com/category/R.html I'm not sure if it qualifies as high power.
RimstarOrg, my question was about transistors, not transformers.
Than Zamp
Oh! The "audio" in front of transistor thru me off. I've never heard them called audio transistors so I've never looked for them. I buy my power transistors at local electronics stores. One store I go to has only NTE 130, which is equivalent to the 2N3055. The other store has 2N3055 and others but I don't remember the part numbers.
RimstarOrg, Thanks for the info about your transistors. I just want to know if any type of high power transistor will work for a joule thief experiment, because I used a lower power transistor taken form a damaged CD-R device, I managed to use it succesfully as amplifier, but in my joule thief experiment with a CCFL transformer *(from a LCD screen)* wasn't working no matter the polarity I used. The strange thing: it is not damaged, because I tested it with my digital multimeter.
Than Zamp
I honestly can't say which ones won't work (well, except for very small ones like 2N4401), only the ones I've mention which I know have worked.
double the turns on the primary with thinner awg , and double the turns on the secondary with thicker awg, 1 awg difference is good , this way use less power and still get high voltage lower frequency may happen not sure lmk
can I use small signal transistor instead of power transistor?
You'll likely burn out the transistor.
so can I use transistor like tip31 or MOSFET like IRFZ44N
The TIP31 might work. Its ratings are a little lower than the 2N3055. MOSFETs probably require changes to the circuit.
Oh, and make sure you use a heat sink for the transistor, otherwise it may heat up very fast and not last long.
Is it possible and how to use joule thief to run a 220v 1.4A fan?
Is there any way to reduce the eneryloss due to heat coming from the components? I think that at least 50% of the energy was wasted in the components.
If you're concerned about energy loss then you might want to start by using a different circuit. The joule thief circuit is horribly inefficient.
I always thought that the joule thief is very efficient at least when powering led`s. Maybe im just gonna buy a 12v cfl and look what`s inside.
Excellent.
I have question to how did you turn the toroidal transformer?? it how be a like a normal transformer turn like primary will be 9 turns and second will be 23 turns to make 23 volt?? I don't get much.
+talastas789 I wound the toroidal transformer by hand. You can see it in the video where I show how it was made ua-cam.com/video/FkLET8MhRbU/v-deo.html.
+RimstarOrg I get it!!! its like a real winding a toroidal one!! but when primary will be much low turn to meet a 9 to 10 input right?? so yeah I did get it !! thank for it!!!
+talastas789 I don't understand your question about the 9 to 10 input. What are you trying to achieve? This will not put out 23 volts DC. A joule thief circuit puts out a voltage that is very spikey, as you can see in some of the oscilloscope clips in the video. If you are looking for how to calculate the number of turns needed for a transformer here's a webpage that seems to cover it www.sayedsaad.com/fundmental/21_TURNS%20AND%20VOLTAGE%20RATIOS%20%20.htm
Maintenance stated they used their hand and had burnt themselves. The normal operating temp of an eddy current brake is 1200 degrees So yeah your going to get burnt bad. I just was wondering how some heat verses a lot of heat effects battery life and current.
sir i have 12V 7.8Ah old desktop UPS Battery with good backup i want to use it for light CFL at least 3 hours continous backup needed using joule thief method . could you please suggest me what kind of material i needed. what rating of register etc .....
sameer khaan I'd go with a power transistor like in this video and as for the resistor, I'd put in a potentiometer (variable resistor) and adjust it, also like I did in this video. That's the only way I know with this circuit to find out what resistance to use.
sir can i use speakers magnet in joule thief method ...............???
sameer khaan Possibly. I haven't tried it. The permanent magnetic field may interfere.
I made a CFL light that runs off 2 C batteries and is powered by a modified camera flash circuit. Runtime seems to be a lot and it's bright enough to see in the dark. the batteries inside seem to be good still. Here's a video. ua-cam.com/video/GvVi5zd57c4/v-deo.html
is pnp transistor gives same efficiency as npn transistor?
+Diwakar Dey I don't know. But off-hand I don't see why there would be any difference in efficiency.
+RimstarOrg ok and thx
Where.
Hi how many turn you make on the toroid thanks.
Jozsef Ienciu It'll depend on what size wire you use and the dimensions of your toroid core. In my case I got 256 turns. See this video of mine showing details of how I made it, including winding the coils "How to Make Joule Thief Light a CFL - Jeanna's Light".
at the end, where the tube is
haha :). I knew it, You are a great teacher!
/watch?v=E3TK7oR2YEE
I Thought it looked a bit like a joule thief, I'm sure it will interrest you!!
Nice
can I make a Taser using this??
The spark is probably a little weak with these specific components but this fly swatter (ua-cam.com/video/sQCKNZMNCo0/v-deo.html) is the same basic circuit with some parts added and can be used as a taser.
ok so if I touch the output will I get shocked??
I'm not sure you'd get a shock, or at least feel the shock, when using the specific components used in this CFL joule thief circuit. But you'd definitely feel the shock from the fly swatter joule thief circuit. It'd be a painful, though harmless, shock (unless you have a heart condition).
Oh no I didn't mean to sound like that hehe. I am not a true expert either , I have read lots and lots about tesla coils though, I think I am 99,9% sure, I think I am. The spike positive spike is the difference. at any rate, My goal is just a continues wave SSTC. Thans
i going to make one like this and i want it to make a lithium ion battery and usb charger for it， heat sink going to be add
mine is not working attempt 2 times its failed like in reality its not possible. high voltage turns all ok transistor all fine its like impossible
How many watts is the CFL that you started with? The longer the coil of the CFL, the higher the voltage you'll need. Maybe that's your problem. You could also try connecting more batteries in series to give it a higher voltage to start with. And the only other suggestion I can think of is to try a lower resistance resistor.
by the way iam working on mosquito zapper lower frequency like 9khz not enough can you help me out how can i get 30khz on that zapper
You'd have to at least wind a new coil for the zapper and possibly replace the resistor. That's assuming your zapper is using the same circuit that was used in a zapper I once took apart.
you helped me lot iam so lucky find you one last thing what about induction module output can you make a vid i think cfl can glow damn brighter then any of these so higher frequency if it douse i will order induction module
I don't know what you mean by an induction module.
i'm sorry, film capacitor not foil capacitor.
add a capacitor