Yep, no transformer and no hard to get IC. But, ... two transistors and other parts are required. I actually found this circuit by accident while roaming Colin Mitchell's Talking Electronics website (talkingelectronics.com). This website is very rich with examples and explanations of a huge variety of circuits. In fact, this website is so abundant in circuits that later on, I couldn't find the circuit revealed here. Anyhow, I believe that the TE circuit can be considered "robust" as it works even if different parts are used.
<p>Hello!</p><p>I designed the pc board for the first circuid.</p><p>You can download this design here:</p><p><a href="http://www.mediafire.com/download/s7lm6f8286t0iay/Joule_thief.zip" rel="nofollow">http://www.mediafire.com/download/s7lm6f8286t0iay/...</a></p><p>It works fine.</p>
<p>I need your suggestion and help, from a source I am getting 100 mV and 50 microAmp (&micro;A) not milliAmp (mA) current continuous mode. Can I use jewel thief circuit to light a LED. if yes please let me know the resistor in ohm, diode specification and led specification so that i can make the circuit and how many wire turns will be required. To enhance the density, can I use capacitor for this small production of what specification. Please help</p>
<p>I don't think that I have any specific answer to your question that will allow you to do what you are hoping to do. But I did make a 0.35 V power supply that you might easily make for further testing. I measured 0.35 V across each of the lower 1 k ohm resistors. When I loaded the lowest 1 k ohm resistor with a 10 K ohm load, the voltage across the middle resistor measured 0.36 V while the voltage across the lowest resistor measured 0.34 V. So,... a 10 k ohm resistor across the lowest resistor should draw about 34 u amps, according to Ohm's Law. This is all that I can offer you at this time. In the worst case, maybe you could show that 100 mV @ 50 uA is NOT a very practical power source...</p>
<p>Is it possible to replace Q2 with a Darlington type of transistor ?</p><p>What value of inductor should be used in case I want to power a 0.5W LED ?</p>
Except for maybe the LED, all the related parts are cheap. Why not start with the parts that work on a regular 3 volt white LED and see what happens. I suspect (but don't know) that you might want a 100 uH inductor instead of 330 uH. Or you could consider putting two higher inductors in parallel to both lower the inductance and resistance. Since the LED isn't ON 100% of the time, you might get away with using parts with a less than 500 ma rating. I don't, however, see any advantage to using a Darlington pair. If I were doing this, I would just start trying things out...
Boost converter-y. Awesome.
<p>can you give me exact circuit which you implanted on breadboard without ....and if possible please send me a jule thief ic</p>
I am not sure of your question about &quot;exact circuit.&quot; Did you read all of my Instructables related to joule thief?<br><br>You must give me your full name and mailing address for me to send you a joule thief ic. You can also try using my email address:<br>theabundance@yahoo.com to communicate with me.<br><br>Dave Kruschke<br>
<p>can you give me exact circuit which you implanted on breadboard without ....and if possible please send me a jule thief ic</p>
<p>can you give me exact circuit which you implanted on breadboard without ....and if possible please send me a jule thief ic</p>
<p>Hi Dave,</p><p>I have looked over and read you comments on the #2 circuit. You spoke about using 3 volts temporarily and found the LED or LED's were brighter. Im looking to light a few LED (more the better). would your circuit stand 3volts on a continuous basis? would I wire the LED's in series or parallel? </p><p>You will see my instructable about charging AA Alkalines. The unit is working flawlessly give it a look. </p><p>Luckily I have access to many &quot;used/dead&quot; AA's, my apartment complex has 350+ units and each unit has 2 AA in the carbon monoxide detectors. Thus I get as many as I wish. Im not sure how long that will last but I'd like to build a few Joule Thief circuits and use them up. Bryan</p>
<p>Thanks, Bryan. I would definitely start out by wiring the LEDs in series, not parallel. I believe that the &quot;final&quot; transistor is rated for about 1/2 watt. This would add up to powering a bunch of LEDs, provided the circuit continued to oscillate. I suspect that a modest amount of hands on trials would quickly answer your questions much better than my speculations...</p>
Winding a toroid is really time consuming , so your instructable saved my day - literally ! - I made it in less than 1 minute and it's more efficient than my toroid based joule thief !
And now, I recently had a reader has ask me where I could put a photoresistor in the two transistor &quot;joule thief&quot; circuit to make the circuit turn off in light and turn on in darkness. I really didn't know but wired a photoresistor between the B plus and the base of the PNP transistor, Q1 (see first schematic below). This sort of worked but light didn't always shut of the LED - not good enough. I then added another transistor to turn on and off the circuit, where this transistor is controlled by the CsS Radio Shack style photocell and a 10k ohm pot (old volume control). This seemed to work fine as the light/dark trip point could be easily adjusted by the pot (see second schematic below)...
And now, I recently had a reader has ask me where I could put a photoresistor in the two transistor &quot;joule thief&quot; circuit to make the circuit turn off in light and turn on in darkness. I really didn't know but wired a photoresistor between the B plus and the base of the PNP transistor, Q1 (see first schematic below). This sort of worked but light didn't always shut of the LED - not good enough. I then added another transistor to turn on and off the circuit, where this transistor is controlled by the CsS Radio Shack style photocell and a 10k ohm pot (old volume control). This seemed to work fine as the light/dark trip point could be easily adjusted by the pot (see second schematic below)...
I recently had a request for info on how to run more than one LED off of the two transistor Talking Electronics circuit using only one 1.5 volt AA battery. <br> <br>I didn't want to use a hook up with LEDs in parallel because I might have to use wasteful dropping resistors. <br> <br>Actually I wanted to make my &quot;new&quot; circuit as much like the circuit from Talking Electronics that I had used for this Instructable. <br> <br>I replaced the single LED with a string of 8 Green LEDs and they all lit up, but they were a little dimmer. So, at roughly 2 volts per LED, maybe the voltage supplied to the LED string was about 16 volts. I then connected a Red LED to the coil-transistor collector connection. I did this to make sure I had DC voltage to measure with my meter. The meter was connected to the free negative end of the LED and also the battery negative. I measured 14 plus volts here. And, if I add 1.6 volts for the Red LED voltage drop, I get pretty close to 16 volts DC. <br> <br>Then I temporarily used 3 volts instead of just 1.5 volts for the battery power and all the LEDs lit up and were much brighter. <br> <br>Finally, I disconnected all the green LEDs, installed the 1.5 volt battery and measured the voltage again and got over 21 volts. Adding 1.6 volts for the Red LED voltage drop, I might have a maximum voltage of 23 volts &ndash; just with one AA battery. <br> <br>Dave <br>
Who is the one who said I was yammering? According to the dictionary yammer is to whine or complain, neither of which I've done. If you're not irascible, you sure have an odd way of complimenting people!
Hello nice writeup but you do know you can achieve this Joule Thief circuit with one Transistor and no Capacitor. Also the &quot;Transformer&quot; you claim not to use is an Inductor in the circuit you made. The only difference is that a person can easily make a Transformer rather than an Inductor and that the Inductor uses less space compared to a Transformer. I myself made an interesting Joule Thief only using one simple IC, a LED, and a source of energy above .5 volts. It can light up several LED's without any problems, too. I would show this circuitry but not a lot of people are into it even though it takes less room than any Joule Thief made so far.
You have some good points. However, one of my main goals with my two &quot;Joule Thief&quot; projects was to find, test and share &quot;Joule Thief&quot; circuits that don't require bulky transformers that has to be wound by hand. In both projects, I used a cheap, &quot;less than a dollar&quot; inductor that comes ready made and is the size of a 1/2 watt resistor. I had thought that some readers might prefer this to winding a transformer. <br>Moreover, It sounds like you also might have a circuit (&quot;simple IC and an LED&quot;) that meets my goal of not requiring a transformer...
I <a href="http://rustybolt.info/wordpress/?p=4479" rel="nofollow">wrote up a blog </a>about how to make a Joule Thief with a single transistor, no coil, no winding anything, and no second transistor and all the other parts that are needed.&nbsp; All you need are two chokes, which meets your requirement of not having to have any transformer.<br> <br> BTW, the 'transformer' really doesn't transform anything, it just inverts the signal so that the circuit will oscillate.
Other Instructable writers have mentioned &quot;Joule Thiefs&quot; that use two inductors side by side but didn't include more detailed information (i.e. dasimpson1981 on June 24th - Reverse Joule Thief Instructable). <br> <br>Now comes acmefixer with a specific example of this &quot;two inductor&quot; circuit. But to try to learn more, we have to go to his &quot;blog.&quot; I did this but found that acmefixer lacked the kindness (tenga la bondad) to reveal a schematic - a common annoyance found on more than a few blogs. Nevertheless, his description and picture revealed enough for me to plunk around on a protoboard with similar but not identical parts and produce a circuit that works. Acmefixer deserves praise here for &quot;Going from the Word to the Deed.&quot; <br> <br>So,... I may have successfully duplicated acmefixer's circuit. But no matter what, the circuit really worked great and also serves as a really good example of a working transformer where mutual inductance allows some energy from one inductor to be transferred to another nearby inductor. To assure some needed mutual inductance, I simply taped my two 330 uH inductors side by side. <br> <br>From a Qualitative standpoint, acmefixer's circuit certainly meets my desire to avoid trying to break apart a CFL bulb to get a toroid that will probably need unwinding and rewinding - OR getting or winding a separate transformer. The side-by-side inductors seem to serve quite well as the transformer needed in this application. Moreover, all the parts are relatively cheap and available from well known US vendors. <br> <br>From a Quantitative standpoint, I don't know if this circuit competes with a more conventional &quot;Joule Thief&quot; or the Chinese circuit in my Instructable, &quot;Joule Thief&quot; circuits, crude to modern. Perhaps this will be determined later. <br> <br>In any case, the Chinese circuit still has the easy option of becoming a solar powered light by replacing the conventional battery with a rechargeable battery and then connecting a solar cell between pin 1 of the Chinese IC and the negative terminal of the rechargeable battery - the circuit currently used in some Chinese made &quot;solar garden lights&quot;... <br>
My error. I forgot to include the hotlink in the text in the first sentence of the blog, which begins with Wikipedia. This link has the conventional Joule thief schematic. In Dave's comment he says &quot;..if this circuit competes with the more conventional Joule Thief..&quot; but with all due respect, this <strong>_is_</strong> a conventional Joule Thief circuit, except that the coil has been separated into two windings coupled by air.<br> <br> Since my previous blog, I have built a second one, using the same value chokes, and a 1k resistor.&nbsp; I used a 2N4401, which is closer to those transistors used in most conventional JTs.&nbsp; At 1.5V the current was 70 milliamps, which is very close to what one gets using the conventional toroid with two windings.&nbsp; The performance of Dave's circuit may be lower because he used a 4.7k resistor.&nbsp; If he changes it to a 1k, or puts a 1300 ohm resistor in parallel, he may get a similar performance as mine.&nbsp; However, as I said in my blog, whatever choke is used, it should have much less than 1 ohm, preferably less than 1/4 ohm.&nbsp; This will minimize losses and increase the efficiency.<br> <br> Some compromise in performance should be expected from the two chokes, compared to a single toroid with two windings.&nbsp; A toroid has very tight coupling between windings and very little magnetic field escapes the core.&nbsp; The two chokes have a lot of magnetic field escaping, that's how they are able to work.&nbsp; So I would expect better performance from the toroid core.<br> <br> And I'd like to remind everyone that the coil is not transforming anything, it's using the feedback winding to invert the signal so that the circuit will continue to oscillate.
Why not just include the schematic in your reply here and save all of us the trouble of finding your schematic? As a simple example, I attached my &quot;joule thief&quot; schematic with the addition of a note that reveals the operating frequency to be 70 kHz. I bet you could do something like this. <br> <br>Yeah, I put another 4.7 k resistor in series with the original 4.7 k resistor and the LED got brighter. Maybe the LED would have gotten brighter if I also had used 180 uH or even 100 uH inductors. But most of my enthusiasm here comes from the circuit simply working at all with such a variation in parts along with the weak coupling between the transformer coils. To me this means that this circuit is somewhat &quot;robust&quot; and might be friendly to readers trying to work with this circuit. <br> <br>I don't understand the concern about the chokes having a DC resistance of less than 1 ohm. If we substitute in Ohm's Law, we get V = IxR = .070x1 = .07. Is a 7/100ths of a volt drop really going to make any kind of visible difference here? I agree that if the current was 1 amp instead of 7/100 amp, a 1 volt drop would have a noticeable effect on this circuit. <br> <br>Finally, the two coils with mutual inductance are certainly behaving like a &quot;transformer&quot; in that some energy from one inductor is being surely transferred to the other inductor. <br> <br> <br>
I'll address your reply in reverse. The conventional JT coil usually has a 1 to 1 winding ratio, so there is no transformation between windings of one voltage to another voltage, there is only phase inversion.<br> <br> I'm not sure why the &quot;.070&quot; quantity is being used.&nbsp; Ideally the transistor acts like a dead short, and when it's on the whole battery voltage is applied across the winding and the peak current is much higher than the supply current.&nbsp; The time that the transistor is on is less than half, typically 40 percent, maybe less.&nbsp; During that time, the current may be 175 milliamps, and 0.175 times 1 ohm is 0.175 volts.&nbsp; So the winding is now dissipating 0.175 volts times 0.175 amps or 30.6 milliwatts.&nbsp; A typical white LED needs 3.2 volts times 0.02 amps or 64 milliwatts to be fully bright.&nbsp; So nearly half that much is being thrown away as heat in the winding.<br> <br> And this assumes a battery voltage of 1.5V.&nbsp; But the JT is expected to put out light when the cell is mostly depleted, when the voltage is 1 volt or less.&nbsp; So the demands on the circuit are even greater.&nbsp; This is why it is important to keep DC resistances very low, in the coil's primary winding and especially in the transistor.<br> <br> As for why I don't post the schematic here, there are several very good reasons, one of which is that if I later find that I made a mistake, I can easily correct it by going to the blog and making the correction.&nbsp; If my schematics are posted here, or in other forums, I have little or no control of them after they're posted.&nbsp; If I can correct the schematic here, I have two locations that have to be corrected and twice as much work to do.&nbsp; Also, I suggest you read the terms of use to find out more about what happens to your intellectual property when you post it here.
A transformer is two inductors coupled together. They have different properties.
Bravo!!<br> Add my sensor version and you could have a free night light out of 'waste'!!<br> <br> reg<br> ketan<br> -----<br> &quot;May the good belong to all the people in the world.<br> May the rulers go by the path of justice.<br> May the best of men and their source always prove to be a blessing.<br> May all the world rejoice in happiness.<br> May rain come in time and plentifulness be on Earth.<br> May this world be free from suffering and the noble ones be free from fears&quot;<br> ---- Vedic blessing
Thanks for sharing your Vedic blessing...
The Joule Thief name was given to the single transistor, single resistor, two winding coil version. The two transistor voltage booster has been around a long time, and was never called a Joule Thief.&nbsp; Only recently have people been adding confusion by calling other V boosters Joule Thiefs just because they happen to act similar to the conventional Joule thief.<br> <br> It might be cheaper to use a single winding coil in a production environment, but the experimenter can wind his own two winding coil using a toroid core from a dead CFL light for nearly free and a few minutes of his time.&nbsp; I think it's good practice for learning electronics assembly.<br> <br> I've <a href="http://rustybolt.info/wordpress/?p=4348" rel="nofollow"><strong>written a blog</strong></a> with more on this subject.&nbsp; Check it out.<br>

About This Instructable


71 favorites


More by Dave Kruschke: Picaxe Tea Brewing PicAxe LED dark detector "Joule Thief" circuit in a Flashlight Bulb? Yes!!!
Add instructable to: