Battery Free LED Flashlight Driver – SJR Looper V4 by lasersaber | May 5, 2015 | Joule Ringer | 34 comments 34 Comments Alejandro on May 5, 2015 at 5:02 pm you tested at least 24 hours? Reply Jimmy C on May 6, 2015 at 7:32 am Sir, I’ve tried building the V3 SJR, and I followed the schematic you have shown. However, I am not successful on what you have done, and I would like to know the wire sizes. Some say 28 AWG or a 30AWG, then on the outside use a 24 AWG. What say you on the transformer-like SJR specs? Reply Sandy Mc Rose on May 6, 2015 at 11:45 pm What component values did you use and what is a litz wire? Thanks Reply iG3 on May 7, 2015 at 12:51 pm I followed the e-mail and saw the adjustable air cap, it is interesting that the circuit exhibits a spike in the output just before draining the energy . . . the fact that you touched the other flashlight circuit and had a great output made me realize that you need to not just very the capacitance for tuning the air capacitor circuit but may also have to adjust resistance as well. Your body has both resistance and capacitance therefore tuning the air capacitor circuit with a variable resistor on the base of your transistor . . . let the sweep(tap) center on the var resistor go back to the primary cap ??? At the pre-flash setting on the cap adjust the variable resistor gradually approaching shutdown of the transistor . . . if this does not help try putting it on the bottom of the LED leg and doing the same . Reply somebody on May 8, 2015 at 6:37 pm What size cap and resistor should be used? Reply iG3 on May 13, 2015 at 2:43 pm Per my posted comment it would be nice if there were a way to send you markups to your schematics that convey what I am saying much more efficiently and provides the technically specific data that is needed. Please send me an address that I can send engineering / technical documents to you. I have a schematic for the led SJR UTILIZING SOLAR PANEL FEEDBACK AND ALSO THE MOST RECENT ADJUSTIBLE CAP SJR SCHEMATIC. This form of schematic markup and documentation is super critical for connection points etc. Reply iG3 on May 13, 2015 at 3:03 pm My BAD! . . . . on a previous post . . .place the Var Pot resistor (0~100k) on the bottom of the transistor and feed it back to 2 diodes or SCRs in series which are connected to the bottom of the primary coil . connect the sweep/tap of the Pot between the diodes/SCRs and the other end of the Pot to the bottom of your Var-cap Reply xenophed on May 16, 2015 at 2:21 pm I have constructed this using a loop-stick ferrite and bifilar winding 2:1 ratio and get about 210 uA current draw. No lenz wire or pot core with a slight change by hooking the other side of the output coil not connected to the base to a 1 M resistor Reply Emon on November 9, 2016 at 3:02 pm Dear sir, I saw your video And enjoy a lot. I ask a one question primary coil howmany turns you put and howmany turns you put secondary coil.please give me a reply Iam waiting for your answer. Thankyou Reply Salvatore de Rosa on May 17, 2015 at 1:03 am 210uA @ 10V (3 white LEDs voltage drop) without considering other resistances or reactances means 2.1mW. Would you measure or calculate the Lumen output? From what I can see the LEDs are on at least 20mA so that is 200mW and probably 60 lumens. Is this correct or am I missing something? It should be totally wrong I believe. Reply bill on May 17, 2015 at 9:51 am did you make the video on the crystal cells? thanks Bill Reply xenophed on May 17, 2015 at 4:10 pm I have continued to work on it and have reduced the current to 100 uA draw. This by far is not the end but I will also note that I rectified the output with a diode and 320uF electrolytic cap and powered the LED that way. By far we have much research to do but if all of us work at it we can get it to 1 – 10uA or even picoamps (just as a side note with picoamp draw the human body would run it no nothing. That is my goal a flashlight that derives it’s power from my hand. I will note the current draw is from the oscillator not the LED or output as that is derived from the collapse of the field. Reply Wildgoose on May 17, 2015 at 4:25 pm Interesting. Picoamps at decivolts tens of femtowatts to drive an LED that requires 30mW at least. There’s more than meets the eye in such a bold goal. Vacuum anyone? Reply xenophed on May 17, 2015 at 4:54 pm 20 uA current draw with hooked to Earth ground see The Xenophed Project Reply iG3 on May 17, 2015 at 5:07 pm Foil in the coil with the new style coil you are implementing is a promising endeavor. This inductor coil style introduces a capacitive feature that increases energy storing capability. remember that finding resonant modes in the final product will greatly enhance the efficiency and drop power consumption inherently. Did you do feedback on one of your other sjr shop lighting applications to supplement the caps/battery life? Did you narrow in on the tuning of the latest flashlight sjr by finding the cap value and feedback the emitter of the MPSA18 thru a adj var resistor splitting the flow between the positive primary coil leg and the negative leg of the power source cap? The idea was to tune both the resistance and the energy flow on both side of the trnsistorand coil. the tap of the var pot landing between the series connected shotkey diodes pointing to the positive primary coil leg? Reply xenophed on May 19, 2015 at 12:01 am People it’s not the coil or really the circuit but the combination of them. I made a coil about 2:1 loop-stick core that lights 12 LED’s @ 2 uA * 12 VDC or 24 uW or 2 uW per LED. So that is no vacuum people just facts and if we reduce this 10 X we get the reward do need to keep trying. Reply lasersaber on May 19, 2015 at 1:52 am @Xenophed Thanks for sharing. I sounds like your are making great progress. Do you have any video of your build? I would love to see it. Today I tested a new pot core. It works better than any of the other ones I have tested. Here is a link to a picture of it: https://www.surplussales.com/Images/Inductors/Ferrite/PotCores/ich-2616-77.jpg Model: (ICH) 2616-77 https://www.surplussales.com/Inductors/FerPotC/FerPotC-1.html It’s performance is much better than my previous cores. It is a very different looking material. It has a super smooth mirror looking finish where the two halves meet. It performed about the same with a current draw of around 2-5uA until I clamped the two halves together with a plastic clamp. As the pressure increased the current draw fell to less than my meter can register. In blink mode it runs on less than .1uA in fact my best digital meter just sets on 000.0uA. I now need to find a much better meter so that I can see improvements. In solid lit mode it runs for a much longer time on one of those small 200v caps than my previous build did on two of them. So some more good progress has been made. Reply xenophed on May 21, 2015 at 10:48 pm Found a 5 uA meter from the 50’s let you know if it reads Reply Vladimir Tolskiy on November 7, 2015 at 8:25 pm Dear Lasersaber, what variable capacitor did you use? What range of capacitances does it cover? I want to buy an air gap variable capacitor of a similar type. This is very important to me. It would be interesting to keep this circuit at its peak brightness with a varicap tuner that has a feedback system. Does peak brightness == faster capacitor charge drain or is it not necessarily related? (working from separate power source) Reply Vladimir Tolskiy on November 8, 2015 at 1:12 am What are the values of the diode – capacitor pair that are used in this circuit? Is capacitor the variable capacitor? Thank you. I want to order parts soon. Reply xenophed on May 19, 2015 at 4:59 pm Video is uploading as I speak at The Xenophed Project. Reply Luta on May 25, 2015 at 6:58 am Hi lasersaber, I made a variation of Your circuit with random pot-core transformer I used in other projects (no copper stripes), 2200uF/50V cap, 1,8V starting voltage, one blue 10mm LED with the effect of over 9 minutes of bright light after initial pulse. Quick contest – the same cap and led directly, voltage 2,6V easily won by looper ( at 1,8V ) – that was really fun of the evening. Thank You very much for sharing Your projects. Greetings from Poland. Reply Travis B. Moore on June 9, 2015 at 11:13 pm Have you tried metal film capacitors from 1oopf to 1000 pf or ceramic caps instead of a tuning cap. And is that diode with the capacitor a germanium diode or will a VARACTOR DIODE AM RADIO TUNING VBR=16V CT=420PF @1V work as well? Would it work with a 3000uf or 2000 uf cap work in that circuit as the power source at 16v+? What are the ranges on the gages of wire that can be used? Reply travisbmoore on June 11, 2015 at 1:48 am would these work? Transistors silicon KT315J analogue 2N2711 USSR 100 pcs by USSR Be the first to review this item Available from these sellers. KT315J Voltage 20V The maximum allowable current 0.01 A Current transfer ratio 20 Power dissipation 0.10Vt Cutoff frequency of the current transfer ratio of 250 MHz Hfe 30-250 There are more than 25 000 items in our stock. The full listings can be found here http://www.amazon.com/shops/A19NX3RFNSYB6R If you can’t find the item you need, you may contact us. Amazon.com source Reply travis on August 3, 2015 at 8:21 pm Just spamming, but I manage to charge a 100 farad capacitor to 3.9v It runs the led for several minutes. I had thought of making a solar led light. And tested a basic capacitor charged with a blocking diode to run a white led. Could your circuit be powered by the sun for a solar version. But I think the 100 farad is alright, it is higher than the cheaper .3F tor .5F 5v capacitors. electronic goldmine had a little battery that was recharable that workes like a PAS cap. I think that may be good for your circuit. I am thinking that simple radio coils may work those tunable coils you find in your radio for a transformer. I am just posting some thoughts. Reply Mgn on June 24, 2015 at 2:27 pm Hello I follow all your progress with alot of excitment… I am curious if you ever tried or thought about powering a peltier using such circuit. Thanks alot… Reply iG3 on July 6, 2015 at 6:05 pm Per a comment on the transistorless carbon rod style ringer: foil-like flat and square interlaced mag wire conductor turns on the primary and the secondary can improve resonant characteristics of ferrite and ceramic pot core transformers where ceramic is used for higher frequency bands. Also adj. cap, and adj. core can help determine exact harmonic bands of operation of a particular circuit even those that are including human body contact. Trying to operate within the optimal harmonic resonance band is key. and not to go into too high of capacitance for the given oscillating power source. On a previous video with adj. cap. it is important to find the optimal capacitance in the freq. band from led intensity rise until mid brightness. That is the working freq. band for that setup . . . tweak forward from there with resistance set, transistor type, led type, feedthru energy coming back from transistor. (Consider most recent energy harvester technology from Maxim, SI, TI, or Mouser or other, that can utilize a single or multiple miniature cell PV to capture ‘droplet lens concentrated’ “back-reflecting or adjacent” light from one or more LEDs in the existing circuit to provide power to itself and supplement mVolt-uAmp regulated power ) These PV & Harvester ICs are tiny and require little or no extra components. Reply Kapanaland on July 11, 2015 at 7:16 pm Hello all @Lasersaber: Can you check this project, i see that is similar to your SjrLopper, but it give continous power. In another recording of this circuit is shown coil winding, if I find him, inserting link next comment. Pls chceck this if you have some free time. https://www.youtube.com/watch?v=FJMif5BOSII I’m waiting for a response, even if you have no intention of checking this, then at least write. Reply Vladimir Tolskiy on September 12, 2015 at 3:39 am Please make a film on how you wound your 4 most efficient transformers by drawing the transformers as you film. Taking them apart, verbal descriptions or written instructions are not very helpful. Drawings are a lot more instructive. Thank you. Reply Roxy on October 28, 2015 at 1:12 pm Hi there! I just want to clarify few points: – the value of used capacitor (I mean capacity and voltage); – Where is the resistor (and used resistance range) on the diagram which you connected on video (for brightness regulating)? Reply Matthew on November 3, 2015 at 2:51 am Instead of a flashlight charging motor, why not just tap a piezo buzzer… One tap would probably zap the circuit but you should add a protection circuit to it, imagine it Flashlight that only requires a tap to become fully lit for hours… Reply flippper00 on November 19, 2015 at 6:53 am Big fan of your work, lasersaber! I built this circuit, and am impressed! I can get 40 minutes of run time on 5 LEDs from 2 x 3300uF caps starting at 15V. Transistor choice does make a noticeable difference. The MPSA18 has a bit higher gain (~1000) than standard TO-92 NPN switching transistors. That makes for a nice bright glow. Hook up xformer windings reversed and it glows for 30 seconds only. Strange. The arched sawtooth waveform with ringing (coil/ferrite resonance) is the ideal waveform. Frequency tends to step at multiples of resonance, not change smoothly, as capacitor voltage drops. This gives it a constant light output for many seconds before dimming an almost indistinguishable amount in a moment. Fun stuff. My longest run time is with an 8050 transistor, less gain, not as bright as MPSA18. Darlingtons have an interesting high power effect at certain low voltages, but tend to switch so quickly that very little energy is expended per cycle, leading to very dim LEDs glowing for a very long time. My setup used 5x 20mA/3.4V/36000mcd white LEDs in series, with a 1:4 wind pri:sec of 26 gauge enameled wire. And every npn transistor type in the house. 90V schottky diode between pri/sec, no cap across it. Fun stuff. Did I say that already? 🙂 Impressive trick to show co-workers – tap the 9V, bright light for minutes. (I also used 18V, but blew a few transistors and 1 LED above 15V on the caps) Reply Carsten on April 30, 2016 at 10:07 pm Hi LaserSaber, thank you for your inspiring work! When do you plan your DIY build instructions video of the latest SJRL? While I was somehow successful in replicating the SJRV 4.1 with the small pot cores from mouser electronics (which you used in the first version of the tesla torch) and using 0.2mm wire and 70 turns + 140 turns, a A 112 germanium diode and the MPSA 18 transistor and playing with involving some resistance by touching several spots in the circuit – I’m still thinking there is room for improvement in my setup because there are so many variables to experiment with. By watching your latest videos I assume that for the latest tesla torch you are using a 1.1M resistor between + and transistor base and for the coil 0.5mm wire. But how many windings (primary/secondary) did you get with that wire onto the bobbin? For the old version of the testla torch which I also tried to reproduce month’s ago the 20:40 ratio with 0.5mm wire did not worked well for me and my runtimes were very discouraging. Which germanium diode you are using in your latest build? All the best for you and your future work! Reply Alex on July 28, 2016 at 1:51 pm Good simulation! I take off a hat! Reply Submit a Comment Cancel reply Your email address will not be published. Required fields are marked *Comment characters available Name * Email * Website Time limit is exhausted. Please reload CAPTCHA. × eight = twenty four Notify me of follow-up comments by email. Notify me of new posts by email.