Introduction: High Voltage Joule Thief!

Picture of High Voltage Joule Thief!

this is how to make a high voltage joule thief!
the ones i made were 90 and 180v but the more
hv turns you do the higher the voltage(about 1v per turn)
this circuit runs best @2.4v-3v

Step 1: Materials.

you will need:
toroid core(ferrite)
2n3904(sometimes others work too)
512ohom resistor
winding wire(0.15mm approx)
0.2-4mm as well!
neon glow lamp (i used maplin rx70)

Step 2: Wind the Hv Windings.

Picture of Wind the Hv Windings.

Get your 0.15mm wire and wind at least 85 turns
through the middle of your toroid like the pic.

Step 3: Wind the Low Voltage Windings.

Picture of Wind the Low Voltage Windings.

twist the high voltage wires together to make winding easier
using your 0.2mm wire wind 5 turns and make a center tap
and then wind another 5 the center tap is the positive 2.4v.

Step 4: Put Together All the Parts.

Picture of Put Together All the Parts.

now put all your parts together .
and test it!


mouse23 (author)2016-10-20

I came to this publication via another youtube video by this author. There You can watch a demonstration of this gismo and the purpose for what it was intended . This dummed down version is pretty much useless.

inspecter gadget (author)mouse232017-01-27

I was sorting through some electronics - and found this circuit again - I could never get it to work..... I had another go this year- 5 Yrs on and still nothing..... being pragmatic and very analytical- I felt the 1amp 2222 transistor, it was very hot -!I tried everything but then I noticed if you flick it on and off there was life -(not oscillating) . I re- wound the primary in thinner wire and in 1 direction not [S] Orientated or bucking and it finally worked!!!

It's a good circuit it's kind of a variant of a slayer exciter - or Tesla HF switched pwr supply ...sweet..........

inspecter gadget (author)2017-01-27

Yes it should, the ferrite creates a good path for the magnetic field, which is setup by the inductor.

Crazyelectroboy (author)2016-10-25

Can someone please answer my question I wound the high voltage winding all around the toroid will that work if I wind the primary over the secondary high voltage winding.

Testas.T (author)2015-08-18

:D really? hv output is very, very low current so it can't blow fuse

mouse23 (author)Testas.T2016-10-15

Voltage divided by resistance gives you the current.

You don't really know what resistance is going to hang on to those HV terminals of 450 volt. And those capacitors pack enough juice to make you see stars.

But still, adding a fuse defeats the whole purpose, which is giving something a jolt.

Testas.T (author)mouse232016-10-16

well. you are wrong. 2N3904 can handle max 200mA. So 2.4*0.2=0.48W (U*I=W). And i don't know from where you took that 450v. author said:

,,the ones i made were 90 and 180v''.

Let's use formula:(I=P/U)

0.48W/450V≈0.001A 1mA

0.48W/180V≈0.003A 3mA

According to

1mA trough body not noticeable

1-3mA trough body is mild sensation, not painful

mouse23 (author)Testas.T2016-10-18

Actually, those transistors will charge capacitors at a rate of milli amps, but the discharge rate of the capacitors are not restricted.

Testas.T (author)mouse232016-10-20

(using your logic)... If i charge capacitor with 2mA, i instantly can get 200A out of it :D? To charge bigger capacitor you will need A LOT of time...

If you will want to argue more, then build this circuit, measure output current and then you can say something. Theories without facts are nothing!

There are no capacitors mentioned in this instructable. (your imagination created 450v and capacitors:) )

legionlabs (author)2013-01-15

This is a perfectly good circuit, but you don't actually need 2 coils.

If you use the original joule thief design, add more turns of thinner wire, and choose a good value for the resistor, you can get the same voltage range. This way you save yourself a little work and wire.

Quite by accident, I got a standard joule thief to produce 100v from a 1.5v cell with about 50 turns of wire, a 2N2222, and a 500ohm resistor. More turns of wire should increase the output voltage.

To those getting 0v output: you probably connected the wrong ends of the coil together. If you were to connect the "start" of one wire to the "end" of the other, that would be correct. Otherwise, your transistor won't oscillate and you get no output.

The usual precautions for working with higher voltages apply.

EleamarN (author)legionlabs2016-10-08

yeah, ur ryt! :D
mine worked so well.

nawg (author)legionlabs2014-04-11

That's actually a bad idea. it exposes the transistor to the high voltage output which will kill it after a while. its better to build it this way and put a varistor on the output to prevent it from getting burned out if you turn it on without a load.

legionlabs (author)nawg2014-04-12

Quite correct! Still quite fun for a 0.07$ transistor though!

dudes (author)legionlabs2014-03-27

can you post the specifics and schematics for you 100v joule thief? it sounds very cool. an ible would be great!

lego mindstorms (author)2016-07-18

I did five layers of 60turns ,that s 300turns right??, but I only get 34v 1a

Houssam21 (author)2016-05-09

Hello... Im really in need for your help! I didnt know how did you wrap the wires around the toroid and how did u connext them in the circuit ? please help

iceng (author)2015-10-26

How about putting the number of turns clearly on your schematic........ Please !

eyal_hadar made it! (author)2014-12-09

is that the same?

AbdulKharis (author)2014-07-30

cool (y)

Effluxmind (author)2013-09-22

Nice! Joule Ringers are interesting little circuits. I use one when I need > 30 volts for finding negative pins on fly back transformers.

BlackWolf1024 (author)2013-07-17

For some reason I can't get mines to work, I measured the voltage it was out of range on my multimeter set it to AC, but it doesn't light up an neon or so. It lits up a LED without a sweat. The parts I use is a toroid core, 85 turns of HV, 10 turns on the LV 1 and 5 turns on LV 2, NPN 2n3904 transistor, 500 ohm parallel resistor. I've tried 5 turns on the LV 1, but it doesn't seem to work. I use 2.4 volt as an power supply.

baven (author)2012-05-21

Hello there i got problem with this circuit. I got exactly same setting as you with same turn ratio same resistor and transistor and voltage too but when i insert the batterys i get no voltage on the hv output. Any suggestions where could be the problem ?

inspecter gadget (author)2011-12-30

Hello, have hit the very frustrated stage?? Have built a tiny 8t Bi-fil version which worked fine -same transistor as in your set up......I have since made your version and have had no luck-only that the transistor starts to head gradually.....I'Ve used [2] different ferrite's and thicker wire----?????????? Anyone have any suggestions

:-Have done some research as some ferrites have low or high Hysteresis -understanding that some saturate with magnetic field and some enhance field strength /Flow.....?

NatureTM (author)2011-12-26

Thanks for the example. Multimeter shows 500V, however not enough current to light the neon bulbs I have on hand. Skin test doesn't produce a shock feeling either. I'm going to school for CE and just finished my "circuits" class. Armed with my incredibly basic knowledge, I would guess this power supply has an extremely 'high internal resistance.'

wateenali (author)2011-11-10

hi there thanks for this great tutorial.I made this circuit and it generates high voltage but the current is too low and it unable to lightup the led bright i cant understand i check the voltage on the meter and it is more than 500v.Please help to solve this problem.

R.A.T.M (author)wateenali2011-11-24

thers no current the thin your hooking it to drops it drasticly giving 3-6 v off

raja26 (author)2011-08-01

I have tried with a thick wire for high voltage output and a much thicker wire for low voltage input.. It works very well.. I didn't connect a resistor though, it can run two 3V LEDs with a single AAA of 1.2V..

The only problem I've got is the heat that builds up in the transistor(2N5294). My another doubt is, whether we could modify the setup to run a dc motor, like simple rectifiers or something.. Thanks for the great instructable. Been searching for a guide to increase DC voltage.

Alex1M6 (author)raja262011-08-28

Adding a resistor to the base might help with lowering the amount of heat the transistor produces.

raja26 (author)Alex1M62011-08-28

Hi Alex, Thanks for replying. I have figured it out by myself. But can you just tell me how I can run a DC motor with it.

Alex1M6 (author)raja262011-08-28

Could you try placing the motor in place of the LED? If it does not run use a 3v input with two AA battery's in series.

raja26 (author)Alex1M62011-08-28

Thats not possible, because the motor runs with just the battery itself, so when using the Joule thief setup it should be working normally. But the output from joule thief is a pulsated DC. I have to convert it to pure DC for making it work. I just want to know what type of capacitor and inductance to use it as a filter for the DC output.

Alex1M6 (author)raja262011-08-28

Oh, in that case you might want to try adding a capacitor across the motor terminals. Try different values and see what results you get.

Robot Lover (author)Alex1M62011-09-27

I believe a simple ceramic disk capacitor will do.

raja26 (author)Robot Lover2011-09-28

Thanks for the relpy Robot Lover. I'm a zero in electronics, so I don't know about these ceramic capacitors and all. All I can do is try some Random Capacitors.

Alex1M6 (author)raja262011-09-28

Hi, Sorry about my comment. I just assumed you had electrical knowledge, I tend to do that sometimes with people.

Have a look at the image I have attached it will show you a ceramic capacitor in parallel with the motor and give you an idea of what to look for.

raja26 (author)Alex1M62011-10-01

Wow, thanks for this quick help.. By the way, what will be is capacitance?.. This is a fair picture.. Much needed help indeed..

raja26 (author)Alex1M62011-08-28

Yeah, thats what I'm looking for but is capacitor alone is enough? Don't we need inductance?. I don't know much about capacitors so please specify some value for that one 1.2 V AAA battery.

PKM (author)2009-07-13

Do you know what the waveform of the output looks like? Are the edges sharp enough to drive something like an ignition coil? I'm thinking of something like this- battery powered Jacob's ladder FTW :)

12V (author)PKM2009-07-13

the first picture is when under no load and the second is with a neon wire ended lamp.

ramatern (author)12V2011-09-04

OK, now I'm interested. What I'm looking at is a sine wave. Why can't I get an inverter to give me a nice sine wave like that? Can I adapt this to function as an inverter? ;-)

PKM (author)12V2009-07-13

Awesome, O-scope traces- I was half expecting a response along the lines of "WTF are waveform edges" or "no" or something. Still, that looks like a "no" because it's near sinusoidal with a load on it. Oh well.

Xellers (author)PKM2009-09-01

555 timer

12V (author)PKM2009-07-13

i do know of another circuit that is a load of dc pulses
but i haven't got o-scope traces because mine broke so i have to make another circuit.
when i made a circuit i will take more o-scope pics.

Adam Manick (author)PKM2010-10-10

That is not battery powered. You plug it in. And yes you can get nice arcs from this circuit with a little improvement.

NcssR (author)2011-03-29

hey guys,at this joule thief works a c945 or bd 139 or sdt9208 power transistors

Mudbud (author)2011-02-28

I can not find a 512ohm resistor anywhere! what else can I subsitute?

P.S no Im not going to get one online, I'm not old enough. :)

sdobbie (author)2011-01-10

Would an n channel enhancement mosfet work instead of the transistor. I have a p60nfo6 mosfet and a k3567 mosfet.

timdgoodpaster (author)sdobbie2011-01-28

I used a 100 volt 3 amp mosfet, had to change resister to 33 ohms and a 9 volt battery. the battery life is short pulling one amp.

andy (author)2009-09-30

I have followed the instructable and found the voltage I'm getting to be a lot lower. I got the circuit resonating at about 24ohms on the resistor, but I always seem to require a capacitor between the Hv outs. The highest voltage I got was 30v out, changing resistor values decreased voltage significantly +/- 10ohms dropped to about 15v and further than that pretty much nothing. Changing capacitor values changed frequency but voltage was effected lesser than by the resistor (+/- 5 volts at the most) 33uF seemed to be giving the best result, any less and the oscillator wouldn't fire. Toroid was 100 turns against 10 tapped in the middle (5/5). The torioid is almost exactly the size of 2 stacked pound coins, 10/12ths of an inch across and 4/12ths of an inch deep. Curiously using a 1.5 volt source (as opposed to three) only dropped the hv out by 5v to (25 volts out). Also the wire I am using is significantly thicker than that shown, at 30SWG (0.315mm). Does anyone have any idea's what could be happening? Would increasing my Hv voltage (by three times) just require (3 times) more turns on the Hv coil, or is it more complex than this? Any help/explainations appreciated, Thanks

rayfalcon (author)andy2010-12-05

try to use a 2200 uF in your project and see how many am/ volts you get then However DO NOT get into the path of it trust me i have done it, my parents picked me up off of the floor.

About This Instructable




More by 12V:ARC tube experiments!     mini 2w arctubehigh voltage joule thief!Simple 12v compact fluorescent tube ballast/inverter only eight components! updated!
Add instructable to: