Big Joule Theif

[winn1971]

hello guys! i'm a newbie in this JT experiments. I've managed to do my first time replication of the basic JT LED lite courtesy of the makezine video. I then moved a step further and tried replicating lidmotor's inverted joule thief circuit using a ferrite rod (aka antenna bar, as in known here in my place) in an attempt to lite a CFL but the secondary winding can barely lite a single LED.
the setup i did is as below:
19T, 19T using a #18 or #20 enameled wire (unsure of size as it came from a small PSU toroid), 500 turns secondary winding (#26 wire wound in 5 layers, i.e., 100 turns per layer wound back & forth)
the primary/trigger and the secondary windings were wound side by side in the greater part of a 8cm x 10mm dia antenna bar
2N2222A transistor
1.5V AA battery source (aslo tried 2x1.5V=3V)
i used a pot at the base of the transistor to tweak the LED brightness.

have also tried modifying the circuit using Slayer's 2N2222 driving a 2N3055 both with only a 1.5V source and a 3V source but just the same, it hardly lites the LED to a decent brightness level.

What am i missing???

i can't get a big toroid locally to try so i'm hoping i could use a rod for my experiments.

i hope you can point me in the right direction and tell me if i'm doing something obviously wrong but can't see or figure out.

thanks.

[BenTec]

Hello,

I've asked a mate of mine if he knew where i could buy some ferrite rings nearby. He told me for what I needed em, so I explained him and send him the diagram we were talking about before (the one of the halo light).

He said that it is very interesting, but there is much to improve and that the circuit is very similar to one of the an emergency light. I can't remember everything he told me that should be changed. He explained a lot of stuff that got me a bit confused at the end of the call, but I'm glad he did, I've learned something.

EDIT: Now I remember, he told me that for each circle the frequency was getting higher, but everytime the frequency goes higher there was an middle frequency between that could blow up some components, and he told me that using CFL is not the way to go. Because it seems like this circuit will each circle highen the frequency, and he told me if it reaches 1Ghz the CFL would emmit microwave, so he told me to use LED's instead. And some other stuff...

I asked him if he wanted to join the forum but, he told me that he was busy at the moment. Hopefully I get him to add his thoughts and improvements to the circuit.

Greets
Ben

[Lidmotor]

Circuits

Yes correct, look at your diagram date in the video! This is exactly the one I meant, see that it's dated 03/03/2009 ! I thought that was the best version of the initical Slayers007's circuit.

I will investigate on this circuit see if I am able to make a improvement to it.

By the way, do you have a link to Ash's Slayer007 Bedini Fan Remix Edition?

Thanks,
Ben

Check out page 14 & 15 of this thread. The circuits got pretty complicated at this point but we were on a quest to get the most bang for the buck. Alot of fun happened along the way.

Lidmotor

[BenTec]

I forgot to say, yesterday I bought

6x TOSHIBA 2N3055 1H (JAPAN)
6x CDIL 2N2222A P6
6x CEDE B4.7K LIN
6x 1N4007
6x 1K Resistors

I was not able to find the exact parts nor the ferrite rings, I hate this island!

Greets
Ben

[BenTec]

I don't know if you would call it the "best" but it was the one that I ended up using in a boxed version on my sailboat last season. I called it the "Halo Light" because I used a round Fl that put out a ton of light. It is a field tested design that worked well in a dark anchorage on a rolling 32' sailboat.

Here is the video I made last year of the finished light:

YouTube - Lidmotor's Channel

In the real world of use on the boat, the light did not work out as a "charger" light. I would turn it on full bright when it got dark and have to charge it back up with the small solar panel the next day. The charging aspect just didn't work as I had hoped. Ash has made a version of this design that encludes a Bedini Fan charger. That one is actually a better and more refined circuit.


Lidmotor

Yes correct, look at your diagram date in the video! This is exactly the one I meant, see that it's dated 03/03/2009 ! I thought that was the best version of the initical Slayers007's circuit.

I will investigate on this circuit see if I am able to make a improvement to it.

By the way, do you have a link to Ash's Slayer007 Bedini Fan Remix Edition?

Thanks,
Ben

[Lidmotor]

The best circuit?

Hello all, I've read the whole thread, yes, right to here were I post now.

The first like 15 pages are all bout Slayers007's JT, and people making improvements on this circuit, which is easy and great to follow since the last diagram always was the best with the latest additions so it was easier to follow and see improvements clearly.

Then people just started mixing and bringing in new circuits new ideas on all these different JT's and Slayers JT seems forgotten by now. I would have prefered people keep the thread on topic. I really don't mean it bad, believe me! But... it's just my personal preference and I find very confusing.

Which is the most efficient JT by now?

Is this the best?
Pulsed Inverted Joule Thief (Lid Motor Modification dated 03/03/2009, but i think he meant to say 2010)

Thank you very much.
BenTec

I don't know if you would call it the "best" but it was the one that I ended up using in a boxed version on my sailboat last season. I called it the "Halo Light" because I used a round Fl that put out a ton of light. It is a field tested design that worked well in a dark anchorage on a rolling 32' sailboat.

Here is the video I made last year of the finished light:

YouTube - Lidmotor's Channel

In the real world of use on the boat, the light did not work out as a "charger" light. I would turn it on full bright when it got dark and have to charge it back up with the small solar panel the next day. The charging aspect just didn't work as I had hoped. Ash has made a version of this design that encludes a Bedini Fan charger. That one is actually a better and more refined circuit.


Lidmotor

[BenTec]

Hello all, I've read the whole thread, yes, right to here were I post now.

The first like 15 pages are all bout Slayers007's JT, and people making improvements on this circuit, which is easy and great to follow since the last diagram always was the best with the latest additions so it was easier to follow and see improvements clearly.

Then people just started mixing and bringing in new circuits new ideas on all these different JT's and Slayers JT seems forgotten by now. I would have prefered people keep the thread on topic. I really don't mean it bad, believe me! But... it's just my personal preference and I find very confusing.

Which is the most efficient JT by now?

Is this the best?
Pulsed Inverted Joule Thief (Lid Motor Modification dated 03/03/2009, but i think he meant to say 2010)

Thank you very much.
BenTec

[FrozenWaterLab]

@morpher44

It may help to clarify what negative resistance is I think, negative resistance by definition refers to a lack of resistance or non-linear resistance. A neon will first conduct in the glow discharge region when there is no visible light and offers a great deal of resistance. Next the excitation of the neon gas produces a rise in conduction due to ionization, this ionization reduces the resistance to conduction of an electric current. If the ionization increases due to extreme voltage a VAD, volt arc discharge will conduct large currents across the neon plates and produce a purple spark many have been talking about.
Basically the confusion over negative resistance started because of ohms law---"Ohm's law states that the current through a conductor between two points is directly proportional to the potential difference or voltage across the two points". This simplistic rule does not take into account the attributes of what constitutes a "conductor", in a neon bulb the voltage can rise to the threshold voltage at which point voltage drops and current rises as it rapidly conducts, this is not negative resistance it is non-linear conduction. "Negative" refers to a negative condition, the opposite condition to resistance is acceleration, can you have an accelerating resistance? It is like saying left is less right or the ball was falling upward, it is non-sensical. I will apologise in advance for the criticism but this wacky term has been driving me crazy every time I see it, I am not sure why engineers have to make up so many non-sensical terms to explain things which are really quite simple.
Regards
AC

Hi AC
Not to tweak your toes here.
My deff (Loosely) of neg resistor has to do with the opposite of what resistance dose in the main stream Electronics world.
Resistance - "It eats Power" or limits the flow of - so to speak.
Neg resistance ""Adds Power or allows Power flow. Whair from? Same place the Quarks come from and return to, where ever that is. haha
FrznWtr

[FrozenWaterLab]

jeanna
(oh, I am posting mybedini progress on the no rotor bedini thread started by jonnydavro.
and, there is a little video too.)[/QUOTE]

Hi Jeanna and All
I'M ON PAGE 13 AND DIGESTING THE THREAD
It would help if when you post something like that, Would you provide a link?
Please. Tnks

PS To all Posters from Slayer007 on down.
Thanks bunch for your open policy and your work.

I have a bunch of 2.4" ferrite cores (4 in-hand & 10 on order) Largest I could find
Here's a Link
Ferrite Core - 2.40" diameter - eBay (item 280319640167 end time Jun-28-10 08:48:47 PDT)
I will be working with these to expand my Knowledge and hope to be able to add to this data base in the future.
I intend to install Earth Batteries on my Lot and see what I can do with them, tworde making my situation "Self Sufficient" haha
What an interesting threed - Like a Novel - Can't stay away
FrznWtr

[Xenomorph]

If the coil itself is too hot by touch, then this indicates that you have an enormous current flowing inside it. Check for short-circuit possibilities, maybe some wire inadvertently touching a transistor pin.
Increasing the base resistance might help if that is not the case.

[PoppyNumber1]

Hi Slayer007...

I tried to replicate your large coil Joule Thief using a bedini style bifilar coil of 600 turns of 21 guage enameled wire.

I connected it to a 12 volt 4 amp hour battery (as run battery) and 8 rechargable AA batteries in series (charge Battery) and the coil got so hot it burned my hand when I tried to see what was smoking.

The diagram I was using had three coils but I chose to use only coils A and B. The diagram was a little hard to follow so I may have hooked it up incorrectly.

Any suggestions?

Thanks
Poppy

[Tectalabyss]

hey folks
this isn't to impressive but i thought it was neat

YouTube - bjt part 2

it has 4 volts going back to the battery on both sides
at 1.0 volts it has 8.0 volts going back
and it is running at 104ma's
it will run alot more leds but once the test started i didn't want to mess with it.. i wanted to see how fast the battery would drain

out
robbie

Hello Robbie
Sounds like you are on to something.Are you still running the test and if so the best of luck to you.Thanks.Tec
PS. I watched your video A big thumbs up

[kooler]

hey folks
this isn't to impressive but i thought it was neat

YouTube - bjt part 2

it has 4 volts going back to the battery on both sides
at 1.0 volts it has 8.0 volts going back
and it is running at 104ma's
it will run alot more leds but once the test started i didn't want to mess with it.. i wanted to see how fast the battery would drain

out
robbie

[mk1]

@Watson I bet you will like this transistor ...

@all

I just made one NEW VIDEO , testing my last unit , the secondary are not yet installed on the second and third core , but it show interesting results and blow out of the water the AG theory ...

You ca see lighting a bulb on a core not connected to anything and leds at the same time one the primary...

Also at the end and attempt to show the sparks from the secondary lighting a 6 volt 250 mili bulb , circuit runs on a 1.25 aa battery.


YouTube - YouTube - abramrk1's Channel

More details Fibonacci, platonic solids and Vortex coils

2x 2 volts 60 mili plus 1x 6 volts 250 mili

I also know how to modify the secondary to fit any configuration rodin to star ship .

[Watson]

One way to keep from frying the transistor

Yes watson, anyone can crank out DC from a dynamo, but if you connect the coil to a reed switch and have the coil close at the peak of your hand cranked AC waveform, now you have radiant too, and can do alot more.

Its not magic free energy, its smarter use of what we have and also learning new ways of getting what we need from previously unknown sources.

Also your choice of transistor is important like you found. A sharp attack and decay time is what you need, and for low voltages you need a specific make for that, which would not work for 9v or more, and there you would use power transistors. How many times I thought I could get it brighter with a 9volt and learn that both tranny and LEDs failed in a picosecond.... So if you want to make a hardy JT that can go from 1v to 12, use a MJL21194 which is one of the better power transistors, but if you know you only will work with 1 to 3 v, then 2n2222 mpsa06 or what have you. I think mpsa06 does not make a good JT transistor because you need almost a whole volt to turn it on, however it might have the sharpest rise an fall time, but I was reading a different transistor only needed .2v (but good luck keeping that transistor from frying with any more) , also I think PNP might use less current.


Also, watson, your discovery of a cap acrost the battery reminds me of a stiffler ladder filter... the first part anyway. just caps acrost the + and - terminals, with ferrites going along bothways. Isolates the oscillator from power supply.

The JT puts some uniquely demanding requirements on the transistor. One obviously is the low voltage, 1.5V or less. Power transistors don't make good JT transistors, partly because their gain doesn't hold up at low voltages. And the 2N3055 for example doesn't have much gain to begin with, only about 50 or less.

And the low voltage means that if the transistor is supposed to switch even a few hundred milliwatts or power, then it has to switch a lot of current, and that means it has to drive a low impedance. For a typical JT, than means 1.5V at about a half amp, or about 3 ohms. It has to be much less than that, a fraction of an ohm, or else it will waste a lot of the power.

The PN2222 or MPS2222 (I'm pretty certain it's not a 2N2222, which has a metal case and is 4 times as expensive and hard to find) and the 2N4401 are better than a 2N3904 which can't handle much current. The BC337 is better, but still can't do as good a job as a transistor made for this job.

One transistor that does a really good job in a JT is the one they use in a flash circuit from a disposable camera. This used to be a 2SD965, but I think many of the circuits now use a surface mount transistor which is really difficult to remove and use. You can get an equivalent one from Fairchild, the KSD5041, or the NTE11 is available from electronics stores but not Radio Scrap. These can handle several amps at low voltages, and put out enough current to drive a 1W LED to decent brightness.

These transistors look just like a PN2222 but handle several amps. They don't handle much power, but if the circuit is switching fast, the transistor doesn't have to dissipate much power. If you just have to have a power transistor there's a special superbeta transistor from Fairchild, the KSD1273, which has twenty times the gain of a 2N3055. With a gain of over a thousand, it's up there with the darlington transistors, but it's not, and doesn't have the inherent limitation of a darlington, two base-emitter junctions in series that have to have at least 1.3V to turn on. It will outperform a darlington at low voltages in a JT.

Finally, to keep from frying transistors, make your adjustment pot go down to a minimum resistance of a few hundred ohms, so it won't draw so much current through the base and burn it out. Put a 220 or 330 ohm resistor in series with the wiper arm of the pot so when it's turned all the way down it can't go below that resistance.

Another way to reduce the current is to use a MOSFET, which doesn't draw any gate current. And they don't go into meltdown as easily as a 2N3055 because the current doesn't go through a PN junction like a regular BJT. And FETs can do some other fun stuff. The 2n7000 is cheap and looks like a PN2222, but it's not. I've put them in a JT and it worked down to 0.27 volt - see my blog here.

And every power source must have a bypass cap across it, especially if it powers a 555 timer chip. This must have a 0.1 uF across the power pins at the chip. Read the 555 application sheet, I think it's AN70, for the reasons.

Have fun experimenting.