Big Joule Theif

[Lidmotor]

Sharing of ideas

I just wanted to say thank you for all the great work you guys have done on this. This is some amazing light output for that much input powerand you still get most of it back in the charging battery. Great stuff

@Jetijs----I don't know about the other guys here but I have followed and admired your work going way back. Your build quality is perfect. If you ever build one of these lights I hope that you will show it. It is kinda like your solid state charger only it makes light along the way.

@Shlodo---Great video!! I especially liked how you showed the "sewing" of the toroid. I almost gave up the project until Slayer told me how to do it. My first attempt was a messy disaster. Your video may become a classic. The only thing missing was the circuit diagram. A little mystery might be OK though.

@ Slayer---All I can say is thanks for all of this. The key piece of the circuit is your trifilar 2" toroid design. Without that it just doesn't work right.

@Xenomorph----This may sound really out there but what if we used two 555s. One to pulse the whole circuit and one of Shlodo's to pules the bulb and charge battery. Tuning that might be a nightmare though. They might get in a 555 timer fight. Things might get ugly.

Lidmotor

[Xenomorph]

@Lidmotor: This is absolutely not too way out for what i can imagine hehe.
I am conceptionalizing a solution for this right now and it looks like this will not even need an additional timer
That all depends on if a cap can be found that charges to a sufficient potential (like above 12v) in around 1/100 sec.
Something like less than 1 microfarad. I will run some cap simulations to find the best capacitance value. Hopefully i can experiment with that tomorrow.
If that will turn out to be not working well, then a 2nd timer that pulses the cap every 10 seconds for example can be added.

[theremart]

This looked interesting..

829-0080-UNIVERSAL TOROID CORE-Electronic Surplus Inc.Offering obsolete IC's, discretes, motors, relays, switches, pots, and much, much more! Always looking for your excess inventory.

it is only the primary wrapped on this torroid, by the look of it one could wrap the top of it with the other and have a dandy of a torroid...
FYI...

[slayer007]

829-0080-UNIVERSAL TOROID CORE-Electronic Surplus Inc.Offering obsolete IC's, discretes, motors, relays, switches, pots, and much, much more! Always looking for your excess inventory.

it is only the primary wrapped on this torroid, by the look of it one could wrap the top of it with the other and have a dandy of a torroid...
FYI...

That ones an iron core theremart.

I dont think it would work quite as good as a ferrite one.
But it sure would be nice to not have to wind the toroid.

I just started winding another toroid.

This one will be 26ga. for the JT and 30ga. for the pickup coil.

Boy it sure takes a long time to wind the 30ga. such a small wire not making much head way.
But I think it would work better with the smaller wire.
Be able to have more turns per layer and the wires a lot tighter on the toroid.

[theremart]

Torroids...

That ones an iron core theremart.

I dont think it would work quite as good as a ferrite one.
But it sure would be nice to not have to wind the toroid.

I just started winding another toroid.

This one will be 26ga. for the JT and 30ga. for the pickup coil.

Boy it sure takes a long time to wind the 30ga. such a small wire not making much head way.
But I think it would work better with the smaller wire.
Be able to have more turns per layer and the wires a lot tighter on the toroid.

Ok, did not look that close I noticed there are worlds of differences in torroids of the materials used. I have no clue as to understand all the different types..

[Lidmotor]

556CN-Dual timer chip

@Lidmotor: This is absolutely not too way out for what i can imagine hehe.
I am conceptionalizing a solution for this right now and it looks like this will not even need an additional timer
That all depends on if a cap can be found that charges to a sufficient potential (like above 12v) in around 1/100 sec.
Something like less than 1 microfarad. I will run some cap simulations to find the best capacitance value. Hopefully i can experiment with that tomorrow.
If that will turn out to be not working well, then a 2nd timer that pulses the cap every 10 seconds for example can be added.

Xenomorph--- I got a timer chip at Radio Shack today that is two 555s in one chip. It is called a 556CN Dual. Part#276-1728. This would probably get the job done----if we need to do this. I hope that we can figure out a way to do it with just the one. Actually I stared boxing the whole thing today. I figured out the (2) DPDT switch circuit and it works great. The two switches have an 'off in the center' position that lets you turn off one battery or the other until you get the bulb on. The way I have it wired you can use both batteries as a source also. I installed a charge port for a wall source input or a solar panel input.

Lidmotor

[seth]

That ones an iron core theremart.

I dont think it would work quite as good as a ferrite one.
But it sure would be nice to not have to wind the toroid.

I just started winding another toroid.

This one will be 26ga. for the JT and 30ga. for the pickup coil.

Boy it sure takes a long time to wind the 30ga. such a small wire not making much head way.
But I think it would work better with the smaller wire.
Be able to have more turns per layer and the wires a lot tighter on the toroid.

Hi Slayer,

Do you always use a bridge and capacitor off your secondary coil, before running it in the CFL?? Have you managed it without?

Thanks in advance and Nice work

PS, Why do you think ferrite cores are more effective than iron cores??? Ive been playing with both myself, and the ferrite ones do seem better for some reason. I saw on OU someone was using an air core. Anyone had any luck with that?

[slayer007]

Hi Slayer,

Do you always use a bridge and capacitor off your secondary coil, before running it in the CFL?? Have you managed it without?

Thanks in advance and Nice work

PS, Why do you think ferrite cores are more effective than iron cores??? Ive been playing with both myself, and the ferrite ones do seem better for some reason. I saw on OU someone was using an air core. Anyone had any luck with that?

Hello Seth

No I don't always use a rectifier off the HV side with a cap.
It's just some thing differant I was trying it did seem to make the light brighter when only using one AA battery tho.
A photo flash cap worked good for that it would keep around 300v in the cap across the light when it was on.
Then I could send that to the base of the 2222 transistor allso to give it a little more boost.

I think the ferrite cores allow you to reach higher frequencies than an iron core.

I did try a couple air cores for just a regular JT with no pickup coil.
It did work ok but I don't think it would work to good with a secondary pickup coil.

[Lidmotor]

"Halo Light"

I finally boxed the thing. I call it a "Halo Light".

YouTube - Lidmotor Halo Light

Lidmotor

[shlodo]

Bravo Lidmotor! and some info about Charging Batteries

I finally boxed the thing. I call it a "Halo Light".

YouTube - Lidmotor Halo Light

Lidmotor

Simply Brilliant Lidmotor!!! I love it, great boxed build! I have been eagerly awaiting to see what u came up with.
Great idea leaving the Voltage regulator in it. Where did u get that thing, its very versatile

wats your amp draw??

I cant seem to get the charging directly to the battery and light the Fluoro at the same time, its one or the other it seems.
Ive built a cap pulser but the light goes down momentarily.

wats your secret to getting the charging and good light on the Fluoro?? Is it your circular Fluoro? Im using a 60cm long tube...
even when i dial the amps up it still wont work??

Battery Charging
Ive had some experiance charging 12V battery with the JT. U might want to monitor the voltages carefully. I find that as the secondary battery charges up its good to lower the resistance (on the JT circuit). it will draw more but u'll charge it higher.
Its all about matching the impediance of the circuit to the battery, which Bedini talks about. When we first start charging the impediance in the battery is low and we can use a higher frequency (more resistance in the JT), but as we charge it the impediance of the battery increases so its good to make adjustments.

The problem with this is that the charge battery wants the circuit to draw more at the end of the charge, but the run battery has lost charge by the end and finds it harder to put out the increased amperage

Hope that helps
-shlodo

[Lidmotor]

Halo Light Charging

@ Shlodo--- Thanks for the tips about charging. That explains a few things that I have seen happen. Impedance is something that Bedini talks a lot about and I need to reread what he said. This flourescent JT is VERY hard to get to work as a charger and light at the same time. The round bulb that I chose really helped for some reason.
On the little LED JT I found that it actually seemed to charge better when the batteries were kept at a slightly discharged level. This reinforces what you said about impedance. The timing of the battery "swap" is important. Do it too soon and the charge battery hasn't had enough time to recieve the 'gift' from the circuit. Do it too late and the source battery gets used up. The circuit draws 30 millamps to 300 millamps depending on the light you want.
I have decided that what I just built will have to be given energy regularly to keep it running. The question is how much? Was this all worth it or not? --- I think so.

Lidmotor

[Xenomorph]

Charge AND light

@Lidmotor:

This flourescent JT is VERY hard to get to work as a charger and light at the same time

I think i have a solution for that. I will test it in the next week, when i get all components required (since using relays is a big waste of energy)
Theoretically, you split the operation of the circuit. In the part of the duty cycle when the CFL fires, you disconnect the charging part, so that it does not draw from the circuit. Then when the coil collapses and the CFL would not light, you connect the charging part to collect the radiant spike.
That way the circuit would not have to provide both parts at the same time with the necessary amperage, hopefully having the CFL light with maximum brightness and provide enough charging when it is off.
Then we would have brighter light and slightly (15%) reduced charging (not sure if it would be even reduced though,depends on when the radiant spikes are moving).

That makes the 555 timer actually a multi-purpose component in this specific circuit.
That could be done in the HV part as well, even charging a 2nd battery with high voltage diodes and a proper sized cap, because there is radiant spikes from the secondary coil that are sitting there unused.

All that is needed is the same timer that pulses the circuit already to manage the switching, which is an awesomely minimal solution.
Again, all theory still. It needs to be tested.

Conceptional circuit:

[Lidmotor]

Xenomorph's new circuit

Xenomorph-----Wow! The more that I studied the circuit the more I thought of a "Singer" sewing machine. If this thing works it will be like watching a marvelously engineered machine run. Beautiful design.
I played around with solar charging today and found out that a solar cell phone charger will run my Fan Bedini SSG. Made a video of it but only posted it at my Youtube channel. I also tried charging up the "Halo Light" with my little 2 Watt panel and it did NOT work. The batteries that I used were for a VHS camera and probably have a circuit in them that didn't like what was happening. They charge up just fine on a wall wart transformer.
I can't wait to see if you circuit works.

Lidmotor

[shlodo]

A Light with significant charging would be ideal!

@Lidmotor
Ok so it looks like i'll have to find a "Halo" tube
You're right, the timing of the "swap" is very important. Its also important to give the batteries a bit of a rest before swapping front to back so they can recover a bit.
So at its maximum 300mA @ 12V that's a 3.6 Watt system plus you're cathing some of the radiant spike which would bring it down to about 2W I reckon. Man that's pretty impressive considering the amount of useable light.

By the way Lidmotor, have u ever seen the "Energy from the Vacuum Series"?? it's were i get all my info about Bedini charging. It's a brilliant resource.
Ive bought a couple of them and downloaded a few as well (im a uni student so I cant afford to buy em all! )
Id highly reccomend Part 2, Part 6 & 7:

YouTube - "Energy from the Vacuum" Documentary Series -Trailer
YouTube - "Energy from the Vacuum" Series - Trailer - Pt. 6

Bedini Solid State Charger:
YouTube - BEDINI RENAISSANCE RADIANT BATTERY CHARGER


@Xenomorph
I like your idea, you're right about the relays. As they make contact a spark appears and you kow that energy is being wasted. What about solid state relay's?
In your diagram what are the switches? transistors, mosfets??

I still see a problem tho with the timer on the front end.... The whole circuit is on for 15% so as I see it, we can only light the light and charge 15% of the time. So u might have to split that 15% into charging and lighting again.
Also as Lidmotor has found out, its very hard to light the Fluoro while charging. I thought swapping between charging (85%) and Lighting (15%) on the back end would be the way to go.
You're right about the secondary coil too, there is energy to be captured, but I wonder if u catch energy off the secondary will be at the detriment of the charging off the collector.

As you said, its all theory, now we have to test it! good luck!

[Lidmotor]

Halo Light circuit diagram

Someone asked me to post the diagram of the "Halo Light" so here it is.

Lidmotor