Joule Ringer!

[Jbignes5]

What about leds?

Jbignes5, Mat

Yes you could do it that way, it's more like a oscillator amplifier and the CFL is the issue. Everybody is using a different CFL so you must tune that device to the circuit. You may find that building a bias circuit like from base to emitter 10 K and then from collector to base say 500K will allow the transistor to start much easier.

Again the CFL looks like a big Zener when it first starts so it just need a little help after that home free. Now the reason I posted that data sheet was to get the Darlington arrangement as close as possible to the gain of that special device.

The SOA curve is also very important. I think that device can take an inrush current of 5 Amps from what I remember from the curves, but not for very long. So it's important that the CFL is glowing right away to cancel that Zener effect ( looks like a short circuit).

I also think that LaserSaber's original circuit used the CFL in the base circuit with high voltage, the reason for the transformer he wound with leaking diodes, but you could use a transistor if you bias it right. When he touches things and it gets bright that indicates a capacitance problem in the circuit, it's not stable.

"It's the high voltage in the base circuit is where the problem is". Just use your current meter for the total current and watch what it does before / after the CFL lights and see if the current drops down to MA , then do the same in the base circuit. That transistor only wants to run in very narrow curve and you must match it and never change that CFL to anything else (not good).

If you then change the CFL it does not operate any more because you have changed that curve again in base current, to much and you cause the transistor to cross conduct at that frequency and it fry's. You can put a Zener diode in the base to limit it, Requires two parts one diode and a Zener of the correct voltage to limit base current.
John B

I know leds are just diodes that emit light. Is this the same thing and why we are able to use them interchangeably? My aim is to make a headlight for my electric scooter that will only need the potential difference the batteries provide to run it.

The motor is another thing and I have stated my direction of interest in other threads of the motor/generator to drive the scooter.

My direction of interest in this is to provide a currentless lighting option so that it will not lead to much drain of the starting battery system of the motor/generator. Ultimately this would be a complete EV solution designed from the ground up once the technology is stable enough to merit attention.

I really appreciate your thoughts!

By the way I am also a John B. lol....

jbignes5

[John_Bedini]

Joule Ringer

Jbignes5, Mat

Yes you could do it that way, it's more like a oscillator amplifier and the CFL is the issue. Everybody is using a different CFL so you must tune that device to the circuit. You may find that building a bias circuit like from base to emitter 10 K and then from collector to base say 500K will allow the transistor to start much easier.

Again the CFL looks like a big Zener when it first starts so it just need a little help after that home free. Now the reason I posted that data sheet was to get the Darlington arrangement as close as possible to the gain of that special device.

The SOA curve is also very important. I think that device can take an inrush current of 5 Amps from what I remember from the curves, but not for very long. So it's important that the CFL is glowing right away to cancel that Zener effect ( looks like a short circuit).

I also think that LaserSaber's original circuit used the CFL in the base circuit with high voltage, the reason for the transformer he wound with leaking diodes, but you could use a transistor if you bias it right. When he touches things and it gets bright that indicates a capacitance problem in the circuit, it's not stable.

"It's the high voltage in the base circuit is where the problem is". Just use your current meter for the total current and watch what it does before / after the CFL lights and see if the current drops down to MA , then do the same in the base circuit. That transistor only wants to run in very narrow curve and you must match it and never change that CFL to anything else (not good).

If you then change the CFL it does not operate any more because you have changed that curve again in base current, to much and you cause the transistor to cross conduct at that frequency and it fry's. You can put a Zener diode in the base to limit it, Requires two parts one diode and a Zener of the correct voltage to limit base current.
John B

[Jbignes5]

I also suggest....

Not only should we look at matching the capacitance but the resistive elements in the circuit as well. Coils have both capacitance and resistance. These should be matched in my opinion so that there is a balance and this energy can flow without problems.

@John_Bedini.. Good to hear about the 2n2222a's.. I will start asap on the Darlington trio. I still have a captret quad so I won't have to put it back together.

The only thing left is to do the bifilar coil big enough to push the current down enough to only have the voltage present. I am thinking that this coil should be matched to the resistance and capacitance after the coil. When this happens I think this is where Lasersaber saw great run times from his setup.

Is this an impedance issue? I am starting to think it has a lot to do with it. when it is just right the run times increase and when it is not power is drawn from the source to make up for it or becomes self correcting. Think of it like a teeter tawter. The system we are looking at is or should be finely balanced so that any change throws it into oscillation. It is in essence growing in strength based on the capacitance and resistive balances and the feedback we choose to re-enter the system. The load once it is turned on (led, Cfl) doesn't need much to maintain that conduction and can run from very minute currents once the voltage potential reaches a stable level. I believe this current we are creating in this system is able to sustain the load for very extended times if the circuit has been designed well.

John, here is a good question. You have been talking about resistors on the base of the transistors. Should it be all the transistors or just select ones in the trio. I have picked out that it should be after the first stage. This agrees with my assessment after experimenting with the Joule Thief when connected to the captret. Since the captret is a current limited device already the first stage is pretty much current limited. Should ever level after the first stage be protected by a resistor on the base? Except for a small voltage drop I don't see how it could have more current available if the captret is driving the circuit.. I am also thinking that we should include a small capacitor across the resistance to allow for the ac signal to pass while the resistor limits the current to the base of the next stage. Am I correct in my thinking here?


on another note I am gonna be using this torrid ring for the bifilar in the Joule Thief:



And a regular ferrite loop stick for the transformer.

I have a few questions about the winding of the loop stick. Is the small windings (1 ohms) bifilar wound? I know the 1400 turns is single wind but the other small winds got me confused.

[Matthew Jones]

Hey John
That helped alot. Smoothed everything out and it got brighter.
I got about 4 minutes, not quite full bright but I see the direction I need to look.

Cheers
Matt

[John_Bedini]

Joule Ringer

Lidmotor,
That device has some gain happy to see it work that easy.
Good work...........
John B

[mk1]

@all

I was wondering if you could charge a cap between the Neg and Top Of the capacitor A la Captret . If that works then maybe then that would work as a input for the loop .

Mark

[Lidmotor]

Closed loop circuit and three transistor Dalington

@All
I worked a little with returning the energy back to the front end by using an AV plug. It didn't work that great but maybe there is a way if the link is perhaps an open link of some kind. On some of the SEC experiments Dr. Stiffler showed it from time to time. It is a capacitive link of sorts.

I also tried the three transistor Darlington arrangement that John B. suggested. That was amazing. I just setup a simple Joule Thief circuit with a loaded 2.3v supercap as the power source and left the base wires open. Lightly touching the wire ends started the circuit running.

Here is a short video of the experiments:

YouTube - Closed loop Joule Ringer & Darlington transistor setup.ASF

Lidmotor

[John_Bedini]

Joule Ringer

Mat, Good that it worked out for you try adding some resistance say 10K ohms between base of first transistor to emitter of the last transistor. Try to look at this like a hybrid amplifier circuit. other tricks are second base to last emitter removing the first resistor. sometimes you need a little bias from collector to base like 500K ohms leave the resistor in the circuit. Also to match capacitance try 3 to 20 PF from collector to base of first transistor to see about stability. I think this one is a matter of gain.
@ Jbignes5 Yes 2n2222 will work just fine almost the same device.
The CapTret touch around the circuit and see if when you touch the top of the cap if it charges. Through some arrangement you can get it to charge up, it may be something that they call a gimmick just a couple of twisted wires for feedback to it.

Good work Mat.
John B

[Jbignes5]

Hmmm...

Anyone had a runaway condition where all of the sudden the cfl goes maximium brightness ,arcing all over the place and then smoked ? well i did using 1.2 volt c cell with two air core coilss(Bi) in series . bummer on xsistor . i need a couple dozen more . this mod i did to lasersabers circuit ran fine for about 40 hours on this battery (still waiting for caps)and i moved two air core coils close together(about 1/4 inch and it ranaway in like 10 seconds . all i saw was bright light ,blue sparks here and there and then smoke and darkness . It was something . .


Albert

Yeah you locked it into a runaway condition. You are changing the resonance as you bring the two air coils together. Since they are bifilar they only emit an electric field. This field has zero resistance. Ok what burned out?

Put a 1/4 or 1/2 inch block in between the two coils and have that as your limiter. The wider, 1/2 inch would be prudent since it ran away at 1/4 inch. Play with it and see what it does...... Well once you get what burned out of the way rofl.....

Don't worry I popped a few tranies already... Oh and a few caps heh... Just play safe. Keep your face at a respectable distance and I usually only use one hand just in case. I always try to incorporate some kind of off switch or wire I know I can pull quickly usually the positive battery connection.

In this case I doubt the battery was even connected or was it?

@John_Bedini hey so what were you thinking about the captret addition? I know the kind of energy coming out of the captret is very very weak but yet it can run an led all the way down to nothing. Amazing how it likes leds... The funniest thing is it takes my dead 12 volt battery (usually 9.2 volts) and lets it run a good amount of diodes depending on the, what I like to call cylinders. You know like a 4 cylinder engine. I had 8 in parallel running a good amount of leds (average 8-10) at medium brightness.

When you mentioned the Triple Darlington transistors and the captret in the same paragraph it got me thinking. hmmm... I'm just forming the picture of the circuit in my mind....

Will 2n2222a's work for that John? I got tons of those... Would be nice if they could be used.... I had my best waveform from a 2n2222a... It wasn't drawing much either... Hmmm...

I was using a Joule Thief on a captret before and got good results. After the captret filled it used very little real power. My weak 12v would dip from 9.2v then quickly back up to the full 9.2v.. I was only using 1 transistor. If there were two more stages that would boost it enough to run without losses right?

[Matthew Jones]

The graphs for capacitor charge and discharge rates are curved not linear. The discharge rate increases as the charge falls, and begins to race, gurgle, spit and sputter like a tipped over container of fluid emptys out towards the end.

No thats not what happens. The thing is going full bright in flicker, but the cap is no where discharged. Usually between 7-8. You get bright flash's from the bulb.

It eventually goes out below 5 volt but its the really bright flickers while the bulb is dimly lit.

The wave form does not change either. Its as even as it can get.

Matt

[synchro]

Capacitor fluctuation.

I get that same thing. I just can't figure out why.

Matt

The graphs for capacitor charge and discharge rates are curved not linear. The discharge rate increases as the charge falls, and begins to race, gurgle, spit and sputter like a tipped over container of fluid emptys out towards the end.

[mikrovolt]

Sharp burst

We understand discharge on the Cap voltage and the proportional decrease in frequency however the sudden increase is not completely clear as it is like inverse log function. Possibly the energy is done resonating it briefly hangs out at one end of the circuit the glowing ions release energy when they shrink back non ionized state. Now we need to use a lot of power to get them excited again. The glow would remain longer if the voltage and frequency were inversly proportional but then we go off topic I think the bifilar mutual capacitance is giving the modified fuji the gradient along with the leaky germanium diodes are what need to be modeled as a subcircuit just the way LS designed it. you make good observations.

[fusionchip]

Was that from the schematic you had previously posted? Sounds like you got a surge of energy from somewhere. Maybe if the positioning of the coils was done very carefully you might get a self runner with excess energy. Maybe move them very very slowly together or find a way to limit the energy once they start (I assume you hit some resonance that looped). Sounds like an exciting find but maybe a kill switch will be good to have.

It was one of the first ones i posted using lasersabers and a hazens drawing except i put two air coils in series . the second one was a 3 pound coil and i hooked its input where the germaniums connect to the + . i am trying increase the output without running more than 1.3 volts and as low as the transistors i have . I have a BUNCH of vintage one and can run it very low but they are rare .The transistor i used was a fugi and all i have of it is a little piece so no number.. it was dang close and the battery still had a volt in it and i moved the coils close and i achieved somthing i did not expect . too much of somthing and POOF !

Albert

[xee2]

I noticed in a couple configurations that the bulb began to fade as the cap ran down, then it would get a bit brighter a couple seconds and then fade again and then in the last second or two got real bright before going out. That point was by far the brightest it ran the whole time. I played around with trying to lower the voltage to the tranny and a couple places but couldn't enhance the effect.

I noticed that also. I think it is jumping to a higher frequency (looks brighter when it gets turned on more often). The reverse current may be too low to return the capacitor all the way to zero volts when collector current (and thus energy stored in magnetic field) gets low. So the capacitor does not need to gain as much voltage before it triggers transistor again. But I am not sure.

[Matthew Jones]

I noticed in a couple configurations that the bulb began to fade as the cap ran down, then it would get a bit brighter a couple seconds and then fade again and then in the last second or two got real bright before going out. That point was by far the brightest it ran the whole time. I played around with trying to lower the voltage to the tranny and a couple places but couldn't enhance the effect.

I get that same thing. I just can't figure out why.

Matt