Joule Thief

[jadaro2600]

Can I rectify the secondary ac and put that into the battery rail, then run that through a toroid without another transistor? (I would like to keep the manufactured parts down) So, will the input of a pulsed voltage take the place of the transistor? and if so, can I use it to amp-up the output to a high enough voltage to be useful?

...

* What this means is that the voltage difference between the 2 separate pickup wires is out of phase so much that there is a wave form that "happens" as a result. It is like the beats you hear when the cello string is not quite tuned to a perfect 5th with the next string. The difference of the differences, if you will.

Well, eventually, the 'positive feedback' will overcome the breakdown voltage of the collector - which is what i've been worried about this entire time.  I'm thought about  doing this very thing, but I can only imagine that because of the nature of the timing, that one would be on and the other would be off, and they would, as you point out, be out of phase - and would be destructive to one another, or worse - they could wind up squeezing the waveform into a higher amplitude, but provide much less useful energy and become even more worrisome.

The idea, I suppose, is that when pulsed DC runs through a CFL, the murcury vapor has a tendency to collect at one end, and so, the poles have to be shifted to keep the bulb working properly - which is why sometimes the weaker power sources will light only one end of the bulb.

Harmonics are symmetrical across a symmetrical body.  I pointed out in my previous post that, since there's a current disparity between the collector and base coils, then the magnetic field is 'whop-sided' and so is the waveform that results between the two, so, in turn, any pickup coils are going to have non-symmetrical bias to them - wanting to flow in one general direction more than another - you can see this by placing a diode across an nonrectified pickup coil in one direction, and then in the other - and make a note that it may be brighter in one direction than another - or in some cases, it might not light up at all, or it might light both ways and there be little difference between the two, which is the end result of a complicated balancing act.

SO - I had this idea that the current flowing through each could be the same, however, the inductance could be different - and in turn , the magnetic flux as well, this would achieve the same effects as a blocking oscillator - I guess it would mean fewer winding on one coil than the other.  This has a bit to do with my resistorless design a few pages back.  My problem is accounting for the pickup coil - which seems to be where everyone is getting their higher voltages from.

I'm getting 40 volts rectified off of three turns going to the collector, and this is only a few turns away from exceeding the capabilities of the 2n2222...

*edit:

I tired rectifying flyback in the coil to the base back onto the coil to the collector and it ultimately lowers the voltage there, so, I'm guessing there's a destructive wave patter in the timing on my model.  This doesn't exclude the idea though, ..on a secondary, if you try to patch back in flyback to the collector, it will probably have the same effect.

[jeanna]

@Jadaro,
Ah yes.
We were typing at the same time when we made our previous posts. I was still typing a bit after you. Our ideas have a lot in common. I love it when this happens.

I am not talking about letting the flyback get to the transistor. This is entirely taken from the secondary and after it passes through the rectifier. (I even stuck it on a separate breadboard.)

This is not where I am going ultimately. I just wanted to see if the pulses could be changed into a good useful waveform and to my delight they can.

I was thinking I needed to study up on the inductors or toroids you have been using, since your approach may be the very thing this needs to finish up and work.

My 7-7-68 from yesterday gives me 85v on one secondary wire, and 22volts on the other. This ought to give me plenty to play with. But not leds. I have no desire to lose any more of them on this. I am looking for some kind of thing that will work on 30 to 60 volts ac. Of course, there is nothing, is there? Why would there be?

Thank you for your input. It has been most helpful.

jeanna

[TheNOP]

But not leds. I have no desire to lose any more of them on this. I am looking for some kind of thing that will work on 30 to 60 volts ac. Of course, there is nothing, is there? Why would there be?

for 60 volts, 20 x 3 volts leds in series will do.

[Pirate88179]

http://www.youtube.com/watch?v=98q4Ic6UL7c&feature=rec-HM-r2

You all HAVE to watch this video.  This is a toroid winding machine in action.  I always wondered how it was done and it seems pretty simple.  We all need one of these machines.  In the beginning, he is just loading the wire, watch as it winds the toroid nice and neat......and fast!!!!

Bill

[gadgetmall]

Well xee2,

It was a while ago so maybe you forgot.
I ran 30 leds on 19mA and they were all very bright.
I could have run more but I ran out at 30.

It works out to 0.63mA each. I am not sure, but I believe it is still the record.

jeanna

  me  0.20 ma . at .20 vdc but i cheated . i used a VERY RARE germanium . so i guess that don't count . Do i hear a Challenge !!! Yea . Im game if you are .
Gadget . lets shoot for a string of 100 leds . white christmas lights .. who can run em the lowest on a Jt with silicon xsistor

GADGET