Joule Thief

[timmy1729]

One more of the ferrite bead JT.  We used to be able to post more than 2 pics/post.  Well, once Stefan gets the new software up and running, maybe we can then.

I am showing the LED from an angle so it does not look as bright.  I believe the extra ferrite mass in the bead helped.  I don't know how to reconcile that with those that are using no ferrite and no core at all.


Bill

@Bill

Well, that should be simple enough to solve. Let's have anyone using no ferrite or no core, including those using the blue inductors(look like resistors) to post their components and the voltage dc with positive going on the positive of the LED after any diode or cap and negative on the bottom of the negative end of the LED.

@All
Using the diode and cap is what doubled my voltage on mine. The size of the toroid didn't make a difference until I added the diode and cap. Before both were reading the same 1.5vdc. Now the small toroid reads 1.9vdc and the big toroid read 3.0vdc. When I added the Darlington transistor to the big toroid, it bumped it up to 3.3vdc. I'm using an NTE Electronics general purpose diode NTE110MP. The cap is 50V and I think 4.7uF. The writing on it is too small. I need to grab a magnifying glass. I just used the diagram that Jesus posted and I reposted a few pages back. Like MK1 said, I will play with the resistance and I also want to play with the cap as well. I'm also going to use a larger diagram of improvements that I saw and report back tomorrow night on how that goes.

@MK1
What would be a safe resistance to try as a minimum without frying anything? I'm using the standard JT 1k Ohm.

[Pirate88179]

Well, I went ahead and stayed up to make a few photos of the double JT.  Yes, all wire gauges and other components are exactly the same as all of my others.  As far as the number of turns goes, these should be close to the same as the ID's are all the same size and I wind as many as I can stuff through the ID.

Yes, ferrite (iron oxide) has many different compositions all of which give different results electronically. (I have machined a ton of wave guides for cell towers years ago)  I have no idea what the composition is for any of my toroids or beads as they were all salvaged.  The other day, I scraped the paint off of one I just pulled out of a CFL circuit and thought was ferrite but was powdered iron, which works in a similar, but different way.  (according to wiki)  So, my control here as far as experiments goes is poor as I don't know what I am working with exactly.  Later, I suppose it would be good to order some stock with a known composition and then if we do something great, we can replicate it a lot easier.  This was a good point you raised.

Bill

:MK1:

The above ferrite comment were directed at you.

@ Jeanna:

Thank you.  Good idea also.

@ Timmy:

Yes, I agree 100% with this idea.  I am just having a bit of a time getting my mind around this.  so the cap and the diode addition made the ferrite size make a difference?  Good to know.

[Mk1]

@timmy

Check the data of transistor online , and check for minimum voltage at the base , and try to get that.

For the core no core i think its a mater of freq , and and different working range.I did not think possible from so simple circuit.

@pirate

You could check the resistance in the coils for idea resistance per winding .

[gadgetmall]

@Goat
re: That is very impressive as far as lighting 300 LED's !!! Pirate88179 is up to 400....but...given what you learned in power recovery from the Bedini circuit do you think that the voltage going to light the 300  + LED's in this project could be recovered to recharge the battery driving the circuit while driving a few less LED's?   

I'm already on this using a toroid off the transformer osc  and rectifying the voltage  with two windings one on top of the other looks good so far . will fire it up in the morning and give results . i think it will need a separate larger winding . Maybe i can wind a Fugi like transformer and an extra winding, large coil with a Big fat 4 inch toroid i got at the junk yard and maybe a 555 timer to play with the Frequency??? We can always step down the 300 +plus ac in another transformer for less volts more amps ... Later .

[Pirate88179]

@ MK1:

Thanks, I didn't realize that about the coils resistance.


@ gadgetmall:

Man, you got a 4" toroid?  That's a $50 item!!!  I need to check my junk/recycling place soon!  Great idea about using another transformer to up the power from the too high volts.  This is getting very interesting!

@ All:

I keep leaving things out of my posts I meant to put in.  My connection of the 2 JT coils was done in parallel using some jumper wires.  Thing is, this now tells me that this is more than just taking 1.5 vdc from the bat and upping it to 4 volts to light the led.  In parallel, the mA's would have increased, but not the volts right?  So why were the 2 superbright leds so bright if the 4 volts were now something like 2.5 or so from my connection?  Neither one should have lit with the volts so low right?  Or do I have Ohm's law confused? (series adds volts, cuts mAs, parallel adds mA's cuts volts?)

Bill