Joule Thief

[altrez]

@jeanna,

Ok so a Permeability of 125 would be better then 5000?

Thanks!

-Altrez

[jeanna]

  Circuit diagram is included, along with two pictures.

I'm NOT using a resistor, only a capacitor, and an axially molded inductor of 15 micro Henrie which contains ferromagnetic material around which is a larger coil.  This simple setup produces 7.5 volts at the test side, without an LED present.

These two elements, or some similar element setup is all that is needed to begin the oscillation process.  This particular setup is running at around 2.5MHz...tomorrow, I will attempt to run many led's off this one circuit.  Then later, I will be using this setup to turn on the oscillations - isolating the blocking oscillation action to this particular two-piece setup, thus freeing up the toroid for primary and secondary winding or some other setup.

Well jadaro, I believe your set up is closer to what Nathan Stubblefield did than any other.

Your explanation is close to how I have explained his generator, also.

An EE friend told me he thought the NS generator oscillated at or in the MHz range.
In fact, this is the reason I did not buy the hand held scope last year. (My friend said 100MHz-200MHz.)

So, I am very interested to find out what is happening with your circuit.

thank you for this,

jeanna

[jeanna]

Ok so a Permeability of 125 would be better then 5000?

I guess. I have never seen any numbers. I only bought one that was called high permeability and it was really bad. (Hazens1 said his same one was fine) Then MK1 made the point that we are looking for high flux and I gather flux and permeability mean opposite things.

Just buy some medium toroids from electronic goldmine.  While you are there, since the is a ten dollar minimum, get a few more types, but save your money and do not buy the bad blue which they call med blue, and don't buy the high permeability, unless you want to see if high permeability will work for you. It is $3.25, I think. I find the filter is a good one.

Get some extra 3904's and some diodes and caps, and a breadboard. It is too bad they don't sell mag wire too. 

@All,
I am going to try the filters that I have wrapped into jt toroid coils in my new circuit today.
The idea is that according to some, the first transformer in the fuji circuit was the one that raised the voltage the most. Then after that, the second transformer could do less transformer work and fill the cap to be ready for the xenon flash.

I have been keeping this in mind, and I will use the higher ratio transformers on the second tier just to see, but I want to put them on the first rail first.

Also, for now, I am continuing to use the XTree (the name derives from its wire colors: christmas tree because it is green and red with shiny red tinsel wire) on the second tier. It may not be the best choice in the ultimate design, but for now I am keeping the second tier the same and the transistor and base resistor on the first tier the same too.

I am recharging some batteries and I will start with freshly recharged ones too. The amps draw is much higher with a high input voltage, which I am used to with all other LED circuits.

OK for now,

jeanna

[TheNOP]

This goes along with my original thinking of not liking resistors.

the base resistor is not only to protect the transistor base, it set the frequency as well.

@all
the frequency is gadgetmall's "secret" of his succes with adding cap on the base to reduce current draw.
it change the frequency as well as the duty cycle, no mystery or unprouvable  things there.

if you understand RC circuit then you can reproduce gadgetmall results easily.
if not, you will need a lots of experimenting and a lots of logic to repete it.
and if you can't think logicaly, you might never be able to replicate it at all.

same thing goes with the cores
if you don't understand core's limitations and how to set the transistor's frequency you will get "bad" cores and not know how you can fix it.


@jadaro2600
i like what i see here: http://www.overunity.com/index.php?topic=6123.msg171923#msg171923
that is an other proof that toroids and ferrites are not needed.
but that depends on your needs.

try replacing the resistor with a bare wire, currents will go up and more leds will light up brightly.
a potentiometer can still be used too, just make one of it lead go through the coil.

don't use any additionnal caps and diode
just the leds at the "normal" jt place
between collector and emiter, between the battery's pole will also work.

[TheNOP]

@All,
I am going to try the filters that I have wrapped into jt toroid coils in my new circuit today.
The idea is that according to some, the first transformer in the fuji circuit was the one that raised the voltage the most. Then after that, the second transformer could do less transformer work and fill the cap to be ready for the xenon flash.

the second transformer is there to trigger the xenon tube, not to fill the cap.
it is an autotransformer type, use to excite the xenon tube, so the cap can discharge through the xenon tube.
it rise the voltage further in the 3kv+ range(no amps) to excite the tube.
the cap won't discharge through the xenon tube without it.


fixed typos: "excite"