Joule Thief

[innovation_station]

@IST: Remarkably complex setup. I wonder what it will achieve!

So you are using MK2, as i understand it has 2 primary and 2 secondary coils on 1 toroid.
Do you use both primaries and feed them from different circuits? I am trying to grasp the necessity
of having 4 coils on there, i am sure it is for a good reason. Would be nice if you could explain that.
Are that 1N60s on your rectifying board?

it does get a little wild for wires soon  lol

the mk2 has 4 secondary outputs plus i use the jt out  each coil of the secondary .... is putting out 106 vdc x 4 plus the jt

we are makeing a magnetic feild real fast and colplasiping it we pay for the magnetic feild but not for the colapse  so the secondaires are fed from the colapse ..  it is best not to take from the source any more than you need

the diode # im not sure what they are ....   sorry

i have found the same results any where in the 0-1k range.... 

not so freq dependent now 

the white bars ... will be 30 outputs...   the 2 groups of 3 rings will be driven from 2 idependent sources ..

wich are my secondaries ....  on my aa powered mk2

it will be a neat unit ..

ist!

i added a 10uf cap across the source on my mk2 coils ...   and i have this out put off a dead battery or almost dead ... 90.9 they all read the same ...

im useing a 2n  4401 npn transistor ...   and 1k resistor ...  there 4 feeds at 90.9vdc ........

have not touched the jt side yet ...

[hazens1]

While testing some other configurations, I have come across a peculair combination..

I have what looks like a 5/8in ferrite toroid taken out of a computer supply wrapped 22 times 2 bifilar wind Green magnet wire from Radio Shack. Using a 2n2222 NPN transistor with a 10ohm resistor going from the coil to the base of the transistor. This is all feeding into a circuit I ripped out of a DOT-it H2O round push light that normally operates on 3 AAA's in series ~4.5v. There is a 24ohm resistor on the board with 3 white leds in series and a push button switch.

I'm using a single AA instead to power the JT to power the 3 led circuit. At first it was just as bright as when hooked up to the 3 AAA's consuming around 200ma. The battery voltage dropped from 1.625v to 0.575v in about 4 hours. The transistor was running a little hot, but no more that 150° F. I was suprised to see it still running with as little as 5 volts.

Then it ran for another 2 hours dropping from 0.575v to 0.425v and the transistor was now cool to the touch. I remeasured the amp draw at about 25ma. I left in on overnight at that point expecting the battery to be dead when I woke up.

Here's the amazing thing. When I woke up the voltage in the battery was at 0.422v after 9 hours and the brightness was the same as when I went to bed which is about 25% on when the battery was fresh. Then I went to work, another 9hours later the voltage is now 0.419v.

I'm pretty new to this so I have a few questions..

How can the battery voltage only drop 0.006v in 18 hours giving out so much light while appearing to run on ~20-25ma according to the meter?

I'm guessing it has something to do with the rate the circuit was oscilating at. Did the oscilation decrease as the battery drained, thus making the circuit more efficient?

How is this thing still running at such low voltage? I was expecting it to cut off at 0.7v-0.9v according to what I have read about other peoples circuits.

How come when I measure the voltage coming off the transistor it is always less than the battery, closer to the battery the smaller the resistor, but always less when the leds need way more than that to run?

Thank you for your time and wonderful research everbody!

[xenomorphlabs]

@hazens1:
That is indeed peculiar.
Battery discharge curves might not be completely linear, but this 18 hour period seems
too unusual.
First i would check the battery status of your voltage-meter (forgive me if you have done that of course)
but some people ran into weird readings just to find out that their meter was not operating accurately anymore.
Second, ... KEEP THAT BATTERY !!! 

Did the oscilation decrease as the battery drained, thus making the circuit more efficient?

Too bad not too many of us have oscilloscopes (me neither), that could show the frequency and
would/would not explain something.

Try to repeat the experiment )

[innovation_station]

so far i have 4 coils wired and working with the 10uf caps  all are putting out between 89 - 100vdc ..


ill keep at it i hope to wire all 120 wires into the rectifier board tonight ... 


ist

[jadaro2600]

I'm not quite using the same setup as you all, and tanking the resistor draws more current - so I don't.  I'm getting reading of 6.4ma.  I use a 9.84k ohm resistor and a 10pF capacitor in parallel with a 9-50pF varicap for fine tuning.  At a total of 24pF, the led seems to be as bright as it has been with other setups.  If these varicaps were performing as they should be, then I wouldn't have to incorporate a 10pF ceramic disk capacitor and I could just tune it using the varicap.  The real dilema is find good varicaps!

Anyone have any links? - these varicaps I just bought perform like crap!

As I said earlier - the toroid I'm using contains no ferrite and the schematic basically sticks to the original diagrams I found posted elsewhere on the internet.  My main innovation is the winding of the toroid - which If i weren't soo ill right now, I might try a few different configurations.

I find that including additonal LED in series - while trying to measure the current useage, causes a low reading and a dimming of the brightness of the LEDs