Joule Thief

[jeanna]

Hi everyone,

I have been studying this new circuit all day.
I want to point out something very interesting about this 2 tier circuit.

There are 2 separate places that have output on the second tier. they are
1- the basic joule thief light place where the legs of the LEDs are at the collector and emitter of the transistor, and
2- the secondary coming off the toroid.

1--, the bjt led spot.
This no longer has a transistor driving the upper circuit and so the place becomes a bit difficult to describe.
The lead from the toroid that normally goes to the collector is still in its place and the LED is directly connected with that.
The other leg of the LED is connected to the negative battery rail .
This is placed exactly as it always is.
Many leds can run in parallel with this bjt LED with NO extra amps draw.
It is usually the brighter one of the two places.

2--, the secondary spot.
This is also not made in the usual way, because only one of the secondary wires is connected to the positive leg of the LED.
The lead from the toroid that is no longer connected to the base and base resistor is connected to the negative of this led.
(This is the one I dubbed orphaned.)

The remaining secondary lead is connected to the negative of the battery rail.

Many LEDs can also run in parallel with this spot.

As far as I can tell these are quite separate spots. The lights have different brightnesses.

They are on distinctly different parts of the circuit.
The only thing they have in common is one lead to the second tier toroid. But it is the opposite lead.
Because of the bifilar arrangement of this toroid, they should be getting their pulses at different times and from opposite directions from each other.

I took a bunch of measurements tonight. The second tier toroid stayed the same (XTREE) as did the base resistor (237ohm) on the first tier.
I tested 4 toroids on the first tier.

In every case, the amps draw was less with the second tier running than with the first tier running alone.
And in all of those I measured, the more lights that were running, the lower the amps draw.
Some of us are accustomed to this phenomenon, but it is here again.

I measured the volts with the scope.
They are never high.
Remember, the lights are in when I am measuring the volts.
This usually lowers the voltage considerably, so the true voltage is probably higher.

OK that's it for now.

thank you,

jeanna

[Artic_Knight]

check this out
http://www.youtube.com/watch?v=1_AGbBsa45E

[TheNOP]

check this out
http://www.youtube.com/watch?v=1_AGbBsa45E

most of the stuff he show is about capacitive coupling at high frequency.
"high" here is relative.

what do you think pc board makers are up against.
at 100 mhz, even 2 traces on a mboard can couples just by being side by side.
carefull layout must be done when creating those board.

you can achive the same thing at any frequencies with capacitors.

i think connecting to an actual component is easyer for most peoples to see things as a closed loop circuit.

wires, "ground", yourself, etc... can become a capacitor plate and the higher the frequency, the lower the capacitance need to be for coupling effeciently.

[Mk1]

@all

I just made a drawing of artic knight idea , i just made it of one wire all credit goes to Artic.

It has not been tested yet.

[innovation_station]

nice  work mk1!!

i cant wate to build this guy   

looks like it is self induceing 


good old otto test with a core  ...   wow

amazing design !!!!!

!ST...