Joule Thief

[NickZ]

   Robbie:
   Glad to hear that you're ready to go on this...
   I think that we should start where you left off, and first replicate the BwJt circuit, or Hartley oscillator that you were using, but with an alkaline battery, as it's the one you feel would be the best bet.  As you can see I've already got my alkaline battery charged up, but I was thinking to use a D cell size Duracell alkaline battery, although I've got AA size alkalines also, and some Ni-cad AA non-rechargeables also. The Hartley type oscillator, or BwJt?  What diodes should be used?  The IN4148 diodes, are what I've been using, what do you think?  In the oven, or BBQ.   
                                                                                         Nick

[kooler]

Robbie:

Well, I had been thinking, (possibly incorrectly) that when attempting a self run device, we need diodes to keep the juice flowing in the desired direction.  In other words....in a simple JT, instead of an led, I feed that back to the battery, how does the battery know which way I want the electricity to go?  I am thinking it doesn't.  Possible diodes on the positive and neg. lines going back to the battery might help with this?  Just tossing it out there.

Bill

bill
the closest i have ever been was having one circuit feeding the other.. or should i say feeding each other.. it takes two batteries.. or two cap's.. i thought of it for awhile like splitting the positive.. when one circuit cuts off it get feed a charge pulse.. if not the pulse go in to the circuit and messes with the freq.. if you try to feed back the same freq on top of the same circuit you will need 3 times more OU to keep it afloat because while one will charge the second mess with timing and thrid will run the circuit.. while the second messes with the timing it will need the charge pulse to keep it going.. the other way is to feed a different freq from the output to source but it needs to be lower.. ex.. lets say the circuit runs at 5khz.. the feed back should come back to source at 400hz or lower so it would get canceled if trying to go back thur the circuit.. which is hard even using filters cause the lost of current needed to run in the start..

robbie

[kooler]

   Robbie:
   Glad to hear that you're ready to go on this...
   I think that we should start where you left off, and first replicate the BwJt circuit, or Hartley oscillator that you were using, but with an alkaline battery, as it's the one you feel would be the best bet.  As you can see I've already got my alkaline battery charged up, but I was thinking to use a D cell size Duracell alkaline battery, although I've got AA size alkalines also, and some Ni-cad AA non-rechargeables also. The Hartley type oscillator, or BwJt?  What diodes should be used?  The IN4148 diodes, are what I've been using, what do you think?  In the oven, or BBQ.   
                                                                                         Nick

nick i wouldn't bother wth cooking .. i just threw 23 germ diodes away yesterday trying make more.. i don't want people wasting money on old diodes even if i have a problem getting it right.. i trying to think of a possible work around.. maybe soon..

robbie

[kooler]

nick
something else i almost forgot was that one circuit it had a lr44 battery.. and i saw once were someone thought in a battery they said the volts come from pos and the current comes from negative.. so i had the special diode going from pos to neg.. it was because the pulses are on the pos side of the bwjt i made .. and they were send to the neg side to charge.. it was then i notice i could charge a battery from the neg or the pos side.. the old man i talked of earlier told me as long as there was pressure on one side or the other it would charge.. he also told me that if i applied enough pressure to a capacitor it would charge.. i sure wish i knew what he was talking about.. he was a big ed gray fan.. i went up to his old house two weeks ago to see if if wife would let go thur his stuff and she was gone and a sign in the yard for sale so i asked the guy next door and he said she died about 5 months ago and said the people that bought thur out everything to clean for resale.. i started cussing so bad he told me to get off his lawn or the law would be here soon.. it's been a bad year for me.. i wanted the motor the george had built years ago so bad.. all i know about the motor was it had three coils and two north facing magnets and two cap's and he could spin by hand it would run the whole time i was there.. he told me once that he couldn't it patent because it come back as something that already was patented..
well i been drinking and talking to much time for bed..

robbie

[gadgetmall]

hey Guys . feel like Sheeeit from doctor abuse but i won't cry on your shoulder so just can't sleep sp . I wanna play too .
Just my three cents . you have to ask your self what is the minimum a transistor need to stay in oscillation . ok say 0.3 volts for silicon  and less that 0.20 for germanium . i would use tiny components with the least amount of resistance . Ok so we feed a standard 2 coil jt 1.2 vdc . i also wind three extra secondary and we control frequency with  the resistor pot cap configuration on the base as usually . ok now we have 4 points that produce voltage . the CE which will be a DC  pulse current that we filter and clean up with electro caps and germanium . we can go both ways with this back to source and run an led . the other secondarys i have found if wound in different spots fire off at different frequency's due to the flux shifting inside the toroid so has to be compensated with our base electronics and the primary windings  . two of the secondarys are wound in different directions while the third is matched to the Jt right side coil winding secondary  (the original two coils that makes jt) ok so now we have two opposite wound coils that produce an ac and a third matched dc pulsed . with ac we use disk caps then electrolytics and then two germaniums to rectify and then a third diode to feed back to out positive source while the third secondary is pulsed dc we filter again with electrolytic and another germanium to the negitive of the source.
am i the only one that tried this ? Kooler does this make sense ? I believe this is what i did playing around on the rouge light  and got the loop going . with just one jt .

I dont know yet . it's late and i need to reread this in the morning and see if i am coherent in my thoughts . but i sure its something i did like this . the flux from the toroid need to be at a certain frequency to fire off those other three secondary s while running led off transistor and also feeding back to the positive source . so three output are feeding the source . at least this is what i think i remember . Ill test it .

Gnite.

gadget

If this is too weird i will draw it up on paper soon anyways 4 you to examine  so if you guys want to work on koolers previous b?jt project then count me in too .I need to build one anyways cause i didnt get a chance last time .

Morning:
reread the last few post and sounds like something very similar kooler .You explained it better than me though ..