Joule Thief

[Artic_Knight]

@ electricme

Photos look great. My interpretation of photo 799 (I assume that has the capacitor across the base resistor) is that the capacitor is slowing down the turn off of the transistor. Since the magnitude of the voltage spike is less when the rate of change of current is less, then this would produce a lower voltage spike. And since the amount of voltage induced in the pickup coil is also less with a slower rate of change of current, there will also be less voltage across the pickup coil. However, I do not see anything that would indicate why adding the capacitor across the base resistor decreases the battery drain (which is the reason I put a capacitor across the base resistor).

ok so the cap that is pushing the transistor to stay open is a bad idea as far as peak output but what good is the cap and resistor in parallel going to do besides lower the resistance of the curcuit? how do you propose that affects the base resistor? it looks like a short to me when the cap discharges through the resistor back to itself accomplishing no work except to allow some of the current to pass through to lower resistance to the base. 

?

[resonanceman]

Wouldn't it be neat if there was an autoadjusting POT to keep up efficiencies as voltage dropped!

Good work

stprue

I think  the pot is actually OK
It is  the  voltages that are changing  as the battery either  charges or discharges
If  we can keep  the  voltage  in a narrow  range  the  problem should  be solved .
I do not  remember  what  your circuit is like
What I am working on is  collecting flyback  pulses  from  JT coils
There are not  alot  of ways to hook up the coils  so that they add  together .....there ARE lots of ways  to connect them so that they  produce less .

Someone  here said  the other day that  a neon light  can  be  used to  drain off  extra  voltage .......Thank you whoever it was .
Now  .....once I get my  setup right .........I will need to add a secondary   that produces  90 V  or so at  the   output  that I want it to run  at .
I will have to make sure that the  feedback to source is MORE than enough to keep the battery  charged .

I will then connect  the neon  between   the output  and  a  place  where it will provide  negative feedback  when it is  conducting .

If I am right the neon  will also  be a good   status indicator
WHen the circuit is   running but not powering anything  the neon should  glow brightly .     when it is   running   well within its limits  the neon  should  glow   just a little ...... (  only  the  very tops of the  waves  will  pass through it )
If it is loaded and the    neon is  not lit at all .......  you are  probably  overloading  your  circuit .  You  are no longer  reaching  the normal  operating  voltage .


gary

[xee2]

The advantage of adding the base capacitor is that it reduces the length of time the transistor is on and thus reduces the amount of current used from the battery. Less battery drain means that the circuit will run longer from the battery. Note, this is still at the theory stage until verified by someone with scope.

EDIT: This theory now seems wrong to me. OK, so the base cap increases the time it takes the transistor to turn on, but once it turns on it should stay on the same amount of time as before. Thus I do not see where the savings in battery current is coming from.

[Artic_Knight]

@Altrez,
You are doing much better with your experiments, than I did with my setup, but I can see you are on the same plane of thought I have had previously.
This brings up a point I have noticed through out this thread, several of us here at times have been replicating without knowing someone else was thinking or had done a previous work, and it is almost identical, I think this is an indication that we are heading in the right directions.
Congrads Al on being able to recharge your batteries, that in itself is a big leep forward.


The cap I used was a 35v DC 1000uF.
I wound a many many turns secondary and made a full wave bridge rectifier using 4 x 1N4004 diodes, the cap went on this output, then I soldered a switch (I just love switches) to the 12vDC CPU fan. When the cap charged up, I pressed trhe switch and the contents of the cap discharged into the fan.

I also experimented with trying to harness some radant energy by dumping the cap contents into a buzzer coil I modified, the coil worked, but not as well as I expected it to, but this has scope for further looking into.

This experiment I made was in the early days of this forum thread, no-one else at that time had progressed thus far and I suppose I just ran outa steem, speaking of steam, I wonder what IST is up to these days?

hooroo Al

jim

not to be a buzz kill but what you did is litterally a knock off of the bedeni school girl curcuit and really doesnt demonstrait anything revolutionary. if you can get a RC electric car to drive for more than 30 minutes off of 2 battery packs that are set to a auto switching and recharging curcuit off of this setup then i will be impressed as i have not seen that done yet!

maybe i should start drawing that up?.... shouldnt be hard to set up a charge and use switching curcuit based on voltage from the batteries.  and this kind of design could make electric cars more applicable... hmmmm 

but honestly to give credit where it is due, you have achieved the lowest draw. if you account for the recharge recovery. if you account for strait draw then i havent seen anything better than what others have done so far.

just a thought.  make it work for a car. hint hint.

[stprue]

stprue

I think  the pot is actually OK
It is  the  voltages that are changing  as the battery either  charges or discharges
If  we can keep  the  voltage  in a narrow  range  the  problem should  be solved .

Yes, and what you said would I feel anyway be ideal!

I do not  remember  what  your circuit is like
What I am working on is  collecting flyback  pulses  from  JT coils
There are not  alot  of ways to hook up the coils  so that they add  together .....there ARE lots of ways  to connect them so that they  produce less .

I am working on Hazens1 style coils since I still don't know how to make MK style coils.  I am getting 355v rectified and can light leds/neons with it.  I am also using supercaps and will soon be using voltage multipliers.  I feel that soon one of us here will feed back to the source or at lease combo with other technologies such as Pirate is doing with his EB/JT combos.  NICE WORK Pirate!

Someone  here said  the other day that  a neon light  can  be  used to  drain off  extra  voltage .......Thank you whoever it was .
Now  .....once I get my  setup right .........I will need to add a secondary   that produces  90 V  or so at  the   output  that I want it to run  at .
I will have to make sure that the  feedback to source is MORE than enough to keep the battery  charged .

To me it seems less of a high voltage problem and more of amperage/Farads problem!

I will then connect  the neon  between   the output  and  a  place  where it will provide  negative feedback  when it is  conducting .

This seems like a very interesting idea.  Please post your results when you can!

If I am right the neon  will also  be a good   status indicator
WHen the circuit is   running but not powering anything  the neon should  glow brightly .     when it is   running   well within its limits  the neon  should  glow   just a little ...... (  only  the  very tops of the  waves  will  pass through it )
If it is loaded and the    neon is  not lit at all .......  you are  probably  overloading  your  circuit .  You  are no longer  reaching  the normal  operating  voltage .

Definetly!


gary

Overall I just though an autoadjusting POT would be an interesting tool