Joule Thief

[stprue]

conradelektro

nice job on the coil.  It is great to see how someone else may do a job differently.  Get some great ideas. Like  (why didn't think of that?)

I have tested the super cap Idea and so far it is a flop and could have a place in my looks nice but don't work pile. 

I have since gone back to my original coil and it will run the circuit and charge its self but not another battery on the chg winding.  I noticed this the other day and seamed to have something do do with the time of day!  Don't know.  Seams goofy but I have seen something like this before in other projects using radiant energy.  Will let run and see if it will do better on charging secondary battery later tonight.  Was working this morning early. I start my day at 3:00 am so was ok then before I disconnected.  If it starts working I will momentarily hook up super cap coil to see if there is any change.  My friend said this morning that his was doing some things it didn't do last night.  He starts at around 9:00 am so I am not sure if it is the time of day,  when that problem starts.  Will keep an eye on results and time of day.

I had hopes for the supercap, so I will keep my fingers crossed.  This particular design (with the battery may still be a success) but just may only work with a battery.  It probably has to do with the size, energy density and that fact it makes energy by chemical reaction.  Time of day can also have an effect as you mentioned for the typical reasons, i.e. temperature, stray radio frequencies etc.  My only other suggestion might be to try a dry cell battery, if you event want to. 

P.S.  You work quick  good job.

[drodenbe]

stprue

Thanks.  I work fast because I don't have to go to a job for 8-12 hrs per day!   that makes a big difference. 

I am still running the first coil and it is 3:20pm and both batteries have settled down.  My run bat is setting at a charged value and the charge bat started to charge but real slow like  one 1 thousandth of a volt per minute.  I have not tweaked the pot because I was building another project while this one did its thing.  Will hook up the super cap later tonight but I don't have much feeling for success.  I believe also it must have to do with the chemical reaction..  Will let you know. 

I saw a video by whoopie may be mispelled but he had a kacher circuit hooked up and had a coil around a 12ah lead acid battery and charging it.  I set up his experiment but had no success that is why I was surprised that this one worked.  I don't really think it is that great but at least it is something.  I was wondering if anyone ever learned what the guy with the ali coil was doing.  He claimed in his video he was charging a cap that it was running on. Never found much else from him but him gloating on his videos and not sharing any info.  I may be wrong.  He may have coughed it up and I just didn't find it.  If any one has something about it worth sharing I would be grateful.

[drodenbe]

one other thing I forgot to mention.  FYI.  You might try looking at Fair radio Sales Company.  www.fairradio.com.  This is where I picked up the HP volt meter.  I got it for 150.00 I believe.  Had to fix it though.  Just a small filter cap.  They are built very well.  It has saved me a lot of time setting around wondering when that second digit is going to change and which direction.  With this thing I can tell with in a few seconds if I am charging or discharging.  That way If it don't work at all I can put it on the shelf faster with the other pretty things that don't work.   

[stprue]

Great, thank you very much for the link!

[conradelektro]

First experiment with drodenbe's JT circuit:

I attach the circuit as I replicated it and a photo of the test which is running since two hours. I already see a small voltage drop over battery (NiCad AA sized rechargeable accumulator, 1200 mAh), but it is better two wait 24 hours to see it more clearly.

My replication has a few minor differences to drodenbe's original:

- 220 µF (instead of 270 µF)
- 0.44 µF (instead of 0.47 µF), I also tried 0.54 µF and I did not see any effect
- 100 Ohm and 25 K potentiometer at the base of the transistor (instead of 27 ohm and 20 K pot).

The big difference is the coil: I used 0.315 mm enamelled wire (and not 0.75 mm as drodenbe), see my Reply #16085above.

One worrying fact is that I do not get anything useful out of the third coil, so I leave it open.

An other strange feature: no matter how I connect the coils (in and out reversed), the LED always lights up. But one particular way makes the LED light brighter (and I use now this direction of connecting the coils; there are four possible ways.) But whichever way I choose, there is almost nothing coming out of the third coil (the one with 130 turns).

I measured the power consumption with a 1 Ohm shunt as indicated by the red arrow in the circuit drawing. I can vary the 25 K pot between a consumption of almost 20 mA (only the 100 Ohm are left at the base of the transistor) and less than 2 mA (about 25 K at the base, in parallel to the 0.44 µF capacitor).

I set the 25 K potentiometer so that the power consumption is about 3 mA. 3 mA seems to be reasonably low (about 18 K at the base).

The circuit seems to swing at about 230 KHz.

Further planned tests:

1) up to 100 K at the base to reduce the power consumption as low as possible (may be 800 µA, the LED just glows).

2) insertion of the NiCad AA the other way round into the coil.


@drodenbe: do you have any suggestions how I should change the circuit? Do you see any errors I made?

Greetings, Conrad