Joule Thief

[jeanna]

Excellent, Dog812!!,
I like the idea of this. There is more room and you are more likely going to want to change the primary than secondary. Having it right there on top is the ultimate convenience.

I think you should start making one.

If you can light 1 you can light 20 in parallel.

I see you are using 2.2k resistor on the base. That ought to limit the amps draw.

You do not need to do any more tests if what you want to do is put lights into your hhoops and start making some money. Go for it.!.

Have you designed the switch yet? the battery holder etc?
I assume you have made a bunch of these in some form, because you showed the video, but I am not sure what more you need to know.

In fact you could try to get more resistance into the base and see how long the battery can last. That is all I would be doing. There is nothing to stop the business now.

Please let us/me know when you have a prototype.
(or, of course if you have any questions.)

great job!

jeanna

[jeanna]

@xee2,
Well, this is really interesting.

I repeated the little test on the frequency and volts in relation to the number of turns of the secondary today. Since you suggested putting a cap in parallel with the base resistor, I did that too.
This is a small preliminary test. I only cut a piece of wire long enough for a dozen secondary turns.
I set the base resistor at 929ohm. and made one turn at a time.
I measured the cps and the volts with each turn and added the cap every other time.
It was all pretty straightforward or straight lines.
The more secondary turns the more volts came out from the secondary and the lower the frequency.

The cap seemed to lower not only the frequency but destroyed the voltage. Then all of a sudden it went very very high. Both frequency and volts jumped. The frequency went from a dwindling
113KHz and 1.6volts at 9 turns to
384 KHz and 3.72 volts at 10 turns and then
416KHz and 4.11volts at 11 turns.

I took it all out and started over in a systematic way. This was very promising.


It occurred to me that a smaller value cap in the beginning with only one turn might be able to boost things, and I might have picked the wrong size cap for the job, so I ran the test again. I pulled the beginning tail tight in hopes that I could get 12 turns in total.

So, this time I started with a 100pF ceramic which did not help until I had 4 turns.

At 6 turns a 1,000pF ceramic made the output improve over the 100pF, then only at the 12 turn did I need the 10,000pF cap to really improve the results over the 1000pF.

I didn't have anymore wire for more turns at that point and stopped for the day.

The way I did it, was I looked at the data without any cap, then added the cap. When the frequency boost was going down too much and the voltage was diminishing I checked the next cap. I will make a table. these words are too hard.

The really exciting thing happened with 12 turns, and I should make a longer wire to see what happens with 15 and 20 turns.

The frequency jumped to 2.2MHz Megaherz!!! and yes the volts dropped to 3v-5v. It was an amazing thing to see. There was this thing trying to be a sinewave. It was made out of little waves that traveled up and down the larger wave. a wave of waves. Wow. Each little wave traveling on the bigger wave is what is 2.2MHz.

OK xee, please say... what is this? what do I do with this.

I will make a table and screenshot and post that soon.

thank you for that suggestion and thanks in advance for any help ...

jeanna
I will just add the table as pic below.

[xee2]

@ jeanna

This is a list of things that effect frequency as I see it:

WAYS TO INCREASE FREQUENCY:
1. increase base resistor value
2. increase turns in base coil
3. decrease turns in collector coil
4. decrease turns in pickup coil

WAYS TO DECREASE FREQUENCY:
1. decrease base resistor value
2. decrease turns in base coil
3. increase turns in collector coil
4. increase turns in pickup coil

Thanks for the data you are taking. I am still not sure I understand the effect of adding the capacitor across the base resistor. I think the basic function is to increase the time it takes the transistor to turn on again after it has turned off. Since the transistor is turning on less often it draws less energy from the battery. But if a capacitor is also put across the battery, then energy is stored up in the capacitor across the battery between the times when the transistor turns on and therefore when it does turn on it has many times more energy available than it would have from just the battery. So in that respect the capacitor should lower the frequency. The extra energy available from the battery capacitor means that the voltage produced when the transistor does turn on is much greater than it would be without the base and battery capacitors. In addition, I think that the capacitor speeds up the transistor turn off time. The voltage generated across the collector coil will be larger when the transistor turns off quicker. Thus I think that adding the capacitor across the base resistor and also adding a capacitor across the battery will increase the voltage produced by a given size battery.

[dog812]

@jeanna
I am still trying to get the LED's brighter.. They last long enough..
But they just arent as bright as using the proper batteries..
That is my next goal.. finding a way to get em brighter

[xee2]

@ jeanna

When I look at your data it seems like adding the capacitor increased the frequency (opposite of my theory - so maybe theory is bad).

I used the base capacitor to get enough voltage across the collector coil to light a neon (about 100 volts). I do not still have my data, but I think I was not able to get that high a voltage without the base capacitor. Also, I was able to reduce the battery drain needed to get a given output voltage by adding the base capacitor and simultaneously increasing the base resistor value. I think it was Gadgetmall who first suggested this and he also noticed that he needed less battery current for a given output voltage when the base capacitor was added.