Joule Thief

[WilbyInebriated]

on a smaller diameter rotor you have to be careful so as not to 'overlap' the magnets fields... you want about a magnet and a half betwixt each magnet. 8 might be pushing it on that vcr rotor, maybe six? i guess you would be the best judge of that. personally, i would play with stacking them first, just because it's easier to do then adding more to the circumference.
more input voltage will give you more rpm, try 20v or 24v on the input. i'd throw a wrap of electrical tape around the edge if you do this, flying magnets hurt
my first bedini sg was on a bicycle wheel which made it really easy to add/remove magnets. played with 1- 12 magnets. i don't have a tach, but if i recall, adding more magnets along the circumference didn't increase the rpm, just reduced the time it took to reach top rpm. damn, i really should keep better notes, sorry.

having fun is what it's all about 

[Pirate88179]

@ Wilby:

Thanks.  That makes a lot of sense to me.  I am going to swap out to two 12 volt lead/acid cells which should, as you say, increase the rpm.  I forgot about the overlapping fields. I did read this somewhere and I am glad you pointed this out.
Yes, my magnets are super glued first, expoxyied second, and have three wraps of tape for safety.  It would be easy to unwrap the tape and try double staking the neos which I will do.  There are so many variables to play with it will keep me busy for a while.  I want to try a JT circuit on the input, and then the output and see what happens,

It spins up pretty fast as it is.  The only thing I have noticed is that if I decrease the resistance in the beginning, and then add it as it speeds up, it goes faster quicker.

Thanks again.

Bill

[jadaro2600]

Just thought that I would chuck this in here: I'm getting better results with 2n2222 transistors than with the 2n3904 transistors.

[Pirate88179]

Jadaro:

That's good to hear.  I had to buy a bunch of stuff to get by the min. and I ended up grabbing a bunch of those as well. I have not tried any of them yet, but this is good news.  They look a little more robust to me with their metal caps. Easier to heat sink if we need to.

Bill

[xee2]

@ jadaro2600

Can anyone tell me why this LED is lighting up like this?

Your circuit is functionally equivalent to a JT with the wires forming the transformer. The capacitor in series with the LED blocks DC but will pass AC and the LED can be lit with an AC voltage. AC meaning any voltage which is not at a constant DC voltage level thus pulsed DC would be considered AC in this case.

As the current starts to flow in one wire it creates an increasing magnetic field around the wire. This increasing magnetic field then creates a current in the second wire. Two wires twisted together make a very simple transformer. Making the wires longer will increase the current produced in the second wire. Adding iron or ferrite near the wires will effectively increase the electrical length of the wires and thus also increase the current produced in the second wire.

You seem to be the only one actually doing JT research. BTW, PNP and NPN transistors are functionally identical except they use opposite polarity voltages and adding two capacitors in series creates a combination with half the capacitance and twice the voltage capability.