JT first attempt

[lygeas]

a lot of notes on experiments I did with different winding ratios...

http://www.overunity.com/11986/what-i-learned-in-joule-theif-101/#.UokacMS23zE


The size of your core will affect your frequency, but the output per turn will remain about the same.
It takes a longer time to build up a field in the toroid, and then inducing directly into the other winding, which causes the transistor to close, and all the magnetic flux in the toroid to stop having a current supporting it, so it falls out into the windings. 


that's how I interpret it anyway.

Dear d3x0r, I ve read your topic and I will try to remember the following

"Fewer windings on the coil connected to the collector increases the overall frequency, fewer windings on the coil connected to the base shortens the length of the pulse generated, but decreases the overall frequency

Frequency depends on toroid size, a larger toroid will be a lower frequency"

Thank you for your reply

[TinselKoala]

Something to stimulate your quest:
(note: no toroid, a single AAA battery and _six_ 90-volt NE-2 neons in series.)

(note2: the components on this side of the battery comprise a wireless power receiver but are not used in this photo, it is running off the depleted AAA battery alone.)

[lygeas]

Something to stimulate your quest:
(note: no toroid, a single AAA battery and _six_ 90-volt NE-2 neons in series.)

(note2: the components on this side of the battery comprise a wireless power receiver but are not used in this photo, it is running off the depleted AAA battery alone.)

Cool circuit Tinsel. I can see on the left of the neons something like a coil. Is that right?
. What are your measures? (mA, volts)

[TinselKoala]

Cool circuit Tinsel. I can see on the left of the neons something like a coil. Is that right?
. What are your measures? (mA, volts)

The inductor is a thing I salvaged from a large CRT television set chassis. It was originally wound with just one winding, around 300 turns (estimated) of fine wire, and measured 10 mH on several of my measurement systems. I wrapped a layer of cloth tape over that winding and then put on 30 turns of heavier wire over that, in two layers. Every TV I've dissected has a similar inductor in it, most of the time smaller in size and sometimes including a biasing magnet in the structure. (Many small inductors of this type use magnets in the cores as a way of "cheating" and getting more inductance in a smaller package.) Try with and without a magnet. This inductor was originally hidden inside a white plastic box housing on the TV circuit board.
The output voltage, with no load, goes over 800 volts peak. The circuit produces very short-duration spikes. The circuit works with an input voltage down to a bit less than 1 volt from the AAA battery. Using the wireless receiver, it gets more input voltage than that and the neons get really bright. It doesn't light the neons unless the diode is across the output in the orientation shown. The variable gate trimpot resistor needs to be carefully tuned, and also tweaked a bit for max brightness as the supply voltage changes. The heatsink on the 2n3055 transistor is just for show, I call it a "jason mask". Not all 2n3055s will work, for some reason; some are better at handling the HV spikes than others, I guess. I don't know what the current demand is but the battery lasts a long time.
The circuit is an ordinary basic JT circuit with the addition of the rectifier diode across the output. Here's the schematic and a scopeshot of the output spikes:

[TinselKoala]

By the way, I got to that point by sort of "random" experimentation with the diode and the inductor, based on what I learned from the "JT Pros" like Pirate and Jeanna and the other experimenters in the various JT threads. I learned about JTs really late in the game and my only contribution to what they've discovered is the use of the neons instead of LEDs. So please keep experimenting, trying your own ideas and variations, while also checking out what has gone before, in the different JT threads here.