Joule Thief

[Pirate88179]

@ Jeanna:

I have been putting all of my caps in the power strip on the breadboard.  I have not tried any on the base yet but I will try that when I can.

@ Mark:

Man, those two coils sound like they work like a transistor.  That is amazing.  They could have been using those instead of tubes prior to the silicon transistor.  Unreal.

Bill

[xee2]

@ jeanna

The cap across the battery should be at least 0.01 uF. If it is an electrolytic then be sure to put the side with the strip to the negative battery terminal. I am using a 10 uF battery cap because that is all I had handy. It does not need to be that large. If you use a larger cap it may take a short time for it to charge up before the circuit will work. The base cap is a good way to get more voltage out with a small battery current. It does not produce an output with more power. The LED is actually only on for a short time during each cycle (it is blinking very fast). The power savings comes from the LED being off most of the time. When you added the cap across the base resistor it caused to circuit to only be on for a short time during each cycle and less power was delivered to the secondary (and less power was taken from the battery). The base capacitor should allow you to get a larger output voltage from the secondary with a small battery current. Do not put a capacitor across the secondary leads or you will drop the higher voltage created. Using the base capacitor should allow you to have your LEDs look as bright but run much longer from the battery.

[Koen1]

@ Mark:

Man, those two coils sound like they work like a transistor.  That is amazing.  They could have been using those instead of tubes prior to the silicon transistor.  Unreal.

I'm not sure but didn't they? Isn't that one of the uses of "magnetic spark quenching"? Well, not necessarily, I admit, but I seem to recall they did use
something like this to either block currents from jumping a spark gap or assisting them in jumping the gap... Or am I confused again? lol

Well anyway, after all the interesting posts you guys put up here, I'll really have to build myself one too!
So, anyone got advice, which is the most efficient/effective setup so far?

Regards!
Koen

[gadgetmall]

@all

I made research on magnetic amplifier

Quote"
The mag amp is a "magnetic field" kind of amplifier and is of Class H type.

Visually a mag amp device may resemble a transformer but the operating principle is quite different from a transformer - essentially the mag amp is a saturable reactor. It makes use of magnetic saturation of the core, a non-linear property of a certain class of transformer cores. For controlled saturation characteristics the magnetic amplifier employs core materials that have been designed to have a specific B-H curve shape that is highly rectangular, in contrast to the slowly tapering B-H curve of softly saturating core materials that are often used in normal transformers.

The typical magnetic amplifier consists of two physically separate but similar transformer magnetic cores, each of which has two windings - a control winding and an AC winding. A small DC current from a low impedance source is fed into the series-connected control windings. An AC voltage is fed into one AC winding, with the other AC winding connected to the load. The AC windings may be connected either in series or in parallel, the configurations resulting in different types of mag amps. The amount of control current fed into the control winding sets the point in the AC winding waveform at which either core will saturate. In saturation, the AC winding on the saturated core will go from a high impedance state ("off") into a very low impedance state ("on") - that is, the control current controls at which voltage the mag amp switches "on".

A relatively small DC current on the control winding is able to control or switch large AC currents on the AC windings. This results in current amplification."

From http://en.wikipedia.org/wiki/Magnetic_amplifier

Hi . ok see if i have this right . we make two toroids . wind them serial with the primary- Power that from lest say a Jt which supply the Dc pulses then wind the secondary serial but either wire the two either series or parallel to each other and power that with say a 555 timer set up to produce a Square wave  . Will that work ? With a Jt Circuit that will only Draw a Variable Current draw of say .20 to 3 ma Leaving some extra 10 ma to power the 555 timer  ?If so i will start on the  555 timer . I think the only other way to get AC would be a using a few transistors to build an astable multi vibrator, Correct ?
AL

[Goat]

@ All

Congratulations to all on your concerted efforts on this thread.

Been watching this thread for awhile and noticed that there are a lot of variations on the circuits so here's one without electronics for you all.

This is using a joke buzzer that gives you quite a jolt btw, I used it to light an NE2 bulb.  I took out the 3 button cells and ran it from a 1.2 V AA battery, I even had it going through a bridge rectifier which would charge a capacitor to over 100 VDC then another NE2 across that cap would bleed the charge back down to 82 VDC and start all over again.

I didn't try lighting LED's because of the high voltage