Solid States Devices > Joule Thief

Pulsed DC JT by TK Labs

(1/3) > >>


This is an unknown device created by TK
It appears to resemble a joule thief (at least that was my take on it)

Im not sure if this device was intended to operate in this mode,
However, attempts to fire this up using constant DC have thus far failed.

So i am using a high frequency pulsed DC i optionally added some external inductance in the power lead using a bi-filar tesla coil that TK made. (im sure any coil would have sufficed)

Heres a short vid

Bob Smith:
Thanks for sharing this. Interesting deployment of the spark gap. Your series wound bifi coil should be upping your voltage tremendously, and taking advantage of it with a spark gap makes sense to me. The spark gap, following the logic of Don Smith, should render your setup an open circuit. Being no longer closed, it can access, via the spark gap, the ambient charge present in the electrostatic environment. I would invite you to consider putting a larger LED load on this - this, I believe, would give you a clue as to whether the bifi coil + spark gap combo has rendered this setup load-driven, versus battery-driven.

If you can put an inline volt meter after the battery's positive lead, you may find the battery draw actually dropping as you light up a load, indicating that the draw is coming not from the battery itself, but from the electrostatic environment.

Again, thanks for this - very refreshing with possibility.

So i took some measurements around the circuit
The bifilar seems to be ramping it up from 0.2v to 0.5v
But only with the gap, if i directly connect the battery
Im reading 0.2v and the device doesnt power up under 0.3v

Theres a 0.1v drop across the shotkey and
im not seeing any voltage drop across the lit LED
at least not that my meter can detect.
0.4v before and after the led to ground, and across it.

So its holding right in its’ operating range.
I have not yet needed to push the ferrite deeper into the variable inductor
We’ll see how long this battery hangs in there


Day 6, the battery is finally showing signs of depletion,
Ferrite is now 3/4 inserted before it kicks on, and not quite as bright anymore

Battery has dropped to 0.18v
(still reading a steady 0.4 across the led)

its an extremely efficient little JT, specially since i can adjust the inductance
and ‘tune’ it back in.

For those that don’t understand what that means:
this is somewhere close to the min cut-on inductance for the transistor w/ the circuit
(generally speaking)
in this range is where you will find the most efficient rlc frequency
brightest light, max output with min input, etc.
and where you will find a resonant node for the inductive part of the circuit.
To visualize this on a scope, a secondary winding is useful to separate the inductive signal
from the digital switching spikes of the transistor.

this can be cleaned up by adjusting the gate resistor, to achieve a nice waveform
(on vs off switching times and waveform symmetry)
Note that doing so will change the frequency requiring an associated adjustment in inductance
to ‘tune’ it back in.


--- Quote ---Day 6, the battery is finally showing signs of depletion,
Ferrite is now 3/4 inserted before it kicks on, and not quite as bright anymore

Battery has dropped to 0.18v
(still reading a steady 0.4 across the led)
--- End quote ---

I don't use batteries anymore and only use input/output capacitor banks. Capacitors do not lie and are natural integrator-accumulators. Whatever the voltage or current going into or out of the capacitor it always shows us the energy based on the voltage change ie. Energy = 1/2 C V^2. We can also get an accurate measure of the energy flow instantly with out having to wait. Ergo... there foolproof.

I use a voltage divider or opt isolator on the input and output capacitors tied to an Arduino to monitor the Energy and the change in energy which gives me power and current flow. I included a picture of my older basic setup below and there's not much to it. I also use a data logger on the Arduino so I can save my test runs however the new Arduino IDE has graph plotting. In the picture below the whenever the total energy in the input/output capacitors changes we know whether were losing or gaining energy within the system.

Another neat trick is using the input/output capacitors to easily loop a circuit and use a separate isolated PMW circuit to add energy to the input capacitor to maintain it's voltage. Thus whatever power(VxI)/energy(VxI)/T is added to the input capacitor to maintain it's voltage is our total circuit losses. As I said it's basically foolproof and if the capacitor(s) voltage is dropping we know as a fact there is no gain to be had...period.

I just got tired of all the BS and wanted a foolproof method to show me where I'm at with respect to an energy loss/gain. That's what all this is about isn't it?, to know where were at and what direction we should be heading. It's strange and while many are preoccupied with current or voltage my only concern is Energy, Energy is what defines what happens in these systems. Understand Energy and Free Energy becomes much easier to understand...



[0] Message Index

[#] Next page

Go to full version