Self Running Micro TPU, with closed loop.

[sm0ky2]

@ groundloop


Basically, we had this equation in one of our EE classes, for coil design.

using 10ga insulated wire, and a  pair of caramic-ferrite rings (that we had stripped the wire off of)
we designed the coil to pick up the ambient 120v/60Hz (U.S.) EMF that was circulating throughout the building.
the rest of the circuit was just a resistor/cap to ground to balance the load and a square-wave rectifier to give us a D.C. output.

i think our final outcome was around 3.5v @ 150ma, hardly any power to speak of, but it lit the 2 led's and made the small fan turn. 

It may have nothing to do with the device in the video, his is a lot smaller than ours, it just made me think of that the way he was holding it close to the wall when its working.

[Groundloop]

[EDIT] Deleted.

[plengo]

I think I figure out how this thing works and why the voltage "seams" to go up. Thanks to Groundloop, the gate of the mosfet is holding a capacitance which holds, as when we touch the gate, some voltage that eventually runs out. As it runs out the drive-source current diminish and current consumption goes down and the voltage goes up (because of the battery "ghost" voltage restoration) and we don't notice anything because those LEDs seams to not change in brightness since they are very good in showing light with very little current.

As the brightness goes down very little per mili-mili volt and it is not noticeble by human eyes but the meter is very sensitive and we only notice that therefore the phenomena seams to be that the battery is recharging itself up.

Now, with all the being said, aside from the fact that I am very piss at it, but hey, that science, inst it?!!! Unless someone come up with a real voltage gain with measurement with the battery disconnected from the device I think my explanation is the final explanation. All the tests that I have been doing, duplications and trials I concluded that thats the case.

Some of my tests demonstrates that with 5 IRF510 with only 2 I was able to replicate well the device, I think there is a huge difference in those parts concerning that capacitance in the gate and its leakage. Groundloop?

Anyone?

Well, at least I am happy that I figure out how to make a device that is definetly hell of efficient in keeping low self-consumption and good light control.

Fausto.

ps: [EDIT] but because I am very persistent and I am to make sure my theory is backup by data. I am now leaving connected the amp meter in series with all the LEDs and make notes of the amperage too, so that brightness of the LEDs will be irrelevant (only for entertainment). Voltage cross battery, current going through the LEDs are now my data. Let's hope for a better future.

[Groundloop]

[EDIT] Deleted.

[Bruce_TPU]

I think I figure out how this thing works and why the voltage "seams" to go up. Thanks to Groundloop, the gate of the mosfet is holding a capacitance which holds, as when we touch the gate, some voltage that eventually runs out. As it runs out the drive-source current diminish and current consumption goes down and the voltage goes up (because of the battery "ghost" voltage restoration) and we don't notice anything because those LEDs seams to not change in brightness since they are very good in showing light with very little current.

As the brightness goes down very little per mili-mili volt and it is not noticeble by human eyes but the meter is very sensitive and we only notice that therefore the phenomena seams to be that the battery is recharging itself up.

Now, with all the being said, aside from the fact that I am very piss at it, but hey, that science, inst it?!!! Unless someone come up with a real voltage gain with measurement with the battery disconnected from the device I think my explanation is the final explanation. All the tests that I have been doing, duplications and trials I concluded that thats the case.

Some of my tests demonstrates that with 5 IRF510 with only 2 I was able to replicate well the device, I think there is a huge difference in those parts concerning that capacitance in the gate and its leakage. Groundloop?

Anyone?

Well, at least I am happy that I figure out how to make a device that is definetly hell of efficient in keeping low self-consumption and good light control.

Fausto.

ps: [EDIT] but because I am very persistent and I am to make sure my theory is backup by data. I am now leaving connected the amp meter in series with all the LEDs and make notes of the amperage too, so that brightness of the LEDs will be irrelevant (only for entertainment). Voltage cross battery, current going through the LEDs are now my data. Let's hope for a better future.

@ Plengo

I am sorry.  This too is what I was speaking of yesterday.  No true charge of amperage.  But it was fun! 

Back to the trenches.....

Merry Christmas!

Bruce