Selfcharging cap circuit from Larskro fake or real ?

[TinselKoala]

OK can anyone give a fairly accurate approximation of how much "energy" can be stored in 20 mAh NiMH battery as compared to a 0.47 Farad capacitor both at 3.7 volts.

A 0.47 Farad cap has 3.22 Joules according to the electronic assistant app.

A 20 mAh battery at 3.7 volts should give 20 mA for an hour at about 3.7 volts ? about 266 Watt seconds ?

Clearly I've confused myself. I think. 

Tinsel, another cool little motor. Strobe attack is neat too.

..

Maybe I should try to test the battery and see what I can get out, try to plot a curve or something.

..

I agree with your numbers, except that the battery discharge rate will affect its "capacity" somewhat. So you've demonstrated that the battery holds a lot more energy than the capacitor does, at that same voltage.
So to get the same energy storage you would need, say, 40 F at 3.7 V. That would give you (40F x 3.7V x 3.7V)/2 = about 278 watt-seconds.
Or you could charge to 8 volts... then you only need about 9 Farads or so.

[d3x0r]

For the short time I played with it, I only saw a rundown effect; the motor doesn't really have a lot of flywheel mass... might be a sphere or differently enginerred in others; Any postive spike was invisible; maybe the captance was too high; but once any pulse from the coil of what should have been more then the cap had no appreciable effect; maybe the resistance of the scope probes is high; because right, without the reed switch there should be no power draw, and there was 100mA at 5V something... 

but I also still have the temp sensor and existing LEDs from the device attached in parallel.....

I like the idea of a 4000 turn coil primary/secondary to see the same thing   It's just going to be a longer slowdown... increasing H instead of F

[d3x0r]

Tinsel, another cool little motor. Strobe attack is neat too.
.

Strobe-o-tach I think.


http://www.allaboutbatteries.com/Energy-tables.html


AAA Alkaline
Long-life      1150   milli-Amp
hours (mAh)   For instance

[TinselKoala]

Strobe attack works for me... 

[synchro1]

Meanwhile, here's MiniPulse II, with bipolar transistor instead of mosfet, running down to the limit of the Hall sensor's switching at just under 2.4 volts. Runs well on supercap, makes spikes 3x input voltage but still no voltage climb effect. I can't find any reed switches in my stashbox....   

http://www.youtube.com/watch?v=gC5GiLVCUPM

@Tinselkoala,


There are several very good youtube videos on how to manufacture your own "Reed Switchs" at home. This is a worthwhile approach, mainly because the switch needs a hair trigger custom installed, and a self build allows one the freedom to experiment with tiny biasing magnets to get something real "Witchy" worked up!


The switch has to linger a little bit to carry the reverse fly back power home to source. You would need a complex arduino program and multiple componants, probably a second transistor, to even come close to matching the tiny switch time intervals the magnet levers allow one to create. It's the amount of time the switch stays closed, after activation, or the hysteresis loop that bridges the reverse return power. Maybe nano-seconds!