Hey guys, just thought I'd add my 2.5 cents about some observations with the captret idea that ibpointless has posted, regarding my own observations from my own experiments.
While it is true the captret loves voltage, it does not like batteries connected in series as the source. I found that the captret can 'see' the source (if it is in series) and will always drain the cap and batteries.
I've found that the captret can power loads for an incredibly long time, but the end result is a totally dead battery.
I will give an example: purchased brand new energizer headlamp consisting of 3 high bright white LED and a single NIGHT Red LED.
It comes with 3x 1.5v batteries unsure of ma/h. It says on the packet that these brand new batteries will power the device on 'White' LED mode for 23 hours.
I connected up a 9v battery reading about 7v to capacitor in normal captret style to power the Light. (powered it in white led mode)
To be fair I'd say that the light coming out of the Device was around about 1/3 of what it would be if it was operated normally with all 3 batteries powering it.
I ran this light at same intensity of about 1/3 brightness for 970 hours, then the light faded. When I disconnected the battery and measured it. It read in the millivolt range giving random readings. It was like every bit of energy was sucked out of it.
I've found the best use of the energy from the captret is connecting a cap directly to the battery as you would normally to charge it up p-p and n-n. What happens here is that yes the battery will drain somewhat then rise back up and then surpass its starting voltage. then you just tap off the positive and can whilst still connected to the battery. The battery will not drain once the cap is full, it will actually charge. If you connect batteries in series the captret will see the source.
Like Ibpointless pointed out, the captret will power small led's ect for a while but eventually the power will dissapate as it needs to rest to recharge. So pulsing the power seems to be the go.
This charging effect in parallel arrangement works well with caps of any voltage and capacitance, however with batteries for the charging effect to work the batteries need to be matched in voltage. I'm not 100% sure about capacity, but have noted different voltage batteries just drain.
Another thought I had was, well 'If only I had a capacitor that held its voltage without draining, we could tap the captret part forever'. So I theorised that If I charge two caps in parallel from a source and leave them connected in parallel, it might exhibit the charging effect and maintain the voltage. It dosen't. The caps are converting energy from the environment, and they will only start drawing this energy in if the are at a lower potential then the enviroment. Ie; mv. the only way I know how to do this is connect a cap to a battery (and leave it connected) and tap off the captret. I read somewhere on the web that eveready makes a 300v battery, which would be good for a source. I'm not sure if it would work as it is not lead acid, but worth a look anyway. I might buy one next week and see.
I connected 4 caps of different voltages (from 10,000uf to 680uf) and capacitances that had been sitting around for a while. (I shorted them all before starting) Connecting them all in parallel and left overnight. I noticed they hit to 200mv range fairly quickly, within an hour or so of connecting up and a test a few minutes ago theyre at 0.34v and climbing, albiet slowly. I don't know how far these caps will charge on their own.
Hypothetically, if they were leftalone and somehow did charge up to 2-3v or more, that would be something, but take an extremely long time. I think if we had 50-100 or more caps in parallel doing their environmental charge thing, we could have a circuit to switch from parallel to series every so often. Sure it would require a battery, but just for switching. That way, with say 100 caps drawing in a mere 50mv , you could have a 5v discharge at a resonably good frequency, or for a rediculous amount of caps, you could probably achieve a constant ouput. (Oh, and for anyone interested, I've tested this cap charging effect in the middle of the desert, no where near towns, radio transmitters or anything remotely electrical, and it still works
Anyways cheers to Ibpointless, and others who have been working on this idea, it has a lot of potential.