Great ideas. I've been inspired, not to be greedy, but to be efficient. As most of my work on this so far has been at a highly experimental level (that's code for hobby/A.D.D. fix/...) I'm trying to get the most out of a micro sized system. My wind generaotr is a 4 phase stepper. I've got each phase pair routed through a bridge rectifier. This gives me 2 seperate, but roughly equal outputs. I've been mulling over weather to connect them in series to boost volts, or to connect in parallel to capture combined amps once I exceed my minimum. I didn't want to waste on the high speed/high energy end, nor did I want to loose all energy produced below threshold.
I must say the relay idea sounds simple enough to work. I like it. I think I could use a DPDT relay that, when triggered, would switch the mill output from a regular charging path to a double then charge path, or back again.
I'm thinking of using a PicAxe chip to control much of my process, primarilly for logging purposes, but this would enable me to do some easy computing to flip the switch. The PicAxe will be monitoring volts, amps, SOC, mill RPM, windspeed. Using these factors, it would be easy to if-then a result. This way it could even decide weather a voltage doubling would get it over the threshold or not. If not, don't bother doubling. Also, it could decide when (after a few calculations and tests I'll know when) to drop the doubling circuit to allow the RPMs to rise in a stronger wind thus resulting in the needed volts, and at improved amps.
On another note, in my bridge (GBL005) I believe I remember seeing that it loses 1vDC on the output (similar to the 1N4001). Anyone had experience with this chip? Sounds like I might be better off making on from schottky diodes. Anyone?