Actually I think this might work rather well...
So, I'm looking into cheap (key word) means of automatic tracking and in the midst, came up with one of them retro-saves...
The fan idea wasn't exactly the best... It helps temp wise, but I'm not sure to what degree (heh get it?) because I don't have an IR thermo. Heat is very evident in the exhaust, although not more than luke-warm-ish in the best of sun... The cells are dropping but I can't tell how far... and if there's output gratuity, I can't find it.
However, it got me to thinking... Why not power the tracker with the aux panel (which was brought up)
but add a battery, dump the excess to the fans, and save my little mishap for one of these here stories.
And, I'd have a small 'reserve' available to me, should I need it...
It kinda feeds into the next part of this...
I set up a rinkydink solar meter using an old button cell tester and a yard light panel. Just so happens that ~1A of buck output coincides with the very onset of full inflection (repulsion type meter). This was moderate overcast. Not bad... IF -
I tipped the panels and pointed them straight up at the sky. Incident light apparently isn't always best...
So long story less long, considering adding a brain to this thing to tell it that if the average of the sky is better than any one direction, use it.
The idea would involve a rather goofy mod to 'reverse' a 70s fiber lamp (funnelled to a CDS cell), inside some kind of clear globe. There would be two CDS cells, to tell the difference between the combined average sky and the incident light value. This would be up above and in addition to the sensors required for the 'existing' XYZ tracking device.
It would also have hysteresis to prevent unnecessary hunting, maybe a couple of minutes worth before heading for a decision. [Probably integrate this into the H bridges actually... ? * for me, not you
]
Maybe, maybe not.
Besides, cooling never
hurt anything...
Steve