I have a dream of outfitting our future home with a solar array and wind turbines, so I spent some of my vacation learning about solar cells and ways I can apply them in the home-building plans.
I wanted to be on the cheap, so I got some $1 solar yard lights to harvest the cells.
To my surprise, I found that they are powered by a 350 mAh Nickel Cadmium AA Battery. There was also some simple circuitry and a small, low watt LED.
I took it all apart, carefully peeling the cell from the hot glue and hooked up a voltmeter. I got about 2.5 volts in overcast and 4 volts in direct sunlight. I discovered that in the sunlight it puts out about 25 milliamps. Not a lot. I guess that makes this a 0.1-0.2 Watt solar cell. Pretty weak, but enough to charge a small battery.
I want to be able to pump out a constant 5v, so I got all my cells out and constructed a small array of three parallel sets of two cells in series. I also made some circuitry to regulate the cells: a 5V regulator, a one way diode, and a few capacitors. In the direct sun, I was able to get 4.7 volts and 100 milliamps. Overcast, the voltage dropped to 2.7.
In only direct sunlight can it get a phone to recognize a charge state, but I would like it to happen during overcast too. I changed up the layout of the solar cells to have 2 parallel sets of 3 cells in series. This upped the voltage in overcast to 4.7 after being regulated, but the amps went way down. I put the array and phone in a place that had good amount of sun (ironically on the watt meter) facing south. Left it there for an hour with a dead battery in phone. Came back and found 2 percent when I turned it on. Not a lot of charge, but got the proof of concept I wanted to achieve.
What would I do differently next time? If I were to do this again (and I might), I would order some cheap cells on ebay. The allpowers 5V 150mA ones look promising. Two of those in parallel would provide enough power to charge a phone while in use. Maybe I will do that next month or so… when the sun finally comes out.