Kinetic Solar Cell

It turns out the bimetal "jumping disks" that used to be sold by the old Edmund Scientific required quite a bit of delta T to pop. But it also turns out that the delta T needed is inversely proportional to the radius of curvature of the disks. A big, shallow-dish shaped one wouldn't need much of a temperature change to pop one way or the other. So apply a piezoelectric coating to one side of, say a 4in. dia. one and paint the other side black. Lay it in the sun, black side up and wait for it to "pop", whereupon it generates a small electric current. If some of this power was used to darken an LCD shutter (used in fast-acting welding masks) placed above it, the disk would cool and pop the other way. Shouldn't take much to turn off the LCD, so most of this half of the cycle is surplus and could be shunted to storage. When it heats up again the cycle repeats. An array would sound like popping popcorn!