An article in NASA Tech Briefs v 35, n 5, May 2011, pp 16-18 caught my attention. The title is "Microinverters are Launching a Solar Renewable Revolution" and it was written by James Washburn, a project manager at Siemens Industry, Inc., Alpharetta, GA. The last time I did anything with PV, people were combining the DC outputs of PV modules and feeding the result into huge inverters that incorporated maximum power point trackers. These had to accommodate an array of PV modules that might not have uniform access to sunlight. PV modules were connected in series, and these series strings were combined in parallel. If one PV junction was shaded, it produced no hole-electron pairs to carry the current of the other junctions in series with it, so each junction required by-passing with a forward-connected diode, and each module in a series string had to be by-passed with forward connected diodes. Then the partially-shaded series strings were combined in parallel, even though they might have different voltages, and the maximum power point tracker had to make the best compromise to match this mess up to the grid. Maybe I don't have this story right, in which case feel free to correct me.
The article shows a cute little Siemens module that connects to a single PV panel and inverts the output, synchronizes the phase with the AC line, adjusts to extract maximum power from the panel, and checks to see whether the AC grid is intact, so that we are not energizing a grid that some service man thinks is dead. I don't know how all this gets done, but the article implies that all these microinverters can be combined in parallel across the AC line without adverse interaction. I understand simpler things, like the large direct currents, DC arcing and grounding problems that these microinverters eliminate. That's it for now on this subject. I just wanted to call attention to some significant progress in PV and power electronics.