The short answer is that I found that roof life expectancy was the real factor in determining whether the system was a good investment. The math is far more fa

The short answer is that I found that roof life expectancy was the real factor in determining whether the system was a good investment. The math is far more favorable if you install a 25 year roof at the same time you install the solar panels (and assuming that you needed to replace the roof anyway, or it was a new install on a freshly built house). Roughly, the math for my situation worked out like this. $20K out of pocket for a 13.5kW system, installed. This should, over a year's time, cover close to 100% of my average consumption. Break-even time was something just over 11 years. Existing roof is 10 years old. Reasonable life expectancy of the roof is 15 more years. Value of panels in 15 years is likely to be dwarfed by uninstall/reinstall costs when roof is replaced, so I call that EOL for the system. The most optimistic scenario I can come up with, assuming PGE raises their rates 2.5% a year, is an ROI of 56K, which works out to about 10% annually. Not bad. But this only happens if nothing breaks, the roof really does last 15 years, the installed cost is only $1.5/W, and I invest $20K out of pocket (i.e. no loans). So the way I see it, this is something to do for the non-economic benefits, e.g. installing battery so it will work when the grid is down.