Solar Net Metering Payout

The following is probably a little geeky, but that is the nature of this subforum:

I have been seriously considering adding about 3.7 kW of grid tied solar to my house. The only place for it is on a SE facing garage roof that is shaded in the morning and is blocked by a neighbor's roof in the late afternoon. I think I can get a good 6 hours of strong sun, but off a bit as it faces SE not due south, and the roof angle is 30 degrees whereas optimum is 42 degrees.

S-E priced up a 3.7 kW system consisting of ten 365 watt panels and ten Enphase Microinverers plus mounts, combiner and disconnect switch which was $8600. Call it $9,000 with cable and misc hardware That is kind of a neat system because all ten pairs (panel and microinverter) are independent and if one is shaded or fails, the others should still work. You tie the 220V L1 and L2 output to an available double pole breaker on your main house panel and the power flows, but.....

Connecticut where I live and Eversource our utility have sensible but not liberal rules for how excess power is handled. Since all of my power will be generated in the middle of the day and it should be several kW on a sunny, summers day, that is more than our home uses. We rarely use the A/C in the summer which would be a potential high power hog but not for us. So all of the excess between what we produce and what we use is "banked" to use Eversource's term. At the end of the year that banked power is paid back to us at some low rate. The terminology is weird but I am guessing it is the cost of Eversource buying or producing power and does not include the higher cost of distributing it. Today the supply is about 9 cents per kW and the distribution is about 11 cents per kW giving a total variable cost for power of $0.20 per kW.

I am not sure that if we use more than we make, such as would be the case when the sun goes down, if we can then draw on that banked power. I suspect not. It would significantly improve the economics if we could.

I ran our situation through NREL's PVWatts calculator which is rather comprehensive and it calculated that we would produce a total of 5,200 kWhrs in a year. If we assume that 1/3 of that is produced when we are using that much and the other 2/3 is banked, then we would save 1733 * $.20 = $347 plus get paid 3467 * $.09 = $312 or an annual total of $659. FWIW our total annual power bill before the solar system is about $1,500.

Presumably I will get a 26% tax credit on the system leaving a net cost of $6,600 since I will DIY the installation. So I get a ten percent return. Not great but not horrible. My installation is difficult because of the direction it faces and the morning shading. There is no other potential location at my house due to tree shading.

It makes me wonder how a perfect installation site would work in our area. Panels would face due south and be oriented 42 degrees (my latitude) not the 30 degrees of my roof, with no shading. Power would probably go up 50% and my return would be a nice 15%. But it makes me wonder how solar installers make out when they get paid by power savings. I bet it would cost $3-5,000 to professionally install such a system. Maybe I would get10% and the solar installer would get paid out in ten years or so. Doesn't sound too lucrative here in CT though.

Our local high school just put in a large, maybe 100 kW system and I understand it is underperforming. No doubt someone was looking through rose colored glasses when they bought into it.

Thoughts?

David