Ok, so my prior wrapup post was primarily concerned with comparing the different arrays I installed, and seeing how much the direction of each array impacted generation of the panels within that array. I recently got the last bill I needed to put together my big overall tracking spreadsheet, so I thought I'd share that here:
For context, I get numbers from a few different places:
1) the Unitil smart meter, which records incoming power, outgoing power, and net power at the meter
2) my emporia vue power monitor, which for this purpose records total power consumption at the electrical panel, and total power production from the solar panels
3) the enphase gateway, which gives me total power production from the panels, as well as per-panel data. I am not using the enphase data for this particular analysis.
Why does this matter? Well, if I only look at the Unitil data, I only see power that crosses the meter. BUT a lot of the time I am making power at the panels, and burning it in my house at the exact same time. This power never crosses the meter. I call this power "concurrent consumption+production" power, and it's part of the bigger picture.
Ok, so walking through the spreadsheet:
Columns A and B are the billing month start and end dates, important for linking meter data to power monitor data.
Columns C, D, and E come from my Unitil bills. I get other Unitil bill data as well (like $), but that's hidden in a separate tab with bill data. I might make a separate post about that.
Columns F, G, and H come from my emporia data export. They are aggregated from daily data into billing-month bins.
Column I is that concurrent consumption + production number. It's calculated with the difference between emporial produced power and unitil received power. Kind of academically interesting.
Column J is variance in the "net" between Unitil and emporia. It is interesting because it shows that where (and when) you are measuring your power in the system tells you different things. I think a lot of it is driven by slightly different time-binning, since you can sort of see row 11 and 12 "make up" for each other...
Columns K and L are how much of the preceding month and year worth of power consumption have been satisfied by solar production, based on my Emporia data. The year data isn't super meaningful until the last month, because I've only had the system turned on for a year.
Columns M and N are the same calculations, but using Unitil's data. The difference is small, but I think important to consider, since Unitil is what matters from a $ standpoint.
Columns O and P use the monthly gross electricity cost per kwh (calculated in my Unitil bill spreadsheet) and multiply it by the monthly power generation recorded by my Emporia monitor. This gives my best guess at the true value of the energy I have produced at a monthly timescale.
Column Q just starts with the total out-the-door cost of the system ($19,300) and subtracts off the monthly generated-electricity value. If we take this as a typical year, we would expect to need another 4.5 years to pay for itself, for a total of 5.5 years to break even (not counting opportunity cost). Instead, I am leaving with a net cost of just over $15k, since we are selling the house :|
Finally, here are some plots.
This one is from the summary data we just talked about:
These 4 are from the unitil billing spreadsheet:
.jpeg)
.jpeg)
.jpeg)
.jpeg){kind=link}