2022 was the fifth year for our rooftop solar system and at 7,569 kWh was the most productive year yet. That’s likely because we added a 4th string of panels to array #2 in October, raising our charging capacity from 6.4 to 7.4 kW. Definitely made a difference in those last 3 months of the year.
We used 3,126 kWh here and sent the excess 4,443 kWh to the grid, saving us a total of $1,240. Electricity costs rose sharply this fall but our existing contract with our supplier didn’t expire until November, but then it rose about 60%, so our rate of savings will increase this year. We’ve earned back $5,618 or 35% of the original investment.
Best month of production in 2022 was July, at 829 kWh, with May a close second. November produced a surprising 600 kWh, a combination of the additional panels plus a lot of sun.
We should see our biggest year yet in 2023 with the extra 1000 kW of charging power added by the new string.
Year 4 is “in the can” as they say, let’s have a quick look to see how the solar system performed in 2021. We generated 7,619 kWh last year, 7% more than last year and the most productive year yet.
We used 3,126 kWh of that power on premises and exported the remaining 4,493 kWh surplus to the grid, slightly more than we exported the first year. That saved us $1,075, the largest savings in one year so far, bringing the lifetime savings to $4,362–which recoups 26% of the system cost.
June was the biggest production month last year at 755 kWh, with March and May close behind. As usual, December was the least productive month although this year it produced a respectable 399 kWh. November was surprisingly strong last year at 552 kWh.
The only maintenance issue that arose this year was the computerized control unit, known as the MATE3s, which had to be replaced after it stopped communicating consistently with the internet (which it does to send data to the cloud-based system monitor).
Hard to believe it’s been over three years since the system went online. Let’s have a quick look at the numbers for 2020. We cranked out 7,125 kWh last year, our best year yet and a 5% increase over 2019. The summer drought, which lasted until late September, probably had something to with that—definitely fewer gray days than normal.
We consumed 3,490 kWh on premises, a big jump from 2019 that can be attributed to the electric car we acquired last year. 3,635 kWh went to the grid, a significant drop from the previous two years again owing to charging the car battery. We saved $1,122.29 on electricity costs in 2020, bringing the lifetime cost savings to $3,306.46, or 20% of our investment. Depending on the price of electricity and other factors, we should recoup our investment around 2030.
Just like in 2019, May was the biggest month of production: 861.9 kWh. As usual, December brought the least production at 295.1 kWh. Interestingly, production was exceptionally high in October, matching June and just behind August. October can often bring gray skies and rain, especially later in the month, but not so much in 2020.
As in past years the system required almost no maintenance—just monitoring and occasional battery checks. Very pleased about that.
Put a bunch of metal-framed objects on the roof, then connect them to the home’s electrical system, and the hazard of lightning becomes potentially quite serious. In doing research I found a wide range of opinions on this matter. Lightning is extraordinarily powerful and to some extent unpredictable–even experts disagree about the most effective mitigation measures. But after studying it I came away with two basic principles that guided my approach:
When lightning strikes an object it will usually take the easiest path to the ground. If the path of least resistance to ground happens to be your array’s equipment ground conductor–or any conductors attached to your panels for that matter–then a massive surge of current could very easily damage or destroy expensive equipment and appliances. Worst case scenario, it could ignite a fire in your house. On the other hand, if you provide an easier path with less resistance directly to ground, bypassing the interior of your house, it’s much more likely the lightning current will take that route instead.
Protect your equipment from current spikes with surge arrestors. Surge protective devices, or SPDs, make a good line of defense for valuable components like inverters or charge controllers. Lightning is potentially so powerful–up to hundreds of millions of volts–that no surge arrestor is 100% guaranteed to protect what’s behind it. But it’s better than having no protection and in most cases will do the job of absorbing/diverting the surge.
How big a risk is lightning? It depends. How prevalent is lightning in your area? And how exposed is your house?
You can answer the question of frequency with your own experience plus data. The map below shows the annual density of lightning by area throughout the U.S. Obviously if you live in Florida or Louisiana the risk is much greater than if you live in Oregon.
This map shows the density of lightning in the United States from 2009-2019, based on the number of flashes per square mile each year.
Where we live in Maine, lightning frequency is moderate but we do get the occasional thunderstorm particularly during the summer. Our house isn’t especially exposed, and there are tall trees near the house (not too near) that make a likelier target than the house itself. But a strike to the house is certainly possible and a nearby strike could still inflict damage.
Guided by the principles above, I decided to take two mitigation measures. First, I installed a 4-gauge solid copper conductor from the rooftop array directly to ground. The conductor clamps tightly onto one of the metal rails supporting the panels, runs down the eave, then down to a copper ground rod driven into the earth several feet from one corner of the house. This ground rod is connected or “bonded” to the rods installed by our electrician at the service entrance–it’s important to bond all the ground rods together (in our case that includes another pair of rods at the location of the inverter and battery bank). This setup provides a direct low-resistance path for the lightning current to reach ground without entering the house/destroying equipment/starting a fire.
A 4-gauge solid conductor runs from the rooftop array rack down the exterior of the house to a ground rod, providing a path of least resistance for lightning.
Second, I installed an SPD at the main service panel to protect against lightning strikes (or other surges) coming from the grid, and will install SPDs at both combiner boxes and at the inverter. For the main panel I used a Siemens product called FirstSurge, which attaches to a knockout on the side of the panel and connects to a dedicated 20A circuit breaker on the main bus. For the other locations I’ll be using SPDs offered by MidNite Solar.
What does the National Electric Code have to say about protecting solar systems from lightning? This is addressed specifically in NFPA 780 “Standard for the Installation of Lightning Protection Systems.” Definitely have a good look at this. Rather than try to summarize everything here, I’ll simply highlight a few key recommendations:
install SPDs close to the array and electrical systems
install an SPD at the AC output of the inverter
ensure all equipment is properly grounded
enclose the PV output circuit in braided wire sheath, wire mesh screen, or bonded metal conduit
ensure all lightning conductors run separately and outside the cable path of PV output circuit
The latest NEC (2020 edition) talks quite a bit about “zones of protection”–although there seems to be some disagreement about this concept in a home solar application. I decided to stick with these two basic measures for now knowing that my system will be much more resilient if a lightning strike every occurs (knock on wood!).
