1 Year with Residential Solar
Last May 17th, I was instructed by my solar installer that my residential solar installation was approved for operations, and I turned it on.
Now, with a full year of operational data, I can review the decision to install solar in the first place and compare how my energy production, energy consumption, and other economic and lifestyle assumptions have met reality. If you don't want to read anything and just want to know the punchline: I'm extremely happy with the results! The most surprising result from the last year isn't how much energy I produced, but how much less energy I consumed. Importantly, if you live in the Baltimore, MD metropolitan area I can also gladly recommend my installer, so if you need a recommendation please reach out and I will introduce offline.
System Details
The Roof
We live in a townhome in Maryland with one due-South-facing roof (well 172˚) and one North-facing roof, both pitched at 24 degrees. There are no trees casting shadows on the South-facing roof, but there is a chimney and the firewalls between the neighboring homes also casts a mild shadow in the early / late hours when production would be low anyway.
The Panels
At the recommendation of our installer, we decided to put 16, 405 watt REC Alpha Pure panels on our South-facing roof. One installer we talked to was brand agnostic and wanted to use panels with lower output and cover part of our Northern roof as well, but the installer we went with had a more thesis-driven perspective that we should 1) use these specific REC-brand panels due to the quality of the manufacturer and their strong record of minimizing ecological impact in the production process (a lead-free process); and 2) that we should only put panels on our high-output South-facing roof. It is noteworthy that the cost of panels and the performance of them has gotten good enough that the North-facing roof idea was not totally bogus (there's a valid movement toward putting panels on building exteriors and fences!), but ultimately I felt that minimizing the installation to a high yield area made the most sense for our budget and our needs.
The Inverter/Optimizer system
Our installer used SolarEdge's Wave inverter, and at the panel level we installed SolarEdge's PowerOptimizers. The bigger installer we did not contract with wanted to use a micro-inverter setup using Enphase, but ultimately the smaller installer convinced me – simply by walking me through line-by-line of their technical specs – that there would be too much “ducking” (a reduction of maximum output measured per the system's “temperature coefficient”) on my asphalt shingle roof during the super hot days of the summer when I would want to be creating the most energy possible. So far I've seen nothing that counters this argument, and I have no issues with the system installed so far.
Sense + Solar Energy Monitor
About a month after commissioning the use of my solar panels I was still frustrated with the lack of visibility into my whole home's energy balance. Using my BGE (local utility) app I could see what my net electricity balance was on a given day (their app shows the data within a day or two) and using the Solar Edge app I could see what my gross production was, but if I wanted to focus on my consumption behavior – or make sure I was time-matching my consumption to my production as best as possible – it was impossible to do. The Sense monitor is something anyone can benefit from, regardless of having solar or not, but it has a great upgrade for people with solar panels that makes all of this data easy to understand at second-level accuracy. As I will detail in my analysis of my consumption metrics later in this post, the Sense + Solar device has been one of the best home improvements investments I've made despite it not being a required part of any solar setup.
Electrical Panel Upgrade
Before doing solar, I had a very odd “split bus” 150 amp panel (no main breaker), despite the house having 200 amp service into the meter. While preferable, I did not need to upgrade the panel to do the solar installation. But given all of the electrification I planned on doing or was in the middle of doing (EV charger, induction stove, dryer, water heater, and heat pump), my solar installer offered to combine the work with the install using the same trusted subcontractors that would do the required electrical work. The benefit of doing the upgrade as part of the solar project was that it qualified for the full, un-capped 30% IRA solar tax credit (versus if I did it as part of the heat pump upgrade or another upgrade where there is a cap to the tax credit). To be clear, I left this upgrade out of my payback calculator (linked in next section) because I felt like it was required for my whole home's benefit, including its resale value later on, and so it was highly amortized across everything.
(No) Battery Backup
Solar systems are largely the same no matter what climate you live in (though, as mentioned above, you want to understand how your panels and their inverters or optimizers perform at extreme temperatures). However, in the US you will see wildly different options about whether or not you should have battery backup be a part of your system, and wildly different rate structures and incentives to encourage them (or not). In Appendix 1 below I detail why I I believe it was not economically practical nor socially beneficial for me to include a battery in this setup.
Economic Assumptions
As I was assessing the decision to get solar, and assessing my 3 main options (1 local installer, 1 regional installer, and Community Solar) I created a simple payback calculator. I've posted a version of it here for review (showing just the installer I went with). At a high level, I imagined that getting solar would be worth it financially (vs joining a Community Solar operation that would have insured a flat 10% discount to the retail electric rate) because my model gave me a <10 year payback period, which I thought was quite reasonable given my ability to make long-term investments. I also believed that on an earlier timeframe (at ~4 years) the investment would add recoverable value to the home, as various Real Estate organizations and companies have published estimates for how much of a premium home buyers might add to a home with solar and none that I've seen indicate lower than a 3% premium. Finally, because we had recently sold our Seattle home and moved into a cheaper market (and smaller house), we had the cash available to pay for the solar up front, as any financing option would have likely been at a worse rate than leaving the cash earning 5% in our high yield savings account. Once all incentives / tax credits were accounted for, the system cost us $15,393.
The model I used ended up being incredibly spot on in terms of energy production (thanks to projections from the installer) but quite wrong regarding how much I ended up consuming (thanks to my poor forecasting and a surprising change in behavior). Here, I'll go line-by-line through my summary (located on the linked sheet, but also pasted above).
Solar Production
My solar installer projected that my 16, 405 watt panels would produce 8,551 kWh of energy in their first year of operation and wow was that accurate: my production was 8,526 kWh! I was actually surprised to see how accurate this was in part because of the Canadian wildfire smoke we suffered through from July through August 2023. The sun was visibly dimmer on the smokiest days of the wildfires and as was widely reported every day of any smoke pollution impacted solar production dramatically during a season when they should have been most productive. I am unsure when in the calendar year we made up the difference, or if the algorithm already hedged for such unforeseen issues, but here we are. Obviously I am hoping for fewer wildfires this year, and I am excited to see if we can beat the projected number next year.
Consumption
I modeled my own consumption taking into account my previous year's consumption of 10,654 kWh (above image from my May 2023 bill), plus an assumption that the addition of an EV and Induction stove – all within a few weeks of getting my solar commissioned – would significantly increase my overall consumption. What happened instead was the opposite: instead of consuming ~10.7 MWh or more we brought our consumption way down to 6.6 MWh. How were we so far off and how did we reduce electricity consumption so dramatically? There are likely 3 explanations:
- Lower utilization of the AC – With no electric being used for any heat process in our home, our heaviest electricity consumer is our AC unit; so the most likely explanation for usage being significantly lower was that we used our AC less. Here are two reasons why that probably occurred:
- In the baseline 2022 year we had an older, sick dog at home that was highly sensitive to heat. While he was still alive, to keep him comfortable, we had to keep the home in the mid-to-low 60s. Last summer, however, he was in dog heaven and we relatively healthy humans had a higher tolerance to the heat, allowing us to more opportunistically cool the home at bedtime or just open windows and run the ceiling fans (although with the aforementioned wildfire smoke there were also times we ran the AC to clean the air and kept windows closed despite temperate weather).
That said, one of the least appreciated benefits to adding rooftop solar is how effectively they cool your home. Think about it: instead all of the sun's radiation penetrating your roof and running through your attic insulation, you are converting ~25% of that sunlight into electricity, while radiating back into space a significant portion of what you cannot turn into energy. In less scientific terms: solar panels are shade placed on the part of your roof that previously heated your home the most. At one point I went up in my attic with my infrared camera. In the image to the right (read it bottom left to top right) you can see that the purple area is the South-facing section covered by solar panels, the bright yellow is the South-facing portion uncovered and 143F (!!), and the orange in the top right of the photo is the North-facing, uncovered roof. While our attic is still under-insulated and unsealed, it's clear that solar has actually lowered our energy burden from its ability to shade and reflect.
- Less incremental electricity usage – I modeled that the addition of our EV and induction stove would add 3,650 kWh of electrical demand (moving over from a hybrid gas car and a natural gas oven and stove-top). In year 1 of owning the EV, we drove 8,994 miles and used 2,633 kWh of energy (~293 Wh / mile); but, because we live in a townhome and do not have a garage, we have to be opportunistic with our charging and sometimes charge at the doctors office or grocery store instead of running our big 240v charging cable 50 ft to the parking lot. We also have put in a good portion of those miles on road-trips, charging either with a fast charger along the way or at our destination. This means only 31% of our charging was actually done at home (33% at Level 2 out of home, and the remainder at fast chargers on road trips), leading to a significantly smaller increase in consumption than modeled. Going forward, we will draw even less from our own meter, as our community gets Level 2 chargers installed nearby. Because of our net metering surplus (making the clunky/extension cable charging method free for us), this will actually increase our overall electricity costs by about $126 next year. Meanwhile, I do not know how much energy I actually used with my induction stove, as the Sense monitor (which has been good at learning the usage patterns from most of my appliances) has had a lot of trouble differentiating consumption from the stove (likely due to the myriad of burners and of energy levels it can draw per burner at each heat level or booster setting).
Behavior change - Finally, I believe we consumed a lot less energy in the past year precisely because net metering (ability to “spend” energy you over-produce at any time during the year) made us think a lot more about our overall energy balance. It felt like on days we produced more energy we deserved to use as much as we needed, but we did not want to ever blow past that amount. The real unlock for this behavior change was the Sense monitor, which gave us the needed real time visibility into how we were doing against that balance at a moment-level (remember, the utilty app was always a few days delayed). Call it gamification or call it obsession, the experience I have had with solar + net metering + Sense is the same feeling I've had wearing a Continuous Glucose Monitor: I felt empowered; I felt like the data helped me make better decisions; and, I felt like a winner whenever I could get through a day a net producer of energy vs a net consumer. An example of how this behavior manifested is that some evenings we'd feel pretty hot and want to turn on the AC, but knowing how much the AC would draw, knowing the grid was especially burdened in the evening, and knowing how much of a balance we still had remaining, I would delay turning on the AC unit until closer to bedtime so that we could still be net positive producers on a given day (image to the right is from the Sense app, showing a recent day where we over-produced 13.4 kWh).
A Note on Future Consumption
While I tried to initially model in an increase in electricity from driving an EV and from the induction stove, I did not try to model in my future electrification or home efficiency efforts. That said, this week I did replace my gas dryer with a heat pump dryer, and I do plan on 1) weatherizing my home this summer (we are under contract to air seal the attic and upgrade its insulation from R12 to R49); 2) replacing my gas hot water heater to a heat pump model; and 3) replacing my gas furnace with a heat pump either later this year or next year. How this all nets out, combined with the upcoming changes to how I charge my EV, is too difficult to forecast. The insulation will definitely reduce our heating and cooling requirements while any additional electricity consumed by heating with a heat pump will be partially offset but the far better efficiency we will get from modernizing our cooling in the same install (our present AC unit is 22 years old!). The water heater will be a net new source of electricity consumption as well, but given the above I think we will still come in under our production numbers. Hopefully I will have a clearer picture when I give my Year 2 update next year.
Electricity Revenue / Costs
Because I had assumed an increase in my electric usage, I modeled that I would save $1,351 via my solar production and spend $909 due to my over-consumption. Instead, I spent $0 on electricity consumption and ended up with a $143 check for the wholesale value of how much I overproduced (Maryland “nets out” at the end of March every year, so really that check only covered 10.5 months of production). To be fully transparent, I still do have about $9.43 in monthly utility fees to be connected to the grid (which is fair, since I use the grid), making me more like $30 on top not $143. Going forward, my model assumed an annual 2% increase in the electrical delivery rate. In December 2023 the Maryland Public Service Commission did approve a substantial rate increase to be spread across the next 3 years with an immediate large 14% increase that went into effect a the beginning of the year and smaller increases after. Recently I also became aware of a service called Arbor that helps put you on a lower rate in deregulated energy markets like Maryland's (to see if your market has such offers, use my referral link here), lowering my energy supply rate (if/when I need supply again) to what would only be a 12% increase. Going forward, my payback calculator's revenue or costs for electricity will be highly subject to how my consumption changes (or doesn't) after further weatherization and electrification.
SREC Revenue
SRECs are Solar Renewable Energy Certificates, and they are purchased by utility companies or other companies who may have a state government or self-imposed mandate to have a certain % of their electricity production created by solar energy. When I put together my project's cashflow model both solar install companies warned me that the SREC values were decreasing significantly and would be worthless within a few years, indicating that I should not use the estimated values too heavily in my model (for reference, in 2015 a 1 MWh SREC could be sold for $180 early in the season, forming a major part of someone's cashflow model, but by 2020 the highest level was $70, and last year it was ~$53). That said, I tried some really crude modeling and expected to see $505 in my first year. Instead I've seen $343, and would guess that the number continues to drop, but have not deeply analyzed what's going on in that market.
Maintenance
One thing I failed to model in my payback calculator was maintenance expectations. While all of my major equipment has really incredible warrantees (12 years for the inverter, but 25 years for the panels and micro-optimizers), I still needed to pay my solar installation company to come out and help remove some panels so that my roofers to address a few leaks unrelated to the solar install (and therefore not covered by my 1 year service agreement). What I've learned from this is that, despite having the roof assessed prior to installation, some issues cannot be predicted and should be expected in any model. The solar company is of course more expensive per hour than my roofer, but my roofer will not touch panels so I need to hire both teams to be on site at the same time to do any repairs. I know I have at least 1 more leak to get fixed and so expect to spend a similar amount this year and should add similar line-items into future years. These charges are meaningful and impact my payback periods (as I detail below).
Additional Considerations
One additional benefit for having installed solar is that it has impacted my ability to advocate for and recommend solar to others. While my neighbor likely would have gotten solar without me, I was very happy to be able to share my experience with her while she explored the idea and recommend my installer to her (again, if you're in the Baltimore area feel free to reach out to me and I will introduce you to the company); and it brings me joy to see her rooftop now participating in the clean energy transition.
As I navigate my own career transition into climate tech / clean energy, another ancillary benefit has been the opportunity this process has afforded me in wrapping my head around the overall economics and logistics of solar energy specifically and “distributed energy resources” (DERs) in general. There's real economic value to this literacy, and I hope that in my Year 2 update next year I'll be able to add a note to the payoff calculator about how this process has benefited me in finding a new job or starting a new business. Stay tuned!
Climate Impact
My payoff calculator was meant to be purely an economic exercise, but of course there are other bottom lines and motivations to consider: mainly, our household's impact on climate and our quest to reduce greenhouse gas emissions. The positive climate impact solar has provided has manifested in 2 ways beyond the obvious first way of simply displacing electricity from a dirty grid to a clean one (our solar production helped prevent 6 metric tons of CO2e)!:
- As mentioned above, solar has encouraged us to consume significantly less energy overall. When we use less energy than we produce, it means the electrons used by our neighbors are cleaner, and we are helping shift the fuel mix for our community not just our home;
- While we were already committed to to electrifying everything in our home and no longer combusting fossil fuels to create our hot water and air, the timing has always been in question, as we still need to make everything make sense economically (especially after spending so much up-front on solar). When I made my model last year I worried that we would easily consume all the electrons we produced, and so I would simply look at the economics of further electrification as the difference between retail fuel costs and overall fuel use. But now that I have an excess of electricity the operational expenses of future appliances can be modeled at $0 until proven otherwise. This hastens our ability to afford that electrification, again saving money in the long-run but importantly helping lower the climate burden of our home in the short-run by bringing in the timelines for our transition.
Conclusion
Getting solar on our rooftop was a marvelous decision. I learned a lot, have been pleasantly surprised by how accurate the installer's production forecast was, and am absolutely floored by how much we've been able to reduce our energy consumption simply by having a daily scorecard telling us whether we “won” the day or not. More than any other behavior change (including eating less meat), rooftop solar has been the most tangible and effective investment in making me more invested in and thoughtful about the energy transition.
Because of my lower consumption than planned, and unplanned maintenance costs, my cash on cash payback with current usage will likely be around year 10.5 or 11 vs the 8.5 I initially modeled; however, given my plans to electrify and weatherize, that calculation is likely to change significantly between this year and my update next year. Even if the payback period doesn't improve, I believe the economic value these improvements have added to my home will show up in any resale price at an accelerated clip, as the energy transition becomes increasingly front of mind for buyers and retail electric rates continue to increase.
Finally, if you are considering buying a solar system for your home I highly recommend it and would be more than happy to chat with you about my experience or talk through any details of your own setup. Feel free to reach out and let me know if I can help.
Appendix: Why I didn't get a battery backup
I decided against having a battery system for 3 main reasons:
- Marginal selfish value - There are 2 self-centered reasons to have battery backup: 1) If you cannot live comfortably with the frequency or duration of grid outages in your area; or 2) If the net-metering economics of your system are so bad, and the ability to feed energy back into the grid so attractive, that you can pay yourself back for the investment of the battery system. For us, the occasional electrical outages experienced in Baltimore aren't that bad: much of the local distribution grid is underground so there are limited storm-related issues, and having lived in this home for nearly 3 years we've had probably a total of 24 hours of outage spread across 3 - 4 events. Because the weather is fairly mild, and because we have a good amount of food in the pantry, we don't worry about not being able to eat or heat when the electricity is out (even with electrified cooking). Meanwhile, the net metering in Maryland makes owning a battery a poor financial investment. With annual net metering in our state, the grid is our battery for economic purposes: When I over-produce a kWh during the day, that energy is credited to my account at a 0% loss rate and my neighbor uses the electrons in their home, reducing the amount of natural gas burned 50 miles away at some power plant. At night, yes I'm drawing on a fossil-fuel heavy grid, but I get to use my “credits” (again at a 0% loss rate) from over-producing during the day. If there are a few days or weeks or months of low production due to weather or seasonality, no worries: the true-up happens on an annual basis, so my summer credits help fill in my winter deficits. If I was using a battery in this way, up to 15% of that energy would just be lost to heat and other inefficiencies. Meanwhile, because renewables are still a very small part of the grid in the Mid-Atlantic, there is not a market for feeding battery electricity back into the grid during peak times at any premium. Even if that market would appear, because I installed a relatively small system on a small house, the cost of a battery system would have ballooned overall project costs and made the economics and pay-back period much worse (again, because there's no revenue upside to having the battery now it would only make it worse).
- Marginal societal value – If you are in California or Texas right now you have a grid where during the day the system is producing too much electricity and then at night things swing back the other way and a bunch of dirty fuel systems need to come online. Thus you could reasonable conclude that it is the most socially responsible thing to have a battery, storing any excess energy you produce during the day, and using that energy (or putting it back into the grid) in the evening and night when there is no sunlight and the electric grid needs support. Beyond it being the right thing to do, and partially because of it, the utility commission in California has made it required to have a battery system in some parts, as new solar installations would continue to overburden the grid when they over-produce. While requiring battery systems increases up-front costs, battery owners can get “paid” by the grid at rates higher than usual retail rates because you are discharging at the highest rate times, making them economical. But here in Maryland we do not have this issue and will not have this issue for another decade at present solar and wind development rates; so rather than helping ease grid congestion and clean it up, installing a battery would only be to my potential benefit during outages, and I'd rather Lithium Ion battery with all its rare earth materials go accelerate the energy transition in someone's new car or a home in California vs using those critical supplies for a small comfort measure. My money is certainly better spent on further electrification in my own home.