If you already know all about legacy AC-coupled, grid-connected Solar power installations you might want to skip down to the paragraph titled “Planning for Upgrades” and avoid all the background stuff.
A somewhat sardonic capsule history: Roof-mounted AC-coupled grid-connected PV systems are now considered legacy systems. Installed by companies at great expense for people who didn’t know what they were doing but didn’t want to pay large utility bills. The systems were sold with optimistic payback calculations from companies that went out of business before the roof started leaking and the systems needed repair, maintenance, or upgrading.
A longer and less sardonic history: The most common PV installation used to be relatively small (250 Watts or so) PV panels on a house roof with Enphase microinverters under each panel. This converted the DC output of the panel to 220VAC and permitted direct connection to the grid incoming service panel so the PV power displaced the grid input power while the sun was shining. If you got the proper permitting and a net-metering agreement with the local utility, the utility replaced the meter with a bi-directional meter than would record the amount of power flowing into or out of the electrical system. If the loads in the home were less than the output of the PV panels the excess power would flow out to the grid.
At night, or any time that the PV system produced less power than the house needed, power would flow from the grid to the home. The bi-directional meter would record all this flow and report to the utility what the net power consumption was. Depending on the agreement between the homeowner and utility the result would be a reduction in the electric bill.
If the grid went down, meaning power was not available to feed the home, the microinverters, which need the grid voltage to synchronize their AC output, would shut down. So no additional switching is needed to keep from feeding a de-energized grid and potentially killing line workers.
Today, utilities enter into a variety of agreements with homeowners who install grid-tied PV systems. These agreements are broadly called net-metering agreements but they vary widely in their financial details. When solar installations were relatively rare and small most utility agreements operated on a one-to-one basis. If you sent one kilowatt of power to the grid when you generated excess power, you could retrieve one kilowatt from the grid later, and incur no cost other than a connection fee which was a modest fixed cost. This treated the grid as a storage device.
From the perspective of utilities, they were receiving power when demand was low, and supplying power when demand was high–they justifiably don’t value that. As home solar became more common utilities successfully argued that ratepayers who did not have solar power were subsidizing those who were supplying power with minimal value. Many public utility commissioners accepted these arguments and the one-to-one net metering became net billing, which sounds similar but isn’t. In most cases utilities paid for power generated (usually with credits) at lower wholesale price and sold it back to ratepayers at a much higher price.
Older installations sometimes are “grandfathered” at a net-metering (one-to-one) rate since these contracts were generally set for 25 years. But grandfathered agreements can evaporate with regulatory changes, and often don’t permit adding more capacity or substantially changing the system. At the very least you need to know the ramifications of making changes.
Treating the grid as a battery made more sense when the grid was highly reliable and the net metering agreement was on-to-one. Electrical grids taken as a national whole remain reliable but in some locations the grid is substantially less reliable than it has been historically. Aging infrastructure, changes in how utilities are governed, adapting to climate-related stresses (like wildfires, floods, et.), and the intermittent nature of large-scale renewable energy sources like solar and wind all pose challenges for grid stability.
The trend over the last five years has been toward systems with battery storage, both in DIY and contractor installed systems.
On the DIY side simpler PV arrays (higher voltage series strings of PV with minimal or no electronics mounted under panels) have become common with All-in-One systems that combine solar controllers, inverters, battery charging, load switching, and sometime management of grid connection. These are often ground mounted to avoid the complexity of meeting RSD requirements.
Solar contractors generally favor roof mounted systems since that is the major market opportunity in more populated areas that are efficient to serve. Regardless of what kind of system contractors install, roughly half of new installations include battery storage, and adding battery storage to existing systems is an obvious opportunity and growth area.
The percentage of new solar installations in the United States that include battery storage has been increasing. In California, more than 50% of residential solar photovoltaic installations include battery storage in 2024, up from just over 20% in 2023. Nationwide, the rate is 25% in the first quarter of 2024, up from 6% in early 2020. I couldn’t find any large scale surveys regarding why people choose battery energy storage systems but for me (and probably you), a solar installation that stops working when the grid goes down is low value, and what value there is rests entirely with what utilities are willing to pay for the excess energy a system generates during the day, which is rarely worth bothering with.
Public policy changes like the Inflation Reduction Act (IRA) investment tax credits for stand-alone storage, and California’s net metering and state incentives for solar-plus-storage accelerated the development of energy storage. Regions with high energy demand and peak demand prices enable excess solar energy with battery storage to be sold to the grid during peak demand periods. In locations with wide variation in Time Of Use (TOU) pricing even an grid-connected battery/inverter system with no solar can deliver an ROI by charging at night when power is cheap (even free in some places) and delivering power when utilities will pay well for it. Such systems also deliver power to the home when the grid goes down.
This is a good time to be thinking about upgrading that ten-year-old system on you roof.
Planning For Upgrades
You can just dive in, take a guess at how much solar electricity you can use and buy that number of panels and a hybrid inverter to deliver your guess at the right amount of AC per day and use the rest of the solar DC power to charge your guess at how many batteries you need to get you through the night and whatever grid outages, cloudy days or winter low solar days you might encounter. You’ll avoid all that pesky math and data gathering (and probably fail to get the full benefit of your upgrades). Or you can geek out, gather data, find out what your utility might offer, and make a more informed decision–as I’m doing.
My initial upgrade ideas turned out to be stupid, would have made for a lot of work and mess (I would have dug a trench through a yard that has artificial grass and a putting green), and would probably have yielded a modest benefit that would have reached breakeven long after I was safely dead. I’m going to dive into the planning, permitting and building process and I hope you’ll come along for the ride.
Ultimately, the decisions mostly concern more solar panels, more batteries, or more efficient use of power. To make the decisions about what you need to add or change it helps to know your energy needs and use patterns. It might seem like utility bills would tell you what you need to know, but it’s really not enough. For example, if most of your load is in the evening you’d usually decide you need more batteries than panels. But if you know exactly what is taking the load you might be able to shift some of that to the daytime, and therefore you might add more panels than batteries. I spent two hundred bucks for a Refoss Energy Monitor to do real-time tracking of energy consumption. It’s got 2 main current transformers (CTs) of up to 200A each) and 16 branch circuit CTs (up to 60A each) so I can monitor each breaker in main panel. It’s also compatible with 3-phase 4-wire Wye systems so when I finish using it here in Maui I can bring it back to the mainland to analyze usage in my 3-phase shop.
One of the upcoming videos will cover the installation and how I’ll analyze the results.
And since I’m slowing down the project to do some monitoring I plan to use the time to investigate what our local utility offers. My current system is grandfathered with 1-1 net metering, and I don’t want to screw that up by adding unauthorized upgrades. But Hawaii has some interesting battery sharing programs. The most interesting program is actually closed now, but it’s worth talking to Hawaii Electric (HECO) about upcoming programs. Most utility programs are not DIY friendly, but I’ll also spend some time with local installers and electricians to see if I can work with them on this upgrade to qualify for worthwhile utility programs.
So I’ll be doing a video and blog post about that.
Here’s the bullet points of the general data we’ll need to make good decisions about updating a legacy system. Some of these points might seem obvious or redundant, but I’m not writing this post or the script for the accompanying video for DIYers who already have a lot of knowledge about solar installations. It’s mostly for folks who have panels on their roof with microinverters who want to make the system they already have be more useful, to power their homes even when there’s a grid outage, and serve more (or all) of their electrical energy needs.
I’ll cover the detailed analysis in (surprise, surprise) some videos and blog posts.
Energy Consumption: Understanding both how much energy you need and when you need it is at the core of guiding these decisions. I’ve been spending a fair amount of time on solar forums, and I see the same kind of questions/complaints over and over. Here’s an exaggerated example:
I have a 3000 watt inverter and 450 watts of solar. I installed seven 280ah, 12 volt batteries and they won’t last though the night even though my load is only 400 watts. Shouldn’t my 23kWh of batteries last longer?
If your energy consumption is high during the day, more solar panels might be beneficial to generate sufficient electricity. Conversely, if you need power primarily at night or during cloudy days, more batteries can help store energy for those times. But more batteries won’t help if you don’t generate enough power during the day to charge them, and more panels won’t help if your batteries are fully charged before noon and you’re dumping power to the grid that utilities will hardly pay for. You need to know: Energy used per hour for 24 hour day, all the details of your net metering or net billing agreements.
Sunlight Availability: In regions with abundant sunlight all year it’s relatively easy to determine how much solar generation is required to meet daytime needs and charge batteries to last through the night. Areas with frequent cloudy weather may benefit from additional battery storage to compensate for lower solar output if other sources of charging are available, like cheap grid TOU electricity. Having more batteries that don’t get fully charged before the sun goes down won’t help. Most places in the USA have data on the peak sun hours for your area.Peak sun hours differ from hours of daylight; the peak sun hour actually describes the intensity of sunlight in a specific area, defined as an hour of sunlight that reaches an average of 1,000 watts of power per square meter. Panels may get 7 hours of daylight a day but for much of the day they won’t generate much. In the continental US peak sun hours are generally around 4 or 5 with the peak occurring at solar noon. The closer you are to the equator, the more peak sun hours you’ll have. You need to know: Peak Sun Hours for cities close to you and current average generation of the system you may upgrade.
Cost and Incentives: The decision may also depend on financial factors such as the cost of installation and available incentives. Solar panels generally have lower upfront costs compared to batteries but batteries can provide additional savings by allowing homeowners to avoid peak pricing periods and or permit selling power to utilities at higher prices. And there are some programs that included credit or payment for upfront costs that are less common for panels alone. You need to know: detailed pricing and incentives as well as upcoming programs or rate changes, especially TOU rates.
Longevity: Solar panels typically have a longer lifespan (25-30 years) compared to batteries (10-15 years). Consider the potential need to replace batteries earlier than solar panels when evaluating long-term investments.
Ultimately, the decision between more solar panels, a solar battery, or other alternatives depends on your specific circumstances, including energy needs, budget constraints, location, and long-term goals. Balancing these factors will help optimize your solar power system’s efficiency and effectiveness. The next video in this series will cover the pesky math required to optimize your decision.