Home water heaters: A new ally in making grids cleaner

Water heaters are a potentially massive resource when it comes to balancing the ups and downs of grids that are increasingly powered by renewables. They use more energy than almost anything else in a home except equipment for space heating and cooling. The question is, how big of a grid resource could they become — and how much should utilities and their customers invest in preparing them to serve that role?

These are some of the questions that Pacific Gas & Electric hopes to answer with its WatterSaver pilot project. Over the next few years, California’s biggest utility intends to sign up between 5,000 and 9,000 customers to install or retrofit electric water heaters that can heat water with electricity when it’s clean and plentiful, and then store that hot water for later so they don’t have to fire up when grid power is scarce or dirtier.

The end goal of the project — up to 2.5 megawatts of grid load-shifting capacity by 2025 — isn’t that big compared to some of the long-running programs tapping water heaters as a grid resource. Southeastern utilities such as Duke Energy and Midwestern rural cooperatives including Great River Energy have hundreds of thousands of electric resistance water heaters hooked up with radio-controlled load switches that can turn them off to ease grid stress or direct them to follow schedules that take full advantage of lower-cost periods for grid energy.

The potential for water heaters as grid assets has long been recognized. Multiple studies have indicated that shifting electricity loads with resistance water heaters — as well as more efficient modern heat-pump water heaters — can yield tens to hundreds of dollars annually in grid-services revenue per water heater, though it’s not always clear that setting them up to store and shift energy is worth the cost.

But PG&E’s WatterSaver project is trying something a little bit different, said Nick Dirr, senior director of programs at the Association for Energy Affordability, one of the project’s partners. Instead of simple one-way on-and-off switches, PG&E will be installing water heaters and control devices with two-way communications via Wi-Fi and cellular networks, as well as on-board computers that can pull real-time data on energy use and water temperatures.

And instead of turning the water heaters off only during grid emergencies, PG&E wants to keep them completely off from 4–9 p.m. every day. Those are the hours when California has faced grid supply shortages and even rolling blackouts over the past two summers — conditions expected to worsen as the state increases its reliance on solar power that fades away in the evening.

WatterSaver will assess how a variety of water heaters work for that task, Dirr said. Some will be existing electric resistance models like those now used by about 50 million U.S. households, which heat water quickly but require big spikes in electricity use. But most will be newly installed heat-pump water heaters, which today make up only a fraction of the country’s water-heater stock but are seen as the future of energy-efficient heating.

California has hundreds of millions of dollars of incentives for heat-pump water heaters through programs to replace fossil-fueled appliances with electric appliances in homes and buildings. About 90 percent of California homes use gas to heat water, making water heaters a big target for carbon emissions reductions — and a major new grid load if they’re switched from gas to electric.

But water heaters that can soak up excess midday solar energy, then ride out California’s grid-stressed evening hours, could be a major grid asset, said Bill Burke, CEO of Virtual Peaker, the company managing the grid-interactive portion of the WatterSaver project. But how big of an asset? That’s where the two-way, real-time data collection from PG&E’s vendor partners comes in, he said.

“What’s really awesome about the new connected water heaters is they tell you how much energy this fleet of water heaters is using,” Burke said. Virtual Peaker is already tapping that capacity in programs with utilities including Green Mountain Power in Vermont, Portland General Electric in Oregon and Sacramento Municipal Utility District in California.

That same data can also check water temperatures to ensure these grid-interactive water heaters pass the ultimate test: reliably delivering hot water on demand. ​“As long as you don’t give customers cold showers”— something that careful data collection and analysis can help avoid — ​“you can shift load as much as you want,” he said.

That’s given supporters of clean energy and building electrification high hopes for the potential for water heaters to help meet California’s growing need to shift its solar energy from daytime to late afternoon and evening. A 2018 study from the Natural Resources Defense Council indicated that heat-pump water heaters could reliably shift about 70 percent of their electricity demand into hours when solar is abundant and use almost no power during evening peaks, as this graphic shows.

Water heaters as thermal batteries: An untapped resource

Matt Carlson, CEO of water-heater control technology vendor Aquanta, sees enormous untapped potential for using hot water to balance the grid. Compared to the smart thermostats now installed in about one-fifth of U.S. homes, water heaters have much more flexibility to help the grid, he pointed out — water retains heat much better than air does, after all.

But to date, deployments of smart water heaters are ​“about an order of magnitude” lower than deployments of smart thermostats, he said. Aquanta’s two-way control devices are deployed by customers like Green Mountain Power and Portland General Electric in the tens of thousands, compared to the millions of legacy one-way load-switch systems out there on water heaters around the country.

Beyond giving utilities much more fine-tuned data and control, Aquanta’s water-heater-mounted devices are linked to a Wi-Fi-connected smartphone app that helps customers save energy on their own, he said. Relatively simple controls, like letting homeowners remotely turn off their water heaters when they’re on vacation, can lower utility bills, he noted. Aquanta also offers more sophisticated capabilities, like analyzing a household’s hot-water usage pattern and adjusting to avoid unnecessary heating, which can save even more.

Aquanta presented data on those efficiency benefits at last month’s Hot Water Forum, an annual event that’s a mecca of sorts for water heater geeks. The results indicated that Aquanta’s ​“active” controls enabled a 19 percent reduction in electric water heater energy usage on an annual basis compared to a ​“baseline” of inactive controls, as this chart shows.

Extrapolated to the country’s total annual electric water heating bill of about $21 billion, these techniques could yield a $4 billion annual savings, he said. And this data doesn’t include the potential for shifting electricity use, which can add even more value.

In fact, smart water heaters can participate in some of the most sophisticated grid services markets out there, said Eric Rehberg, chief engineer and co-founder of Armada Power. The startup’s technology, developed at U.S. Department of Energy laboratory manager Battelle before it was spun out in 2015, is being used for PG&E’s WatterSaver pilot.

Armada’s control system can react within subseconds, allowing it to serve in frequency regulation markets that help maintain transmission grid stability, he said. Many of Armada’s devices deployed in apartment buildings are actively bidding into the frequency regulation markets of mid-Atlantic grid operator PJM, according to Rehberg.

The next frontier for Armada’s control platform ​“is direct renewables firming,” he said. That entails giving grid operators access to sources of electricity that can quickly turn on and off to absorb shifts in wind and solar power output. The market structures for enabling this kind of moment-to-moment renewables firming are not yet fully developed. But Armada has shown it can be done in its pilot project with wind- and solar-rich utility Hawaiian Electric, he said.

Finding multiple ways for water heaters to earn money doing work for the grid is vital to making them a money-saving proposition to customers, said John Reinhart, demand-response technology lead at Great River Energy, a Minnesota-based generation and transmission cooperative.

Out of the roughly 720,000 customers at the 28 member cooperatives GRE serves, about 100,000 have water heaters equipped with one-way load controls, he said. About 40 percent of those are enrolled in what are called ​“interruptible” programs that turn off water heaters when grid energy costs are spiking, and about 60 percent are in a program that controls them to charge overnight, when grid power is cheaper and more plentiful. A growing portion of that overnight power comes from Midwestern wind farms, he noted.

Under a year-old program with Aquanta, GRE and its member cooperatives are switching over to two-way water heater controls that provide ​“a new layer of data we never had before,” Reinhart said. That’s an important step in getting a clearer picture of exactly how many megawatts of controllable load Aquanta’s system can provide for the grid.

“That unlocks all of the market products, whether it’s capacity, energy, frequency regulation,” he said. ​“They’re all enhanced with better accuracy.” 

Getting the regulations and market structures right for smart water heaters

California is an interesting market to test the capability of smart water heaters to deliver both energy efficiency and carbon-reduction benefits, said Amber Wood, buildings program director with the American Council on an Energy-Efficient Economy (ACEEE).

ACEEE hosts the Hot Water Forum along with other building efficiency events, and it has been promoting the value of smart heat-pump water heaters for years. The first test for any water heater program is whether it improves overall efficiency, Wood said. ​“If you do energy efficiency first, then you don’t have oversized equipment, and you can decarbonize more effectively.”

But as power grids start to face hour-by-hour shifts between clean and dirty power, load flexibility becomes a more pressing issue, Wood said. California, which is reworking its energy-efficiency programs to value reducing and shifting energy use to match the ups and downs of its increasingly clean electricity supply, is a good example of a place ​“where energy efficiency and decarbonization really cross over, when you’re aligning the demand side and the supply side.”

Most utility programs and regulations aren’t designed around these shifting grid carbon metrics, she noted. ​“One of the questions is, are there rate structures, or other ways to incentivize more efficient water heating that also ends up decarbonizing, because it aligns with how clean the grid is?”

PG&E’s WatterSaver program will direct $6.3 million toward these kinds of incentives, starting with money to get people to buy heat-pump water heaters, which are still more costly than traditional water heaters. Beyond the $3,000 in state incentives now available to help a customer install a heat-pump water heater, WatterSaver is offering an extra $50 to those who enroll in the program, Dirr said. (This $50 incentive is also available for resistance water heaters.)

It’s also offering a $200 incentive to install the thermostatic mixing valves that allow overheated water in tanks to be brought down to temperatures safe for use in homes, he said. That’s a vital part of any thermal storage system, since overheating water in tanks ​“doubles or triples the load-shift potential” by expanding the stored heat capacity.

California’s electric-appliance-friendly building codes adopted last year also offer additional efficiency scoring benefits for water heaters that can shift loads, Dirr said. The state’s Self-Generation Incentive Program, which now mainly pays for batteries in homes and buildings, also has $85 million available for heat-pump water heaters, along with performance structures that reward load-shifting that reduces greenhouse gas emissions.

As for rate structures, PG&E and California’s other big utilities already have time-of-use rates that make electricity more costly during peak afternoon and evening hours and cheaper during off-peak hours, as this chart from a PG&E presentation at last month’s Hot Water Forum indicates. 

But all these theoretical money-saving and grid-supporting features need to be tested in the real world, Dirr said. Technical issues include maintaining communications links through everyday disruptions, he noted. ​“What happens when a customer’s Wi-Fi drops or they change their router password? What happens in an apartment when the old tenant moves out?” In some cases, direct cellular connections between smart water heaters and the utility and customers may be more reliable than Wi-Fi, he said — but cellular is a more expensive option.

Then there’s the issue of how California’s regulatory framework measures the value of the ​“permanent load shift” capacity that the WatterSaver program is testing, he said. ​“This program is weird because it’s not efficiency, nor is it demand response,” he said. ​“It’s changing the load shape. We’re not reducing kilowatt-hours; we’re just moving it.”

That complicates how utilities are allowed to consider the costs and benefits of a program, he said. ​“With energy-efficiency programs, there’s a whole established cost-effectiveness tool, and it’s all about reducing kilowatt-hours over an existing baseline,” he said. But because the WatterSaver program is designed to start its water heaters on their 4–9 p.m. load-avoidance patterns from day one, there’s no baseline to measure against in this case.

This cost-benefit measurement issue is actually one of the most complicated parts of the WatterSaver program, Dirr said. It’s made even more complicated by recent changes in the avoided-cost calculator, the methodology California uses to measure the cost-effectiveness of energy efficiency, solar net metering and many other aspects of utility energy policy.

“We’re in an early stage here,” he said. ​“A lot of what this will be testing is how much can they load shift and what are the variables that influence that.” 

Heat-pump water heaters, for instance, are so much more efficient at using electricity that it’s not clear how much more grid value is to be gained from the additional costs involved in setting them up to store and shift loads, he said.

Setting up rules and regulations for valuing what these load-shifting water heaters can do for balancing the grid and reducing carbon emissions ​“is the next step,” he said. While investing upfront in getting a major class of all-electric appliances may not make as much of an impact in the early stages of deployment, he noted, ​“at scale, it probably does matter — once everyone in the state is using them.”