After decades of being considered “alternative” energy fit for hippie communes and maybe a few crazy utilities in California, wind and solar power are now booming, mainstream businesses in the United States. Last year they accounted for almost two-thirds of new electricity generating capacity. Iowa generated more than one-third of its electricity from wind, and deep red Texas produced more wind power than any other state.
Nonetheless, because these power sources only produce electricity when the sun is shining or the wind is blowing it is widely assumed that they can only supply a limited proportion of total electricity or that expensive batteries will be needed to provide backup power. As battery prices decline this option becomes more attractive, but there is a cheaper, ubiquitous energy storage device available in nearly every American home: water heaters.
These humble appliances, sitting largely forgotten in more than 100 million basements, garages, and closets throughout the country, store energy every day in the form of hot water. Fifty million of them already run on electricity and only need a way to communicate with the grid and a mixing valve to deliver water at a constant temperature in order to make a significant contribution to matching electricity demand with supply. Converting natural gas water heaters to efficient, grid-interactive heat-pump electric water heaters would reduce carbon pollution in most regions today. For households that are currently using lower efficiency electric resistance water heaters, making them grid interactive (which has a considerably lower installed cost than switching to a heat-pump water heater) would reduce electricity system costs and pollution in many areas now. Electric water heaters of all types will become increasingly beneficial and necessary as we transition to a 100 percent clean energy system. Residential and commercial gas water heaters emit about 120 million tons of carbon dioxide per year according to Department of Energy figures, not including the very significant methane leaks in the gas system, so this is an important near-term opportunity to reduce emissions and an essential long-term one.
How much energy storage do all these water heaters represent? Conveniently, the definition of a BTU is the amount of heat it takes to raise the temperature of one pound of water by one degree Fahrenheit. So assuming that the average household heats 50 gallons of water per day by 100 degrees Fahrenheit, and that a heat-pump water heater delivers two BTUs of heat to the water for each BTU of electricity it uses, you can use the back of an envelope to calculate that an average heat-pump water heater uses 5.9 kWh per day:
Or you could look in the appendix of a recent Brattle Group report on using water heaters for electricity storage and find that a fancy computer simulation model estimates that a heat-pump water heater will use 2009 kWh/year, or 5.5 kWh/day. That’s pretty good agreement with the envelope’s back. Let’s go with the fancy estimate of 5.5 kWh per day.
Now suppose that as water heaters wear out over about 10 years they are replaced with grid-interactive units designed with sufficient spare capacity (e.g., larger tanks and higher internal temperature set points plus a mixing valve to deliver water at a consistent desired temperature) to allow them to draw power only when storing electricity is desirable. We could then have a national battery pack consisting of 100 million water heaters, which would be able to store 200 TWh of electricity per year, or almost 5 percent of total electricity generation. Assuming this energy is drawn over a twelve hour period, the equivalent power absorption capacity is 46 GW, or 35 times the ground-breaking storage mandate adopted by the California Public Utilities Commission.
And this form of energy storage is far cheaper than most other options. Assuming an incremental cost of $250 for a larger tank, a mixing valve, and control circuitry to allow waters to be used for energy storage, the cost comes out to $45/kWh. By comparison, Tesla’s much-touted Powerwall costs $3000 for the 6.4 kWh model (not including installation), which comes to $469/kWh. Of course the Powerwall has the ability to supply electricity as well as soak up available power, and such two-way operations are essential in some applications. Moreover, the cost of battery storage is falling rapidly and other types of batteries designed specifically for stationary applications may be significantly cheaper than Lithium-ion chemical batteries. Nonetheless, flexible demand, such as smart water heaters, can go a long way to integrate large amounts of wind and solar into the grid at a very low cost.
By combining flexible demand, enhanced transmission and distribution capabilities, some conventional batteries, and other techniques, we can build the 100 percent clean energy system we need to solve climate change. To do so cost effectively, new thinking about how to use the equipment we already have will be just as important as new hardware.