When will we get the heat pump thermostats we need?

It was a dark and stormy night. In a house near Denver, Colorado, the indoor temperature dropped below 68℉, and the thermostat activated the heating system. This particular heating system combined both a heat pump and a natural gas furnace into a configuration known as a dual-fuel system, which is gaining traction as an archetype for inexpensively electrifying existing homes.

This home was also equipped with a thermostat typical for duel-fuel systems that determined which heater to run based on outdoor temperature. As is common, the outdoor temperature control was set at 35℉, just to make sure the house would stay comfortable. Above that temperature the heat pump would run, and below it the furnace would. No one knew for sure that was actually the right temperature to maintain comfort. It was based on a rule of thumb, which in-turn was based on the performance of heat pumps made decades ago.

When the outdoor temperature dropped below 35℉, the thermostat fired up the natural gas furnace. Unbeknownst to everyone sleeping peacefully in the house, operating the heat pump at that moment, instead of the furnace, would have caused less carbon dioxide to be released, and it would have cost less to run. The thermostat wasn’t smart enough to figure that out. As a result, the family wasted money and an opportunity to reduce CO₂ emissions.

It’s likely that the family would have wanted a smarter thermostat if they could have gotten one. When consumers are asked what motivated them to buy heat pumps, two of their top answers usually are that they wanted to cut their energy costs and reduce their environmental impacts. Given these sentiments, you’d think that for dual-fuel systems, you’d be able to equip them with thermostats capable of either minimizing CO₂ emissions or utility bills. Think again. They’re not available at any price.

In theory, we already have most of the technology needed to make such thermostats. To minimize emissions, the thermostat would need to know how clean the local electric grid is at any given moment (it changes frequently over the course of the day), and how efficient the heat pump is (which also varies). The thermostat would also need to know the emissions associated with the natural gas grid and the efficiency of the gas furnace (both are relatively constant). Provided with such information, the thermostat need only make a simple calculation.

Determining the least cost heater requires a similar calculation, with one major difference. Instead of knowing grid cleanliness, the thermostat would need to know the cost of electricity and natural gas at any given moment. Combining that cost information with equipment efficiency, the cost of operating both heaters could be calculated, and the least-cost one identified. It’s not possible to minimize both emissions and costs simultaneously, since only one thing can be minimized at a time. It would be possible for such a thermostat to include an algorithm that balanced emissions and costs based on the user’s preferences.

Performing such calculations wouldn’t require technology much more sophisticated than current smart thermostats, like those from Nest and Ecobee. Instead, we don’t have CO₂- and cost-minimizing thermostats for two reasons. First, several of the needed information streams don’t yet exist. For example, there isn’t a good way for thermostats to know what the actual energy prices are at any given moment. Second, it’s hard for manufacturers to build markets for products that are so new consumers don’t even know they need them.

There are folks working on developing smarter heat pump thermostats, both in government laboratories and in manufacturers’ offices. Because there’s so much untapped value for these controls to access, it’s likely they will eventually come to market. That emergence will come a lot sooner if: utilities and equipment manufacturers get to work on developing the necessary information streams; the federal government invests in research and funds manufacturers’ development efforts; and lastly, environmental advocates get the word out on how these thermostats would save both money and emissions.

Why dual-fuel systems matter

The gas furnace is the most common heating device in the US. Over 40% of US homes are equipped with one. Millions of these homeowners are going to want to take advantage of the Inflation Reduction Act’s heat pump tax credits, but will find it’s expensive to retrofit heat pumps, with costs often running as high as $30,000.

One trick to cut these costs down to just a few thousand dollars or less is for homeowners to wait until their air conditioners fail, replace them with heat pumps that cost little more than new air conditioners, and leave their gas furnaces in place as backup. This strategy avoids the expense associated with upgrading air distribution ducts and electric panels, both of which are often required. It might not be the ultimate long-term electrification strategy, but in the short-term, while we build public support, workforce volume, and manufacturing capacity, it’s an attractive one. To learn more about this strategy, see Replacing a central air conditioner? Get a heat pump instead. As more homeowners learn about it, they’re likely to drive a lot more heat pump installations.

If we’re going to maximize the benefits we get for our investments in these heat pumps we need better thermostats. The way virtually all dual-fuel thermostats currently work is that when there’s a call for heat, they turn on either the heat pump or the gas furnace based on either indoor or outdoor air temperature. These controls are fine for homes, where heat pumps are backed up by electric resistance heat, propane or fuel oil. These fuels are so expensive and carbon intensive that simply maximizing heat pump operation will come close to minimizing energy costs and CO₂ emissions.

Natural gas is different in that it’s much less expensive and carbon intensive than those other fuels. Whether or not it’s beneficial to run a heat pump instead of a gas furnace at any given moment depends on several factors that can vary by the hour, including heat pump efficiency and energy costs. Making a decision on which heater to run based on a single temperature setting can’t possibly account for such variability.

This lack of effective thermostats is not a reason to delay heat pump purchases. A dual-fuel system installed in most locations in the US, using controls readily available now, over its lifetime, will likely cost less to operate, and cause fewer CO2 emissions, than a standalone gas furnace. The problem is that when people install dual-fuel systems equipped with current thermostats, they’re not getting all the money and emissions savings available. Indeed, we’d all benefit from those savings.

The quest for a dual-fuel heat pump thermostat that minimizes greenhouse gas emissions

A dual-fuel thermostat that minimizes CO₂ emissions is actually not that far from fruition. That’s because an information stream for quantifying marginal carbon dioxide emissions currently exists. It’s offered by WattTime, a nonprofit organization founded by UC Berkeley researchers. WattTime puts out an Application Programming Interface (an arrangement that enables multiple computers to communicate with each other, usually over the internet, abbreviated as API) containing real-time electric grid marginal emissions factors, updated every five minutes, for virtually the entire continental US, and most of Canada, Australia, and Europe.

As the electric grid goes through the day, solar power ramps up and down, wind power does the same, and grid operators respond by firing up and shutting down various fossil fuel and other generators. The mix of all these different electric power sources also varies depending on where you are on the grid. WattTime figures out, for numerous locations, what generation sources will power the next devices to draw electricity from the grid, and quantifies their emission intensity in pounds of CO₂ per Megawatt-hour of electricity consumed. See the graph just below for an example of what an ordinary day of such data looks like.

Courtesy of WattTime

Already, several companies offer products that make use of WattTime’s data to reduce CO₂ emissions. Enel X Way enables electric vehicle owners to charge only when the cleanest power is available. The recently released Google Nest Renew thermostat is able to shift heating and cooling operation to times when less CO₂ would be emitted.

Since a CO₂-minimizing thermostat would only be a bit more complex than what Google is doing with its Nest Renew product, it seems that it should be feasible. I asked Geoff Hancock, a WattTime product manager when he thought it would come to market, and he responded “This is inevitable. My feeling is that a WattTime-enabled heat pump could be commercially available in two years.” Currently, similar concepts are being investigated by several groups, including Oak Ridge National Laboratory and the Western Cooling Efficiency Center.

Accessing heating equipment efficiency

Whether you want a dual-fuel thermostat to minimize operating costs or CO₂ emissions, it would need to know what the heat pump efficiency is whenever there was a call for heat. That’s tricky to do, as heat pump efficiency varies largely based on outdoor temperature. The colder it is outside, the harder the heat pump has to work to extract heat from outdoor air. For example, at 40℉, some heat pumps provide about 4½ units of heat to a home for every unit of electricity input, but only about 2¼ at 10℉. These numbers vary quite a bit between different heat pump models. To determine heat pump efficiency at any given moment, the thermostat would need to know both the outdoor air temperature and the characteristics of the installed heat pump.

Knowing the outdoor air temperature is the easy part. Some thermostats already include outdoor temperature sensors. Others are able to get the local temperature from weather services over the internet.

Knowing how the efficiency of a heat pump varies with outdoor temperature is more complicated. Heat pumps are currently tested for their heating efficiency at a few standardized outdoor air temperatures between 17℉ and 47℉, with the option to also test at 5℉. The results of these tests are compiled by the Air Conditioning, Heating, and Refrigeration Institute, an industry trade group. One way to enable thermostats to access equipment efficiency information would be for AHRI to put out an API containing those efficiency test results. Installers would need to simply enter the heat pump’s make and model number into the thermostat,

Once the thermostat could access both outdoor air temperature and the efficiency test results, it could readily determine moment-by-moment heat pump efficiency. The thermostat would also need to know the furnace efficiency, but that’s a much simpler proposition. Furnace efficiency ratings are also compiled by AHRI and they vary little with external conditions. Installers would only have to program a single efficiency number into the thermostat.

Accessing real-time marginal utility costs

Most people never know what the actual price is of the electricity and natural gas they consume. Mostly they just pay the bottom line number on their bills. I have occasionally figured out my marginal cost of energy—the cost of the next kWh and therm we consume—and it’s tedious work. Even then, I only know my cost of energy at the end of the billing cycle, well after I’ve actually consumed it.

A cost-minimizing thermostat would need to know the marginal cost of electricity and natural gas in real time, and both can change monthly, or even more frequently. For example, between November of last year and this March, my gas marginal rate changed by 34%. My electric rate changed only by 1%, but soon my utility will switch me over to a time-of-use rate. When that happens, my electric rate will change several times a day, by as much as 67% during the heating season.

A cost-minimizing thermostat would need to access real-time utility costs via an API. Such an API doesn’t exist now and it would be a major undertaking to develop one. There are hundreds of utilities in North America, and most offer multiple rates. An API administrator would need to continually monitor all those utilities and rates for filings and notifications, and then quickly make hundreds of updates. This might be a job for artificial intelligence.

Once a thermostat had access to both equipment efficiency and real-time utility rates, it would be a simple calculation to determine which heater to run at any given moment.

The state-of-the-art for dual-fuel heat pump thermostats

The most advanced dual-fuel heat pump thermostat that I’m aware of is the one that Mitsubishi Electric provides with its intelli-HEAT heat pump. According to Mitsubishi, this heat pump can be combined with most natural gas furnaces, and it uses artificial intelligence to reduce both energy costs and greenhouse gas emissions. Obviously, it’s not capable of the cost minimizing calculations described immediately above—many of the needed information streams don’t exist yet—but it does some impressive things nonetheless.

For example, if the thermostat senses that during the heating season the room temperature is dropping, even with the heat pump operating, and that situation persists for nearly a half-hour, it will switch to the gas furnace to take over. Its artificial intelligence feature over time learns the conditions at which the heat pump can’t maintain comfort, and switches more quickly to the furnace. This is a much more effective means to increase heat pump operating hours, while maintaining comfort, than to use a single outdoor switchover temperature.

Mitsubishi Electric is working to improve the intelli-HEAT thermostat’s capabilities to reduce operating costs, anticipating that eventually the information streams needed will become available.

The old chicken and egg problem

To gain more insight on when a a CO₂-minimizing thermostat might come to market, I contacted Daniel Meyers, the founder and CEO of Flair. His company makes several innovative thermostats, including one that enables mini-split heat pumps to be controlled from smart phones. He told me “I'd love a business justification to incorporate a carbon driven algorithm for selecting equipment but I don't think there are currently any market mechanisms that are easy to plug into.” In other words, just like any manufacturer, he’s reluctant to invest in a product for which market demand doesn’t yet exist.

In the annals of corporate product development, this is a classic problem: What comes first, the market demand or the product? Customers can’t demand a product they don’t know they need, and they often don’t know they need it until someone makes it and markets it to them.

Building a market for cost- and CO2-minimizing thermostats is going to take some nudges from policymakers, manufacturers, utilities, environmental advocates, and investors. For example, utilities and heat pump manufacturers could get to work developing the APIs these thermostats need. The US Department of Energy could kick in money for research, and for manufacturers to develop and test prototypes. State and municipal heat pump program administrators could signal to manufacturers that when these thermostats are available, they will require their installation for dual-fuel systems. Environmental advocates could begin educating consumers about the benefits of these thermostats. It’s time for all these groups to get to work.

And now a word from our sponsor

This post originally appeared on the Energy Technology Revolution site. You can subscribe to ETR by clicking here. Also, tell your friends and family how much you enjoy ETR. Lastly, I’m sure you’ll want to read these other posts on heat pumps:

• Big Changes Afoot for Heat Pumps
• Replacing a central air conditioner? Get a heat pump instead.
• The More Heat Pumps the Merrier