Lower-than-expected efficiency of heat pumps.

People, including me, other people I know, and academic researchers, have reported lower-than-expected efficiencies on heat pumps. The efficiency of heat pumps is measured using a confusing number of terms including COP, or coefficient of performance, HSPF or heating season performance factor and SEER or seasonal energy efficiency ratio. Let’s explain each one of them:

COP is the heat energy delivery by the heat pump divided by the electrical energy used by the heat pump. If the heat pump delivers say 4kWh of heat by using 1kWh of electricity, then the COP is 4. People are often familiar with using kWh to measure electricity because that is how their utility company delivers and bills them for electricity. People are less familiar with kWh to measure heat but both heat and electricity are forms of energy and so can be measured using any unit of energy such as kWhs, BTUs, therms or joules. Using these different terms for measuring energy is similar to using centigrade and Fahrenheit to measure temperature or miles and kilometers to measure distance, they are just different units for measuring the same thing. People are more familiar with using BTUs to measure heat because that is how the energy in natural gas is sometimes measured (though it is also measured in therms, one therm is equal to 100,000 BTUs). One kWh of energy is equal to 3,412 BTUs of energy. As long as you use the same units, (either kWh or BTUs) for the electricity used by the heat pump and the heat output of then you will get the right COP. COPs for heat pumps are usually between 2.0 and 4.0. The higher the COP the more efficient and the more money it will save you.

HSPF is similar to COP, except it measures the heat output in BTUs and the electricity used in watt-hours. There are 1,000 watt-hours (Wh) per kilowatt-hour (kWh). Hence, in the above example with 4kWh of heat output for every 1kWh of electricity used, there are 4 x 3,412 BTUs = 13,648 BTUs of heat delivered for every 1,000 Wh of electricity used. So the HSPF is 13,648/1,000 = 13.6. Mathematicians will quickly realize that HSPF can be converted to COP by dividing the HSPF by 3.412. The higher the HSPF the more efficient and the more money it will save you.

SEER measures the efficiency of the heat pump in cooling or air-conditioning mode. Now the output is the amount of energy removed from the room (which cools it down) in BTUs and the input is the amount of electricity used by the heat pump measured (just like HSPF) in watt-hours. 

I believe that consumers would be far less confused by these terms if HSPF was simply called “heating efficiency” and SEER was simply called “cooling efficiency”.

For SEER, HSPF and COP higher is better. Manufacturers are required to state the SEER and HSPF on the units in the energy label which is a bit like EPA’s the miles-per-gallon sticker you see on the windows on new cars.

However, the SEER and HSPF that are required to be stated on the equipment’s label are measured under continuous usage under ideal laboratory conditions. Under real-world conditions, HSPFs can be far lower than those stated on the equipment. It is as though the EPA measured the mpg on cars when they were all going downhill with a following wind.

In the real world, heat pumps often work in conjunction with a fossil-fuel heating systems and have to heat the house in short bursts in the spring and fall whereas the tests are done over long time periods of continuous operation, in other words, the tests simulate winter or summer operating conditions but not the conditions in which the heat pumps operate in spring and fall. Both the integration with the fossil-fuel furnace and the operation in spring and fall reduce the efficiency of the heat pump. Hence, the real-world, year-round COPs, HSPFs and SEERs are often lower, sometimes much lower, than those advertised by the manufacturers.

This comment is in reply to data shared by someone who had installed Mitsubishi Hyper Heat heat pumps and, like me, was not getting the performance he had expected.

Hi, XXX,  

This is great data and very consistent with both my own experience, that of academic researchers and the members of the Heat Smart Alliance. If you did a year-round average COP calculation (weighted by the energy used) then I would guess you would come out to a COP of between 2.5 and 3.0. When I do consulting work for other people, I use 2.5 in my calculations for ductless systems and 3.0 for ducted systems. Both numbers are far below the claimed COPs of manufactures which are in the 4-5 range. COPs, HSPFs and SEERs are all good ways to compare heat pumps from different manufacturers, but they are highly misleading if you use them to predict your energy-bill savings.

I agree with you that the short cycling reduces COP. Short cycling is not only a problem in the shoulder seasons but is also a problem if there is a back-up furnace that comes on with an outdoor temp of say 40°F. This mean that both the furnace and the heat pumps are short cycling which kills the efficiency of both. From the heat pump’s point of view, it does not care if the outdoor temperature is mild or if the back-up furnace is coming on to help, it leads to short cycling either way.

I believe that this effect is why people who have only a heat pump (and no back-up furnace) in cold locations (and have enough insulation to enable the heat pump to maintain 70°F year round) actually get better year-round COP than those with back up furnaces and / or milder climates. I have had such people on my webinars, and they cannot believe that other people (like you and I) aren’t overjoyed with our heat pumps. 

These short-cycling issues lead to lower COPs in practice than the manufacturers advertise. Academic research shows that the year-round actual COP of Mitsubishi Hyper Heat units in MA is about 2.5. If you are paying 23c/kWh (current Eversource or National Grid rates) then the 2.5 COP translates to a cost per kWh of heat in the house of 9.2 cents. Oil heat costs about 8c/kWh of heat in the house. This is paying $2.75 a gallon with an 85% efficient furnace. This is a 15% increase in your heating bill. If you are heating with natural gas, which costs about 5.2c per kWh of heat in the house (paying $1.45 per therm with a 95% efficient furnace), then your heating bill will almost double. This is why I disagree with the common conventional wisdom in MA that heat pumps are for everyone. The only way (in MA) to make heat pumps pay is to install solar at the same time (in which case, if the solar generates electricity at 8c/kWh of electricity – which is easy to do) then with a COP of 2.5 the cost of heat in the house is 3.2 cents or a 40% cut in your heating bill if you are heating with natural gas and a 60% cut if you are heating with heating oil. This cost saving can pay back the cost of the new heat pump in around 10 years for a return on investment of about 7% per year. It is only in these far more limited circumstances that I recommend heat pumps to people. My advice often comes as a nasty shock to the proponents of “electrify everything”. This is why I think it is essential to take “whole house” approach that looks at the financial returns on these investments. The momentum to cut carbon will be stopped dead in its tracks as soon as stories of doubled heating bills with heat pumps start to spread.

So, if you are suffering from high bills after installing a mini-split heat pump, here is my advice on how to improve the performance:

  1. Do not have upstairs heads and downstairs heads on the same outdoor unit. The upstairs heads will sometimes be trying to cool while the downstairs heads are trying to heat. This gets very confusing for a heat pump with very little brain.
  2. If the heat pumps have enough heat output to heat the house at a very low temperature, like 0°F, (this should have been determined in a Manual J calculation before the system was installed) then it will likely be more efficient to set the temperature at which the back-up furnace comes on to a very low temperature like 10°F rather than the common practice of setting it to 40°F which leads to short cycling (and inefficiency) in both the heat pump and the furnace. 
  3. If it gets too cold in the winter with the heat pump thermostat set at say 70°F, turn up the thermostat rather than turn on the furnace or turn on an electric fan heater. Keep turning it up until your feel comfortable. Only when this fails to keep the house warm, turn on the furnace.

And let me know if you have found any tips for getting better performance out of heat pumps!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.