GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Building Science

Will a Heat Pump Water Heater Freeze Your Basement?

Study demonstratesthe effects of heat pump water heaters on surrounding air temperature

Are you considering replacing your current water heater with a heat pump water heater but worry that it’ll freeze your basement in winter? Well, enough homes have this new technology that we now have some data on this topic. A new paper by Slipstream is just out with some results that may interest you. Titled “Installed Performance of Heat Pump Water Heaters in a Cold Climate,” the paper covers many facets of heat pump water use. I’m going to discuss only one here: the effect on air temperature.

The homes in the study

First, note that I’m going to show data below from two different parts of the study. One was a field study with 9 homes. The other part was a survey that included the 9 field study homes plus another 72 homes, for 81 total.

The 9 homes in the field study were in rural Michigan. The heat pump water heaters were all in partially conditioned basements. Seven were in the main part of the basement; two were in a separate room. All of the water heaters were 50-gallon models from Bradford White, Rheem, and AO Smith. The Slipstream researchers monitored water temperature (hot and cold), water flow rates, electricity consumption, inlet air temperature, and indoor and outdoor temperatures.

The objectives of the study were:

  1. HPWH Performance: Characterize measured energy factor and effective capacity at different operating modes.
  2. Economics: Calculate cost and energy savings compared to an electric resistance water heater.
  3. Space heating impacts: Quantify changes in basement temperature and estimate HPWH’s impact on space heating and cooling operation.
  4. Customer satisfaction: Survey on experience and satisfaction with installation and performance.

(I encourage you to read the whole study, if you’re interested in this topic. There’s a lot in it!)

Effect on basement temperature in the field study

First, the field study monitored the basement temperature at four heights at a distance of 5 to 10 feet from the heat pump water heater. The graph below shows the cycles of the heat pump water heater at the top. In this case, there were three cycles during that 28-hour period.

The lower part shows the air temperature in the basement at the four temperature sensors. One interesting feature of the temperature data is what they show about the cycling of the water heater and the furnace. The floor sensor shows the water heater cycles most clearly—the big dips coincident with the water heater cycles. The ceiling sensor shows the furnace cycles most clearly—the choppiness of the curve, especially between about 1 a.m. and 6 p.m.

Changes in air temperature near heat pump water heater
Changes in air temperature near heat pump water heater (Fig. 19 from the Slipstream paper)

In terms of the effect on basement temperature, the data above show a drop of about 4°F near the floor as the heat pump water heat operates. Then the temperature recovers pretty quickly. For the 9 homes overall, the average temperature drop during a cycle was 2.3°F. After a 4-hour recovery period, the average basement temperature was 0.1°F lower than it was before the cycle. The study didn’t measure the long-term effect on basement temperature over the course of a whole winter, but that would be interesting to see.

Note that these basements aren’t really cold. The range of temperatures at the beginning of a heat pump water heater cycle was 62 to 65°F. My unheated, uninsulated (for now!) basement in Atlanta has been about 60°F for the past 6 weeks. The basements in the field study, recall, are partially conditioned, but they aren’t independently heated. In other words, the thermostat is upstairs.

Satisfaction with effect on temperature and humidity

The other part of the study was a survey, and one of the questions participants were asked was, “Do you like the changes to temperature and humidity?” The results, separately for summer and winter, are shown in the chart below. In both seasons, the majority had no opinion. Seventeen people liked the changes in winter. Only 9 people didn’t like the cooler basements.

I have no idea why anyone would like their basement to be cooler in winter. Then again, I don’t understand why someone would choose to live in a cold climate that’s not near a mountain with 2,000 vertical feet of ski runs.

Heat pump water heater survey results on the question: "Do you like the changes to temperature and humidity?" (Fig. 25 from the Slipstream paper)
Heat pump water heater survey results on the question: “Do you like the changes to temperature and humidity?” (Fig. 25 from the Slipstream paper)

It’s easier to understand the summer results for this question. If a heat pump water heater keeps a basement cooler and drier in summer, that’s a great thing. One of the survey respondents wrote: “One added benefit: I do not have to run my dehumidifier in the basement anymore.”

Factors to consider

If you’re on the fence about putting a heat pump water heater in your basement (or elsewhere inside your house), here are some things that can affect how much impact it will have.

  • Amount of hot water use. The more hot water you use, the more the heat pump water heater will run and cool off the house.
  • Water heater mode. These things come with a heat pump and electric resistance heating (strip heat). You can operate them in one of three modes: all heat pump, all strip heat, or a hybrid mode that supplements the heat pump with strip heat as needed. If you use heat pump only, it will do more cooling.
  • Size of the basement. The bigger the basement, the more air there is to dilute the cooling effect.
  • Basement use. If you have an office in the basement, you may need a way to heat the space independently of the upstairs. If your main use of the basement is for exercise, you may welcome the cooling.

The Slipstream heat pump water heater study shows that fears of freezing your basement are overblown. And if you find that your basement does get too cold, you can always turn it to strip heat mode.


Allison A. Bailes III, PhD is a speaker, writer, building science consultant, and the founder of Energy Vanguard in Decatur, Georgia. He has a doctorate in physics and writes the Energy Vanguard Blog. He also has written a book on building science. You can follow him on Twitter at @EnergyVanguard.


  1. Tim_O | | #1

    The study goes a bit further into this, but leaves it as inconclusive - if you have a water heater in 50-55* dirt (Michigan basement in the winter) and the water heater is pulling from 60* basement air, those BTUs are coming from your furnace. The study doesn't have enough fidelity to show this it seems like. But HPWHs don't get to bypass the laws of thermodynamics. The heat comes from somewhere, and as long as the temperature around the HPWH is above the earth or outdoor temperature, it came from the home's primary heat source.
    Maybe an uninsulated root cellar or other section of the basement that is isolated from the insulated basement would work? Then the heat pump can pull air that is heated directly by the concrete walls to the 55* dirt temperature. I assume the efficiency of the heat pump would drop, but the efficiency of the house as a system would rise.

    Simply put - in a heating dominated climate, to take advantage of the efficiency of a HPWH, it needs to be decoupled from the homes heat. Otherwise, a water heater of the same fuel as the primary heat source will likely be most efficient. In the case of a heat pump heated house, this still is a HPWH, but overall COP is reduced.

    EDIT - I'm not disagreeing in any way with the authors. They tried to prove this out, but weren't able to get the data. Appendix B talks further on that.

    1. Expert Member
      Michael Maines | | #2

      I have read other articles that argue that although HPWHs do indeed "steal" heat from the space-heating system, because they don't operate continuously, don't draw a large amount of heat and provide desired cooling and drying in warmer months, they are still a net positive investment.

      1. Tim_O | | #3

        Yes, you'd have to balance out your warm months and see if the gain there is worthwhile.

        From a comfort perspective, you won't really notice them in your basement much, which I think this article seeks to prove. But from a BTU perspective, 100% of the energy used by the heat pump is from your home's primary source of heat. Unless you can thermally decouple it.

        1. chimewind | | #8

          Also not having a full winter of data makes it hard to make a judgement. My basement does not have a heat source. It is considerably colder near the end of winter than the start. Having anything remove heat in the latter half would be detrimental.

      2. StephenSheehy | | #11

        We installed one a few months ago, replacing an electric resistance water heater. I'm looking forward to seeing what difference it makes in overall power use, since it's an all electric house and the water heater is the only thing that has changed. I figure that after a year, most variables from weather should balance out. If I'm still around, I'll report back.
        The HPWH is in the mechanical/laundry room.

  2. aaron_p | | #4

    Having owned a GE Geospring in climate zone 6 for a couple years - I can confirm that the argument for them stealing enough heat to offset their savings is more of a theoretical exercise. In the real world we had fairly similar results to the article - it was hard to quantify the losses as it seemed to be within the noise of normal usage.

    We had a townhouse primarily baseboard electric heat in most rooms with a gas fired fireplace on the main living level and a gas fired heater in the primary bedroom. The heat pump water heater was located in a closet off the full bathroom with louvered doors - that bathroom only had electric baseboard heating. The utility bill savings from the switch from an old electric tank water heater to a heat pump unit saved considerably on our electric bills as a whole. The gas bill usage saw savings from the previous usage, but that was likely due to prior occupant usage (programmable thermostats vs manual ones). Overall we saw about a 50% reduction on the overall electric usage.

    HPWH either completely offset or lessen the dehumidifier usage and that with the cooling benefit in summer (vs a heat load from a dehumidifier) will likely be a wash in my experience with any heat stealing losses in the winter. The net result is still somewhere close to the COP difference between the resistance only tank and the heat pump tank. There is an argument that we probably also lessened the standby losses from the old tank due to newer insulation requirements, but I think the end result is still the HPWH will be a significant savings for most people.

    P.S. - I think the bigger issues to discuss are the noise issues and the longer term maintenance costs. If the heat pump technology gets a bad reputation from the general public it can be hard to convince them to adopt it as the technology progresses in the future. Also, if the ROI calculations are disrupted by the loss of the heat pump unit due to complex repair issues, consumers might be better served by the lifetime tanks. I know when we put it in the utility rebates made it a no brainer as the HPWH was about the same cost as a conventional electric tank, but servicing the unit was the big wildcard. No plumbers around us knew what the were at the time and I doubt HVAC techs would want to service/diagnose them either. (we didn't have any issue, but I had worried what we would do if we did).

  3. USER_1977 | | #5

    I am glad to see this recent study discussed; it helped inspire me to log temperatures in a few locations in my basement earlier this fall. In 2015, Ecotope performed a similar study in the Pacific Northwest looking at temperatures in basements (as well as garages) with a HPWH, finding relatively modest impacts on unoccupied basement temperatures. They also assumed that the HPWH has a negligible impact on space conditioning loads, although the Pacific NW Regional Technical Forum now assumes that ~50% (25 to 75% range) of the heat extracted by the HPWH is supplied by the space heating system.

    The question I have is: what happens to basement temperatures in colder climates when you deploy a HPWH in a basement *without* space-heating systems, ideally both for homes with and without basement wall insulation? This is the case of deploying ductless minisplit heat pumps as the sole source of space heating in existing homes in colder climates and could have nontrivial HPWH COP impacts (both from a lower cold-side temperature and, potentially, compressor lock-out), as well as an increase in space heating loads.

    ROM calculations show that a HPWH extracts ~(120F-50F) * (55 gal * 8.31 pound/gal) * 1 Btu/pound * (2/3) = ~21 kBtu/day from the basement air to the water (the 2/3 is estimated portion of heat from ambient to total heat into the water assuming no electric resistance operation; the rest is the electric power supplied). If a 70kBtu furnace or boiler loses ~2% of its heat input into the basement over 6h runtime/day, that is ~8 kBtu/day of heat that would no longer be available that has to come from the first floor or the basement floor (once/if the basement temperature dips below that of the basement floor, ~ground temperature).

    I know one practitioner who is reluctant to insulate between the first floor and unoccupied basements specifically to avoid colder basement temperatures, and would be very interested in hearing about others' experiences with this HPWH use case.

    1. Tim_O | | #6

      Put it next to your chest freezer in the basement. In this report, it was a Michigan house, so there is a 50/50 chance there is a chest freezer full of venison. I had a Kill-o-watt on my fridge for a while, and it used about 200watts. That would be about 16k btu/day if continuous. And that doesn't take into account the COP of the compressor.

      1. hockipuck | | #7

        Tim, we've had a Rheem HPWH for almost 4 years and I've come to using it as such in spring, summer, and fall but switch it to standard for the winter. It's in our laundry room which is in an occupied walkout basement. The basement is cooler than the rest of the house without the water heater (2-3F). In the summer it keeps the basement around 68F with the upstairs at 74-76F. In the winter it drops the basement temperature to 60F, which is outside of the comfort range for us. I'm happy with the significant cost savings 3 seasons of the year and no worse off than before during winter. If we get a mild week I'll switch it back to heat pump only for the dehum help, but I doubt if most people would change their settings regularly. One caveat is that our heating heat pump is undersized for loads below about 30F OA-T and I have the heat strips turned off so we may not notice the discomfort as much if the heating kept up.

  4. oceaneric | | #9

    Not a a high tech. reply; but a real life test with the same concern.
    I've been running a A.o. smith 60 gallon heater pump water heater for 3 years in my personal home with a family of 5 in Maine:Two adults and three kids aged 3 to 11. There is baths and showers and laundry. The heat pump water heater is in an unconditioned and old corner of a walk out basement .
    Simple math: there is a huge tank of hot water in the space. I have a digital thermometer metering the temp around my unit and next to the fan. It is 50-60 degrees all winter here in Maine.
    Freeze stuff-If you ran it in a chest freezer, maybe??

  5. USER_1977 | | #10

    @oceaneric and @hockipuck- Thank you for sharing the experience for your basements. It would be helpful to have more context about them to better understand basement heat flows, to wit:

    1. Is there a furnace, boiler, or heat pump in your unoccupied basement?
    2. Are the basement walls insulated?
    3. Is the basement floor/slab insulated?
    4. Is there insulation in the basement ceiling?

    FWIW, the basement in our two-family in Massachusetts has two boilers, two gas-fired WHs, and lacks insulation. Given that it is pretty open, I would expect replacing the gas water heaters with HPWHs should not create a problem, but I am more concerned about the HPWHs in January and February if we replaced both boilers with ductless minispilt HPs.

    1. oceaneric | | #13

      User 1977, In my case there is no heat source at all in the walkout style basement. I do have a forced hot air oil furnace, but I only use it for power outages with a small generator as the two upper floors are heated with a mini split. The basement walls are insulated, but to fairly less than desirable standards and one larger wall with none yet. The floor slab is not insulated and the ceiling is with just fiberglass between joists.

      This space will be fully upgraded to better energy standards, but what really impressed me is that I have not yet seen the unit switch out of heat pump mode to resistant mode, It may when I am not checking, but on the coldest Maine mornings I have yet to see it.

    2. Duck_Bill | | #15

      I have a HPWH in an old, uninsulated basement in cold RI (design temp = ~10F). The house is now heated only with mini-splits, so no heat source in the basement. In mid-winter, the steady state of the basement tends to be about 50F without the HPWH working, and >45F with. Remember that the basement wants to revert to the temp of the surrounding soil, so any reduction from the HPWH working is quickly overcome by the surrounding walls/floor temps. I have yet to see the HPWH revert to all electric.

      One other consideration is the HPWH will make your basement ceiling that much colder; if it's not insulated, the first floor will be cold!

  6. paulrey27 | | #12

    I'm intrigued by the idea of heat loss... because I have a partially finished basement through which all my HVAC ducting runs - all uninsulated and leaky. Once our gas-fired tank dies, I'm very interested in an HWPW.

  7. Matthew_M | | #14

    Rheem's hybrid water heaters throw an error code if ambient temps are too cold, and reverts to resistance operation. A home owner reported this to me, and if I recall correctly, the temperature was around 38F in the basement.
    This was in a particularly leaky basement that was since addressed.

    So, the Rheem is designed to not freeze your basement.

  8. PAUL KUENN | | #16

    We've now had our heat pump WH for over 10 years and love it. No resistance heaters allowed. All run on solar PV along with the HP dryer. 1960 basement with cold lower walls (read my GBA blog about adding 6" of EPS to top 5' of walls on the exterior). There are two larger 80 gallon tanks for solar thermal floor heating storage next to the DHW tank. We see no difference in the temp of the basement and do keep a thermometer down there just to see where we're at. Zone 6 near GB, WI.

  9. rwsac | | #17

    Something that I have not heard or read on this topic is that, yes, the heat pump is pulling heat from the basement, BUT any standby loss from the tank is also going back into the basement. The net loss seems to me to be only from the water that is used. Even heat loss from hot water in the distribution pipes in the basement (even if they are insulated) goes back into the basement. I look at the HPWH as a form of recovering some of that heat, which other DHW methods cannot do. I have had a HPWH in my conditioned basement in Zone 5 for 4 years and would never go back to resistance or even gas.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |