Cold Climate Heat Pump Article

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.

peakbagger

Minister of Fire
Jul 11, 2008
8,845
Northern NH

It pretty well backs up my recommendations to have a backup to cold climate minisplits. Plenty of folks getting paid to say otherwise but more than few folks have learned the hard way that the actual performance ont he ground just doesnt work very well when its below zero. A great secondary heat source, but make darn sure there is cold temperature backup.
 
Last edited:
  • Like
Reactions: woodgeek and EbS-P
It doesn’t get that cold down here so we can use resistive strips until the power goes out. I agree no way your appropriately sized heatpump will maintain set point at record low temps.
 
In NC I expect you dont see below zero and folks in CT near the ocean probably do not but up in Northern NH we can have several days and nights in row where it does not get over zero. Its reportedly also tough on the equipment, I saw a recent article in VT about techs reporting compressor failures for sustained low temps.

As I mentioned before a friend of the family sells and service vented oil heaters for cold temp backups. They used to be Monitor brand that only run Kerosene and he keeps them running but now sells and installs Toyotomi units that run on heating oil. I think some folks are trying to run the rather sketchy chinese clones of German Wesbato air and water heaters used in semis and campers. The chinese clones are cheap but I have seen more than few people having issues with the clones and they definitely are not listed for residential use.
 

It pretty well backs up my recommendations to have a backup to cold climate minisplits. Plenty of folks getting paid to say otherwise but more than few folks have learned the hard way that the actual performance ont he ground just doesnt work very well when its below zero. A great secondary heat source, but make darn sure there is cold temperature backup.
I'm glad that we see below 20º weather only rarely here and below 15º almost never. Performance in that range makes heat pumps a good fit for a large chunk of the country. Insulation, better glazing, sealing, and smaller houses can help make a heat pump a better fit in colder areas. That said, we will always have a backup, and a backup to the backup because nature can throw a curve ball once in a while.
 
And the article refers to this document, (talking about the cold snap in February 2023) with consumer experience blurbs:


And I quite from that:

“I live on the top of a mountain in Waterford, Maine, where it
gets pretty windy. That’s not a challenge for my heat pumps,
which during the February cold snap kept me warm without
backup even during -49 °F wind chill! I’m also saving around
$300 a month using heat pumps instead of propane.”

And

“We live in a small 1930s cape cod and it routinely gets down to -20 °F in the winter. We replaced our failing,
25-year-old oil furnace with heat pumps. When we had forced hot air with an oil furnace, we were always
adjusting the temperature, and now we set the heat pumps and forget them. The heat pumps are also saving us
money. This past February when it was -25 °F, our electric bill was only $281 for the month, and we were very
comfortable. It’s also really inexpensive to run the air conditioning. This technology is great.”
Cathy and George H., Dexter, M

I would say that this does not confirm your point but instead provides anecdotal evidently to the contrary. (Of course wind chill temps are not relevant.)
 
  • Like
Reactions: pinetop12
And the article says
And in the single digits or low negatives, heat pumps can be closer to the 100% efficiency of an electric baseboard heater. Costs at this level are closer to that of oil heat, which usually has about an 87% efficiency rating.

I particularly note that last sentence.
 
And the article refers to this document, (talking about the cold snap in February 2023) with consumer experience blurbs:


And I quite from that:

“I live on the top of a mountain in Waterford, Maine, where it
gets pretty windy. That’s not a challenge for my heat pumps,
which during the February cold snap kept me warm without
backup even during -49 °F wind chill! I’m also saving around
$300 a month using heat pumps instead of propane.”

And

“We live in a small 1930s cape cod and it routinely gets down to -20 °F in the winter. We replaced our failing,
25-year-old oil furnace with heat pumps. When we had forced hot air with an oil furnace, we were always
adjusting the temperature, and now we set the heat pumps and forget them. The heat pumps are also saving us
money. This past February when it was -25 °F, our electric bill was only $281 for the month, and we were very
comfortable. It’s also really inexpensive to run the air conditioning. This technology is great.”
Cathy and George H., Dexter, M

I would say that this does not confirm your point but instead provides anecdotal evidently to the contrary. (Of course wind chill temps are not relevant.)

Wind chill temps are not relevant, but at the same time, wind can be. I have a 1960's house that is 2x4 construction and no house wrap. The heat always runs more on windy days, regardless if it is 45* or in the negative numbers. For homes that have better insulation and tightness, there is probably less impact.
 
I do not disagree that cold climate heat pumps are great things if sized and installed properly and even for my climate they may be suitable for 80% of the heating load plus our limited cooling load to boot. Mine has been my sole source of heat for a month or so despite occasional overnight temps in the 30s and day time in the mid forties during a recent colder than normal wet stretch. A normal home owner without wood with a "hot" boiler system would be running the boiler with a pretty good standby loss so less carbon would be going up into the air by far with a minisplit assuming a reasonable renewable mix on the gird. Here is snap shot of ISO New England's mix as of 8:20 AM Monday 5/1 . the Least Marginal Price (LMP) on todays grid is natural gas so as the demand goes up during a normal New England work week, additional power is going to come from natural gas.

1682943606047.png


The one caveat is -10 F is -10 F no matter how tight a house is and that is where folks even in superinsulated homes are getting burnt. If a house is built tight it has an air to air heat exchange and in many cases, the contractor with experience oversizes an electric duct heater on the incoming air supply and that may be enough to carry the house through a cold snap. On the other hand, the state like VT and Me are really pushing them on older homeowners in older housing stock as a replacement for oil and the owners dont have the money to keep a spare tank of oil and the equipment maintained and that usually means a tough first or second winter when the oil runs out or will not start. Even with high oil costs, electric costs are frequently higher than oil on a BTU basis at a COP of 1.0.

With respect to those claiming unusual great performance in sub zero conditions, it calls into question what they have for a system setup. A typical Maine home has 100,000 btu oil (or rarely gas). In many cases, in very cold conditions the boiler is running almost constantly to keep up with the heat loss. Very few of the minisplit installs I have seen have the equivalent of eight 1 ton units hung off the house to supply the same load. In order to get the best performance at low temps, single head cold climate minisplits are required (multihead units are noticeably less efficiency on a COP basis). Unlike much equipment, there are no economies of scale for minisplits, the best COPs are consistently in the smaller sizes. In most cases I see one maybe two units usually 1 ton units hung off a house. That means that there is going to be another source of heat in the house for very cold temps or the homeowner has modified their heating usage substantially. Given that most older heating systems in Maine were installed cheap with at best one zone per floor, the whole house gets heated to one uniform temp. In the case of minisplits, the recomendation is to keep one "warm room" supplied by the minisplit and then keep the rest of the house at a much lower temp. Thus the heating comparison is not apples to apples. Talk to any heating tech and they spend a lot more time these days chasing frozen lines in homes that try to keep the heat down low in most of the house while keeping warm rooms. There are fixes but it is not publicized until after the freezeup occurs.

Both sides in the "heating wars" in Maine and VT have their incentives in bad mouthing the other and both adopt comments from the general public that support their side. I know the friend that sells and services vented oil heaters (Monitors) is over 70 and was only in business to support long term customers and assumed he would have retired long ago. As it is, his daughter with a liberal arts degree went out and got her heating license and has taken over the business with him working part time to cover the recent demand.

I keep looking for independent numbers based studies making direct comparisons on truly cold climate operation (-10 F or lower) of heat pumps but most of the time what I see are studies by groups advocating for one source or the other and usually they are tweaked to favor their side. The problem is with heating using seasonal numbers just doesnt tell the story, heating systems need to be designed for the worst case scenario or keeping the house at a comfortable temperature for a 3 to 5 day period of sub zero temps. AGW is making those scenarios less often but they happen on occasion even in relatively modern homes compared to the average housing stock of the region as evidenced by a shot ai took in my bathroom one very cold morning this past winter.BTH Ceiling.jpg

VT homeowners are soon to become the guinea pigs for the experiment, the Affordable Heating Act soon to become law effectively put in significant and increasing penalties to not go to electric based heating (or biodiesel) in existing homes. This is in state that has a known substantial deficiency in the regional and many small local distribution grids where rural winter outages are expected and on occasion last for days. The relatively well off folks in the Burlington and Montpelier areas probably will be fine but the folks living in the old drafty housing stock in the rural areas are going to have to take what is free and subsidized and learn to live with it.
 
Last edited:
  • Like
Reactions: sloeffle
That may all be true, but it does not explain how a 1930's home can remain comfortable at -25 F, and run (in Dexter, Maine) only $281 for the month of February.

Either you are saying that is a non-truth, or it's simply possible despite engineering concerns.

I agree a heat pump is best when it's not as cold. But the above shows it's good when it is that cold (too).
I agree that it's great to have another source of heat (especially with electric cables above ground...) - I have heat pump, oil, and a wood stove.... I added the heat pump, notably without taking away the oil heat.
But the constant "it won't work when it's cold" discourages people from installing them (and leaving the current heating system in place), and does not seem warranted given the attestations quoted in that document. Unless those are lies.
 
Wind chill temps are not relevant, but at the same time, wind can be. I have a 1960's house that is 2x4 construction and no house wrap. The heat always runs more on windy days, regardless if it is 45* or in the negative numbers. For homes that have better insulation and tightness, there is probably less impact.
Our old farmhouse is the same way. Windy days definitely take more heat to keep the interior warm.
 
  • Like
Reactions: SpaceBus
That may all be true, but it does not explain how a 1930's home can remain comfortable at -25 F, and run (in Dexter, Maine) only $281 for the month of February.

Either you are saying that is a non-truth, or it's simply possible despite engineering concerns.

I agree a heat pump is best when it's not as cold. But the above shows it's good when it is that cold (too).
I agree that it's great to have another source of heat (especially with electric cables above ground...) - I have heat pump, oil, and a wood stove.... I added the heat pump, notably without taking away the oil heat.
But the constant "it won't work when it's cold" discourages people from installing them (and leaving the current heating system in place), and does not seem warranted given the attestations quoted in that document. Unless those are lies.
IMHO its a non truth by an unqualified individual homeowners promoted by organizations that have financial or organizational interest in getting cold climate minisplits installed. BTW I am a PE that used to do third party performance testing of power cycles to ASME and other standards so when it comes to testing, I do have more experience than a typical homeowner. Our tests were on occasion subsect to litigation and we made darn sure that we were comparing apples to apples down to the smallest statistical degree that the client was willing to pay for or within the available instrument accuracy.

Years ago, Professor Hill (now deceased) at U Maine during the first energy crisis was sick of hearing all the claims about certain types of building constructions being "more efficient" to heat. He was mostly irked at the log cabin folks who constantly pushed that log cabins had more energy density and far more efficient. He wanted to come up with a way of proving that a lot of it was BS so he had built a series of identical footage buildings in an open field with different types of construction including a log cabin. Each had the same interior volume, sun and wind exposure. I think most people on the project referred to them as the "outhouses" as that was what they looked like. Each had an exterior door, a temperature controller on the interior and some 100 watt light bulbs connected to the controller for heating each building with a power meter on each building. They also did not have an occupant to screw things up. He set the temps inside to the same value and used his grad students that worked for him and lived in his apartment building nearby to record the data and left them over a course of year. It was a very easy to understand study and disproved many fallacies including a major hit on log cabin efficiency claims.

In the case of minisplits in very cold climates as the only heating source, show me the unoccupied "outhouses" study. To date all the so called studies I have seen are observational studies based on occupied structures of varying construction, exposures and location that probably have a fair share of bias and placebo effects. My contention is that no one in the business is really interested in funding true third party testing unless the results line up with their commercial and organizational interests and biases.

To paraphrase Tom Cruise, show me the data!
 
Last edited:
That may all be true, but it does not explain how a 1930's home can remain comfortable at -25 F, and run (in Dexter, Maine) only $281 for the month of February.

Either you are saying that is a non-truth, or it's simply possible despite engineering concerns.

I agree a heat pump is best when it's not as cold. But the above shows it's good when it is that cold (too).
I agree that it's great to have another source of heat (especially with electric cables above ground...) - I have heat pump, oil, and a wood stove.... I added the heat pump, notably without taking away the oil heat.
But the constant "it won't work when it's cold" discourages people from installing them (and leaving the current heating system in place), and does not seem warranted given the attestations quoted in that document. Unless those are lies.

So, we just have in the example is that it is a 1930's house, with no other information. Although the testimony might be truthful, it's tough to know what it actually means or use it as a comparison to any other house.

We don't even know if they had additional work done on their house that would skew the numbers on what they say they are saving. Were their ducts newly insulated along with the heat pump install? Also, their oil burner was 25 years old. Who knows what the difference would be between a new oil burner and the heat pump?

Oh, and then there is the chimney (I assume there was one since it was an old oil burner). Was that capped off and insulated - or maybe even removed? And if that chimney was freestanding in the middle of the house (like mine), that allows a whole lot of cold air to come in. So that is a whole nother discussion.

Maybe, nothing else was done to the house. The point is the blurb does not have enough information to actually tell us anything.
 
Last edited:
  • Like
Reactions: stoveliker
Good discussion. I'm sympathetic to @peakbagger's skepticism, but would just add that there is a large variation in heating loads from home to home, due not only to the variety of construction styles and vintages, but also to seemingly minor details of airsealing that are baked in during construction. Including his mention of 'the occupant screwing things up' by not putting down storm windows, closing flues, etc.

I am also convinced that many heat pumps (especially split systems) are installed incorrectly in ways that impact their performance relative to laboratory conditions.

So it seems to me likely that two similar houses next to each other might have one that 'works' on a heat pump, and the one next door really 'doesn't'.

All of the above is for conventional heat pumps in a mild climate. Add in the fancy tech in the compressors to allow more BTU output at low temps, and I expect the factors I mention get even worse.

Houses and HVAC systems are not built on an assembly line with QA qualification. They are built piecemeal by a poorly coordinated team of low bidders who know that their hasty work only needs to pass a cosmetic inspection. HP installers I have dealt with (years ago now) will _underpromise_ the performance of their wares so that they can say 'I said it wouldn't work' when you call them back later. Which is a travesty compared to, I dunno, providing a warranty that their install will perform at the specifications of the same unit, or even, gasp, every install being qualified with empirical measurements to be functioning as intended (which would require visiting the owner on a cold design day).
 
I’m convinced that the average consumer has no ability to tell if HVAC work is done to a minimum standard and following manufacturers directions/guidelines. To push equipment to perform near the edge of its design envelope the install matters. Sure you will have equipment failures. Manufacturers probably have the data.

I do think one needs to figure the annual cost of maintaining your secondary heat source and not just the cost of the fuel. Heatpumps need regular maintenance too. Indoor coils need cleaned at so point. Outdoor too.
 
I do think one needs to figure the annual cost of maintaining your secondary heat source and not just the cost of the fuel. Heatpumps need regular maintenance too. Indoor coils need cleaned at so point. Outdoor too.

Um, not in my experience. I replace the high MERV filter on the return a few times a year. And my indoor coil when I have looked at is clean as a whistle after 15 years. I hose off the outdoor coil (and cut back the plants trying to smother it) maybe every year or two?

I called a company for a discounted $50 'check up' once, and he told me I was dumb for having a heat pump, my system would never work, and that me calling him for a check up was a waste of time and money, bc there was no maintenance to do on them. Hahaha.

Versus my oil boiler that needed an annual cleaning (and sometimes coked up in less than a year) and often repair calls on top of that. Figure $200-300 year on top of the $2000-5000 fuel bill.
 
Um, not in my experience. I replace the high MERV filter on the return a few times a year. And my indoor coil when I have looked at is clean as a whistle after 15 years. I hose off the outdoor coil (and cut back the plants trying to smother it) maybe every year or two?

I called a company for a discounted $50 'check up' once, and he told me I was dumb for having a heat pump, my system would never work, and that me calling him for a check up was a waste of time and money, bc there was no maintenance to do on them. Hahaha.

Versus my oil boiler that needed an annual cleaning (and sometimes coked up in less than a year) and often repair calls on top of that. Figure $200-300 year on top of the $2000-5000 fuel bill.
You change the filter regularly what nonsense ;) And you cut the weeds back and hose it down???? Unheard of! ;)

All kidding aside. That’s really all it should take. But they can get dirty.
 
  • Like
Reactions: woodgeek
You change the filter regularly what nonsense ;) And you cut the weeds back and hose it down???? Unheard of! ;)

All kidding aside. That’s really all it should take. But they can get dirty.
That is all the maintenance our HP system has had after 17 yrs of operation. The prefilter gets changed every year and the big 4" merv filter about every 3 yrs.
 
That is all the maintenance our HP system has had after 17 yrs of operation. The prefilter gets changed every year and the big 4" merv filter about every 3 yrs.
My Gree and Midea minisplits have been running non-stop for 3 years set to 70. I have literally never turned them off since the electricity to power them is essentially free from my solar.
 
In NC I expect you dont see below zero and folks in CT near the ocean probably do not but up in Northern NH we can have several days and nights in row where it does not get over zero. Its reportedly also tough on the equipment, I saw a recent article in VT about techs reporting compressor failures for sustained low temps.

As I mentioned before a friend of the family sells and service vented oil heaters for cold temp backups. They used to be Monitor brand that only run Kerosene and he keeps them running but now sells and installs Toyotomi units that run on heating oil. I think some folks are trying to run the rather sketchy chinese clones of German Wesbato air and water heaters used in semis and campers. The chinese clones are cheap but I have seen more than few people having issues with the clones and they definitely are not listed for residential use.
We hit -5 here on the shoreline in CT during that cold shot in Feb and my Gree and Midea that are rated at 80% capacity at -22 farenheit performed flawlessly.

Here's my energy usage from that day. My entire house and both mini splits were powered from my solar for most of the day as well.

Screenshot_20230210-055515_Monitor.jpg month_2__year_2023__station_HVN__network_CT_ASOS__dpi_100.png
 
If someone has pets, definitely change the filters. The condensate pan can also get pretty ugly.
 
We hit -5 here on the shoreline in CT during that cold shot in Feb and my Gree and Midea that are rated at 80% capacity at -22 farenheit performed flawlessly.

Here's my energy usage from that day. My entire house and both mini splits were powered from my solar for most of the day as well.

View attachment 312542 View attachment 312543

Nice.

There were 5 days in February where I had absolutely zero solar production. Running both pellet stoves for all my heat in February 2023, I produced a total of 251 kwh and used 305 kwh. February of 2022, I used the minisplit for heating when it was at 15* or higher and pellet stove on main floor when lower temps (basement pellet stove is only heat source so it is always in use), and 542kwh was my electric usage.
 
We hit -5 here on the shoreline in CT during that cold shot in Feb and my Gree and Midea that are rated at 80% capacity at -22 farenheit performed flawlessly.

Here's my energy usage from that day. My entire house and both mini splits were powered from my solar for most of the day as well.

View attachment 312542 View attachment 312543
And yet I keep hearing this is not possible....
 
  • Like
Reactions: Highbeam
We hit -5 here on the shoreline in CT during that cold shot in Feb and my Gree and Midea that are rated at 80% capacity at -22 farenheit performed flawlessly.

Here's my energy usage from that day. My entire house and both mini splits were powered from my solar for most of the day as well.

View attachment 312542 View attachment 312543
Is this supplemented with wood or another heat source?
 
Is this supplemented with wood or another heat source?
No supplemental heat was used. Its very rare to get subzero temps here so I used them as a test and didn't have a fire to see electricity consumption and see what indoor temp they could hold. No issue holding their 68 degree set point.

They were running at full out max capacity at the peak cold but I was still probably getting close to a cop of 2 even at -5. It's rated a cop of 2.08 putting out 13k btu at 5. Cop is I believe around 1.76 at -22 with 80% capacity.


GreeSUbScreenshot_20230306-170145_Reddit.jpg