Heat pumps can't take the cold? Nordics debunk the myth | Hacker News Reader

Heat pumps can't take the cold? Nordics debunk the myth

techxplore.com

139 pointsthunderbong149 comments

Wow, I was not even aware this myth was a thing. I'm thoroughly confused. Here in northern Sweden, geothermal heat pumps are seen as pretty much the only viable option for houses that aren't connected to district heating. We get -25 degrees in winter here and it's simply a non-issue. Heat pumps are so dramatically cheaper and more efficient than any other option in cold climates that no one even considers installing electric radiators or oil/gas furnaces. The only other real option that people actually use is a wood or wood pellet furnace.

It feels like near enough every home owner I talk to either has or is planning to drill for a geothermal heat pump. Air-air heat pumps are popular too for secondary spaces like garages or if you're on a tight budget.

The top five countries in terms of heat pumps per capita all being in northern Europe* should tell you something about this myth:

1. Norway

2. Finland

3. Sweden

4. Estonia

5. Denmark

* https://www.theecoexperts.co.uk/heat-pumps/top-countries

Here in the mid-Atlantic, my air-air heat pump struggles once things get well below freezing. Which it does a few days a year.

On those days, there's a backup resistance heater. Which is vastly more expensive, but since it's only run on occasion, it wouldn't be worth it install a geothermal heat pump.

I would love to, but the additional cost is tens of thousands of dollars. The since the alternative costs perhaps a couple of hundred dollars per year, the time to breakeven is many decades.

It's still quite clear that a heat pump is a very good all-electric option, especially since it provides cooling in the summer. Which is absolutely mandatory around here, and will get more and more so over the years. Which, ironically, will also reduce the use of the resistance heater from "very rare" to "practically never".

Why/how does it struggle?

The new propane or CO2-based pumps the temperature it keeps working is also lower so it isn't much of a problem for most places if you are in market for one currently.

What type of heat pump do you have? I have a ducted split system with a HSPF of 9.5, and I'm actually pretty disappointed with how it performs. In our new-ish, better-than-average insulated house, we've had some insane electrical bills over the last couple winters compared to the gas bills we had in previous years (prior to renovating).

Sometimes I wonder if the installer didn't really know what he was doing.

What heating elements or output temperature are you runnng your HP at, and do you have buffer storage or not?

It's an air-air ducted system. I live in the North East US, so winters are generally cold, with a few days around 0F/-17C. During heating season we keep it at 67F/19C.

That's a COP of 2.8. You have to check the efficiency graph and your mean winter temperatures, because I suspect it operates fully resistive at below -10°C or so. I also noted underwhelming performance and increased faulty compressor incidence for air source heat pumps with the external unit mounted on Northern walls (in the Northern hemisphere). I suspect it doesn't get sunshine and the defreezing resistive element is operating a lot, driving up energy costs.

How old is your aircon? I bought some cheap Chinese "mini split" 3.5kW units locally for about 350 euros that go down to about -15 at about 2.5x effective heating to input (watts). They're closer to 4x at higher temps. Cooling runs at about 7x.

Aircons have become much cheaper and more effective over time. It might be worth looking at replacing your unit, assuming it's not something much more elaborate.

Old enough that I'm not gonna say out of sheer superstition. Old enough to raise eyebrows on the repair guy.

When it finally has to be replaced the new one will assuredly be more efficient. But since it's already quite cheap to run (the very most expensive bill is perhaps $200 during a very hot August with me working from home), I'd love for this one to keep going as long as possible.

I'm cooling an approx 30m2 apartment with brick (cinder block) walls that are not insulated and I run the aircon continuously 24/7 all summer and use about 200kWh per month (approx $35 for me). Daytime temps are usually between 31 and 37 celsius and nighttime around 21 to 25.

> Here in the mid-Atlantic, my air-air heat pump struggles once things get well below freezing. Which it does a few days a year.

How far north are you (IECC climate zone)? What is the make and model of your unit? Is it a cold climate air-source heap pump (ccASHP)?

* https://ashp.neep.org/#!/

* https://neep.org/heating-electrification/ccashp-specificatio...

Because there are folks in Climate Zone 7 in the northern tip of Maine (Caribou) that use heat pumpss

* https://www.youtube.com/watch?v=OcwIz6heDss&t=43m21s

Mitsubishis can work down to -13F/-25C and still have a COP of 2.08 (1=resistive heating):

* https://www.mitsubishicomfort.com/articles/keep-warm-this-wi...

* https://mylinkdrive.com/viewPdf?srcUrl=http://enter.mehvac.c...

I live in Maine, but not as far north as Caribou.

The units we use up here will work flawlessly down to about 10f, and keep working until approximately -20f. Thanks to climate change, I haven't seen -20f at night in several years, but the last time it happened, the heat pump struggled to keep my house warm but still managed to do it.

"Most" of Canada only gets about this cold, where "most" is the majority of the population that lives near the border.

THere's something wrong with your heat pump, or you needed/need a different unit.

I live in New Mexico @ 6000', I have Mitsubishi Hyper units, and it has to get down to below 10 degrees before you can even see any difference in the heat pumps' operation. Even so, they still get the house up to temperature, just takes a bit longer. The difference in cost over regular Mitsubishi's was low single digit thousands.

But geothermal heat pumps are interesting in cold climates because air-air and air-water heat pumps (the most common and affordable types) are less efficient below freezing, so I feel like that confirms rather than dispels the myth!

Air-air heat pumps have gotten a lot better in the last 10-20 years and a modern one by a reputable brand will absolutely still work at Nordic winter temperatures. People buy those too, but not generally to heat entire houses.

Air-water is not popular here, I know they exist on the market and some people probably have them but I don't think I've ever seen one.

My air-air heat pump works down to -13 fahrenheit. Its less efficient but still more than efficient enough to heat my house, and more economic than anything else when its below freezing.

Geothermal is very different than air. In my state that hole will run you about 20k alone, if they let you drill it the first place. I have an air-air heat pump and i hate it, mostly due to air flow issues, but they really don’t work well in low temps.

It runs you $20k here too (well, the hole plus the pump plus installation is about 20k) but still pretty much everyone does it, because it's so dramatically more efficient that you're going to make that money back before the pump is toast -- and when it is, it's much cheaper to upgrade to a new one because you already have the hole ;)

Yes they do. I have no idea what heat pumps you have, but air-air heat pumps can be found in every other house here in Norway.

My main unit (live in the northern part of the central US) can operate below -10F with no problem heating my house. It is a dual fuel system (natural gas + HP) and, normal operations I don't, normally, run the pump at those temps. I generally lock the HP out around 10F and let the system choose the heat source based on the conditions.

Air source heat pumps are very viable in the northern climates of the US.

In the '00s, HVAC people in my area of the US were recommending against heat pumps because they'd not be able to run in heat-pump mode enough to make it worth it, and they were more complex, so had more ways to fail. And more expensive up-front, since you'd still have to have a furnace anyway and they were pricier than a plain central AC unit of similar size. Ones I spoke to said the new houses installing them and touting the energy efficiency to buyers were all under-sizing and those folks were going to have really crappy heating and cooling until they eventually replace the units (but, to be fair, new construction also tends to under-size furnaces and regular AC).

From what I could tell, they were mostly right about all that. You'd be lucky to break even on overall costs if you went with a heat pump in that decade. The ones that weren't crazy-expensive only worked down to (IIRC) 20F or something like that, and we just had too many days under that in a year for it to make much sense.

Newer ones are better, but I bet "heat pumps are a scam" remains an attitude in many circles, because at one point they kinda were. At least in our climate.

Other countries may have had bigger government incentives that offset the costs of higher-end heat pumps back then so the economics worked out better, I dunno, but we didn't.

[EDIT] Air-air, that is. Practically nobody does geothermal heat pumps here. The up-front cost is waaaaaay higher than any kind of normal (for us) set-up, not many people do it so that makes the cost even higher because the installation market is tiny, and energy costs aren't high enough to make the extra tens of thousands of dollars a good investment.

> Practically nobody does geothermal heat pumps here. The up-front cost is waaaaaay higher than any kind of normal (for us) set-up, not many people do it so that makes the cost even higher because the installation market is tiny, and energy costs aren't high enough to make the extra tens of thousands of dollars a good investment.

This isn't true in all parts of the US. A family member of mine has had an open loop geo-thermal system for almost a decade now, and they're definitely not the only one in the area.

I'd say about half of my neighborhood has air source heat pumps and about a quarter of the remaining have geo-thermal (vertical loops).

There are two HVAC businesses in my area that specialize in nothing but ground source heat pumps. If there was no market, they'd be out of business.

I think, where I am, it's harder to find a cognizant air source installer. Most of the HVAC businesses here are still under the misconception that air source isn't viable in our northern locale and often don't try to sell, or try to talk people out of them. But my air source is able to heat our house easily at 0F and is rated to run below -10F. It's currently my main heat source. Over the course of winter I'd guesstimate 60-70% heat pump heating overall.

Finland runs ground source water to water (geothermal) heat pumps on district heating carrying lower temperature waste heat from industrial processes. If one does that they don't need a borehole, but it needs the waste heat infrastructure and it only works where the population density is higher (apartment and office buildings). Also if one has a lake or high enough groundwater, it's easier since one only needs to put the outside pipes in the lake or pump water, or dig shallower trenches or boreholes. So the choice mostly depends on your location and what energy resources you have around you.

Historically electrically powered convection radiators were very common in northern Sweden and plenty of one-family houses still have them. Perhaps their relative inefficiency was good enough in earlier times when power in northern Sweden was very cheap. Nowadays it wouldn’t make much sense to install them. Both because electricity is more expensive and there are more efficient options.

"...geothermal heat pumps"

There's your answer, isnt it? With geothermal it doesn't really matter how cold it gets outside. But you have to dig a really deep hole, fill it with pipes and refrigerant, and offer sacrifices to Thor so that the pipes never leak.

The "myth" in the US refers to air-air heat pumps which do have a significant degradation is performance as it gets colder.

Not necessarily. There’s also something called air-to-water heatpumps which work like a mix of geothermal and air-to-air.

My house (~1900sqf/200m2) is currently warmed up by a 10kW Nibe 2025 that’s connected to a 300L buffer tank. It does it by its own down to -5C, after which a 3kW resistance heater in the tank adds the missing effect. During the winter it uses ~1000-1500kWh for both heat and warm water.

We don’t have those in the US either.

I have a heat pump that used 200 kWh of electricity in a single day! Granted, it was the coldest day of the winter.

Is your house especially insulated to get that low?

I think that in Europe traditional building materials - like bricks are natural insulators and our modern glass panels are 3 layers by default. So it may come as normal house in Europe is especially insulated in US.

200 kwh is quite a lot ... but also in US it looks like that the houses are bigger.

Not OP, but 200 kWh sounds like A LOT.

I have a 7 kWh Vaillant Heat Pump in Denmark, and the highest consumption i saw last winter was just over 60 kWh per day, which was with -16C.

My total consumption for a year with heat and hot water was around 3600 kWh ( 2200 ft2, 4 people).

COP was around 3.7 on average for the entire year, so ~13,500 kWh of heat produced by the heat pump. For comparison, one m3 of natural gas is around 10.5 kWh of heat, so we produced the equivalent of 1,285 m3 of gas.

a lot of heat pumps in usa are air to air and have an electric heater in the heat pump to generate the heate when temperatures get so cold that air to air is not effective. without a geothermal system to extract the heat from, they get very expensive to run in very cold Temps when compared to gas hearing. Hensel why the myth that they don't work. no one thinks they don't work, they just think it's a worse solution than natural gas.

My Air to water heat pump also has a backup resistive heater built in, which kicks in around 0C. It doesn't fully switch to it until -15C, but uses it for assistive heating for things that cannot wait, like hot water.

Still, i use way less energy than i did with natural gas. The original estimate from the installer was that the backup heater would consume around 100 kWh during normal year, and considering my estimated "heating needs" of 13,000 kWh per hear, that's just a tiny fragment of that. Resistive heat is (almost) a 1:1 conversion of energy from electricity to heat, and the heat pump delivers a COP value of between 3.5 and 4, meaning i spend around 3500 kWh on the rest.

For comparison, 1m3 of natural gas contains roughly 10.5 kWh of heat energy when burned in a gas boiler, so to deliver 13,000 kWh of heat, i would have to burn 1250m3 of gas.

Where i live, in Scandinavia, electricity is around $0.13/kWh and natural gas is around $2.2/m3 (both including taxes), so in total it works out to :

- Gas : 1,250* 2.2 = $2,750 / year

- Heat pump : 3,500 * 0.13 = $455 / year

Gas is about 6 times as expensive as a heat pump here.

That'd be €100 of electricity right there - in one day?? Damn

We pay 12 cent/kWh in France. Still a lot.

> But you have to dig a really deep hole

Well, you can also dig a relatively shallow long trench, if you have space for it.

Different temperatures.

Sure, but the max frost-line even in Anchorage, Alaska is like 10-feet down. A 5-foot deep hole is gonna get you temperatures that a heat pump can function at 365 days a year in all of the continental United States, and most of Canada and Northern Europe.

It can go nearly 10 ft in some places in Minnesota, for comparison ...

There are mini oil-rig-style drilling machines that can drill a 1000 foot deep hole and put a pipe in it all in a few hours.

The whole kit fits in a truck and can be operated by 2 people. Makes it pretty cheap.

The pipes that run into the ground are not filled with refrigerant, but with a water and antifreeze mixture.

This is borderline a tangent, but I think manufactures should be required to provide data regarding efficiency at specific temperatures. We moved into a house with a heat pump almost 2 years ago. For the most part it is great, but I noticed that in cold(ish) temps it can't keep up. I'm not talking single digits temperatures.

I called the manufacturer and asked for data about my model. They refused to give it to me because I wasn't an authorized installer.

It took me a long time and many sensors and calculations later but I have learned that below 36F it can't pull heat fast enough to raise the temperature of my house. The house doesn't get colder - that is the "break even" point. To be fair, it was installed in 2010 and I'm sure it was good for its time.

Why can't they just provide a spec that says how much heat/kw can be created at given temps? Even in steps like 10, 20, 30, 40 etc... I think that would go a long way to debunk the myth

Sounds like it was undersized for your house (unless your doing something crazy like trying to hold 80+). FWIW I was able to get efficiency charts and such at various temperatures from my salesperson, but I gather most don't bother. He was pretty surprised when I asked about them and more surprised when I took the time to (kind of) understand them, which was a huge pain in the ass. I certainly understand why most people would pay attention to the SEER values and just let the salesperson size the thing.

> Sounds like it was undersized for your house (unless your doing something crazy like trying to hold 80+).

It was set to 69 and held steady at 67 running for 8.5 hours. It wasn't until the temperature outside broke 36 that the inside temp slowly started to climb. It is a 3ton system for a 2500sqft home. From what I have read it should be enough. I think the problem is that it is an older heat pump and just doesn't perform well below 35.

> FWIW I was able to get efficiency charts and such at various temperatures from my salesperson, but I gather most don't bother.

I think this is my point? This was here when I moved it. I shouldn't have to go to a salesperson. I should be able to look it up in the specs or even the manual. At a minimum the manufacturer should provide it when asked.

> It was set to 69 and held steady at 67 running for 8.5 hours. It wasn't until the temperature outside broke 36 that the inside temp slowly started to climb.

This may be normal. Standard practice, per ASHRAE, is to design for 99% of days, and for the remaining 1% (~4 days per year) the system may have to run 24/7 (either for heating or cooling):

* https://www.energyvanguard.com/blog/we-are-the-99-design-tem...

* https://www.ba-inc.com/hvac-systems-what-is-a-design-day/

This allows for equipment that is "right-sized", as issues can arise (including premature death of equipment from short-cycling) if they are oversized:

* https://www.energyvanguard.com/blog/why-an-oversized-air-con...

* https://www.cooltoday.com/blog/3-problems-caused-by-an-overs...

* https://bryantlincoln.com/understanding-the-risks-of-oversiz...

By designing for 99% of the time, your gear should be just enough to handle things.

I live in the pacific northwest. Winter are low 30s at night for 3 months out of the year :( Granted, highs are mid-high 40s so it works fine during the day 90% of the time

30F is hardly anything.

Mitsubishis can work down to -13F/-25C and still have a COP of 2.08:

* https://www.mitsubishicomfort.com/articles/keep-warm-this-wi...

* https://mylinkdrive.com/viewPdf?srcUrl=http://enter.mehvac.c...

Heat pumps that look like standard ACs often can run at 100% capacity down to 5F and still have a COP>2 (1=resistive heating).

Also worth perhaps looking into how leaky your house is:

* https://www.energy.gov/energysaver/blower-door-tests

and then after that maybe insulation.

If the unit is sized correctly for the appropriate climate, then its broken. There's no reason the system cant deal with 0C, the COP should still be well above 1.

Get the refrigerant level checked by someone that didn't install the system.

For what its worth this is why I don't like to rely on heat pumps. Furnaces are really primitive, simple devices: little more than a heat exchanger and a blower. A heat pump has a hermetically sealed volatile liquid undergoing phase changes by a compressor with seals.

They're cool. They're efficient. They're great.., but they're also far more complicated and we're living in an era of cheaply built consumer goods

Think you should consider changing out your heatpump. My heatpump is said to deliver 3800 W at 5 fahrehheit.

I'm considering it. I have an energy audit scheduled next week. Depending on what they find I may consider buying something new. This one is 13 years old and I'm sure they've improved a lot since it was installed.

Heat pumps are over a century old; 13 years ago had great heat pumps. They didn't have fancy motor drives ("inverter technology"), but they probably had better refrigerants (we keep banning them).

You system is old, though, and the coolant might have leaked. Or it was never properly installed to begin with.

The COP numbers for Mitsubishi and Fujitsu are all online and readily accessible. If you're not using one of those brands, I suggest doing so.

It is a Trane. Sadly switching isn't like distro hopping :/

> This is borderline a tangent, but I think manufactures should be required to provide data regarding efficiency at specific temperatures.

These are available, and there are standards for EnergyStar approval and for being able to call yourself a "cold climate" air source heat pump (ccASHP):

* https://ashp.neep.org/#!/

* https://neep.org/heating-electrification/ccashp-specificatio...

Mitsubishi data sheet with COPs at various temperature:

* https://mylinkdrive.com/viewPdf?srcUrl=http://enter.mehvac.c...

Mine isn't listed. My takeaway is that it isn't rated for cold climates. For what it is worth, I know the HSPF rating is 8.8 but I don't know at what temperature that is calculated at.

The article lumps a lot of things together and lacks nuance (I have two heatpumps).

Wall mounted heat pumps are definitely not great to heat your house because of their placement higher on the wall means that your head is getting blasted with hot air while the air at floor level remains cooler. As a retrofit for a house with electrical heating it's just not as comfortable.

As central heating in a newer construction or as a replacement for an older gas furnace it is definitely awesome as it reuses the existing ducting.

Finally the "heatpump doesn't work in the cold" is definitly mostly a myth. There are several models that are spec'ed to go all the way down to -25C.

I bought property thinking I could retrofit a heat pump air handler onto the existing ducting. Turns out it's all undersized for heat pumps - the air handlers don't heat air as hot as a furnace does, so you need more volume to make up the difference.

Thanks for your insight on the wall units- my installer recommended wall units for the bedrooms for some reason (individual control) but I think I'd rather keep everything ducted, and I neglected to remember the difference having a ducted return would make in comfort.

What’s your plan? I’m likely to be facing the same situation soon, so any insight would be valuable.

Reno'ing a 2 unit house in NE USA, 3300 sqft~

Gut rehab, all new MEP and interiors except floors.

Going all-electric. HVAC to be 3x condensers (1 per floor). One air handler in basement, two in attic space, with the entire system being ducted, or if somehow my installer convinces me, all ducted with exception of wall units in 3 of the bedrooms.

Going for state and federal tax credits, my state has list of approved equipment. I'm going with Mitsubishi hyper-heat for the low temp performance. Even though it should be fine even w 0F winters they quoted resistive backup heaters for the air handlers. State is going to get more strict about what equipment meets their criteria in 2024, I'm getting in under the gun.

Why all electric? Your plan seems big enough that you'll have multiple outdoor units.

Why not have a gas furnace instead of two heat pumps? If the power goes out for an extended period of time (not improbable in the NE) a furnace will happily run off of a cheap generator.

Im not hating on HP, I installed one in my old house (now rental) and will probably add one to my current house. But Im to freaked out of burst pipes to drop a furnace.

Subsidies. So many subsidies. IRA, state. And I get A/C for free, and get to ditch the insane duct system that was built into the house and reclaim almost the entire basement.

I'm in metro Boston which is reliable as far as power goes. Last big outages were 2003 and 1965? My friends who live here longer say the longest they've personally experienced was ~12 hours.

I am considering adding batteries to the house eventually, although I would need to figure out smart load management. Also consider that you can easily make your own electricity (I've got a good south facing roof) but not your own NatGas.

Good answer, but don't forget the 90s Quebec ice storms weren't far from where you are. A week or two w/out power would suck.

In Maine, ice storm of '98. Can confirm, did suck.

Also got to ice skate down the road, across the lawn, and, well, anywhere.

Interesting I had not heard that existing ducting could be incompatible with heat pumps.

For the wall units, one of mine has a better placement that the other so it's not all bad, but the one that is between my living room and dining room is a bit annoying (definitely 1st world problem I know) because my head is really at the right height to get too hot while the rest of my body is still cold.

In other words, wall units can still be an option but double beck the placement so that you don't end up constantly crossing the "stream" of hot air.

In my house they did some math on my ducts, asked me about cold spots in the house, increased the size of the ducting/vent to the bedroom and added one in the kitchen. Didn't add much to the installation cost but I imagine that depends on how the house is constructed.

heat pumps can easily handle your ducts. However, heat pumps designed to work with minisplits, which are non-ducted and do not use a return ... not so much.

It is non-trivial to get the sort of heat pump you would need to retain your ducted system and have it work well, because very few people are going down this path.

You can get units that look like registers, especially if you have attic access. Mitsubishi has them.

The only problem with them is that the installer has to do a good job with the condensate line (for when you run the AC)

I have thought of those as well. They'd have to be on the approved equipment list for my state to get those sweet rebates. They've got pretty routine service required as well yes? Not gonna lie, being able to touch just a couple of air handlers instead of deal with cleaning and changing filters for every room is a plus. Whether it works out I think depends on how well the rooms get balanced, and whether everyone can agree on what temperature their bedroom should be.

> Wall mounted heat pumps are definitely not great to heat your house because of their placement higher on the wall means that your head is getting blasted with hot air while the air at floor level remains cooler. As a retrofit for a house with electrical heating it's just not as comfortable.

I don't get this at all. Hot air rises and cold air falls so you have that problem with all heating and cooling. Maybe if you're talking about a very confined space, but I run my aircon continuously to get the temperature I want and it's generally not blowing on anyone. If it is I turn the fan down, which works just as well but is slower when starting off with a hot or cold room.

> you have that problem with all heating and cooling

This is probably regional, but in Canada it is common to have heatpumps/AC for cold and electrical heating/furnace for heat.

That means your cold is coming from the top of the wall (where the wall unit is) and your heat is coming from near the floor with baseboard heaters. That gives you a better mix of air and a better overall comfort. Furnace exits also tend to be embedded in the floor at the bottom of a wall/window.

Wall mounted heatpumps mess with that paradigm because the air, whether hot or cold comes in near the ceiling. That actually works great for cold, but whn heating the hot air simply won't come down. If you have a second floor it will instead heat up that second floor.

It's not the worst thing in the world, but you definitely have a large temperature gradient from the first floor floor to the second floor ceiling that is noticeable. Especially when going down the stairs. In my case it will be around 19-20C at the lowest point and 23C at the highest.

You generally want air mixing, so it doesn't necessary matter where the units are.

There's an entire procedure that HVAC+D contractors (should) follow when designing a new system (and possibly for retrofits/renos) on just selecting the vents at the end of your ducts for this:

* https://www.acca.org/standards/technical-manuals/manual-t

* https://www.nrel.gov/docs/fy12osti/53352.pdf

What is the D for?

The "+D" in HVAC+D is "dehumidification".

There are ducted units available:

* https://www.mitsubishicomfort.com/residential/products/intel...

I am finding this. The room heat has a distinct vertical gradient. So what is the retrofit that ameliorates this ?

Drawing cold air upward from the floor thru a duct and to the intake near the ceiling won't work passively. Putting a small fan in the duct also seems bogus.

Heat pumps can heat in cold temperatures, but at decreasing efficiency or COP, which is how much heat you pump from the outside divided by how much heat it took to run the compressor. The theoretical COP is infinite if $\Delta T = 0$, and you can use Carnot to calculate the upper limit otherwise.

Now, the whole point of having a heat pump is to have a COP (much) greater than 1 (an excellent furnace has a COP just below 1). Once the difference in ambientT-outsideT becomes large, you approach COP = 1. The heat pumps manufacture will report the actual COP vs $\Delta T$; that is including losses, vs theoretical.

Now, a heat pump runs (almost certainly) on main power, and the alternative (usually) on natural gas.

It's a simple matter, therefore, to calculate the CO2 and financial cross over point for a heat pump vs. a furnace. I did that calculation when I had one installed in my house (for ductless AC; heating is an added bonus). It's somewhere between about -5 or -10 C, if I recall correctly.

Sure, a heat pump can work at -25C but it'll cost more and emit more CO2. This is because, dumb Joule for dumb Joule, natural gas piped to my home is cheaper and cleaner than the same natural gas piped and burned in a thermal power station [1]

In my small corner of the world this means that the heat pump is usually better to run than my furnace. But I keep my furnace on anyway, basically set on idle because there is no way that a heat pump can compete with a furnace on reliability. I don't want burst water pipes.

In Europe, heat pumps make 100% sense. Even though Europe is significantly north of the US, the winters are mild with few stretches below 0. In contrast, in some parts of the midwest long stretches of -20 C are not impossible. A week at -20 is not unheard of. In this situation Id be loath to have to rely exclusively on heat pumps.

[1] actually coal in my case so I fudged it in favor of the HP. Coal is also the case for many of you as well because coal is typically used as marginal power during the winter since it's easy to store and gas pipelines' pressure drop in the winter due to home demand.

> It's a simple matter, therefore, to calculate the CO2 and financial cross over point for a heat pump vs. a furnace. I did that calculation when I had one installed in my house (for ductless AC; heating is an added bonus). It's somewhere between about -5 or -10 C, if I recall correctly.

How can I do this? I know my heat pump can only maintain (not raise) my house temp at 36F/2C. But I have yet to figure out at what temp it is cheaper to run my natural gas furnace. I suspect ~40F/5C but that is just a guess (see my comment on parent)

If you happen to have an ecobee thermostat, https://beestat.io/ is an amazing tool for figuring out how your system performs.

I don't :/ I do have a smart home setup but it is local-only. I think I am capturing most of the data I need.

I know when my HVAC is running. I know how much power it is using. I know the temperature of every room, I have a stat outside so I have realtime temp/humidity of my location...I just don't have a way to capture and record how much heat is being generated. I'm wondering if putting a temperature sensor directly on an air vent would let me capture it over time

Simple in the sense that high school chemistry/math is all you really need + definition of COP. Its a few steps.

1. You need to get your pump's COP vs T from the manufacturer. This is the hardest part.

2. Now find out what chemistry is used for your mains power (nat gas, coal, etc).

3. Assume (or get) the efficiency of your furnace as COP = 0.9

4. Calculate for each delta T how much input each heating Joule of energy takes both methods.

5. Calculate the cost of each Joule for each method at each T. If you are charged electricity based on consumption, use the marginal rate.

For CO2 emission:

6. Apply the appropriate generation/transmission loss for electricity.

6. Calculate the CO2 from chemistry.

The manual for mine just gives two COP values: 3.63 at 47F, and 2.44 at 17F. That puts the cutoff for me at 30F, which is absolutely as cold as it ever gets in the bay area, so I only have the heat pump.

This is a longer explanation I wrote up when I replaced it: https://www.reddit.com/r/bayarea/comments/10lxyso/replacing_...

What's up with HN and heat pumps? I've seen this multiple times over multiple years. Yes, heat pumps works when it's cold, is that really news? Why?

HN is weird when it comes to heat pumps.

Here is a fun one: You can increase the overall efficiency by using a heat pump on the exhaust air. That way heat from e.g. your PC can be used multiple times instead of just being thrown away. And yes, active ventilation is needed in well insulated houses as they are quite airtight.

In the US, the limiting factor in heat pump adoption isn't technical. It's cost.

$18-25K for a minisplit with 2 heads is the norm in Seattle. Likely cheaper in other locations, but not by much.

Apparently it only cost Oyvind $2,630 including installation? I'd do that tomorrow if that were the case. I'd love a heat pump with a gas backup when we got our week or two of < -30C.

Seriously. I'd be screaming from the rooftops for everyone to get one.

> The state now subsidizes heat pumps, which in 2022 were used in just three percent of homes, and sales are beginning to take off.

That's probably not the only reason for the difference, but it is probably one reason.

That's not a part of the reason. Air-air heat pumps, exhaust air heat pumps and air-water heat pumps is so common that they are excluded.

Only new and emerging tech on the market are subsidized. When adoption increases they remove it as the goal is to decrease cost of newer and less common methods.

Apparently it only cost Oyvind $2,630 including installation?

Beware that the article is not very clear and talks about "heat pumps" as a general concept when the term actually encompasses a bunch of different and very differently priced things. What Øyvind gets for $2,630 including installation is a wall-mounted air-to-air heat pump like this one:

https://www.polarpumpen.se/varmepumpar/luftvarmepump/mitsubi...

Something like this will be great for heating a single large room, a small house with an open floor plan, or a secondary space like a garage, but is not a system that will handle an entire normal-sized house. But yes, you can get it for $2,630 including installation and it is very efficient at what it does.

The HVAC industry in the US enjoys massive gross margins on this equipment. Prices are supported primarily by licensing laws.

You can't legally break into the business, but DIY on your own house is entirely possible: the equipment is available for 1/4 to 1/3 of the "installed" price and some of the more climate-friendly refrigerants do not require an EPA license to purchase.

Yes, the work involves electrical, light plumbing and welding, a specialized tool or two, attention to detail, and definitely physical labor, but for a lot of engineers that's just fun stuff to do on a weekend - really fun stuff when you save over $10K per installed system (at the 3-5 ton unit capacity level.)

Not just electrical, 240 V electrical and a possible 200 Amp service upgrade. The really fun "it'll kill you dead" electrical.

The 240V electrical isn’t as scary as it sounds; individually, the hots are no more “hot” relative to ground than a 120V hot (still demands respect!)

Certainly it’s not for everyone or even every engineer but I guess my point was that if you’ve already done most of these things yourself it’s not a huge leap to DIY your own HVAC install compared to the financial incentive, especially if your house has more than one system.

120V will kill you dead, so will 60V.

True.

so will 12 volts if conducted through your heart in the right place

Checking the prices here in Norway a Mitsubishi Uwano 8700 (Mitsubishi's most high-end model) costs 2700 USD before any government rebates. It guaranteed to work down to -30C, but has been tested down to -36C as well. At -25C it produce 6000W. The "SCOP" value of the Mitsubishi is 5.2, so over the heating season it should produce 5.2x as much heat as you otherwise would get for the same amount of electricity.

Most of the heatpumps available are about 1200USD cheaper, but only go down to -25C, which really is plenty where most people live (not too far from the coast). Installation typically costs around 600-900USD it seems.

Given our 25% sales tax and high wages it should be considerably cheaper in the US.

Ah, that's a single room heater, not a whole house, eh?

It has a max capacity of 8700W, if that is just enough to heat a single room you have to look into insulating house better.

Well, our whole house needs about 70,000 BTU, which is ~20,000 W or 10,000 W (assuming it's still 2x efficient) input at -35 C, but our basement and second floor won't care if the living room is the only area getting heat.

What makes you think it's a single room heater? It looks like a mini-split system to me.

A ducted air-source heat-pump shouldn't cost more than 10% on top of a regular AC system. In my case, the heat pump was the same price as the AC only system. Paid about $8k for a new 1.5ton Lennox condenser and furnace on a townhome. Only options are for electric heat on this property. Re-used existing ducting, line set, and power. Old condenser was toast and an r-12 system.

Dual fuel/hybrid systems will be pricier, but not needed in Georgia. If designed right, the aux heat will be more than enough to supplement.

Poorly sized installations in very poorly insulated homes will need more supplemental heat.

This. Heat pumps are just AC run backwards, and it doesn't take much to make an AC able to run in both directions. The only thing is if you need a lot more heating than you need cooling you may need a larger system. But because they're new a different, as well as in demand, you get price gouging on installation compared to an AC in a lot of places.

Subsidies are a real problem, the installers are jacking up the prices, instead of the real subsidies being passed to the consumer. Same thing happened with EV subsidies. You have Kia dealers marking up by exactly the amount the tax credit returns to you.

If you need a lot more heat than cooling, you can't just get a larger system, you have to also go from single-stage to 2/multi stage or inverter-based, so that you can scale down in the summer so you don't cool too quickly and not dehumidify enough.

How did you just decide every sales man in North America is limited to a 10% markup on heat pumps?

I've seen similar prohibitively expensive price quotes. While not identical, there are many <$1000 mini split heat pumps on Amazon (DIY installation of course).

Highly recommend DIY'ing it.

MrCool et. al are all within the range of someone willing that can handle electrical work on their home.

3 head minisplit (also in Seattle) shipped for ~4.6k

Jeez...I bought a hybrid gas + heat pump system a couple years ago for my house in Portland and it was $17K.

It's an overkill system, for sure, but the gas was useful when one of the pressure switches in the heat pump failed (covered under warranty!) so I had no heat when it was < 50 degrees outside. Just switched the thermostat to "emergency heat" to force the use of gas, and I stayed warm.

This article is trash. I can't believe it made it to the front page.

It includes so many factual errors that it cannot possibly have been generated by AI.

1) I would like to point out that Oslo does not have harsh winters. If you wish to publish an article regarding heat pumps and harsh winters travel a lot farther north. Get into the -30C to -60 territory.

Otherwise, you can say "Heat Pumps work in the southern parts of Norway where it gets cold in the winter".

I live in Oslo and my heatpump works well.

2) Actually, good parts of Oslo are heated by district heating, Such as my house.

Statnett thinks district heating in the northern most part may be more efficient than electricity for heating

And some is already up and running. https://www.statkraftvarme.no/om-statkraftvarme/fjernvarmean...

3) It is highly common in the North to use good old fashioned wood for heating which is a rich source of pollution. https://eng.mst.dk/industry/air/air-polluton-from-stoves

4) Given the photo of the installation and the brochure the person in the picture is pointing to this is probably geothermal heat pump not an air to air heat pump֫ Those are far more energy efficient than air to air.

> 1) I would like to point out that Oslo does not have harsh winters. If you wish to publish an article regarding heat pumps and harsh winters travel a lot farther north. Get into the -30C to -60 territory.

What is the proportion of population that lives in IECC Zone 7 and above (or equivalent in each country):

* https://basc.pnnl.gov/images/iecc-climate-zone-map

There are folks in Climate Zone 7 in the northern tip of Maine (Caribou) that use heat pumps:

* https://www.youtube.com/watch?v=OcwIz6heDss&t=43m21s

Folks running stuff in Alaska:

* https://cchrc.org/air-source-heat-pumps/

The design standards are that your system should be able to handle things for 99% of the days (on average), so during maybe up to 4 days your system would be running 100% of the time and still keeping up. The rest of time it's not running as hard. Look up ASHRAE's "design days":

* https://www.airequipmentcompany.com/2021/what-does-design-da...

* https://www.ashrae.org/technical-resources/bookstore/weather...

Are there some portions of the population where a ccASHP won't make sense? Sure: but what population percentage, especially if you invest in some good air sealing and insulation first.

I did not question the assumption that heat pumps work in cold weather.

I questioned the methodology / logic used to "prove" this point in the article.

Works in Oslo = Works in harsh winters is false. Works during harsh winters in Siberia = Works in harsh winters is true.

I did not realize this was a myth, but to the extent it is I'm glad it is being addressed.

Everyone I know (mostly US midwest) who has one can't say enough good things about them. The main barrier to more widespread use seems to be upfront costs for a a new installation, not concern for efficacy.

I mean, kind of. Mine is rated for -20C and despite it being cheaper than resistive[1], it definitely needs a lot of help both to cool baseboards and to deal with temps even at -15C. It still does provide some heat as low as -25C, but not enough to keep the house at +20C. It also doesn’t qualify as a primary heating system where I live due to this — it can’t deal with the week and a half we sit at -30 lows, so you still need a rather heavy resistive heating load.

[1] due to prices set by the government, electricity is cheaper than gas for heating in Quebec, BTU per BTU, before any equipment or fees come into play. 6.5c per kwh for the first 40kwh per day is unbeatable.

This is my experience as well. At around -15C my heat pump will struggle to keep the place as warm as we would like. But, for those rare occasions I still have the gas furnace. Overall, the heat pump has been a money saver.

One issue is the "spark gap" in prices, which occurs when taxes and policies are misaligned.

It's purely a policy thing but you can end up in silly situations where heat pumps and EVs, which directly replace fossil fuels are getting taxed as if every unit of electricity they use was increasing carbon.

Ramping carbon prices and returning the money to people so they have a financial incentive to do the right thing is the pure solution, but subsidies achieve similar effects, and seem to be politically easier.

Some detail on the UK spark gap:

https://www.linkedin.com/pulse/price-cap-spark-gap-richard-w...

I want to see heat pumps with multiple heat sources and smart switching between them.

For example, a house might have a ground well, a panel on the roof (collecting sun-heat), a thermal storage tank in the basement, an air heat exchanger and a pipe wrapped around the sewage outlet (sewage is warm).

The system would then auto select the best place to get heat from to minimise operating costs.

This allows time-shifting of heat demand (transfer roof heat to the storage tank in daytime, then into the house at night).

It also allows generating electricity when moving heat in the same direction as a heat gradient (eg. Heating the house with water from the storage tank)

Anyone using a geothermal heat pump with radiant baseboard heat? I have an existing oil boiler that I would love to upgrade to geothermal, but I don't want to install air ducts. I know some systems exist for this (like WaterFurnace [0]), but I have some concerns about whether it can get hot enough for this kind of system. Haven't spoken to the company who would actually install it, but I'd love to hear any success or failure stories.

[0] https://www.waterfurnace.com/residential/products/geothermal...

So many people see old heat pumps struggle and make the assumption that all heat pumps struggle but recent model heat pumps can definitely operate will in sub-freezing temperatures.

Gas Furnace went out in September so I thought I'd get bids on a heat pump to replace it. All 3 places said I was going to have to have a backup heating system for when the temps go under 35F (!). I was kind of surprised by this. Apparently this is because it's a forced air furnace replacement (as opposed to a mini-split). Others told me that if I got a minisplit they'd be good down to around 0F. But since I've already got all that ducting for the air why would I get a minisplit?

Minisplits are more efficient than ducting because of energy loss from coils/burners to conditioned space (where as minisplit cartridges are typically directly in the room to be conditioned). In renos I've done with minisplits, you seal off the ducting and leave it in place (or if convenient and accessible, remove it, depending on objectives).

I have the impression that going mini splits would be a lot more expensive than just a drop in replacement for the NG furnace because you've got coils&fans in every room and the tubing needed for all of them.

Just depends on the property, hard to generalize. Running the math both ways is straightforward prior to implementation.

My mini split works to -13F but at 85% of its normal efficiency once it drops below zero. A lot of Canada is on heat pumps and definitely deal with below zero

But very few Canadian homes are heat pump only.

Modern heat pumps are great in spaces that are insulated well. Even when very cold outside they can get enough into the space to keep it warm. In spaces with poor insulation then a furnace or boiler is better as they can blast a space full of heat quickly.

In my very old home I have areas that are not insulated and a furnace heats them. I have had renovations and those areas are well insulated and a heat pump does great.

It's not a myth. The heat pump at my house works great, but it shuts down around 12F (-11C). We don't often hit that temperature in a MA winter these days, but it does happen. Oil burning for backup.

How old is the heat pump and is it a central air heat pump or a minisplit?

Air sourced heat pump?

Mitsubishis can work down to -13F/-25C:

* https://www.mitsubishicomfort.com/articles/keep-warm-this-wi...

* https://mylinkdrive.com/viewPdf?srcUrl=http://enter.mehvac.c...

What’s the COP at -13F

Per the second link PDF, under "Efficiency", 2.08.

The article only compares heat pumps with electric radiator heating though.

They're cheaper than the most expensive heating solution and that's it?

It's my understanding that, in ideal but realistic conditions, it's more efficient to burn fuel in a powerplant and use the generated electricity to run a heatpump, than it is to just burn the fuel for heat. Pretty neat!

More efficient but more expensive, depending on the rates you pay for electricity vs. natural gas.

Exactly, efficient for who?

If it's so much more efficient why is a local burner (natural gas, wood whatever) a couple times cheaper for the person living there?

In a cold climate like Norway, they are literally the cheapest source of heating full stop*, and it's not close.

* Except if you own woodland, chop your own firewood, and burn that for heat in a central furnace, which is not actually that uncommon in the rural Nordics

There's a bit of strawman here. The article is comparing air source vs ground source heat pumps without mentioning the major differences.

Yes, heat pumps work in the Nordics ahem with the additional CAPEX cost of installing ground loops. A ground source heat pump is NOT a drop-in replacement like an air source heat pump is.

Interestingly, Google spun out a company called Dandelion Energy which focuses on installing ground source / geothermal systems.

Good for Norway, but is not directly transferable to other countries. Norway has still one of the cheapest electricity in the world, because of hydroelectricity, oil to electricity and other sources. They heat with electricity since decades, where this for example was unthinkable in Germany. Yes, heat pumps are an improvement compared to radiators of course. Now, in Germany, we powered up our coal power plants again to generate electricity, feeding heat pumps. This is just ridiculous, because we have no hydro power like Norway has. And wind energy isn't enough.

Nuclear power is a great carbon-free option.

Nuclear power - at least when talking about nuclear fission, not fusion - is a great way of decarbonizing the grid, but I don't know about the long term. Neither I'm sure if renewables are a sustainable way to generate energy. Maybe a mix of both would bring the best benift if we manage to develop the power grid towards a decentralized and smart grid fueld by renewables and backed by nuclear (fusion) power.

> This is just ridiculous

Ridiculous compared to what? Burning fuel at every home?

I get the distaste of needing to power up coal plants, but if you want to look at CO2 output of a mixed grid with fossil fuels + renewables vs direct fuel consumption you'll find the CO2 output is going to be much lower on average for the heat pump + seasonal coal plant fire up.

That's because even if wind can't fully power all your heat pumps, it is powering a large chunk of them.

Even without renewables, burning gas for electricity and using that to power heat pumps can easily be more efficient than burning that gas for heating homes directly, in most climates with commonly available heat pumps. Coal is a bit dirtier, though.

> Now, in Germany, we powered up our coal power plants again to generate electricity, feeding heat pumps. This is just ridiculous, because we have no hydro power like Norway has.

So what. A heat pump makes 3-4 kWh of heat energy out of 1 kWh of electricity, and a coal plant, especially a combined district heat-power coal fire plant, has a far higher efficiency than any domestic gas/oil boiler can achieve.

> coal plant, especially a combined district heat-power coal fire plant, has a far higher efficiency than any domestic gas/oil boiler can achieve.

This is bullshit. Domestic boilers have a thermal efficiency of around 93%. The most modern coal plants have electrical efficiencies around <47% on 100% and around 43% on 50% output. Combined heat-power plants have a combined efficiency of around 90%. And the centralised versions have the transport loss.

Getting all the calculations right is really complicated.

Edit: the annual mean of the heat pump COP is not 3-4, either.

Norway has still one of the cheapest electricity in the world

Norway is actually pretty middle-of-the-road, in line with France and Greece for example.

https://ec.europa.eu/eurostat/statistics-explained/index.php...

You need to compare this with incomes and GDP! Norway is one of the richest countries in the world per person.

I can guarantee you that oil power plants is not making electricity cheaper in Norway. Mainly because there is not a single one in Norway...

I didn't talked about oil power plants. But 2022 Norway got 100TWh from oil, which is a total of 20% from the energy mix! 300TWh came from hydro power. Energy is in the wider sense, transferable. So it does not matter, if you have an oil power plant. https://ourworldindata.org/grapher/energy-consumption-by-sou...

How do you conclude that electricity prices are low in Norway because of oil, when oil is not used for electricity? That makes no sense.

Electricity prices in Norway is not low because of oil. Oil/gas is not used for heating, and prices are higher than other places. So if anything the electric prices are higher as oil products are more expensive in Norway compared to most other places.

Yes, look, if you have 20% oil in your energy mix, this is used somewhere. For example by ships, cars and also some industries. If you would not use this 20% for some energy, you would need to burn wood instead or maybe better use electricity. :) So burning oil for some extend keeps electricity prices overall lower.

If Google is telling me correctly the kwh price for electricity in Norway is around 12 Eurocents. This is very cheap! In Germany it is currently 30 Cents, last winter even 50 Cents. And Germans getting lower salaries..

Reply on news.ycombinator.com