Hacker News new | past | comments | ask | show | jobs | submit login
Heat Pumps Work Miracles (ucsd.edu)
251 points by mhb 55 days ago | hide | past | web | favorite | 243 comments

I build controls for many heatpumps (https://flair.co/pages/mini-splits-and-window-units) and follow the space pretty closely. They are truly magical with respect to pumping heat from the cold outdoors into a warm indoor space.

That said, the efficiency gains on the energy consumption side are admirable. From a global warming standpoint however, heat pumps are actually still pretty bad because of the refrigerants.(https://www.drawdown.org/solutions/materials/refrigerant-man...) This is not an issue with more traditional electric heaters (resistive or otherwise).

The real calculus for those concerned with global warming (I understand this was not the topic of the writeup) is the net effect of these systems over their lifetime - aka, Energy Saved and its GHG effect via leveraging a heat pump vs the GHG effect from the release of the refrigerant. Perhaps the real win would be a refrigerant recovery business that is...air tight :)

I think when I did the numbers for our heat pump, a catastrophic leak that would release all its refrigerant would result in a GHG contribution equivalent to a year of emissions from our (previous) natural gas heating. Can't find the figures now so do check before quoting me on it ;)

Alternatively, Mitsubishi has been selling a first-generation product that uses CO2 as the refrigerant (the QUHZ model). It posts a COP of 3 for producing domestic hot water (that's pretty good) - I'm not sure why they don't post the COP for room heating, maybe it's not great. But I'm curious what the second-generation product will do!

I've done some theoretical work in this area. The main problem is that most hydronic (water based) heating solutions return fairly 'hot' water after heating the space. In order to get the most energy out of your CO2 refrigerant gas, you need your other heat exchange fluid to be as cool as you can manage. A designer could choose to use a different style of radiator which would allow the water to get closer to room temp, but that is not the type of system currently being manufactured. This would also not work for in-floor heating, as you would have parts of your floor very close to room temperature, not what people expect (warm floors).

These problems really aren't specific to CO2 hydronic systems, and also appear with HFC refrigerants when heating water. CO2 is actually very good for domestic hot water production, where the hot water is not returned.

The problem is really that hydronic heating is a bad fit for heat pumps in general. You'd much rather heat air, and skip the secondary fluid loop.

I recently toured a municipal heat pump plant. It took cleaned sewage at IIRC 12C, dropped it to 6C or abouts, and the heat was pumped into the district heating network at about 90C.

The plant had 6(?) pumps, each producing 18 MW heat while pulling 6 MW electricity (and also some district cooling). Not CO2 though, each pump had 9000 kg R134a as the working fluid.

Interesting, sounds like a water to water heat pump (takes heat from sewage and pumps it into district heating). Just to dig into your example a bit: It kind of depends on what the return temperature is of the 90C heating fluid is. Based on the refrigerant and COP of 3 (18/6), its probably around 50C. If they were able to find a use for the process fluid so that it returned at 20C, they would probably have a COP closer to 5. You can still build 'relatively' efficient heat pumps for hydronic heating, but they would be significantly more efficient if we did a better job dumping all of the heat from the water (which would require a larger, more effective heat exchanger).

Edit: I guess what I'm trying to say is that most of the radiator technology we use (like the in room part of the heating system) was designed with combustion-based heating in mind. Combustion based heating sort-of works just as well if the return water is 50C or 20C, but you would get a substantial boost in heat pump hydronic heating performance if you were to redesign heat exchangers to get as much temperature out of the fluid as possible. This is an optimization problem: as you increase the size of the heat exchanger, it gets more expense, but improves performance. There is an optimum in there somewhere...

Here's a brochure from the heat pump manufacturer about the plant: https://www.friotherm.com/wp-content/uploads/2017/11/katri_v...

In general, the evolution in district heating networks is towards lower temperatures, allowing better efficiencies with things like heat pumps. As you say, integrating heat pumps into existing district heating networks designed for combustion based heating is suboptimal, but redoing the network for lower temperature operation is incredibly expensive as well.

Currently, in the case referenced above, during the winter the heat pumps are AFAIU the "first leg" when the cold water returns to the plants; the output of the heat pumps is then routed via the fossil fueled CHP plants for topping up (depending on how cold it is, it's heated up to IIRC ~120C) before it's sent out in the network again. The challenge is that due to climate change, the fossil fuel plants need to be replaced with something else (no, biofuels are not an environmentally sensible solution). What that something else is, is not entirely clear. Moar heat pumps + wind + heat storage? Or geothermal? Or nuclear?

Thanks for the link!

There is a such thing as a solid state heat pump which operates without refrigerants or any moving parts. They are not as efficient as regular heat pumps, though, at least currently. But maybe that could change.




You reminded me of Lonnie Johnson's "Johnson Thermo-Electrochemical Convertor", or JTEC for short. On their webpage they say they are aiming for 85% Carnot efficiency for converting heat to electricity, and it can run in reverse to become a heat pump[0][1].

Sadly, I can find no third-party measurements of how well it performs in practice, so I guess it's still in the R&D phase. I did find a nice talk by Johnson about the project though[2].

[0] http://www.johnsonems.com/about/

[1] http://www.johnsonems.com/benefits/

[2] https://www.youtube.com/watch?v=F6FUk2E9-68

Thermoelectric Coolers (TEC) or Peltier devices.

You can buy small ones and play with them for very little money. They're used all over the place.


Are there any that are powered by natural gas instead of electricity?

If I'm understanding your question correctly, yes. The term you want to search for is "Gas Absorption Heat Pump". Instead of compression, it uses heat (which can be from burning natural gas) to drive a gas (usually ammonia) out of solution. They can reach about a 1.5 COP: https://aceee.org/sites/default/files/files/pdf/conferences/....


So I found the GAHP-W LB but the price seems to be a state secret.

Also the Lochinvar Optimus, but again, no prices.

But thank you - I don't understand why these types of units are not standard, are there downsides I'm not aware of?

Aside from the low COP, the market is much smaller (utilitity gas availability).

Heat pumps use actual pumps to pressurize a refrigerant. Could you use some sort of natural-gas internal combustion motor instead of an electric motor? Sure, why not. You could use the waste heat from the combustion as a heat input, but the cost of the complexity and achievable efficiency are a big question.

Modern heat pumps are truly a marvel, especially when you get one designed for heating and not just an AC-unit running in reverse. They're very common in Sweden and can easily warm an entire house even at -20 C.

Whenever this topic appears on HN I want to get on my soapbox and yell "INSULATION PEOPLE, INSULATION!". Most houses I've seen outside of the Nordics have shit-all insulation and that's depsite it working both ways - house stays cooler in the summer. It's such a small cost to add during construction and it pays off in a year or two, tops.

Modern, newly built houses in Sweden, way above the arctic circle, require less than 5000 kWh per year to heat and generate hotwater.

> It's such a small cost to add during construction and it pays off in a year or two, tops.

Classic market failure: operational costs don't factor into home buying decisions prominently, so whatever small added construction cost is borne by the homebuilders (the cost of added insulation doesn't quite result in an equal increase of selling price).

Short of instituting a carbon tax that would make operational costs of heating suddenly more prominent and priced accordingly, this is a great case for efficiency standards in homebuilding. Suppose your home came with an "energy star" HVAC rating...

You can actually get your home Energy Star certified.


Some states have even offered tax credits for doing so.

Thanks for the link, didn't know about that!

They an energy label here in the Netherlands, its required for all sales.


I'd love to install a ground heat pump but the installation requires digging up our fairly newly made garden. We'll move in a few years and buy something which has an A label with a pump and panels.

It’s required to even get a building permit in Lithuania

In France, before buying or selling a house, you need to contract a company to do an Energy consumption diagnostic, which gives you a rating between A to F, F being the worst.

It also gives you an estimate of the amount Energy used to heat/cool your dwelling as well as the average yearly cost.

Obviously, this system is not perfect but its something.

Based on that you can work out if buying a cheaper and less insulated place is actually worth it or not.

You don't even need a tax, you just need to make this information visible. Like when you buy an electrical appliance and it comes with a label that tells you how much it consumes and costs per year. Give people better information and they will make better decisions.

You can build a LEED house the same way as any other building: https://freshome.com/2014/10/09/what-is-a-leed-certified-hom....

Yes, of course you can find anything if you're looking specifically for it. At HN we talk about scale, though, so how do we make everyone look for efficiency?

Tell them it's cheaper to live in.

Consumers are notably horrible, and for good reason (neuropsych/behavioral economics has studied this in depth), at making rational total cost of ownership calculations.

You can argue all you want that it SHOULD be enough to share lifetime costs, but it doesn’t work. Just look at people who still buy incandescent light bulbs because they are cheaper at the cash register, even though they will save a multiple of the cost of the bulb over its lifetime.

Consumers in the US seem to be swayed by monthly costs rather than total cost. You can get them to buy expensive things if they are financed and the monthly cost sounds reasonable. I wonder if you could say - well the house is $3000 more, but the monthly savings are XXX which is like getting your iPhone for free. It might work at the time of sale, but will be forgotten later when they have to actually pay their phone bill.

What do you mean exactly by "financed" in this statement? It feels like you're using it in a technical way, such as thinking of the purchase as having its cost spread over multiple time periods.

Financed means bought by the means of credit.

Over on today's $35k Tesla Model 3 announcement thread, there's a whole bunch of "Dark Pattern!!!" pitchforks being jostled about because Tesla threw fuel savings into their "effective cost" price tag (the MSRP is still $35k, but the subsidies and fuel savings bring the "effective cost" down to the mid/upper-20's).

We as consumers seem to ignore or discount future pain for lower prices now, which is why operating costs are ignored by average consumer if they mean higher upfront costs. And when marketers try to bring future savings into the present and get you thinking in terms of total cost of ownership, even the rational-types at HN cry "dark pattern!"

All of this is to say we haven't yet found how to make "it's cheaper to live in" an effective message.

Solar folks are getting there, though ("the system pays for itself in 5-7 years"), and with the rise of EV's, I think this reasoning will become more common.

I read through most of those earlier, and I'm not sure that you're being charitable in your interpretation of the conversation.

The "dark pattern" complaint is that actual realized savings (tax credits) are mixed with hypothetical savings (future cost of gasoline) and taken off of the sales price. Meanwhile, things like delivery fees, sales taxes and maintenance cost differences are not added back in.

It's one part standard marketing, one part underhanded marketing imho.

> And when marketers try to bring future savings into the present and get you thinking in terms of total cost of ownership,

That's because it is fundamentally inaccurate, and therefor counter-productive. I'm all for demonstrating TCO differences, but when you show a "price" of $26k and the actual price is over $40k with taxes and everything added back in, suddenly you get struck with a bit of cognative dissonance.

LEED is like an energy star rating for a home. Mine got a silver I think (low 50's). So it could be mandated, display the houses result alongside the listing perhaps.

Yes, insulation, but that's not without its problems either. If you're adding insulation to an old house you have to consider that the ventilation system in that house is probably gravity-based. What that means is hot air moves upward, through an unheated attic in the type of house I'm thinking of, and replacement air is pulled in through the breathable structures of the walls. When you insulate such a house heavily, that air intake is compromised and you lose ventilation. So for an insulated house you may need to add mechanical ventilation, which underpressurizes the house. You now need to add air vents, but if you don't have enough of them, replacement air may get drawn e.g. from the understructure of the house, pulling soil microbes into the air with it. Also, if you insulate in such a way that the breathability is compromised, you risk moisture problems, which lead to mold. In particular, adding modern fibreglass insulation to an old house is very problematic, because that type of insulation should not be allowed to draw moisture as it will not dry out and will grow moldy, and wood structures in contact with it may rot.

Basically, moisture control and insulation is a tough problem. In modern houses the solution is to have an airtight layer of plastic sheeting between the insulation and drywall, but those houses should also have been designed from the ground up to have an active, mechanical ventilation system.

Based on experiences here in Finland, a lot of such buildings built in the 60's and 70's have developed severe problems, and frankly even new projects seem to have issues all the time. The more you insulate, the more you have to take care that moisture does not gather in the insulation because you no longer have airflow that would dry it out.

> In modern houses the solution is to have an airtight layer of plastic sheeting between the insulation and drywall [...]

In a building that will be air conditioned, putting plastic sheeting between the insulation and drywall is a bad idea. The problem is that any moisture in the (comparatively warmer) insulation cavity will condense on the (comparatively cooler) plastic sheeting, wetting out the insulation and causing mold and rot. They also prevent the wall from drying to the inside.

It works fine in cold climates that are do not install air conditioning, as I assume is the case in Finland. In the majority of the US, even up to the Canadian border, many new houses have AC.


The vapor barrier should be placed on the 'warm side', i.e. the side where the temperature is higher (as you say). So yes which side that is (what side of the insulation) depends on whether you're mostly cooling or mostly warming your house.

There are moisture regulating sheets on the market that allow vapor to pass in summer. It is many times more expensive than a simple plastic foil, but still a drop in the construction budget bucket.

I've gone through the exercise of retrofitting insulation. Unfortunately there are a few things that are extremely hard to get past. 2x4 walls can only be insulated to R-13 or so unless you're replacing the siding. Subslab insulation can pretty much only be done when it's poured. Joist insulation & air sealing can be very hard to get to.

I can only hope 2x6 framing becomes more common, the subslab insulation has started to become code.

2x4 can be insulated much higher - around R26. You need to use expensive closed cell foam to do this. It is much cheaper to build with 2x6 which gets you to R19 with cheap insulation.

You can't easily retrofit spray foam though. Either the drywall or sheathing needs to come off, unless the cavity was empty, and even then it doesn't fill evenly if sprayed in a single hole.

2x6 framing should be nearly universal on new construction, which calls for R19 according to the international building code.

Ahh, so this is why building my 3 season shack is damn near impossible financially.

> I've gone through the exercise of retrofitting insulation.

Based on your experiences what do you recommend? (or what sources do you recommend consulting to learn more about retrofitting older buildings?)

finehomebuilding, familyhandyman, and greenbuildingadvisor are all valuable resources with a different bent to each.

Counter-intuitively, air sealing is usually the #1 improvement. A good audit with thermal scans is worthwhile, especially if your muni offers a subsidized audit.

Tough to go into more detail without a little more information.

> Counter-intuitively, air sealing is usually the #1 improvement.

I think the hard thing for people to think about is that heat transfer is exponential. But we don't measure it that way. In fact R values are really misleading[1].

Meaning transfer coefficients can vary by orders of magnitude. So it's easy for one source like air leakage to totally swamp every other source.

[1] Figure of merit is really 1/R (smaller is better) not R.

Thanks for the recommendations.

> Tough to go into more detail without a little more information.

Have a ~1800 square foot single family Colonial located in Northeast USA built around 1910. It was partially renovated ~2000 but most of it's still uninsulated (afaict, still very new to this!).

It seems residential natural gas prices in Sweden are almost 4 times what they are in the US! Show what a carbon tax can do in changing economic decision making.


Homes are heated with communal hot water, wood gassification (pellet burners), heat exchange with the ground or deeper bore-holes, or electricity. Not much gas, heating oil was gone by the 80's sometime.

This might be because no one in Sweden knows what “residential natural gas” is. I haven’t heard about anyone heating their house with natural gas. Residential gas is used for gas stoves in some old (exclusive) areas of stockholm. Oil used to be a thing but these days it’s usually electric (heat pump) or geothermal.

This demonstrates the danger of assumptions. As others have commented, "residential natural gas" simply isn't a thing in the Nordics.

And how fearful people are of depending on Russia.

Air sealing is also critical. The insulation can be defeated if unconditioned air can leak into the house and conditioned air can leak out. It's a difficult thing to get right because it requires attention to detail and if done poorly it can result in moisture problems.

Are there no health issues associated with a house being that well sealed, or do you have some sort of air exchanger that is temperature efficient?

Edit: e.g. googling brought up this article - https://www.architectsjournal.co.uk/buildings/airtightness-b...

You insulate and air seal properly, then make sure that you bring in outside air in a controlled fashion (enough CFM, filtered, etc):

* https://www.youtube.com/watch?v=CIcrXut_EFA

I wonder what carbon dioxide levels get up to in a well-sealed house.

You’ll want mechanical ventilation (with heat recovery) to prevent that: https://en.wikipedia.org/wiki/Heat_recovery_ventilation

Heat recovery + well sealed house with some humans inside can potentially mean no heating at all!

Some good comments on that in this HN thread from a few weeks back. In short, not good.


Thanks for the link.

shouldn't houseplants be sufficient?

For anyone interested, Building Sciences Corp is a great resource on different methods to build functional, airtight, well insulated walls. Here's a starting point:


I'll add https://greenbuildingadvisor.com for a site that has a lot of quality discussion and analysis on energy efficiency for buildings. The forum and comments can teach you a lot.

I wonder how feasible maintaining that is? I feel like it's very common to find people who work on your house will mess this up (cable or satellite dish installers, for example). That seems to happen for everything, but it depends on the resiliency. I bought a house where satellite dish installers drilled right through the shingles, likely compromising the water permeability of the roof. Ironically, I had heat pumps installed last year. Soon after I had rodents. I feel like they did a good job of weatherproofing exterior holes they made, but after patching up the crawlspaces the rodents disappeared.

People are grossed out by rodents and a leaky roof has to be addressed, but if their heating/cooling cost goes up slightly because a cable installer messed up are they going to hire someone to hook up a blower door, patch things up, and charge this back to the cable company?

It's hard, but the BTU loss from a 10'x 10' uninsulated wall is equivalent to a 1"x 1" hole in an R-31 wall.

There really is just a severe lack of building science education in American construction, because the margins are so thin and it is still dominated by blue collar "man in a van" companies. So you have boneheaded contractors drilling holes into air barriers because that's how it's always been done.

Interesting thing is a lot of sound deadening practices also apply to insulation. I don’t know if this statistic is accurate in the audio case but it’s not far off.

Heat conduction through studs in an insulated wall also carries sound. A 2x6 plate wall with 2x4 studs on alternating faces (think slalom) helps with both.

I just had a minisplit head installed at my house, and when they used a hole-saw on my external wall, it basically trashed the fiberglass insulation in that stud cavity.

Fortunately this was part of a project to finish my home office, so I still had access to the space between the internal and external walls. Not only could I see the insulation damage, I could also repair it without too much effort.

But in general, I agree there's a responsibility gap when it comes to these installers. I think the best practical solution for now is to factor into such projects the cost of restoring a building's thermal envelope.

EDIT: I should mention that a blower-door test might not have detected this damage. The cavity whose insulation was damaged has pretty good draft-blockage from the rest of the house, due to fire-code requirements.

> But in general, I agree there's a responsibility gap when it comes to these installers. I think the best practical solution for now is to factor into such projects the cost of restoring a building's thermal envelope.

What, $5 for a tube of exterior grade caulk? Vapor barrier seal tape is 10 cents a foot. Add in $10 in labor per penetration and you're done.

You could leave it out of your estimate entirely and not even notice if your guys did the work.

edit: I'm a commericial construction estimator

Maybe I used the term "thermal envelope" too loosely?

I agree that restoring an air seal is trivial on a per-hole basis. My main point was restoring the lost insulation performance that comes from trashing one stud cavity of fiberglass batt insulation when the hole is made.

Based on my limited knowledge, I don't know what impact that might have on the overall structure's thermal efficiency. I would also worry about risk to any water-carrying pipes running through such a section in a cold climate.

If the drill snagged the fiberglass and spun and compressed it creating a large void, it's certainly going to be much more than caulk and tape to repair and restore.

Right. In my case the strip of fiberglass batt was basically flopped over in that section. (It's a double-framed wall, so there was space for the batt to fall away from the outer wall.)

That high efficiency envelopes are brittle is something that concerns me. Cold air and water needs to be kept out, but more importantly water needs to get out or you get major damage that will completely negate any energy savings.

In the Midwest there is an utter catastrophe where modern high efficiency houses are rotting away because construction and design errors let water in or water to condense with no way out.

It is indeed critical and all new houses in Sweden are tested for leaks under pressure. Those that leak sufficiently little get to call themselves Passivhus and you have some taxation benefits.

A handful of US states have also made blower door testing mandatory in new homes, but we still have a long way to go. A few states don't even have state-wide building codes, much less efficiency requirements.

How do they test that?

At least in the US, blower door testing:


I've had this done before, very, very informative. Those machines are like $3K to buy, you can hire someone to come out and do it, but this seems like a perfect DIY project.

An other very informative thing to get done is thermal imaging of the walls (from inside and outside both). Very good at uncovering damp spots and water infiltration too.

UK is finally getting there, but we have a very long way to go. Our housing is mostly bloody awful for insulation, draught proofing, and sensible forms of heating. The very worst is anything 60s-90s, particularly if there's a flat roof.

We've only been hearing about how well Sweden and other Nordics do insulation, community heating, heat pumps etc for about 40 or 50 years. Various combined heat and power, insulation, passive houses etc would turn up on things like Tomorrow's World (60s-90s technology TV series) fairly often, like it was some sort of breakthrough new discovery. Well it was, to the UK, I guess.

There's a joke in Sweden:

"Hey, do you know why the English put their pipes on the outside of their houses?"


"It's so they can reach them easier to thaw them when they freeze!"

And then everyone laughs.

...and then you learn that it's not a joke, and then you cry.

The UK's deeply dysfunctional planning/housing system contributes to that. Land is expensive, building your own home on purchased land quite rare due to bureaucracy of planning permission.

I live in a flat that was built mid 90's and it's a disaster from a heating and cooling point of view.

Ventiliation is shit, one side never gets the sun so the outer wall is cold which means condensation is a daily battle, I basically run a big heavy duty dehumidifier from mid-autumn to mid-spring (at least its not a net loss, it dumbs a fair amount of heat as a byproduct so I can run the heaters less).

It's rented though so not something I can fix until I buy a property in a few years.

Insulation is probably the most important mechanical energy saving feature of a house. Unfortunately it is behind the drywall and as they say "out of sight and out of mind". We built a very large cottage (3000 sq/ft) in Northern Ontario, Canada. The entire house had a continuous envelope of a minimum of 4-5 inches of Closed Cell Foam (2 lb) ( https://www.demilec.com/ ) . Installed a Geo-Thermal Unit with in floor heating through out the basement and forced air everywhere else. We also have a high efficiency fireplace stove. It can be minus 40 C outside and you only need a couple of logs in the fireplace to keep the entire cottage very warm.

You can change out your furnace/AC/Heat Pump anytime, but try tearing off all the drywall in your entire house to replace the fiberglass insulation that doesn't work or was installed incorrectly. Spray Foam Insulation pays for itself and to think it's only around $3.00 a sq/ft at 3 " thick installed. People have no problem paying more for carpeting or tile floors, strange....

I have a 2 story house with a heat pump for the upstairs...and I hate it. It causes so many problems that we just keep the heat very low up there so that the heat drift from the 1st floor furnace does most of the work.

When it runs, you can never set it higher than 2 degrees F above the current temperature and if you do that when it's even close to freezing outside the unit will build up a think layer of ice around the casing...which just makes it spin constantly until we turn it off to go manually de-ice it. It's the most hated piece of equipment in my entire home.

Do you have a reference for one of these more effective modern heat pumps? I've been sincerely planning to install geothermal when this thing dies.

These are the two biggest complaints about heat pumps: one, they blow out air that is only marginally hotter than the room air (unlike a furnace, which pumps out hot air, or a scalding radiator), and two: heat pumps take a very long time to warm up a cold space. If you try to ramp up rapidly, you’ll often trigger the “emergency heat” which is usually electric strip heat and thus VERY inefficient and expensive.

Heat pumps work best when you maintain a room at a stable, comfortable temperature.

You don't want to use a heat pump to blow hot air. You want to use a heat pump for low temperature underfloor heating. The heat is right where you want it, the heat pump doesn't have to work as hard (because of the lower temperature). And everyone is happy.

"Heat pumps work best when you maintain a room at a stable, comfortable temperature."

Right, but for this to be economical, you need to have a well-insulated house, which is pretty much impossible to retrofit. So practically speaking, only newly build houses (and then, only those to the highest energy standards) should use heat pumps for heating. If we go around promoting them for other purposes, they will build up a reputation for being a crappy solution and people will shy away from them. That's my main problem with people promoting them - they often promote them for retrofits.

Groundwater systems are much better. My parents house in Ottawa has one, and it works fine through -30 C nights. There's a source well about 30' deep with a pump at the bottom, and a return well some distance away. That provides a constant supply of water around 5-10 C for the heat exchanger.

It is a complicated system -- lots of valves and reservoirs and so on, but very efficient and doesn't need much maintenance.

I grew up just outside of Ottawa and we had a system like this 20 years ago. I don't think you can get a permit for them anymore because they don't like you moving water from one water table to another because of contamination reasons. Nowadays, it's mostly ground loops being installed. In rural settings where natural gas isn't available, and you have the space to run the horizontal loops, they're a good option.

Worse than that, they won’t allow direct expansion systems. So you need a glycol loop going through the ground which has another heat exchanger to the refrigerant loop.

I wonder how we could work this for an urban setting - could we use the water lines?

It consumes a lot of water, so it'd be expensive if you send the outflow down the drain. If you pump the outflow back into the fresh water supply, you have to worry about contamination from everyone's heat exchanger.

In a dense urban setting, you can heat homes with water heated at the local power plant, either fossil or nuclear. https://en.wikipedia.org/wiki/Cogeneration

There's closed loop ground source heat pumps. It's all very situational as to whether they are practical to install. The vertical ones are more expensive but don't need a whole lot of ground area to be installed.

In my city, about 10% of the district heating load is covered by heat pumps where the heat source is municipal waste water.

Toronto does this for district air conditioning with incoming city water.

I don’t think they’ve bothered to calculate the effects of pumping warmer water into everyone’s air conditioned home...

If we think insulating a single home is bad, image trying to build the infrastructure for that throughout a city.

You could do some heat recovery from the sewer, but it would be a yucky affair - they already tend to clog even without heat sink fins in them :)

Here are some calculations suggesting that the numbers simply don't add up for ground-source heat pumps at urban population densities: https://www.withouthotair.com/c21/page_152.shtml

A similar idea that's somewhat popular in Sweden is to add a heat-exchanger between outgoing sewer water and incoming tap water. You don't need to put fins into the outgoing water, it's simply a piece of regular straight copper pipe, but it's thicker, and the incoming water pipe coils around it a bit.

No need to muck around with fins in the sewage, or trying to convert to electricty, simply heating the incoming water a degree or two saves energy for your water heater.

Yes, we've got this on our shower drain. In this configuration it does better than a degree or two, recovering some 50% of the heat, in fact.

You normally have to drill one of more well holes. If you can't go horizontally you have to go vertically.

Maybe you have a really old one or an ac-unit in reverse? My pump works perfectly well in -15 C and I have no problems with ice.

In Norway most pumps are Toshiba and Mitsubishi. F.ex Mitsubishi FH35.

that sounds like it needs maintenance. I can't remember what it was but something was causing that to happen our ac unit in the heat and we paid like 200 to get it fixed and it stopped happening.

Heat pumps all have defrost cycles where it'll run like a regular air conditioner to heat up the outdoor coils temporarily to melt off ice build up. It turns on auxiliary electric heating inside in order to maintain heating while it does this. Yours clearly is broken in some form or another, it's either not going into a defrost cycle at all or even a full defrost cycle isn't enough to melt all of the ice and it's building up faster than it can be defrosted. Just call an HVAC company and get it fixed.

> I want to get on my soapbox and yell "INSULATION PEOPLE, INSULATION!".

I'm joining you in that. I'm in Dubai at the moment thanks to airspace closure over Pakistan.

I noticed that richer Middle Eastern countries are the few places outside of the ex-USSR where foamed concrete picked up popularity as an insulation material. Seems to be a very popular choice for detached housing here.

I also noticed that despite the fabulous wealth, glass curtainwalls are not so widespread here. Even the most "skyscrapery" skyscrapers here have some solid walls and/or external shading (even Burj Khalifa on some levels.)

Many Swedish suburbs are supplied with communal hot water for heating (fjärrvärme) produced by burning non-recyclable rubbish or maybe a local factory. The last stage of houses built near us was so thermally efficient that the supplier refused to extend the heating network, as there was not enough consumption to generate a profit. When I lived in New Zealand I could see my breath indoors in the winter. Insulation? I'm a convert.

Are houses in the Nordics framed with wood? I was having a discussion with someone the other day about houses in the U.K. They felt stone houses were far superior to the wood ones in the US. I tried to convince him about insulation being much better in wood frames homes in mild and cold climates to no avail.

My exposure to the Nordics is limited but in the pictures I’ve seen there were lots of wood single family houses.

No new construction will be stone, but it will most likely be brick, with an inner of cement block.

I think relative cost of timber is a bigger factor than you might think - the US has endless forests, while the UK came quite close to deforestation round about WW1 (earlier in Scotland). Traditional Scottish cottage construction had essentially only one structural timber piece - the roof beam. That was the most difficult part to source.

In all seriousness, what are the advantages of stone? It doesn't hold up in an earthquake, it doesn't insulate, and it can't be renovated after the fact. I've encountered this mentality before and am absolutely baffled by it.

The old way of heating homes in the UK was to provide little insulation but lots of thermal mass. A strong fire would heat up the mass in the chimney and around the home. The fire could be damped down over night and relight it the morning.

There is also a mentality of homes being forever. My house was built originally in 1870 and that's not particularly old. In the US old homes (except perhaps in a few stylish old areas) are something which need torn down and rebuilt.

> In the US old homes (except perhaps in a few stylish old areas) are something which need torn down and rebuilt.

I don't really understand where this belief arises from. Most homes in the US are built on undeveloped land. It's not typical to tear down an existing house to build a new one.

To the extent that it happens, it's generally because an area is essentially undergoing redevelopment. i.e. a rural area is being consumed by an urban city. In cases like that, the purpose is to increase density. It's pretty rare to tear down a single family home just to construct an equivalent but newer one on the property.

In rural marion county Kansas, it's fairly common for a family to tear down their 80-100+ house and build a new one. I'm living in a 6yr old house that replaced our 100+ yr old house.

I'm just arguing anecdote against anecdote at this point, but I question the frequency of this, mostly because it's so absurdly expensive to tear down a house to rebuild (because you have to buy the house before you tear it down). In my experience, I see houses torn down to rebuild only to increase density (two homes replace one) or to replace with a fundamentally different type of house (3500 sq ft house replaces a 1200sq ft house, modern energy-efficient house replaces unmaintained home on the verge of falling down). The value gap between the new home(s) and old home has to be very large to justify a teardown.

To the extent that teardowns are more common in the US than the UK, I imagine it's driven mostly by the ability to massively upgrade in the process. If you live in a 150-year-old stone house in the UK that's been updated over the years so that it's got modern electrical and plumbing and already maxes out your square footage (meterage?), what would you gain by tearing down except a lot more debt and some better insulation? With that said, I think this math generally doesn't work out for Americans, either, who generally trade up homes rather than rebuild.

only in high property value areas does it usually happen. say the house was built in 1930-1950 and is 2000sqft yet the 1/4 acre property is worth $650,000

Raze, rebuild, repeat: why Japan knocks down its houses after 30 years


> In all seriousness, what are the advantages of stone?

If done right, it is likely the most durable material to use for siding. Will last for hundreds of years. Same with bricks if well maintained.

All other kinds of siding deteriorate quickly in comparison.

I am keeping an eye on the new cement siding that's now on the market, that's very interesting.

"Nordics" covers a pretty diverse area regarding climate and quality of local timber. Most houses in Denmark (southernmost part of "nordics") are built with bricks because wood has a tendency to rot in our climate. Also, we experience almost no earthquakes.

Or because clay is cheaper than wood in Denmark.

Bricks are not just clay. Ask yourself: How does clay become bricks? Hint: it has a cost.

I know you are asking generally, but the parent comment mentions the UK; more or less, the UK doesn't have earthquakes. Yes it has some, but:


The British Geological Society lists "Significant Earthquakes since 1932" and there are twelve, but they're dated 1580, 1816, 1880, 1884, 1896, 1931, 1984, 1990, 2002, 2008, 1974, 1979. Three are offshore, three in the Scottish Highlands.

But wood in the UK was in demand by the Royal Navy, to the point where they depended on imported oak from the Baltics, forest cover in the (small) UK was down to 5% around the turn of 1900.

https://www.theguardian.com/travel/2013/jul/27/history-of-en... and a skim-read of the start of http://www.wou.edu/history/files/2015/08/Melby-Patrick.pdf

I wouldn't be surprised if the advantages of stone in the UK were/are availability until it became the default.

Familiarity. People like the construction they're familiar with. There are pros and cons to any material.

Probably an emotional thing, "set in stone", a sense of permanency, etc.

Yeah well, the pyramids still stand today 4000 years later. So the "emotional thing" might have some ground in reality.

It insulates sound very well, which is important to some people.

It doesn't burn.


Mindful of how flames [Great Fire of London] had destroyed a third of the largely wooden city, building with timber frames was banned in the years after Sir Christopher Wren's 1667 Building Act. [..] until 1999

Ah, probably a lot of that involved.

95% wooden frame, typically 45mm+190mm+45mm insulation.

For small houses wooden frames are most common by far.

One thing to add to insulation, make sure your cooking appliances have decent internal insulation too; otherwise you may find your gas oven overwhelming whatever natural radiative qualities your home has, even with all the windows open. Not a fun thing to learn when your AC gets knackered in the florida heat.

Is -20 a practical limit for them working in your experience or is this just a convenient low number for the sake of an example. I assume that is Sweden the temp can drop below that at times in the middle of winter. Are housing using them using a backup heat source as well?

On the west coast of Canada we'd probably never hit -20, but lots of other parts of the country do regularly.

It's a nice round number that the manufacturers use a reference point, it's not a hard limit. In 99% of the cases, the out door unit will be connected to a indoor counterpart that has a 9kW resistance heating element what will aid with heating as the outdoor unit loses efficency with dropping temperature.

You get lower efficiency for lower temps, so for a typical Swedish house you just need to consider how many days in the year you expect sub -20-25 temps. For e.g Stockholm this is only very few days per year so a good idea is to have a fireplace (which is nice anyway) as a backup in the very rare cases it’s -25 or colder.

In northern Sweden you can’t be surprised to see a month of -20 some years, so you probably want either a bigger heat pump, expect to use more direct electric in the pump (the heat pumps are air/water so the houses are heated with circulating water, not hot air) or something else entirely like geothermal.

The problem with heat pumps in cold temperatures is that the compressor (outside) unit has to go below the air temperature in order to absorb heat. I'm not sure the difference, but I'd expect it to have to go to -30, -40, -50, at those temperatures its difficult to build mechanical devices that are going to work reliably for 10+ years.

There are some Mitsubishi and Panasonic units that (IIRC) can work in -30C.

> Modern, newly built houses in Sweden

Also worth mentioning is that Sweden pushes a lot of retrofitting of older houses, not just new ones. In the 90's, triple-pane windows became a legally mandated standard, so every old apartment building that only had double-pane windows had to add a third pane somehow, or change out all the windows or whatever.

A lot of older buildings are also adding extra insulation, one complex I used to live in that was built in the 1940's did that after I moved out. They stripped off all the outer layers to get to the brick, added foam insulation blocks, and redid the plaster finish.

When I lived there, the complex did the mandatory energy certification and got a shit score, so that's why they had to do it.

The Google X spinoff Dandelion Energy is now testing in my area, and I have been experiencing frequency illusion with geothermal heating since I saw an ad for it.

I do wonder how X's technology compares to existing household products, though.

My dad is such a pain in the ass about this - he'll turn the heat down to 15 degrees in any room that's not currently occupied, and whenever they leave the house. So the place is always uncomfortably cold, there are hot spots/cold spots and drafts. Meanwhile there's minimal insulation in the attic and zero insulation under the floors, and a huge basement that's always cozy and warm all winter, despite the fact that there's no heating in there.

Then he complains about the heating bill and blames me for it (because I cranked the heat when I came over for dinner one night). What a mystery.

I experienced a modern heat pump in action for the first time about a year ago, when I was visiting family in the very cold US state of Maine -- creaky old [and ill-insulated] house in the woods heated up in a jiffy! Hats off to the engineers who work on this stuff.

p.s. even though the refrigerants are bad, still a mile better than the old oil heaters everybody used when I was a kid

What about at the other end of the spectrum? We wanted temperature control in our garage so we had a heat pump installed. The company told us that heat pumps stop working for cooling around 38 C which kind of sucks because that's when we want it the most.

In cooling mode, heat pumps work by heating the outside of the garage/house. To do this, the refrigerant has to be warmer than outside temperature. For a given fluid, you get an upper limit temperature like you mention, where efficiency drops to zero. This is because the fluid has a critical point, as you approach critical temperature the achievable enthalpy change in the condenser drops to zero.

Is there fluids that are optimized for more extreme climate on both ends?

For industrial use, there's heat pumps optimized for pretty much any temperature range you can think of. We have one in the lab at work that goes from 40 C to 130 C, another (cascade unit) that goes down to -200 C, etc. But price goes way up and efficiency way down as you increase the range.

Can you please direct me to a supplier for the kind of heat pumps you have in your lab? I've been trying to source a high temperature heat pump for an experiment of mine, and I haven't really found what I am looking for yet. I'd appreciate any leads you could give me.

I'm afraid the high temperature one is a custom built unit, actually for an R&D project of some colleagues of mine that aims to demonstrate the efficiency and reliability for high temperature applications. You can email me to discuss, if you want.

Technically, any standard AC unit is a heat pump, so that doesn’t make a whole lot of sense.

Isn't the refrigeration loop "tuned" for a particular operating range? Like, if I want a heat pump for extracting heat from cold winter air, the operating envelope will be set up differently than for a unit designed to dump heat into hot summer air.

I'm guessing here. I just know that the outside condenser and inside evaporator (in a cooling setup) need to match each other pretty well to avoid damage to the compressor or inefficient operation.

I'm sure there are many parameters to tune, although I don't know the details. But a "heat pump" is just a machine that pumps fluid in a certain way to make one end hot and another end cold. With a traditional AC, the hot end is placed outside and the cold end inside. What the HVAC folks call a "heat pump" is a particular variety that can reconfigure itself to put the hot end inside. If you were so inclined, you could heat a space by taking an AC and turning it backwards so the hot end was inside. (This would be pretty dumb unless you needed to heat stuff up in the summer, since the temperature range it's built for would be all wrong otherwise, but it works.)

I'm sure there are heat pumps that don't cool well when the temperature is too high, but I don't think it's a general property of all of them.

It mostly depends on what refrigerant is used. All heat pumps need to be able to change the phase of the refrigerant, which becomes too hard above or below certain temperatures. You can extend the range by having a higher pressure pump, but that reduces efficiency.

FWIW, our central air struggles above 100F has well. On our hottest days (105 F) it can barely hold 80 F inside.

Have been suffering in our apartment with single-pane windows this winter in Los Angeles. :-(

Has it been a particularly cold winter there?

We're on track to have the coldest February since the 1950s, so by that set of standards it's particularly cold. However I didn't get a lot of sympathy from a friend of mine in Minneapolis when I told him how miserable it was walking to my car when it was 50° outside

Reasonably certain there are no apts in MN with drafty windows from the 60s. Thanks, NIMBYs.

You would be wrong.

Folks living there would be dead.

I had some parts of my windshield frozen a while ago. That is really rare.

Parts of LA just got snow recently.

To be fair, here in California, all newer homes have insulation preinstalled. I think there may still be a tax credit to help have your home insulated if it doesn't have any insulation.

> require less than 5000 kWh per year to heat and generate hotwater

For what size house, if I might ask?

What sort of insulation is in use in Sweden?

Around 300mm for new construction (a foot in freedom units). Glass wool or styrofoam. Slightly less in older houses.


Nitpick / clarification: this is not 'EPS' which is what is commonly known in the US a styrofoam (although according to Wikipedia, the trademarked name 'Styrofoam' actually referes to XPS); so not the white grainy stuff e.g. some disposable coffee cups are made from. Modern insulation is done with PIR, PUR or XPS.

It's worth noting that the temperatures at which heat pumps function well is more of a design choice than a fundamental limitation. You can make a heat pump which functions down to -20C with HFC refrigerants, maybe lower with some other types. However, you may choose to optimize 'mild weather' performance by forgoing low temperature performance. Modern heat pumps work pretty darn well across a range of temperatures.

What we really need are heat pumps which do not use HFC refrigerants - they are a pretty significant cause of global warming, and there are other options (CO2 being one of the best, in my opinion). https://www.drawdown.org/solutions/materials/refrigerant-man...

Heat pump water heaters have efficiency of 3x+ vs .9 or so for gas or electric resistance. If you have have an electric resistance water heater get a heat pump model. It costs more upfront but will pay for itself in not too many years, especially considering many places have incentives. In the US natural gas is so cheap the economics are not nearly as good, it takes a long time to pay back the difference but does eventually. With even a modest carbon price obviously this would be different.

If you are going to replace your furnace or AC get a quote for replacing with a heat pump instead, either central or multiple mini-split heads. Heat pumps have gotten a lot better in recent years, particularly being able to work in cold weather down to 0F or below, expanding northward the places they are practical.

Electric resistance heat is 1.0 or 100% efficient, all electricity put into the heating element is converted to heat. Not sure if your trying to include power plant conversion and transmission losses, but that gets very complicated.

Gas units range in efficiency based on how much heat they recover from the exhaust, 80% is typical, 90+ for a condensing heater with two stage heat recovery. The problem with condensing is the exhaust gases fall below condensation point for the water vapor created during combustion requiring a condensate drain complicating the install location and cost significantly.

Heat pumps are over unity they use some energy to move more energy from the environment into the water, they are basically solar assisted resistance heaters.

The energy cost difference is not as clear when comparing a heat pump tank storage unit to a gas on demand tankless they are both typically more cost effective than tank storage electric or gas.

I have a 80% efficient tankless natural gas, it is wonderful having endless hot water at all faucets concurrently with a family of 4 plus occasional guests. I wanted to go condensing 90+ efficient but it was to complex based on its install location.

In typical mounting of electric radiator elements, some of the heat might be absorbed in the walls and windows that are facing outdoors. Then the temperature gardient in the wall/window insulation is steeper, leaking more heat outside. So yes, all the energy is converted into heat, but some of it may be heating the outdoors.

This has nothing to do with the type of heater but rather the placement. When talking about relative efficiency you look at the heater itself and not common factors as they can be eliminated from the equation.

Speaking specifically of heat pump water heaters - I will absolutely cede that they are more efficient, but holy _crap_ they take forever to recover.

My previous water heater was a 70 gallon conventional electric. It didn't have quite enough hot water to get everyone through showering when we had extended family over, but it recovered reasonably quickly so that we only had ~ 30 minute gaps or so in waiting between the person who felt it get cold first and the next in line. Pretty reasonable.

Ours was about 20 years old and we also wanted a bunch of other plumbing work done, so we decided we may as well bite the bullet and get a new one. We replaced ours with an 80 gallon hybrid (electric + heatpump). We figured the extra 10 gallons would probably actually completely negate the need to wait between showers for those last people, so we were super stoked.

The reality is it exacerbated matters drastically. The refresh rate on our old water heater apparently was almost good enough to keep up with sustained usage between 7 people. The current one runs out of hot water out of 2 and takes well over an hour to even have enough hot water for a single next shower.

I tend to take pretty long showers in the morning - like 20 minutes or so. I use a CPAP and clean my mask, I brush my teeth, I brush my invisaligns (which is why I do my teeth in the shower - it's impossible to scrub invisalign braces and not fling infinite toothpaste flotsam everywhere), and I wash myself (which honestly is the fastest part - I keep my hair next to non existent and just do a body wash from head to toe, rinse, and leave. The point is that the 20 minutes I spend in the shower, with a low flow water reducing showerhead, still means my wife gets to get a cold shower an entire hour later.

However, from what I was told, there are new regulations regarding electric hot water heaters; namely, you can't get over 60 gallon tanks with a conventional hot water heater. Apparently people are getting around this by buying light industrial ones, which don't have this limitation. I wanted to try to cut down on energy usage AND I was keen on the idea of having a heat pump in my garage, transferring the heat in said garage into the water, thus leaving the garage cooler. It's like a really inefficient air conditioner for my garage, but given how hot that gets in the summer and how little I spend time in my woodworking shop, I wanted to leap at the chance to maybe make it a better environment.

All of this is a long winded way of saying it may win on efficiency, and it does have a number of compelling secondary features about it, but it also still suffers from poor refractory performance. If you have one as well - do you notice issues like this? Perhaps the one I got is faulty or just ... not good.

Have you heard of drainwater heat recovery? One of the side benefits is dramatically improving the recovery rate of your water heater.

(Unfortunately it doesn't work well with certain plumbing layouts, and the rising price of copper has reduced the cost effectiveness)

Just moved into our "forever" home and our setup is very similar to isbjorn16's (7 in the family, HPWH "80G", but I have an 8" PowerPipe from these folks: http://renewability.com/#learn-more. (It's a monster - plumber wanted to give me a clean out at the bottom - but that meant some jackhammering to set it low enough.)

So far so good - we do a bunch of baths/showers at bed time and no one runs out of hot water. Also - I'll add that we accidentally had the plumber remove all the low flow "plugs" from the showers so there's a lot of water flowing.

The pipe is pretty fun - not only is it fun to look at - but when someone is taking a shower it's fun to feel the heat at the top of the pipe and as you move down the pipe it gets colder until it reaches the temp of the cold water coming into the house.

Since it was new construction - we'll never know how much of a different the drainwater heat recovery pipe is making - but it is definitely doing work.

I can imagine these pipes becoming more popular (though rising prices will kill off any efficiencies real fast). Maybe code?

I have not! https://www.energy.gov/energysaver/water-heating/drain-water... seemed to make a fair amount of sense for me, so if anyone else is curious, there's a link!

My plumbing layout actually might be darn near ideal for this. I'm not super keen on spending another thousand or so on getting it installed - at least not prior to the "planned but not scheduled" bathroom remodel, but that is at least worth considering when we do do the major remodel and complete bathroom reconfiguration. Our master bath is on the first floor and we don't have a finished basement so making something like this work should be comparatively easy. Thanks for the idea!

FWIW they aren't difficult to do yourself, as long as you can sweat pipe. It's a very simple device after all.

Is it possible that the setpoint on your old water heater was higher than the new one, so that you used less litres of heated water for the same shower (since it mixed in more cold water)?

Actually, come to think of it - it might just be that it's pretty cold in my garage right now, ~ 40F: maybe it just isn't getting much help from the heat pump and the conventional electric heater isn't doing a great job on its own?

I'll have to pay attention into summer when it's like 95 in my garage. Maybe I can presume raging hot showers -for days- at that point :D

That's definitely part of it. I'm actually surprised it's working in heat pump mode at all. We have a hybrid and it will run pure-resistive if the ambient temperature is below 50. Now, it's in our basement, so even in the winter that's a reasonable cut-off - not sure if you got a specific model for being "outside".

Speaking of which, does your model not have a resistive backup? Ours has several modes:

- only use heat pump, unless not safe to do so (ambient too cold) - use heat pump normally but fall back to resistive to recover - use resistive only

We only have 2 people in the house so we are able to leave it on the first mode, but in the second mode it would, presumably, be the same performance as our old resistive-only heater.

Solid questions - I'm going to take a peek at it tonight. I honestly haven't given it a lot of thought, but literally just this morning my wife went off on a rant about it for a good 15 minutes. It suddenly bubbled up my priority queue pretty quick - and then out of nowhere, this thread on heat pumps and water heaters! Crazy coincidences sometimes.

I use a showerhead shutoff valve -- when I'm shaving in the shower or doing anything that doesn't require a steady shower of water, I turn off the water at the showerhead, then I can flip it back on when I want water, and it continues at the same temperature/pressure it was before.

I also use a 1.5gpm low-flow showerhead (many "low-flow" heads are 2.5gpm).

That would be a mitigation strategy, but my main point is my old conventional had significantly fewer and less lengthy hot water outages compared to my new fancy hybrid with a larger tank. To be clear, we make this work - it isn't a catastrophe by any stretch, but it is a marked surprise when you expected superior hot water uptime and you get distinctly WORSE hot water uptime, you know?

I can tell you that my heat pump set up at my condo is absolute trash here in Northern Ohio. Running 24/7 it is unable to get my condo above 62 deg F. It blows cold-ish air under an outside temperature of 40 deg F. Windows are new, doors are weather stripped. I'm having it run it in supplemental mode just to get it above 65 deg F.

My HVAC is utter crap ≠ HVACs are crap. This is a great example of a worthless anecdotal comment that distracts from the point. Give us some more detail like model number or install date? Is it a modern unit not coming close to its advertise performance? Is it appropriately sized for your condo?

Your heat pump is probably about 20 years old and on its last legs. Newer ones are amazing. We lopped $40 off of our electricity bill each month in the winter and $20 in the summer after we replaced ours. It pays for itself after a few years.

There are some very bad heat pumps out there. The difference in usability varies dramatically. Some heat pumps will produce heat even below 0F. Others are crappy at anything under 40F.

From what I understand, cold weather air source heat pumps are relatively new to the market, at least in the US. Previously they were designed more for cooling loads and were essentially space heaters with an COP of 1 in cold weather.

Many manufactuers seem to offer warm and cold climate models. For example, pulling a random manufactuerer, Mitsubishi brands their cold climate line "Hyper" and specs out much better efficiency down to -13F temperatures.

Something must be wrong. My heatpump will run you out of the house at 40F outside.

Below 20F it puts out so so heat but can maintain. Below 10F electric heat is needed from time to time to maintain. Below 0F heat pump is worthless.

Edit: Wanted to add my is air heatpump, not ground source. Also reading other comments looks like heatpumps range from amazing to utter crap.

Your heat pump just does not have sufficient capacity. It's not problem of heat pump per se.

1. Have an HVAC tech inspect it (~$100-150).

2. If it's not a coolant or compressor issue, you might need to replace it (either because it is undersized or end of life). Your local government or utility might have energy incentives for replacing it. There are also federal tax credit incentives [1].

[1] https://www.energystar.gov/about/federal_tax_credits/air_sou...

A heatpump blows warm air not hot. Most people expect hot air from their vents and so are confused by not getting it.

My heat pump in Iowa worked just fine to keep the house comfortable down to 14F (the system automatically switched to backup heat colder than that). Something is wrong with your system if it can't do that. I couldn't begin to guess what though.

Big issue in the Netherlands (and probably more countries) will be noise pollution though. Houses here are built really close to each other and the combined noise could be an issue. Neighbours are already complaining about the noise from my parents heat pump.

Modern heat pumps are extremely quiet. I have one right outside the bedroom window and never hear it. Little to no vibration either, despite it being mounted directly to the house. Perhaps older ones are noisier.

I had a few units installed in my house last year. I was a bit confused when they described the "low-noise" mode because each indoor unit has a separate remote, but the "low-noise" applied to the shared outdoor unit, not to the indoor portion.

Do you have any centralized district heating in NL similar to in Sweden?

I expect this to become more common in cities in the Netherlands for existing neighbourhoods built between the 1920s and 1970s. These older houses (of which there are a lot) are difficult, if not economically infeasible, to convert to heat pump heated houses.

Insulation can be done to some extent — I've managed to insulate my 1930s house to a thermal resistance of R⒞ = 4,5 m²K/W in the attic and parts of the level above the ground storey, but this is not realistic for the ground floor — but it is not feasible to reach the R⒞ ≥ 6 m²K/W mandated for newly built houses.

The noise is a serious problem too. In terrace housing you simply can't expect everyone to install these things and maintain a sane level of quiet.

Better to solve this problem by insulating as much as possible, and to provide heat by collective solutions such as geothermal heat on a neighbourhood scale. The current 'let-the-citizens-solve-this-themselves' approach of our government is making homeowners insecure about what to do though.

In some newly developed areas we do, but most of the heating in the Netherlands runs on natural gas, since we have so much of it.

...or rather since we had so much of it. The Dutch government is, after all, making noises about the country going 'off the gas' to become 'gas-free' around 2050. As to whether this really will happen and what it actually will entail - are they going to tell industry to stop using gas as well? - remains to be seen but heat pumps are being pushed as one of the alternatives to gas. Given that nearly all central heating in the Netherlands is water-borne and most systems are designed for a high working temperature - between 70 and 90 degrees - the switch to a heat pump will need more than just a simple swap of the prime heater. Low-temperature (possibly fan-assisted) convectors can be used to replace high-temperature radiators, these can also be used to cool down the house in summer. More insulation is needed in most Dutch houses, even newly built houses sometimes don't have more than 10 cm of foam insulation in the roof and less in the walls.

The most recent season of This Old House, Jamestown Net-Zero House[0], is worth a watch for folks interested in learning more about home efficiency techniques in general.

[0] https://www.thisoldhouse.com/watch/jamestown-net-zero-house

I'll add that to my must watch list - I live in an old (1930's) home that leaks heat like crazy, so I'm interested in possible retrofits, though I'm a bit concerned that it'll be a monstrous task.

Yeah, in this case they spare no expense with the upgrades. It's really cool to see and learn about but probably not realistic for anyone not ready to invest a lot of money.

As much I love TOH, I do wish they would take on more "budget" projects or at least talk a little bit about costs. But ATOH kind of fills that gap.

Pretty much anything in a pre-1980 house will a nightmare to retrofit. Houses in the 1945 to 1965 era (in the US) especially were mostly built as quickly and as cheaply as possible.

The concept of energy efficiency didn't really exist before the 1973 oil crisis as oil use per capita was steadily going up and up until then.

Energy Star, for example, was only established in 1992.

Energy efficiency is very very recent.

Oh yeah, my home was built strictly to keep the frozen death out, I have to throw a big furnace at it to be livable while it exhausts everything through its impossible to insulate roof.

Good channel on building science:

* https://www.youtube.com/user/MattRisinger

He does renos and new build in Austin, so mostly worries about keep heat out, but air tightness and insulation apply everywhere.

I had a "geothermal" heat pump put in to my house via Google's Dandelion and it works great so far. It's been -30F at my house but the ground sourcing renders that moot. As far as I can tell their innovation was going vertical so they didn't need to dredge up a huge loop of land. The pump has two channels that go down 300ft each and the fluid runs down and back up each bore hole.

It provides heat (working great so far), hot water (but I already had an electric inline heater, so I can't speak to the slow-refill issues), and A/C in the summer. I'm curious to see how well the AC performs!

I don't see Google mentioned anywhere on their site. What's the relation?


Alphabet spinoff, looks like: https://dandelionenergy.com/about/

Is your house entirely electric? That's a dream of mine but I don't know how realistic it is, especially in my state where there aren't good incentives

No, we have a propane fireplace that used to heat the whole house (terribly inefficient, the previous owners used it as a summer-only property) and we now run that just on colder evenings to make the house a bit cozier. We also live way out on a dead-end of the grid, so outages are common and sometimes long enough to need backup heat to keep the pipes unfrozen.

I also eventually want to swap out my electric stove with propane, though my last experience with induction stoves has me considering them too. Electric conduction stoves just aren't fun to cook on...

I'd love to convert the fireplace to run on hydrogen that I create myself via electrolysis one day (and have 0 emission tiki torches on the porch) but I'm not sure how safely I could pull that off! It's a bit of a moot point until I get solar installed anyways, since my power isn't coming from a clean source as it is.

Maybe I'm missing something, but no matter how efficient a heat pump is, doesn't it have a limit on how much heat it can pump as ∆T increases?

Even here in western NY we occasionally have to deal with -10 F (-23 C) weather for a week and so that's a pretty big temperature delta if you want a comfortable 68 F (20 C).

Combine that with the fact that gas is still quite a bit cheaper per unit of energy than electricity (in my area) and a heat pump just doesn't make any sense.

The ability to have a single unit heat and cool is definitely nice, but without a geothermal-like ambient temperature to work with, I just don't see this working out, even with all the insulation in the world.

Elsewhere in this thread fastbeef says "They're very common in Sweden and can easily warm an entire house even at -20 C [-4F]." So, it can work though at -10F you would probably need supplemental heat.

Of course in Sweden residential natural gas prices are almost 4 times what they are in the US so its a very different economic decision. Shows what a carbon tax can do.


Keep in mind that 1 kW worth of gas gives out 0.9 kW, while 1 kW of electricity entered into a heat pump gives out AT LEAST 3 kW of heat.

Also, no one uses gas for heating in Sweden. Very old houses down town use it for cooking only.

Aren't you supposed (at least when you have a single house) to couple it with a geothermal energy so that your external temperature is always the years median of around 9 C?

Geothermal or some other relatively stable source. If you live near a lake you can drop a water sling on the bottom and use that as a source, it is also possible to bury a sling in a field next to your house at a frost-free depth (2.5 m or deeper in the north of Sweden) and use that. In built-up areas with many houses using deep wells for heat extraction there have been cases of those wells freezing up, this can also happen if the system is under-dimensioned.

Question to other engineers here. We were designing a "smart wine chiller" recently. Ran into a problem - small size sealed compressors are very, very, hard to come by, and if you want one that is also quiet, prepare your arm and a leg.

We ended up having the compressor custom made by local fridge factory. Engineers there told us that designing such a small piston compressor and hoping it to be reliable and efficient is a futile effort, but they still did their best.

Is between piston driven compressors, and peltier coolers, is there just anything on the market? More quiet and robust than small piston compressors, and more efficient than peltier?

Sure. Look at fridges built for RVs and the like. The ones that can run off of propane as well as electricity. They use the absorption cycle, so while they have worse efficiency than a compressor-based system, they will easily beat thermoelectrics from a cost/BTU standpoint. Many have no moving parts, so noise and vibration are minimal.


Have you looked into compressors for mobile applications, such as a Danfoss 35F? I mean, technically its a reciprocating piston compressor, but the smaller ones can be quite quiet.


Not small enough. The one we custom ordered is like 6x6cm (piston itself is only 9mm,) and they even managed to make it run on DC.

The factory and their tech level truly surprised me despite them being complete nonames, who mostly do fridge part kits for local production in 3rd world countries.

This reminded me yet another time of just how underappreciated are those no-name factories. They truly are the real unsung heroes of our globalisation era.

Adding to that, the most expensive single part in all at happened to be the specialty BASF o-ring for the piston at $6 dollars...

Body, piping, insulation, every other part came at less than that

Stirling engine?

Slightly OT, but..

I live in a temperate climate. I've sometimes wondered if it would be worthwhile to adjust house construction so that:

- In the summer, the kitchen refrigerator's heating coils touch outdoors air instead of indoors air.

- In the winter, the refrigerator's cooling is provided, at least partially, by the outside air.

- In the summer (only), the kitchen oven vents into the outdoors rather than into the kitchen.

- In the winter (only), a heat exchanger recovers some of the heat from our electric clothes dryer.

One idea i've had is to just make a central heatpump system.

Why does a fridge need its own compressor that vents heat back to the house even in the summer?

Why can't your water heater use that same heat instead?

This all started because my brother used to redirect dryer air back into his apartment during the winter, using a saw dust sock to keep the lint contained.

I started thinking about how heat could be better managed if everything used a central ac/heat pump. heat generated by the fridge, your computer, the ac, etc could be used by your water heater, your heater heater, your dryer.

Add heat recovery systems to your dryer, and your drains, and bam super sustainability.

There are AC units that do this: pipe the heat to the hot water tank first.

* https://www.rheem.com/innovations/innovation_residential/int...

* https://techcrunch.com/2011/06/12/air-conditioning-water-hea...

Once the water tank reaches its max temperature, then heat is dumped outside.

The practical problem is that you need to have the different parts work together, and fridge makers don't don't to hot water makers.

>The practical problem is that you need to have the different parts work together, and fridge makers don't don't to hot water makers.

You don't need a fridge or a hot water heater or an ac or a furnace at that point, because the primary function has been offloaded to the central heat pump. you need heat exchanger attached to water, heat exchangers attached to a cold room, heat exchangers attached to a central air system.

You don't have to have the parts work together if the most expensive part of each appliance has been centralized. At that point you've now reduced the appliance to where to put the heat exchanger.

In the apartment I live in now in Hawaii, the utility closet has the AC unit mounted right above the water heater.

They are not connected at all. And this is in a building that is about ten years old.

I have no words, it's so fucking stupid.

There are AC units that, instead of sending heat to the outdoor device, pipe it to the hot water heater.

There are also hot water heaters that, instead of burning something to heat water, have an heat exchanger on top of them to extract the warm from the surrounding air. Usually called "hybrid".

I known Rheem makes such units, but there may be others.

You’d think any A/C tech could loop around the outgoing AC pipe to the cold water intake, and insulate it.

It won’t be anywhere near 100%, but the extra $50 in copper pipe could pay for itself.

Recovering heat from a dryer is not trivial: the output air from a dryer has a bunch of lint in it (yes, even with the lint filter in there!), so either your heat exchanger or the filter in front of it or both will get clogged pretty regularly.

There are several electric tumble dryers which use a heat pump to recover heat:


Which also don’t require a vent.

I just hang my clothes and let the evaporating water cool the place a bit.

The most commonly used heat pumps utilize the external air as a source. Which is quite silly on a sunny but chilly day. Even 2m^2 black & isolated panel exposed to the sun will be tens of degrees warmer. Just putting the heat exchanger in some micro glass house with a valve would increase the efficiency.

We have a heat pump and I love it, but for the longest time it was beyond me how the thing works. When it finally clicked, it blew my mind. The idea that there’s heat to be found, even on a cold day, is amazing. It’s actually what made me want to understand thermodynamics more deeply.

Also heat pumps work a miracle because the fluids they use are NOT perfect, platonic fluids that obey the ideal gas law.


This was a really lucid and engaging piece on the mathematics of these systems. I wish more math teachings would be done in long form like this instead of the cargo culted textbooks that are put out by major publishers.

Talking about fridge, do you guys know if there's a use for heat collector out of the fridge radiator ?

Right now it's mostly wasted heat, or IIUC even worse, in closed spaces it forces more energy consumption from the fridge.

It’s not a lot of heat in residential use cases.

Fridges for enclosed spaces should have a fan to push the air around.

The more space you give a radiating fridge, the more efficient it’ll be.

I do wonder if a small fan on a radiating fridge would pay for itself if controlled by temperature probes.

BTW the whole blog is a goldmine. Rim-filled of evidence of how our energy-hungry civilisation can't last :)

Heat pumps based on air do not work economically below 40F. Neighbor has one and his electric bill is out of control in the winter. Drop below freezing and the system will pump heat out of the house to de-ice the external unit, and electric resistance heat will kick in.

We have dual fuel, so we burn oil below 40F. Works great.

If you can afford it, putting in a heat pump that works off of ground will eventually pay for itself.

Modern heat pumps are still highly efficient at 40F (4C) the coefficient of performance is 4+ at this temperature. We switched from an hybrid set to a full-electric setup (also PV and solar thermal) and are super happy with (live in Amsterdam, so regularly goes below 40F/4C). There's probably something wrong with your neighbors setup.

I'd say practical limits of current models are probably around -20C (-4F)

If you're interested here is an example datasheet[1] of an air to water heat pump (2018) (using Celsius here):

Mitsubishi PUHZ-SHW112YAA Guaranteed operating range (outdoor): Heating: -28 to +21 Domestic hot water: -28 to +35

On page 108 you can find the actual co-efficient of performance (COP) for different outside temperatures and inside water heating temperatures. As an example: at -10 and an water temperature for heating at 25 you still get a COP of 3+. That means that 100% electricity moves 300% of heat into your house.

[1]: https://www.mitsubishi-les.info/database/servicemanual/files...

Outside temperature definitely affects an air heat-pump's performance, but the ones common in sweden stop being effective at -10c or even -15c.

I have a ground-source pump and its brilliant. But plenty of my neighbours are happy enough with air-source pumps.

Recent conversations with net-zero builders indicate that ground-source pumps are being pushed away for most single-family residential projects in favor of the new-generation, highly-effective, cold-climate air-source heat pumps.

Mitsubishi and Fujitsu seem to have a large lead in cold-climate air-source heat pumps that are rated effective to -13F (anecdotal reports greenbuildingadvisor.com indicating effectiveness down to -20F). Yet, talking to builders, contractors, friends, and family, you will often still hear old tales about how "heat pumps aren't effective below 40F". Heat pumps come in a wide range of flavors. It's clearly possible to heat your house exclusively using an air-source heat pump in most USA climates.

There's a huge difference between air-source heat pumps and rebranded AC units running in reverse. We have a Mitsubishi FH35 6kw heat pump that easily heats 80 m2 (800 sq feet) down to -13 C this winter.

Do you recommend the model you quoted in your comment?

I do, it's fantastic! Just make sure to get a good installer for it.

As a fellow northern climate person, now in Arizona, have you any idea its efficacy in the 45 celsius normal summer heat, for cooling?

Doesn't work below 40F? That's just not true unless you're using a very poor, cheap heat pump.

Modern, high efficiency air source heat pumps can work well even below 0F.

They do work, but with decreased efficency, which in the end line has an effect on total power output (not enough to satisfy heating capacity/needs of user’s installation) and increased energy consumption.

Most of modern heat pumps (air to water) can work without issue all they way to -15/20* C.

Yeah, they have two-stage pumps for cold weather. I thought about it when I upgraded my heat: gas furnace and electric heat pump had about the same running cost, but heat pump was much more expensive to install and much less reliable / more expensive to service.

How new is the heat-pump? We put one in our house this year and been in the 20-40F range for the last 2 months. So far we haven't seen resistive heating kick in more than when it's above 40F.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact