Ask Hackaday: Is It Time For Waste Heat And Cold Area Heating To Shine?

It’s difficult to escape the topic of energy supply at the moment, with the geopolitical situation surrounding the invasion of Ukraine leaving the natural gas supply to an entire continent in jeopardy. Fortunately we’re watching the green shoots of an early spring here in the Northern hemisphere so the worst of the winter weather is behind us, but industrial customers can take no such solace from the season and will have to weather whatever price hikes are to come. Every alternative idea for energy supply is on the table, and with the parallel imperative of decarbonising the economy this goes beyond the short term into a future without so much need to rely on gas.

The Future is Cloudy

A district heating plant in Vienna, Austria.
A district heating plant in Vienna, Austria. Joadl, CC BY-SA 3.0 AT

A collaboration between a Finnish district heating network and Microsoft caught our eye because the location of a new data centre for the tech giant was chosen specifically to supply waste heat to the network, rather than releasing it to the environment. It’s not uncommon at all for European cities to use district heating networks but they are normally supplied by waste incinerators, boilers, or combined heat and power stations. The use of data centre waste heat is a novelty, as is in particular the siting of the data centre being dictated by the network.

Individual gas boilers are economical and convenient, but despite being better for the environment than the coal fires they would have replaced when first introduced, by today’s standards they have a higher carbon footprint than is ideal. Governments around the world are encouraging their replacement with more efficient air-source heat pumps, but as we read about the Microsoft waste heat deployment we couldn’t help wondering whether it’s worth considering the cutting edge of district heating systems — the cloud is other peoples’ servers, after all.

No, Maybe the Future is Cold

When we think of a district heating system, our minds move immediately to a scaled-up version of that domestic heating boiler. A very large boiler or other heat source heats up water, and this piping-hot water is sent down underground pipes to our homes where it flows through our radiators and keeps us warm. It’s simple to understand and has stood the test of time, but as anyone who has walked around a centrally-heated school or hospital on a frosty morning and seen the melted spots on the ground where the pipes are can tell you, even the best-insulated pipes will waste some heat into the soil. The lower the temperature in the pipes the less energy transfer there will be to the soil, but as the pipe temperature approaches ambient temperature of course it would not be enough to heat a home. At this point we return to those heat pumps mentioned above, and encounter the so-called cold district heating system.

Unless we happen to exist at absolute zero, everything has some heat. A freezing cold lake has some heat in that it can be made colder by taking some of its heat away, and that’s exactly what  a heat pump does. The air-sourced ones take heat from cold air, but heat pumps can use any medium as a source. A “cold area heating network” pipes water around at close to ambient temperature to be used as a source for heat pumps on customer premises, which might seem like a pointless exercise until we consider that while some heat pumps can take heat from the system, others can put heat into it. Industrial users can pass their waste heat into the pipes allowing consumers to recover it, and the network becomes a lot like a power grid with many smaller source nodes rather than a simple distribution system with only one large node.

It’s probably too late for a momentum shift towards cold area heating for many cities, but we’re curious to hear from any readers with knowledge on the subject. Do any of you live in a city with cold area heating? Or perhaps you’re aware of another data centre area heating project? We’d be fascinated to hear in the comments.

Banner image: “Steam Heat Plant” by Greg Habermann, CC BY 2.0.

82 thoughts on “Ask Hackaday: Is It Time For Waste Heat And Cold Area Heating To Shine?

  1. I’m glad to see systems like this being discussed more often. It’s annoying to see a bigass plume of steam being dumped into the air during the winter by some industrial process right across the road from a block of homes.

    A really efficient air source heat pump is great, but you’re attempting to dump heat from your house into outside air at the worst possible time: when it’s the hottest part of the day. (Similar situation in reverse for wintertime.) I’d like to know where the breakeven is between heat pump efficiency gains and the cost of an efficient thermal storage volume. One of these days I want to try building a small system just for the hell of it to see what happens.

    1. A possible way of storing energy is by using a heatpump to freeze a basin of water in the winter and unfreeze in the summer. That basin ,good isolated, can be located under the house. The enery to (un)freeze water (state switch) is about 320 MJ / m3. When you then also ,mostly, power the heatpump from solar it should reduce the heating / cooling bill.

      1. Ground or ground water loop cooling is using the ground under you for that purpose. Solar thermal collectors would do most all heating in most areas so a small woodstove could back up if needed. This assumes a hyper insulated building envelope. Passive solar winter gain would also be free if planned aread.

        1. Looking at EU building codes in few years every new house will be in passive house standard or very very close to it. Problem is what to do with all those old buildings, heat pumps don’t scale up to city grid size, and while we have heating grid its coal and gas reliant (one gas boiler for electricity and hot water in summer and two additional coal boilers for heating in winter), getting waste heat from factories and data centres in city would be good option to lowering usage of gas/coal but its still miles from stopping usage completely. I have hopes for small modular reactors, they hooked up to grid and heating water while they produce electricity would be great thing to cut off completely but thats a impossible investment to do on city budget.

    2. I think like all community challenges, the positive changes rarely occur from top down policy. There is also the practical limits of inflicting ones own compromises on others, rather than showing a reduction in peoples gas bills. Also, solar heated buildings may work it some places, but the campus green-building up north famously had profs with electric space heaters in their offices.

      Heat exchangers coefficient of performance is usually around the 2 to 5 range, and thus saves on the amount of energy being purchased to run the system. They are reducing in price, luckily many models do qualify for government green-grants in many areas, and air-to-liquid systems are most efficient. However, the upfront investment into a whole-home central climate control solution can be $6k to $24k, maintenance even with warranties for some brands is costly, and various financing scams pollute customer goodwill. Thus, the calculus for people who do not plan to own a property for over a decade means the cost of such systems is difficult to justify. YMMV

      As a side note, did you know a goat wool sweater is very warm… goat drawn carts are the future given the ubiquity of the technology in our past.

      1. Top down changes are often required – or at least some entity with lots of money has to put in the initial investment, as in the case of M$ siting a data centre to supply waste heat to the district heating system… Without it the individual usually can’t effect any great change – the only thing we could do to our home is replace the boiler with a heat pump, we can’t magically generate the investment needed to use all the waste heat from the local universities computer racks, the sewage works – that has to be a wider area thing and paid for at least in part by government or business (presumably at least in part for their own befit).

        With the way gas prices are likely to go, and stay for a very long time I think the calculus even for shorter term expected ownership will tilt in favour of fitting the greener cheaper to run system – its a selling point when it comes to sell the house, that might well end up paying for itself in the sale price…

        Also Wool might be warm, usually naturally fire resistant, and so very vastly underused and undervalued today, but as quite a large number of folks have wool allergy it can’t be entire future – presumably linen will be making a comeback too.

        1. Current heat pump technology for residential heating and cooling is only good down to 24 degrees Fahrenheit. Once that point is reached the heat pump no longer produces heat and the back up heating system kicks on. Usually resistance electric. At that point the homeowner is running both units. In my latest home I installed a dual fuel system that at the point the heat pump becomes inefficient it is shut down and natural gas takes over. If one also installs a ground loop geothermal system the natural gas is almost never used. 2800 square ft with high ceilings with a total utility bill of 67 us dollars a month is the best I have been able to do consistently. Before the additional dual fuel system average utility cost was 189 and after the dual fuel went to 125 after adding the geothermal went to 67. Additional cost for dual fuel is minimal. Adding the geothermal was not quite 7000. So it is doable today. I keep my home cool in the summer and warm in the winter.

          1. Those numbers must be outdated or something. Or maybe you missed a minus sign in front of that 24°F number.

            Here in NZ, air-source heap pumps are the norm. I recently had a self-contained studio/workshop built in the back yard (best decision ever!) and fitted it with a pair of modern but not-high-end heat pumps – they’re rated for operating at down to -15°C (5°F) with no drop in efficiency.

            Thankfully it doesn’t get that cold here. But newer so-called cold-climate heat pumps will efficiently extract heat from an outside temperature of down to -30°C (-22°F)

        1. People voting with their feet say you are dead wrong.

          Which isn’t even starting the times the USA had to be the adult in the room after ‘top down positive policy changes’ led to world wars.

    3. At least in the commercial space the break even point for a thermal storage systems was about 50 tons (a ton =2000btu). The complexity of the system often meant that any savings were quickly lost in maintenance costs.

      One system that I worked on would freeze an underground tank of water overnight while it was cooler and energy rates were lower, then use glycol flowing through pipes to dump heat into it during the day. The tank developed leaks, which wasn’t too bad until the glycol pipes started leaking into the tank, which meant glycol was leaking into the ground. They ended up draining the whole thing and pumping it full of concrete, and switching over to a conventional direct expansion refrigeration system.

      This was also partly due to the energy rates completely changing due to the closure of the San Onofre nuclear power station (which caused overnight rates to increase)… Coupled with and caused by the explosion of solar power which caused daytime rates to plummet.

    4. In the US, much of housing in single family on lots of in suburban or rural settings and district heating is simply unobtainable.
      While traveling Europe I noticed a much higher percentage of people in high rise housing (yuck) but district heating becomes efficient and effective.
      In the US I would venture to guess solar thermal home heating and DHW would have the quickest payback especially north of the “snow line” (Des Moines Iowa and north in the midwest) ground source cooling)

    5. Hello, nice to read about your interest,

      I would love to read more about actually doing and rethink the „breakeven“ thinking in the field of energy. The only reason why energy is/was cheap is it was produced a long time ago. I know it is all about money but what’s the breakeven of a car, a wedding dress and so on?

      And I totally agree to your thinking that heat is wasted while it could be used across the street.

      All the best peter

  2. Here’s a “novel” idea: Heat people, not buildings. Yep, that winter coat you have … can be used indoors as well.

    “You’ll have nothing and you’ll be happy.” Don’t worry, our “elites” will still have heating at 23 C.

      1. People sat on a sofa / at a desk are generally stationary. Take my wife on the sofa last night, freezing whilst i was quite literally naked and comfortably warm. I fetched the duvet from the bed and she was happy.

        My parents have central heating which for me in the morning is a little cold after its been off for a few hours, but in shorts & tshirt in winter is acceptable. Give it a few hours of the central heating on and whilst they’re now comfortable, I’m unbearably hot for the rest of the day. So yes, I’m all for personal heating – it’s much easier than personal cooling!

        1. Damp and cold will still eat into and destroy brick and stone buildings – with moss and molds helping water get into a crack and freezing, playing landing spot for seeds that will sprout roots that tear stuff up, and lots of wooden elements to these buildings too, sure it won’t destroy the whole structure but its not going to do it any good…

          1. pyramids survived thousands of years in a jungle… and according to some, actually utilized water to generate energy…. those “some” includes Tesla himself. the sphinx shows signs of rainfall if you ask any given geologist. so some stones are better than others, however you are correct that modern building techniques will always succumb to water erosion, expansion and contraction.

            so i think the takeaway os that the choice of building material should suit the environment and use case.

          2. @dudenamedben there are no pyramids that have survived thousands of years in a jungle. And who cares if Tesla had some strange religion about pyramids?

          3. @Shannon, the ancient Mayan might disagree… Lots of their buildings are very pyramidial in shape, its definitely a jungle, and they are still here a very very long time later (not sure of the dating for any of the pyramid structures they built, but at least has to be before 500AD)…

            Of course define survive, after centuries of abandonment they don’t look half as good as they once did, and if they had never been abandoned they might well still look good as new, folks do that to the heritage they deem worth upkeep…

            That said I have absolutely no idea what @dudenamedben is on about, there is no building method of the past to my knowledge we can’t match or beat for endurance if we want to spend enough upfront to build something to last hundreds of years longer than its actually likely to be useful… Sure the Romans had some damn impressive concrete/mortar technologies, heck really good quality and recyclable glass too – stuff that isn’t trivial to match, and remained unmatched as far as I’m aware for centuries after the fall of their Empire, and there are many other examples I’m sure through history, but ultimately EVERYTHING will always succumb to water, at least if it survives everything else long enough for water to wear it down…

        1. Dehumidifier is just a heat pump + air exchange. When you exchange air in heated home, your home is a dehumidifier. If you don’t exchange air but use internal dehumidifier, you will just have co2 increase.

          1. But a standalone dehumidifier is nothing more than a fridge open to the room. The waste heat generated provides some heating to the room. The whole cycle looks very inefficient from the outside, spend energy to turn water vapour to ice, then spend more energy to melt ice back to water which is drained away. The moisture removed from the air allows you to have a cold(er) room without mold worries. Surely this is more of a CO2 reduction that heating the room to avoid mold, especially if the room isn’t inhabited?

            I get there’s a different type of dehumidifier (can’t think of the name) which may be more efficient in that your room heats up less. Whether than’t an advantage or not though…

          2. What if you use silica gel in a bag and some solar concentrator outside to dry it out?

            It will need heating to about 120C?

            Have bags inside while outside bags are being dried.

            Swap over at dusk.

          3. @abjq that is actually what we do here partly – old house that doesn’t breath well enough to avoid damp with all the modern windows and better insulation, at least for now, so some big sacks of silica gel passively keep the problem down and then get cooked off on the bright sunny day – Solar PV and electric oven (as we have those), rather than by solar Concentrator though. So yeah not quite as free or efficient but still not bad and a simple and silent fix (who wants a rattly loud refrigeration pump on the normal dehumidifers running all the time)…

    1. Rather impractical for far to many tasks to wear the full on winter coats indoors…

      Might work in some places – like the UK where the temperature doesn’t drop bellow freezing very far or very often, as that is still warm enough a thin jumper is probably comfortable for most indoors (and personally I’m still a shorts and t-shirt person in the snowy icey temperatures (with a coat/jumper should the windchill be really bad or I’m going to be very inactive outside for ages) – always been rather more comfortable in the cool than the warm, once it gets to 20c outside I really don’t do well)..

      And as Twisty Plastic says no heat in a building is almost certain damp problems – you don’t however have to heat up to 23c…

    2. Heat them by setting them on fire?

      Warm for the rest of their lives! Best thermal efficiency!

      (Para) Pratchett

      I still have my ‘arctic tundra’ coat from growing up. I curl it whenever CA gets too nuts, until the loonies seem slightly less nuts.

  3. Interesting. And there is already a water pipe connecting every building in a city…

    The one entering my house, wide-open, delivers 100 L/min, dropping 500 kPa in the process. It could easily handle 10 L/min with much lower pressure loss.

    That can only deliver about 5 kW before delta-T starts getting big. 5 kW is just barely adequate heating or cooling power for my house. An additional 10 kW would be needed on cold winter nights.

    So, what *would* be the pump power associated with supplying every piddly source or sink with sufficient water?

    This city uses 200 kilotons of water per day, serving a population of 400,000, and requires about 10 MW of pumps to supply it.

    To supply every resident with enough water to supply or remove 5 kW of heat will multiply the water requirement and pumping power by a factor of 30, to 300 MW.

    Not completely insane, but a heck of a capital investment and ongoing cost.

    1. Using the drinking water supply to transfer heat is perhaps not a good idea either – you don’t want luke-warm water circulation around growing nasty bugs..

      If you were to use it however the pumping requirements are likely to go up a vast amount from all the drag caused by the heat exchange surfaces.

    2. If you are talking about drinking water supply, optimum temperature for delivery of water is 8-12°C. This temperature will prevent bacteria growth. Acording to our reqired standards, temperature of water can’t exceeds 25°C.

  4. Feels like some enterprising BitCoin miner could create a furnace-sized heater made of GPU’s that could then be offered to homeowners for free, and the miner and homeowner could split the cost of operation or some similar arrangement.

  5. As mentioned, it takes a lot of water (and pump power) to move significant quantities of heat, especially when the delta-T (required by heat pumps) is so small. Local loops heat with hot water using reasonable flow rates because the water is, indeed, “piping hot”, and the delta-T is large.

    But even that doesn’t scale so well beyond single-building size. The way to move a lot of heat is, of course, with steam, but that doesn’t work well to cool servers, because they don’t work well at the boiling point of water.

    And steam has its own issues. A local hospital is heated by steam from a not-so-nearby gas turbine co-gen plant. The pipe is 3 km long, and there is *always* someone working to maintain it somewhere along the line. For the lousy few megawatts they get, they must spend a megabuck a year in maintenance. But it means the hospital doesn’t need to maintain its own boiler and stack. But it also means the hospital is susceptible to disruptions of supply through that one long skinny pipe.

    1. Sweden has good fvärrvarme – ‘central’ (as in heated somewhere in the city) heating where hot (near boiling?) water with a green dye is pumped around. Seems to work well, the green dye looks cool when a leak springs up and helps to show it’s a heating leak and not hot water leak (which is also centrally distributed if you want it). On cool clear nights they’ll go round with a helicopter equipped with an IR camera mapping out the grid, making it easy to spot any leaks that haven’t yet broken out to the surface.

      1. An interesting solution, but a total-systems analysis would have to include the cost (in various senses) of the helicopter airtime (and resulting per-flight-hour maintenance costs).

        1. About $10/minute to fly a single turbine helicopter. Much more for a twin engine of any kind.

          I see no reason they aren’t doing this with a fixed wing or a drone, except it isn’t their money. Good enough for government work.

          Goes double for cops and ghetto birds. Take that money away, yesterday.

          1. I suspect a Helicopters very slow flight, tight turns, and hovering abilities when you want to circle and zoom in on the potential problems vastly outweighs the saving of flying fixed wing here – you spot something even in a very low stall speed fixed wing and you end up having to circle around a heap to actually get good views of it from every angle and be sure.

            Multi-rotor could work, but the very limited flight times make that less practical and I suspect you really want at least 3 operators with the quadcopter – and people usually cost more than equipment anyway, so might as well use the Helicopter with people on board and get better results much faster as you can fly for so much longer…

          2. Helicopters have to maintain forward speed for safety anyhow. ‘Dead man’s curve’ reflects minimum combined altitude and speed need to autorotate.

            They hover very little and orbit just as you describe.

            It’s just an endless money pit. The vast majority of that $10/minute is maintenance, not pilot cost. IIRC Helicopters are completely rebuilt every 2500 flying hours, by law. $Millions each time.

          3. HaHa, there is a rather big difference between the orbits of a fixed wing and helicopter, the most acrobatic planes still can’t turn in as short a distance over the ground, certainly not in a way at all useful for surveying said ground, and that is if you assume the helicopter has to keep moving forwards at all times, which it really doesn’t…

            There is a reason why so many places still use Helicopter for camera work, including all those private enterprises that only exist to make bucket loads of money, and its because they have a very useful flight profile and can actually stay in the air more than 10 mins (plus I expect its vastly easier to actually survey and film stuff if the operator can feel the crafts movements, use their peripheral vision/move their head to keep camera on target and the pilot has that same awareness of the surroundings – even the biggest best multirotor can’t give you that, just ask yourself how many times even great FPV fliers have crashed because they have no feel for the acceleration or awareness of anything not in that tiny cone infront…

  6. Data center waste heat is still way too cold for district heating. All of these installations require massive high-temperature heat pumps that have an efficiency of about 50% and are very maintenance intensive. You have to invest huge amounts of electric energy, making this concept way too expensive at average European electricity prices. Oh, and you’re still burning lots of coal and gas to generate that power.

    1. Remember, they were talking about Microsoft being involved and therefore Windows systems. Windows on those Intel CPUs are cooking hot when they are running more than a couple of things at a time. More than a couple of services running means they have to run Virtual Machines, data-center was mentioned too, so that’s Windows n times on Windows. Smoke’n

    2. The cold aisle in a new data center can these days be 30-35 C, some DCs wants to operate at 45 C, and that is on the “cold” side of the servers. On the warm side things can be 10-20 C warmer. (I am not envious of the technicians servicing these servers)

      Now yes, putting this heat into district heating pipes that usually operate in the 70-90 C range isn’t all that possible without a heat pump.

      But for local heating demands it is more than adequate, since the water might only reach 35-50 C and for a lot of heating demands this is more than warm enough.

      For underfloor heating one rarely needs more than 30 C, and air handling units can do decently at sub 40 C, though radiators is a different beast that wants a fair bit higher temps.

  7. We had the same here in NL.
    Microsoft made green promises as well. The greenhouses next door would be heated with the waste heat from their data center. End result: it was too costly to use it because of the low output temp and nothing came from it.

  8. Central city steam heating and later steam/electricity cogeneration has been used in may cities. Indianapolis (https://en.wikipedia.org/wiki/Perry_K._Generating_Station) still has the second largest steam plant in the US. Fortunately they phased out coal and switched to natural gas in 2020. May I live long enough to see nuclear fusion heating. Initially constructed in the 1890s it later added electricity generation when light bulbs became the new thing. Maybe when we have sufficient renewable electricity generation the steam district can be phased out with electric heating in individual buildings.

    We do have a steam clock that plays “Back Home Again in Indiana” next to the State Museum. https://www.youtube.com/watch?v=GuVZ1Pq7NBE Several other cities also have steam clocks.

    And after mentioning “Back Home Again in Indiana” I would be negligent to not reference Jim Nabors (Gomer Pyle of the Andy Griffith Show) singing at the Indy 500. https://www.youtube.com/watch?v=7wcgaRkHAWY

  9. I work in a datacenter where my office is located. This DC is part of a larger facility where I have my own hvac system for heating and cooling. During the winter time, I have no need to run the gas furnace as it gets quite toasty with just the server farm and switches running. Although in the summer time (Texas…) with the AC running at 65F the fans still spin up a notch….

  10. Low levels of waste heat, which is typically not captured today, could be used to help our friends “yeast” make “green” fuel for long term storage.
    Sunlight will re-captures any carbon dioxide released, so overall if designed well, it could be totally carbon neutral.

    It is a simple technology that has never really been pursued on a massive distributed scale, mostly due to complex laws, which basically make it a nightmare to even think about implementing.

    1. or we could turn said yeast into beer via heating greenhouses, then make a bunch of giant circular treadmills hooked up to a generator in the center and install them into gyms, thusly using beer to power the world! of course, the beer would have to be free or cheap and only get it while running the treadmill. 😋

      …talk about a convoluted pavlovian response to generate energy.

  11. Wow, no mention of (closed-loop) ground source heat pumps? You were so close, you even called out air-source heat pumps specifically! Even though air-source units have reasonable coefficients of performance, ground source units are now available with COP’s in the 20+ range (20x the heat transfer for a given input energy).

    GSHP units take advantage of the fact that the soil temperature in many areas is remarkably consistent throughout the year, once you get a meter or so below grade. When you need cooling in the summertime, you can transfer the heat from the hot indoor air and reject it to the relatively cool ground. Conversely, the ground temperature in the winter may be close to a tolerable indoor air temperature, only requiring the GSHP to extract a bit of extra heat to bring the temperature up to a comfortable level.

    Installation is the primary cost with such units, unsurprisingly, but depending on the local geology and site details different types of loops can be used to optimize the performance/cost ratio. Vertical wells are common, but there are also horizontal bores, trenches with ‘slinky’ loops, and a design called a “shallow heat rejector” that works well on flat areas. There’s also what’s called a “pond loop” that is, well, a loop of pipe in a pond; this is actually the most efficient design, if circumstances permit it.

    My department head in college was the chairman of the International Ground Source Heat Pump Association (IGSHPA), all of the students in my degree program were able to take the certified installer’s course for free. RIP Dr. Bose.

      1. The other advantage is that the CoP is much less dependent on outside ambient (and hence air) temperature, whereas ASHP’s start to struggle below -10C and most don’t work below -20C. Not very common here in the UK, but exactly when you need the heating most, in a cold snap, it struggles and hence needs more electricity, or packs in completely. Multiply that extra demand by 30 million when they are all rolled out….

    1. Did a DYI version of this in Wisconsin. Cooled my home from a exchanger loop in a very cold springs piped though furnace A-coil. But heating was through cast iron radiators (salvaged) and solar thermal+outdoor wood boiler booster for 6weeks per year average.

  12. At my company we wanted to use 100% renewable energy datacentres. Ironically, Azure isnt and they’ve had the lame target of 2025 for a while now. We found plenty who are run on 100% renewable.

    I sense a bit of greenwashing

  13. the waste heat can be offset by burying the lines well below the frost line. most locations maintain roughly 55 degrees F at or around 6 feet underground. given you can calculate the year round temperature, an energy loss calculation can be made. as for waste incinerators, if the energy differential is high enough, it may be better to heat water for steam as to turn it into electricity to be easily sent into the grid. data center water will never get that hot. I can see using this to hook the line directly into the boiler or hot water heater of large facilities so that not as much energy is used to heat up what would otherwise be cold water, of course that is dependent on the distance. another use case could be just to run the lines under roads, so when its cold ya simply open the lines to de-ice the roads.

  14. This is nothing new, Swedens most controversial Internet provider (controversial in the way that they do not track or cencor any content or users, so there have been Illegal activitys on their servers) is heaing a church.
    The server hall is located in a former military bunker, and the church is on top of it.

  15. Hm, no I don’t think this is a sound idea. Much more practical just to capture heat from the sun, and the heat rejecting pumps, used for refrigeration, just reject heat to the atmosphere. The amount of energy you are going to save by connecting the output of the refrigerators to the input of the heating heat pumps is minor, it would be expensive and heat from refrigeration is already rejected into the building envelope in most cases

  16. It is batshit insane that people focus so much on solar electricity and yet use natural gas for heating. 1 kWhr is 3.6 megajoules. One cubic meter of natural gas yields about 40 megajoules upon combustion. A square meter of solar collector, optimally oriented but not moving, gets between 2 and 6 kWhr per day of energy hitting it.

    Clearly there is vast potential. That’s like between like 1/9 and 1/2 cubic meters of natural gas that just falls out of the sky.

    Capturing it should be dirt cheap. Some translucent insulator like freeze dried gelatin, which is a type of aerogel and is already used in transport trucks, could work better than glass or polycarbonate greenhouse glazing.

    Even just triple glazed polycarbonate window glazing is about $20 USD per square meter. The collector would pay off in one or two seasons, probably.

    1. Yes. South-facing windows do that as well. (West-facing windows do that also, but only during the summer). It’s called passive solar design. There’s been a lot of experimentation on solar collectors since the 1970s, but when considering the conflicting needs of solar capture, power distribution to the home, insulation, regulation, and of course weather protection, my personal assessment is if you’re looking at your roof, electric solar + heat pump are worth evaluating. Modern solar panels are incredibly efficient, and they also produce electricity when you’re not heating. And if you’ve done proper insulation, you don’t need that much power for heating to begin with.

  17. I will have to respectfully disagree with almost all above comments ;-)

    The future is in insulating, not heating.

    To illustrate that point, let me describe a house I designed and self-built over a decade ago in France (continental climate). With the walls made of straw bales (50cm thick insulation), a compact shape (octagonal), and a thorough passive solar design (30% of the floor surface in south-facing triple-glazed krypton low-e windows with a low solar ratio, outside shades), mechanized ventilation, top-performance windows and average air-tightness, we were able to save on the heating almost entirely.

    We went through the 1st winter (-15°C for a few weeks) on only a refurbished 2kW electric radiant heater on a wall plug, for the whole 150m2 house, more or less making it to “passive house” standards.

    My conclusion being that, if you take good care of the insulation and thermal design, heating becomes a non-issue. It doesn’t really matter how you heat, because you heat so little.

    Fast forward to this year, I moved countries and am having a house renovated in Shanghai (hot summers, damp winters). With only 10cm of insulation outside the walls and roof (that’s around 10cm more than anyone else on the block), I should more than balance my standard-issue heat pump’s consumption with my roof’s worth of solar panels.

    The saying goes that a house in Geece consumes more per sqm than a house in Sweden (for heating & cooling). That’s a building codes problem, and an insulation problem, not a machine problem. As we hackers tend to relish on complex machines for moving heat around, it’s easy to overlook the designs that do away with the machines.

    One more tought: if you work >30km away from home, then your car commute is consuming more power than your house; so living near your workplace trumps the perfect house. And housing blocks are far more performant than private houses (power dissipation *and* cost of insulation are both proportional to the outer wall surface). So while we might relish on beautifully-designed passive solar houses, urbanism is where it all begins.

Leave a Reply to Jenny ListCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

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