Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Featured HVAC Using the thermosyphon effect for a floor water heating system

  1. Dec 10, 2018 #1
    I've moved to a "new" house - built about 350-400 years ago. Firewood in a stove kind of works for heating a few rooms, but not the full house. So far it seems like the grid power tends to die about twice a month. Temperatures is expected to reach -25C (-13F) for the coldest couple of weeks every winter. Hence I need some kind of backup power and heat source.

    The house is located in a small hill side, facing south-west. From about 15th of November some near by mountains is blocking the sun, so solar energy will be zero for at least 2 months/year.
    On the property trees grow faster than I'll ever be able to burn them, so there is unlimited access to firewood.

    Energy required to heat the house seems to be about 3kw at 0C outside. That should indicate the need for 7kw for the coldest days.

    I have an idea: What if I install water tubes in all floors and build a wood burning waterheater in a "stove house" somewhere down hill from the house. Could this be made in such way that water will circulate from thermosyphon effect, without any pump (needs to work even when grid power dies). A convenient place to put the stovehouse will be 16m from the house. Then the base of the stove will be 5,5m below the floorlevel. Could this be enough to make the thermosyphon work? Will it be able to transport 7kw of heat? (What is required in order to make such system work?)

    I'm thinking of making the stovehouse burn logs, maybe 2m long, maybe some kind of autofeed. Also I'm thinking of storing logs on top of the stovehouse to let the waste heat help to dry the logs. Hence the time consumed to deal with firewood should be minimized.
     
    Last edited by a moderator: Dec 10, 2018
  2. jcsd
  3. Dec 10, 2018 #2

    anorlunda

    Staff: Mentor

    In the 90s, there was a surge in popularity of external wood-burning boilers. They burned logs, and piped steam to the house for heating. I believe that they used the circulation from steam pressure rather than electric pumps or fans, but I may be wrong. My neighbor had one.

    But their popularity declined because they created so much smoke and localized air pollution, that many towns changed the rules to forbid them. On some days with temperature inversions, my neighbor's boiler blanketed the neighboring farms with smoke.

    Be advised that designing a boiler is no DIY project. There are many codes and laws because of numerous boiler explosions in the past that kill people within a large radius. To do that, you must buy a boiler certified to meet all applicable codes.

    My father in law once built a 3 story house with hot water heating and baseboard radiators. The water circulation was done entirely by "natural circulation", no pumps. A google search for natural circulation hot water heating provides lot of hits.
     
  4. Dec 10, 2018 #3
    In theory, this should work. There is just one practical issue: since this method of heating is a bit out of being common right now, I guess it will be a bit difficult to find somebody who can and: will build it...

    Masonry heaters, thermal mass stoves, rocket mass heaters has a big popularity these days, maybe it would be better to look into them instead.
     
  5. Dec 10, 2018 #4

    berkeman

    User Avatar

    Staff: Mentor

    One other safety consideration is being sure that carbon monoxide from the combustion will not gather in your home. Is the breeze typically downhill during the times when you will be doing the most burning (I would guess so, given your description of the cold hillside environment without much sunlight). 16m sounds like a good separation, and it may help to be sure that any air inlets into the home are on the uphill side, away from the stove house. It may also be smart to put a couple CO detectors in your home, just to be sure. They are inexpensive and run on 10 year life batteries.
     
  6. Dec 10, 2018 #5
    Given that OP wants to use the floor as a heating surface I don't think there is much reason to worry about boilers and steam o0)
     
  7. Dec 10, 2018 #6

    berkeman

    User Avatar

    Staff: Mentor

    Because that would burn their feet?
     
  8. Dec 10, 2018 #7

    russ_watters

    User Avatar

    Staff: Mentor

    The system can be unpressurized water to avoid hazards associated with pressure, but I see some difficulties:

    7kW seems low, though it would depend on the size of the house. My townhouse has a 20kW heater.

    Circulation by convection is unlikely to work for this application. for one thing, you need the water to circulate relatively fast for even heating.
     
  9. Dec 10, 2018 #8

    Klystron

    User Avatar
    Gold Member

    With the original questions answered above about heating, consider alternate power sources.

    Create a local grid node. Store electrical energy similar to how you store dry wood for burning. Consult local electricians. Monitor usage; estimate future requirements. Filter house electricity through a UPS (uninterruptible power supply) system. Batteries in the UPS supply power during municipal lows. If you add a generator, an automatic transfer switch (ATS) can sense prolonged outages, start the generator, and switch to the local grid.

    Design your electrical upgrdes to favour critical systems in emergencies such as communications, pumps, refrigeration. If and when you add power sources such as windmill, solar, biomass; you can combine these sources with municipal. In western USA where the sun shines bright <humor>, smaller households equipped with passive solar panels generate enough electricity during a summer day to charge their UPS and trickle power back to municipal; with the inevitable corollary that the household must be air-conditioned.
     
  10. Dec 10, 2018 #9
    That's an often missed point. The peak power of a heating system is not based on the ability to keep an acceptable temperature on the expected coldest day, but on the ability to heat up the house within acceptable time on the coldest day. Quite the difference I would say...

    These kind of heating systems can work, but careful engineering is needed. In the house of my parents we had one such system with ~4m distance between the hot and the cold pipe. The main pipe was 2.5" in diameter. It took an hour for the radiators to get warm. (Yes, it was an 'open' system on atmospheric pressure.)
    Pairing this kind of setup with floor heating... Well, I would not do that.
     
    Last edited: Dec 10, 2018
  11. Dec 10, 2018 #10

    russ_watters

    User Avatar

    Staff: Mentor

    Yeah, mine is definitely sized for rapid warm-up, but even then it's probably 50-100% extra, so still a lot more than the OP. And my house is relatively new, well insulated and only has 3 walls and not many windows.

    ...for AC, they don't generally oversize, and mine takes forever to cool.
     
  12. Dec 10, 2018 #11
    7kw is the required constant heat radiation inside the house.

    16m from the house is actually more than the distance to the neighbors chimney. Although I'm not planning to build the stove house as high as the neighbors house. I'm not too worried for the smoke as I'm thinking of using a stove that burns hot and clean. They say a perfectly designed rocket mass heater emits only CO2. How hard can it be to make it perfect? (Probably extremely hard)

    "natural circulation" was a nice term to search. Actually better than "thermosyphon".

    I'm thinking of designing the stove similar to a rocket mass heater - let water be the mass. I have to build it my self, as such thing cannot be bought in Norway (where I live). That (fire) is also one of the reasons for having this in a building some distance from the house rather than in the basement. (Ease of loading the logs, bugs that might live in the logs, no need to carry logs up the hill, are other reasons for not bringing the stove into the house.

    I was considering steam, but I don't want to create local steam clouds. I want this all to be a closed loop system. So, IF I'd go with steam, it would have to condensate back to water somewhere. The steam route would first power a steam engine. Steam engine exhaust then goes into coil inside a large water tank inside the house. Here the steam should condensate and return to the boiler. The steam engine should power a pump that circulates the water from the tank into the floor heating pipes. I think this would turn out quite expensive, and steam is a significantly higher risk level than hot water. The steam system would have a major advantage though - it would generate electricity. I could cancel the grid power!

    The hot water thermosyphon approach seems safer and way cheaper. The only major risk, as I see it, would be if the water circulates too slow, or circulation stops for some reason, it will boil. This system will require a pressure valve that will let water/steam out if boiling happens. From how I understand thermosyphon the circulation will be faster the more heat/cooling is applied to the loop. I can apply more cooling to the loop by letting it run outdoors with little or no isolation. In my mind it all comes down to scaling the various parts of the loop properly. How to scale them, I have no idea.
     
  13. Dec 10, 2018 #12
    I once helped a friend design a heating system that transferred heat from the wood fired boiler to an open storage tank by thermosyphon. The storage tank held 1100 gallons of water. That was the amount that would allow him to build a fire in the evening, and heat the house until the next evening at 0 deg F. At design temperature of -20 deg F, he had to build a smaller fire in the morning. The wood burner was located next to the tank as shown in the simplified sketch below.
    PC100049A.jpg
    Unfortunately, he chose to build the tank from concrete block and seal the inside with epoxy. He had to fill the tank three times and drain it twice to find and fix the leaks, but finally got it to hold water. He then learned that the thermosyphon effect stops when the bottom level of the hot water in the tank gets down to the level of the heat exchanger in the boiler. The top half of the tank was hot, the bottom half stayed cold. He only had 50% usable heat storage. The next learning experience was that hot concrete blocks expand, cold concrete blocks do not, and the cold concrete blocks at the hot/cold water interface fail in tension, causing cracks and new leaks. The tank was insulated on the outside. He fixed the leaks (third drain down, fourth fill) and added a circulating pump. The rest of the four zone hydronic system worked properly. The gas fired backup boiler worked properly.

    The maximum water temperature in an open system is about 210 deg F. When the tank reaches that temperature, there will be evaporation. The top of the tank needs to closed tight against vapor, but such that it will not hold against pressure, along with an open vent to atmosphere.

    The usable heat storage is the total amount of water times the maximum temperature difference. The maximum temperature is 210 deg F. The minimum temperature is the minimum temperature that heats the house on the coldest day. Most hydronic systems are designed for a water temperature of 180 deg F. A hydronic system designed for 180 deg F water would have a maximum temperature difference of only 30 deg F, while a hydronic system designed for 120 deg F water would have a maximum temperature difference of 90 deg F.

    Thermosyphon flow can be calculated. Start with the supply side temperature and the estimated return temperature. The density difference between the two temperature times the vertical distance gives you the total head (pressure difference). From that, you can calculate the flow rate. The heat transferred is calculated from the temperature difference and flow rate. The return temperature is calculated by standard heat transfer calculations. Iteration is necessary.
     
  14. Dec 10, 2018 #13

    Tom.G

    User Avatar
    Science Advisor

    Don't forget antifreeze in the system, otherwise you can't leave for a couple days without draining the whole system.
     
  15. Dec 11, 2018 #14
    These kind of systems with natural circulation phased out exactly because they are hard to design (properly) and they requires too much material (pipe diameters are usually several times larger than for new systems with forced circulation).
    Floor heating systems requires careful design and control to be effective and comfortable.
    You want to pair them together, without real control (so it can be operated without electricity).
    I think this project (as it is now) is just 'asking for trouble'.
     
  16. Dec 11, 2018 #15
    I kind of know how to make trouble. I'm asking for control ;)

    Jrmichler, your experience is valuable. Did your friends system not create steam in the boiler?

    Delta T in the loop can possible be quite large. If the loop is created like this: Coil around the stove - isolated pipe to the house - coil in hotwater tank located high up inside the house - floor heating - outside driveway heating (the driveway has a 21% slope at the steepest and a sharp turn) - return to the stove. This loop should make it possible to have a delta-T maybe as large as 80C (176F). With the hot water tank high up in the house, the total high difference should be near 10m, although the return path from the lowest part of the driveway to the stove will need to go 3m up, 20m horizontal, 3m down.

    I guess some kind of control could be added - battery powered. Then, how much energy will a pump require? (How to calculate?)
     
  17. Dec 11, 2018 #16
    In case you can't stop immersing yourself in dubious DIY projects around floor heating, then you would do better with a diesel based heatpump system. Might be hooked as an emergency generator and by utilizing the waste heat it is possible to aim for a nominal 100+ % efficiency.
    Still plenty of weak points, but at least there is hope.
     
  18. Dec 11, 2018 #17
    Heatpump (air - air or air - water) doesn't work well with outside temps below 15C (5F). That is, you get a COP<=1. If you're lucky the generator provides 30% of the diesels energy as electricity. You would be better of circulating the engines coolant via the house than using the electricity it produces. Or even better, put a diesel driven heater in the house.

    Half the purpose of this project is to make use of the resources that exists on this property. I also was thinking of making a water driven generator. It would probably work very well, except the the water also stops moving in these temperatures.
     
  19. Dec 11, 2018 #18

    anorlunda

    Staff: Mentor

    If you're really interested in the simple solution; multiple wood stoves. But is sounds like you're not searching for heat, you're searching for a big project.
     
  20. Dec 11, 2018 #19

    Klystron

    User Avatar
    Gold Member

    When I posted about battery backup and local power sources, water generation was the first method I considered after diesel. It seems an ancient even medieval method, but when the water motion generates electricity instead of grinding flour or pounding ore...

    Wind turbines or old-style windmills could add another unpredictable weather-dependent source to your energy independence.
     
  21. Dec 11, 2018 #20

    russ_watters

    User Avatar

    Staff: Mentor

    I don't know how long these power outages go, but a battery operated pump might get him through. If not, a small gas generator would be enough for a pump and to keep the lights on in the house.

    So yeah, the intermittent power is the problem I'd try to solve, not the loss of pump problem.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?