GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Guest Blogs

Goodbye Radiant Floor

An in-floor hydronic heat distribution system may be overkill for a tight, well insulated house

Radiant floor heating systems don't make much sense for superinsulated Passivhaus buildings, which are often heated by a few ductless minisplit units. [Photo credit: Amanda Devries]
Image Credit: Amanda Devries

[Editor’s note: Roger and Lynn Normand are building a Passivhaus in Maine. This is the second article in a series that will follow their project from planning through construction.]

Goodbye radiant floor. Though we never really knew you, we are sad to see you go away.

Our original vision included radiant floor heating. Though we’ve never lived in a home with radiant floors, we had read numerous articles celebrating its comfortable, silent, even heat with no blowing air currents. According to many articles we’ve read, it’s also a good match to the output temperature of a solar thermal system, which is some 30 degrees below the input temperature for baseboard radiators.

Dad’s house in southern Maine has oil-fired, cast-iron baseboard heat, and we’ve always admired how comfortable the house felt, even with the thermostat set several degrees cooler than our own home with its natural gas-fired forced-air heat. We figured radiant floors would be even better than baseboard heating. We would not miss seeing dust and pet dander (we have one dog and two cats, and frequently host other dogs as part of a vacation dog-sitting exchange) wafting in the air currents whenever the heat comes on. We would have concrete floors for both thermal mass and embedded PEX tubing for radiant floor heating.

Ductless minisplits are all the rage

Alas, it seemed most articles we read about heating/cooling a Passivhaus or other very tightly insulated homes relied on ductless minisplit systems. These systems consist of a condensing unit outside, connected via a small copper tube that circulates refrigerant to an indoor wall-mounted blower unit. The indoor component is touted as being whisper quiet. Some manufacturers claim their units can heat effectively down to a very cold -13 degrees Fahrenheit outdoor temperature.

One such article was the one by Martin Holladay in the February/March 2011 issue of Fine Homebuilding, in which the author writes that radiant floor heat is a good match to heat poorly insulated homes, but overkill and a waste of money in a small tight house that needs only a small annual heat load. At some 4,000 square feet equally divided between a main floor and a basement, our planned home isn’t small, but it will be very tight.

We’ll await to see the results of the heat load analysis. At this point, however, we’re figuring that, sadly, radiant heat probably won’t make sense. We’re still not enamored with the forced hot air approach of a minisplit system, and are dubious of its prowess when the thermometer plummets outside. On the other hand, it would provide an inexpensive way to provide air conditioning in the summer, which is something Lynn absolutely wants. I’m not a big AC fan for Maine, though I will admit there’s been more uncomfortably warm summer evenings when we’ve visited Dad, which has no air conditioning. Lynn says when the temperature and humidity get high, it’s time for AC, no matter where you live.

Just a few radiators?

We’ve started thinking that perhaps a small forced hot water system with a few carefully placed baseboard radiators would make sense for heating our new home in the winter. It would still be radiant heat, perhaps fired by a tankless water heater or very small boiler that takes preheated water from a solar thermal storage. We just hate the thought of giving up some form of radiant heat.

Stay tuned!

The first article in this series was Kicking the Tires on a Passivhaus Project. Roger Normand’s construction blog is called EdgewaterHaus.


  1. user-1140531 | | #1

    Heat Distribution Systems
    There are several points consider:

    In my opinion, a clear distinction can be made between radiant heat and forced air, with radiant being deemed superior. However, I am not referring to radiant heat as being limited to warmed floors, as is the common assumption.

    Radiant heat transmits heat directly to the occupants without the need to wash them with air. It also allows the air temperature to be lower than that of air being used in a forced air heating system. Radiant is more comfortable, and it is not just related to the warming of people’s feet by a warm floor of the so-called radiant system.

    However, both types of systems create a blend of all three mechanisms of heat transfer, so for practical purposes, there can be no direct comparison between radiant heat and forced air.

    The so-called radiant systems are widely assumed to be warmed concrete floors. However, they transfer heat to people’s feet by conduction. They transfer to the ceiling by radiation. So the color and texture of the ceiling play a role in either reflecting heat back down; or absorbing the heat, warming the ceiling mass, and re-radiating heat back down. If the ceilings absorb heat and re-radiate it, the insulation above them is especially critical.

    Heat radiating upward and downward will warm all the objects in the room. As heat spreads by conduction in those objects, it reaches surfaces that face the walls, so it will radiate to walls from those surfaces, and heat the walls. The heated walls will re-radiate back and heat more of the objects that present larger surfaces to the walls. Some horizontally radiated heat will go right out the windows through the glass and heat objects that it strikes outside.

    All of the objects warmed by radiation will heat the air that contacts them by conduction, and that will cause the air to transfer heat to other objects by convection.

    And the warm floor itself will heat air contacting it, and cause that air to rise and transfer heat by convetion to the ceiling, walls, and all objects in the room.

    In a house heated only by a forced air system, all of the air-warmed objects will radiate heat in all directions, and transfer that heat to any surface that is at a lower temperature. So radiant transfer is actively working even though the primary system is forced air.

    The ideal radiant heat distribution would be to radiate from the floor, ceiling, and all the walls. This would apply the radiant heat to all parts of a person’s body. But such a system of radiating surfaces is impractical to produce. The floor is the most practical radiator.

    However, a system radiating from the walls is also quite practical. Traditional radiators are actually largely convectors designed to heat air and cause it to circulate. The true wall radiator would be a flat panel tank or tube grid panel, which would transfer more by radiation than by heating air by conduction and causing it to convect. These flat panel radiators are becoming popular, but the ones I have seen do not cover entire wall surfaces, which would be ideal for radiant distribution. But, unlike floors, walls are interrupted by doors and windows, so it makes sense to spot and cluster the radiant sources on a wall where needed rather than try to cover the wall area 100%.

    Baseboard heat transfers largely by convection of the air that is warmed in the baseboard elements. However, that rising air washes up the walls and comprehensively heats the walls. The walls then become broad radiators, which warm all objects in the room by radiation traveling horizontally. So baseboard heat might be said to be more of a radiant heat system than is a system using conventional radiators at spot locations functioning primarily as convectors.

    In a tight, superinsulated house, any type of heat distribution is going to work better than in a more typical house. With slower heat loss, the heat has more time to spread out by radiation, conduction, and convection to all the mass inside the house. This results in even heat, which can then be set as low as is comfortable. There is no need to try to compensate for natural cold spots, because there aren’t any.

    I regard radiant floors as a sort of novelty. They have some benefit in added comfort, but some disadvantage in the large mass that responds slowly. My ideal system for a superinslulated house would be either baseboard hot water, or flat radiator hot water, or a combination of the two. In any case, it would all be on the walls.

  2. wjrobinson | | #2

    Roger and Lynn, nice project.
    Roger and Lynn, nice project. Radiant heated master bath tile may be doable. Limit the heated area and set up to warm the bathroom and adjoining bedroom via proximity. Use with the splits and a gas or pellet or wood stove coldest days backup. That's my present PGH HVAC model plan.

    Ron, your post, filibuster practice?

  3. LukeInClimateZone7 | | #3

    I Second AJ Builder (and good summary Ron!)
    Good idea AJ,
    Whether or not radiant floors are expensive and overkill is a subjective question. Granite countertops I also find are expensive and overkill (and dull my knives) but lots of people want them because they think other people want them at resale. So be it.

    To AJ;s idea: I find the important expectation that needs to be met for radiant floors is the notion of Toasty-Toe. In a superinsulated house, a whole floor radiant floor system would necessarily be much cooler. Temperature is sorta the 2nd law indicator of "Quality" of heat, and when it comes to my preferences, I like a minimum 80 degrees (strictly personal preference) of Quality. Anything less than that is like taking a lukewarm shower. In the realm of 1st world problems, this is a big one.

    Limit the area and focus on the habitated areas like kitchen/bathroom/ and living room (or wherever you'd likely hangout). For response time, go with a low-mass solution like an electric mat, or a Warmboard kind of solution.

    As far as capacity: again, a paraphrasing of the 2nd law of thermo-- Heat is the sewer of energy. It's up to you whether or not you install a system that meets the load 98% of the time, or a couple of simple systems where one meets your needs 80% of the time, and another (cheap) system meets your needs the other 18 or 19% percent of the time. In a passivhaus, maybe you could just leave the lights on an electric hour, or have a hair-drying party? (the extra-humidity might be nice too!)
    In the realm of Thermodynamics-- 70-80 degrees really is low-grade heat. If I'm going to be using a high quality energy input like electricity, I'd prefer it to go someplace with significantly higher productivity like feeding some kind of process that I'm doing. Maybe I kick on the XBOX for a couple of hours, or make some chicken soup on the range to eat the next week. Why use Evian for toilet water, when you can first make something with it first, and then piss away the scraps into a toilet-greywater system (please pardon the rather obtuse simile-- I hope it made even a little sense).

    On one of our projects-- they are electing to go with a 15 person Dance-Party anytime it get's really cold. Dance/exercise first, then the house coasts through the night more comfortably on the waste heat.

    Personally, I find bundling up in pajamas and a nice hot cuppa works well for me in the wintertime.

    The point is-- for heating you have virtually limitless options. It is entirely up to you which solutions you would find in that subjective space-intersection of 'elegant' and 'practical'.

  4. Expert Member
    Dana Dorsett | | #4

    Tankless & radiators is too much
    The minimum-fire output of most tankless hot water heaters is several times the design-condition heat load of a PassiveHouse (and more than 10x the average heat load.) If you have thermally-massive radiators (let alone a massive radiant floor) you end up with uncomfortable overshoots in temperature running with water at domestic-hot water temps, further complicating a tankless approach. To stay comfortable with a water-heater + radiant-baseboard approach, use a condensing tank-type hot water heater and a heat exchanger to isolate the heating end, and use a thermostatic mixing valve on the radiation loop to run at a lower temp. (The actual temp will depend on your heat load analysis and how much radiation you have out there, but in a PassiveHouse it wouln't take a lot of cast-iron baseboard to be able to deliver design-day heat with 100F water.)

    But when it's said and done, a mini-split will likely cost less than a hydronic baseboard/radiator solution, and provide comparable or BETTER comfort, and air conditioning:

    The reason radiant floors don't make sense at PassiveHouse heat loads is that the temperature of the floor only needs to be a degree or two above room-ambient to deliver the design condition heat to the house, and most of the time it will be less than 1F above room temp. While a 71F floor in a 70F room is comfortable, it's hardly the cozy-warmth you're looking for. With radiant-baseboard (not fin-tube) or panel radiators you can run with warmer water temps and get at least a hint of the warm-glow effect, but it's still nothing like what you experience in code-min or older homes with radiators. Mini-splits always put out a warmer-than-body temp flow, and a minimum compressor & blower speeds have BTU/hour outputs more in line with the actual heat loads experienced in a PassiveHouse. They have very stable temperature control and you won't get signficant under/over shoot of the setpoint temperature. It's not the cozy warmth of a radiator, but since you won't have cold floors, walls or windows absorbing your radiated body heat in a PassiveHouse, the comfort factor of a mini-split's very stable air temperature may be more aptly thought of as an absence of discomfort, which is different from the cozy-glow of a radiator in room with cooler surfaces elsewhere. There's never a wind-chill effect or drafty feel, and when idling along in their low range, the interior heads are quieter than most refrigerators. The fact that they're fully modulating and can be had in sizes within the design-condition heat load of a PassiveHouse make them a very appropriate solution.

    At the current price (or even the 5-year average price) of heating oil, you might think about talking your dad into adding a mini-split to cover most of the heating of a large zone in his southern ME house too- it'll pay for itself in reduced oil use in remarkably short years. Many have had to resort to maintaining much lower temperatures (and lower comfort ) in recent years to make oil heat affordable, and there is nothing to indicate that the price of heating oil should stabilize or fall over the next decade. Heating with mini-splits is about 1/3 the cost of heating with oil in a southern ME climate. Even if sized to only 1/2 the design condition heat load a mini-split can cover the bulk of the seasonal heating, and cut the total heating bill roughly in half.

    1. user-3697170 | | #51

      Question about slab temperature in well insulated buildings, in response to Dana Dorsett's comment. "The reason radiant floors don't make sense at PassiveHouse heat loads is that the temperature of the floor only needs to be a degree or two above room-ambient to deliver the design condition heat to the house, and most of the time it will be less than 1F above room temp. While a 71F floor in a 70F room is comfortable, it's hardly the cozy-warmth you're looking for." Agreed!, but what will the floor temperature be if you are heating with heat pumps as opposed to radiant tubing? Assuming slab on grade construction with a fairly aggressive thermal envelope, R-30 below slab, R-40 walls, R-60 roof, 25% wall glazing at u-.18, might the slab be 5 degrees or more below air temperature? Scott Gardner Burlington, VT

      1. Expert Member
        Michael Maines | | #52

        Scott, in your example the surface of the floor will be very close to room temperature, 1-2 degrees cooler than if it had in-floor heat. Your feet are somewhere around 88°; they aren't going to notice the difference between a delta T of 17° vs 19°.

  5. DC_Eakin | | #5

    Moving Air Always Feels Cooler
    This is a common complaint among people who are used to hydronic or high-temp forced hot air heating systems when they switch to a heat pump system. "The air feels cold!" As long as the air flowing over your body is at or near body temperature, it will "feel" cooler due to perspiration evaporation. If you want to construct a solar-heated hydronic wall radiator for max cold days, you will probably find that the inherent costs of install will exceed an electric radiant wall panel. How about planning for passive solar gain in the lower-level slab? Or (ironically, in a Passive House) an incandescent bulb would provide inexpensive direct radiation when/where needed.

  6. wjrobinson | | #6

    Dana... wrong my man, small
    Dana... wrong my man, small areas of radiant... work. The difference is just sizing down the square footage of warmed floor for PGH builds. I think you are one of the best posters in regard to HVAC but you are not thinking out of the box and in future tense. Stuck in the past my man...

  7. user-1127834 | | #7

    Solar Air pannels to HX and space comforts
    Common well designed (Solstar tm 1980 copies) AIR-Solar collectors of just 2 x 10 x 3ft baked a 1700 sq ft pantek 4" foam wall x 8" ceiling (half vaulted) over a 6" insulated flooring over cold crawl---
    baked- the home well until a very cold Jan-Feb week where 2 kw of a 3-stage 6kw heat strips engaged once in a while. Adding ERV added to hit a 4kw usage a few days, b/c house was too tight.

    But HW was wanted from other means too through summer so Solr-AIR-to-HX to HW systems were then commonly installed with the 1980-1985 tax credits being 50% (then).
    So perhaps there are still some very practical (and qualified for Solar crediting) - reviewing HW needs with possibly a best AIR SOLAR to HX and any small radiant HX to a floor.

    no one ever need "feel cooler" in properly distributed , highest Efficiency Mini's or any forced air...

    Many homes with 6" r(4) flex joist-ducting of runs to 25 ft (in-box) lengths on 2x12' registers, and a few more for 3- 3.1/2 air changes on highest stages designed, are UN-notibly "air-washing". It can feel just like a BBD Radiant.

    Basement floors with well-insulated pads under carpet have shown NO NEED for radiant heat, with only a vapor barrier in under the concrete, unless the slab area is for sleeping rooms. Those designs have the warm air just billowing straight downward to the floors, comforting all sitting nearby.
    Even there Air-Solar -to-HX Radiant tubing (a little) can be more cost effectively used, I would believe.

    GeoThermal 100% variable programming of set-bonnet temps like 98-100'f selected keeps 'heat-pumping' warming at any registers for 98% of the people.

    High efficiencies, like MINI's are set still near 100 deg.f at the furnace/blower discharge. More runs simply make it like radiant comfort over MINI's. The GT Priority 100% HW may be sized first and only use minimal staging DOWN for space heating and cover all cooling and ERV-of cooling to make 100% HW, too, loop pumps off in that Cooling mode. Slower 325-340 cfm/ton for DeHumidification can be programmed. Various zone "idle" cfm can keep near perfect peace of mind throughout.

    A ground loop for the GT of matched to building loads + HW Priority is crucial for say above 35-deg entering fluids from the ground loop at 3GPM per equipment ton, usually in 50-deg.f Earth-Coupling.

  8. GBA Editor
    Martin Holladay | | #8

    Response to Ron Keagle
    You wrote, "Radiant heat transmits heat directly to the occupants without the need to wash them with air. It also allows the air temperature to be lower than that of air being used in a forced air heating system."

    Neither statement is true.

    Except on the coldest day of the year, radiant floors are usually off or operating at a low temperature. (If a radiant floor has a constant circulation pump, it is a true energy hog, wasting lots of electricity.) The floor isn't really a radiant source for the occupants in the room, since the floor is usually at room temperature. Even when the floor heats up, what it is mostly doing is heating the furniture and ceiling by radiation and the room air by convection. (And occasionally, heating a human being by conduction, if the human is barefoot.) Unless you are lying naked in a mesh hammock, with the radiant floor on full blast, you aren't being heated by radiation.

    I have never seen a study that shows that homeowners with radiant floor heat set their thermostats lower than homeowners with other types of heat. One Canadian study that looked into the issue found the opposite -- homeowners with radiant heat set their thermostats just a little bit higher than their neighbors.

  9. Robert Swinburne | | #9

    radiant ceiling heat
    I live in a small, poorly insulated house with radiant electric heat in the ceiling. This is only a backup to our woodstove although we run it in the bathroom more often. the ceiling heat makes us feel like we are having a hot flash or a head rush when it is on. Our feet remain quite cold.

  10. Expert Member
    Dana Dorsett | | #10

    Response ot aj
    It's not that small-area radiant doesn't work, it's that small radiant doesn't work well at the min-fire output (or water temp) of a tankless in a house with a design-condition heat load of less than 10KBTU/hr. It's hard to rationalize the cost a radiant system for only the most modest of comfort improvement of comfort in a house that by passivehouse specification has no cold interior surfaces, and an a very even radiation temperature of the walls/windows/ceiling/floor. Sure you could heat the place with small panel radiators or a much reduced radiant floor area, but it doesn't really increase comfort much except at the outside design temp or lower. How much are you willing to spend for that tiny improvement in comfort for the 1% of hours, the bulk of which occur while you're in bed?

    If you're willing to spend the money for that small luxury, a condensing tank heater can be tapped at much more appropriate BTU/hour rates than a tankless without resulting in short-cycling or temperature overshoots, but you're still stuck with paying for the (very necessary in New England climates) air conditioning. The heating/cooling balance point of a PassiveHouse tends to be in the mid-40s Fahrenheit, and while cooling can be achieved with ventilation during the shoulder seasons, some mechanical cooling (particularly dehumidification) is necessary for at least 3-4 months of the year in a superinsulated house, which tips the financial argument toward combining heating & cooling into the same ductless mini-split unit.

    The (Shrewsbury MA) Beaton PassiveHouse near me runs mini-splits at least half as many hours and kwh) in cooling mode as it does in heating mode due to excess plug-loads lowering the balance point temp:,%20Paul%20-%20Ad-Hoc%20Passive%20House%20-%20The%20Beaton%20Residence.pdf

    In a PassiveHouse heating & cooling power use is less than half the total power use, and the real future-view for the HVAC systems in such a house is to keep them cheap & efficient, since the method of heating has very little impact on the actual comfort realized. It's the "just reduced the size" hydronic heating approach based on experiences with houses of more conventional heating loads that is proving to the real "inside the box" thinking, since it doesn't get down to the "what for?" The hydronic approach buys next to nothing on comfort in a house with such even radiation-temps on all surfaces, and adds expense.

  11. wjrobinson | | #11

    Dana, we agree. Heat a PGH
    Dana, we agree. Heat a PGH home with a split as primary. For the love of Christ and good beer and warm toes, add a section of warmed tile in the bath for the morning routine. We have done this and customers love it. Very little cost to install and to run monthly. Either electric mat or PEX loop off water heater.

    I am not talking about a full house radiant floor heat system with boilers and 2000 feet of PEX.

    I do like PEX in cellar floors that are living space. We have one here and you wouldn't know it was cellar space as there is no cold floor discomfort, just the opposite. Warm comfortable space as if not even basement space. And you know most basements feel like basement space (not so nice).

    And Bruce Brownell definitely knows how to pour concrete that is comfortable to live over.

  12. user-1140531 | | #12

    Response To Martin Holladay

    With all due respect, both of my statements that you say are incorrect are correct.

    I said that radiant heat transmits heat directly to the occupants without the need to wash them with air.

    I don’t see your logic in refuting that statement by imposing a condition that the radiant heat is turned off.

    I am speaking of radiant heat in the fundamental sense, and it does indeed heat people directly, however, for that statement to be technically correct, the person would have to be naked. Otherwise, the radiant heat would heat their clothes and the clothing would heat their skin by conduction. Certainly radiation heats people directly without the need to wash them with heated air, as I stated. It could heat people even if they were in a perfect vacuum.

    When I spoke of radiant heat, I was not speaking of radiant floors per se, or of peoples’ habits and preferences for how they set their thermostats.

    In your example of a radiant floor at room temperature, if it is colder outside, I assume that the floor is putting out heat to maintain the room temperature. And if it is putting out heat, a portion of that heat is being transmitted by the radiant mechanism. And it will warm every absorptive surface in the path of that radiation.

    When I said that radiant heat allows the air temperature to be lower than that of air being used in a forced air heating system, that is also fundamentally true. It goes back to the fundamental truth that radiant transfer does not require the presence of air. Basically, if you could raise the temperature of the walls, floors, and ceiling of a room, you could lower the air temperature correspondingly, and the body perception of wamth would remain the same. So, for example, you could make room air feel like 70 degrees when it is actually at 30 degrees if you raised the surrounding radiant surface temperature.

  13. jklingel | | #13

    numbers are welcome
    I confess that I do not look for studies specific to this topic, so if someone can line me out on one, please do so. I often read about "ductless mini is in, radiant floor is out" because of the "huge expense" of radiant, etc. OK, maybe so, but what I would like to see are the numbers. Given a "fairly good house" (just shy of Passivhaus standards) in zone 6, 7 or 8, where radiant heat may make the most sense (my guess), what would it actually cost per btu to run a mini-split(s) vs radiant, including all associated factors; equipment, labor, maintenance, operational costs, etc? If that info is on anyone's radar, it would be great to see it. Thanks.

  14. GBA Editor
    Martin Holladay | | #14

    Response to John Klingel
    When it comes to operating costs, fuel costs matter (of course). These two systems use different fuels.

    Minisplits use electricity, which can be cheap (6 cents a kwh) or expensive (18 cents a kWh).

    Radiant floors usually use natural gas (which is usually cheap), propane (which is usually expensive), or oil (which is usually expensive).

    So the answer is: it depends where you live.

  15. GBA Editor
    Martin Holladay | | #15

    Response to Ron Keagle
    You're right: a radiant heat source (for example, a wood stove or a campfire) can heat a naked person directly by radiation, even when the air temperature is low.

    However, if these statements don't relate to hydronic in-floor heat, they aren't particularly relevant.

    Hydronic in-floor heating systems do not primarily provide radiant heat. Nor do such systems allow homeowners to lower their thermostats.

  16. user-1017420 | | #16

    Hydronic Heat
    Hi folks. Good luck on your project. We are just finishing up a CasaClima house (sort of like PassivHaus) in Northern Italy. We have used hydronic heat in houses in Canada, and let me tell you, a nice warm floor is nothing to sneeze at! My wife can't say enough positive things about it, especially in the bathroom. Our house in Italy is tighter and better insulated, and the hydronic heat is completely overkill. In fact we found (last winter -15C) that ANY form of heat was overkill. We simply used no heat last year. Having said all this, I will tell you that we installed hydronic heat in this house. I could not talk my wife out of it, and in truth there simply are not enough people out there that would believe that the house could be heated without a heating system. So 1) to keep my wife happy, and 2) for resale value, we installed the hydronic system. It's there, we don't have to use it and of course it is on zones, so if Sara wants warm floors in the bath, so be it.
    Not germane to this subject, please make extra special double sure that your blower door test will pass. I do blower door tests in Italy and believe me when I tell you, I speak from experience.
    Again good luck

  17. jklingel | | #17

    Response to Martin
    Thanks for the response, but I now see that a better question would have been this: How many kilowatts would a mini-split system use if a hydronic floor system were to use 700 gallons of heating oil? That eliminates the money, climate, etc. Is there any data on such? I guess it would require a house w/ both systems used on alternate years and the heating degree days taken into account; probably asking too much, but it would be interesting to know.

  18. user-1031655 | | #18

    The problem with radiant floors
    As much as I love radiant floors, the real problem as I see it is that if you are in a climate with both heating and cooling needs, you still have to put in a forced air system for the cooling component. This is the major downer for those of us who love the lack of dust and pet dander being blown around that radiant systems provide. I have seen a radiant cooling system advertised, but it seems to be mostly for commercial settings as a secondary system, but even if it was primary I would still be worried about the condensation and mold issues.

    Of course with the minisplit systems, you may still need a backup heating system, and if you don't want to go with a portable space heater or wood/pellet stove type system, you are going to have substantial installation costs for that backup system. I suspect there are still a large number of areas where nightly lows drop below -15 on a regular enough basis that the minisplits wont cut it.That cost seems to be overlooked.

    As I mentioned in another thread, I also worry a bit about life safety issues revolving around installation strategies that assume bedroom doors being left open all the time to allow heat to circulate. This runs counter to good fire safety practice of closing bedroom doors at night to slow the spread of fire/smoke. Adding additional head units presumably would add cost to the mini split systems.

  19. GBA Editor
    Martin Holladay | | #19

    Response to John Klingel
    It sounds like you are looking for a fuel cost calculator. BuildingGreen has a good one available online:

    The calculator allows you to compare the cost of heating using a variety of appliances, including ductless minisplits and boilers. You can adjust the COP, boiler efficiency, and fuel costs to apply to your equipment and your location.

  20. user-1004247 | | #20

    Radiant not needed
    In 2005 I built my first well sealed and insulated house which included a sealed insulated crawlspace and attic and it was in a cold northern climate where ALL better built houses installed radiant.

    The HVAC system was forced air through uninsulated ducts in the crawlspace and the crawlspace had a supply and return in it (3,800 SF 2-story house). The wood floors were warm and comfortable throughout the winter and there were no "cold" or drafty spots in the house.

    I don't think I would ever build any other way again.

  21. Expert Member
    Dana Dorsett | | #21

    Response to John Klingel in-re oil-gallons vs. mini-split kwh
    The seasonal coefficient of performance (COP) of a mini-split will vary with the specific climate and the sizing relative to actual average loads. In broad field performance testing in the Pacific Northwest covering US climate zones 4, 5, & 6 with mini-splits covering only part of the design-condition load the seasonal COPs varied from the mid-3s in zone 4 to the high-2s in zone 6.


    (See Table 24 for seasonal COP averages for the different regions. The Eastern Idaho region is US climate zone 6.)

    But in lab testing under the same program it's clear that under part-load performance there is a significant boost in the average COP at any temperature, and if the unit is actually oversized somewhat even for the peak loads (as would be the case in a PassiveHouse) it appears likely that it would increase the average COP by more than 0.5. See figure 5 in this document:

    Since these are fully modulating systems, at part load they will run mostly in their low compressor & blower speed ranges. The seasonal average COP will run somewhat ahead the performance at the mean January temp for the location due to the VERY high COP during the warmer-temps. In a shoulder seasons. So in a zone-6 location such as Minneapolis, with a mean January temp of about +15F, looking at Figure 5 the mid-winter COP might average below 2.5, but the Nov-Dec and Feb-Mar mean temps is closer to 30F, and the heat loads lower, averaging closer to 4 for those periods, bringing the full-season average to about 3.

    If the load of a PassiveHouse at the outdoor design temp in zone 6 is say 10,000BTU/hr, with an average mid-winter load of something less than half that, a mini-split that can deliver 12,000 BTU at the outdoor design temp will deliver a seasonal average COP of 3 or slightly better. A 1.5 to 2 ton (nominal) mini-split would likely fill the bill at an installed cost of $4-5K. (eg: The 1% outside heating design temp is -11F, a temperature at which a Mitsubishi H2i is specified to deliver over 70% of the nominal heat rating, so a 1.5 ton unit would still have margin.)

    With an seasonal average COP of 3, every kwh of input, you get 3 kwh (=3412BTU x 3=) ~10,000 BTU) out.

    Oil has 138,000 BTU of source fuel energy, and burned in a 85% burner delivers ~117,000 BTU per gallon used. (Mind you even the smallest oil burner is on the order of 10x oversized for the design condition heat load of a PassiveHouse, which results in extreme cycling inefficiencies unless there is substantial heat storage/thermal mass somewhere in the system.)

    So the average gallon of oil is equivalent to (117,000/10,000=) ~11.7 kwh of power used by mini-split in climate zone 6.

    In zone 7 the average COP will be lower by about 0.3- 0.5, and there will be significant numbers of hours when the outdoor temp is below the operating temperature of a mini-split. This may be a real problem for less-well insulated homes, and even a PassiveHouse may need some resistance heating backup to make it through the decade-coldest cold snaps, even if it might coast OK through most design-condition days with the mini-split putting out zero during the coldest hour or three of the day.

  22. bmunsell | | #22

    Small super insulated house in Alaska
    Small - 860 sq.ft. - really well insulated houses in Alaska with maybe a max heat load of 40,000 btu add to the challenge. Because of extreme low temps, possibly -40, and no need for cooling. Mini-splits are out. Forced air systems seem to start at 60,000 BTU and price wise are overkill. Standard boiler operated radiant floors are over kill and expensive. Direct vent wall heaters would work and can be sized appropriately, but small houses don't really have the wall space to spare.

    By the way, this is a gut job. The house was heated by a toyo oil stove, but will now be on natural gas and also have a small wood stove. 2x4 construction.

    Going back with 3" closed cell foam in the walls, 5" in the roof with 2" ventilation space under the roof sheeting and then a exterior foam wrap of still to be determined thickness.

    Heat options?

  23. Expert Member
    Dana Dorsett | | #23

    Response to Barry
    Without a better description of the building envelope it's hard to say if a heat load of 40K @ -40F is really the right range. Assuming you can get 800' of radiant floor that's 50BTU/hr per square foot at design condition, which would require a floor temp of foot-roasting temps and higher water temps than domestic hot water. A combination of radiant floor + radiant ceiling could probably get you close at pretty good condensing temps though. No matter what, if your heat load is really on the order of 40K, a radiant floor makes for huge improvement on comfort, even if it needs more radiation than that to meet the load.

    A heat load of 40KBTU/hr is not a tiny heat load (it's more than the heat load of my not-so-superinsulated house), and using a 50-60K modulating/condensing boiler with outdoor reset control may be the right approach.

    With a design-day heat load of ~40KBTU/hr you can do fine with a condensing gas hot water heater designed for low-temp hydronic heating such as the Polaris or Vertex. Even the smallest-burner Vertex has a ~75K burner with over 65K of output, and if you have sufficient radiation that you can deliver design day heat to the room at hot water temps it may be a relatively simple way to go. The nice thing about using condensing hot water heaters for space heat houses with very low heat loads is that the rate at which you draw heat from the tank has no impact on the burner efficiency, and you can sip heat out at arbitrarily low rates. It's probably more appropriate for homes with heat loads well under 30K though.

    Calculating the real heat load (Manual-J methods or similar) on a room by room basis is an all-important first step in any heating design. Without that information you have no idea what the radiation/water-temp requirements would be.

  24. user-1005777 | | #24

    Response to Barry
    It seems that your heat load is high for that size house. Maybe it is because of high wind speeds in your area. We live in a 1035 sq ft mini home with a heat load when built of 300 btu//HR/F. We have baseboard electric heat with programmable thermostats. When I retired, I realised the programmable thermostats would not save us that much. We air sealed the house and installed inside stretch film storms. Last October we had a 12,000 BTU Mini split installed. We set the baseboards to come on at 61F. They never turned on. The minimum outside temperature was -14F. If we had more mass in the structure we could run the mini-split more during daylight. Anyway, at $0.0985 per KWhr our usage for a year is well below $1000 for all our electricity. This does not include monthly charge and taxes. They now make up 35% of our power bill.

  25. user-659915 | | #25

    " It could heat people even if they were in a perfect vacuum"
    If they were in a perfect vacuum there'd be little point in heating them.
    They'd be dead. ;-)

  26. user-1096505 | | #26

    Passive house changes how we push BTUs
    I am in exactly the same boat with our off grid passive house now under construction. Mini splits are out of the question so the make-up BTUs have to be from propane or wood when the sun don't shine or when I don't have a party during a blizzard. I think underfloor radiant works but we don't need to go 12" oc and overkill the system.

    One benefit I see is to be able to run a very small circulation pump off a solar electric panel which can distribute the heat from where the slab gets plenty of sun to where is does not, thus improving the overall 'thermal battery' and potentially getting us through a couple cold days we would other wise need to heat via carbon fuels. Does anyone have real world experience using this technique?

  27. user-1140531 | | #27

    Reply to James Morgan
    " It could heat people even if they were in a perfect vacuum"
    If they were in a perfect vacuum there'd be little point in heating them.
    They'd be dead. ;-)

    Somehow, I knew somebody would say that. But my point was not to suggest that radiant heat would work well, even for those people who choose to live in a perfect vacuum. Most of those people heat with wood.

  28. jklingel | | #28

    Just some numbers
    First, thanks for the links and responses to my questions and to other folks' questions as well. Just an FYI to add to the numbers being discussed. I went through Siegenthaler's room by room heat loss spread sheet and came up w/ a 20,600 btu/hr heat loss with a 3600 sf house in Zone 8 (Fairbanks). I am hoping that some passive solar heat gain will reduce that. Floor water temps ranged from 73 to 84 F. I am leaning toward a propane mod/con boiler and radiant floor heat (perhaps solar HW heating too) but am still working on that. These mini-splits sound interesting, but I don't know if they are viable up here; more reading to do.

  29. GBA Editor
    Martin Holladay | | #29

    Response to John Klingel
    A heat loss calculation is a peak load calculation. It never considers solar gain, because peak load is calculated for the coldest night of the year. The sun doesn't shine at night.

    For that matter, the sun doesn't shine during the day either in Fairbanks during the coldest weeks of the year. (I'm only exaggerating slightly.)

  30. wjrobinson | | #30

    John, Alaska... Your oil
    John, Alaska... Your oil should be free, tap that pipe up there, the Arabs like that trick there way. Splits to me in Alaska would be "split" to Hawaii for the winter. Install airline ticket November 1st and send pics to the non-splitters.

  31. MICHAEL CHANDLER | | #31

    responses to John K re BTU/ HR
    My engineer and I have been using historical weather data from Climate Consultant software to adjust the design heat loss for extremely well insulated, high mass, radiant heated structures to adjust the design coldest hour to reflect a weighted average coldest temp over a 48 hour worst cold snap. We do tend to have a lot of sun on our coldest days around here though.

    We also are using primary and secondary manifolds, the water or glycol is warmed at the heat exchanger and then goes into the floor in primary coils where we will value a warmer floor - mud room, entry, center living room and dining, master bath and dressing room, then comes out of the floor and re-enters into a secondary manifold of rooms where warm floor is less significant. We get the passive solar thermal flywheel with out having to settle for tepid floor heat.

  32. jklingel | | #32

    back to Martin
    A heat loss calculation is a peak load calculation. It never considers solar gain...•• True. That would be impossible to figure in. I am just saying that I hope the passive solar reduces the calcs; I imagine it will.

    For that matter, the sun doesn't shine during the day either in Fairbanks during the coldest weeks of the year. (I'm only exaggerating slightly.) •• You are not far off, and what we do get for a month or so is trivial IMO for passive gains. We get down to 3 hrs 45 min (approx) on 12/22 or 24... somewhere in there. That is when Thorsten cranks up his massive masonry heater.

  33. jklingel | | #33

    back to Michael and AJ
    Michael: That primary/secondary loop is quite an interesting concept. I will run that by my plumber. AJ: Funny guy, and that concept crosses my mind often. No snowbird yet.

  34. Roger Normand | | #34

    Yes, we have (so far) waived goodbye to radiant hot water
    As the author of this blog that GBA reprinted, It’s truly an honor to have en entire consulting committee helping brainstorm on EdgewaterHaus! I am always amazed at the insight and depth of responses posted on GBA. I wanted to address a few of the comments.

    I was aware of the minimum fire output of tankless heaters that Dana mentions, which soured me on using that approach.

    I toyed with several ways to use solar thermal, including an air to water solar thermal system, as David Eakins suggests. Thought about a sand bed, but rejected it due to inability to regulate the temp. Solar thermal is expensive compared to solar voltaic.

    I thought about geothermal, with ground loops around the foundation excavation, but an ongoing study I read (Solar Today?) was inconclusive. It's also an expensive option.

    It took awhile, but our energy consultant Marc Rosenbaum convinced us that we just are not good candidates for any type of radiant hot water heat, regardless of source heat, and ideal candidates for mini splits. He convinced us that adding more solar voltaic with a condensing DHW was the better approach. Air conditioning is a non-negotiable need for Lynn, and a mini split system with a COP of ~ 3 can provide both heat down to -15 degrees F outdoors and cooling/dehumidification. The mini splits also eliminate the need for oil or propane and its associated infrastructure. That will leave us fossil-fuel free, at point of use anyways. We will use a small pellet stove with a low BTU output for ambiance and backup heat. Pellets are a renewable resource and a very common heating fuel here in ME.

    The main floor is a 1 1/2” thick lightweight concrete with tile. We will add electric radiant floor mats under the tile in the bathrooms as AJ Builder suggests to give us the toasty toes. Might add it to the foyer also.

    So that’s the current plan as we await completion of the foundation. Any further thoughts are welcomed.

    Of course, our ultimate backup heat will likely be booking an immediate cruise to the Caribbean!

  35. user-1115524 | | #35

    Radiant Options
    If you can afford it, I would consider roughing in the in-floor piping for future use. Try the mini split during the first heating season and see if it feels comfortable, or drafty and dry.

  36. Perry525 | | #36

    Radiant Options.
    I wonder how many correspondents have lived with both baseboard heating and in floor hydronic systems? I first fitted both in the same home some 40 years ago. I found that baseboard heating existed without problem, we never knew it was there, the home was zoned with each room controlled by its own thermostat and motorized valve, the system was turned on when fitted and off when we moved out.
    I fitted in floor hydronic in our bathroom and toilet, a light weight system between the joists, it took up no wall space, it was invisible and worked perfectly again un -noticed.Our ground floor in concrete slab was different, yes it was (almost) always on, it was fitted as we had floor to ceiling twin pane windows, this gave us invisible heat offsetting the cold winter air dropping down the windows but, when the sun came out the heat was still escaping from the concrete slab, sometimes the rooms became unbearably hot, we opened the doors and windows and we could not loose the heat quick enough. On the other hand, it was wonderful for the rest of the time.

  37. mscannell | | #37

    Radiant not needed by Bob
    Bob, I like that approach. It may be the answer for a client I am currently working with. We are going double wall etc, and he doesnt want to see a wall coil unit, so he is opting for high efficiency gas forced air. we talked about keeping the ducts in the envelope.
    What did you do for ground insulation? We have a fairly level lot with a crawl space. But I can see insulating around the footings could be truoblesome

  38. GBA Editor
    Martin Holladay | | #38

    Response to Michael Scannell
    Who is Bob? I'm not sure who you are responding to.

    If you are looking for information on insulating a crawl space, here's an article that may help you: Building an Unvented Crawl Space.

  39. user-946029 | | #39

    Comment #20

    I believe Michael was responding to comment #20.

  40. GBA Editor
    Martin Holladay | | #40

    Response to Mike Collignon

  41. marcopolo10 | | #41

    watts to btus
    Hi, not sure this question fits here but haven't found a good answer elsewhere yet. I have a 2500 sq ft ICF home (ICF basement and main floor, stick walls upstairs) 1,000 sq ft main floor, 1,000 sq ft basement and 500 sq. feet upstairs. I have electric baseboard throughout the house but only use those on the main floor, 2 at 1,500 watts each, 2 at 1,000 watts and a small 500 watt unit in the bathroom. Total of 5,500 watts. This is sufficient to heat the entire house, although the basement stays cooler. I have an LP fireplace in the living room that kicks in when the electric (off peak) goes off for four hours during the off peak periods from 4 p.m. to 9 .m. The fireplace is variable up to 38,000 btus and does a good job of heating the house during the time the electricity is down. Central Minnesota. Question is this, if I wanted to get a small LP forced air furnace to replace the electric baseboards, what size furnace would be the equivalent of the 5,500 watts of baseboard that currently do a good job of heating the entire house. Please note, I have a very open floor plan with big open steps going downstairs and big stairwell going upstairs. I'm thinking of just putting the furnace in the basement and putting some floor registers on the main floor and letting the warm air move up through the house. I'm thinking if a 38,000 btu fireplace can keep the place warm, a small 40,000-50,000 btu furnace in the basement would also do the job and more cheaply than the electric baseboards. I don't want to use the fireplace as a primary heat source. Thanks for any ideas. Is there a simple conversion table that compares watts to btus put out by a forced air furnace?

  42. GBA Editor
    Martin Holladay | | #42

    Response to Dan Anderson
    Here is the conversion:
    1 watt = 3.4129 Btu/h
    5,500 watts = 18,771 Btu/h

    A furnace with an output of 18,771 Btu/h has the same heat output as 5,500 watts of electic resistance baseboard.

    However, it's certainly possible that your peak heating load is less than the capacity of your electric baseboard units. The first step to sizing your furnace is to perform a heat loss calculation for your house.

    For more information on this topic, see:

    How to Perform a Heat-Loss Calculation — Part 1

    How to Perform a Heat-Loss Calculation — Part 2

  43. badgerboilerMN | | #43

    Trashing radiant floors
    " The floor isn't really a radiant source for the occupants in the room, since the floor is usually at room temperature. Even when the floor heats up, what it is mostly doing is heating the furniture and ceiling by radiation and the room air by convection. (And occasionally, heating a human being by conduction, if the human is barefoot.) Unless you are lying naked in a mesh hammock, with the radiant floor on full blast, you aren't being heated by radiation."

    So outrageous that one hardly know where to start. But it is a new year.

    The floor is either a heat source or it isn't. Lacking any other source of heat, one must concede that if the indoor temperature is above the outdoor temperature the floor must be giving off heat. It will certainly convect but most of us who design and install and live with radiant floors know that the ambient temperature in a radiant floor heated room are often below both the floor and the ceiling temperatures. Yes the ceiling by a certain amount of convection but the bulk from being exposed to the floor below. The same is true of radiant ceiling thus proving the the effectiveness of radiant panels regardless of orientation.

    The statement that floors are usually at room temperature could only be true if there were a source of heat below the floor or direct solar gain. In all other circumstances the floor of any heated structure will always be the lower than the ambient air above (this is where your convection comes to play). Regardless of the quality of construction the windows will present a constant load (ever-increasing as the rest of the envelope improves and the cost of quality windows continues to be relatively high). This load impacts human comfort in the form of cold air falling from window to floor and to the feet of those nearby. The typical un-heated floor temperature of a 70° room is in the mid-sixties regardless of construction practice.

    They fact that you "feel" warmer in a well constructed house than in a leaky and poorly insulated one, is a factor of AUST, which can be had by elevating any surface thereby making the average higher and the human more comfortable.

    As for large, full-time pumps my own is on nearly full time following outdoor reset and using an average 8 watts. I will concede that a true Passivhaus will not be the perfect place of radiant panels, but with the current US standards you only kid yourself with forced air. I use a mini-split for my own cooling and will use it for heating until the outdoors goes in below freezing and then the warm floor come on and the mini-splits waits for spring.

    We design and install many systems for cold climates and move air for cooling but radiant for warmth. It is a matter of comfort and makes even more sense when integrating domestic hot water and space heating.

    Finally, the beauty of radiant floor heating is it's versatility. Any fuel may be used to drive a radiant floor boiler including ground source heat pumps, solar or yes, natural gas.

  44. GBA Editor
    Martin Holladay | | #44

    Response to Morgan Audetat
    I stand by my statements, which you call "outrageous."

    If you design a radiant floor heating system to meet the peak heating load -- a load that only occurs on the coldest day of the year -- then it won't be necessary to use the heating system for 24 hours a day. Most of the time, it won't be needed. That's why the "warm floor" phenomenon only occurs for a few hours a day, at most.

    I don't recommend that designers use circulating pumps to keep fluid flowing for 24 hours a day -- in other words, 4,000 to 5,000 hours per year. In one house I reported on, the two pumps drew 173 watts. You can read more about that case here: Near-Zero-Energy In New England.

    I am skeptical that your circulating pump draws only 8 watts; that number is very unlikely.

  45. badgerboilerMN | | #45

    You are likely not familiar
    You are likely not familiar with ECM pump technology. I have hundreds of radiant floors in the field, starting in 1987, and few draw more than 80 watts of power. One poor design (173 watt pump for a residence) does should not kill a category in anyone's mind.

    Many in the southwest use PV powered circulators to drive radiant floors. Whether circulators operate 24/7 is more comfort than power concern but since the weather is ever changing matching design water temperature to outdoor conditions (ODR) is our bread and butter. I am not talking about the warm floors, but efficiency and comfort together i.e. real world.

    Running circulators full time, modulating on flow or delta T with thermostat controlled thermal actuators will soon be common place in many hydronic heating designs, as they are in Europe and a perfect way to transfer heat from a room with solar gain to one without a window.

    Depending on the source of fuel there are few heat transfer systems as efficient as a slab already in the specs.

    But hey, if you have never heard of a circulator heating a home with 8 watts draw (the circulator has an LED readout in real-time) then what more is there to say...stand where you please.

  46. GBA Editor
    Martin Holladay | | #46

    Response to Morgan Audetat
    Please tell me the manufacturer and model number of this 8-watt pump.

    The reason you can't use a PV-powered pump to circulate hydonic fluid for 24 hours a day is that there is no sunshine available at night.

  47. tos | | #47

    One type of low power circulator

    I am using 2 of the 12V ELCID circulators in my radiant floor system; they are currently running off a small wall wart 24hr/day; yeah, they work fine. I expect to eventually run them off a battery kept charged by solar panels (even in these gloomy Vermont winters!).

  48. GBA Editor
    Martin Holladay | | #48

    Response to Terry Steiner
    I like the El Sid pump -- I've been using one for years on my solar thermal system. Mine is PV-direct, without any battery.

    Two El Sid pumps draw 20 watts. That's good, but it's not 8 watts.

    I strongly advise off-grid homeowners to abandon any idea of running circulators for 24 hours a day during the winter. Your two El Sid pumps will draw 480 watt-hours a day, or 14.4 kWh/month. That's a lot. To get that much electricity in November or December in Vermont, you need a 275-watt PV array that costs $1,200 (not counting batteries).

    And you'll find that there are long periods in November and December with almost no sun at all.

  49. tos | | #49

    Vermont insolation
    I agree that there are several months in Vermont where sight of the sun is rare; I have been observing that depressing fact since about 1975. That is one reason my primary heat source is a wood boiler and wood stove. I want to eventually put in a solar thermal system as a supplement for when we do get sun ( and your PV direct circulation works well and is simple). My reason for the battery driven circulators is for our frequent power outages when I struggle to keep the whole 'mess' from freezing. I like the 24/7 circulation since it is quiet (no creaking tubing as it heats and cools) and, currently, a more uniform temp in the conditioned space since my insulation envelope is still incomplete. (Yes, once I get that insulation envelope complete, there will be much less risk of freezing, and, likely, less need for 24/7 circulation; which, sort of, gets back to the original point of the starting blog!).
    I am not trying to live off grid at all, as tempting as the concept is, I feel it isn't really practical yet, at least in my circumstances.

  50. badgerboilerMN | | #50

    The energy draw of any
    The energy draw of any circulator is dependent on the requirements of the hydronic system is serves and the design condition. It is not A pump rather the specific pump and design parameters.

    Since my expertise, and trade licenses are related to plumbing and HVAC I have made an extra effort over the years to learn more of the details of the thermal envelope (important in performing Manual 'J' heat loads). I would encourage you to do the same as it relates to mechanical systems.

    All the relatively new ECM wet rotor circulators will run full time and modulate between 2 and 35 watts depending on the hydronic system demand. The solar systems of the southwest are easily driven by PV panels with battery backup and have been for many years.

  51. Trevor_Lambert | | #53

    I know this thread is ancient, but there's a glaring error in the assertion that a "warm" floor will heat a human body radiantly that no one mentioned. Radiant heat transfer is from hot to cold. So unless the floor is above the temperature of your skin, the radiant heat flow is actually from you to the floor. I guess you could make an argument that the "warm" floor is radiantly cooling you less than it otherwise would, and that is the same effect as heating. But it's definitely technically incorrect. And when you're talking about a floor that's only a degree warmer than the air, even the "cooling less" claim is pretty preposterous.

    1. maine_tyler | | #54

      > "Radiant heat transfer is from hot to cold. So unless the floor is above the temperature of your skin, the radiant heat flow is actually from you to the floor."

      The net flow, yes.

      A net flow of energy from human to floor doesn't imply that there isn't energy radiating towards the human though. But you are right that one is essentially being 'cooled less.' That's true for many things, including furnace hot air once it mixes.

      I see it being a vaguely similar concept to a blanket, which 'warms you' even though it is only reducing your heat loss. A player here in both cases is our own internal fires.

      Any 'reduced cooling' reduces the burden on our metabolic furnace. The more energy we receive, the less work our bodies have to do to maintain temperature ('more' energy not needing the qualification of being 'above' our body temperature).

      This is all just to quibble about the science. I agree with the general sentiment that a radiant slab in a well insulated house isn't gleaning the benefits most are looking for (and comes with further drawbacks). If the slab temp is truly only a degree or two above what it would become via room air heating (is that only true of Passive level performance?) then it's true that it won't likely add very noticeable radiant influx.

      It is worth noting— even if its a bit trivial— that concrete being a few degrees warmer than air isn't the same as air being a few degrees warmer than air, given heat capacity and such. This translates (or at least correlates) into higher emissivity; a 70 degree cloud of air encircling me will not radiatively 'cool me less' than a concrete box at 70 degrees surrounding me.
      But this certainly helps to explain WHY the slab would likely need to be only a degree or two above set temp, whereas a heat-pump head will spit relatively warmer air.

  52. Trevor_Lambert | | #55

    "A net flow of energy from human to floor doesn't imply that there isn't energy radiating towards the human though"

    You could substitute the word "floor" with "block of dry ice" and it would still be true, since it's above 0K. Everything on earth is emitting energy. The relative temperature determines what is heating what. Yes, you're correct that the air coming from a furnace, once mixed with ambient air, is not directly heating the person either, only reducing the cooling from person to the surrounding environment. The difference is in what is being claimed; people are claiming that radiant surfaces are warming people via radiation. In reality, the warming effect has little to do with radiant transfer; perhaps close enough to nothing to just call it nothing. Even the cited wood stove or campfire, I would argue aren't significantly warming a person by radiation. The heat perceived from a hot stove is almost entirely due to the hot air in direct contact with the person. How far away do you have to be from an outdoor campfire before you cannot perceive any warmth from it? It's not very far, maybe 20 feet. I would suggest this is because the temperature differential is not high enough. You can feel the heat from the sun because it's millions of degrees, and the distance doesn't matter (except of course due to the amount of radiation hitting you is a function of the angular size you represent).

    The effect of a fire at a couple of thousand degrees differential is perhaps arguable that the radiant energy has a noticeable warming effect. A temperature differential of a couple of degrees difference in a floor? There's no way any human could stand the same distance away from two surfaces, one at 21C and one at 22C and tell you which one is which. If they think they can, they are kidding themselves. By proxy, this has been proven by the fact that people with "radiant" heating systems do not actually keep their homes at lower temperatures; in fact it appears the opposite.

  53. maine_tyler | | #56

    Thanks Trevor, that response cleared things up for me.

    My initial response was mostly a quibble with the specific phrase I quoted. I felt it implied that energy was only flowing one way. I also have questions about the threshold at which a radiating body can be considered to be an effective source of perceived warmth (see * below). Overall, I see your points and pretty much agree with your assessment, even if I quibble a bit with the precise language.

    I do think fires warm via radiation, but in terms of whole house heat distribution, yes, it is probably mostly via convention as you suggest.

    The point I was driving at is that a body/object will radiate energy according to its own temperature, not the temperature of other objects around it. It's always an exchange. A radiating body doesn't 'care' what it is radiating towards. It only 'cares' about what it is getting back. Differential temperatures matter in terms of the tendency towards equilibrium. I don't think there is disagreement here; I'm just clarifying.

    And we agree that air at 70 degrees also isn't 'warming' a person. So the goal of conditioning a house is really to slow body heat loss.

    *It starts to get complicated (and I confused) when we compare radiative exchange concepts with air temperature exchange concepts. In normal operating temperatures, clearly both are only slowing our heat loss. We readily ascribe value to 'warm air' being effective to 'slow heat loss' but is the same not true of 'warm objects'? In other words, does an object really need to be above ≈98°F for any 'perceived' warming (less cooling) to equal or surpass the perceived warming (less cooling) of ≈70° air? That seemed to be the implication you made (correct me if wrong) and was perhaps my main gripe.

    Air is a tough substance to analyze in thermodynamic terms (I'm not up to snuff on it, that's for sure). How does it actually transfer heat to us, or we to it? Obviously convection moves it, but then how is that energy actually exchanged at the boundaries.? Is that still convection? Is it radiation, conduction, something else, or all? Maybe this doesn't matter for the purposes of this discussion, but I have a feeling it might—peripherally at least.

    I stumbled across some concepts in search of this understanding. I haven't had time to digest them, but they are: mean radiant temperature, operative temperature, and physiological equivalent temperature. Perhaps you're already familiar and have insights.

    I'll take any blame for making this more complicated than it perhaps needs to be... but I can't help myself from going down most any rabbit hole I find.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |