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

Community and Q&A

Can I combine floor and wall heating in a solar passive house to avoid overheating?

user-6593244 | Posted in Mechanicals on

I have consulted many “experts” on highly insulated houses, most of them recommend to use ductless heat pumps to heat the house. But I am a radiant heat addict…
Apparently the problem with radiant floor and passive solar is that when the sun comes out, the southern side of the house is too hot because the concrete slab has heated during the night and is warm and when the sun comes out it can bring it to a much higher temperature.
To try to balance this, I have been suggested to put pex in the walls instead of the floor with an opening at the bottom and at the top for convection. Like that the change of temperature would be much faster and we would put radiant heat only in the northern half of the house where the windows are small .
Our house will be a 78 ft long duplex.

Do you see any inconvenient in having only half of the slab heated? could it cause it to crack because of the difference of temperature?
It will be insulated with 4 inch of spray foam underneath and will also be insulated all around with Styrofoam and a membrane over it.

The heating system should be an air to water heat pump (we are looking into geothermal… but not too sure because of the cost)

Any input will be greatly valuable

Thank you!

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member
    Dana Dorsett | | #1

    It's fine to zone slabs even in code-min houses, and in a high R house the temperature differences of the slab from zone to zone will be even smaller.

    But slabs are fairly unresponsive, with long thermal lags. Radiant walls & ceilings are more responsive, but whether that alone is going to be enough to eliminate all temperature overshoots in high solar gain rooms depends on a lot of factors.

    All decisions around heating systems are a bit premature until & unless a careful heating & cooling load calculation has been performed.

  2. user-6593244 | | #2

    Hello Dana
    I am not sure what you mean by code-min houses, which other factors to minimize solar gain overshoots are you thinking about?
    What I am most concerned with is can the slab have a tendency to crack due to an uneven temperature?
    It does that with glass, but I don't know much about concrete, the slab is supposed to be 5 inches thick.

  3. GBA Editor
    Martin Holladay | | #3

    You don't have to worry about the slab cracking. But Dana gave you good advice on basic design principles.

    For more information on this issue, see Reassessing Passive Solar Design Principles.

  4. user-6593244 | | #4

    Thanks Martin!

  5. Expert Member
    Dana Dorsett | | #5

    "code-min"== "built to the bare minimum code requirements for R values & window performance", in other words, not an exceptionally high thermal performance house with a exceptionally low heating load.

    If you are designing & building a new house it can be cost effective to build in thermal performance higher than code-min, lowering the heating & cooling requirements, and the size/cost of the mechanical systems needed to manage the heating & cooling loads. The cost of up-sized mechanical systems can be very substantial if going with a ground source or air-to-water heat pump.

  6. user-2890856 | | #6

    Zoning for any radiant project must be considered very carefully based upon individual room needs . Doing radiant on only the North side quite frankly is silly . In a home such as you describe the water temps would be relatively low , thinking 100* or less at design day conditions . When done properly solar heat gain will keep a zone from requiring heat thus no water would be flowing through said floor . Thermostat placement should also be well thought out , especially in this type of building .

    More often than not the water temps required will actually remove heat from the slab . Overshoot , while not a wives tale is created by poor design . When zoned properly and designed well using a tight outdoor reset curve these factors are easily handled .

  7. user-6593244 | | #7

    Thanks Richard for your input
    We have already zoned each side of the house with 4 zones each (it is a duplex)
    Zone 1= 2 bathrooms North side
    Zone 2 = 2 bedrooms North side
    Zone 3 = Home office south side
    Zone 4 = Open living, dining, kitchen area south side

    I am with you, the water should normally be low, (I thought it was about 120ºF)
    If we put pex in the floor on the south side and the temperature is already at 68ºF, the slab is already warm so if the sun comes out it will keep on accumulating heat (even though it is actually not heating)
    and the room temperature will start rising to end up being may be 75º or more... I don't know...
    and this is the part I am trying to figure out...
    If the floor does not contain pex... and requires no heating, because the sun is out... it becomes more responsive ( radiant floor that we have where I actually live take about 12 hres to change) therefore it does not overheat.
    Am I thinking right? is this silly?
    Where should the thermostat be placed?
    Which part am I missing?
    So you suggest to put pex all over the floor and forget the wall heating?
    make sure that the slab temperature is monitored? and it should do the job and avoid overheating?
    Thanks for your precious input


  8. user-2890856 | | #8

    The radiant wall detail you say someone described to you is hard to picture working well , if at all . Bad idea , in my opinion . An often overlooked method of radiant heating is radiant ceilings . Probably because most people think heat rises , in fact heat energy is omni directional , hot air rises . Radiant systems do not heat air , they heat surfaces and objects very similar to how the sun heats the same things . Is the top of the Sun always toward Earth ? Ceilings also offer less resistance (R Value ) for the energy to travel through . They are also very responsive , unlike any
    floor detail that can be constructed . That heat is not effected by any carpeting or furniture . It also does not suffer from the same surface temperature limitations and considering all these things also offers the ability to use the lowest possible supply water temps . Since this energy moves to the cold surfaces in the same manner as systems built in the floor that energy will still move through air to the colder floor and negate the cold floor feeling , it will be neutral , not hot or cold , warm floors are a myth in these types of homes anyway . If done properly with a foam based product such as Sun Board ( sunfoam) it will also lessen if not eliminate any thermal bridging from framing members . Something to think about .

  9. Jon_R | | #9

    I'd think more about how to move heat from south side rooms to the rest of the house (so you don't waste heat by opening windows). A floor fan moving air through an open doorway will have a much larger effect on overheating than a concrete floor being at 70F vs 80F.

    Even so, a non-heated concrete floor is better (although perhaps not in $ terms). A normal radiant wall has a ~34% better heat transfer coefficient than a ceiling (due to the convective component). Convection on both sides would increase that, further improving heat pump efficiency.

  10. user-2890856 | | #10

    A well designed ventilation system would do a fine job of moving that excess heat around the house Jon , without floor fans and such .
    I disagree that a normal radiant wall has 34% better heat transfer than a ceiling . Where would one verify that myth ? Both have drywall covering them , Walls offer less sf for actual active panel and thus would require higher supply water temps which in turn uses more energy while lowering many source type's efficiencies . By the way , what is a normal or abnormal radiant wall ?

  11. user-6593244 | | #11

    I had not thought about the ceiling, so used of hearing that heat rises...
    It would be a very simple solution that would only require some kind of opening maybe with artsy metal sheet that would allow the heat to circulate.
    So far the back side of the wall are sealed, it is a double wall structure.
    May be I could put wood shutters over the pex in the walls and the convection would be natural..

    Thanks for your contribution, I will be able to bring my project one step further

  12. user-2890856 | | #12

    An assembly that has vents added at top and bottom would require much higher temps than what I am about to show , especially in a double wall assembly since the air would be circulating through a colder space between the walls . Not sure it would be a good thing either from a building science perspective . Maybe others would like to add why circulating warm air through that assembly would be a bad idea . Besides the fact that this assembly is no longer a radiant wall , it is a convective emitter and as such would be a waste of money since you could get better output from simple baseboard at a much lower cost .

    I question why you linked to Uponor's cooling manual , maybe you meant to link to the CDAM , .

    I am attaching reports for 3 assemblies , Uponor reports , all placed in the same room to demonstrate the differences between ceiling , wall and floor assemblies . Please remember , I design for comfort , when you achieve that efficiency follows . I also design systems that operate as systems and maximize both comfort and efficiency . Output is only part of system design and installing something because it may or may not have 34% better transfer is faulted in most cases , if not all .

    Similar to installing more insulation at a one time cost that pays for years , designing systems that optimize comfort and component efficiency is done best when all parts of the system are considered , especially the comfort of the end user .

    Micheline , Being a radiant addict you must take care to whom your purchase your addictive material and word of mouth is subjective .

  13. Jon_R | | #13

    How many CFM would such a ventilation system move?

    See here (or numerous other places) for wall vs ceiling coefficients. Table 2-5, heating wall vs heating ceiling. "Values Reflect Both Radiant and Convective Transfer".

    Normal radiant wall - air flows on one side, backside is sealed up. Alternative that I think Micheline is proposing - vents are added such that air can flow over both sides of the drywall or plates. Of course radiant transfer still occurs from the room facing side.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |