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

Community and Q&A

The Need for Supplemental Heat with Radiant Floor Heating

jason_v | Posted in General Questions on

Thank you in advance for your time and consideration.

We are building a custom home in climate zone 7a. The home is 3000 sq feet of conditioned space with lots of glass (triple pane tilt and turn), most of it southwest facing. It’s 2×6 construction with 1 to 2 inches of continuing insulation on the outside walls (depending on exterior material). The roof is a non vented assembly that will be approximately R35.

According to the Manual J Heat Load calculations (done by a professional engineer) the total load for our house is approx. 85,000 Btu. We are going with slab on grade with a hydronic  system embedded in the slab.

The company that did the loop layouts (and also a heat load calculation which came up with the same numbers) reports that “some rooms may require supplemental heat supply to meet the design load”.

Has anyone ever heard of this situation? Why wouldn’t the radiant heat be able to meet heating demands?

Thank you, everyone here has been greatly helpful.

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. Jon_R | | #1

    Does the system treat each room as a separate zone?

    1. jason_v | | #3

      Thanks for the reply.

      There are several zones, yes. master bedroom, ensuite bathroom, main living area, a guest bedroom "wing". entry and laundry.

      1. Jon_R | | #5

        Without per room zoning, the ratio between rooms won't be perfect or constant, leaving one room in the zone either too hot or too cold. Per room supplemental heat with its own thermostat would address this.

        It could also be a boilerplate excuse to point to if there is ever a problem.

  2. Expert Member
    Akos | | #2

    Would be good to read through this thread:

    Doesn't sound like your insulation level is as high, but definitely high enough that there is very little chance your place will have an 85000BTU heat load unless somebody forgets to install a couple of windows.

    You have to check through the design details and find where the error is.

    Sounds like you are also trying to design for a passive solar type setup with floor heat. This sounds like something that should work but very hard to get right in practice, most places end up with significant overheating problems on sunny days. Combine this with floor heat which tends to be slower response, you are asking for a place that will always be either too hot or too cold.

    Stick to reasonable glazing, have this facing the orientation that gives you the best light and views. Design in exterior shading as needed.

    Floor heat generally does not give the perceived benefit most people expect (warm toes) in well sealed and insulated house even in zone 7. It is a lot of cost for not much benefit. I'm in a big fan of floor heat, it just has to be applied properly. Installing 3000sqft of full radiant slab is a waste of money.

    1. jason_v | | #4

      Thanks for the reply.

      We're not counting on any passive solar. We have ductless minisplits for cooling. All glazing has significant overhangs (4 feet average). We have had 3 different manual J calculations done by professional engineers as well as energy modelling and they all come to 85K, so there is a high probability it is correct.

      I have lived most of my life with radiant flooring so I am familiar with the control issues. We heat over 300 days a year up here.

      After looking closer the calculations are counting a 17K heat loss via "infiltration" but i don't know how the came to that number. We are aiming for a 0.5 ACH 50 via construction techniques are using Aerobarrier so I'm going to have them done using that alteration.

      thanks again

      1. Expert Member
        Akos | | #7

        If you can post your modeling report including the input section, we can quickly go through and find where the issues are. The 17k of infiltration for 0.5ACH house is the first one.

        The table is also missing heating loss associated with ventilation, that is not something you can ignore in a sealed house.

        Your slab losses seem high, you might need to add more insulation under the slab.

        1. Jon_R | | #8

          > heating loss associated with ventilation

          +1 on this. A 200 CFM of kitchen exhaust alone will sometimes exceed 17K btu/hr of heating. And you should have more CFM. Since the heat from cooking is exiting out the exhaust, it does nothing for the distant rooms where infiltration is being induced. Balanced kitchen makeup air will greatly reduce the peak heat load.

          Windy days also push infiltration rates far beyond average rates.

          If you are using a modcon boiler, then proving a lower load (likely) may only be important for the emitter sizing.

        2. Deleted | | #10


        3. jason_v | | #11

          thank you

          1. Expert Member
            Akos | | #18

            All of them have issues. I think the simples is to fix one, easiest is the Wrightsoft one from Acacia.

            Can you post their detailed report.

            Quick look at the summary:
            - design temp is off (Red Deer should be -20F), inside should be 70F.
            -ACH of 3.6, well above your target
            -no load for ventilation, this needs to be added in and designed into your HVAC plan

            Looks like a lot of load for heating a garage. Is this something you want to do all the time or occasionally? Heating a garage full time is a big energy cost, even the best garage doors leak like a sieve.

            The Rehau report shows R10 under slab. I would bump that up at least to R15, R20 better with heated slab. You want the heat from the slab to go into heating the house not the soil bellow.

        4. jason_v | | #19

          Thanks for the reply. The Acacia one is posted in full, it's just those 4 pages. We planned on in floor heat int he garage because it seemed to make sense, the numbers are supposed to be designed around keeping it at 10 C. We will look at adding more sub slab insulation for sure. I would also consider going to a unit heater or something for the garage but that doesn't really affect the current issue. I would like to know if the "may need supplemental heat" would be resolved with a lower infiltration rate and more sub slab insulation. Thanks again for your input I really appreciate it.

          1. Expert Member
            Akos | | #21

            The short answer, based on the description of the building, there is zero chance that any of your rooms will need supplemental heat.

            The Acacia report is the summary. They should be able to provide a full detailed report, should be around 15 to 20 pages. This has all the details of the assemblies listed in there such as what R values of walls and roof and windows.

            You do want to get these numbers correct, there is no point in installing larger boilers, pumps or extra tight pipe spacing.

            Floor heat is very slow, you don't want it for a space that will be only intermittently heated. A single larger unit heater is what you want, something that can get the garage up to temperature quickly if you want to work there.

          2. Expert Member
            NICK KEENAN | | #22

            Floor heating isn't inherently slow to respond, but heating embedded in concrete is. Floor systems that use aluminum tend to be more responsive, which provides better comfort. Concrete is a lousy flooring for residential use. Plus it's environmentally harmful in its production.
            If I were designing a luxury home concrete floors is the last thing I'd want. Well insulated wood framed floors with aluminum heat transfer plates will give much better comfort.

        5. jason_v | | #23

          Thank you. Acacia used the values from the energy modelling, which i think i attached, which has all the values for the assembly and windows from the vendors and architectural designers. I will attach again. Regarding the garage, it is just a home for vehicles and storage, its will not be generally used as a work shop. Thank you again for your input I really appreciate it.

          1. jason_v | | #24

            Thanks for your input. My daughters wheel chair destroys wood floors so they aren't an option. Polishing the slab on grade is the best and most economical option for us, we are doing an engineered frost protected monolithic slab to reduce the amount of concrete.

          2. Expert Member
            Akos | | #26

            Been a while since I had to look at HOT2000 inputs, not the easiest to go through.

            The design heat load output is 38000BTU in there though, so that is in the ballpark. The other heat calculations showing 70000BTU+ are out to lunch.

            A couple of items that will reduce the load further. It looks like the calculation is using 2.5ACH. The R value of your EIFS walls seems low. 2" EIFS over 2x6 studs with R20 batts is around R27 (RSI of 4.8) not 3.26.

            Looking at the summary, most of your losses are through your slab, bumping that up will make a big difference.

            I would also aim to get some rigid under the exterior walls with the metal cladding. A big surface area to build with a low R value assembly.

            Ventilation is about 8% of your annual heating loss with the 70% efficient HRV. In cold climate like yours, it will probably be worth your while to invest in a higher efficiency unit.

            Also, how exactly is your ceiling built? The HOT2000 summary is showing 2x4 with 6" of ccSPF plus 2" rigid. Something doesn't add up.

          3. jason_v | | #30

            Thanks, it looks like some stuff got mixed up. The exterior is either 2 inch EIFS or Zip R6 (1.5 inches) under metal cladding.. The ceiling is open web trusses with spray foam underneath, 3/4 inch plywood and a sloped EPS package. The effective R value was guestimated somewhat based on an averaging the depth of the sloped EPS package calculation and an estimated depth of the spray foam. We will go deeper on the foam if we can (i understand its not green). We decided to input the roof conservatively. We are not sure how much space will be consumed in the trusses yet.
            We will definitely go to R20 under the slab based on advice from here, it was previously recommended but i was trying to cost save. Dumb now that i think about it.

            We are now considering adding heat to the HRV and possibly putting hydronics into some the walls as well, depending on how all this shakes out.

            thanks again for all your help.

  3. BirchwoodBill | | #6

    Did they do a room by room breakdown at different temperatures? By supplement heat did they indicate additional heat emitters? Or did they need additional capacity at the manifold? I would ask how many BTUh per square foot did they use in the model? For warmboard, I used 15 to 20 btu per hour per sq ft. With water temps ranging between 83f to 103f. That allowed me to perform a heat balance for the micro zone.

    Which software package was used? Ms-excel, wrightsoft, loop cad?

    Was the calculation used to size a boiler or a heat pump? Heat pump performance drops off below -7f.

    1. jason_v | | #9

      I don't believe the manual J includes different temperatures, they used the lowest temp the house will be exposed to less than 1% of the time (in my case -27 C). I am not sure what software was used by each professional. The calculation would be used to size the heat source but we haven't gotten that far yet. What it looks like to me is they are saying that at max floor temp (85 F) the radiant floor would not be keeping up to the BTU's lost by the house, resulting in the need for an additional (supplemental) heat source in order to make up the lost BTU's the floor cannot provide. This is the part that makes no sense to me.

      1. BirchwoodBill | | #27

        I am in Minnesota zone 6a, with design day at -13f. My heat exchanger is set with outdoor reset from 83f to 103f which keeps things comfortable. Per warmboard spec, the floors are insulated with R20 and the rim joist is air sealed with 2 inches closed cell foam R13. Most of the time the boiler is modulating at 20%. What did you spec out for slab insulation? Also, did the PE break out the DHW load? Gary Klein has new guidelines for DHW usage, that may reduce the heat demand.

        1. jason_v | | #31

          thank you. We are going to bump up the sub slab to R2o, the perimeter is 4 inches of EPS. I will look into those guidelines, thanks.

  4. Jon_R | | #12

    85000/3000 = 28 btu/sq ft. This is about the limit of what full coverage radiant floor heat can provide without being excessively hot. But some rooms might require more - so the warning makes sense. At this output, you will have warm toes with bare feet and slightly warmer than optimal feet with normal footwear. Lower floor outputs (likely) will produce comfortably warm toes with normal footwear.

    Fan coils are a supplemental heat option with a response time advantage. And they can be located under windows, increasing comfort. Radiant walls or ceiling are also possible.

    1. jason_v | | #14

      thank you

  5. Expert Member
    NICK KEENAN | | #13

    With 3ooo square feet and a heating load of 85k you're looking at 28 but per square foot on average, which is a lot. Some rooms will be more.

    The design limit for floor heat is surface temperature, beyond 85F or so it starts becoming uncomfortable and can cause long-term problems in people's feet. You get about 2 btu/SF for every degree of temperature difference between the air and the floor, so that 28 btu/SF means a 14 degree difference, or a maximum room temperature of 71F with a floor temperature of 85F.

    That doesn't sound too bad to start, but the problem is those are averages. You're never going to get heat under 100% of the floor. And if your average temperature is 85F you might have a situation where part of the floor is 90F and part is 80F. The 90F part is going to be unacceptably hot.

    In the real world you have to assume that the average temperature is more like 80F, and you might be able to heat two thirds of the floor. That gives an upper limit in the 10-15 btu/SF range.

    1. jason_v | | #15

      thank you

  6. charlie_sullivan | | #16

    I don't know what jurisdiction your zone 7A build is in, but I doubt an R35 roof is to code. If it's a custom build, I would think you would have a choice to do better. R70 would be a good target, but at least meet code.

    1. jason_v | | #17

      Thank you. code for an unvented flat roof assembly with HRV is R28.5.

      1. charlie_sullivan | | #28

        That is probably be a misunderstanding of the code. You might be looking at Table R806.5 (2012 IRC) which is about roof assemblies with some of the insulation above the deck and some below. It lists, for those assemblies, how much needs to above the deck. You still need the remainder to meet code requirements for total insulation, added below.

        I'm not sure what the HRV has to do with it.

        1. jason_v | | #32

          The code specifies different levels of insulation in some circumstances wether you have an HRV or not. Why this is the case is above my pay grade.. Our roof assembly is in fact how you stated. Thanks for your reply.

  7. Expert Member
    Michael Maines | | #20

    You live in a place with over 10,000 heating degree days and a design temperature of -25°F and you're ok with R-20 walls and R-28 roof? Wow, I wish I had your money to burn for energy every year! R-30 to R-40 walls and R-50 to R-60 walls make good financial sense down here in climate zones 5 and 6. You would significantly reduce your heating load (and associated cost) if you used a reasonable amount of insulation. Even the southernmost US states have more stringent envelope requirements. I'd say no offense intended, but I'm just shocked that anyone would build a new house with such a miserable envelope.

    1. jason_v | | #25

      Wall assemblies will be closer to R 30. roof will be closer to R35. Currently administration fees and taxes are over 75% of any energy bill, so reducing the actual energy used has limited benefit.

      Also, we know that adding insulation past a certain point has a steep reduction in benefit as showed by

      With the cost of construction supplies right now going double stud etc and framing a super deep roof assembly simply does not make any sense. I feel like we landed in a happy place for us considering all the factors.

      Thank you for taking the time to reply.

      1. charlie_sullivan | | #29

        If you are worried about insulation costs, moving from EPS + spray foam to EPS plus filling the cavity you have with fluffy insulation will be cheaper overall and get you to a higher R-value.

        1. jason_v | | #33

          Thanks, we will consider it.

      2. Expert Member
        NICK KEENAN | | #34

        With 10,000 HDD, and a 70K load at design temperature of -20F, I get 18,666 Btu/HDD. So a back of the envelope calculation is 190 million BTU per year. I pay $1.12 per therm, which is 100K BTU, which would give an annual heating bill slightly over $2,000. So yeah, improvements that cost tens of thousands of dollars aren't going to pencil out. But there is some room for improvement.

        This is where having an energy model really pays off, you can punch in a change and see how it affects your annual energy bill.

        Two things I would look at: first, more insulation under the slab. Second is making the garage "cold," ie outside of the heated part of the house. So the intersection between the garage and the house is completely air sealed and insulated. This is a good idea anyway to keep fumes from the car out of the house. If you also insulate the exterior walls of the garage it will gain some heat from the house and stay somewhat above exterior temperatures on the coldest nights.

        1. jason_v | | #35

          Thanks, we have energy modelling it is attached above. We are approximately 27% better in terms of energy use than the average house up here currently. We are going to go up to R20 under the slab. Since our garage is attached it will be insulated to the same degree as the house and sealed etc as you suggest as per code. Again the garage is not our primary issue, the issue is the radiant heat not being able to supply enough heat in certain rooms, which another fellow on here has been greatly helpful with.

          Thanks again for taking time to reply.

  8. Expert Member
    Akos | | #36

    Once you get the inputs in Hot2000 fixed to match your actual wall assembly and update the slab insulation, your heat load will be somewhere around 30000BTU which is about 10BTU/sqft. This means that your floor heat setup will easily carry the house, there will definitely never be need for supplemental need.

    You'll find is that if you go with a fully hated slab, the slab will never be warm. Even at your design temperature, it will only be around 75F, which is too cold for barefoot comfort with concrete.

    I would look at optimizing the layout so that most (if not all) of the floor heat is under areas with high traffic such as hallways, kitchen, bathrooms and under windows/doors. This would let those areas run at higher temperature which will be much more foot friendly.

    If you are doing a hydronic heat for the HRV, make sure it is a post heat (it heats the fresh air feed of the house), not a pre-heat. You can also design out the need for this with good register layout (register up high, blowing along the ceiling and away from people) and higher efficiency HRV.

    1. Expert Member
      NICK KEENAN | | #37

      Akos, the one caveat I would add is you still have to look at the room by rooms. If you have a room that has a lot of window and not a lot of floor that room may fall short.

    2. jason_v | | #38

      Thanks. Yes we are going to do post heat if we go that route. We are not concerned about the floor not being warm, we are doing radiant because of my daughters allergies (and its what i prefer) and we are polishing the floor because its durable and i don't want to spend 40K on wood or tile. Thanks for all your help.

    3. jason_v | | #48

      You (Akos) were correct. With inputs corrected via meeting with the builder and designer and supplied to another engineer, we got a total heating load of 42K btu.

      thanks everyone for your help.


  9. paul_wiedefeld | | #39

    Jason, have you decided what your heat source will be? Earlier you alluded to lower energy costs, is that true for electricity?

    1. jason_v | | #40

      it will be a boiler, gas up here is much cheaper than electricity.

      1. paul_wiedefeld | | #41

        Can you share the prices of the two?

        1. jason_v | | #42

          natural gas is 3.30 a GJ, electricity is 0.07 per kWh

          1. charlie_sullivan | | #44

            That gas price is 35 cents/therm for people like me who think in old fashioned units. Both are quite cheap!

          2. paul_wiedefeld | | #45

            So cheap! Do they cover distribution with a fixed fee?

          3. Expert Member
            Dana Dorsett | | #49

            At 95% combustion efficiency that's 95KBTU/therm or (1,000,000/95,000=) 10.5 therms/MMBTU (million BTU), which at 35 cents costs ~$3.50/MMBTU.

            At a seasonal average COP of 3 with an air source heat pump that would be 3 x 3412 BTU/kwh= 10,236 BTU/kwh, and (1,000,000/10,235=) ~98 kwh/MMBTU, which at 7 cents cost $6.86/MMBTU, nearly twice as much.

            It's hard to make a financial argument for going with heat pumps, but at such a low price point one might still consider pricing for yourself the currently un-priced externalities involved with natural gas & coal extraction & combustion vs. the lower net externalities of electricity and refrigerants used in the heat pumps, if one assumes that Alberta/Canada get on board with international targets for reductions in fossil fuel consumption by 2030 and 2050.

            At the moment Alberta has some of the highest-carbon electricity in Canada, but with the available wind and solar it doesn't have to remain that way:



            Southeastern Alberta has some of the best wind & solar potential of ANY location in Canada, and better than much of the US:


            Wind & solar are already cheap and getting cheaper. By 2030 it would be cheaper than keeping existing coal & gas fired equipment going, especially when the fossil burners are running at a reduced capacity factor due to zero marginal cost renewables. The levelized cost assumptions in most analyses assume unrealistically high capacity factors for existing or newly built fossil plants, all of which are doomed to becoming stranded assets well before 2050 (if not by 2030) due to the still rapid learning curves and already low cost of solar + wind + batteries. Solar/wind + batteries are already cheaper & better than fast ramping peakers for maintaining grid stability, and will be cheaper than a new combined cycle gas plant by 2030.




            With such a cool 99% design temp one might assume that an average COP of 3 isn't very likely in that climate, but the average mid-winter binned hourly temps in Calgary are a relatively temperate ~+20F, a fairly comfortable spot for cold climate air source heat pumps & reversible chillers. With sufficiently low water temp requirements it's not really all that hard, even if one has to plan on electric boiler backup for the few hours of extreme Polar Vortex disturbance coolth.

        2. jason_v | | #46

          yes a fixed fee for distribution.

          1. jason_v | | #50

            Awesome thanks for the calculations and perspective. We are considering a heat pump water heater for domestic, and are running chases and wiring for future solar panels on the roof (flat roof assembly). Thanks for your reply i really appreciate it

  10. user-2890856 | | #43

    Sounds alot like someone has a data entry issue . 85k seems almost impossible for a 3000 sf home with that type of insulation .

    You stated that several people did the load calc and came up with similar results which leads you to believe it is correct . Don't believe it is correct , expert or not , not very long ago and in fact still to this day , engineers and contractors oversize systems and equipment regularly , the number is almost definitely wrong .

    Find a new company to design your system , it may cost you some money , but free stuff is worth exactly what you pay for it . My point is that any company doing this correctly or well should not have a disclaimer about this room MAY REQUIRE , it does or it does not , Period , they are suppose4d to tell you , not guess .

    If you'd like to discus , my e mail is [email protected] and I speak to folks for free at least 1 time .

    1. jason_v | | #47

      thank you

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |