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

Passive solar and flooring choices

mrigney | Posted in Energy Efficiency and Durability on

I suspect I know the answer I’ll get from this crowd, but I’m going to ask it anyway. I’m in Climate Zone 3. Have a passive solar design that I’m about to build. Southern glass exposure is about 7%, so pretty low for a passive solar design due to me being relatively far south. Originally was going to do a finished concrete on the first floor south side. For various reasons having to do mainly w/not having many (any?) folks in the local area who I felt comfortable with doing interior concrete work, we decided to look at wood-look porcelain tiles. Well, found out today that if we want planks 3′ long or longer, installation is ~$8.15/sq ft (slab on grade + long tiles really jacks up the price). So, it just got me to thinking. Will I really notice much different if we went with wood floors?

For some reference numbers, we’re shooting with our builder for ~1.5 ACH50 and a HERS score in the 50s to give an overall idea of construction. We’ll have Zip sheathing, external rigid foam board, etc.

The south facing area is a kitchen/dining/living great room, about 700 square feet total. My assumption is that the area in the kitchen wouldn’t really do much from a thermal mass perspective since the floor in there will be shaded by cabinets/island/etc most of the day. That leaves me w/about 540 sq feet of dining/living area. Of that, some portion of the floor will be covered (couches, area rug, coffee table, piano, etc). If you estimate ~150 sq ft for that, you’re down to about 390 sq ft of exposed flooring.

I’m not sure that the difference in thermal mass/heat capacity of wood vs tile will be enough over that 390 sq ft to pay for the increase in cost of the tile (~$13/sq ft material + labor). Am I missing something? Don’t get me wrong, I still want to leverage the sun for as much winter warmth as I can, but it seems foolish to break the bank over it. Any thoughts?

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. this_page_left_blank | | #1

    The amount of extra useful heat you'll get from the tile vs wood would never, ever pay back the increased cost to install it. And in your climate, a good portion of the time that tile will be costing you rather than saving you anyway. Wood will be far nicer to walk on as well. This is from someone who lived on an uncovered concrete slab for a whole year, which is pretty much how tile on concrete will feel. It gets old very fast. Tile should be relegated to bathrooms (and commercial spaces). I wouldn't even put it in a kitchen, let alone a whole floor.

  2. GBA Editor
    Brian Pontolilo | | #2

    Hi user-7697971.

    I wish I was more informed about the nuances of thermal mass, but my understanding is similar to Trevor's and from what I have read and heard experts advise, flooring over a concrete slab has a minimal effect on it's ability to act in the way you want it to. Maybe we'll here from others with a better understanding.

  3. Expert Member
    Akos | | #3

    You are in warm enough climate that with any reasonable amount of insulation, your wintertime heating costs will very small. Passive solar just doesn't do much down there.

    Be careful with shading as the south facing windows in the shoulder season can add a LOT of cooling load.

    Thermal mass of the slab does help. On that front the extra bit of insulation of hardwood flooring (~R0.75) makes very little difference.

    If you really want to reduce that, you can go with one of the thinner glue down engineered floors, it would still be much cheaper and way more comfortable than tile.

  4. mrigney | | #4

    Thanks, guys, for their response.

    @Trevor Lambert - the payback thing is basically what I told my wife (mind you, I'm the physics guy in physics, develop/use thermal simulations--not of buildings--for a living). I told her that ultimately in this case it seemed (despite the architects recommendations), this was a case where the differences saved in heating cost during the winter would be miniscule compared to the extra paid for the tile, especially since it seems to me that a pretty decent engineered hardwood for ~$8.50/sq ft installed.

    And beyond that, per Martin's article on re-evaluating passive solar and some other things I've read, it seems that the general comfort (regulation of temp in the house, humidity, air quality, etc) is largely controlled by the quality of the build vs the flooring type.

    @Brian Pontolilo - How do I go from my "user-7697971" name to something more....identifiable?:-)

    1. GBA Editor
      Brian Pontolilo | | #5

      In the upper right hand corner, next to the search icon, there is a drop down. Choose account settings and you will be able to update your screen name.

      1. mrigney | | #6

        Got it fixed up. Thanks.

  5. Jon_R | | #7

    > general comfort (regulation of temp in the house, humidity, air quality, etc) is largely controlled by the quality of the build

    It's largely controlled by the HVAC system. On the other hand, passive thermal storage only works if you allow variations (usually large) in the temp of the house.

    1. mrigney | | #12

      Ok, I agree with that. I should've worded my original statement better. Maybe a better way to say that is that the ease with which your HVAC system can regulate your general comfort level is more strongly affected by things like air sealing, insulation, and other envelope choices made than by whether I have hardwood or tile floors.

  6. Expert Member
    BILL WICHERS | | #8

    The ability of a material to hold heat like a battery is what is known as the “specific heat” of a material, usually given in calories of thermal energy to a gram of the material with a temperature delta. I’ve worked with this before to make “thermal batteries” to backup large chilled water plants in Datacenters. Big chillers can’t restart immediately after a power interruption, so we need to limit the temperature rise in the equipment rooms during the time between the power interruption and the time the chiller can restart. The amount of time is known (it’s given in the datasheets for the chiller), and we know how much energy is going into the equipment rooms from their power consumption. That gives a total number of watt hours of energy that needs to be absorbed by the “thermal battery”.

    Using the specific heat of water, it’s possible to work out how many gallons of water are needed to absorb that energy to limit the temperature rise of the equipment rooms during the restart interval for the chillers. Why do I like to use water tanks instead of battery backup for the chillers? Water tanks don’t wear out like batteries do, they’re cheap (relatively), and there are no special safety requirements for storing large amounts of low pressure water. There is also essentially no energy loss with the water tank, but the batter backup system loses about 6% — 24 hours a day. It’s essentially a greener AND cheaper way to accomplish the goal.

    Hopefully my analogy allows you to visualize the energy storage physics a little.

    Wood won’t add much in terms of energy storage. Wood is low mass (compared to concrete or tile), and my guess is cellulose — the material wood is made from — has a low specific heat too, so a low capacity for thermal energy storage. What wood WILL do is slightly insulate the concrete (about R1 per inch), which will serve to slightly slow down the energy absorption and radiation to and from the concrete. Basically that means the wood will cause the temperature peak and valley (minimum) of the concrete to shift a bit later in the day compared to what the ambient temperature and sun cycle are doing. Since it’s only a little R value with maybe 3/8” to 1/2” or so of wood for a wood floor, you will see a very minimal impact on the thermal performance of your thermal mass material.

    The short answer is I wouldn’t worry about the floor much. Using a darker color for the floor might be enough to offset the slight insulating value of the wood.


    1. mrigney | | #9

      Good analogy, Bill. I have a pretty good grasp on the physics of the problem...I'm a physicist who builds radiative transfer models and thermal simulations for a living, but don't have experience applying it to this situation. Trying to get a handle on the "how much". I've run back of the envelope numbers, but it's not an overly simple problem. After all, whether you have wood or tile as your surface layer, you still have a slab underneath, so there is heat transfer between your surface and your slab.

      But yes, seems that the general consensus is it won't make a huge difference. Appreciate all the input! It was more or less the conclusion I had come to, but it's always good to hear it from someone else.

    2. mrigney | | #10

      Also, surprisingly, wood/cellulose has a higher specific heat than porcelain or concrete (about 1500 J/(kg*K) vs ~1000 J/(kg*K) for porcelain and about 1000 J/(kg*K) for concrete depending on your source).

      Conveniently, assuming a 3/8" thickness for the tile or wood, 1075 square feet of flooring gives you one cubic meter of material. This gives you about 2400 kg of tile or 750 kg of wood. So, for the entire floor unit, multiplying Cp by the total mass gives 1.125e6 J/K for the wood and 2.4e6 J/K for the porcelain.

      Fair to say then, it seems that it is the difference in mass more than the specific heat that aids the tile. There's just a whooooole lot more of it in a given space.

  7. Jon_R | | #11

    With passive solar, the challenge is getting the heat in and out of the thermal mass without an excessive delta-T (ie uncomfortable room). Say the sun is hitting the concrete directly, so R = 0 (to the top, a little more further in). Now add a R = .5 wood floor between the sun and the concrete. That's an infinite % increase in resistance to heat flow - not a slight one. Big differences also occur without direct solar heating (eg, 11F rise in room air temp vs 6F to move the same BTUs into the concrete). Thermal mass shouldn't be covered with anything - and a fan to reduce the air film R value is helpful.

    Despite the big difference, even uncovered interior thermal mass isn't worth much in a house that is kept comfortable - so go with the lower cost floor and use the savings more effectively elsewhere.

    1. burninate | | #14

      Sometimes it becomes quite important not to assume zero or infinity for certain things.

      150pcf standard concrete is R-0.07/in

  8. burninate | | #13

    Thermal mass works pretty well in the desert southwest because every night is cold and every day is hot and there's barely any moisture in the air; In climates without much water, there's a huge day-night delta, and odds are the desired room temperature is somewhere in that delta. Additionally, it just doesn't get that cold.

    In other places, it's questionable.

    In places with high humidity, like most of zone 3, day-night temperature delta is fairly small. Cooling using thermal mass in a humid climate requires precipitating water out of the air, which causes damp building materials and mold.

    Places with some moisture where it doesn't get very hot in the summer can take a thermal mass approach and add passive solar, and they're fine. Barely anyone in Germany has air conditioning. That's not the case for people dealing with the Gulf of Mexico maritime airmass.

    Passive solar is a double-edged sword; Some kind of active management is preferable, and lots of expensive, thermally leaky glass is required.

    PV has become so inexpensive that most people here have abandoned the whole passive solar paradigm. Even people with solar thermal collectors are dropping that idea in favor of slapping a PV panel up there and using electric resistance heat. It's so much easier to implement.

    The idea of thermal stores is, to me, one that still maintains lots of possibility, but instead of "Ten trucks of extra concrete and a window wall", they look like a roof full of solar panels and a pair of 200-gallon to 5000-gallon insulated water tanks, one kept below room temperature, one kept above room temperature (or above hot water temperature, depending on your goals), attached to thermal loops that give the house air conditioning (in a proper fixture, with a drain!), space heating, and hot water. Why? Because water is about a thousand times cheaper and considerably less explosive than large quantities of lithium ion batteries. A smaller one will get you through the day-night cycle, a larger one will get you through a hundred-year blizzard or even an entire season. Every junction of heat transfer to and from these tanks can be throttled at will, eliminating all the coupling in a passive PV system. If (as most engineers appear to) you want to make your own custom solution, I recommend looking in this direction.

    1. Expert Member
      BILL WICHERS | | #15

      Water also has the advantages of being safe, having no limit on the number of charge/discharge cycles, and basically being absolutely reliable. The downside is it’s not particularly space efficient — you need a LOT of tank capacity to do much. A typical wood boiler heater home with water storage may need 500-1,000 gallons or so to hold enough heat to last for a day. In my earlier example using chiller water systems, we usually need around 10,000-20,000 gallons to limit the system temperature rise to about 10 degrees in about 15-20 minutes, which was the approximate time needed for the chillers to come back online.


Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |