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
Musings of an Energy Nerd

All About Thermal Mass

Interior thermal mass can sometimes help lower energy costs — but in cold climates, it’s won’t help much

Promoters of high-mass walls, like this wall built of autoclaved aerated concrete (AAC), often boast that materials with a high thermal mass perform better than their dismal R-values indicate. The AAC blocks in this photo were manufactured by Aercon. According to Environmental Building News, an 8-inch-thick Aercon wall is rated at only R-11.5.
Image Credit: Martin Holladay

UPDATED on December 4, 2013 with a citation of recent research findings.

What’s the deal with thermal mass? Since manufacturers of materials that incorporate concrete often exaggerate the benefits of thermal mass, it’s easy to get cynical and conclude that the buzz around thermal mass is all hype. But in many climates, it’s actually useful to have a lot of thermal mass inside your house. Just keep in mind that thermal mass may not be as beneficial as its boosters pretend.

Thermal mass is a solid or liquid material that can store heat. Most of the objects inside your house can be considered thermal mass, including plaster, furniture, books, and canned tomato soup.

The specific heat capacity of building materials varies. In general, denser building materials have a higher specific heat capacity per unit of volume than less dense materials, which is why concrete, stone, and gypsum wallboard are more likely to be used to provide extra thermal mass than wood.

Three analogies: cistern, frying pan, truck

A building with lots of thermal mass on the interior side of the insulation may have lower energy bills than one without as much thermal mass, for reasons I’ll explain soon. But it’s important to point out that thermal mass can’t heat or cool your house. It’s just plain old concrete. To heat and cool your house, you still need HVAC equipment.

Here are some analogies:

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. user-1068982 | | #1

    Excellent article.
    Thanks for clearly illustrating the importance of a 24-hour temperature cycle when considering thermal mass. It makes alot of sense!

  2. Expert Member
    Dana Dorsett | | #2

    Clarification of IRC prescriptive interior vs. exterior R
    Martin writes:

    " For example, wood-framed walls in climate zone 5 need to be insulated to R-20, but concrete block walls can get away with R-13 (as long as “more than half the insulation is on the interior” of the wall). "

    Under the IRC there are TWO prescriptive R values, and the higher value applies when more than half the R is on the interior, thus by default the lower value applies when half or more is on the exterior. For zone 5 mass walls are specified as R13/R17, so the R13 only applies if half or more of the insulation is on the EXTERIOR, and R17 is required if more than half is on the interior, which is the converse of Martin's interpretation. (And yes, parsing all of that CAN be confusing!)


    Read note "i" carefully.

    Also note, that for framed walls the prescriptive values for zone 5 is R20 (cavity-only) or R13 cavity + R5 continuous insulation.

    At a typical 25% framing fraction the whole-wall R of the sheathing + studwall for R20 cavity fill comes in around R14 as does R13 2x4 construction with R5 of continuous insulation (unbroken by studs) on the exterior.

    That means that a mass wall with more than half the R on the interior is required to have a substantially HIGHER whole wall value than a stick built wall (R17 rather than R14), but if at least half the insulation is on the exterior it can be modestly lower-R (R13 rather than R14.) There actually IS method to the madness in the IRC prescriptions, since with higher interior R on a mass wall uses more energy in a zone 5 climate, as noted.

    Curiously, that's at variance with TABLE N1102.1.3, using the whole-wall U-factor method, where U0.057 (R17 whole-wall) would be required for low mass construction, but U0.82 (R12 whole-wall) with half or more of the insulation on the exterior, or U0.065 (R15) for mass walls with more R on the interior. The methods are not totally in sync.

    But in the case of ICF, most current offerings start at R20 (R16 if you really hunt), and none of the of the lower-R ICFs have more than half of the insulation on the exterior- they would all meet code min for zone 5 using either method.

  3. GBA Editor
    Martin Holladay | | #3

    Response to Dana Dorsett
    It looks like I was correcting my editing mistake at about the same time that you were composing your comment. I recently corrected the bad sentence -- you'll see that it now reads, "For example, wood-framed walls in climate zone 5 need to be insulated to R-20, but concrete block walls can get away with R-13 (as long as at least half of the insulation is on the exterior of the wall)."

    Thanks for your comments, and for your sharp eye.

  4. Expert Member
    Dana Dorsett | | #4

    The other common exaggeration...
    ICF companies are also prone to mentioning (without qualifying) the 40F center temp R values of EPS rather than the 75F number. This is because Type II EPS runs about R4.5/inch when it's average temp is 40F, about 7% higher than the R4.2/inch when it's average temp is 75F. Were it being sold as insulation rather than a wall system, they would be required to label at the 75F number. What's never stated in the ICF literature is that in order for the foam to average 40F in an 50/50 ICF, the temp of concrete would have to average 40F. In a house that's maintained at 70F it means the outdoor temp would then have to average +10F (!)

    While that's a mid-winter average one might expect in US climate zone 7 or maybe the cold edge of zone 6, nowhere is that the outdoor average temp across an entire heating season in ANY lower 48 states, or for most of the population Canada & AK. Using the 40F R-values for EPS without clearly qualifying what that means is another sleight of hand to add another R1-2 of implied performance (that isn't really there) to an R20-ish ICF. They seem to get away with it because they're selling a wall system, not insulation, but it's pretty sleazy marketing, IMHO.

  5. dankolbert | | #5

    Excellent post
    It's one of those subjects people tend to get all gooey about - the perpetual motion machine of building science. Thanks for demystifying it.

  6. DennisDipswitch | | #6

    Windows and Thermal Mass
    Every time I read about thermal mass and south facing windows,I always wonder what the net effect is.Doesn't seem to work too well during the night time hours.And it would seem that you get an added counter productive effect during the seasons that you do not want any added heat.

    Martin,you give a quick mention of this in the article,but is there ever a circumstance,that the south facing glass/thermal mass floor and wall scenario is ever a net positive in energy usage?Or is the best possible scenario a situation where the view is to the south and a lot of glass would be included there anyway,so might as well capture some solar gain?

  7. GBA Editor
    Martin Holladay | | #7

    Response to Dennis Dipswitch
    Q. "Is there ever a circumstance,that the south facing glass/thermal mass floor and wall scenario is ever a net positive in energy usage?"

    A. Yes, of course. If you choose the right glazing, a window can gain more energy than it loses on an annual basis. (This is what you want in Minnesota, but it probably isn't what you want in Miami.)

    The trick is to make sure that you have the right type of glazing, the right amount of glazing, and the right roof overhang -- so that the heat gain occurs at the times of the year when you want the heat. If the house has solar heat gain in July, it's usually unwanted.

    These elements of passive solar design can be tweaked with a good energy modeling program like PHPP. For more information, see Windows That Perform Better Than Walls.

  8. watercop | | #8

    ICF in the humid south
    Lots of good stuff in the original article and follow-on comments. Something I haven't seen mentioned that I would like to run up the flagpole for reaction / analysis:

    ICF seems confer an extra benefit in the hot humid south. We here endure about 4 continuous months, call it 5/15 - 9/15, where the overnight lows are around 70-75, and that low is dry bulb, wet bulb, and dew point; in other words, nearly 100% RH obtains from midnight until dawn.

    With outdoor temperature and indoor setpoint about matched, AC runs little during those hours, because, by midnight AC has removed most of the solar heat gained during afternoon by a typical low mass frame house as well as the internal load from the evening's cooking, bathing, media. and lighting activities.

    From about midnight onward, indoor RH rises owing to infiltration and respiration by sleeping occupants, some of whom may react to the humidity by reducing AC setpoint, causing the house to transition from hot and muggy to cold and clammy; still uncomfortable, but with the added zest of mold risk in some building assemblies.

    Typical AC may cycle only once or twice after midnight, and barely long enough to make a dent in RH, since the thermostat tends to be satisfied just as the coil gets cold enough to dehumidify.

    ICF seems to help with all that, in two ways:

    One - In the manner of "traditional" thermal mass, it shaves the peaks from torrid afternoons during summer and frigid nights in winter, allowing HVAC equipment to be a bit smaller, costing less up front and operating more efficiently owing to longer on cycles.

    Two - specific to humidity - I suspect, but haven't the data to prove, that the concrete "meat" between the foam "bread" within ICF stays at at temperature of 80-90 all summer, west walls at the higher end of that range, and north walls at the lower end.

    Running some numbers, suppose an ICF home (such as mine) has 5000 SF of wall with average 85*F concrete in contact with R10 worth of foam and drywall (ICF mfgs might argue the number is closer to R12, but there are no end of small bridges and penetrations, so I'm going with R10 for the example...maybe R8 is closer to the mark...)

    With cooling setpoint of 75, that translates to 5000 Btu / hr sensible moving through my walls, 24/7, all summer. That, in turn, results in about a minimum 20% duty cycle for my AC, 24/7. That effect, coupled with the traditional thermal mass of the home, causes the AC to run for 10-20 minutes 3-4 times during the midnight - dawn near-100% RH period, resulting in superior humidity control and comfort compared with typical frame homes.

    Right-sized two stage AC coupled with reasonable infiltration (~3 ACH 50, tight, nowhere near PH standard) results in RH being held to 45-50% throughout cooling season, which in turn allows drybulb setpoint to rise a bit above typical 73-75 to about 77-79, reducing HVAC cost even more and reducing building assembly mold risk as well.

    Thoughts, anyone?

  9. user-942951 | | #9

    Thermal mass of lightweight 'insulating' concrete products?
    Am I correct in my understanding that lightweight concrete and CMU products (such as the AAC illustrated) are less effective thermal mass than regular concrete or CMU, due to their lower density? So should these products be called 'high mass' or not? More 'medium mass' maybe?

  10. GBA Editor
    Martin Holladay | | #10

    Response to Curt Kinder
    Your theory is intriguing. You may be right, but it would take some research to verify it -- for example, a research project involving an ICF home with embedded sensors.

    It's certainly true that a high-mass wall tends to shift some of the air conditioning load from the late afternoon to nighttime hours.

    I think it's ironic that you are proposing a somewhat counterintuitive justification for the use of ICFs in a hot climate -- namely, "You'll have hot concrete in your walls, all summer long. In fact, the concrete will be so hot that it will make your air conditioner kick on in the middle of the night!"

    In Houston, researchers trying to come up with a solution to high indoor humidity problems had good success by simply installing a $250 stand-alone dehumidifier somewhere in the house.

  11. GBA Editor
    Martin Holladay | | #11

    Response to Andy Parkinson
    In general, you're right. Denser concrete products have a higher specific heat capacity per unit of volume than less dense concrete products.

    Two observations:

    1. For any given application, there is an optimum level of thermal mass, beyond which any additional mass provides no energy performance benefit. There's no need to pay for concrete that doesn't do anything. Modeling the optimum level of thermal mass for any given application gets complicated, but often rules of thumb and experience give results that work just as well as modeling.

    2. The exception to the "denser is better" rule of thermal mass concerns phase-change materials.

    Phase-change materials used in construction are usually some type of paraffin or wax. (Small spheres of these paraffins or waxes can be incorporated into gypsum wallboard or cellulose insulation.) These materials have the ability to absorb and then release heat by changing phase from solid to liquid and back. The materials are chosen because they have a melting point that is close to the temperature setpoint of the typical heating or cooling systems.

    During this phase change from solid to liquid, a material absorbs heat from the surrounding environment. Conversely, in cooling from liquid to solid, heat is given off to the environment. The energy that can be stored and released during the change of state occurs over a very narrow range of temperature. During the change in physical state, the material itself remains at nearly constant temperature until the phase change is complete. (Remember 8th grade science class, when you put a thermometer in a pot of water, and then brought the pot to a boil?)

    Phase-change materials only provide benefits for indoor environments where occupants are willing to allow indoor temperatures to range above and below the thermostat set point.

    Phase-change materials behave like thermal mass, but store more heat per unit of volume or per unit of weight than concrete or water. So these materials are an exception to the "denser is better" rule.

    For more information on phase-change materials, see Storing Heat in Walls with Phase-Change Materials.

  12. Expert Member
    Dana Dorsett | | #12

    Response to Martin...
    "In Houston, researchers trying to come up with a solution to high indoor humidity problems had good success by simply installing a $250 stand-alone dehumidifier somewhere in the house."

    The definition of "...good success..." get's a bit fuzzy when you're converting a latent load directly to a sensible load, inside the thermal & pressure boundary of the house, in a climate where you'd rather be dumpng that latent heat outdoors rather than raising the indoor temp with it (along with the heat dissipated in the motor & compressor, which are all indoors.)

    I like the Daikin Quaternity mini-split approach of having independently settable relative humidity and temperature setpoints, and proprietary valving on the interior coil which allows it to hit it's marks on humidity even without sensible cooling. (I'm told there are some ground source heat pumps with similar functionality, but have yet to look into them in any detail. ) In this case the latent heat and compressor-motor heat are all sent outdoors where they belong, not adding to the sensible cooling load.

  13. GBA Editor
    Martin Holladay | | #13

    Response to Dana Dorsett
    It's all fine and good to like the Daikin Quaternity, which I don't doubt works well. But the Daikin Quaternity is hardly an affordable solution for an existing house with high indoor humidity.

    It takes energy to remove moisture from the air, and there is no way around that fact. Running a stand-alone dehumidifier takes energy, and of course it dumps a little extra heat indoors. Most central air conditioners are oversized, however, and easily handle the extra load from this small appliance. Moreover, if the central AC runs for a few extra minutes each cycle, that helps to lower indoor RH as well.

    No solution is perfect. I don't doubt that a Daikin Quaternity is more efficient than my suggested solution. But for $250, the stand-alone dehumidifier approach is hard to beat.

  14. Expert Member
    Dana Dorsett | | #14

    I'm familiar with the approach....
    Since I have effectively no sensible load most of the summer, even with the ever-tightening house the place still needs dehumidification (especially in the basement where there's an as-yet uninsulated slab), and the $250 standalone dehumidifier does the trick (to the tune of a measured >300kwh/year- it's a significant plug load.) It's contribution to the almost non-existent sensible load isn't something that needs to be actively pumped out of the house- the basement rarely breaks 72F even with the dehumidifer and other plug loads.

    In TX the both the latent & sensible loads are much higher than near me in central MA, with no "passive cooling" slab to soak up the additional sensible load. I get that the central AC can handle it, but was pointing out that it's less than ideal. (To be sure on new construction you'd like to be able to do something different.), but it's certainly true that from a retrofit point of view a best-in-class mini-split (or GSHP) isn't going to be in the cards. In TX there's a pretty good case for applying the $250 cost of the dehumidifier toward a heat pump water heater, pumping the latent and compressor-motor heat into the hot water tank where it does some real good, though it's not exactly under dehumidistat control. Every house will be different...

  15. Tedkidd | | #15

    Mass effect offers almost no benefit in cold climates?
    Very interesting topic, thanks for posting this information!

    Is load linear to temperature? Or does it look like a mpg curve (as speed increases mpg goes down), having sort of geometric relationship?

    Mass seems to have this flywheel or momentum effect. Doesn't thermal mass provide a buffer to worst case load? If load is not linear, then wouldn't the buffering effect of mass create lower load experienced by buffering temperature spikes?

    If the temperature ranges between 10f and 30f, would a high mass structure experience a load of 20f?

    Also, can you change design temperature for high mass buildings and install smaller equipment, providing further benefit of the type Curt suggests?

  16. GBA Editor
    Martin Holladay | | #16

    Response to Ted Kidd
    Q. "Is load linear to temperature?"

    A. Good question! As far as I know, it is. But one has to consider temperature like a physicist, and think in terms of degrees Kelvin, with absolute zero as one's starting point.

    So when the outdoor temperature is 40°F, and you want to heat your house to 70°F, you have to remember that the outdoor temperature is 278°K, and you are raising the temperature to 294°K.

    When it's cold outside -- let's say that the temperature is -5°F, or 253°K -- the outdoor temperature is 91% as warm as it was when the temperature outdoors was 40°F (278°K).

    Q. "Doesn't thermal mass provide a buffer to worst case load?"

    A. Yes, and this benefit occurs with all houses -- not just houses that include concrete. Every house includes partitions and drywall, and most include furniture and books and kitchen cabinets and appliances. Because of this thermal mass, you never really need HVAC equipment to handle the worst-case load. In theory, Manual J knows this.

    However, Manual J includes a fudge factor, so it is extremely rare for anyone to actually specify HVAC equipment that is comes close to the actual load required on the worst-case day. For one thing, you can't even buy a furnace small enough to match the Manual J load on a well insulated house. So it's really hard to take any advantage of the load-shaving mechanism you describe.

    Where I live, the temperature occasionally gets down to -40°F. But that happens rarely. Many engineers might use a design temperature of -30°F for this part of Vermont. On the rare morning when it is -40°F, the thermal mass of the house prevents occupants from being cold. With any luck, the temperature will warm up to -30°F before long.

    With lots of thermal mass, you might think that you could use a design load of -25°F. And you probably could. However, there will be an occasional day when the thermometer starts out at -40°F, and warms up to -18°F at noon, and then drops down to -28°F by dinnertime. If you have many days like that, you might be getting chilly after a while.

    But really, since equipment is always oversized, it's very hard to save money with this trick. The concrete is expensive, and there aren't any cheap, small furnaces to save you any money.

  17. watercop | | #17

    Good building science rarely lends itself to sound bites...
    ...But I guess I asked for it.

    ICF causes some highly welcome late night HVAC operation during hot weather. I suppose that could be viewed as a negative, except that the extra hour or so operation after dark is more than offset by the ton or so shaved off the design load and also by the substantial delay and reduction of operating time during sunny afternoons.

    I'm familiar with the Houston study. I agree that portable dehus can solve certain problems at certain times but I cannot advocate their continuous operation unless all other humidity control strategies and techniques have been exhausted or ruled out: We advocate a ten point defense-in-depth approach to controlling humidity (WITHOUT separate active dehumidification appliances), to wit:

    • Minimize enclosure air infiltration
    • Properly size HVAC system to minimize short cycling - endeavor to install the smallest feasible system.
    • Ensure ductwork is configured to provide individual room design air flows - stave off thermostat wars
    • Ensure ductwork is within conditioned space or at least minimize duct leaks to / from unconditioned space
    • Avoid single stage HVAC systems that short cycle during part load conditions
    • Include controls that modulate system airflow (CFM / ton) in response to humidity
    • Manage point sources of humidity with ventilation - encourage use of bath vent fans by installing very quiet models controlled by timers and / or motion. Ensure range hood is properly sized, selected, positioned, and ducted so that it both works well and is reasonably likely to be actually used.
    • Discourage use of continuous fan in cooling mode.
    • Discourage use of natural gas or propane for cooking.
    • Evict panting dogs, sweaty children, and thirsty houseplants to the extent possible during summer months.

    Specifying a dehumidifier before or in lieu of implementing the ten steps above amounts to an expensive, irresponsible band-aid.

    In my experience portable and central dehus tend to have flaky, inaccurate controls, consume inordinate amounts of energy, add substantial sensible load, and confer additional system complexity. All that and they are loud and unreliable. No worthy building scientist would specify one before attacking the roots of a humidity problem as outlined above.

    Note to Dana - my HPWH just finished a continuous run from 7:30 until 10 PM recovering from the evening's multiple showers and along the way made the basement zone the driest in the house.

    I'm not totally opposed to dehus...we just used one to drive RH down from 70% to 40% in a space being drywalled during 3 days / 8 inches of rain. 40% RH greatly reduced time between mudding cycles, allowing drywall crew to stay productive during the storm. As I wrote earlier, dehus do have their uses...

  18. GBA Editor
    Martin Holladay | | #18

    Response to Curt Kinder
    I can't argue with any of your suggested measures -- they all make sense. You're right, of course, than any suggested solution to an indoor humidity problem must take a whole-building approach.

  19. Expert Member
    Dana Dorsett | | #19

    Linearity of loads
    Heating loads are only approximately linear with temperature differences between interior & exterior temps, with MANY secondary & third-order aspects to screw it up. The U-factors of low-E windows is VERY non-linear, with the radiated portion of the loss being a function of the difference of the fourth powers of the absolute (relative to absolute zero) of the temperatures on each side of the low-E surface. The U-factor of insulation materials is also non-linear with temperature & delta-T. Mid to high density fiber insulations increase in R with lower outdoor temps. Foam insulations also have highly non-linear R with temperature & delta-T- polystyrene goods increase in R with falling average temp, falls with rising temp, and it's by more than 10% across the annual temperature ranges seen in most of the US. Polyiso falls somewhat with both rising OR falling temp around a middle sweet-spot. Then there are the not strictly air temperature aspects like wind-washing, wind driven air infiltration, passive solar gains, etc.

    But despite all of these effects, linear approximations based on temperature alone are still pretty good across typical temperature ranges one would expect, though not better than 10% accuracy at any particular set of indoor & outdoor temperature conditions across the spectrum of other possible second & third order effects. It's definitely NOT a geometric or exponential fit, more of a linear fit to the average, but the noise coefficients are large.

    For example, in a house with a lot of south facing glass the heat load at 30F outdoors, 70F indoors may be near zero or even NEGATIVE at noon on a sunny calm day when there's a reflective snow cover boosting the solar gain, but quite substantial a 2AM during a 40mph blizzard. The temperatures are the same, the heat load isn't.

    Cooling loads are even less linear with temp, with solar gain through glazing being a sometimes dominating secondary effect, followed by outdoor air humidity & air infiltration & ventilation rates.

    Curt: Heat pump water heaters are the best dehumidifers EVER, eh? In a steamy gulf-coast climate you just can't beat the double-duty aspect!

  20. watercop | | #20

    Yes times two down here
    Windows nearly always dominate cooling load down here, unless a home is hopelessly leaky. Window cooling loads are about on par with penalty caused by ductwork in unconditioned attics.

    While I routinely specify / retrofit heat pump water heaters into unconditioned garages, saving clients approximately $100 per person per year, the absolute best place for an HPWH down here is somewhere within the conditioned enclosure, where the home can make use of the free cooling and drying occasioned by water heating.

    My own HPWH routinely maintains our basement 5*F cooler and 10% RH dryer than the rest of the home during shoulder seasons while it totes the whole water heating load (~4 hours operation / ~3 kwh / day) We are in the process of finishing the basement for a teenaged child, but I routinely threaten to move down there out of envy for conditions best suited for a displaced New Englander...

  21. lutro | | #21

    50% improvement is significant
    This article says several times things along the lines of "Studies have shown that thermal mass can provide heating energy savings in only a few areas of the country." However, both the linked ORNL and "Mass Confusion" articles say that in the worst climate modeled, Minneapolis, high thermal mass will give a roughly 50% improvement in the Dynamic Thermal Performance, with an R-17 wall. 50% improvement is a lot, particularly since this is the worst example. The improvements are even better in Denver, Miami, Washington, and Atlanta, peaking with Phoenix, as mentioned. The effect is far better than negligible in all locations, which represent a variety of climates across the country. These figures appear to be averages for the whole year.

    The advantage of the high mass wall is greater in all locations for an R-17 wall than for an R-9 wall. Figures for low R-value walls (R-1.6 to 2.3) wall show a negative effect for high thermal mass in all locations except Phoenix. This might imply that advantages of high thermal mass could be even more significant, with levels of insulation higher than R-17, as many of us would prefer. However, the numbers for the R-13 wall are similar, and often slightly better than those for the R-17 wall.

    I would appreciate your comments on each of these issues.

  22. GBA Editor
    Martin Holladay | | #22

    Response to Derek Roff
    Here is a summary of the conclusions of the ORNL research: "Potential whole building energy savings, available when lightweight walls are replaced by massive walls of the same R-value, were calculated for 143 m2 (1540 ft2 ) one-story ranch houses located in Minneapolis, Minnesota and Bakersfield, California. For high R-value walls, up to 8% of the whole building energy could be saved in Minneapolis and 18% in Bakersfield when wood-framed walls were replaced by massive wall systems. Thermal mass layers must be in good contact with the interior of the building in these walls."

    You're correct that 8% energy savings in Minneapolis are not insignificant. My main unanswered question concerns the assumptions made for the control house -- namely, the "one-story ranch house with wood-framed walls." What rate of air leakage was assumed for this house? (For that matter, what rate of air leakage was assumed for the house with high-mass walls?)

    In other words, if this one-story ranch house wasn't built very well, the results may partly reflect the fact that most one-story ranch homes with wood-framed walls are poorly built.

    My second question is: what would be the most cost-effective way to lower the energy bills of the wood-framed Minneapolis house by 8%? Here's my guess: a little bit of air sealing would save 8% of the heating energy -- for a much lower cost than building concrete walls.

    When I have the time, I'll contact Jeff Christian or Jan Kosny at ORNL and ask about the assumptions made for the lightly built ranch house.

  23. user-1119494 | | #23

    how about longer thermal cycles?
    RMI claimed that their HQ, build with huge thermal mass, could even out seasonal fluctuations. Same claim was often made for other underground/earth-sheltered homes. This seems intuitively reasonable, even though I have not run the numbers...

    At the very least, larger thermal mass (like a large battery bank) would allow one to coast over extended extremely cold or hot periods.

  24. GBA Editor
    Martin Holladay | | #24

    Response to Dustin Harris
    Count me a skeptic when I hear that someone claims that their house has so much thermal mass that it evens out seasonal fluctuations.

    Of course, an underground house can take advantage of the fact that the soil deep underground has a temperature equal to the annual average air temperature. But if you are in a northern climate, and your house is surrounded with dirt at 48°F, you'll still need to heat your house all winter. (However, you will probably be able to get away without using air conditioning, as long as you minimize your windows -- which, if your house is truly underground, shouldn't be that difficult.)

    One thing is for sure: if you are worried about power outages, fuel shortages, or natural disasters, then lots of thermal mass on the inside of your home will help keep your pipes from freezing as you wait for the linemen to repair the electrical lines. But it's possible to achieve the same result with lots of insulation and attention to airtightness.

    There is no doubt that thick concrete inside your home has some advantages. For the average homeowner, however, the question is simple: are the advantages worth the high cost of the concrete?

  25. lutro | | #25

    Seasonal thermal storage
    Responding to Dustin's comment, I share Martin's skepticism. I have run the numbers on several different thermal mass systems that claim seasonal heat storage, and my calculations have never come close to matching the claims. In most cases, this is a good thing, from a comfort perspective, in the given designs. If they could really store significant heat energy from late summer to mid-winter, in their passive thermal mass systems coupled to the living space, then that living space would be overheated through October and November, and probably December, as well. I don't see how you could do seasonal thermal storage without an active system, and I haven't seen any of those that look cost effective.

  26. JIM BAERG | | #26

    Comfort too
    I really lucked out. I bought a 100 year old house 7 years ago. Storey and a half with double wythe brick walls on the main floor, so no stud cavity and 2.3-3" air gap between the bricks. It's always been cold in the winter, wonderfully cool in the summer (8400 HDD, good diurnal swings, but we can get temps up to 100F in the summer). Last fall, I blew the wall cavity with dense pack FG. Immediate noise reduction and comfort. We turned off the furnace a month ago, and are increasingly leaving the doors and windows open. Come summer, we'll regulate the doors and windows and stay cool.
    Can't beat it.
    Here's a question. Can you recommend any software that has good building input capability (with thermal mass), uses hourly weather data and does hourly output, so that I can mess with daily and weekly temperture swings? Annual energy summaries are useful, but I'm often concerned about worst case scenarios in my design work. Thaks.

  27. GBA Editor
    Martin Holladay | | #27

    Response to Jim Baerg
    I'm not really sure which energy modeling program will meet your needs; perhaps another GBA reader will make a suggestion.

    In the meantime, here is a link to an article that provides an overview of energy modeling programs: Energy Modeling Software.

  28. Perry525 | | #28

    Thermal mass - not really for homes?
    The thing with thermal mass is, it can only emit heat when the room air temperature drops below that of the emitter.

    When you use concrete as a thermal store, the store has to be completely insulated and separate from its surroundings, otherwise the input heat disappears into the far reaches of the structure and is so diluted as to be irrecoverable.

    Concrete mass is not really suitable for homes, it is mainly used in glass office building where the concrete absorbs heat and delays the moment when the air conditioning comes on, this is usually about six hours.

    The system works best where you have regular hot days and cold nights. Where the mass can be cooled overnight by opening all the windows making it ready for the next hot day.

    When you have several hot days and nights in a row, the mass doesn't cool and the system breaks down.

    There is no point in having dense walls or floors that are not warmed by the sun, they merely add
    cost and contribute nothing.

    If you decide to have mass you need floors that can be warmed in summers direct downward sun and perhaps, walls that are positioned to be warmed by the sun in spring and autumn. Keep in mind that the days are shorter now, the sun is low and it does not have much heat in the morning, and your windows will bounce a lot of the available heat as the suns angle sharpens.

    My sun lounge has floor to ceiling windows facing south and smaller windows facing east and west, this is because we like to see the sun rise and sun set, not because the sun adds much heat in the morning and evening. Morning sun lifts the temperature from 22C to sometimes 25C. Daytime sun can rise 22c to 40C or more - far too hot.

    We then retire to our north facing lounge where the temperature is a more comfortable 22-25C.

    With concrete floor and brick walls the residual heat from the sun in the sun lounge and other rooms often keeps the heating off for the whole of our home until after midnight.

  29. watercop | | #29

    Another aspect of thermal mass difficult to assign value
    When commissioning HVAC in a high mass home, I typically adjust the system for less frequent, longer cycles. Within the Honeywell thermostats we often use is a parameter, CPH cycles per hour, that governs this. It defaults to 3, but I'll set it down to 2 or even 1, taking advantage of the lower rate of temperature change characteristic of a high mass house.

    Longer less-frequent operating cycles by the HVAC system confers 3 benefits:

    1) System spends more time in steady state operation where it operates closest to rated efficiency

    2) Improved dehumidification - full dehu only begins when indoor coil drops to steady state temperature, and that takes many minutes, especially with higher SEER systems whose coils are larger.

    3) Fewer starts / stops - presumably less wear and tear on system.

  30. gman571 | | #30

    We're all cavemen
    Nice article. I like that it distinguishes the two properties. I think in the years to come, the building "science" industry may start talking about the thermal time constant, which is a metric that helps describe how quickly a house heats up and cools down.

    Drawing analogies from electronics, the time constant is influenced by both thermal mass and R-value. Although the time constant is the product of the two, it's relationship to the temperature of house is described by more complicated equations, which is beyond the scope of this thread. However, understand that as you increase either R or the thermal mass, you increase the amount of time the house will take to respond to a temperature difference.

    This is why people often confuse the two concepts and talk about an "effective" R value of thermal mass. However, they're different physical properties and every material has both. Most of us have a good feel for R value, which influences the speed of heat transfer. Thermal mass acts like a battery to store and release heat. It is more than concrete walls and earth married structures. Everything has mass, and it affects the thermal response (ie comfort) of the structure. The mass of an asphalt roof, for example, will keep your attic uncomfortably hot for many hours after the sun sets. To complicate things, a material's R value influences it's "effective" thermal mass and vice-verse. The fact that building codes meld the two together shows just how primitive the building industry is in their understanding of these principles.

    The challenge we have is that as we super-insulate a house, we drastically lengthen the time constant and our historical understanding of how a house should behave may no longer be accurate. For example passive gain becomes a fleeting commodity, and allowing your house to cool down during the overnight hours may not be practical.

    This illustrates that heating and cooling a house is a dynamic process. The demands constantly change with varying outside and inside temperatures, wind and solar load, etc. Modeling them with static conditions and linear equations can only be so accurate. More complicated models based on differential equations are used in other industries because they more accurately account for this. In the years to come, the building "science" industry will start to explore these concepts.

  31. Jon_R | | #31

    Good article
    I'd like to see more data on the effect on nighttime thermostat setback. Up to 15% used to be quoted as the savings for a low mass home, but with better insulation, this figure needs to be revised downwards. Is it still viable and does its benefit outweigh interior mass? I should add that using % is misleading - btus saved would be much better.

  32. jackofalltrades777 | | #32

    Desert SouthWest
    As mentioned here, the desert SW is the perfect place for mass walls like ICF. Hence one of the reasons why I am building with ICF. We get vast diurnal swings out here at 4,800 feet elevation (Northern AZ). During winter It can be 65F by 4PM and then by 6AM it will be 25F. During summer it can peak at 95F at 7PM and then by 5AM it will be 55F.

    My house design will be geared towards Passive Solar/Passive House with exposed concrete slabs on the south exposures to benefit from the thermal mass. It is a Zone 4 climate. The summer sun is brutal but 30" roof overhangs and thermal curtains will help curtail the summer sun. I was also going to incorporate "throw rugs" on the exposed slabs on the south elevation. Any direct sunlight from the lower windows will not hit and warm the slab since the rugs will prevent the slab from warming and keep it cool during summer months. Then when fall - spring comes, one can simply remove the throw rug and let the sun do its thing and warm-up the slab.

    I also don't mind the vast indoor temperature swings. Warming up to 80F during the day and then down to 70F by morning is not a big deal to me. I would rather wear shorts and then put on a sweater than pay high utility bills to the energy monopoly known as APS.

    The Native Americans understood thermal mass and the adobe dwellings lets them survive the brutal summers and cold winter nights. I think ICF gets a bad "rap" but some of that is deserved because they made some outrageous claims in the beginning. ICF offers a lot of advantages that wood structures don't and vice versa. Out here builders don't know how to seal a wood frame home. They build them "leaky" so they can breathe. At least with ICF the walls are air tight and as long as you pay attention to the windows, doors, and roof transition areas, an ICF home can hit PH Standards of <0.60 ACH more easily than a wood frame home. A lot more to detail in a wood framed home than stacking and pouring a 6" monolithic concrete ICF wall assembly. Builders would think you are crazy if you asked them to tape the OSB seams. Shoot, they don't even put OSB on the exterior walls, it's mostly open framing (2x4 @ 16" o.c) with no OSB sheathing out here.

    ICF/thermal mass works but is very location dependent and design dependent. If you get both right, you can have a home that performs better than a wood framed home would. Getting the wood frame home to perform as well as the ICF home in that scenario would require more money and out-of-state expertise that would break the bank.

  33. GBA Editor
    Martin Holladay | | #33

    Response to Nathaniel G
    You wrote, "...unless you're willing to go 'far out' and build a passive annual heat storage type house which is so massive that is releases stored summer heat in the winter, but most in the building trades and residential real estate market are too conservative to even consider such a thing."

    Conservatism has nothing to do with it. The issue is economics.

    The cost of thermal mass is so high than the energy savings associated with the investment have an extremely long payback. It makes more sense to invest in measures like improved insulation and airtightness improvements than it does (in most climates) to invest in extra thermal mass.

  34. iLikeDirt | | #34

    Bias against thermal mass
    The initial examples reveal a bias for "on-demand" thinking regarding thermal comfort. For example, the point of having a cistern is not so you can back up a truck and buy your water in bulk from the water merchants; it's so you can hook your gutters up to it and capture the torrents of rain that drench the land only a few times a year, after which point you drink for free for several months while everybody else is lining up at the water merchants' tent. The insulation equivalent would be installing low-flow appliances throughout your house so you don't need to buy water as often. It should be obvious the the two strategies can work together to produce an incredible efficiency of water storage and consumption. It should also be obvious that the required size of the cistern decreases the more regularly it rains. If it rains every week, a big cistern is a waste of money. If it rains every day, you don't need a cistern. But then again, you don't need low-flow taps either.

    The entire point of thermal mass is to smooth out irregular flows of heat, just like the big cistern smoothes out irregular amounts of free rain. The more irregular the flows, the bigger the advantage, but the more mass you need to smooth them out. Anywhere and anytime mechanical cooling is required but the time periods we're talking about here are measured in hours, thermal mass can work really well and can often eliminate the need for mechanical cooling. When it gets colder and you're not in the southwest, the time periods are measured in seasons, and that's why thermal mass doesn't work as well during that time--unless you're willing to go "far out" and build a passive annual heat storage type house which is so massive that is releases stored summer heat in the winter, but most in the building trades and residential real estate market are too conservative to even consider such a thing. But even with relatively conventional construction, any house can benefit from thermal mass wherever cooling is typically required to bear the heat.

  35. iLikeDirt | | #35

    building conservatism
    I think conservatism does have something to do with it. Thermal mass is only outrageously expensive because we short-sightedly insist on getting it from concrete, whose price rises constantly and requires skilled, experienced labor to work with to avoid the potential for disaster. But alternatives exist that we don't even consider; a compressed earth block, for example, is an inexpensive, durable, building-code-approved structural mass material that we ignore because of our tradition of wood-framed buildings and perceptions that earth is used only to build "poor people's" houses. But there is absolutely no reason why CEBs could not be used to build a structural wall that is them covered in rigid insulation on the outside and drywall on the inside. You just screw those materials right into it and finish using conventional methods. CEBs with rebar and grouted cores through the walls just like CMUs become seismically-resistant reinforced masonry. Nobody would know that there was a "poor people's" wall inside the granite-bedecked kitchen.

    Unskilled, barely-trained, inexperienced laborers can run the machines with an extremely high degree of success, and in many parts of the country, suitable soil can be taken from the dirt excavated to dig the foundation. Depending on the level of mechanization of the process, and the potential speeds of construction enabled by interlocking, mortarless lego block style blocks, I suspect the cost could be reasonably close to the price of a wood frame, and certainly cheaper than an equivalent mass of concrete.

    Is it economically worth it? Let's say I'm wrong and CEBs increase the wall expense by 50%. That probably raises the total building cost by maybe 8 or 10%. And if this highly-insulated CEB wall gives you improvements in thermal comfort and ongoing utility costs that are always positive, and strongly positive in many climates, and it improves the house's soundproofing, and it makes the air sealing easier and less likely to be full of holes, then that seems like a pretty good hypothetical deal to me.

    Not that I am in love with CEBs or anything. I just think they're an example of a promising technology that the building industry has yet to set its sights on improving its efficiency and reducing its cost, due to prejudices and the weight of tradition.

  36. blueyebear | | #36

    Radiant heat thermal mass combo
    I grew up in a home with a concrete slab foundation with hot water radiant heat pipes (fired by an oil boiler). There were advantages and disadvantages to this system...however, it seems to me that a concrete (or other thermal mass material) wall, or floor, exposed to the interior and insulated properly on the outside, warmed by solar hot water during the day would make for a very effective daytime storage and nighttime release system for heating. Not passive solar, but perhaps more effective. Does anyone have any thoughts or experience with this? Or, considering the cost of building such a system, is a straightforward forced air geothermal system preferable?

  37. GBA Editor
    Martin Holladay | | #37

    Response to John Christman
    You wrote, "A concrete (or other thermal mass material) wall, or floor, exposed to the interior and insulated properly on the outside, warmed by solar hot water during the day would make for a very effective daytime storage and nighttime release system for heating."

    The main problem with your suggestion is that a superinsulated home with good windows doesn't need any space heat on sunny days. Enough solar gain should occur to keep the house toasty.

    You suggest installing lots of solar thermal collectors on the roof and running hot water through a concrete slab. There are two problems with this suggestion: (1) Solar collectors, tubing, and pumps cost more than windows, and (2) The hot water would cause the house to overheat.

    The second problem can be solved by storing the hot water in a large insulated tank and using it later. The main problem with this approach is that it costs even more money than your suggestion. Moreover, the heat can only be stored for a few days using this method, and the economic value of the collected heat is too low to justify the cost of the expensive equipment required to collect it.

  38. Mountainguy | | #38

    Thermal Mass in the best possible area
    We live in a great place to utilize thermal mass, the mountains over 5000" in southern California. Our last house was a "mid century modern" ranch design built in 1960. It had plaster walls and a lot of thick paving tiles on the floor. It was also slab construction. This was well built house with a good number of south facing (single pane) windows. The south side of the house had a well engineered overhang that eliminated direct summer sun. Simply opening the windows at night allowed for NO air conditioning in the closed up house during the day (90+ degrees outside). Nightime summer temps were never above 70º and mostly in the 50s and 60s (sometimes in the 40s) In especially hot weather our DRY climate allowed for us to dump water on the tiles right before we went to sleep. Brought the internal humidity up (from 0 to 10 percent) and the tiles were COLD in the morning.

    In the winter the sun came in the windows under the overhang and warmed the house up. Our heat use was low because of this. Anyhow, I would encourage people with older houses to think about their situation. In our case the one size fits all mantra , "insulate, insulate, insulate and seal the house for zero airflow to the outside" which does a very good job everywhere was not the best. Our airflow method allowed us to have a more comfortable and healthier house than those with super insulated sealed up houses . It would have been foolish and wasteful indeed to tear down the interior plaster walls and insulate better, tear out the windows and replace them with double pane and install an efficient HVAC system. We did upgrade the heater.

  39. Jeffrey_Savage | | #39

    Phoenix sounds like a terrible place for ICFs
    Im planning a build in PHX and was considering ICFs. Our main concern is cooling in the summer. Its common to have 110 highs and 90 lows. Since the lows never come close to below interior temps its sounds like thermal mass will have no benefit and I would be left with just the poor insulation of the rigid foam. Maybe Im missing something here as people say ICFs are good for hot climates.

    Im trying to decide between a 2x6 with open cell foam and 2 inch polyiso on the exterior vs a double wall of 2x4's (blown cellulose or foam but it's getting prohibitively expensive at that thickness), or ICFs.

  40. GBA Editor
    Martin Holladay | | #40

    Response to Jeffrey Savage
    You are correct that in the weather conditions you describe -- 110°F highs and 90°F lows -- there is no advantage whatsoever to having thermal mass. Under those weather conditions, what you want is (a) adequate R-value, (b) a thermal envelope that is as close to airtight as possible, and (c) windows with a low SHGC, with most of the windows protected by shade.

    Two of the wall systems you describe -- 2x6 walls with exterior polyiso, and a double-stud wall -- are commonly used by builders in search of a high-R wall. However, these wall systems are expensive, and may not make sense in a hot climate.

    Make sure that you focus on methods that really lower your energy bills. Often, a hot climate design needs things like adequate roof overhangs, smaller windows, or windows with the right glazing -- not a high-R wall. If you get a high-R wall, but mess up the roof overhangs or window specs, you're barking up the wrong tree.

  41. Natchorx | | #41

    location home
    Hi, I've been involved in the design of an ecologic home for seasonal locations (no full time people living in it) in Quebec which is zone 8 I believe for people who want to visits the country but don't want to rent an hotel. The building I thought would be the most appropriate would be a passive solar house with south facing windows. I also thought that thermal mass would be a great plus for the building as it would reduce the peaks of heat and stabilize temperature during the night. My question is : Do you think it could be possible to make the house 100% passive or semi-passive meaning nobody would be there and everything could simply be controlled with sensers like hiding windows with curtains during night and maybe using geothermal heating and solar liquid heating system?

  42. GBA Editor
    Martin Holladay | | #42

    Response to Nathan Chagnon
    In your climate, it isn't possible to build a house that doesn't require a heating system. In most cases, local building codes require you to install a heating system, for one thing; but even if these codes didn't exist, you still would be unable to get through a Quebec winter without a heating system. There are too many stretches of cloudy weather to coast through a winter with just solar heat.

    That said, it is possible (but difficult) to build a house that will never freeze, even during long electricity outages like the ice storm that hit Quebec in January 1998. To build such a house, you would need to pay close attention to airtightness, aiming for the Passivhaus target of 0.6 ach50. You would also want to install triple-glazed windows and above-code levels of insulation.

    In your climate, there really isn't any benefit to thermal mass, as this article ("All About Thermal Mass") explains.

    The passive solar approach also doesn't work very well in your climate. Here is a link to an article with more information: Reassessing Passive Solar Design Principles.

    Geothermal heating systems (more accurately called ground-source heat pump systems) are almost always too expensive to make sense for a single-family house. There are better heating systems out there. For more information, see Are Affordable Ground-Source Heat Pumps On the Horizon?

    Bonne chance, Nathan.

  43. MeInKiev | | #43

    This is an excellent article, and I have read it a couple of times with great benefit.
    I am about to build a house in Albury, NSW, Australia. We are in a Zone 4 region, which I assume is the same zoning system for the States? The January average minimum temp is 59.2 F, and Av. Max. 87 F (but for two weeks in a row this January we had daytime maximums between 104 - 118 F).

    Historically, Rammed Earth houses were common here, and I am still deciding whether to use Earthbag walls (similar but thicker) or Double brick with the cavity insulated (with R 2.5 foam boards). If using Earthbag I would insulate the exterior of the 380 mm (16" after plastering) Earthbags with around 50 mm (2") of Closed Cell Spray Foam (XPS). I understand Earthbag walls are only around R-1.5, and the XPS adds around R-6.5 per inch.

    I don't think I will need any A/C during summer with either of these wall types, but I am sure I will need to heat at least the living areas during winter, as I have experienced a Rammed Earth house in Portugal that was VERY cold in the bedrooms.

    I know for sure earth walls will dissipate heat very quickly in winter, so there will be a central heating wood stove in the kitchen supplying an in-slab hydronic system in the lounge, study and kitchen-dining rooms (but not in the bedrooms, which we traditionally don't need to heat). Two of the three bedrooms will have some northerly sun heating them, and partially from another ceiling heat re-circulation system.

    What I am wondering is if you think external insulation on such massive Earthbag walls will hinder natural cooling during summer months? I am pretty confident an earth wall without insulation will perform well during summer as some of the internal heat will escape outwards and ventilation will take some internal heat out during the night.

    I suppose with external insulation the walls should not allow much external heat to enter, so there may be less heat to dissipate with night ventilation? What is your opinion?

    During the winter, with external insulation, most of the heat applied will be retained in the walls to even out the day night differences. I suppose the main problem would come if the house needs to be heated from a standing start after a few days away.

    My architect has run many variations through his NatHERS energy efficiency software (the standard in Australia), and we are confident about the double brick figures, but the software does not do well with high thermal mass walls like Rammed Earth or Earthbag, as it cannot account for time lags and comfort factors.

    I suppose what I am also asking is if there may be too much thermal mass with Earthbag walls? That is, would I get enough benefit from the internal layer of a double brick wall; and whether the 16" insulated Earthbag walls will be too much thermal mass, or if they will still offer a benefit?

    Anyway, I will be happy for any thoughts you may have, even simply by having a (knowledgeable) sounding board?

    1. Expert Member
      MALCOLM TAYLOR | | #44


      Martin is away for the month, so probably w0n't reply for a couple of weeks.

  44. Expert Member
    DCcontrarian | | #45

    One of the issues I have with "thermal mass" is that discussions of it are never quantitative. No one ever measures it, or calculates it, or estimates how much would be added by various assemblies. One of the ways I like to tweak proponents of "thermal mass" is to ask them what the unit of measure is, I've never gotten a response to that question.

    There's also the related issue that different people have a different notions of what property of matter they're talking about when they use that term. Most commonly, they mean the property that is properly called "heat capacity." That raises a third issue, which is that the entire discussion begins with a presumption that conventional construction results in houses that are somehow deficient in heat capacity -- an unsupported presumption.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |