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
Musings of an Energy Nerd

Hot-Climate Design

Appropriate details help keep a house cool

A Cracker-style roof. Researchers from the Florida Solar Energy Center built this PV-equipped model home in New Smyrna Beach, Florida. The home’s white metal roofing and wide roof overhangs help keep the interior cool on hot summer days.
Image Credit: Florida Solar Energy Center

People who live in Florida or Texas often accuse energy-efficiency experts of having a cold-climate bias. They’re right: most energy-saving tips are written with cold-climate buildings in mind — perhaps understandably, since Americans spend about twice as much for residential heating as they do for cooling.

Whatever the origins of this pervasive cold-climate bias, it’s time to rectify the situation with a few hot-climate design tips.

We’re not in Kansas anymore

Most builders know that house designs need to be climate-specific. In areas of the country where air conditioning bills are higher than heating bills — as they are along the Gulf Coast and in much of Florida — homes should be designed to reject exterior heat.

So what are the most important factors governing hot-climate design?

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. homedesign | | #1

    the walls
    I think that delta-T may be a good tool for sizing equipment...not for determining insulation levels.
    Heating Degree Days,Cooling Degree Days and Energy Balance seem more relevant to me.

    The Best-Practice measures that you cite should be the first steps to designing high performance homes in hot climates.

    I think that once the basics have been taken care of .... then "the wall" and the windows become the weakest links.

    I don't think we can build much better than HERS 50 homes in Hot climates if we continue to ignore or diminish "the wall"

  2. homedesign | | #2

    cathedralized attics
    Concerning cathedralized attics... do not forget the roof rafter....
    Spraying foam to the underside of the roof sheathing does not address the thermal bridge at the rafters.

  3. homedesign | | #3

    Advice from Researchers
    Which research report concludes that R-10 Wall insulation "is probably plenty"?

  4. GBA Editor
    Martin Holladay | | #4

    FSEC document
    See "Zero-Energy Homes: Lakeland, Florida."

  5. homedesign | | #5

    Not a good Project
    You and I both know that your example is not an example of good enclosure design.
    They have no concept of the importance of a good air barrier...
    How did they come up with R-10?
    Perhaps in their leaky house...going beyond R-10 would be useless.

    I can not believe that PHIUS felt that this was a good example of hot climate design.

  6. GBA Editor
    Martin Holladay | | #6

    Not a good project?
    1. By "PHIUS," I assume you mean the Passive House Institute U.S. I'm not sure where you got the impression that anyone claimed that "PHIUS felt this was a good example of hot climate design." I never made that claim, and I don't think that the researchers from the Florida Solar Energy Center made that claim.

    2. On what basis did you conclude that the FSEC researchers who built the Lakeland House "have no concept of the importance of a good air barrier"?

    3. You will never hear me criticize a builder who wants to install thicker wall insulation. If you prefer to build an R-30 or R-40 wall, that's great. Clearly, there will only be half the heat gain through an R-20 wall as there is through an R-10 wall.

    What I have proposed is a package of measures that is appropriate for air-conditioning climates with very little or no heating season. Any builder who follows the guidelines I propose will be building a house that is vastly superior to most houses built in Florida. As you point out, though, it's always possible to imagine a still better house.

    Here's the thing: I'd hate to see a Florida builder install 4 or 6 inches of wall foam to make a high-R wall, but run out of money when it's time to buy windows. If your budget is rich enough to afford the entire package of measures I propose, and you still want to go further, then by all means beef up the wall insulation. But I still think my proposed measures are head-and-shoulders above most Florida homes.

  7. homedesign | | #7

    Lakeland ZEH
    1. I am fairly sure that this is the project that Katrin Klingenberg (PHIUS) used in her Book "Homes for a Changing Climate" as an example for a hot climate.
    2.Martin's Comments from Jun 4 (passivhaus for beginners)
    "Here's what jumped out at me about the Lakeland near-zero-energy home: its air leakage is 4.9 AC/H at 50 Pascals. The researchers noted that "much of the leakage to the outside appears to be from the 30 recessed lighting cans in the ceiling."
    Back to the drawing board, I guess."
    3. No one said you have to use expensive spray foam to achieve a good air barrier or good R-value.
    You should recognize that both are important in Hot climates.

  8. GBA Editor
    Martin Holladay | | #8

    Good catch, John!
    You made me laugh. Caught me sleeping, for sure. Here I was, expecting you to quote some source, and you end up quoting my own words. You have a mind like a steel trap.

    You're right -- the Lakeland house evidently didn't have a great air barrier. But that hardly undermines my suggested package of measures, since I made it quite clear that I advocate a tight air barrier.

    So, to be clear: Build better than FSEC! No recessed cans! Keep your ceilings tight!

    You're right, you don't have to use expensive spray foam — and I never said you did. If someone wants to cathedralize an attic, though, it's the most typical way it's done. It's also possible to cathedralize an attic by installing rigid foam on top of the roof sheathing. Another perfectly good option is to build a house with high ceilings, and box in the ducts within the thermal envelope — and to leave the attic unconditioned.

  9. Michael Blasnik | | #9

    good list
    I agree with pretty much your whole list, except I'm wondering why you need radiant barrier roof sheathing and a highly reflective roof if there are no ducts in the attic and you have an R-30 ceiling? I can't imagine any significant cooling savings, although I can see that the reflective/cool roof may provide other benefits. Also, you don't mention HVAC equipment -- SEER 15 too obvious? SEER 17 too questionable, ductless mini-splits too expensive?

  10. GBA Editor
    Martin Holladay | | #10

    Radiant barrier roof sheathing
    You caught one inconsistency in my list, but there's another. Here it is: if you manage to design a house that has all of its windows shaded — say, a house with very wide porches on the north and south sides, and no east or west windows — then you don't really need to worry about window SHGC. Yet I recommend a very low SHGC for windows.

    I thought about clarifying the inconsistency, but here's the thing: almost nobody takes a list like this and follows it to a T. They end up picking and choosing, or compromising. That's okay. After all, it's pretty tough to keep EVERY window in the shade, all day long, in all seasons. So it makes sense to consider SHGC.

    Now, getting back to the question of radiant barriers: you're right. For years, I've been writing about the uselessness of radiant barriers. The one exception always raised to my anti-radiant-barrier rants: what about radiant-barrier roof sheathing for new construction in hot climates? After all, the incremental cost compared to regular roof sheathing is relatively small.

    So I have given in on this point. I'm willing to say, go for it. Get the RB roof sheathing, and remember to install it shiny-side down. It doesn't cost much and it can't hurt. And if you fail to follow my list, and stupidly put a few ducts in your attic — which you shouldn't — or some stupid subcontractor disturbs your attic insulation, leaving a thin spot — which he shouldn't — then your attic will be a little cooler than it would have been if you had conventional roof sheathing.

  11. GBA Editor
    Martin Holladay | | #11

    High-SEER air conditioners
    Concerning high-SEER air conditioners: again, you're right. It should go without saying that efficient HVAC equipment is a wise investment; if you are in a cooling climate, get a high-SEER air conditioner.

    My column concerns design. I want to be sure that builders get the basic shell and orientation done right. Air conditioners come and go, after all -- they're not a permanent part of the building -- but the shell endures. Get the shell right, and you're most of the way there.

  12. homedesign | | #12

    Rethinking Cathedralized Attics
    I am a former fan of cathedralized with one myself.
    For sure mechanical and ducts must be inside conditioned matter what.

    Problem with CA(Cathedrizied Attic) = too much surface area and it almost always limits us to expensive spray foam.
    I am looking to avoid the CA ...create a damn good air barrier at the attic floor and then use AMPLE levels of less expensive attic floor insulation.
    Forget about the radiant barrier roof deck and use those dollars for more Wall R-value...perhaps I might get radical and go for R-11 walls ;-)

  13. GBA Editor
    Martin Holladay | | #13

    I agree
    Certainly here in cold-climate Vermont, I've concluded that the best attic is an unconditioned attic. Detail the perimeter to allow room for deep cellulose (raised-heel trusses or rafters on stacked plates on top of ceiling joists) and blow the cellulose deep. Air barrier at the ceiling plane.

    Like you, a few things about cathedralized attics make me nervous: thermal bridging through rafters must be addressed, and roof leaks above spray foam may cause extensive sheathing rot before they're noticed.

    Nevertheless, I've been astonished to learn how reluctant southern builders are to abandon attic ductwork. Sometimes it's easier not to fight the status quo -- just advise southern builders to cathedralize their attics, because it seems they just won't stop putting their ducts up there.

    1. Jbaemom | | #51

      We put it in the attics because we don’t have crawl spaces, for the most part. In Texas it’s slab on grade. So it’s either low ceilings or use our attics.

      1. MartinHolladay | | #52

        J. Bae Mom,
        Here is a link to an article that lists 8 different solutions to the attic duct problem: "Solutions to the Attic Duct Problem."

  14. user-716970 | | #14

    Wind driven snow
    Here on the Canadian prairies, we usually get quite a bit of cold windy weather that gets snow drifting and blowing, sometimes causing white-out conditions. I think most homeowners here would be shocked to find out how much snow can pile up in their vented attic spaces during these conditions. This rarely causes any damage as the snow seems to evaporate or sublimate (like an ice cube in your freezer). Kind of risky though...further north, in the arctic territories, it does indeed cause damage.

  15. GBA Editor
    Martin Holladay | | #15

    More information on optimal wall insulation
    I decided to look into John Brooks’ question of whether thicker wall foam makes sense in Florida. It turns out that doubling the thickness of the wall foam from R-10 to R-20 would save only $20 per year.

    I discussed the issue with Danny Parker, a senior researcher at the Florida Solar Energy Center. Using EnergyGauge software, Parker calculated the savings for a 1,600-square foot single-story slab-on-grade house in Tampa, Florida.

    He assumed that the house had fairly good specs, roughly along the lines of the house I described above. He assumed that ducts were inside the thermal envelope, the house had 3 ACH @ 50 Pascals, the windows were U-0.30 with a SHGC of 0.30, and the thermostat is set at 77°F in summer and 68°F in winter. He assumed that electricity costs $0.15 per kWh — higher than the current retail rate there of $0.12 per kWh. (Assuming a higher electricity rate tends to exaggerate the value of increasing insulation thickness.)

    If the house has 2 inches of rigid foam (R-10) on the walls, the heating and cooling costs [Later note: should be "total household energy use" — see discussion below] will be $1,794 per year. If the wall insulation were doubled to R-20, these costs go down only $20, to $1,774 per year.

    Is it worth it to double the thickness of the foam to save $20? Well, it depends. You could also save the same $20 by buying a photovoltaic (PV) array costing $857; that 114-watt PV array would produce about 166 kWh per year, valued at about $20. So, if you can double the foam thickness on a 1,600-square-foot house for $857 or less, the foam insulation will cost less than PV. If doubling the foam thickness costs more than $857, then PV is cheaper.

  16. 2ohMKKQPcu | | #16

    Designing efficiency in various climates
    Designing Energy efficiency

    There are some great points made in this article and create food for thought when designing a home. However….
    The single biggest factor responsible for superior energy efficiency in a home is its shape. Shape makes all the difference in the world. Of all possible geometric forms the pyramid has the least amount of exposed exterior surface area per square foot of heated or conditioned floor space. Heat loss (or gain) is a direct mathematical function of surface area.

    A common term used in construction today is "R" value when talking about insulation in the roof or walls. This is a very misleading term and I suggest with good evidence that it not be used. It is supposed to indicate the relative energy efficiency of a material in stopping the flow of heat through a wall or roof. The coefficient of heat loss (which is a small decimal point number describing the heat flow in BTU’s per hour per inch of thickness) is the "U" factor and the reciprocal of that decimal number (or 1 divided by the U factor) is called the "R" factor (which is a whole number). The larger the "R" value the better the home’s efficiency, right? Not exactly! Insulation’s "R" value is only one small part of the formula used in calculating a building’s total heat loss. A claim of any percentage amount of comparative energy efficiency is misleading as well. A humorous but true old adage says that "figures can lie and liars can figure" should suffice to remind us to examine the whole picture when it comes to claims of energy efficiency. A truer statement proving energy efficiency is the actual energy bill history paid by the homeowner.

    Those are the numbers that really count. An analogy in the automobile industry would be to say that "this vehicle gets 15% better gas mileage than a comparable sized auto".... we all can measure "miles per gallon" and judge for ourselves the relative efficiency of the vehicle.

    A building’s exterior surface area measured in square feet "multiplied" times the material’s "U" factor and the design temperature difference between the interior and exterior surfaces (Delta T) will yield an overall heat loss in BTU’s per hour. A perfect vacuum in the exterior shell would lose zero BTU’s. A building that leaks air like a sieve through all the joints would lose all it’s heat energy in a short time. The point is that the shape of the building and the air tightness are much more important to energy efficiency than simply the "R" rating of the insulation.

    We often forget about the principles of earth contact homes and how efficient they are if we incorporate the lower floor to be in contact with the cool earth 55 degrees. My first solar home was an earth contact with collectors, storage tanks and heat pumps and a solarium.

    Designing the window locations and amount of them is another important factor,and it is paramount that each design be done to fit the clients needs, location and climate criteria, stamping out cookie cutter designs and placing them at random without consideration of solar orientation will not achieve the goal of energy efficiency we wish to attain. My next pyramid home project just happens to be in Kansas, so to make Dorthy happy in her mixed climate, We have to design accordingly.

    Designing with SIP panels, ICF foundations using cool roofs and the technologies and building material advances in todays market that are available to us, is a major step toward efficient design, we have to sacrifice tradition of style and construction if we are to make a major impact in design efficiency. Or we can continue like we have for the last 300 years of post and beam construction without regard for energy usage and our impact on our invironment.

    Less is more as Mies Van Der Rohe put it, and Neccessity is the mother of all invention. Bucky Fuller had the right idea decades ago when he changed the shape of housing design. He was long ahead of the times and we are slow at catching on.

  17. GBA Editor
    Martin Holladay | | #17

    Dennis, you're preaching to the converted
    I think it's fair to say that any readers who have made it this far into our discussion of these issues are well aware that wall, floor, and roof R-values are not the only factors affecting heat loss and heat gain.

    Of course air leakage matters. Of course window size, window placement, and glazing specs matter. Of course house shape matters. Compact house designs are always easier to heat or cool than complicated designs with numerous ells, cantilevers, dormers, and doo-dads.

    That said, I predict that few Americans want to live in a pyramid -- for the same reasons that few Americans want to live in a 1970s dome. It's hard to put furniture up against a sloping (or curved) wall.

    Good luck, though, with your pyramid scheme.

  18. homedesign | | #18

    Fairly Good Specs?
    I am startled by the projected heating and cooling costs for the house designed with your specs.
    Those numbers sound very high to me... more along the lines of a house barely built to current energy codes...
    It sounds to me as though there is plenty of opportunity for improvements to the enclosure.
    Where IS the major "weakness" in the house that Danny Parker modeled?
    Does Danny have any insight concerning the weakness?
    Is it the windows?
    Any chance you can post the specs and output from the modeled house or email it to me?
    Tampa climate does not sound THAT severe to me.

  19. GBA Editor
    Martin Holladay | | #19

    Send me an e-mail
    Send me an e-mail ([email protected]) and I'll try to get you in touch with Danny Parker.

    Of course every energy simulation depends on the specs used; I'm sure you can design a better building than Danny used in his example. Here's the point, though: if you assume a better envelope or better equipment, you'll end up with lower energy bills -- meaning that R-20 wall insulation will save even LESS energy than the example provided. That undermines your main point, it seems to me -- that thicker wall insulation is justified in Florida.

  20. Michael Blasnik | | #20

    Tampa house run -- typo?

    John brooks has a point. I think you might want to double check with Danny Parker about the energy usage of that house in Tampa. The $1794 might be reasonable for the total electric usage of the home, but is out of line for just the heating and cooling loads for a house like the one described. At 15 cents/kwh, $1794 is nearly 12,000 kWh/yr. There should be virtually no heating load and so the cooling load would need to be about that large yet a home like you describe (inside ducts, good windows, tight shell, presumably new equipment), should use maybe 5,000 kWh for cooling and probably less.

    The cooling climate in Tampa is not much worse than Houston, where I have data on tens of thousands of new homes electric use and the cooling loads are typically around 5,000 kWh for larger homes with worse specs than you describe.

    My own quick and dirty modeling of a Tampa home like you describe estimates a cooling load of about 3600 kWh with R-10 walls and 3400 kWh with R-20 walls. The overall conclusion is similar to yours -- the wall insulation only saves 200 kWh ($30) per year -- but the usage levels are way different..

  21. homedesign | | #21

    ONLY 200 kWh
    Only is a matter of perception...
    200 kWh is an 18% improvement.
    Michael Blasnik's numbers are more what I was expecting...
    If you take Michael's R-20 example and upgraded the might be able to ditch the heater altogether in Tampa.

  22. Michael Blasnik | | #22

    double check math
    200 / 3600 = 5.6%, not 18%. The windows already are very good (check the specs U 0.30, SHGC 0.30) and the heating load is essentially zero.

  23. GBA Editor
    Martin Holladay | | #23

    Thanks, sharp-eyed readers
    John and Michael,
    I'm happy to have your sharp eyes to double check the math.
    Right now I'm double-checking with Danny Parker to be sure I don't compound any errors by jumping to premature conclusions.

    However, it looks like we're achieving consensus on at least one point -- doubling the wall insulation from R-10 to R-20 only yields annual savings in the range of $20 to $30.
    I will post more later.

  24. homedesign | | #24

    yep and DOH
    Yep.... and DOH too

  25. homedesign | | #25

    wishful thinking
    I must have been practicing wishful math

  26. GBA Editor
    Martin Holladay | | #26

    Should have written "total household energy use"
    John and Michael,
    Thanks again for catching my mistake. Danny Parker had his numbers right; the fault was mine.
    His annual energy bills were the total household energy use, not the heating and cooling costs, so I misreported his simulation results. My apologies.

    However, the bottom line (the difference between the two simulations, one with R-10 and the other with R-20 walls) remains the same. Using the specs provided in my earlier post, the difference is only $20.

  27. Danny Parker | | #27

    More details for discussion and clarification
    Michael and others:

    Sorry for the confusion, the $1974 annually @ $0.15/kWh in the previous analysis referred to all household end uses and thus, not just cooling and heating.

    The building simulation I ran today (did not save the one yesterday), showed a space cooling energy use of about 4400 kWh/year-- not far from your numbers, although this simulation today is for a 2,000 sqft home.

    This is a newer house with high efficiency levels that I happened to have up. I also ran the R10 vs R20 walls question on that house. I attached the details and results for the house ran today. Again, total energy use is much different from heating and cooling! Lots of other end uses.

    We do know, from 171 monitored homes in a statistical sample monitored for Progress Energy Florida (PEF), that the average sub-metered cooling energy use in Florida single family detached existing homes in 1998 was 6,400 kWh. Measured total electricity consumption average 17,130 kWh. We still have the 15-minute data on the metered end-uses as well as the audit characteristics for all the homes in the study. See the section on cooling below:

    Also, we have similar sub-metered data from the other large Florida utility, FPL, should you be interested in those numbers. Averaging the total 2 million FPL residential single family accounts shows an average total annual consumption of 17,688 kWh/yr in the most recent year (We have a plot showing the five-year trend in Figure 4 of this report):

    The corresponding sub-metered data on cooling from FPL (a much smaller, but sub-metered statistical subset) showed an average in their Central Florida sample of 6,507 kWh/yr for air conditioning. Thus, they are virtually identical to the PEF numbers.

    I have performed still another simulation for an existing Florida home, which while I can't show here, indicated cooling at about 6,900 kWh for existing homes with about 16,800 kWh used overall each year. Here is the breakdown for those so inclined:

    Cooling: 6925 kWh
    Heating: 677 kWh
    Hot water: 2652 kWh
    Ceiling Fans: 580 kWh
    Clothes washer: 79 kWh
    Clothes Dryer: 970 kWh
    Dishwasher: 145 kWh
    Lighting: 1736 kWh
    Range/Oven: 447 kWh
    Refrigerator: 600 kWh
    Miscellaneous Electric loads: 2000 kWh

    Total: 16,802 kWh

    (Actually I need to revise this building as it has a new vs. existing refrigerator; the old ones in FL typically use about 1000 kWh each).

    If we compare that with measured average end-use consumption, you can see they are pretty close. Thus, the simulation and the measured data are not too far from each other. And since in the same monitoring study, we monitored each of the end uses (Figure 1 in the report), we know that the simulation results are close to what we measured.

    We also agree that the added wall insulation in a new home in Central Florida in a new building will only save about $30, but do note that this does vary somewhat with thermostat assumptions, wall solar absorptance and location around the state. Also, as John Brooks point out, it doesn't mean that this is not worth going after-- particularly if PV is being considered.

    On the other hand, measures, not typically considered, such as purchasing an 42" energy efficient LCD TV that draws 160 Watts vs. a plasma one that draws 300 Watts for the same picture may be an even more pressing consideration-- particularly since this average TV is on for six hours in American homes which now have more TVs per household (2.5) than people (2.4). Go figure.

    And relative to the envelope, the simulation suggests that in our climate that a tile floor vs. a carpeted one is more important to reducing cooling than is the R-10 to R-20 wall insulation push. Recall that in Passivhaus (designed for cold climates), the slab is uniformly insulated underneath. In hot climates, a powerful earth contact is a good thing. Insulating the slab (or substituting a crawlspace) are bad in that they decrease free earth contact cooling. And in Florida, carpet= dust mites and a difficult to dry-out home after a hurricane related flood. Thus, the blanket Passivhaus recommendations may not be fully optimal across climates.

    The new 2,000 sqft home above using 4,400 kWh for cooling per year can be made even better-- possible to reduce that to less than 2,000 kWh per year using a variety of technologies, but more efficient system than the standard SEER 13 AC simulated, being a key one. And again, these miscellaneous electric loads, water heating, clothes dryers etc. remain very, very important.

    Sorry, no silver bullet. Each end use has to be addressed.

    In one sense, the confusion we had over the cooling numbers really highlights a key reality: heating and cooling are just part of the overall energy picture in homes. In Florida, these heavily focused-upon HVAC end-uses comprise only about 40% of the pie. One ignores the other 60% at their peril.

    See Figure 1 in the monitoring study cited above.

    In a real sense, that means that if you are really intent on getting to zero energy, you MUST look very carefully at all the end uses and not just heating and cooling. Also, as heating and cooling shrink down with better buildings, this residual become steadily, more and more important.

    While I think the Passivhaus PHPP process is admirable to reduce heating and cooling, I do think that many focus obsessively on the building and miss the "60%-Other" reality more often than not.

    And we haven't even discussed the biggest issue of all-- how consumption varies with behavior and what to do about that. I'll stop there. Its a another huge topic, but a vital one-- particularly when you consider that evaluations of how energy use in buildings show (from evaluation of movers vs. stayers) that about 60% of energy use is due to the intrinsic equipment and building; the other 40% is due to occupancy behavior patterns. I have the feeling now, that modern home electronics are pushing the later figure up even further these days.

    In any case, I cheer everyone's effort and apologize if this addition rambles.

  28. homedesign | | #28

    Slab Edge and Slab perimeter Insulation
    thanks for your comments...please Ramble anytime
    What are your thoughts about slab edge and slab perimeter insulation?
    Is this a significant "thermal bridge" in Florida?
    No-one in North Texas has attempted slab edge or perimeter insulation as far as I know....(termite concerns)

  29. Danny Parker | | #29

    Slab edge gains
    Give me some time....very busy this afternoon. But yes, we have data on this from experimental facilities we have here at FSEC showing edge heat gains on the interior while cooling. I also have IR images, but I don't think any way of posting them here...

  30. Danny Parker | | #30

    Slab floor temperatures in NightCool test buildings

    You can see the interior tile floor surface temperature in the Nightcool online data. I would place the images here, but don't know how:

    Go here:

    There is data for two separate test buildings; one a control and the other with the Nightcool technology. Click on the INTERIOR TEMPERATURES to see that for both buildings.

    And then click on the FLOOR SURFACE TEMPERATURES to see how the interior tile floor center temperature relates to the temperature at the edge (about 4" from the edge).

    Right now, we are using a daytime set up in the buildings to see how that influences Nightcool. If you want to see a level playing field (78 F all day) select July 15th before that was altered.

    Anyway, in July you'll see that the slab edge temperatures are about 2-2.5 degrees warmer on the inside than the center of the slab. You can also see that with the thermostat set up during the day, that we are getting a lot of thermal storage in the slab itself. IR images makes this really clear.

    Yes, edge insulation seem like a bit of a challenge these days. One thing we have seen is that planting Lirope muscari or similar landscape plants that shade the ground just outside the slab may provide much of the benefit of insulating-- at least for cooling.

    We know that is likely since we saw an anomaly (cool spot in the edge) in the IR image on the inside of the one of these buildings and went out to see what could be causing it. Turned out to be a large weed on the south that was shading the ground there...

    Moral of that story from Yogi Berra who would love IR cameras: "You can learn a lot, just by watching"

  31. homedesign | | #31

    Slab experiment
    I was looking at the nightcool project..July 14 and July 15
    Very interesting that the slab edge of the experiment house is much cooler than the slab edge of the control house.
    I am assuming that both houses are the same from the ceiling down....
    Does the R-30 SIP cover the top of the wall plate on the experiment house?
    I assume that the control house does not have R-30 above the top plate.

  32. Danny Parker | | #32

    Slab edge surface temperatures

    Yes, the buildings are identical from the ceiling down. The R-30 SIP does cover the top plate, but the walls fare frame construction: R13 batt + R6 exterior sheathing covered by hardiboard.

    I saw the difference in the experiment and control, but should take a look at the floor thermocouples (which are surface mounted). I'll go out and check if the are adhered similarly. They are in the same location.

    Should be something nice for my daily walk...

  33. Rick | | #33

    vapor barrier
    John mentioned a good vapor barrier at the attic floor. I'm at ground zero for the conditions you're talking about, Fort Myers in humid southwest Florida. All, and there may be a few exceptions but very few, homes here are built with trusses. How would a vapor barrier be installed over the insulation and around the truss webs?

  34. GBA Editor
    Martin Holladay | | #34

    An AIR barrier, not a VAPOR barrier
    Be careful — you definitely don't want a ceiling vapor barrier in Florida. John Brooks never recommended one. He recommended an air barrier.

    A ceiling air barrier is definitely necessary. This can be provided by ceiling drywall, as long as there are no unsealed penetrations, and as long as all electrical boxes are special airtight boxes. Of course, the attic access hatch must be carefully weatherstripped, and caulk or gaskets must be used between the top of the partition drywall panels and the partition top plates.

  35. Rick | | #35

    Air barrier

    Thanks for the clarification. Air barrier it is then. Detailing ceilings should be easy as long as all the subs are on the same page. Insulation is not a big topic of discussion here as people seem to tie insulation more to heating climates. It's very hard to get even basic batts done correctly, even the insulation companies don't care.

  36. carpeverde | | #36

    Best Bang for the Buck

    I always enjoy the lively discussion on this blog, and you touched on some critical issues we are having at Build San Antonio Green in discussions with affordable builders. They must keep housing prices down below $110K to fall within terms with City programs, and in order to earn Energy Star rating and a certificate from us must demonstrate a max. HERS Index of 85. Using urethane foam or SIP construction is out of the question for them as is likely BIB or sprayed cellulose. They often have very little ability to orient the small 900-1100 sf houses with the major axis east-west. For 2010, the City and electric utility are offering decent incentives to these builders if they can build homes with a max. HERS Index of 75. They are almost there because some have been able to reach Indexes of 80 this year. My question is whether you think they might be able to reduce the Index by 5 points if they were paying closer attention to air infiltration opportunities during framing. Are there critical locations where they might add sealants at all envelope penetrations, windows, doors, etc. to get them to HERS 75 by improving their blower door test? Are there any other low cost aspects to consider? They are finally using the white asphalt shingle roof rather than the dark color shingles, but it's the best roof they can install for their budget. Our building codes are getting better with regard to requiring better energy efficient choices starting in 30 days, and will only get more aggressive in 2011 and beyond. I'm trying hard to help increase the option for home ownership for our lower income residents whose other option is rental properties. Any ideas you might have would be appreciated. I must be overlooking something. Thanks.

  37. GBA Editor
    Martin Holladay | | #37

    Air sealing and the HERS Index
    There are really two components to your question:
    1. Do air sealing measures represent a good bang for the buck?
    2. Does the HERS Index give adequate credit for air sealing measures?

    Concerning question 1: Generally, air sealing measures represent the best bang-for-the-buck out there. Of course, with any measures, there comes a point where it gets very expensive to make a very small incremental improvement. What level of air tightness are your builders achieving now? Those who have tried to reach the Passivhaus standard of 0.6 ac/h @ 50 Pa. find it very challenging. Most new home builders in Minnesota and Vermont find it fairly easy to achieve 2.5 ac/h @ 50 Pa. If you can achieve 1.5 ac/h @ 50 Pa, you are doing better than the average new home builder.

    Concerning question 2: I certainly hope that the HERS Index provides appropriate credit for actual measured air tightness. That's the whole reason for performing a blower-door test -- so that the HERS rater has a number to enter into the software. But since I'm not a HERS rater, I'd love to hear a HERS rater address question 2.

  38. carpeverde | | #38

    What Tightness Are Builders Getting Now
    Thank you for your reply. I certainly agree with your assessment, and I had a meeting late last week with the best of the area's affordable builders just to ask what he's doing to address air leakage in new homes. He read off a fairly substantial set of taping and sealing practices that sounded pretty stout for homes destined for fiberglass batt insulation. He received ENERGY STAR certificates for all of his homes, but he never asked for the blower door results, so I've asked him to check with his RESNET rater to include that information and let us know how his previous ES homes scored. If he's getting very low ac/h now (<2.0), the "miracle" of spray foam insulation may not even earn him additional reductions in the HERS Index even if he could afford it.

    I've been hearing from my pals closer to the RESNET/DoE/ENERGY STAR work going on that the housing evaluations for 2011 are going to be very different from the way they're evaluating homes now, and that the new protocol will (hopefully) eliminate the HERS performance bias toward the larger homes. It's more difficult now for very small homes (<1100sf) to earn ENERGY STAR even with near identical construction practices. Maybe everyone here just has to tough it out for 2010, but as requirements push energy efficiency further, it will take another mechanism to encourage low income families to live in a new home. Hopefully, there will be a happy combination of stimulus fund $$, combined with green jobs, combined with low-interest loans for high-performance low-income housing, and other factors to help this along for everyone's benefit.

  39. Riversong | | #39

    A Plethora of Errors
    I just read through most of the comments and Dennis Hayes Oct 12 post demanded a response.

    In addition to the demeaning lecture style, which assumed that none of us understand R-values and the myriad other issues relevant to energy efficiency, his statement is rife with errors and contradictions.

    "The single biggest factor responsible for superior energy efficiency in a home is its shape."

    It's size, not shape. Shape is secondary. And simplicity of shape is often more important than the geometrical configuration.

    "Of all possible geometric forms the pyramid has the least amount of exposed exterior surface area per square foot of heated or conditioned floor space."

    Comparing 2-storey structures with the same footprint, then it's true that a pyramid has 35% less exterior wall to floor area ratio any other shape (because it has only 75% of the floor area of a rectilinear house, some of that unusable). But if the first floor area is included as part of the exterior shell, then the improvement is only 15%, and the exterior wall to volume ratio and the total surface area to volume ratio are 46% and 92% worse than a simple rectangular house. It's surface area to volume which typically defines the efficiency of a shape.

    "Heat loss (or gain) is a direct mathematical function of surface area."

    Yes, if only conductive transfers are considered. But not if radiant transfers (much higher at roofs, and a pyramid is all roof), doors & windows (how do you shade roof windows in a pyramid?), and air exchange are considered.

    "A common term used in construction today is "R" value when talking about insulation in the roof or walls. This is a very misleading term… It is supposed to indicate the relative energy efficiency of a material in stopping the flow of heat through a wall or roof."

    While "whole wall R-values" are a very useful metric, R-values are meant only to compare insulation materials to each other, and only in terms of conductive transfer. R-value is not at all "misleading" to those who understand what it measures.

    "A truer statement proving energy efficiency is the actual energy bill history paid by the homeowner…An analogy in the automobile industry would be to say that "this vehicle gets 15% better gas mileage than a comparable sized auto".... "

    Not at all. Just as a heavy-footed driver can get lousy mileage with a 4-cylinder car, the actual energy consumption in a home is at least as much a factor of use patterns as it is of the objective efficiency of a building. The only accurate comparison is objective energy analysis based on "typical" occupancy.

    "A perfect vacuum in the exterior shell would lose zero BTU’s."

    No, it would conduct zero BTUs, but it could lose or gain considerable heat by radiation.

    "We often forget about the principles of earth contact homes and how efficient they are if we incorporate the lower floor to be in contact with the cool earth 55 degrees."

    Sure, except deep earth temperatures vary dramatically in different climate zones. The earth temperature is within 1 or 2 degrees of the average annual air temperature. So, while the earth might be 55° in the lower Midwest, it can be anywhere from 35° to 80° in the lower 48. 80° earth in Florida doesn't help much with cooling.
    "We have to sacrifice tradition of style and construction if we are to make a major impact in design efficiency."

    Not at all. Super-insulated passive-solar homes can be designed to match almost any vernacular architecture, though some details – such as no roof overhangs on traditional capes – might have to be altered for both durability and efficiency.

  40. Riversong | | #40

    One Disagreement
    While most of Martin's suggestions are right on, there is one that may be problematic - not just in Florida but in northern climes as well.

    The detail that could result in catastrophic roof failure is "cathedralizing" the roof with closed-cell urethane foam insulation, particulary if a non-permeable roofing or roof membrane is used.

    An article, "Compact Asphalt Shingle Roof Systems: Should They be Vented?" by Peter E. Nelson and Jason S. Der Ananian, recently published in Journal of ASTM International, Vol. 6, No. 4, reported on WUFI analysis of cathedralized roofs in a cold (Boston) and warm/humid (Maimi). Their model house had asphalt shingles on either felt or SRAM (bituthene) membranes with either fiberglass & poly VB, open-cell foam with variable perm vapor retarder, or closed-cell foam insulation, and either vented or not.

    Their modeling was for a five-year period with a small wind-driven roof leak in the third year.

    Their conclusions for the warm/humid climate included:

     The least tolerant roof assembly in either climate is the unvented closed-cell polyurethane insulation roof assembly with SRAM applied over the sheathing. This roof assembly creates a vapor trap and is slow to dry although the SRAM is supposed to prevent leakage from wetting the sheathing. Additionally, the closed-cell polyurethane foam will not allow leakage water to filter through and can promote deterioration of the wood roof structure with no visible indication of a roof leak.

     In hot humid climates, the most durable roof assembly is the vented open-cell polyurethane with either felt or SRAM applied over the sheathing due to decreased drying time of the interior gypsum wallboard when compared to the unvented roof assembly. However the addition of venting in a hot, humid climate further complicates the location and construction of the air barriers and vapor retarders. Functionally, the foam needs to be applied to a second layer of sheathing, which also serves as a redundant air barrier. A layer of blocking would form a vented cavity between two layers of sheathing. The first layer of sheathing could also be changed to glass-fiber faced gypsum sheathing, to improve the fire resistance of the assembly.

     Unvented permeable shingled roofs are also a viable consideration in hot humid climates, although it would be slightly less durable. The unvented roofs would be less expensive and simpler to construct that the vented roofs. These unvented roofs result in increased drying time of the inboard gypsum wallboard. The insulation must remain permeable to avoid a vapor trap.

  41. Bob Guthrie | | #41

    opinion from Australia on Hot Climate design
    The research here says that reducing radiant heating input is THE issue for hot climates.
    Reflective insulation is the best way to achieve this - typically level and just above the ceiling.
    Other forms of insulation, and roof ventilation, do almost nothing to reduce radiated heat input.
    And, in the evening, this other insulation retains heat gained during the day, when the ideal is to dump it as rapidly as possible.
    Most houses here are not air-conditioned, and are not occupied during the day. Evening comfort is a priority.
    Bob Guthrie

  42. Mario Carballo | | #42

    Houses on stilts
    Many coastal communities in Florida have houses on stilts with no insulation under the floor, just plywood and flooring material leaving the space open to the elements. Many contractors here propose using close/open cell foam under the floor but I always wonder the durability and effectiveness of spray foam when exposed to Florida humidity, heat and tropical force winds (never mind hurricanes).
    Are there any studies done in such situations...what is best way to insulate floors on stilt homes?

  43. GBA Editor
    Martin Holladay | | #43

    Spray foam durability
    I wouldn't hesitate to use closed-cell spray foam to insulate the floor of a stilt house. The foam will actually add to the structural integrity of the floor system, making the house better able to withstand hurricanes.

    I would protect the cured foam with OSB or plywood after it is installed.

  44. ColinFlorida | | #44

    I see a lot of builders in Florida use foil backed rigid insulation on the interior CMU walls, yet J.Lstiburek of Building Science corp, even Florida builders code, says do not use vapor barriers on the interior of walls. Yet, the foil backed XPS or Polyiso is a class I vapor barrier. What gives? Did I miss something?

  45. GBA Editor
    Martin Holladay | | #45

    If you install fluffy insulation like fiberglass along with an interior vapor retarder like polyethylene, you can easily end up with mold problems associated with inward solar vapor drive. For more information on this problem, see "When Sunshine Drives Moisture Into Walls."

    These problems only occur with air-permeable insulation materials like fiberglass batts. Inward solar vapor drive won't occur through a continuous layer of rigid foam (especially foil-faced foam), however, because rigid foam is an air barrier as well as a vapor retarder.

    In general, thin vapor barriers without R-value can be dangerous, since they prevent walls from drying out. Vapor barriers that have R-value (like continuous rigid foam) are generally safe, since the surfaces of these vapor barriers never get cold enough to allow condensation. (The R-value prevents condensation.)

  46. ColinFlorida | | #46

    Thanks Martin. I just bought a house and discovered all exterior walls of the house has a polypropylene (inside) vapor barrier ...I got Rogered. so yeah, my whole house is now musty since rain season started (Florida).

    Thanks for filling in the missing fact! I figured I was missing something. I've read a lot of datasheets, articles and books and nowhere have I read that CI changes the rules on vapor barriers. It makes sense though since there is no direct contact between the high RH side and low temp.

    My new wall assembly will be std.stuccu-CMU-CI-Furring-Drywall. I can see how the foil backed polyiso could be ok since on one side of polyiso the CMU can dry to exterior and CMU is not affected by water as it normally carries it. On the other side of the polyiso is wood firing and drywall which dries to the inside. So as long as the CI itself isnt getting wet via capillary action it would be ok. For such a wall assembly would unfaced XPS still be better or does it really matter?

    Thanks, Colin!

  47. GBA Editor
    Martin Holladay | | #47

    Q. "For such a wall assembly, would unfaced XPS still be better?"

    A. No. Green builders try to avoid the use of XPS, because it is manufactured with a blowing agent that has a high global warming potential. For more information, see "Choosing Rigid Foam."

  48. mills9867665789760 | | #48


    If you were to write this article today (12 years later), would your recommendations be the same?

    Any new techniques or products you'd recommend for hot climates, especially hot and humid CZ-2?

    1. GBA Editor
      Martin Holladay | | #49

      Two things I noticed:

      1. Minimum code requirements for wall insulation and roof insulation are more stringent now than when this article was written. It should go without saying that insulation R-values should not be less than minimum code requirements (in Zone 2, often R-38 for ceilings and R-13 for walls).

      2. If you want to construct an unvented conditioned attic (a "cathedralized" attic), it's possible to achieve that goal without spray foam. For more information, see "A New Look at Conditioned Attics."

      1. mills9867665789760 | | #50

        Thanks for the reply!

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |