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

Superinsulated House Specs

Two cold-climate designers choose optimal wall designs, roof designs, insulation R-values, and window glazing

R-49 walls and an R-66 roof. Jim Riggins is building this superinsulated house in Monument, Colorado.
Image Credit: Jim Riggins

Designing a superinsulated house can be tough. How much insulation should you install under a slab? Should your walls be sheathed with rigid foam, or should you go with double-stud walls? Could SIP walls save you money? Does the added cost of triple glazing make sense?

The answers depend on your climate, your performance goals, and your budget. Coming up with an optimized design requires careful heat-loss calculations, multiple energy simulations, and construction cost estimates. For those who haven’t yet struggled with these calculations, it can be instructive to compare the conclusions of thoughtful designers who have gone through the exercises.

Energy-efficient homes in Massachusetts

Two recent magazine articles describe cold-climate superinsulated homes designed by experienced professionals who did their homework. Their conclusions are instructive.

The first article, “High-Performance Homes on a Budget” by John Abrams, appeared in the January 2011 issue of the Journal of Light Construction. In his excellent article, Abrams provides the specifications for a cluster of superinsulated homes built on the island of Martha’s Vineyard in Massachusetts. The designs were perfected with the help of energy consultant Marc Rosenbaum. (For a photo of one of the homes, see Airtight Wall and Roof Sheathing.)

Abrams built his walls with double rows of 2×4 studs. He explains the decision this way: “After years of experimentation with a variety of approaches, we have settled on cost-effective wall and roof assemblies that meet our insulation standards … and minimize thermal bridging — and that our carpenters and other tradespeople are comfortable with.”

Minimizing residential energy use in Colorado

The second article, “Heading for Zero: Smart Strategies for Home Design” by Jim Riggins, appeared in the February / March 2011 issue of Home Power. Riggins describes the multiple calculations and simulations he used to design a superinsulated house in Monument, Colorado.

When Riggins began the…

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

    Exterior Foam = Too Tricky
    The thing I like about both of these projects is that they both concluded Exterior Foam is not-so-user-friendly.
    I would highly recommend purchasing online access to JLC and Home Power to view these 2 projects.
    Then you will have access to the Archives.

  2. homedesign | | #2

    More Links?
    lots of good stuff in both articles....
    do you have any links to more details,plans or photos ?
    besides JLC & Home Power

  3. GBA Editor
    Martin Holladay | | #3

    Another link
    Here's another article with information on the Martha's Vineyard project:

  4. homedesign | | #4

    not just cold climate designers
    I think the articles would be "worth reading" for designers from other climates as well.

  5. GBA Editor
    Martin Holladay | | #5

    Thanks, John
    I always keep you in mind as I worry about our struggles addressing GBA's cold-climate bias. I'm glad these two houses (one from Colorado and one from Massachusetts) get a thumbs-up from Texas.

  6. homedesign | | #6

    the thought process not the houses
    I appreciate the thought process of both designers/builders
    Not that the homes should be duplicated in Texas.

  7. a5WwLPKdXb | | #7

    rough rules of thumb for central texas

    I was wondering if you have rough rules of thumb for depth of insulation (I'm assuming cellulose) in both walls and (flat) ceilings, in either your neck of the woods or Austin (where I'm located).

    Do you feel double stud walls are overkill in our climate?

    Some of your other postings seem to indicate that you think the accepted standards, even for green building, may be on the low side.

  8. homedesign | | #8

    Response to Paul M.
    In the past my homes(N. Texas) have been not-so-thick open cell Spray Foam.(or SIPS)
    Not-so-thick because Spray foam is VERY expensive per volume.

    I think the high cost of sprayfoam makes it too costly to go very far beyond code minimums.

    I am planning to switch to Cellulose.

    I do believe that it will be worthwhile to create a very good air barrier with an Airtight Drywall Ceiling Plane and Airtight Drywall (or Sheathing) Walls......
    and then either cross hatch walls or perhaps even double walls...with densepack Cellulose.

    For the ceiling I think large quantities of loose blown cellulose should be not-so-costly.
    I have not done any real budgets or energy modeling yet... but hope to soon.

  9. user-836662 | | #9

    Response to John Brooks
    From 2/11: ":Martin,
    lots of good stuff in both articles....
    do you have any links to more details,plans or photos ?
    besides JLC & Home Power"

    I don't have more details published anywhere for the Monument home, but feel free to e-mail me with questions or request specific details at: [email protected]

    Jim Riggins

  10. a5WwLPKdXb | | #10

    Thanks John
    Cross-hatched 2x6 walls and cellulose sounds right. I'd be very interested in hearing about what your modeling shows. The kicker is always hitting that sweet spot, where you're getting the most bang for the buck. I appreciate this site and the people who take the time to post.

  11. 5C8rvfuWev | | #11

    John Brooks off topic

    Your comments re crosshatching are something I've been thinking of for N GA. Could I email you with a couple of questions to avoid hijacking here?

    Apologies, Martin ...

    Joe W

  12. davidmeiland | | #12

    Martin... about your bio
    ... sez your first PV panel from 1980 is still working after 29 years. Isn't that now either 30 or 31 years? Or did it retire?

  13. GBA Editor
    Martin Holladay | | #13

    Response to David Meiland
    I guess I should figure out how to update my bio....

    The PV panel is still working: Testing a Thirty-Year-Old Photovoltaic Module.

  14. user-947651 | | #14

    Mass house unvented roof insulation?
    I couldn't access the article but based on your spec list it looks like a cellulose 'hot 'roof system is being promoted. In light of the BSC research:
    could you explain in more detail the workings of that approach relative to the moisture potential. My work is with existing homes but I learn much from the experience of approaches in new construction. Thanks

  15. Kopper37 | | #15

    High R Enclosures
    Martin - You've probably seen this one, but I thought I'd post it for others to view.

    It's a link to a recent Building Science Corporation research report on "superinsulated" construction techniques for ALL CLIMATES.

    Having emphasized that their methods can be used in a variety of climate zones, the authors did recommend some climate specific insulating values, based on some assumptions of cost effectiveness. See p. 10 of the full PDF document.

    Also, there is a revealing graph on p. 39 that shows the energy loss relationship between air leakage and R-Value. My take away is that superinsulated structures need to be built very tight in order to gain the full benefit of the increased insulation levels.

  16. Kopper37 | | #16

    Bottom Plate - Photo #2
    Regarding the Colorado house (James Riggins), it appears that there are (2) 11 7/8" LVL plates at the bottom of the wall.

    Is this correct? If so, why not just one? Seems like a significant thermal bridge for a double-stud wall. I'm wondering if it was a local code requirement . . . ?

  17. GBA Editor
    Martin Holladay | | #17

    Response to Mike
    Here's what John Abrams has to say about his unvented roofs: "These homes are designed and built with 'hot' roofs -- that is, there is no structural ventilation. In our experience, good insulation and air-sealing will keep excessive heat and moisture out of the roof system, so there's no need for vents. In fact, we've used this roof system for 30 years; it wasn't part of the code until recently, but long ago we convinced our local building inspectors of its effectiveness. We have never had performance or moisture problems with hot roofs."

    Not all building inspectors agree that the code allows unvented rafter bays to be filled with dense-packed cellulose; many inspectors insist that such roofs are only allowed when a sufficient amount of rigid foam insulation is also installed above the roof sheathing. When in doubt, discuss your plans with your local building officials.

  18. user-836662 | | #18

    Response to Daniel Ernst
    RE double LVL sill plate: You have hit upon one of the handful of pitched battles I waged with regional building engineers. Their rationale: the inner 3 - 4 inches of sill plate sits over foam which insulates the slab perimeter. And even though the outer 2x6 wall is the load-bearing wall, the inner 2x4 wall will still experience a load, especially in the 2-story portion of the house. My argument: the load on the inner wall is minimal and the added strength of LVL over dimensional lumber is more than sufficient for support. I lost. (I did win on the argument to not double the top plates anywhere in the house, plus another issue on being able to use insulated headers over the windows and doors rather than solid)
    After fuming for a few days over the thermal bridging penalty, my framing contractor suggested we surround the base of the wall on the outside with a 2' high, 1.5 inch thick layer of XPS foam as a decorative bump out skirt before stucco. That, plus the spray foam on the foundation wall, offers a thermal break for the massively over-engineered sill plate. This is what I plan to do.

  19. user-836662 | | #19

    Unvented Roofs
    Interior wall and roof condensation was also a concern during design of the Monument house because of the very thick walls (12") and roof (18"). The vapor and dew point profiles for our climate showed the potential for condensation on the inside face of the exterior sheathing. I addressed this by specifying the 3 to 4 inches of closed cell spray foam on the entire sheathing inside face, both wall and roof. At greater than 2 1/2" thick, the foam provides a very low perm vapor barrier at the sheathing, and protects the sheathing from condensation.. By using vapor semi-permeable drywall with clay finish as the interior air barrier, I allow the wall and roof cavities to dry to the inside in the event of moisture infiltration. Additionally, the spray foam was a major part of our strategy for ultra-air sealing.

  20. Kopper37 | | #20

    Response to James Riggins
    Thanks for the explanation. Although not in your original game plan, it sounds like you resolved the issue. Good luck with the remainder of the project!

  21. user-945392 | | #21

    Response to Mike & Martin re: Un-vented Roof Insulation
    I'd like to reiterate Mike's question regarding the use of unvented cellulose-only roof construction.

    John Abrams' success with the system seems very contrary to Joe Lstiburek's experience with the system.

    As far as I can tell, the 2009 IRC precludes this approach (my understanding is that cellulose does NOT meet the IRC definition of "air-impermeable insulation," such that exterior rigid insulation IS necessary).

    Am I missing something?

  22. GBA Editor
    Martin Holladay | | #22

    Response to Will Tinkelenberg
    No, you're not missing anything. This is a long-simmering debate, not yet resolved.

    In New England, some cellulose installers have been dense-packing rafter bays (without any ventilation chutes) for 20 years. Joe Lstiburek is just the latest expert to call it risky. Plenty of people report success with the technique, though.

    Of course, all codes are local codes, because it's up to your local code official to decide whether any technique you propose meets the intent of the code. Believe it or not, many practices that appear to violate the code are routinely approved.

    Welcome to building science -- always a field full of lively debate.

  23. user-945392 | | #23

    Response to Martin
    Hi Martin, thanks for your response. I'm working on developing small house plans for selling via the Internet, so I'm planning on following the IRC. Hopefully this will make my plans more readily approvable in more locations. (I'm even planning on at least showing the fire sprinkler systems... how's that for a debate well beyond the 'simmering' stage?!... ). Currently, I'm basing my construction details on the use of a fair amount of insulating sheathing, both for walls & roofs, even though the simplicity of a cellulose-only system has its appeal.

    Thanks again; thanks for confirming that I wasn't completely missing something (at least as far as this topic goes!...).

  24. MPFadR5c6L | | #24

    I'm currently trying to build an energy efficient home (virtually by myself--no contractors). As the home isn't getting built very quickly, at least I have at put together an impressive library! Anyway, I have alot to talk about, but my question at this time pertains to insulation, specifically the insulation in my ceilings. To be very brief, I have icf walls, so the wall insulation is mostly under control. All my floors are concrete with embedded 3/4" hydronic tubing. My full basement slab was relatively easy to insulate (I used 2" rigid foam sheets), but I need to figure out what my options are for insulating my ceiling cavities.

    The floor decking material is 22 gauge galvanized 2 inch corrugated, which sits on 16 gauge 12 inch c joists on 12 in OC. There is an average of 5 inches of concrete on the decking (4 inches above the top corrugation). The purpose of the insulation is to keep the radiant heat from passing through the floor .
    I like the idea of spray foam, expanding in every crevice as it's sprayed, but i am unsure of all my pros and cons. Would a wet blown cellulose or something else provide me with the same or more benefit? Will a soy based spray shrink? I heard that there is a spray that actually increases R value over time--any input? does the steel decking decrease the R- value? Can the spray foams have any affect on people with allergies? Are there any vapor permeable (breathable) options with high performance #'s? I'd really be interested in this.

    Most importantly to me is the life cycle. I am aware that foam has come some ways since the 70's, but I am no expert and don't know how. I remember poking my finger through some old spray foam and it basically turned to a fine dust upon contact. Are there any guarantees in terms of R value degeneration? For example, if I spray an R-4.5 per inch today, what will the R factor be in 5 years, 10 years, 20 years?

    I am not an internet guru, and I therefore have a hard time finding unbiased information. I would like to do more homework about assessing all my options and weighing the pros and cons of each. Are there any reliable resources which might not be funded by companies?

    Finally, about spray foam. I would be interested in attempting it myself, but obviously I will need to get more into the science of mixing the chemicals, etc. What resources might be available for this direction--i.e chemical formulas, tanks, spray equipment, etc. ?

    Before I make an uninformed guess, any input would be greatly appreciated. Thanks.

  25. GBA Editor
    Martin Holladay | | #25

    Response to Deniz
    I doubt if you will be mixing your own chemicals for spraying your own spray foam. The equipment used to install spray polyurethane foam insulation costs tens of thousands of dollars.

    I don't think you need to insulate between floors. As long as your heat is within your conditioned envelope, the heat isn't lost.

    It's too bad that you only installed 2 inches of foam under your basement slab, because that's where your insulation belongs -- not between the first and second floor. Four or 6 inches of foam would have been much better under your basement slab.

    If you are planning a vented unconditioned attic, then the best insulation for your attic floor is a deep layer of cellulose.

    I hope you have carefully detailed the insulation at the perimeter of your floors -- that steel decking and the embedded PEX tubing are setting you up for a lot of heat loss at the perimeter of the floors unless you take care to insulate your perimeters well.

    Remember, put your insulation where it matters -- at your thermal envelope, not between floors.

  26. MPFadR5c6L | | #26

    Thanks Martin. How do you
    Thanks Martin. How do you find the time for all the blogging? I certainly appreciate it.
    I thought about 4 inches beneath the slab, but I figured that since the temperature 5.5 ft below grade was pretty constant and well below frost (Atlantic City area, NJ is no deeper than 24"), it wouldn't make a big enough difference. I guess I was wrong. I did, however, take great care around the footing. There is 12" of crushed stone containing a perimeter drain and filter fabric on both sides of the footing. The 12" stone was hand tamped and was at an even height to the top of the footing. Then I installed a 10 mil poly vapor barrier, taped at all perimeter corners, plumbing penetrations, and coming up the wall approx 6-7 inches (just above the finished slab height). Then came the 2" closed cell rigid foam board (R-10), taped at all penetrations and seams. At the wall edge, the rigid foam sat on top of the top of the footing and was directly bonded ( using canned foam) to the icf wall foam.

    As far as the perimeter edges go, that's a bit tricky. The way the project turned out, I wasn't thrilled with the Faswall system I used (as they were ALOT more fragile and much more wood content than I thought), and in combination with the amount of window openings I had, I didn't trust the final wall product to support the concrete deck. I wound up running an 8" steel I-beam around the entire perimeter of the house, welded to metal plates I embedded in the concrete at the top of the walls. The beams, which I welded together and also to plates in the walls, provide the structural integrity of both top plate and header systems. Also, the joists are welded directly to the i-beams. Hopefully this all makes sense.

    The faswall block is 12 inches wide, with roughly 1.5" webbing and 3" mineral wool insert. There is about 5-6 inches of concrete filled cavity. The beams are centered on this concrete-filled cavity, which leaves about 5 inches to the outside of the wall. There will be a 2" foam insulation piece on the outside edge of the entire perimeter, "glued"to the top of the faswall block, and extending to the top of the finished slab height (8" for the i-beam, 12" for the joists, 2" for the decking, and 4" for the slab on top of the deck) for a total of 26 inches. The concrete will be poured over the edge of the decking, catch about 1-2 inches of the joists, and cover the front edge of the i-beam. The thinnest portion of the concrete will be about 2-3 inches thick.

    There will be no attic, as the second floor roof will be of the same -- a flat poured concrete roof , which will also contain 6" of living roof + drainage. As the living roof on the 5 inch (avg) slab will avoid temperature swings, would the cellulose be better here? if not, would thicker open cell work better? under what circumstances would the cellulose work such that it is more effective than spray? or is it just cost?

    Back to the floor insulation: Uninsulated? I understand that there won't be heat loss through the building envelop, but won't I be losing more heat from one floor and gaining it in another--transferring too much heat to the center floor(where both the floor and ceiling is putting out heat). I have no justification, either from experience or theory, but i can't help but feel that by not insulating the floor, it would take more time to heat it because more heat is lost to the room below.
    Sort of like a stone thrown in a lake and watching the rings radiate from it. if there is insulation, it would bend the "ring" so transfer more heat in the concrete before proceeding to the floor below.

    Just trying to figure out how this works. Thank you.

  27. GBA Editor
    Martin Holladay | | #27

    Response to Deniz
    It sounds like you are building a very durable (and fairly expensive) house.

    You have posted your question on a page called "Superinsulated House Specs," but your house is not superinsulated. You have 2 inches of foam under the slab, and 2 inches of foam on your walls. (I don't think that half of the thickness of the Faswall blocks will be providing much R-value at your floor perimeters, because the Faswall blocks include wide webs that are thermal bridges.)

    Many cold-climate builders are using much higher R-values than you are planning.

    I don't know if it's too late for you, but you seem willing to invest a lot of money in concrete floors and PEX tubing, but not much money in insulation. If I were building a house, I'd spend more money on the envelope, and less on the PEX and concrete.

    Your roof assembly doesn't sound like a good match for cellulose. I would advise using EPS or polyiso. It's important to be sure there are no thermal breaks at the intersection of your wall insulation and your roof insulation.

  28. sunplans | | #28

    House design, climate zones and passive solar
    After reading yet another insightful article by Martin, I am inspired to comment on several postings at once. A full disclaimer, I was the architect (and I hope one of the exceptions to the postings on “Are architects the problem?”) who created the original house design from which Jim Riggin’s design evolved. (Home Power has generously taken off the log-in requirement for the article for at least a while so that I could post it on my web site and allow others to read it without a subscription.)
    Comment 2 asked about links including plans. Jim Riggin’s Colorado design evolved from our Northern Sun design. Jim purchased our CAD (computer aided design) Files that come with a copyright release that allow for modifications by an experienced residential design professional such as a builder, designer, engineer or another architect. In this case, Jim himself was qualified to make the adaptations to the plans and he included bits and pieces of several of my Northern Sun designs.
    As part of my CAD files (or Construction Prints), I provide Custom Energy Specs that include among other things recommendations on how to adapt the home to the climate in which it will built. In Jim’s case, my specs included several options for achieving the high R-values for each of the walls and roofs. I list several options for each because I believe a cost-effective choice can only be made when the builder and their framing and insulation subs participate in the decision. Jim then evaluated what became his Heliospiti design with extensive energy modeling in order to make his choices.
    Comments 4, 5 and 15 mention other climates. I have also prepared Custom Energy Specs for the same house design in milder climates. While my recommendations for insulation values vary based on a combination of climate and customer personal values, the same design can typically be used for a wide range of climates. Yes, I might recommend different glass –especially on the south wall. Yes, I might recommend adjustment to the overhang lengths above the south windows. And yes, sometimes I recommend that a customer reduce the amount of south glass, but that is rarely the case and if so, typically in Climate Zones 1, 2 or 3. (Typically I do not recommend increasing south glass since on my designs it’s already high, and balanced with both the thermal mass in the design and the structural integrity of the south wall.) There are other things that I also review for outside the scope of this posting.
    The passive solar design guidelines that I have been using for over 20 years evolved out of the earlier “Passive Solar Guidelines for Builders” (no longer in print) published by the previous Passive Solar Industries Council which is now the Sustainable Building Industries Council. Their book Green Building Guidelines has an abbreviated version of the passive solar guidelines. Doug Balcombe with NREL had told me that the guidelines were developed based on the performance of hundreds of passive solar homes. Before I posted some of my house designs to be sold over the internet, I ran calculations with their then Excel-based software of the same design in 6 different climate zones to see how both the heating and cooling loads were affected. The results were such that I felt comfortable that the home designs could be adapted through varying the specifications (building details that mainly addressed changing insulation levels) of the home more so than the design itself. This allows home owners semi-customization of a stock home design when they do not have the time or inclination to work with an architect. (Yes, I often think I am shooting myself by offering both stock plans and a custom design since the latter then appears expensive, but other professionals in the passive solar industry convinced me in the past that there was a need for more passive solar designs.)
    Comment 14, etc. regarding Cellulose insulation in-between rafters: I too am uncomfortable with unvented cellulose packed between rafters UNLESS there is closed-cell foam insulation on top of the sheathing or sprayed below to the sheathing – both which have the amount vary with the climate. However, I am also comfortable with cellulose alone in the cavity IF a metal roof is installed over 1x4 nailers over 30# felt since there is an air space that can be designed to allow for venting on the outside of the assembly.
    Comment 23: While in general I would encourage others to design small passive solar homes to sell over the internet since people are asking for more smaller designs, I have found that copyright infringement is rampant (people have even emailed me with photos of the house that they built by copying the design off of my web site or from my book and not purchasing a plan) and would say to do it because you have a passion for it and not as a money making investment. I sometimes feel that the entire house design profession needs to move away from stock plans and more towards creating right-sized custom homes where the builder can be brought in from the beginning and not after a house plan is purchased or designed, or at least prior to modifying a design.
    And with that in mind, I wonder if Green Building Advisor could initiate a Google Map where builders like the ones found on this site and that read Fine Homebuilding, Journal of Light Construction, Home Power, etc. could post their locations. It seems like it could really help designers and architects, as well as home builders get a head start on start getting a builder involved early. I would gladly send my potential customers there since one of the frequently asked questions is where can I find a builder that is familiar with the low-energy, passive solar concepts that your designs incorporate. Although I do not have details, I remember going to one link for instance where an organization had posted people’s favorite green vacation spots. I wonder if this is the way to get one started? Or this Google Earth Gallery? Here is an example one:
    Ok, enough subjects!
    Debbie Coleman, Architect, Sun Plans
    All climate zones in North America

  29. GBA Editor
    Martin Holladay | | #29

    Response to Debra
    Thanks for all the helpful information -- and thanks for the Google Map suggestion, which we'll take under advisement.

  30. John Straube | | #30

    Durability needs to be considered.
    Martin, thanks for some of the most relevant blogs/articles in the business these days.
    The concern I have with a lot of the discussion is that people talk about what is the best way to get a high r-value (SIPS cost X% more, or exterior foam Y% more) without comparing durability. A double stud wall will be less expensive if labor and wood costs are low, and always uses lower cost insulation. Of course. But how does it compare in terms of durability? While people seem concerned about unvented roofs, and calculating where condensation might occur, few seem to do the same with double walls, which are very similar to unvented roofs dense-packed with cellulose in terms of risk. Of course they work often, but they also fail more often than most would be comfortable with.
    Double walls solve thermal bridging at the studs. They often do not do a good job of solving thermal bridges at floor penetrations or the connection to roof and wall. They certainly can be made pretty good in this regard, but the success will depend on the details.
    But double stud walls are tough to build to the same level of moisture risk level as exterior insulation. (And the exterior insulation can be rockwool if you dont like foam.) I think we can do it by using exceptional airtightness testing with IR cameras, a vapor permeable sheathing, or spray foam to the back of the sheathing, but the comparison is far more complex than comparing R-value of Wall A vs Wall B, and such simple, middle of the wall comparisons can often lead to very wrong conclusions, comparing apples to turds rather than apples to apples.

  31. GBA Editor
    Martin Holladay | | #31

    Apples to turds?
    I very much appreciate your comments.

    I love data. Do you have any? I'm specifically intrigued by your statement, "Double walls ... are very similar to unvented roofs dense-packed with cellulose in terms of risk. Of course they work often, but they also fail more often than most would be comfortable with."

    Do you have failure stories? Failure studies? Failure statistics? Or just a gut feeling (backed by WUFI modeling) that these cellulose-insulated double-stud walls make you nervous?

  32. John Straube | | #32

    Apples to polished turds
    I have data of one building (MC measurements of outdoor sheathing over a few years) I am not able to share but it showed that the exterior sheathing MC rose well over 20%. I also have met folks over the years in Western Mass and Eastern New York state, and Alberta at seminars who tell me that they had problems. On the other hand, BSC has looked at countless examples of air leakage condensation rot in standard 2x6 fiberglass batt walls, which should, based on physics, be safer than double stud walls. It is this later experience which first prompted the exterior insulation recommendations: the energy saving benefits of thick layers came later.
    Failure statistics are not available for window leakage in standard tract housing: expecting stats for such an odd and custom type of assembly like a double stud wall is asking for building science nirvana. Hence why we need to rely on traveling around the country and talking to people, answer calls for forensic investigations, etc.

  33. GBA Editor
    Martin Holladay | | #33

    Thanks, John
    I appreciate the further information.

    OK, GBA readers, here's the challenge: who will be the first to build a home with Roxul mineral wool on the outside of the wall sheathing? We need volunteers. We need good details. We need photos.

    We're ready to publish anything you send.

  34. albion10 | | #34

    insulatiing slab verses basement ceilings
    Reviewing many articles: installing foam under slabs- the norm but very little about insulating the basement ceiling. Is it because the basements are finished? Moreover, little about crawl spaces an insulation.
    With millions of existing homes, so little mention of superinsulation renovating why? Is it used is not as cool as new?

  35. GBA Editor
    Martin Holladay | | #35

    Response to Tim Miller
    It is usually advisable to include a basement within a home's thermal envelope. So you want to install insulation on the basement walls and under the basement slab, not at the basement ceiling. That way the home's furnace, water heater, pipes, and ductwork are all within the home's thermal envelope.

    There are many articles at the GBA Web site about insulating crawl space walls. Use the "search" function at the top of every GBA page.

    When it comes to superinsulating existing houses, we have also covered the topic at length. Do a search for "deep energy retrofit."

    Here are some links to get your started:

    Energy-Efficiency Retrofits: Insulation or Solar Power?

    Best Construction Details for Deep-Energy Retrofits

    The History of the Chainsaw Retrofit

    Roofing and Siding Jobs Are Energy-Retrofit Opportunities

    What Is a Deep Energy Retrofit?

    Deep Energy Retrofit: Focus on the Envelope

    Remodel Project: Deep Energy Retrofit


  36. MPFadR5c6L | | #36

    Roxul--I'll give it a whirl
    " a home with Roxul mineral wool on the outside of the wall sheathing? We need volunteers. We need good details. We need photos." ---> I don't know how to get the purple box with previous post quotes.

    Anyway, I've posted a couple times here, thinking that the house I'm currently building is insulated well (I'm in Atlantic City, NJ area), but this site is definitely reshaping the way I think about insulation.

    I switched from the foam ICF foundation/basement wall to the Faswall block system for my first floor because I feel a home needs to breathe. Now that I agree that the block itself does not provide a sufficient thermal break, I would like to add additional insulation but still retain the breathability of my wall system. This Roxul product may be the answer to this situation. I know nothing about it other than it's not foam, but if I can install it on the outside of the exterior walls and prepare it for stucco, we have a winner.

    Thanks again for all the resources from this site.

  37. sQNDdNGbLN | | #37

    Using the Kraft paper under Thermo-ply or not?
    Martin, as a published author with Taunton Press and 30 years under my belt in the construction arena, I finally met my match where I don't have a clear answer to a project I'm currently working on. It's my hopes, that with your experience you will be able to shed some light on the following:

    I'm currently working on a home that was built in 1932 as a 3 level home in the Northwest (Spokane, WA [zone 5B]). However in 1957 two levels were removed to build two other homes in the area and the first floor was moved to its final location on a daylight basement also in 1957. The second floor still in place with hardwood has the original set of stair located in a closet. There's a pocket door to the closet and a wooden Pella folding door to the second floor entrance, which is now a semi standup attic space. Any cellulose insulation that was up there has been removed.

    The attic space is stick framed with 2x6's 2-feet on center to create a Dutch hip roof. I'm making this space a conditioned attic space. So insulation between the living space and conditioned attic space does not have any vapor barrier to meet the City Building Code. And the living space is the same sized as the conditioned attic space, so there are no knee walls in the attic to insulate.

    So with that said, the 2x6 rafters will be insulated with R21 fiberglass insulation. In order to do this, the City wants me close off all soffit vents, remove all roof vents, and install an attic roof fan with humidistat control. The insulation would then be installed directly up against the roof sheathing.

    Now here's where the dilemma comes into play. We are not going to cover the insulation with wallboard because of the added weight on the rafters. Instead we are going to use a new product called Thermo-ply sheathing by Berry Plastics Tapes and Coatings Division to protect the insulation, because of the low headroom. Thermo-ply, however was designed for use on the exterior of a structure (housewrap).

    I chose the Structural Grade Blue grade because it's lightweight, has a thickness of 0.137, and is a 100% recycled material (a "Green" building product). It has a perm rating of 0.4 making it a vapor retarder that's semi impermeable. All Thermo-ply seams will be taped once installed.

    The City says, that with the Thermo-ply attached to the bottom of the roof rafters, it does not matter if the fiberglass insulation has a vapor barrier or not. The Thermo-ply will act as the air-impermeable barrier and if there is another vapor barrier on the backside of the Thermo-ply, that vapor barrier should not have any vapor reaching it. Plus the vast majority of insulation is between the roof sheathing (the surface that you do no want to be a condensing plane) and the vapor barrier. If the vapor barrier is kept above the condensing temperature, then there will not be any condensation.

    So the question is, would you keep the insulation face, with asphalt-impregnated Kraft paper, under the Thermo-ply or would you remove it? If you keep it, do you think any moister would get trapped between the Kraft paper and the Thermo-ply? Of course, by keeping the Kraft paper it would make it easier to install the installation. So what's your thought on this dilemma? Any input would be most appreciative, thanks, the toolguy.

  38. GBA Editor
    Martin Holladay | | #38

    Response to Leon Frechette
    1. Your plan is a bad one. Using fiberglass batts to insulate an unvented sloped ceiling is never a good idea.

    2. Your plan violates the 2006 IRC, section R806.4, which requires the use of air-impermeable insulation when the insulation is applied directly to the underside of the roof sheathing.

    3. Thermo-Ply sheathing is not a new product; it's been around for years.

    4. There is no reason why you can't install Thermo-Ply sheathing over kraft-faced batts; there is no incompatibility resulting from these two vapor retarders being back-to-back.

    5. To make your insulation plan work, you need to switch to an air-impermeable insulation (spray polyurethane foam or rigid foam sheathing), or you need to include ventilation baffles from soffit to ridge.

    6. Don't forget, your minimum insulation R-value must meet code. In Zone 5, the code requires a minimum of R-38 ceiling insulation.

  39. GBA Editor
    Martin Holladay | | #39

    Response to Leon Frechette
    If you choose to build an unvented roof assembly using closed-cell spray foam between the rafters, then it doesn't matter whether or not you have a ventilated soffit. There won't be any moisture generation in your cornice. Most people wouldn't vent the soffit in that case. If the soffit is already vented, though, there is no harm in leaving the ventilation holes.

  40. sQNDdNGbLN | | #40

    Soffit vents or not?
    Martin, thanks for your comments in regards to my question using Kraft paper under Thermo-ply or not. If I decide to take your suggestion of spraying the rafters with spray polyurethane foam (closed-cell), do you recommend in enclosed eves to install vents in the soffits to help with air circulation in this area should moisture come about during the winter months and/or heat build-up during the summer months?

  41. user-460363 | | #41

    Faswall webbing material is NOT a thermal bridge
    I am writing this to “set the record straight” about thermal bridging in wallforms made of cement/wood-chips. I’m a homeowner doing research on what to use for my own house. And I’ve noticed a few places on GBA where there exists this notion that the webbing material of Faswall is a thermal bridge. This is a pretty serious claim, but it turns out that it’s not true, and I’d like to offer an explanation. Both for my own benefit, and also for the benefit of other homeowners doing their own research in the future.

    Faswall is on the opposite side of the continent from me. However,a comparable product, Durisol, is located close to us. So my data and research comes from looking at Durisol. They both use a mix of petrified wood chips, recycled from softwood lumber production (pallets, off-cuts, scrap). I believe the webbing material for both products is very similar, but I could be mistaken. Let’s assume it’s similar.

    I’ve seen the lab report that tested the webbing material to have an R-value of 1.75 per inch.

    It is fair to say that a material with an R-value of 1.75 per inch is a fairly poor insulator. Especially when compared to XPS at R-5.0 or mineral wool/Roxul boards at R-4.0.

    However, it’s NOT fair to say that a material with an R value of 1.75 per inch automatically becomes a thermal bridge in a wall assembly. It depends on what else is going on in the wall assembly.

    A thermal bridge is an area in a wall assembly that is RELATIVELY or COMPARATIVELY conductive to heat energy when compared to the other areas in a wall assembly. The classic and notorious example is wood or steel studding. A softwood lumber stud has a value of around R-1.25 per inch, and it’s typical fiberglass batt neighbour is around R-3.1.

    A 2x4 wall has some of it’s surface area represented by a 3.5” deep piece of wood, for an R-value of (3.5* 1.25 =) R-4.4. Then it has a greater surface area represented by 3.5” of insulation material, say fiberglass batt (3.5”*3.14=) R-11. Yes, there is sheathing and siding and drywall, but I’m trying to isolate the concept of thermal bridging and keep it simple.

    Part of the wall is R-4.4 and the other part is R-11. That represents a huge disparity in the relative insulating value of each. So thermal energy finds it much easier to flow through the weakest link in the chain, being the studding, and you therefore have a wall assembly with tons of thermal bridges. It performs nowhere near R-11 when you look at the whole wall assembly.

    Let’s pretend for a moment that we had the option of building a wall of ONLY softwood lumber. And that we could get it 12 inches thick, using really big, perfectly square logs. Rather like a log home. Such a wall would have an R-value of 12 * 1.25 = R-15. But because the wall assembly was all the same material, across the whole surface area and depth, there would be NO relative bridging. There is NO relative weak spot. So the relatively poor insulator, wood, can be a thermal bridge in ONE assembly, but definitely NOT a thermal bridge in another assembly.

    If you look at a typical Durisol wallform – let’s look at the 12” thick form – see the photo attached. It has 4 dimensions in the photo:

    T – total thickness of the form, being approx. 12”
    t- thickness of the webbing on the inside (and the outside) in the area where the concrete goes, being 1 ¾”
    I – thickness of the insulation insert, being 3”
    H – thickness of the concrete fill, being 5.5”

    Across the wall’s surface area, you will have places where there is a sandwich of webbing-insulation-concrete-webbing. Or you will have webbing only.

    If the webbing is R-1.75 per inch, and there is 12” of it, then its R-value is about R-21.

    The concrete sandwich part is: 3 ½” of webbing (R-6.1) plus 3” of Roxul board (R-12) plus 5.5” of concrete (R-0.5) for a total of about R-18.6.

    This makes a wall assembly that is relatively FREE of thermal bridging, because the different areas of the wall have relatively similar R-values. There is no obvious place where the heat is going to tunnel out, like with a highly conductive steel stud, or a typical softwood stud. And when you look at the webbing in isolation, you actually see that it’s the strongest part of the wall, at least thermally speaking in R-values. So an R-21 web inside a wall that’s rated R-21 is definitely NOT a thermal bridge in that wall assembly.

    Plus, once you get into relatively thick walls, the “rules” about thermal bridging get more obscure. Because thermal energy doesn’t just take straight paths, but rather it interacts and dissipates inside a wall – especially a thick one. The bottom line is that “thin” walls like a standard 2x4 or 2x6 wall are vastly more susceptible to obvious bridging.

    The thermal imaging photos also show a distinct lack of thermal bridging.

  42. GBA Editor
    Martin Holladay | | #42

    Response to Zenon Tymosko
    You've done a good job of explaining thermal bridging. You understand the principles well.

    However, your analysis of heat flow through a Durisol wall is incomplete. You haven't made a 2-dimensional analysis, much less a 3-dimensional analysis. You have made a modified one-dimensional analysis, which is of limited use.

    Your analysis would be a pretty good way to estimate the R-value of a Durisol wall if the voids were filled with air. Once the voids are filled with concrete, however, your analytical method becomes incomplete.

    I don't know whether Durisol expects the insulation insert to face the interior or the exterior; but for the purposes of thermal analysis, it doesn't matter much. Let's assume the insulation inserts face the interior.

    There is 1.75 inch of material (wood chips plus additives) between the concrete and the exterior. (I'm ignoring the stucco). That material has an R-value of about R-3. That's not much. That's all there is separating the concrete from the outside world. When the outdoor temperature is cold, the concrete will be fairly cold too -- definitely colder than the indoor temperature (although not, of course, as cold as the exterior air).

    The interior heat will leak through the Durisol block and be absorbed by the cold concrete. The maximum heat flow will occur at the inside corners of the cold concrete. At those points, there is only 4.75 of Durisol material between the cold concrete and the interior. That's something -- and it may be enough to avoid cold stripes on your wall. But analyzing these two-dimensional heat flows is tricky. You need a good software program like THERM to see the isotherms created by this type of heat flow.

    Once you introduce three dimensions into the equation, with details at the bottom of the wall and the top of the wall, the thermal analysis becomes still more complicated.

    Most building scientists find the the best way to determine the thermal performance of this type of wall is to measure heat flow through a completed wall assembly measuring 8 feet by 8 feet or 8 feet by 12 feet. Oak Ridge National Laboratory has a calibrated hot box that is large enough for this type of measurement, but as far as I know, no one has ever performed this testing on a Durisol wall.

  43. kerrar | | #43

    For those reading this relatively old article, you'll probably find that the links to the two articles no longer exist. I couldn't find the first article, but the second article is still on the Internet Wayback machine and can be found here:


    1. GBA Editor
      Martin Holladay | | #44

      Thanks for letting me know about the dead links. I have corrected the links to both articles.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |