GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Choosing a wall assembly: discussion

blamus20 | Posted in Green Building Techniques on

I really tried to not ask this question since I’m aware that it has been discussed a ton, and I have read a ton on the subject. And even if I’ve accepted the fact that there’s no clear answer, I still have to choose a wall assembly – double wall, 2×6+4″foam, SIP, or some hybrid. Around R40 (I have 12″ to work with)

The top priority here is actually budget and ease of construction/quality control.

I’ve gone through many articles from BSC and GBA amongst many other case studies. This is how my thought process went:

At first most sources seem to suggest the most cost effective way to do it is double stud wall. That works with me too but my exterior wall would be the structural one. Just as I got comfortable with the concept, I start reading about all the “potential” moisture issues. And then I see the “ideal” double stud wall with interior wall sheathing (vapor) AND exterior (structural) wall sheathing. And since I’m framing this outside-in, I didn’t really like the need for the interior OSB layer from a construction sequencing perspective. I then see that people spray CC foam on the inside of the exterior sheathing, but I didn’t like that because it adds a different type of insulation in the form of foam adding complexity and cost, so I might as well go exterior foam sheathing.

And then I see that a lot of people are saying that exterior foam is the best way to go in terms of thermal, air leakage and moisture control anyway, and everything else is inferior in their own way. I was concerned with the extra complexity of windows, doors and siding installations, but maybe I am over thinking it – 2×6+foam seem to be the standard amongst contractors anyway. But I thought it costs more (foam is expensive).

Then I read conflicting reports – “practical residential wall systems – R30 and beyond” ( show a generally higher cost for 2×6+foam compared to double walls (my calculations show 1.2-1.8 $/sqft more than double wall), while Moisture management for high R-value walls” ( suggest that 2×6+foam is the cheapest way to go. (1.4 $/sqft LESS).

At least one thing they relatively agree on – that SIP incremental cost as compared to their reference wall of 2×6+batts is about the same in both articles.

In terms of cost 12″ SIPs don’t seem like a bad option either compared to the other 2, and the extra speed is money, but then there are horror stories about moisture issues at the seams.

So what am I to think!? Whats the most simple, budget friendly, trouble free way to get close to R40? Of course I take these cost estimates with a lot of salt, but I thought at least they would agree with each other in relative terms and trends. Which doesn’t seem to be the case.

background – House is simple rectangle, 2 floors, 4 corners (65’x26′) supported by 2 concrete basement foundation boxes (2 of 22×22, one box as garage, other is movie room+entrance) so 3 floors total. Mono pitch shed roof. Climate zone 5 @ 6000′. See attached illustration.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. JTyler | | #1

    Bernard - I am in the process of building a 2x6 house with 6" of foam. There is a list of materials to consider when pricing this type of assembly; the insulation itself is only part of the battle. I went with recycled foam, and couldn't be happier with the decision. Once I realized it was possible to get perfectly good foam for roughly 40% of the cost of the sticker price, going with a REMOTE assembly was an easy choice for me. If you haven't already checked it out, I recommend reading the REMOTE wall manual published by the Cold Climate Housing Research Center. This document should answer most questions you have about siding and penetration details. Good luck with your project!

  2. GBA Editor
    Martin Holladay | | #2

    If you haven't read it yet, I suggest that you read this article: How to Design a Wall.

    In my article, I suggest two approaches. Either approach works fine. If you have trouble making decisions -- and I'm guessing you do -- talk to a builder. (Eventually, you're going to have to choose a builder for your project, so you might want to start on that process.) If your builder prefers one of the two options I describe, that's the way to go.

  3. SwitchgrassFarmer | | #3

    That is a nice looking home Bernard, great site too.

    Some thoughts on SIPs:

    1. I noted that the upper floors are cantilevered. SIPs require good bearing on inner and outer skins, this was an issue at our home built over the top of an ICF foundation. Can be managed with sill plates, slabs, etc, but perhaps slightly more critical for SIPs than other construction methods.
    2. You have a lot of windows, I can only imagine the view. However that probably means that your structural engineer will make you insert a ton of LVLs and other lumber in your SIPs to carry the needed headers etc. That negates some of the SIP thermal advantages, and diminishes any speed/cost advantage too. (Been there.)
    3. Seems like you do have an opportunity for a SIP roof, depending on span/support underneath. I am thinking the scale of your home is such that the depth of a SIP roof (with an added cold roof above) wouldn't be out of character. (We camouflaged the depth of our SIP roof with layered fascia and metal roof trim.)

  4. STEPHEN SHEEHY | | #4

    I would assume that the difficulties of working on a steep slope will determine what sort of wall you can build. Can you get a crane to the site to deliver and install SIPs? Can you erect staging on the downhill side?

  5. user-5574861 | | #5


    I heard Joe L. speak last week and his "Perfect Wall" assembly still remains the same, cladding, rain screen, exterior insulation, sheathing, insulated 2x6 wall, gypsum. It works all over the world according to him. He says don't fret over the insulation, pick want you want, just do it in the right order.

  6. Toyoland66 | | #6

    SIPs give me concerns about longevity, even if they are perfectly installed and air sealed you have an assembly with cold sheathing, and if there ever is damage it will be more difficult to repair than a stud wall.

    The difficulty of penetrations in walls with exterior foam is overblown I.m.o., decide where your drainage plane is and align your doors and windows with this plane.

    I think an assembly to consider is a double stud wall with the structural sheathing on the interior face of the exterior wall. The stud cavities can be insulated with a permeable insulation such as mineral wool which will keep the sheathing warm and allow it to dry to the exterior.

  7. Jon_R | | #7

    but then there are horror stories about moisture issues at the seams.

    When the SIP is properly air sealed and vented, are there any more issues than other partitions using OSB?

    I don't think there are any problems with double walls with reasonable interior humidity and pressure, a smart interior vapor retarder, cellulose, exterior plywood and vented cladding. Even more so if you use breathable, non moisture susceptible external sheathing.

    Some things to consider about fiber filled 2x6s + external foam:

    a) the minimum thickness of foam prevents most but not all condensation conditions. Even more common is sustained high humidity at the sheathing layer.

    b) these walls often can't dry to the exterior. And for long periods, the interior is the source of more moisture (so it won't be drying that way either).

  8. blamus20 | | #8

    Thank you guys, that was a lot of food for thought.

    So SIP walls are out, and it makes sense.

    Indeed the upper 2 floors (the main house) cantilever over the foundation, there will be posts INSIDE the house spanning both floors to transfer the roof and floor loads to the foundation, strategically placed to not interfere too much with furniture placement and aesthetically add to the architectural details.

    I'd like either a conditioned attic or no attic at all, so maybe SIP roof?

    The concern for staging of the construction is very real. As you can see the property is very steep (and rocky). I've attached the site plan and see what you guys think.

    Its very easy to get mixed up with REMOTE, PERSIST, Joseph L's perfect wall, etc. I assume that they are all variations of what I was referring to as 2x6+4"foam? Except apparently I can just do 2x4 if I choose to?

    What they all seem to agree on is to put the vapor/air barrier on the sheathing (OSB), and not the outside of the foam.

    So it does seem that exterior foam walls are the norm and more popular than the double stud. And my fear for windows and door details, cantilevering outtie windows etc is a non issue.

    So far we think, metal roof, and cement board cladding (in big squares), maybe some corrugated metal. and again, apparently my concern for siding attachment on the foam is moot.

    There seem to be more ways to "get it wrong" with the double stud, and most builders are very familiar with the exterior foam approach. (its code where I'm building - R19 between stud+R10 continuous)

    And the cost difference is negligible even with the simple shape of the structure. Well, I guess that settles it then! Thanks guys!

    ....should I do 2x4 16" OC or 2x6 24OC? Or even 2x4 24OC? I do plan to insulate between the studs with batts or cellulose just for a bit of extra R.

    And there does one source recycled foam? Where do they come from?

  9. GBA Editor
    Martin Holladay | | #9

    For more information on PERSIST, see Getting Insulation Out of Your Walls and Ceilings.

    You wrote, "What they all seem to agree on is to put the vapor/air barrier on the sheathing (OSB), and not the outside of the foam." In most cases, the membrane on the outside of the OSB is called a water-resistive barrier (WRB), and it is usually vapor-permeable, not a vapor barrier. That said, this layer can be a vapor barrier if you follow the PERSIST approach.

    For information on a related question -- whether the WRB belongs on the exterior side of the rigid foam, or the interior side of the rigid foam -- you may want to read Where Does the Housewrap Go?

  10. user-5574861 | | #10


    Whatever assembly you go with the keys are to 1) keep your control layers in the correct order for your climate zone (vapor control layer location and type can vary), 2) make sure the control layers are continuous and 3) make sure your ratio of exterior to interior insulation is appropriate for you zone if using rigid exterior insulation.

  11. SwitchgrassFarmer | | #11

    Stephen Sheehy's points about the site are indeed very important. Our home is on a hill top/side, and it greatly complicated things. From having to offload the tractor-trailers with SIPs at our neighbor's and store them there too bringing up a few at a time, to having to put in temporary fill for crane and telehandler pads. Scary times with cement trucks, wind problems, you name it. These issues lengthen and increase the cost of a project. (It is neat when it is done though...)

  12. blamus20 | | #12

    Hi Chris,

    "I think an assembly to consider is a double stud wall with the structural sheathing on the interior face of the exterior wall. The stud cavities can be insulated with a permeable insulation such as mineral wool which will keep the sheathing warm and allow it to dry to the exterior."

    I've been leaning towards the double stud wall from the beginning, thinking that its the best budget way to do it, until I read all about the moisture studies. Installing the sheathing on the inside of the exterior wall is interesting, that puts 4-6" of insulation on the outside. Would there be something on the outside of this wall to hold the insulation? How would that interfere with the rain screen gap? But the moisture risk is still there, as I'll still have more insulation on the inside of the sheathing. Or have I missed something?

  13. Toyoland66 | | #13

    In that assembly you could use a WRB on the exterior side of the studs, solitex mento plus is an example of one made for this sort of application. Using a permeable insulation on the exterior side of the sheathing gives it drying ability that you wouldn't have with foam exterior insulation making the interior to exterior insulation ratio less critical.

    On my own house I am using 4" rigid mineral wool over 2x6 walls, it has some advantages. But disadvantages include the rigid mineral wool is relatively expensive, there's added cost of labor to install furring, and 8" screws to attach the furring are also expensive. There's no perfect system.

  14. blamus20 | | #14

    Hi Chris,

    Could fiberglass batts be used on the exterior side as insulation in the wall you described?

  15. Toyoland66 | | #15

    Probably could, I think MW batts are a better material, many on this site would agree.

  16. user-2310254 | | #16

    I don't think fiberglass batts have the compressive strength for an exterior application, or am I not understanding the stack up?

  17. JC72 | | #17

    I suspect your climate at 6,000 ft in climate zone 5 is generally dry. With that amount of windows area I would consider a wall where the flashing and rainscreen details are relatively straightforward. Perhaps something like R12 Zip-R followed by whatever insulation of choice on the interior side of the wall. Rainscreen gap could be a simple wrap with a product such as Home Slicker.

  18. blamus20 | | #18

    Indeed its very dry in colorado in general, we get a lot of sun, even in the winter, and a decent amount of snow. Its alpine desert.

    What I liked about double walls is its simplicity - framing the way everyone knows it outside and in, and a cheap insulation.

    what I dont like about double walls is that now we know that we need to complicate things to deal with a moisture problem - add a sheathing on the inside, or a membrane, or spray foam the inside of the ext wall, or move the sheathing and add another type of insulation etc. - to the point where I wonder if its just easier and cheaper to just deal with the exterior foam on a single wall. And thats before I start thinking about the sequencing!

    alright lets try this -

    1. Frame exterior walls, floors, and roof. All exterior OSB and house wrap.
    2. Install windows, doors, rainscreen gap, siding.
    3. Build interior walls on the floor panel by panel, add membrain to the outside of interior wall (air seal layer and vapor retarder), seal around windows and doors openings.
    4. erect walls and secure to ceiling and subfloor
    5. blow in cellulose (through hole in membrain?) and reseal
    6. finish taping caulking and air sealing
    7. batts between 2x4s, run electrical etc.
    8. drywall, normal primer and paint.

    Does that sound good? So this way the membrain will be the air control and vapor control layer. Is it difficult to air seal the membrain this way?

    An alternative would be to forget the cellulose and just use batts all the way through. This way I wouldn't have to blow through the membrain and then reseal.

    Can the windows/doors be installed before inner walls are up?

  19. exeric | | #19

    Bernard, I don't think you really registered Chris M's idea.You should do some research on the ZIP system with R12 for yourself before just excluding it. It has a lot going for it in terms of simplicity.

  20. blamus20 | | #20

    ok i might have miss read about the interior Zip R. You mean zip-R instead of OSB as exterior sheathing right? which adds R12 to the 12" cellulose wall on the outside, and I wouldn't need to use house wrap, just tape the zip-R and that will be my air control layer and I'll have no need for vapor retarder layer?

  21. exeric | | #21

    Yep, that's about it. Also, unless you need a wall wind rating greater than about 100 mph, (don't know where you live) the framing needs no seismic reinforcement.

  22. blamus20 | | #22

    design windload = 131mph and snowload 30psf.

    Is the OSB/ZipR as structural sheathing sufficient?

    12" cellulose + R12 Zip-R will be almost R55? is that a little overkill? Is the point of the Zip R that it is not susceptible to moisture issues like OSB would be? I wonder because the actual wood part of the ZipR is still on the cold/outside like OSB would be. How is ZipR different from polysio/sprayfoam+OSB+tyvek?

  23. exeric | | #23

    Here's the structural specifications for the Huber R-panels.

    I would say that the way the structure is designed and the 131 mph wind loads required that the R-12 might be pushing the structural shear limits. There's a lot of windows in your design and I think the R-panels might be only designed for two stories. I'm not sure your plans would be a good place to use them on second thought.

  24. blamus20 | | #24

    Hi Eric, thank you for looking it up for me, I dont really understand it but I'm sure my engineer will.

    On a different note, I had a play with the Beopt energy modelling software. I modeled many different scenarios - and whats suprising (to me) is that when using a)R39 double wall + ZipR12, b)R19 2x6wall + zipR12, c)R19 2x6wall + OSB + 4"XPS, my yearly heating is between 69.3-75.9MMBtu, which translate to a difference of $54 per year in electricity use!? And thats with a minisplit + baseboard backup.

    Does that sound right? Maybe I did something wrong somewhere? Also I played around with the windows. People comment how I have so many windows, actually its only about 21% of floor area (750/3500sqft), but more than 50% are directly south facing, and reducing my window areas doesn't change my energy bills much at all, literally just 10 or 20 $ here and there. At least this I can explain by the passive solar effect?

    If going from code minimum R30 to R40 or 50 really changes so little, it will be hard to justify a double wall, or even 4" of XPS...............

  25. JC72 | | #25

    Bernard Lam,

    IMO building codes with regards to levels of insulation have really caught up these past 6 yrs, so you're right that it may not make sense to build a wall with a R value that's significantly higher than code.

    BTW..with the low humidity levels in your climate using cellulose in a double wall might be just fine because what little moisture that it might accumulate will quick dry out. Like you said, the house will be build in the high plains desert.

    KISS (Keep It Simple Stupid).

  26. STEPHEN SHEEHY | | #26

    Reply to #18: Bernard- Your proposed double wall assembly is exactly what we did in my house in Maine, except we installed the windows after both walls were built, since the windows are in the center of the wall.
    Installing the membrane on the outside of the interior stud wall was simple. Just leave enough to overlap the ceiling membrane. We did tape the exterior sheathing and the wrb as well. We ended up with a very tight building (.59ach50)

  27. srenia | | #27

    r20 walls equal about .05 U-value or 95% heat retention. R40 is about .0025 U-value or 97.5% heat retention. So for all the added cost you are getting around a 2.5% gain in heat retention. That's fine if you want that. I would work on air sealing and very high quality windows if you want greater heat retention. More ROI. Of course the greatest energy saver is smaller footprint.

    In Colorado you have to worry about fire so rockwool insulation is excellent when faced with that. So many choices so find the one that works for you. Cement fiber/steel/brick siding with rockwool will help protect your investment.

  28. GBA Editor
    Martin Holladay | | #28

    If an R-20 wall is a code minimum wall, that's the starting point for comparison, because building a wall with less than R-20 insulation is illegal (at least for builders following the prescriptive code).

    So here is a better way to make your point: If you upgrade your wall insulation specification from R-20 to R-40, you will cut the rate of heat flow through that wall in half.

  29. blamus20 | | #29

    Ok, now that I understand source energy vs site energy. Here are my "2 extreme cases":

    R31 (R19 insulated 2x6 + R12 ZipR) needs 18.7 (heatpump) + 5.4 (electric resistance) MMBtu/yr
    R51 (R39 double wall + R12 Zip-R) needs 17 + 5 MMBtu/yr

    So the difference is 2.1 MMBtu/yr and about $50 a year

    Other numbers:

    Hot water = 13.3
    Lights = 6.5
    Large appl = 7.9
    Misc = 21.4

    Whats the MMBtu/sqft for heating in zone 5 should I expect? Do my numbers make sense?

  30. Reid Baldwin | | #30


    The fact that a relatively large improvement in R-value of the walls yields a relatively small decrease in heating energy is an indication that conduction through the walls is a small fraction of your heat loss, even for the baseline. Other areas, like better air sealing, improved windows, and better floor and roof insulation might yield better payoff.

  31. srenia | | #31

    True. But just the wall portion, not the windows, doorways and other weak points of the assembly. The whole wall assembly is less r-value and only a fraction of the total heat loss. Even greater is that the wall assembly has only a few peak months where the greater thickness is really key. So the savings from going from r20 to r40 is less than what the r-value number would suggest to a consumer.

    Saying all this I have over r--80 insulation in the attics because it cost under a couple hundred dollars to do and it pays itself back. Whereas a wall assembly would add significate cost in construction and have little payback. Air sealing also has a greater impact as does making a smaller house.

    You are right that r-40 reduces the heat flow in half from r-20. But the savings to environment and the pocketbook are questionable. So at that point it becomes a designer choice and not necessarily a heat savings choice.

    The perfect wall doesn't exist. But keeping it simple with r22 rockwool and a advance framing techniques with zip wall isn't a bad way to go. With the wild fires I would stay away from foam in this instance. Zip is easy to air seal and quick to construct. Something the contractor will like.

  32. blamus20 | | #32

    Reid is right (duh!)

    I played around with a few other improvements, and got my site energy use down to around 13 MMBtu/year, which is just less than my hot water (4bed4bath) of 15.6.

    Walls stayed the same with 2x6R19+OSB+R12. The only change that made a significant difference was air leakage. I changed it from 3ACH to 1ACH - this doesn't "cost" me anything, but I can't really know if I will be able to get to 1ACH or not. 3ACH is the code minimum, so thats why I used it in the first place. went from 21.3 to 13.7 MMBtu/yr for heating.

    Other changes I made that didn't do much:

    2. Interzonal floors from R19 to R30 (to match the ext walls) - I thought this would make a difference, because I have alot of this in my design, I have a tuck under garage as well as cantilever all around.

    3. Roof from R50 SIP to R60 - I thought the roof could do with more insulation.

    between 2 and 3, I reduced 1MMBtu/yr. Double walls reduced 2MMBtu/yr, Doing triple pane U0.19 instead of code minimum 0.30 double pane also only made 2MMBtu/yr difference. Using a HRV also didn't do much compared to just mech ventilation.

    Sure they all add up, but the expensive upgrades are obviously windows and going to a double wall. Adding more insulation to the roof/floors is relatively cheap.

    And my aim was to get the heating down to the hot water level, which I did, so maybe getting a heat pump water heater will make more sense from here.

  33. exeric | | #33

    Hi Bernard,

    After thinking about it some more I think you might ask your engineer what would actually be required to make the framing of your home fit for 131 mph winds that are independent of having sheathing attached. It may be that having let-in wood bracing or metal Simpson ties of some sort in appropriate places would give the structural integrity needed at minimal extra cost. I just don't know.

    If that was feasible then using the ZIP R-12 would definitely be the way to go. Nothing else would come close in simplicity for insulating and air sealing the walls. That's especially important since the less time working up on scaffolds the better.

  34. exeric | | #34

    I should add a caveat. You will have so much radiant energy streaming through all those unshaded windows on the south side that your house might just be one big heat trap. Remember, the radiant energy at high altitude is much greater than at sea level. It may be a nightmare to keep your house cool so having the remainder of the walls adequately insulated could be the least of your problems. At least as your planned home is currently designed.

  35. user-1072251 | | #35

    If it is a windy site (sure looks like it), I'd go with the exterior foam.

  36. exeric | | #36

    "If it is a windy site (sure looks like it), I'd go with the exterior foam"

    Yes, that's probably easiest in this situation.

  37. blamus20 | | #37

    I think the site itself is not so windy, its low in a canyon, rather sheltered. But obviously the building department assumes the worst in that area. If I was building just half a mile north on top of the ridge line then yes it would be many times more windy. You know how it goes, they just draw a line and say everything in this area need this much snow and wind load, and the further you go into the mountains (eventually end up in rocky mountain national park), the more it need. My site is just on the edge of the city, I can't imagine seeing anywhere near 100mph wind. But I still have to comply to their code.

    Irrelevant fact - it really is incredibly windy and dry up in rocky mountain national park. The frozen waterfalls dry out and disappears from solid ice in the winter because there is so much wind constantly blowing on it.

    "If it is a windy site (sure looks like it), I'd go with the exterior foam"

    Do you mean OSB+foam would be a better choice even if the zipR system can satisfy the structural requirements? Or are you saying you doubt the ZipR will will be enough?

    The solar gain will also be very real. I'd be (pleasantly) shocked if the solar gain can provide too much heat in the winter - but then I have not lived in a well insulated house before. In the summer, I'm going to have to use blinds and shutters. In my energy model, it didn't show that much cooling needs at all. But removing the southern windows didn't change the heating needs much, and may have increased it a little. Having said that, the windows design/rearrangements will likely change. My architect is working on it.

    Maybe a heatpump hot water heater?

  38. exeric | | #38

    "Do you mean OSB+foam would be a better choice even if the zipR system can satisfy the structural requirements? Or are you saying you doubt the ZipR will will be enough?"

    No, I was just trying to be agreeable because I personally don't have the knowledge about the structural requirements to do that and how hard it would be. If you have someone who can tell you about the modifications required AND it's not expensive then sure, ZIP R would be the way to go.

    I'm pretty sure the structure would require modifications in a 131 mph wind zone in order to use ZIP R.

  39. exeric | | #39

    I just realized you were addressing Bob Irving. But since I agreed with him I guess my answer shouldn't complicate things.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |