GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon

Community and Q&A

Won’t double-stud walls have a condensation problem?

Michael Roland | Posted in Green Building Techniques on

Using exterior rigid foam solves thermal bridging and prevents condensation within the batts in the wall cavity. Double-wall construction also solves thermal bridging, but what about the dew point within the batts? Won’t there be a condensation problem?

I’m designing a new house and need to choose a wall assembly: double-wall with batts or loose fill vs. single-wall w/ exterior rigid foam.

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. GBA Editor
    Martin Holladay | | #1

    The question you raise is usually called the "cold OSB" problem. To read all about this question, see How Risky Is Cold OSB Wall Sheathing?

  2. Expert Member
    Dana Dorsett | | #2

    Condensation doesn't happen in batts: Because batts are extremely vapor permeable, and low-mid density batts are so air-permeable, whenever the coldest surface of the cavity reaches the dew point of the entrained air in the cavity the moisture condenses on that surface, not in the fiber itself. The colder it gets, the more moisture gets concentrated on that surface- lowering the dew point of the air in the cavity to the temperature of that cold surface. At any other point within the batts the fiber temperature is ABOVE that temperature, and no condensation occurs.

    If that surface is a hygroscopic material such as OSB sheathing it doesn't condense either, but instead adsorbs into the material, never achieving a true liquid state (unless there is so much moisture entering the cavity from air-leaks that the OSB saturates.

    If a hygroscopic INSULATION is use such as cotton batting or cellulose fill, the insulation in contact with the cold surface will adsorb moisture too, which results in lower moisture accumulation in the sheathing. Cellulose can take on quite a bit of moisture before saturating and losing R-value, and can store then re-release the moisture as conditions change. This protective characteristic makes it a favorable material to use in thick high-R assemblies where the volume of entrained air is larger, the moisture content of which might otherwise concentrate in the sheathing. Dense-packing the cellulose is also protective, since it limits the air-permeability, and roughly doubles the amount of cellulose at the same time it increases the moisture buffering capacity.

  3. Jesse Thompson | | #3

    Our opinion?

    Double-stud wall with any type of batt insulation is a high risk wall system in a cold climate, due to the cold sheathing issue. Even so, there are many versions of these walls still standing that were built in the 80's & 90's. We know people who built houses like this, and their buildings haven't fallen apart yet. However, too risky for us to recommend.

    Well air sealed double stud wall (better than 1.5 ACH50) with dense-packed blown in cellulose (not loose fill), or perhaps blown in fiberglass seems to be a low risk wall system. The worse you air seal, the more the risk goes up with this wall system.

    Thick enough rigid foam over 2x seems to be a very low risk system, even with mediocre air sealing. That's why Building Science is such a fan of exterior foam, you can be a pretty sloppy builder and still not get building failures with this system.

  4. GBA Editor
    Martin Holladay | | #4

    Here's my opinion: if you want a double-stud wall, lower your risk with the following details:
    1. Use plywood sheathing or diagonal board sheathing, not OSB.
    2. Include a ventilated rainscreen gap between the siding and the sheathing.

  5. Expert Member
    Dana Dorsett | | #5

    +1 on the plank sheathing and rainscreened siding recommendations! OSB is mold-food, and needs to stay much drier than solid wood to avoid damage. Moisture gets into end grain much easier and wicks further than the side-grain of planking, and the processing heat that uses the wood's own lignin as the "glue" converts a good fraction of that lignin into simpler wood-sugars more readily fed upon by fungus & mold.

    (Not trying to pick a fight, but...) I don't quite understand how Robert Riversong's no-sheathing with ship-lap siding smack up against the insulation approach to truss & double-wall high-R construction is going to hold paint, (or hold UP) over time. He usually has substantial roof overhangs to limit direct wetting, and always uses cellulose, but I'm skeptical that it has sufficient resilience to withstand the test of time.

  6. Michael Roland | | #6

    Martin, Jesse & Dana,
    Thanks for your responses. I had intended to use plywood and build a rain screen, but I have three more questions. First, for a WRB is a house wrap like Tyvek the best way to go, or would traditional asphalt/felt paper work. Second, can these double for exterior air sealing, or would interior drywall sealing be adequate? Third, would mineral wool batts like Roxul provide significant enough reduction in air movement compared with fiberglass to be worth the cost, and would it have any advantages over dense-packed cellulose? Thanks.

  7. GBA Editor
    Martin Holladay | | #7

    Q. "For a WRB is a housewrap like Tyvek the best way to go, or would traditional asphalt/felt paper work?"

    A. Either one will work.

    Q. "Can these double for exterior air sealing?"

    A. Asphalt felt is certainly not an air barrier. Some people claim that housewrap can be used as an air barrier, but I'm skeptical that it can. On the exterior, use the plywood as your air barrier; tape the seams.

    Q. "Would interior drywall sealing be adequate [as an air barrier]?"

    A. If you follow the Airtight Drywall Approach, drywall is a perfectly good air barrier. You can have an interior air barrier, an exterior air barrier, or two air barriers; it's your choice.

    Q. "Would mineral wool batts like Roxul provide significant enough reduction in air movement compared with fiberglass to be worth the cost?"

    A. Switching from fiberglass to Roxul won't provide much of a difference in air leakage rate.

    Q. "Would mineral wool have any advantages over dense-packed cellulose?"

    A. No.

  8. Lucas Durand - 7A | | #8

    I'm not calling you out to the bike racks ;-)

    I don't quite understand how Robert Riversong's no-sheathing with ship-lap siding smack up against the insulation approach to truss & double-wall high-R construction is going to hold paint, (or hold UP) over time. He usually has substantial roof overhangs to limit direct wetting, and always uses cellulose, but I'm skeptical that it has sufficient resilience to withstand the test of time.

    What is your source of concern?

  9. Expert Member
    Malcolm Taylor | | #9

    Lucas, I would imagine the concern varies by region. Riversong's maxim that the wall exterior should have five times the permeability of the interior might work well for interior moisture migrating through the wall but, at least here in the PNW, I can't see how you would detail and flash the siding to manage rain driven wetting without sheathing or a rainscreen.

  10. Lucas Durand - 7A | | #10

    Rainscreens are a good idea on the wet coast but I'm not convinced they're needed everywhere.

  11. Lucas Durand - 7A | | #11

    Window flashing sans exterior sheathing (not-so-wet climate):

  12. Expert Member
    Malcolm Taylor | | #12

    Lucas, It's an entirely academic debate for me as our code mandates both rainscreen and sheathing for seismic shear walls, but I'm interesting in learning how you deal with a couple of situations: How do you handle head flashing on openings and the horizontal flashings that divide differing siding materials? Do you rely on tape? Does the absence of sheathing preclude exterior trim that butts against rather than laps the siding? I also see in your photos what look like tapered shims on the window sill. Our code requires a sloped sill but in addition that the window be mounted to allow a gap for any moisture to drain between the window flange and sill. This creates a real problem for insulating. Any thoughts on how detail that condition to increase its efficiency?

  13. Expert Member
    Malcolm Taylor | | #13

    Lucas, thank you for the detailed reply. I don't seem to be able to access the one link above to your project, which I'd really like to see. Could you repost it?

  14. Lucas Durand - 7A | | #14

    I am using structural plywood sheathing but it has been located on the interior side of the assembly and also functions as the air barrier and vapour retarder.
    I elected to omit exterior sheathing because, under the circumstances, it seemed excessive.
    Details of the assembly can be found here.

    Regarding window openings:
    I used beveled cedar siding to slope the membrane at the bottom of the openings.
    The shims you see on the sill were cut from the same siding so that they sit level from front to back on top of the sloped sill.
    The gap around the frame is treated first with an EPDM gasket which is designed to push into that space.
    To allow for the drainage space you mentioned, I didn't push the gaskets up against the flange - the frames are 3 1/4" deep so I left ~2" between the gasket and the flange.
    When it gets a little warmer, I plan to foam the remaining space on the interior side as usual.
    The windows are flanged and include a slim brickmould, so I flashed over the flanges at the jambs using tape.
    Head flashing is pretty standard since the exterior framing provides enough lumber to nail the flashing to.

    Regarding other flashing and cladding:
    In general, the exterior framing provides enough nailing opportunities to hang horizontal flashing between different cladding types (ie, siding, ledger, siding).
    Vertical joints between cladding types are backed with felt (and shingled with any horizontal flashing that it may intersect).
    The biggest challenge is the trim because, depending on width, it doesn't always leave a nailing opportunity at the end of any abutting cladding.
    I still haven't fully commited to one approach yet for this problem, but have several options I'm considering.
    I would like to go with trim that I can butt the siding into, but if I can't make it work, I'll probably end up just lapping the trim over the siding.

  15. Lucas Durand - 7A | | #15

    No worries.
    I fixed the link in my earlier post but just in case, here it is again:

  16. John Brooks | | #16

    Excellent Pressure Boundary and Illustrations.

  17. Lucas Durand - 7A | | #17

    Hi John,

  18. Patrick Walshe | | #18

    You could do both a double wall and exterior insulation, though of course it would be more expensive. We did a 9.25" staggered stud wall and 2" rigid rock wool on outside on Vancouver Island - see and

  19. Expert Member
    Dana Dorsett | | #19

    Here out by the bike rack...

    The permeability of 1x shiplap siding with a couple coats of exterior latex paint is about 1 perm, and any wind-driven rain that gets by the ship-lap gets redistributed by the cellulose, since it can't vent-dry in that assembly. Repeated dampening/wetting of cellulose (even 4lb dense-pack) causes it to settle, but in the meantime it's drying rate through the siding is pretty slow. Yes, it can hang in there for quite awhile, but if the only drying path is through painted ship lap, any errors in flashing or even changes in wind patterns that increase the frequency or intensity of wetting events can affect the long term resiliance of the assembly. But that all goes away with even a 1/4" of rainscreen between the siding and the drain-plane material. I suppose if built to perfection (yeah, right, PERFECT) and have big roof overhangs it'll outlive the builder, but perfection is a lot to ask.

    Cellulose can buffer HUGE amounts of vapor-diffusion moisture and even a fair amount of air-leak moisture if dense packed. But asking it to handle repeated bulk wetting, however minor, isn't really going to cut it over decades of (ab)use, and all walls in the real world see bulk wetting events- it's only a matter of how wet and how often.

    Rainscreens may not be absolutely needed everywhere, and maybe not ANYWHERE, (not even in Vancouver, if you can build perfect walls that require no maintenance), but anywhere it rains or gets cold for extended periods they offer a huge resiliance margin from real-world construction errors.

    After this past weekend's li'l Nor'Easter I ended up having to dig down to below the bottom clapboards on three sides of my central MA house, despite it being essentially a hip-roofed 1-story with 24" overhangs. Leaving the snowdrift up against the siding over time would surely have wetted both the siding and the sheathing, and gotten into the subfloor, etc. raising the average moisture content. Whether this approached rot-levels of moisture is academic, but it's seen snow levels like that dozens of times in the past century, and I'd be surprised if it had been dug out every time. In Riversong's stackup digging it out every time would be critical, lest the bottom layer of the cellulose become saturated. But with rainscreened siding (on my house or his) it wouldn't much matter- even if the lower planks of siding remained buried in snow for a few weeks, the sheathing wouldn't be much affected, and I could have spent my time skiing instead of digging.

  20. Lucas Durand - 7A | | #20

    It seems like your concerns are mostly theoretical...
    I don't think any of the concerns you've expressed about Riversong's design have ever been observed.
    He uses a WRB behind the siding to mitigate against bulk wetting, if not some dampness.
    And I think all but his earliest designs allow the wall to dry in both directions.

    Somewhat ironically, Robert has often argued that rainscreen cladding creates an opportunity for the types of flashing errors you refer to - that can make the "perferct wall" not-so-perfect after all.

  21. Expert Member
    Dana Dorsett | | #21

    Sample size = small

    Time scale < 30 years

    We'll see. I don't suppose any of his houses have been instrumented & monitored for moisture content of the cellulose (particularly under windows), and siding over several seasons(?).

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |