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
Q&A Spotlight

Choosing the Right Wall Assembly (2013)

In a double-stud wall, will batt insulation accumulate moisture?

One wall option

One possibility for a double-stud wall assembly is to locate the structural sheathing on the inside, as Lucas Durand did on this house in Ontario.
Image Credit: Lucas Durand

Michael Roland is designing a new house and trying to choose the right wall assembly. It’s down to a choice between a double-stud wall filled with fluffy insulation, or a single wall wrapped in a layer of rigid foam insulation.

“Using exterior rigid foam solves thermal bridging and prevents condensation within the batts in the wall cavity,” he writes in a Q&A post at Green Building Advisor. “Double-wall construction also solves thermal bridging, but what about the dew point within the batts? Won’t there be a condensation problem?”

That’s the topic for this Q&A Spotlight.

Moisture won’t condense in batts

Don’t worry about condensation on batts, writes Dana Dorsett, because it simply won’t occur.

“Condensation doesn’t happen in batts,” he says. “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.”

While the condensing surface is picking up moisture, Dorsett adds, “at any other point within the batts the fiber temperature is above that temperature, and no condensation occurs.”

“If [the condensing surface] is a hygroscopic material such as OSB sheathing, [the moisture] 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),” he writes.

The real risk is that sheathing made from oriented strand board will rot. This problem is known as the “cold OSB” problem, according the GBA senior editor Martin Holladay. When OSB gets wet…

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. davidmeiland | | #1

    SEE stud
    I hadn't heard of the "SEE stud" mentioned by Peter Yost, so I read through the BSC article that he links to. Here's a screen shot from that piece showing assembly of non-load-bearing studs from scrap lumber and sheathing.

    I don't think we throw any lumber in the dumpster on my jobs. Small pieces get used as blocking, and what can't get used can go home as firewood. None goes to landfill. We could probably make some studs out of scrap using this method, but not a whole lot. Ordering the right framing lumber and using it efficiently is important. Sheathing scraps are another story--you have to cut out your windows...

  2. jinmtvt | | #2

    What is it with risk u guys like ?
    Has anyone done cost calcs on double walls with stuffing VS single wall + ext board insul ??
    Is there a large difference? does the difference justify the risks ?

    My 2 cents ( which are worth probable nex to nothing .. )

    If you really wish to stick with stuffed insulation and or batts,
    use enough exterior insul. boards so that the dew point will never reach
    on the sheathing used on the single wall with its cavity filled or partly filled.

    I understand the reason of using double wall to get to ridiculous R values such as R50+,
    which then is really hard and expensive using rigid boards

    Using a double wall stuffed in zone 5 or less seems much safer than 6-7-8

    What about fully opened up wall like the one on mr Chlupp's house, no sheathing removes the problem ??

    Still don't understand why someone would want any water vapor to move around/through a wall ...
    when we all know how to avoid it theoritically.

  3. user-757117 | | #3

    Structural lumber in non-load bearing walls.

    What is a shame is to use structural 2-bys for the non-structural wall. It’s just not resource-efficient.

    2x4s were selected for the exterior frame because it provides the right depth for "off-the-shelf" Roxul batts.
    But I should have taken more care when I ordered my lumber to specify lower quality 2x4s for the exterior frame.

  4. user-757117 | | #4

    Response to Jin Kazama.
    I'm not sure I understand...
    There are many ways to stick build a wall that are "low risk".
    Risk management is in the details of the design and the quality of construction - there doesn't have to be a single "perfect solution" for all climates.

  5. user-1140531 | | #5

    Quote from article above: “While the condensing surface is picking up moisture, Dorsett adds, at any other point within the batts the fiber temperature is above that temperature, and no condensation occurs.”

    Certainly it is possible for the cavity temperature to be lower than the dewpoint. So, if the cavity temperature is below the dewpoint temperature, how can the fiber temperature within the batts remain above that temperature?

  6. jinmtvt | | #6

    Lucas Durand: i apologize for
    Lucas Durand: i apologize for generalising, but what i really meant is that it sounds like many here and elswhere are going at great lengths to use a system that "might" be safe.

    With the current knowledge floating around it is easy to design to push the dew point to the exterior
    design temp or ever lower, why still have the dew point before the sheathing ?

  7. user-1140531 | | #7

    Dewpoint Location

    Why is it necessary to push the dewpoint to the exterior? If you use fiberglass, you still need an air barrier on the interior, so vapor will not be transported into the cavity by airflow. And we are also told that diffusion does not matter because there will be so little of it. Therefore, diffusion will not lead to significant wetting from condensation in a cavity that is below the dewpoint. So why care if there is dewpoint temperature inside of the wall cavity?

  8. jinmtvt | | #8

    Ron Keagle: because ur
    Ron Keagle: because ur situation assumes for a perfect installation

    But i am referring to instalation of a very large R wall with an exterior solid sheathing .

    Also, if dew point is exterior of sheathing, u do not need/want an airbarrier inside, because ur sheathing becomes the air barrier and it should all be air sealed neway.

  9. Expert Member
    Dana Dorsett | | #9

    Answer for Ron
    Ron Keagle asks:

    "Certainly it is possible for the cavity temperature to be lower than the dewpoint. So, if the cavity temperature is below the dewpoint temperature, how can the fiber temperature within the batts remain above that temperature?"

    With the gypsum side air-tight (but semi-vapor permeable) to the cavity, at equilibrium the dew point of the air in the cavity is the same as that of the conditioned space air. But when you cool the exterior side to below the dew point of the interior air, the sheathing reaches that temperature first, and adsorbs moisture from the air in the cavity, bringing the dew point of the air in the cavity to the temperature of the sheathing (whatever temperature that might be.) Unless there is a significant air leak in the gypsum, there is never condensation in the fiber layer until the sheathing is fully saturated and can't take on any more moisture. While some of the fiber may be below the dew point of the conditioned space air when the sheathing is cold, none of the fiber is below the temperature of the sheathing. The cold sheathing is dessicating the air in the cavity to below that of the conditioned space, which creates the vapor-pressure difference that drives moisture diffusion through the wall. Wood can take on quite a bit of moisture at vapor-diffusion rates before saturating, even through un-painted gypsum, keeping the fiber dry. But at air-transport rates through leaky gypsum it saturates much quicker, but the slower the leak, the longer it takes.

    If you let air leak in at high rates from the conditioned space the sheathing can and will saturate, eventually leading to condensation in the fiber itself, but that's at the air leak extreme (it happens often in sloppy construction, but would likely be avoided if one meets IRC2012 levels of air tightness.) Both the air-retardency and hygric buffering capacity of cellulose can mitigate quite a bit against more typical less-than-perfect air tightness at the gypsum, providing considerable resilience to the assembly. Even with minor air-leakage, the rate of air movement and therefore moisture transport is limited by the air retardency of the cellulose. With fiberglass it takes a density of about 1.8lbs per cubic foot to reach the air-retardency of even ~2.5lb damp-sprayed cellulose, and it takes ~2.2lb fiberglass to beat 3.5lbs dense packed cellulose for air-retardency. ( High density fiberglass batts run about 1.4-1.5lbs density, a bit below Spider or Optima minimal dense-pack spec.) But at any density the buffering capacity of fiberglass is nil, whereas cellulose can take on quite a bit without damage or loss of performance, especially in thick double-studwall assemblies.

    Building in a vented rainscreen gap between siding and sheathing allows the sheathing to shed some of it's winter-time moisture accumulation to the exterior too. Any time the outdoor dew point is colder than the interior face of the sheathing (which is nearly 100% of the time in winter), at least some drying can take place, as long as the exterior side of the sheathing is sufficiently protected from direct-wetting from the exterior. Rainscreens ROCK as a means of providing moisture resilience to less-than-perfect wall assemblies, since they limit direct wetting, and provide better/quicker drying capacity toward the exterior for the sheathing.

  10. jinmtvt | | #10

    Dana : considering gypsum
    Dana : considering gypsum plane seal as a vapor barrier is crazy and only temporary when regular folks live in the building without supervision.

    How many pins and tears do people make during the first 10 years of a house ?
    Then, i have yet to visit a recent building in which none of the gypsum joints have cracked
    due to wood drying/settling etc..

    Still trying to patch the basic deficiency.

    We are discussing about R40+ walls, using 12" of EPS under slab, using super high performance
    ( read $$$$$ ) windows, but skimping on exterior rigid insulation is being considered??
    What are you guys trying to achieve here with your exterior sheathed vapor barrier walls ??

    Lucas: tell me please, on the project picture used on this blog header, was an exterior sheathing installed after filling up with Roxul ??

    All of this is the same as trying to save with some PV arrays on top of a Zone 3-4 building that has unshaded south windows during all of the cooling season ... ( go through green homes section, and probably more than half of the projects there have some very basic shading deficiencies )

    Please pardon my "WTF" !!!

    But again, i might be just going nuts and probably overlooking some basics here ...
    And i do not have much experience with all the vapor stuff, but what i am typing and thinking sounds logical to me with my very limited knowledge, don't want to insult anybody here ! ( serious )

  11. heinblod | | #11

    dew point
    The dew point is the point where dew forms.

    If the dew point is within the insulant or wall then there will be dew in the insulant or wall if vapour gets into the insulant or wall.

    Use wufi. Purchase materials with guaranteed behaviour.

  12. jinmtvt | | #12

    Hein: nice software, i'll
    Hein: nice software, i'll keep it bookmarked!!

    What do you mean by " garanteed behaviour" ??
    examples "bitte" .

  13. user-757117 | | #13

    Response to Jin Kazama.
    No need to appologize for anything.

    There is no exterior sheathing used on the exterior of the wall shown in the photo.
    After insulating, housewrap (Typar) is installed as a WRB, then solid wood siding.
    The siding is a full 1" thick "dolly varden" planed from local Tamarack (Eastern Larch) and stained all six sides with solid latex stain.

    All of this is the same as trying to save with some PV arrays on top of a Zone 3-4 building that has unshaded south windows during all of the cooling season ...

    In what way?

  14. user-1140531 | | #14

    Wintertime inward vapor drive
    In another discussion, information was posted that states the total water amount passing through 32 square feet of drywall by diffusion, for the entire winter season is 1/3 quart. The interior and exterior humidity and temperature conditions were also stipulated, but I forget what they were. In any case, they were selected as being typical.

    How much water is transferred into the insulation cavity per 32 square feet of cold OSB during the heating season under the same conditions of interior/exterior humidity and temperature?

    If your OSB has turned to oatmeal, how do you know that the moisture originated due to falling sheathing temperatures as opposed to outward air leaks from the interior?

  15. Expert Member
    Dana Dorsett | | #15

    and more...
    Ron correctly observes:

    " considering gypsum plane seal as a vapor barrier is crazy and only temporary when regular folks live in the building without supervision."

    Which is exactly the problem with using batt facers or interior poly as the vapor retarder- it is guaranteed to be violated at some point (or many points) by random electricians, plumbers, and picture-hangers.

    Which is why the air-retardency (and buffering capacity) of the fiber layer actually matter, along with providing better exterior-drying capacity via rainscreens. OSB is on the order of 1 perm when bone dry- but rises to 2-3perms when wet, so with rainscreened siding it has substantial capacity to dry into the dry winter air, and can keep up with minor air leaks like picture hanger nails in most US climate zones, and in climates where it can't, the buffering capacity of the cellulose can pick up a huge amount of slack. There are clapboard sided houses in Saskatchewan insulated with cellulose more than 80 years ago without moisture problems in walls, in part due to the buffering capacity of the cellulose, but also the lower mold/rot potential of plank sheathing vs. OSB. It really works.

    If your OSB turns to oatmeal 99% of the time it's from bulk water intrusion from the exterior (and bad-flashing technique.) It simply can't saturate and fall apart from the moisture contained inside the wall cavity- it's own buffering capacity is quite significant, but you could get areas of mold and rot going from air leaks even in US zones 3 & 4, or even vapor diffusion in colder climates, if the vapor permeance of the interior side exceeds that of the OSB and housewrap etc on the exterior. In most of the US and Canada a 1-perm interior side vapor retardency would be sufficiently low to prevent OSB from getting moldy from interior vapor diffusion drives alone. Air tightness is a far more critical factor to control, and since air tightness is never guaranteed forever, the air retardency and buffering capacity of the fiber makes a difference in real-world situations.

    Hein: In het engels "dew point" betekent alleen de temperatuur punt van condensatie (en platz/locatie punt niet dus.) Binnen en hout bedekt lucht-dicht muur opbouwing de "dew point" van de ingehouden lucht is nooit groter dan de hout temperatuur, sinds hout opdrinkt het water wanneer it kouder dan de dew point van de vlakbij lucht is. Maar de cellulose of fiberglass is zeker warmer dan de dekking (tijdens winter, tenminste), en geen condensatie gebuurt daarin. Het water als "adsorb" blijt alleen in de hout, en niet as vloeibaar water in de insulatie dus. Maar met en lager dewpoint binnen de muur dan binnen het huis, water als gas komt laagzaam door via diffusie. (Snelheit een functie van de binnen kant materiaal van de muur en het verschil tussen binnen lucht vs. muur lucht dew points.) Duidelijk?

    OK enough of my bad swamp-German for one day... ( and that's even zonder een pilsje of drie- het gaat iets erger daarna! :-) )

    But WUFI does indeed model moisture transfer issues quite well, with good correlation between experimental measured data and the simulations for peak moisture content of different structural elements, drying rates based on real weather & climate data. It takes much of the guesswork out of Ron's questions.

  16. jinmtvt | | #16

    LUCAS: what i mean is that
    LUCAS: what i mean is that one needs to address base of the problems before going to secondary solutions ... if condensation point is move outward of the sheathing for design temperature,
    not need to worry too much of the rest..that is the basic.

  17. user-757117 | | #17

    Response to Jin Kazama.
    If you are wondering about the design of the house I'm building, the "base of the problems" is taken care of and there are no "secondary solutions" being used.

    There is more than one way to skin a cat.
    The location of dew point temperature within the wall assembly is (in this case) made essentially irrelevant by design.
    By locating the sheathing on the interior side of the assembly, the sheathing will be at room temperature with 100% of the insulation on the outboard side keeping it warm.

  18. jinmtvt | | #18

    Lucas : that is exactly what
    Lucas : that is exactly what i was talking about ..

    You addressed the basic problem here, which is a cold vapor retard/barrier.

    When i said "base of the problems" etc.. i was not referring to your picture of course.

    This here is the same as a recent blog, discussing about thermal breaks in stud framing.

    Dana : i love your choice of word: " it is guaranteed to be violated "
    ahhahhah ... puts the user back in the owner

    please explain "swamp-german" ??

  19. Expert Member
    Dana Dorsett | | #19

    Swamp-german (way off topic...)
    Swamp German is a light-hearted reference the dialects of the Germanic tribes who live at or below sea level (the swampy river deltas), which includes Flemish & Dutch, (but not Frisian, which is as much like Old English as it is to modern Dutch). There are many differences in dialect & accent, but it's pretty much the same language from Belgium & the Netherlands to some parts of Germany. Though some draw the line at the Netherlands/German border, the differences are more academic than real. To unify the language in spelling & grammar there are conventions agreed upon by Belgian and Dutch academics that define "algemeen beschaafd nederlands", (ABN) or "general civilized netherlandish", which most English speakers simply refer to as "Dutch." ( Zie ook: ) Low German speakers in Germany or South Africa don't necessarily use the same spelling & grammar conventions of ABN, but it's still more similar to ABN than it is to high German, and not hard very hard to follow. Local dialects and idioms can stray pretty far from ABN sometimes, but all native-speakers seem to understand & use ABN if the conversation gets too confusing.

    I mostly get the north-Hollands accents & idioms, but Flemish farm town dialects can lose me pretty fast sometimes (especially on beer tours :-) ) even though native north-Hollands speakers seem to follow along pretty well. But if I ask a Flemish local to re-phrase it in ABN it's usually comprehensible. (Following a Flemish newscaster is much of a problem, but sometimes a local guy at a pub may as well be from Mars- or maybe even France :-) , and I've even heard native Dutch speakers ask "what is that in ABN?".)

    Hein claims facility with both Dutch & German and his name translated from Dutch to English literally would be "Henry Blood", but idiomatically could also mean "Grim Reaper's Blood" in some contexts (nice web handle, dude!), but there is also a German comic-strip character with that name, ( Captain Bluebear's sidekick: or ).

    I'm not exactly sure which cultural references he is trying to hit, but I thought I'd take a stab at it, despite not having used Dutch very much in 20+ years. You can probably find an online translator tool to copy and paste if you care for the details. I don't use web-translator tools often enough to know which ones work reasonably and which don't, but since I didn't use a lot of idioms even a dumb word-by-word translation would probably be clear (especially since I probably anglicized the grammar a bit, making it somewhat broken, odd, or awkward Dutch, being an American dialect speaker myself. But there are probably spelling errors too... ) My grasp of Dutch is still way better than of Japanese (a-so deska!? :-) ) or Hindi (kiya?). It was mostly a re-hash of parts of my prior postings in 'merican.

    OK, back to the usual energy-nerd stuff.

  20. user-1140531 | | #20

    Is Moisture in Walls Inevitable?
    I am a little confused by the premise of this topic as originally posed by Michael Roland a few weeks ago in the Q&A section.

    He asks if moisture will occur in double stud walls (without added foam board on the sheathing side). Under the terms of this discussion, it seems like moisture will indeed occur. How could it not? There are three different mechanisms for introducing moisture.

    1) Outward vapor diffusion.
    2) Outward vapor transfer due to air leaks.
    3) Inward vapor drive caused by cold sheathing.

    There seems to be a forgone conclusion that vapor retarders, vapor barriers, air barriers, or airtight drywall will be violated or have defects from the start. In addition, the inward vapor drive due to cold sheathing will be unstoppable since nothing is contemplated to eliminate it other than adding foam to the exterior to keep the sheathing warm.

    So, overall, double stud walls without foam board on the exterior will always have moisture accumulating in the wall cavity due to the three separate causes listed above.

    But even if you do incorporate exterior foam, might there not also be a forgone conclusion of violated or defective vapor and air barriers as it applies to the exterior foam board installation? Joints will open up; tape and caulks will degrade and fail over time. Some foam joints will be missed during the sealing process. Therefore all three of the above mechanisms for vapor transfer will occur despite the use of exterior foam.

    Also inevitable, will be defects in flashing that will let rain and snow enter the wall cavity. Therefore, it seems that all walls, and not just double stud walls, will have moisture entering the cavity under the terms we have been discussing. The only remedy is the intermittent drying potential coupled with moisture buffering to prevent water damage when drying is insufficient.

  21. Expert Member
    Dana Dorsett | | #21

    It's not at completely foregone conclusion. (Especially the flashing details, which are all-important to get right in the first place, but will will go 100+ years without failure if done right on day-1.) There's good-better-best, but no "perfect", and it's always better to strive for "best" on your construction details, but also design for resilience.

    Yes, moisture will get into the wall cavity from the interior during winter, but it also gets out on the exterior, if you let it, even with OSB sheathing. The vapor permeance of OSB goes up as it accumulates moisture, and with any sort of exterior gap between layers at all it dries toward the dry winter air. It's a matter of rates, and whether the interior moisture drives from air leaks and vapor diffusion exceeds the winter-drying rate to the exterior enough to cause a problem. In spring the wall can also dry toward the interior, if you don't make it super-vapor-tight, and wood that dries rapidly enough before it reaches mold-growth temperatures also survives without problems, even in less-than perfect assemblies.

    The higher-R you go, the colder the sheathing, which increases the vapor pressure across the interior air/vapor boundary layer, so more moisture gets transferred. In low-R structures the sheathing temps are warmer all winter, and rise more quickly in the shoulder seasons, leading to more rapid drying. With super-insulated R-values rainscreen gaps on the exterior the sheathing has pretty damned good drying capacity year round, but it still pays to have some buffering capacity to the fiber.

    Your point #3 seems turnd on it's head, and makes no sense. Cold sheathing does not generate a moisture drive toward the interior, but hot-wet sheathing can. With cold sheathing the vapor pressure is always going to be driving toward the cooler exterior, not toward the warm conditioned space. Even cold sheathing will dry toward the exterior (if you let it), if the entrained air in the cavity has a higher dew point than the exterior air (which is most of the time in winter. )

    During a hot humid summer with a dry air-conditioned interior, the moisture drives reverse. But in most cold climates the dew points of the exterior air rarely exceed 75F for even days, let alone weeks on end, and a dry 75F interior is a very comfortable conditioned space. Only when you have siding that stores water (like brick) that gets released at very high rates when warmed by the sun would it experience adsorption or condensation events on the interior side of the cavity. Interior side poly vapor retarders often have summertime condensation problems in brick clad structures, even in Canadian-coolth. But with even a half-inch of XPS between the brick and the studwall as a vapor retarder, it's usually won't happen. And if the interior vapor retarder is 2-3 perm paint rather than poly the drying rate to the air conditioned interior is usually enough to keep up.

    Using the term "vapor transfer due to air leaks" while mostly technically correct (yes, the water in the air IS usually in vapor form), it's typically referred to as "air transported moisture", since air could arguably contain micro-droplets of liquid water, and does, say in steamy shower locations.

    But your basic conclusion is correct:

    "Therefore, it seems that all walls, and not just double stud walls, will have moisture entering the cavity under the terms we have been discussing. The only remedy is the intermittent drying potential coupled with moisture buffering to prevent water damage when drying is insufficient."

    Yes, all walls leak some air, and diffuse some moisture via diffusion. You can limit air leaks & vapor diffusion, and enhance the drying capacity in climate-appropriate directions, but building with materials that buffer the moisture without damage makes it more resilient against deviations from "as designed" perfection. But that doesn't mean you should just shrug and relax on the air-sealing and flashing details- a creative idiot can and well break the most resilient of designs!

  22. user-1140531 | | #22

    My Item #3 in post #20.

    Thanks for your comments on this topic. From above, you said this regarding my point #3 (post #20) in my list of ways that moisture gets into walls:

    “Your point #3 seems turned on its head, and makes no sense. Cold sheathing does not generate a moisture drive toward the interior, but hot-wet sheathing can. With cold sheathing the vapor pressure is always going to be driving toward the cooler exterior, not toward the warm conditioned space. Even cold sheathing will dry toward the exterior (if you let it), if the entrained air in the cavity has a higher dew point than the exterior air (which is most of the time in winter. )”

    Well it does indeed seem turned on its head, as you say, but it is not my idea. This has clearly been explained here in several recent discussions and specific blogs such as this one:

    Here is the explanation that has been given here: When insulation began to become widely used, it was observed that insulation cavities were suddenly accumulating water and frost during the winter, and paint did not stick to the exterior as well as when the buildings were un-insulated.

    The problem was blamed directly on the insulation with the conclusion that insulation “drew moisture” to it. Failing to understand the cause of the problem, it was wrongly concluded that outward vapor diffusion was causing the wetting toward the exterior side of the wall cavities. This conclusion lead to the remedy of using a poly vapor barrier on the warm side to prevent outward diffusion. This lead to the present codification of the warm side poly vapor barrier.

    Then, building science made the contradictory claim that outward vapor diffusion was an insignificant vapor drive and was not the cause of the wall wetness associated with insulation. Instead, it was found that the real cause of the wetness was cold sheathing, siding, and exterior portions of studs. These features had been made colder by the addition of insulation, whereas, they had previously been warmer due to heat loss from a lack of insulation.

    The explanation is that the hygroscopic exterior wall materials will take on more moisture from the outdoors during the winter because of the seasonal drop in temperature. It is based on the fact that cold materials can hold more moisture than warm materials. The increased wetness of the exterior materials may dry to the exterior at times during the winter when the atmosphere loses relative humidity, but it also dries to the interior side as indicated by the presence of frost and water inside of the wall cavity. So it is indeed an inward vapor drive during the winter, and its cause has nothing to do with inward vapor drive of summer.

    This wintertime inward drive caused by chill down of exterior materials has been referred to here at GBA as the “cold OSB” problem, however, it is not just limited to OSB. The problem is named in association with OSB because OSB is relatively more hygroscopic, and therefore capable of producing the greatest manifestation of the problem. In fact, this moisture increase due to chilling is said to be one of the causes for OSB “turning to oatmeal.”

    Therefore, this certainly is a wintertime inward vapor drive toward the warm interior. Moreover, it is strong enough to create water and frost in the insulation cavities and completely destroy OSB sheathing. It is said to be enough of a wall wetting problem to require prolonged spring/summer drying in order to normalize the wall.

    This is the explanation for my item #3 in post #20 above.

  23. jklingel | | #23

    Well done
    Good blog and discussion. After a fast read thereof, I am even more convinced that Thorsten C's, and apparently Lucas', design of: stud (eventually insulated), plywood (air barrier and structural), lots of insulation, another stud, WRB, rainscreen, then siding, is quite a good build. Hang pictures where you want to, etc, and sleep well. On the note of WRB, I have seen some Siga MagCoat (?), and it is impressive stuff. Thorsten said he is now using something even better.

  24. user-757117 | | #24

    Response to John Klingel.
    Are you getting near to pulling the trigger on your own project?

  25. jklingel | | #25

    Lucas: I will be dense packing the shop walls (just 2x6) and blowing in the lid Saturday. I worked on the shop all winter; very cold, and very slow. Once insulated, I will install the heaters and then the garage door will go on. That will be a huge step forward. I can finish the rock in the warmth. I hope to have it all finished in a month or two. After breakup, I will start on the house. I am seeing Thorsten once again for a quick visit next week, mostly about having him design the solar part. I just can't learn about everything.

  26. user-757117 | | #26

    Response to John Klingel.
    I hear you on "very cold, and very slow".
    We've had some big winds and big windchills - good times.

    Re: your house...
    Have you decided on an envelope?

  27. user-1140531 | | #27

    Request for Clarification

    To the best of my knowledge, what I described in my item #3 and the following elaboration in my post #22 is a basic building science axiom being espoused here by GBA. It is not a theory that I came up with. The principle of this wintertime cold sheathing inward vapor drive is claimed to be the main reason for omitting a warm side poly vapor barrier. It has been described in great detail by Martin Holladay and William Rose in several recent discussions.

    So I would really like to hear your explanation for why you reject the theory when you say that it makes no sense in your post #21 above.

  28. Expert Member
    Dana Dorsett | | #28

    show me
    Show me anybody's description of wintertime moisture drives from the sheathing toward the interior. It defies the laws of thermodynamics for moisture to drive from lower humidity to higher. Yes, weather wetting from the siding can drive toward the sheathing, but the sheathing doesn't dry toward the interior in winter, only in spring/summer/fall.

  29. user-1140531 | | #29

    Wintertime Inward Vapor Drive

    I will see if I can dig up the links, but to explain for now; the moisture drive I am referring to is not from low RH to higher RH. It is also not from direct wetting of the siding from rain or snow. It arises from wetting of siding, sheathing, and the outer parts of studs due to their temperature falling.

    This relatively recent historical discovery of this behavior is said to have been based on the finding of water and frost inside of the wall cavity. I do not know to what extent this wetness spreads into the insulation. However, if the back of the sheathing is wet or frosty, it will be at a higher RH than the rest of the cavity. Therefore, it will dry in that direction, so I would call it inward vapor drive.

    This discovery said to have led to the following points:

    1) It explains the existence of cavity moisture that had been previously erroneously been attributed to outward diffusion.

    2) It eliminates the role of the warm-side poly vapor barrier in preventing outward diffusion; and calls for the omission of the that vapor barrier in order to double the drying potential by adding inward drying to the usual outward drying.

  30. jklingel | | #30

    yes. feel free to email me at jolinak at gci dot net. no sense in filling up a blog with our chatter. i would love to share details w/ anyone knowledgeable, so fire away! also, send the link to your blog; i have stolen ideas from thorsten shamelessly, and will steal anyone else's if they look better than what i have. i think it was your site where i saw rebar hoops to hold the 4 rebar in an edge beam. duh. simple. i did that in my shop and it was way easier than tying rebar sticks to everything. thanks. john

  31. lutro | | #31

    Dana, and wet equals cold quotes
    Dana, I am grateful for your comments and explanations on this thread, and many others. Both Bill Rose and Joe Lstiburek have published statements on GBA and elsewhere, saying “Cold, wet. Warm, dry,” in several similar variations. You will find a couple of examples, if you search Bill Rose and/or Joe Lstiburek on this blog entry: Very similar statements appear on several other GBA blogs, and are repeated by Martin Holladay. I've read their articles and explanations, and I consider the statement to be an over-simplification, which is right some of the time, but far from always. I don't agree with Ron Keagle's summary of his/their position, however, which he has labeled "point 3" in this thread.

    Dana, you have said at least twice in this thread, statements along the lines of "sheathing... has substantial capacity to dry into the dry winter air." As you point out, sheathing doesn't constantly dry to the outside air in the winter, regardless of other variables, but in many conditions, it does. While I agree with you, this is in opposition to the generalizations and blanket statements made by Bill, Joe, and Martin. I'm very glad that you have added your knowledge and perspective to this question. I think you are providing a more accurate and less confusing way of describing the issues, conditions, and processes involved with wall moisture variations in the winter.

    Thank you for taking the time to help educate me, and other readers, who are striving to understand these questions, and apply the knowledge appropriately.

  32. user-1140531 | | #32

    Point #3 in post #20

    I don’t doubt that cold siding and sheathing absorbing water can and does dry to the outside.

    According to my understanding, the cold siding and sheathing will absorb water just because it temperature drops. It can absorb this moisture from the exterior, the interior, or both simultaneously depending on the moisture available. And it can dry to the exterior, the interior, or both simultaneously depending on how dry those regions are relative to the wet siding and sheathing.

    When it dries to the interior, it is inward vapor drive as I stated in my item #3.

    According to Martin Holladay, the behavior of cold sheathing gaining moisture due to falling temperature was first discovered as water or frost inside of the insulation cavity on or near the back side of the sheathing. It seems obvious that that wetting would be wetter than the regions further inward toward the heated interior. If so, why would that outer moisture not dry inward?

    Furthermore, this cold sheathing wetting has been cited as a reason to omit a warm side poly vapor barrier in order to facilitate inward drying. Is inward drying not the same as inward vapor drive?

    Actually, when I read Dana’s comments in post #21, I get the impression that he is not addressing the principle of the sheathing gaining moisture due to its temperature dropping. It seems that he is only addressing outward vapor drive due to diffusion and air leaks, and the ability of the sheathing to adsorb that moisture and re-evaporate it to the outside. And in that sense, the “cold sheathing” simply represents a condensing or adsorbing surface for outward vapor drive should there be any due to air leaks from the interior.

    But with the issue of “cold sheathing” that we are discussing, even without air leaks, the siding and sheathing might gain moisture from the outside if the relative humidity goes up. In fact, Martin Holladay has said that the OSB sheathing can acquire so much moisture by this falling temperature adsorption that it can disintegrate the OSB. In acquiring this moisture, the OSB then becomes a source reservoir of moisture independent of the interior humidity of the heated space. As such, the siding and sheathing could easily acquire more moisture than what is contained in the insulation cavity and the interior heated space. Therefore this moisture will move inward from a source of high humidity to lower humidity, as Dana says must always be the case.

    I certainly welcome a less confusing way of describing the whole phenomena. I see vapor drives alternating in both directions at different times, depending on exterior humidity, interior humidity, air leaks, diffusion, and fluctuating outdoor temperatures. I am working on an elaborate graphic diagram to detail all of the possible phases of these vapor drives and their conditions with indications of flow and explanatory notes. When I get it ready, maybe I can post it here so we can analyze it as a fixed reference to the discussion of all these relative variables.

  33. mscannell | | #33

    Vapor barrier= inward vapor drive
    If you're placing a foam board outside of your exterior sheeting aren't you creating a vapor barrier?? How will moisture ever diffuse outward in a cold climate? You're not going to create a vented air gap between the sheeting and foam panel. The moisture is going to stay right there, unless you have a long dry summer, but not warm enough that you want to use your A/C. I'm thinking that might also be your so called "inward vapor drive" in disguise .

  34. GBA Editor
    Martin Holladay | | #34

    Response to Ron Keagle
    You are correct that cold wall sheathing collects moisture during the winter. However, there is no justification for describing this phenomenon as "inward vapor drive."

    The sheathing is damp, but there is no force (during cold weather) driving vapor inward.

    In the spring and summer, the temperature warms up. The sun shines on the exterior of the wall, and the temperature of the sheathing rises. If the house has air conditioning, the interior air is cool and dry. Under these conditions, the damp sheathing will dry inward. This inward drying doesn't occur during cold weather, however. It occurs during warm or hot weather.

  35. user-1140531 | | #35


    If this exterior source, “colder-wetter” moisture gain in the sheathing during winter is so much that it can present water or frost on the interior side and even disintegrate the sheathing, would it not dry to any region where the moisture concentration is lower?

    If the wall cavity is air sealed from the interior, and if outward diffusion is insignificant, wouldn’t that wall cavity be dryer than wet, saturated sheathing?

  36. GBA Editor
    Martin Holladay | | #36

    Response to Ron Keagle
    During the winter, the OSB sheathing will be at a much lower temperature than the drywall. The air and the fiberglass fibers will be at a temperature (or temperatures) that is higher than the OSB temperature but lower than the drywall temperature. The OSB will be the coldest surface in the wall assembly. That's why the moisture accumulates there.

    As you explained and as you know, the moisture in the (relatively) warm air between the studs is likely to accumulate in (or on) the (colder) OSB. The direction of the vapor flow is from the warmer air toward the colder OSB. So there won't be any flow of water vapor from the very cold OSB to the warmer air.

    During the summer, the flow is in the other direction.

  37. user-1140531 | | #37

    Response to Martin Holladay

    Yes, I understand exactly what you are saying. That is the classic outward vapor drive in winter, condensing when it reaches the dewpoint temperature, either in the insulation or at the back of the sheathing.

    However, in reading everything at GBA about the cold sheathing problem, I concluded that it was an entirely different source of moisture whereby the temperature drop of the sheathing made it thirsty, so it gathered moisture from any source, including the outdoors, and then dried that moisture back to wherever conditions were dryer on a periodic basis.

    As I recall, when discussing this with Bill Rose, we were talking about the outdoor wintertime humidity levels as determining the extent of the cold sheathing problem. And as I recall, you and Mr. Rose had explained that the source of the cold sheathing moisture was originally thought to be from outward diffusion or air leaks, and that led to the conclusion by the insulation industry that a poly vapor barrier was needed on the warm side to stop the outward vapor drive, thus eliminating the wetting of the sheathing.

    However, Mr. Rose said that this was an erroneous conclusion because the source of the observed moisture on the sheathing was not the outward vapor drive from the heated interior, but rather, it was the chilled down sheathing absorbing moisture mostly from the outdoors (because there would not be any coming from the interior, as the insulation industry had erroneously concluded).

    Referencing your blog here:

    In that blog, you say this about Mr. Rose’s description of the discovery that discredited the insulation industry’s flawed conclusion as to the source of sheathing wetness:

    “Rose wrote that Teesdale, Rogers, and Rowley “created a version of hygrothermal building science for the United States that focused on moisture conditions in exterior materials during cold weather. The version they created was partial, and it was biased: It highlighted the importance of vapor transport, while it obscured the importance of temperature impact.” In other words, Teesdale, Rogers, and Rowley promoted the idea that the siding was getting damp because moisture was traveling through the wall assembly by diffusion from the interior. While this diffusion does occur, the amount of moisture transported via diffusion isn't that significant; the governing factor determining the moisture content of the siding is its temperature, not the rate of diffusion through the wall.

    Rose continued, “They produced prescriptive recommendations that later became code requirements, and these prescriptions embodied the incomplete and biased nature of their analysis. They supported their argument with a flawed and misleading analogy.”

    Now, however, from your above explanation in your post #36 here, it sounds like the insulation industry was correct in that the sheathing wetting was coming from outward vapor drive which would be stopped by a vapor barrier.

    According to your explanation here in post #36, IF you had zero air leaks, and IF there was no outward diffusion; then there would be no “cold sheathing problem.” There would be no wall cavity wetting regardless of how cold the sheathing became.

    Is that correct?

  38. GBA Editor
    Martin Holladay | | #38

    Response to Ron Keagle
    You quote me correctly: "Teesdale, Rogers, and Rowley promoted the idea that the siding was getting damp because moisture was traveling through the wall assembly by diffusion from the interior. While this diffusion does occur, the amount of moisture transported via diffusion isn't that significant."

    The cold sheathing takes on moisture from the air between the studs. It also takes on moisture from the exterior or the backside of the siding, especially after rain.

    You wrote, "It sounds like the insulation industry was correct in that the sheathing wetting was coming from outward vapor drive which would be stopped by a vapor barrier." The insulation industry was partially correct; some of the moisture originates from the inside of the house. However, the moisture transport mechanism is not diffusion; it is air leakage. Most of the interior moisture that reaches the cold sheathing is carried by exfiltrating air. Diffusion has very little to do with this problem.

    As you well know, a vapor barrier does nothing to address air leakage.

    Finally, you wrote, "IF you had zero air leaks, and IF there was no outward diffusion; then there would be no 'cold sheathing problem.' ” I might add: if pigs could fly, and unicorns laid golden eggs, I'd be a rich man. Or something like that.

    If you manage (somehow) to build your wall like a submarine wall, with no leaks, you'd probably be OK. But as you know, the OSB would still get damp every winter -- so your proposed submarine-style wall better have a way to allow the OSB to dry to the exterior in springtime.

  39. user-1140531 | | #39

    Reply to Martin Holladay

    From earlier descriptions, I had understood this cold sheathing problem to be independent from the performance of the air and/or vapor barrier. That is why I stipulated no air leaks and no diffusion as a premise to my question seeking to clarify cause for the cold sheathing wetting. As I understand it now, the cold sheathing issue is simply that the sheathing is the first condensing surface encountered by outward wintertime vapor drive.

    But I do not see how this is fundamentally different than what Teesdale, Rogers, and Rowley postulated. You say that they highlighted the importance of vapor transport, while it obscuring the importance of temperature impact. What evidence is there that they obscured the importance of temperature impact? Their explanation had to include a condensing temperature in order to make their observation about vapor transport relevant. The symptom they were diagnosing was wetness. Without a condensing temperature, the vapor transport would not result in wetness.

    So I do not see the cold sheathing problem as something that evaded Teesdale, Rogers, and Rowley; or that they reached the wrong conclusion about the remedy. They proposed a solution consisting of stopping the outward vapor transport. If that were successfully accomplished, then it would prevent the cold sheathing wetting, possible disintegration, and paint failure. If you assume that air leaks and diffusion are inevitable, then wintertime sheathing wetting problem will inevitably occur due to outward vapor flow from the heated interior, resulting in possible sheathing disintegration and paint failure.

  40. GBA Editor
    Martin Holladay | | #40

    Response to Ron Keagle
    We're going in circles here. You wrote, "Teesdale, Rogers, and Rowley ... proposed a solution consisting of stopping the outward vapor transport. If that were successfully accomplished, then it would prevent the cold sheathing wetting, possible disintegration, and paint failure."

    No. Stopping vapor diffusion with a vapor barrier is different from stopping air leaks with an air barrier. Kraft facing on fiberglass batts is a vapor retarder (and its existence can be traced back to the work of Teesdale, Rogers, and Rowley) -- but it isn't an air barrier.

  41. user-1140531 | | #41

    Response to Martin Holladay

    I appreciate your information and explanations, and have no desire to go in circles, as you say.

    In your blog writings, you said that there are two basic points which the historical explanation of moisture in walls offered by Teesdale, Rogers, and Rowley, got wrong. The two basic points are:

    1) They highlighted the importance of vapor transport while they obscured the importance of temperature impact.

    2) They focused on diffusion rather than air leaks as the motivation for vapor transport.

    I see these two points as being independent and unrelated.

    I understand the distinction between air leaks and diffusion as a vapor transport mechanism, but did not see them as being anything new. However, in discussing the importance of temperature impact and cold sheathing effect in one or more blogs, plus Q&A threads, I was led me to believe that it was an entirely new and different explanation for the origin of wall moisture. I interpreted the new explanation to cite sheathing and other cold side hygroscopic materials as the localized source of moisture; as opposed to the historical explanation whereby those materials were merely the condensing surface recipient of outwardly transferred moisture.

    This understanding arose on my part because in our initial discussions about this, you explained that “temperature impact” referred to the tendency of hygroscopic materials to gain moisture as their temperature drops, i.e. “colder-wetter.” We discussed the outdoors as being the source of this moisture and it was not from rain wetting, but rather, it was simply due to the falling temperature of the outer materials causing them to absorb moisture from the outside air. This is why I concluded that the old Teesdale, Rogers, and Rowley explanation of the heated interior being the origin of moisture by outward vapor transport had been replaced with the new explanation of the cold sheathing /siding being the origin of moisture.

    As I now understand it, you are saying that old explanation by Teesdale, Rogers, and Rowley was primarily (if not entirely) wrong because it cited diffusion as the force for the outward vapor transport, whereas the new explanation cites vapor being transported by air leaks and discounts diffusion.

    But when you say that the old explanation obscured the importance of temperature impact, I fail to see that. No matter whether the outward vapor transport is by diffusion or by air leaks, the temperature impact determines whether the vapor condenses. Therefore the temperature impact would have been obvious to Teesdale, Rogers, and Rowley as a basic component of the problem of water in the wall.

    Therefore, I conclude that if Teesdale, Rogers, and Rowley were wrong, it was only in that they cited diffusion rather than air leaks as being the motivating force of vapor transport. The temperature impact is the same either way.

  42. GBA Editor
    Martin Holladay | | #42

    Response to Ron Keagle
    Your account of your evolving understanding is somewhat interesting, I suppose. I'm glad to hear that your understanding is evolving.

    Your conclusions about whether scientists' current understanding of the moisture performance of walls is dramatically different or only slightly different from the understanding that was common in the 1940s is a matter of perspective rather than fact.

    Your conclusion that "the temperature impact would have been obvious to Teesdale, Rogers, and Rowley as a basic component of the problem of water in the wall" is speculative. I'll leave it at that.

  43. user-1140531 | | #43

    Reply to Martin Holladay

    I understand your point that Teesdale, Rogers, and Rowley highlighted vapor transport, but what do you mean when you say that they obscured the importance of temperature impact?

    One minor point: I need to clarify my comment that I see vapor transport and temperature impact as being independent and unrelated. Of course they are directly related in terms of creating the wall moisture because that effect requires both vapor transport and a condensing temperature. What I meant was that they are two separate and independent issues. You could have one without the other, or have neither, or have both.

  44. GBA Editor
    Martin Holladay | | #44

    Response to Ron Keagle
    As I explained in my blog, Do I Need a Vapor Retarder?, it was Bill Rose who wrote the words, "The version they [Teesdale, Rogers, and Rowley] created was partial, and it was biased: It highlighted the importance of vapor transport, while it obscured the importance of temperature impact.” Bill Rose is the person that you should direct your questions to.

    To the best of my knowledge, Bill Rose reached his conclusion by researching original documents published i the 1930s and 1940s.

  45. user-1140531 | | #45

    Thanks Martin. I will check
    Thanks Martin. I will check with Bill Rose.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |