Double-Stud Wall
A Habitat for Humanity house in Wheat Ridge, Colorado, used double-stud walls and raised-heel roof trusses to reduce heat loss through the thermal envelope. Although the strategy works, a potential complication involves local code requirements for air and vapor barriers.Image Credit: National Renewable Energy Laboratory
More Q&A Spotlight
Erik Olofsson is planning a small house in the Rocky Mountains of British Columbia. Ideally, he’d like to get the walls close to R-40. The question is how.
“Seeing that the received opinion around GBA is the tandem of polyethylene sheeting and exterior rigid foam is not ideal, what do the builders on this site recommend?” he asks in a post at the GBA Q&A forum. “Larsen trusses seem fairly labor-intensive and rigid foam is expensive … Is a double-stud wall the answer?”
A complication is a local building code that apparently calls for a polyethylene vapor barrier on the warm side of the insulation. Although once a common building technique, it’s no longer universally accepted by building scientists as the best practice in all climates. Many builders have abandoned the use of interior polyethylene, even as some building inspectors continue to insist on it.
Olofsson’s quest for high performance at a reasonable cost, while solving the riddle of air and vapor barriers, is the topic of this month’s Q&A Spotlight.
Double-stud walls a good option
Double-stud walls are designed to provide lots of exterior wall volume for insulation while sharply reducing thermal bridging. John Klingel and Albert Rooks are among those who think that building double-stud walls is a good approach.
“I built a double-stud in 1980 and have never regretted it,” Klingel writes. “New house will be the same, but thicker, and with dense-packed cellulose instead of fiberglass.”
“A double stud with a plywood exterior and interior poly and ADA [the Airtight Drywall Approach] will work,” Rooks says.
GBA senior editor Martin Holladay doesn’t push the double-stud option, but he does point Olofsson toward a number of GBA…
Weekly Newsletter
Get building science and energy efficiency advice, plus special offers, in your inbox.
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 TrialAlready a member? Log in
66 Comments
Peter nailed it (of course)
Yes, I think we all know that there are plenty of writers and enforcers of the building codes out there - as well as plenty of builders, trade contractors, HERS raters, & others in the home building industry - who don't understand the purposes of the various control layers. That's especially true for air barriers and vapor retarders. There are also some folks who do understand these things and know how to apply them, and their number is growing.
Regarding the control of moisture, Peter nailed it at the end of the article with his hierarchy of the forms and paths of moisture control. Last month when I spoke to the Building Science Discussion Group in Portland, Maine, the discussion turned to the topic of condensation and moisture damage in building assemblies. As I recall, not a single one of the builders, contractors, designers, or anyone else in the room could think of a single case where it was diffusion rather than air leakage that caused moisture damage. And they're in IECC climate zone 6!
We spend a lot of time talking about vapor retarders, and the discussion seems to have evolved from "Make sure you put in a vapor barrier" to "Be careful about putting in materials with permeance ratings that are too low." That's a good thing. Just ask the folks in North Carolina who don't have to put plastic in their walls anymore - and repair the resulting damage.
Diffusion vs. Infiltration
For those who may not have seen these yet, here are two diagrams showing how much more water vapor gets into places where it shouldn't by infiltration (second image) than by diffusion (first image). These are from the Builder Guides written by Joe Lstiburek and based on research done over a whole heating season.
Barriers
I have been following this revised thinking concerning the function and need for a warm side vapor retarder to prevent vapor transfer by diffusion. However, I am not quite sure what conclusions to draw for practical application to a new house project in cold climates such as northern U.S.
1) If codes commonly requires a vapor diffusion barrier such as polyethylene film, is it worth it to fight this by omitting and diffusion barrier and only providing a air exfiltration barrier?
2) Is it easier to include a film diffusion barrier than to omit it and provide an exfiltration barrier by meticulously sealing the drywall or other finished surface?
3) Can I believe the science that suggests that drywall is sufficient to prevent condensation problems from diffusion?
Intelligent walls
Double stud walls are a cost efficient way to build a super insulated house. To ensure a good airtight seal an interior airbarrier is ideal as it can actually be checked an verified by a blowerdoor test. Hence making an uninterrupted airtight layer out of plywood, or even better an intelligent vapor retarder like Intello Plus - which are taped together can assure that your insulation stays free of condensate in the winter in your walls, floor and roof.
The reinforced Intello membrane (disclosure I import this with http://www.foursevenfive.com) is like membrain - is vapor closed in winter and vapor in summer, however it has as added benefits that it:
- is extremely strong
- can be used as dense pack cellulose mesh
- part of complete airsealing solution line from Pro Clima that includes tapes, gaskets, adhesives and high quality exterior membranes
As shown in this 3d section, a double wall does benefit from a service cavity, which minimizes service penetrations thought the airtight layer and thus avoids vapor issues in the wall.
Diffusion v. Infiltration
To set the matter in context, the 1 square inch hole in Dr Joe's second example is the equivalent of a gap right round the edge of an 8ft x 4ft board which is 3.5 mil wide. Which is, quite literally, a hair's breadth.
This discussion seems to be edging its way, ever so slowly, to the wall construction described as "A New USA Wall" by Greg La Vardera. Disclosure - I have no commercial ties with Greg La Vardera.
Response to Ron Keagle
Ron,
Q. "If codes commonly requires a vapor diffusion barrier such as polyethylene film, is it worth it to fight this by omitting and diffusion barrier and only providing a air exfiltration barrier?"
A. As far as I know, no U.S. building code requires the use of interior polyethylene. In some climates, codes require an interior vapor retarder with a permeance of 1 perm or less. This can be satisfied by the use of kraft-paper facing or (in some cases) vapor retarder paint.
Requirements for air barriers are a separate issue from requirements for vapor-diffusion retarders, so it's important not to confuse the issues. All walls need at least one, and in some cases two, air barriers.
Q. "Is it easier to include a film diffusion barrier than to omit it and provide an exfiltration barrier by meticulously sealing the drywall or other finished surface?"
A. Again, you are confusing two issues (code requirements for vapor retarders and the need for an air barrier). Your question is a little like, "What is easier -- installing a window or installing roof flashing?" These are two different issues.
Q. "Can I believe the science that suggests that drywall is sufficient to prevent condensation problems from diffusion?"
A. That science doesn't exist, since drywall is vapor-open. Unpainted drywall does little to limit vapor diffusion. However, your question raises another issue, namely: Do homes generally have any moisture problems that can be traced to diffusion from the interior outwards during the winter? The answer is: No. So don't worry so much about outward diffusion.
Combination air/vapor diffusion barrier
Martin,
Thanks for that information. I am very interested in the issues surrounding vapor barriers/retarders, and particularly this new turnabout in thinking about the need to combat vapor diffusion.
Regarding my question:
"Is it easier to include a film diffusion barrier than to omit it and provide an exfiltration barrier by meticulously sealing the drywall or other finished surface?"
What I was getting at was this: Rather than omitting a film vapor barrier under the premise that it is not needed; and then providing the needed air barrier by making sure the drywall is sealed; I would rather include the film vapor barrier to function both as a vapor barrier and as an air barrier. That way, in addition to having an air barrier, I will also have a barrier to vapor diffusion just in case there is any that is worth preventing. By taking this approach, I will not have to air seal the drywall because the film barrier will perform the function of air barrier.
I realize that the mostly impermeable film vapor/air barrier will not let the wall dry to the inside, after being wetted from solar vapor drive from the outside. But I do not want conditions arranged for wet walls to dry to the inside. Moreover, I do not want walls getting wet in the first place, so I would avoid this whole issue by preventing the walls from getting wet from outside sources.
To accomplish that objective, I would avoid the use of reservoir cladding and limit rain exposure by a large roof overhang. I am not fully certain of the exterior detailing, but the point would be to use a house wrap that would prevent the intrusion of rain in conjunction with an air space. So the wall interior would not get wet from rain, or from outward condensing vapor.
However, the wall system would be subject to the same humidity level as the outdoors, and I can see how cooling the building interior with air conditioning might chill the film vapor barrier enough to condense exterior vapor on the exterior facing surface of the vapor barrier. However, in the northern U.S. climate, I would not expect the outside dewpoint to reach the interior air-conditioned temperature very often, if ever.
If any vapor does condense in the wall cavity, it would be cyclic, and I would expect it to dry to the outside in the opposite cycle.
Response to Ron Keagle
Ron,
Plenty of people use what you call an "interior film" as an air barrier. However, I would strongly urge you to avoid the use of polyethylene for this purpose, especially if there is any chance that the building will ever be air conditioned.
If you want to use a membrane that comes in a roll as an interior air barrier, you should use MemBrain, Intello Plus, or Siga Majpell.
Strapped 2x6 wall
After doing many variations on the double wall theme, for my next project I am contemplating a 2x6 wall with 2x4 horizontal straps every 2 feet on the exterior. 1.5" taped foam and housewrap sheathe these straps; 3/8 lath and vertical B&B siding on exterior, drywall on interior, with moisture retarding paint. Cavity insulation is densepack cellulose @7". Rack bracing is let in. R value is over 30, and it is a pretty simple and inexpensive assembly. It also has a redundant air barrier just in case the detailing is less than perfect.
air barriers
I'm wondering about the placement of the air barrier. On a double stud wall does it matter whether it is on the inside using the drywall or on the exterior using taped plywood? Can both locations be equally effective?
Response to Bryan Shephard
Bryan,
There is no consensus on the answer to your question; opinions vary.
Here is an article that discusses the relevant issues: One Air Barrier or Two?
Response to Bryan Shephard
Bryan,
In my (not-so-empirical) opinion, if there is only going to be one air barrier, it should be on the warm-in-winter side of the envelope in heating climates.
And possibly on the exterior side of the envelope in cooling dominated climates.
My reasoning is that there are often interstitial "3D airflow networks" within an envelope...
A good air barrier should be located so as to prevent the air with the higher moisture content from entering the interstitial space (in addition to the more straight forward job of preventing air exchange between the interior and exterior environments).
For example, in a heating climate in winter:
cold, relatively dry air that contacts a warm air barrier via the "3D airflow network" is lower risk than;
warm, relatively moist air that contacts a cold air barrier via a similar "3D network".
I think this is what has prompted some double wall systems to be refered to as "turds" despite possibly having good blower door test scores...
Response to Bryan Shephard
I think that Lucas has it right.
I like to think of it this way: An air barrier will keep the wind from blowing in and out of your envelope. For that simple barrier, placing it at the exterior sheathing or the interior will do the trick. However, life is full of improvements... Placing said barrier on the warm side of the insulation in a heating climate will keep the water vapor that travels with the interior air -inside the habitable space and out of the wall.
On a thick wall in a cold climate, the exterior sheathing is cold since it is well removed from the warm interior. By keeping warm air out of the wall, it is argued that you are reducing chances of condensation on the interior of the exterior sheathing. Hence, an upgraded air barrier ( if you will).
Location, Location, Location ...Location matters
Opinions about Best Single Air Barrier Location Vary.
I agree with Lucas & Albert
IF...IF...IF there must only be ONE Air Barrier in a Heating Climate...
The Best location is on the Interior side of the Cavity (3-D Network)
Attached is my version of a John Straube Illustration ...
Showing that even a perfect Exterior Air Barrier can be a "Fail" IF there are any voids in the cavity.
I think a Better solution for walls (and Vaulted Ceilings) is TWO Air Barriers
One Interior + One Exterior
Water Vapor Control Fundamentals:
This is my analysis of the entire water vapor issue for buildings:
Water vapor can enter the wall from either side, either by diffusion or as a component of moving air. If that vapor encounters a cool enough surface inside of the wall, it can condense into water. This problem can be prevented by preventing vapor from entering the wall; or by making sure that entering vapor cannot contact cool enough surfaces to condense.
To prevent water from entering a wall by diffusion, a relatively non-permeable barrier is needed. Such a barrier will also prevent water from entering a wall as a component of air. However if the objective is only to prevent water from entering as a component of air, a less impermeable barrier will suffice.
If vapor does condense into water inside of a wall, it may be removed by evaporation if vapor is permitted to escape to either the interior or exterior, either by diffusion or as a component of moving air.
Barriers that prevent the passage of vapor into a wall by diffusion will not permit drying evaporation of the wall interior by diffusion.
Barriers that prevent the passage of vapor into a wall as a component of air will not permit drying evaporation of the wall interior by moving vapor as a component of air; however, such barriers will permit the drying evaporation of the wall interior by diffusion.
Response to Ron Keagle
Ron,
Sorry -- you're still not quite there.
You wrote, "To prevent water from entering a wall by diffusion, a relatively non-permeable barrier is needed. Such a barrier will also prevent water from entering a wall as a component of air."
There are many perfectly acceptable vapor retarders that are lousy air barriers. So it's simply untrue to write that a barrier chosen to limit vapor diffusion will "prevent water from entering a wall as a component of air."
If you want to limit vapor diffusion, there are paints that do a perfectly acceptable job. The kraft facing on fiberglass batts also works well as a vapor retarder. However, neither product will "prevent water from entering a wall as a component of air." For that, you need an air barrier.
Martin,
I do not follow your
Martin,
I do not follow your comment. How can something be a lousy air barrier while being a perfectly acceptable vapor retarder? A lousy air barrier would permit vapor transmission as a component of air. And it would also offer no resistance to vapor diffusion. Diffusion could simply occur directly through air, even if air was not moving. So how can a lousy air barrier possibly be a “perfectly acceptable vapor retarder”?
Response to Ron Keagle
Ron,
Q. "How can something be a lousy air barrier while being a perfectly acceptable vapor retarder?"
A. It's easy. Vapor diffusion is a totally separate phenomenon from air leakage. When exfiltrating air carries interior moisture into a cold wall cavity, it can lead to moisture problems. But this moisture transport mechanism has nothing to do with vapor diffusion.
Q. " A lousy air barrier would permit vapor transmission as a component of air."
A. That's true.
Q. "And it would also offer no resistance to vapor diffusion."
A. On the contrary: bad air barriers can be effective vapor diffusion retarders. A wall finished with painted gypsum wallboard might leak a lot of air, especially if electrical boxes aren't air sealed. However, if there are several coats of oil-based paint on the drywall, you'll get almost no vapor diffusion.
Q. "Diffusion could simply occur directly through air, even if air was not moving."
A. I'm not sure what you mean. Vapor diffusion is a different phenomenon from air leakage. For example, vapor can diffuse through unpainted drywall, even if the drywall is airtight.
Q. "So how can a lousy air barrier possibly be a 'perfectly acceptable vapor retarder'?"
A. It's easy -- all it has to do is stop diffusion.
I think you are hung up because you think that vapor diffusion matters. It actually doesn't matter very much -- especially vapor diffusion outward during the winter. What matters is air leakage.
Martin,
I see what you mean
Martin,
I see what you mean by the possibility of a lousy air barrier being a good vapor retarder. By vapor retarder, you are referring to retarding vapor that is moving only by diffusion. So you could have a near perfect vapor retarder/barrier, even if it were full of holes because transmission through the holes would not be diffusion. According to that definition, a membrane could be nearly perfect as a vapor barrier, but nearly worthless for stopping the flow of vapor.
Response to Ron Keagle
Ron,
It's not my definition; it's the definition established by building codes. For decades, U.S. building codes defined a vapor retarder as "a material having a permeance rating of 1.0 or less when tested in accordance with ASTM E 96." There is no mention whatsoever of whether or not the material is a good air barrier. (More recent building codes have complicated the definition by breaking vapor retarders into three classes -- Class I, Class II, and Class III -- but they are all based on permeance testing, not air leakage testing.)
Plenty of materials can meet the definition of a vapor retarder without being effective air barriers.
When you write, "According to that definition, a membrane could be nearly perfect as a vapor barrier, but nearly worthless for stopping the flow of vapor," you are not quite correct, but you are getting closer. Actually, any membrane that is nearly perfect as a vapor barrier is, by definition, nearly perfect at stopping the movement of vapor by diffusion.
I would express your idea this way: A membrane can be a nearly perfect vapor barrier but nearly worthless at stopping the flow of air.
It's also possible to say: A membrane can be a nearly perfect vapor barrier but nearly worthless at stopping the transmission of water that accompanies air moving through defects in the membrane.
Martin,
Setting aside the
Martin,
Setting aside the issue of a good vapor barrier that has holes in it, when we speak only of continuous material, what materials would be a good vapor retarder/barrier while being a poor air barrier? I can’t picture a material that would be impermeable enough to stop the diffusion of vapor, and yet be porous enough to permit airflow.
Response to Ron Keagle
Ron,
Materials used to build buildings always have seams, so your theoretical question is interesting but possibly irrelevant.
A wall of metal shingles, installed like cedar shingles, would probably be vapor-impermeable but air-leaky. So would a section of wall with lapped kraft paper.
Many painted materials could pass the permeance test without passing an air leakage test.
In any case, building codes require vapor barrier materials to pass a permeance test. Once installed, however, all of these materials have seams and laps (as well as penetrations), and all of these materials, even those that pass the permeance testing with flying colors, are air-leaky as usually installed.
Here's a sample job-site dialog to help you understand what I'm saying:
Martin: What are you using as your vapor retarder?
Ron: I'm using vapor-retarder paint.
Martin: That makes sense. What are you using as your air barrier?
Ron: I'm using vapor-retarder paint, because as far as I know, any material that passes the vapor retarder test should be a good air barrier.
Martin: But...
My vote is for closed cell spray foam
Without starting the "R" value debate, I say closed cell SPF is the best choice. 2 inches is a vapor barrier, while spraying and quoting spray foam you base it on an average, so I always recommend 3 inches. The perfect situation and budget would be all closed cell, but we all know in the real world thats not always in the budget. 3 inches of closed cell, than cover with open cell for R value.
Personally not a fan of flash and batt, and mineral wool is pretty exspensive as well, and if the cavities are not perfectly sealed we all know air will flow through it, and where there us air there will be moisture.
Spray foam in my eyes is the solution.
Martin,
You mentioned that
Martin,
You mentioned that U.S. building codes defined a vapor retarder as "a material having a permeance rating of 1.0 or less when tested in accordance with ASTM E 96." And you went on to say that “Plenty of materials can meet the definition of a vapor retarder without being effective air barriers.”
When I asked what materials would have those characteristics, you say that materials having those characteristics would be materials meeting the definition of vapor retarders, but being installed with open seams and other defects. I don’t see that as being a material that qualifies as a vapor barrier but is not a good air barrier. The material is one thing, and the quality of installation is another thing.
So, I am standing by my assertion that materials meeting the definition of vapor retarder/barrier would also be effective as an air barrier. I don’t see how possiblities of leaks or other installation defects changes that fundamental truth.
Can’t a vapor retarder also be an air barrier, so one membrane performs both functions? That was my original point in this discussion.
Response to Ron Keagle
Ron.
You wrote, "I am standing by my assertion that materials meeting the definition of vapor retarder/barrier would also be effective as an air barrier."
OK -- then go ahead and use paint as an air barrier. I'm going to choose different materials for my air barrier.
Martin,
Doesn't vapor
Martin,
Doesn't vapor retarder paint need to be used in conjunction with a substrate such as sheetrock? Wouldn't the sheetrock be a suitable air barrier with or without the paint?
Actually, what I am considering for use is DURA-SCRIM 6WW 6 mil four-layer reinforced extrusion laminate with a perm rating of .07. I would install it with no leaks, and expect it to function as a vapor retarder and air barrier combined.
Response to Ron Keagle
Ron,
Q. "Doesn't vapor retarder paint need to be used in conjunction with a substrate such as sheetrock?"
A. Yes. But the drywall can't meet the permeance requirement of the vapor retarder provisions of the code. The paint is a vapor retarder -- not the drywall.
Q. "Wouldn't the sheetrock be a suitable air barrier with or without the paint?"
A. Yes, if the drywall is installed with the necessary gaskets and with airtight electrical boxes, following the details required by the Airtight Drywall Approach. If that's done, the drywall and gaskets are your air barrier, and the paint is your vapor retarder.
Q. "What I am considering for use is DURA-SCRIM 6WW 6 mil four-layer reinforced extrusion laminate with a perm rating of .07."
A. The product you are describing is a type of polyethylene. It is not a "smart" retarder. Your approach has been successful in many Canadian buildings for 30 years, but interior poly can still lead to disasters. Interior poly has been blamed for many of the "leaky condo" problems in Vancouver, since it prevented wet walls from drying inward. You can also get into trouble using interior poly in Ontario or Quebec if the house is air conditioned.
In short, good luck.
Location, Location, Location
John Brooks, we wholeheartedly agree with your analysis - and John Straube's diagram is very informative.
To reiterate and expand a bit on the idea: First priority should be an air barrier at the interior, preventing possibility of bulk vapor carried by air leakage from inside. Second, particularly if one is using cellulose or mineralwool or fiberglass or other fibrous insulation, is an air-barrier on the outside, to optimize the insulation - like a windbreaker over a sweater.
I would add to this that airtight drywall is an unreliable air barrier and there should be a dedicated air barrier (either rigid or fabric) protected by a service cavity, behind the finish drywall. We have an informative blog post on the use of service cavities here: http://bit.ly/LDMUR4
But also protect the exterior air barrier with a vented rainscreen/roof, as we describe the benefits here: http://bit.ly/MvD4lM
Finally, as my colleague Floris noted above, if you are doing an interior air barrier in a cold climate it may make sense (particularly at roofs) to use an intelligent vapor retarding air-tight membrane that is essentially vapor closed in the winter and vapor open in the summer, thereby minimizing the wetting and maximizing the drying potential.
Reply to Martin Holladay
Thanks Martin. I understand your points about the vapor barrier, and have been considering all of them as my project develops. For the exterior, I would use a vapor permeable air and water barrier combined with flashing that would prevent any water intrusion.
I would try hard to avoid the issue of inward vapor drive by using non-reservoir siding and a large roof overhang that would substantially reduce rain contacting the siding, and even reduce the storm saturation of the ground right next to the house. My objective would be to limit the in-wall humidity to the level of the average outdoor ambient.
The poly vapor barrier on the warm side would not permit drying the wall cavity to the interior, but I don’t anticipate much need for drying. The only need for wall cavity drying that I can see is the possibility of condensation on the exterior side of the vapor barrier if the outdoor dewpoint exceeds the vapor barrier temperature during days of the highest humidity and maximum air conditioning. But I don’t anticipate that overlap occurring very often, and when it does occur, it might be with a dewpoint of 78 degrees and an interior temperature of 76 degrees, for example.
I understand the new thinking that vapor diffusion is not a problem in buildings. Nevertheless, I cannot verify that for my application, so I would still want to protect against diffusion condensation as insurance. That is my reason for using the polyethylene vapor barrier on the warm side. I would expect the vapor barrier to perform as an air barrier as well.
Any condensation that does happen to occur within the wall cavity during air conditioning during high humidity would dry to the exterior. I am aware of smart vapor barriers that would permit this potential moisture to dry to the interior while preventing outward vapor flow during the heating season. That performance would ideal because the inward drying would be into the air-conditioned space, which will be very receptive because the air conditioning would be removing water vapor from the interior. On the contrary, drying the in-wall moisture to the exterior will have to wait until after the air conditioning/high humidity phase ends. But again, I do not see much need for drying the wall cavity because condensed moisture will be relatively unlikely to be in the cavity.
The only problem I see with these smart vapor retarders is the possibility that they will not do what is claimed for them. And there is no practical way to verify their performance. Basically, you have a membrane that is acting as an automatically operating valve. I simply would not take that on blind faith that this type of barrier would continue to cycle functions effectively as it was intended and claimed. Once installed, there would be no reasonable possibility of correcting it if it failed, and it would also be very difficult to detect a failure.
No blind faith required.
Ron,
The science behind the intelligent vapor barriers is fairly straight forward, well established and tested. I encourage everyone to take a closer look at a Pro Clima study, here (pdf download): http://download.proclima.com/en/int/study.pdf
Note: While we are in the process of translating all the units in this document to "American", (Perm = 3.28/Sd value) the physics remain unchanged.
Keeping It Away From A Cold Surface
I have the understanding that the science allows (even if the building code doesn't always) one to forgo a warm-side vapor barrier in a cold climate if there is a certain amount of insulation exterior to the sheathing. The idea being that the exterior insulation raises the temperature of the sheathing enough to keep any vapor that does get into the wall from the interior from feeling the need to condense on it. The number of inches of exterior XPS/EPS/Mineral Wool/etc that you would need for this depends on the coldness of the climate and on the thickness of the wall in question--but the requirements go up if you have a darn thick wall like a double stud wall. Am I understanding this correctly? Is avoiding condensation on colder exterior surfaces the only reason for vapor control in a heating climate?
If I am understanding that correctly, then I wonder if one could substitute a flash-and-fill style insulation system into a double stud wall in place of adding many inches of exterior insulation. For instance, if design conditions say you need 2" of XPS on the exterior of the wall to negate the need for an interior vapor barrier, but you instead install a certain number of inches of spray foam on the inside of the wall and fill the rest of the cavity with something fibrous, are you not accomplishing the same purpose with that interior spray foam of warming up the sheathing? Further, would it create a bad "vapor sandwich" problem for the sheathing to have some exterior insulation AND a flash-and-fill system in the cavity?
Smart vapor retarders work
Ron, if you doubt reports supporting the performance of smart vapor retarders, go on Certainteed's web site and order a sample kit of their MemBrain product. At least when I got mine several years ago, it came with two plastic pouches, one of MemBrain and the other regular old polyethylene, and two thin wood strips. The instructions said to soak the wood strips to saturate them, then seal them inside the two pouches and leave them around for a number of days. Initially you see beads of condensation on the inside of each, but less so in the case of MemBrain after a few days. After perhaps a week, the MemBrain sample was totally dry inside, while the polyethylene pouch allowed no drying of its wood piece at all. I was happy, and I used it under the drywall in my own house, a double frame, cellulose-filled wall structure. I detailed it properly, to tie into the floor system and everything else so as to form a continuous air barrier as well as vapor retarder.
Response to Leigha Dickens
Leigha,
Q. "The number of inches of exterior XPS/EPS/Mineral Wool/etc that you would need for this depends on the coldness of the climate and on the thickness of the wall in question--but the requirements go up if you have a darn thick wall like a double stud wall. Am I understanding this correctly?"
A. Yes. More information here: Calculating the Minimum Thickness of Rigid Foam Sheathing.
Q. " Is avoiding condensation on colder exterior surfaces the only reason for vapor control in a heating climate?"
A. I don't think there is much reason for vapor diffusion control on the interior of a house. However, houses with reservoir sidings that have air conditioning need vapor diffusion control on the exterior of the walls, especially if the interior finishes have a low vapor permeance.
Q. "If I am understanding that correctly, then I wonder if one could substitute a flash-and-fill style insulation system into a double stud wall in place of adding many inches of exterior insulation. For instance, if design conditions say you need 2 inches of XPS on the exterior of the wall to negate the need for an interior vapor barrier, but you instead install a certain number of inches of spray foam on the inside of the wall and fill the rest of the cavity with something fibrous, are you not accomplishing the same purpose with that interior spray foam of warming up the sheathing?"
A. Yes.
Q. "Further, would it create a bad 'vapor sandwich' problem for the sheathing to have some exterior insulation AND a flash-and-fill system in the cavity?"
A. Yes. I would advise the use of exterior mineral wool in this case rather than rigid foam.
Smart Vapor Barriers
Yes, I understand the science behind smart vapor barriers, and I don’t doubt that it has been have proven and demonstrated to perform as promised. Really, the only question I have is how long they will perform. I simply do not know the answer to that question. When products are promised to perform on a molecular level, I think it is an act of faith to accept the manufacturer’s promises. I see it as an issue like paint and stain performance promises.
There is one interesting point that I noticed about the Certainteed smart vapor barrier information. The barrier opens up in high humidity under the assumption that high humidity means it is summertime, and drying to the interior would be needed. Any yet, it is humidity that the barrier is intended to stop as the humidity moves outward during the winter. The product information cautions not to use the product if there are continuous high humidity conditions inside the house because the smart vapor barrier would open up and vent the humidity during the wintertime when it is supposed to stop the vapor flow.
The information says that temporary high humidity would not be sufficient to trick the smart vapor barrier into opening when it is needed to be closed. But that seems like somewhat of a fine line to walk.
But my main reason for not using the smart vapor barrier is that I do not believe I will need that type of performance, so I would choose a fixed phase vapor diffusion barrier. I am in Minnesota, so I am not much concerned with air conditioning promotion of in-wall vapor. I have built one house here with the same type of superinsulation and poly vapor barrier system as the new one that I am planning now.
cost of double wall w 2" rigid rock wool vs 2x6 w 4" rockwool
I'm curious if anyone has worked out the cost difference between doing a double stud cellulose wall with 2" of rigid rock wool on outside versus a 2x6 wall with rockwool batts and 4" of rigid rockwool. Let's say the first wall assembly is a staggered 2x4 stud wall on a 9.25 " plate so it all goes together at the same time. Both walls have airtight drywall on inside and plywood sheathing, house wrap, rainscreen, and fiber cement siding ( this is Vancouver Island BC).
Fiberglass Batts & Minneral Wool Batts
Has there been any detailed discussion of the pros and cons of fiberglass batts and mineral wool batts that I could link to? I am interested in the pros and cons of each material relative to all other types of insulation, plus a direct comparison of the two materials.
Response to Ron Keagle
Ron,
Here are three articles to get you started:
Installing Fiberglass Right
Installing Mineral Wool Insulation Over Exterior Wall Sheathing
Batt and Blanket Insulation
Fiberglass Batts Fair Shake
Martin,
With great interest, I read your piece on fiberglass batts. In the comments, I am stunned that people call for a ban on fiberglass batts because people fail to install them correctly.
When you state, “Of all the commonly used types of insulation—including cellulose, rigid foam, and spray polyurethane foam-- fiberglass batts perform the worst,” the clearest implication is that you are comparing performance of the insulation materials per se.
However, you then base this conclusion that fiberglass batts are the worst performer by assessing their actual performance after they are installed on a number of actual jobs, and therefore include poor installation quality in your assessment of insulation performance. I believe it is unfair to conclude the material performs poorly because it is often improperly installed.
You mention the problem of leakiness, but again, you link it to poor quality installation. You also link the problem of leakiness to the insulation material itself by saying that the material is permeable to airflow, which can degrade the insulation performance. But how does that discredit the fiberglass batt? It is intrinsically air permeable, so an air barrier has to be part of the insulation scheme as it does with any other air-permeable insulation.
Response to Ron Keagle
Ron,
I have never called for fiberglass batts to be banned.
My statement — “Of all the commonly used types of insulation — including cellulose, rigid foam, and spray polyurethane foam — fiberglass batts perform the worst” — has been amply documented by many research studies. As typically installed, fiberglass-insulated homes are leaky. Whole-wall R-values for fiberglass-insulated walls are significantly lower than the R-value on the fiberglass package.
As I wrote in the article, it's possible to get fiberglass batts to perform close to the R-value shown on the package, but it takes a lot of work. It is extremely rare to see anyone perform that work. In fact, it takes so much work to try to get fiberglass batts to work the way they are supposed to that almost every builder who has looked into the issue closely has decided that it's easier, faster, and cheaper to use cellulose, rigid foam, or spray foam insulation than it is to carefully detail a fiberglass batt job in hopes of achieving the R-value on the package label.
Fiberglass batts
Martin,
I understand that you mean to say that, in an average sample of actual installations of the common insulation materials, installations using fiberglass batts perform the worst. I suspect that would be true. And I did not mean to suggest that you were calling for a ban on fiberglass batts.
I think that you have fairly covered the difficulties of installing fiberglass batts. It is an interesting problem. Perhaps an underlying reason for poor installation quality is influenced by the physical irritation caused by fiberglass unless a person is wearing full body protection. Most people installing it want to get done as quickly as possible.
Perhaps because the material is so soft and compliant, it encourages deforming and using torn pieces to fill gaps. Fiberglass batts need to be measured, cut, and fitted as though they were as solid as wood. Installation truly does require a fully professional, and conscientious attitude. Apparently, this is lacking in most cases of installation. In regard to the fiberglass batt insulation material alone, I see nothing indicating that it performs poorer than other types of insulation.
I found the comments on fiberglass compression to be very interesting. The golden rule to never compress fiberglass has been long shouted from the mountaintop. But the truth is that limited compression can raise the R-value of the cavity albeit using more insulation than the cavity calls for. But another value of compression is elimination of voids.
The manufacturers finally began to clarify the advantages of compression, however, they worry that too much compression will interfere with the sheetrock work. Now that they offer a high-density fiberglass batt, they seem to no longer promote compression. They will tell you that the high-density material performs the compression for you.
I do not understand what you mean when you say that “whole-wall R-values for fiberglass-insulated walls are significantly lower than the R-values on the fiberglass package.” The package only stipulates the intrinsic R-value of the insulation. The package labeling does not account for thermal bridging, insulation gaps, excessive packing, or other installation defects that altogether will affect the R-value of a whole wall.
Response to Ron Keagle
Ron,
Several factors explain the low performance of fiberglass batts:
- The presence of insulation voids in most framing bays, due to the presence of wiring and plumbing vents, as well as installer sloppiness (careless cutting and haphazard installation).
- The fact that the insulation is very air-permeable, which explains why, on average, homes insulated with fiberglass batts have worse blower-door results that homes insulated with cellulose or homes insulated with spray foam.
- The fact that fiberglass batts can only achieve the R-value on the package when they are surrounded on all six sides by an air barrier -- an installation method that very few builders honor.
Sealing penetrations in walls that use exterior mineral wool
Thanks to everyone for this very informative thread.
I am left wondering how people who decide to use mineral wool outside housewrap and sheathing, e.g., as specified in the USA New Wall design, will seal window and door penetrations?
I don't have any experience working with Roxul, but the number of enthusiastic posts about it suggests there must be fairly straightforward solutions to this problem.
Response to C Talwalkar
C Talwalkar,
There are two kinds of sealing required at a window penetration: sealing against water entry and air sealing. If the wall has exterior Roxul, neither type of sealing differs from ordinary practice.
Typically, the air seal is established between the window frame and the window rough opening.
The water-entry sealing occurs at the WRB and related flashing.
Drying To the Interior Question
Consider a design where you want to stop outward migrating vapor at the interior wall. The insulation will be air permeable, and outer wall will be allowed to breathe with house wrap.
Suppose you forego the use of a polyethylene vapor diffusion barrier under the drywall, and use the airtight drywall to create an air barrier. The theory is that vapor diffusion is so slow and limited that its role in creating condensation of outward migrating vapor in the wintertime is so small that it need not be addressed.
However, if airflow is allowed to pass outward through the interior wall, it will carry water vapor into the insulation cavity and pose significant condensation wetting problems in the wall. So, while the diffusion barrier can be omitted, an air barrier is essential.
Not only is a diffusion barrier probably not needed, but also if there is significant air conditioning in the summertime, a diffusion barrier at the interior wall may condense vapor in warmer exterior air entering from the breathable exterior. So by using only an air barrier in the form of airtight drywall, exterior vapor condensing on the cooler back of the drywall during air conditioning will be able to dry to the interior because the airtight drywall is not a diffusion barrier.
Here is my question:
If diffusion is so slow through an air barrier of airtight drywall that it is inconsequential, how can it be relied upon to dry condensed moisture out of the wall cavity through the interior wall? Because the interior wall is an air barrier, there can be no air movement to carry vapor into the interior to aid drying. All such drying would have to rely on diffusion, which is said to be a relatively poor mechanism for vapor transmission.
It seems to me that if you were to cool the interior space on a daily basis for a few weeks or more, the inward drying would not be able to keep up with the accumulation of condensing moisture inside of the wall cavity.
So how can airtight drywall be a suitable diffusion barrier to prevent outward movement of vapor during wintertime heating, but not be a diffusion barrier that would impede inward movement of vapor during summertime cooling?
Re: Martin Holladay
Thanks, Martin. So let's say you're using 2" Roxul (boards, right? The batts don't seem appropriate to use as outsulation).
Does the window have to project past the Roxul, the furring, and the cladding? If not, what would be the preferred way to carry that penetration past the cladding layer? How would this work with the 10" thick Roxul product Albert mentioned?
Sorry to be so dense. The earlier posts refering to Roxul's compressibility and low adhesion to tape made me wonder how to seal and trim penetrations through a squishy, potentially very thick substrate.
Thanks for your patience!
Response to Ron Keagle
Ron,
Q. "How can airtight drywall be a suitable diffusion barrier to prevent outward movement of vapor during wintertime heating, but not be a diffusion barrier that would impede inward movement of vapor during summertime cooling?"
A. Airtight drywall is not a diffusion barrier; it is an air barrier. It does not prevent outward movement of vapor during the wintertime. Luckily, outward diffusion of vapor during the wintertime is not really a significant problem.
In the wall you describe, the biggest risk is connected to the cold wall sheathing on the exterior. The best way to lower this risk is with a ventilated air gap between the siding and the sheathing. For more on this risk, see How Risky Is Cold OSB Wall Sheathing?
Response to C Talwalkar
C Talwalkar,
Q. "Does the window have to project past the Roxul, the furring, and the cladding? If not, what would be the preferred way to carry that penetration past the cladding layer?"
A. It's possible to do it both ways. For more information, see ‘Innie’ Windows or ‘Outie’ Windows?
Our GBA detail library includes details for both innie windows and outie windows. Links to these details can be found here: Building Plans for a Deep-Energy Retrofit.
Why Is Diffusion Not a Problem?
Here is the point that I am not understanding: Why is outward diffusion in the wintertime not a problem?
Is it because there is not enough vapor available ahead of the membrane to cause a problem once it goes through the membrane? Or is it because the membrane slows the passage of vapor enough to minimize the problem once it gets through?
Certainly there is a difference in vapor pressure from interior to exterior, so this would result in diffusion if a barrier were sufficiently permeable.
My impression was that drywall is a good air barrier, and a good enough barrier to slow diffusion to the point where it is not a problem. My understanding was that poly would be a much better diffusion barrier, but airtight drywall is good enough. Moreover poly has the drawback of stopping inward drying.
I understand your point about the risk of outward vapor condensing on cold sheathing on the exterior. But it can also condense in the insulation before it gets to the cold sheathing.
Response to Ron Keagle
Ron,
Q. "Why is outward diffusion in the wintertime not a problem?"
A. Because the amount of water vapor that diffuses through materials used to finish walls is too small to cause any problems.
Q. "I understand your point about the risk of outward vapor condensing on cold sheathing on the exterior. But it can also condense in the insulation before it gets to the cold sheathing."
A. No it can't. As William Rose has explained, "The language ‘reaching dew point’ seems to indicate that one could plot a temperature profile through a wall, find the point where that profile intersects a horizontal line indicating indoor dew point temperature, and expect burgeoning water at that location. This impression is decidedly incorrect. If water accumulates, it does so on the surfaces of materials, not within the thickness of materials.”
Anton Tenwolde agrees with Rose: "The perceived importance of condensation has been bolstered by the wide misuse of the dew-point calculation. … Many of you are familiar with a chart like this: you project the temperature profile through the wall to calculate saturation vapor pressures. Then you calculate vapor pressures based on the permeance of the materials, and you come up with a profile like this. I have seen hundreds of these profiles, and many seem to show condensation occurring in the insulation. This has encouraged a lot of research into the performance of wet insulation. But the picture is wrong, because the vapor pressure has to be below the saturation pressure. You need to make a correction, and if you do that, if you redraw it, the condensation does not occur in the insulation. We thought there would be a problem with condensation in the insulation, but all the action happens on the sheathing and the interior vapor barrier. We’ve confirmed this by opening up walls. The action is never in the insulation."
For more information, see Are Dew-Point Calculations Really Necessary?
Clarification Please
Martin, regarding your answer to my question:
"Ron,
Q. "Why is outward diffusion in the wintertime not a problem?"
A. Because the amount of water vapor that diffuses through materials used to finish walls is too small to cause any problems."
But why is it too small? Is it because there is not enough vapor available to pass through the drywall; or is because the drywall slows the vapor transmission to the point where it does not accumulate in the wall cavity?
Response to Ron Keagle
Ron,
The amount of water that diffuses through a material is a function of several variables, including the vapor pressure and the permeance of the material.
If you wanted to make a wall fail, you could change some of these variables to make it fail. Fortunately, residential builders don't have to worry about outward vapor diffusion through walls, as long as they aren't building a sauna or a swimming pool.
Vapor movement
Martin,
I conclude that even if you are not building a sauna or swimming pool, there may easily be sufficient vapor in the living space to cause wetting in the wall cavities if that vapor were allowed to go into the wall cavities, especially if the living space is humidified. Moreover, features with vapor producing capability similar to saunas and swimming pools may indeed be present in a house.
Therefore, if vapor wetting is not occurring in the wall cavity, it must be due in part to the diffusion retardation of the drywall, even though drywall is not considered to be a diffusion barrier according to the technical definition. Certainly drywall has some ability to slow diffusion even though its perm rating does not place it in the category of diffusion barrier.
So that goes back to my point that a barrier slowing the outward diffusion during wintertime heating will also slow inward diffusion associated with summertime drying of the wall cavity to the interior. I don’t see how you can have one without the other. So you would either have outward diffusion and inward diffusion or no outward diffusion and no inward diffusion.
Ron Keagle, this is a pretty old thread so not sure if you'll ever see this comment/question. I'm struggling with the same types of questions that you often ask, but I'm still a bit confused. Do you have a satisfactory answer for your question about air and vapor barriers, because I'm still confused. How can drywall be a good enough vapor retarder to limit interior moisture passing into the wall in the winter, and yet in the summer it's plenty permeable enough to let any outside moisture entering the wall to dry to the interior? Now that 6 years have passed since you've asked that question, have you found satisfactory answers?
Diffusion is still enough to become a problem at the cold edge of US climate zone 6 or colder, but it's pretty slow compared to air leakage. A couple square inches of air leak from the interior moves more moisture than diffusion through a whole side of a house. The diffusion of moisture from the indoors into the wall cavities only occurs when the exterior sheathing is colder than the dew point of the indoor air, which in most of the US would be less than 4 months out of the year, which means that 8 months or more of the year. If it's air tight the amount of time for drying the winter accumulation is much longer than the amount of time moisture is being accumulated. And in most wall stackups there is also some amount of drying toward the exterior.
It's possible to design a wall stackup that doesn't need anything other than standard latex paint on drywall (3-5 perms) even in climate zones 7 & 8. In zone 5 or lower those stackups are pretty simple & cheap.
> The diffusion of moisture from the indoors into the wall cavities only occurs...
"occurs" should be "accumulates significantly". You will have vapor pressure differentials that cause outward moisture movement without the sheathing being at dew point temp.
To be even more picky, on a porous surface, this accumulation starts somewhat above the dew point (perhaps 80% RH).
Part of the answer is also that latex paint on drywall is a surprisingly smart vapor retarder. See Figure 10 here:
https://buildingscience.com/documents/building-science-insights/bsi-099-its-all-relative
Thanks Dana and Jon for the response. That bsi-099 article is interesting to me since I've taught high school chemistry and am well acquainted with hydrogen bonds and the dipole nature of water. But the subject of sorption is a bit more esoteric. I'll definitely go back and slowly read that article so I can absorb it (pun intended)
Dennis,
Q. "How can drywall be a good enough vapor retarder to limit interior moisture passing into the wall in the winter, and yet in the summer it's plenty permeable enough to let any outside moisture entering the wall to dry to the interior?"
A. Unpainted gypsum drywall is vapor-permeable. Water vapor can diffuse easily through drywall in either direction.
That said, almost no one leaves drywall unpainted. Once you install a few coats of latex paint, the vapor permeance of drywall decreases. Painted gypsum drywall is still vapor-permeable, but less permeable than unpainted drywall. The more coats of paint you add, the less permeable the drywall becomes. Over the years, drywall tends to become less and less permeable, as homeowners redecorate by painting their interior walls.
Painted drywall tends to slow the diffusion of water vapor into a wall assembly compared to unpainted drywall, and in most cases the rate of outward vapor diffusion through painted drywall is slow enough to prevent moisture problems in walls.
An important point to remember, however, is that outward vapor diffusion through drywall is almost never the cause of moisture problems in walls. Moisture problems in walls are mostly caused by rain penetration. The second most common cause of moisture problems in walls is air leakage (exfiltrating air in winter). Vapor diffusion is way down toward the bottom of any list of concerns.
So here's my advice: Pay attention to airtightness, and stop worrying about diffusion.
Thanks everyone for the quick comments. I totally understand that air leakage is by far the biggest contributor to wall moisture and that diffusion is small if not negligible by comparison. I totally get that. But as I read scores of articles and the comments my head swims because there are so many choices for insulation, wall depth, air barrier locations and climate zones so that I can't decide which factors are relevant and to what degree. I just know that when I spend maybe $300K or more on a house, I don't want any mold and rot, so I'm a little nervous reading about scenarios where this has occurred.
Perhaps it would be better for me to describe my particular scenario. I'm in Lancaster Co, PA climate zone 5 although the counties on three sides are zone 4. My plan has been to build a double stud wall about 11" with dense pack cellulose. Vinyl siding is super popular in these parts and that's what we're planning to use. I think the vinyl needs a good solid surface for attachment so I was intending that the outer wall would be sheathed in plywood. I don't really want to add a second layer of sheathing (as in the Lstiburek ideal wall). Summarizing, the layers are vinyl siding, permeable WRB, plywood sheathing as bracing and air barrier, double stud wall w/dense pack cellulose, then interior drywall using airtight techniques for a double barrier (or at least an attempt at it).
I think in the winter I'm not worried about diffusion and would rely on my double air barrier to prevent exfiltration. And the WRB is permeable so that sheathing can dry to the exterior through the vinyl siding. So I think that's good. In the summer I'm a little concerned about infiltration of humid air being a problem if it can diffuse to the interior air conditioned drywall. Is this fear unfounded? Is this moisture level extremely low? Or since my wall is relatively permeable all the way through, the vapor can dry in any direction and the wall is safe? Or should I consider a vapor retarder mid-wall (which is partly the function of the sheathing in Lstiburek's wall). If the summer humidity diffuses inward and dries to the interior, would this be a noticeable rise in indoor humidity? These are questions that taunt me. Comments to soothe my troubled mind...??
Dennis,
I suggest you visit this page: Category: Walls. Click on all of the articles on double-stud walls. GBA has published many comprehensive articles on the topic over the years, and you'll find your questions answered if you read the articles.
Probably overkill, but consider structural fiberboard {exterior} sheathing and/or {interior side} Membrain for additional double stud {Winter wall moisture} robustness.
> summer humidity diffuses inward and dries to the interior, would this be a noticeable rise in indoor humidity?
Under 5 perms, I consider it insignificant - it's mostly about air sealing and ventilation air.
Reply to Jon R
Where are you suggesting to put the fiberboard and/or Membrain? In the middle of the double stud wall somewhat like Lstiburek's. Or in lieu of the plywood? I didn't want to have two layers of sheathing materials due to cost and labor. I'm not too concerned about additional structural robustness, or are you referring to robustness against vapor?
And am I understanding you correctly that if my wall permeability is less that 5 perms then you think the infiltrating moisture is insignificant for my climate? I know I'm not Georgia where the air-conditioned walls can easily sweat. But we do have some miserable high humidity sometimes.
OK, now clarified.
Insignificant is a different number for everyone, but at < 5 perms, I wouldn't worry about the diffusion driven latent load effect. Summer moisture in the wall due to diffusion is a different question and depends on ratios (and siding type). High perms to the exterior and much lower perms to the interior are probably a mold concern. Wall moisture due to air movement is yet another (bigger) question - air barriers on both sides and very good blower door results are beneficial.
Jon's suggestion is to use asphalted fiber board would be the exterior sheathing instead of CDX, the MemBrain (2-mil nylon) goes under the wallboard.
Fiberboard is extremely water tolerant and very vapor open when the humidity is high, which maximizes the drying rate toward the exterior. Unfortunately even 3/4" fiberboard doesn't quite cut it at dense pack pressures. In Europe and in Quebec there are thicker fiberboard sheathing products available that might cut it, but not in PA.
Vinyl siding is inherently back-ventilated, and the very definition of "Vented cladding" built into the code prescriptive (though alternative vented cladding can also be used.)
https://up.codes/viewer/wyoming/irc-2015/chapter/7/wall-covering#R702.7.1
In a zone 4/5 location vinyl siding over CDX would be sufficient for typical studwall assemblies even without the MemBrain. Higher-R assemlies aren't really covered in the IRC, but with MemBrain and cellulose cavity fill (which shares the moisture burden due to it's high adsorption capacity compared to other insulation) you'll be just fine.
Typar housewrap plus 1/2" fibreboard is plenty strong enough for dense pack cellulose, as long as you put the furring strips over the wrap first. I've even seen houses with just Typar holding in the cellulose. I wouldn't recommend that though, as it does bulge quite a bit, potentially compromising the rain screen gap.
> My plan has been to build a double stud wall about 11" with dense pack cellulose.
> 3/4" fiberboard doesn't quite cut it at dense pack pressures.
See below for a case where damp spray cellulose (presumably with adhesive) was used at 10 1/4". Lower chance of settling makes sense to me - but I don't know of long term performance data regarding damp spray vs dense pack to prove that.
https://www.greenbuildingadvisor.com/question/dense-packed-cellulose-might-go-extinct
But this is mostly theoretical - for colder climates or high interior humidity where cold plywood might get too damp.
Log in or become a member to post a comment.
Sign up Log in