Image Credit: Images #1 and #2: Kohta Ueno - Building Science Corporation
More Musings of an Energy Nerd
Most wood-framed buildings have no insulation on the exterior side of the wall sheathing. That means that the wall sheathing gets cold and wet during the winter.
Whether or not this common situation is a problem depends on who you talk to. Monitoring studies show that the moisture content of the OSB or plywood sheathing on some homes with 2×6 walls rises in February (especially on the north side of the house); fortunately, however, the sheathing dries out in March or April. As long as the sheathing stays dry for most of the year, it can usually endure a few weeks at an elevated moisture content without developing mold or rot. In most cases, sheathing won’t rot as long as the wall’s drying rate exceeds its wetting rate on an annual basis.
Not all builders are comfortable with this analysis, however. Some builders prefer to install rigid foam or mineral wool insulation on the exterior side of their wall sheathing, to keep the OSB or plywood above the dew point during the winter. Warm sheathing is dry and happy, so installing an adequate thickness of exterior rigid foam is one possible solution to cold sheathing worries. (The extreme version of this approach is called PERSIST. PERSIST homes put all of the wall insulation on the exterior side of the wall sheathing, and leave the stud bays empty.)
In theory, a very thick wall with lots of insulation on the interior side of the wall sheathing — for example, a double-stud wall — is riskier than an ordinary 2×6 wall, because the high-R insulation reduces heat flow through the wall, making the sheathing colder and wetter than ever. Some hygrothermal modeling programs, including WUFI, show that wall sheathing on a 12-inch-thick double-stud wall insulated with cellulose can have an elevated…
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
102 Comments
helpful summary, a couple of questions
This is something that's been nagging at me too, the difference between models and experimental data. It is great to see an summary of what we've seen with some real structures so far.
I take Ueno's experiment to mean he looked at one 12 inch double stud wall with cellulose and one double-stud wall with open cell spray foam. That foam is vapor open too, so I'm very curious what the sheathing moisture content was in the case of that stud bay.
Also, what was the thickness of the exterior XPS and Roxul on the 2x6 walls in Straube's study?
Waterloo Study
Leigha -
I can speak to your last question. The walls that we are studying have 3 inches of Rockwool (Density of 8 PSF), 2 1/2 inches of XPS and 2 inches of foil faced polyiso. All were installed in two layers. The idea was to create 3 walls with the same nominal R-value to get a direct comparison.
Response to Leigha Dickens
Leigha,
Q. "I take Ueno's experiment to mean he looked at one 12-inch double stud wall with cellulose and one double-stud wall with open-cell spray foam. That foam is vapor open too, so I'm very curious what the sheathing moisture content was in the case of that stud bay."
A. Kohta Ueno has two winters of data. During the first winter, the moisture content of the sheathing on the spray-foam wall was higher; during the second winter, the moisture content of the sheathing on the cellulose was was higher. He told me, “We had two different walls that we were monitoring, one with cellulose and one with open-cell foam. The cellulose-insulated wall was 12 inches thick. The other wall had foam that was 5 1/2 inches thick. The first winter, the moisture content in the sheathing on the wall with open-cell foam wall was a little higher than it was on the wall with cellulose. That result confused me at first, but then I realized that the foam provides a little bit of vapor resistance. The second winter was more interesting. All of the walls were high. In the wall insulated with cellulose, the peak moisture content was in the range of 25%, maybe up as high as the low 30s, on the north side. The spray foam wall was in the 18 to 20 percent range. All of the walls, by the end of September, were back down to 10 to 12%.”
Q. "What was the thickness of the exterior XPS and Roxul on the 2x6 walls in Straube's study?"
A. It looks like Trevor Trainor just answered your question. Thanks, Trevor.
I have edited the article to include the exterior insulation thicknesses -- information which John Straube provided in our phone conversation, but which I wasn't typing fast enough to catch.
The Cold Sheathing Problem
This “cold sheathing problem” is one of the most interesting issues that I have encountered regarding superinsulated construction. I welcome the elaboration on the topic, and I have a few fundamental questions and observations to consider:
What is the source of moisture that accumulates in the sheathing?
When measuring moisture content inside of walls or in sheathing during the winter; and finding an elevated moisture content; is it known whether the source of that moisture is the interior or the exterior?
If the source of moisture is the interior, why would it get past the proper air sealing of the interior side of the wall?
If the source of moisture is the exterior, why would it not continue drying inward and be absorbed, retained, or buffered by the cellulose? Indeed, why would that inward moving moisture not just continue drying inward all the way to the interior, and not accumulate in the sheathing?
If the wintertime moisture rise in sheathing is from exterior moisture, I assume that it is a one-time adjustment because it is only induced by a lowering of average temperature. However, if the wintertime moisture rise in sheathing is from interior moisture moving outward, the source will be self-sustaining with a continuous feed of moisture generated in the interior living space.
Therefore, if the source of moisture is the exterior, it is supplied as a higher moisture content of air, and will be more readily absorbed by the sheathing simply because of the sheathing’s lower wintertime temperature.
However, if the source of moisture is the interior, it reaches the sheathing as a continuous supply of condensed water. In that case, the lower average temperature of the sheathing would not be necessary for it to absorb a continuous supply of actual contacting water resulting from an outward moving supply of vapor from the interior living space.
An important note about the Waterloo Data
Martin -
When comparing our Waterloo data to the other studies it should be noted that the peak moisture content levels occurred during a period of intentional, controlled air exfilatration.
We intentionally injected interior air into the walls at a rate that we determined as a realistic natural air leakage rate for a building that meets Canadian Energy Star standards (0.2 CFM per sq. ft (or 1.02 L/s/m2) at 50 Pa). We used a rate of 40 CFH per 4 X 8 wall panel (the equivalent of 0.2 L/sec/m2). Depending on the geometry of the house, this roughly translates to 2.5 ACH50.
Obviously a house that is significantly tighter than this should perform better - depending on the distribution of the leakage. I think that is what Dr. Straube means when he states that these type of walls are on the edge and that you have to get the details right. I would say that this type of wall is sensitive to construction quality, where as the exterior insulated walls that we studied are more robust - just my 2 cents.
Response to Ron Keagle
Ron,
Moisture flows through walls are a dynamic phenomenon. The interior surface of the sheathing will take on moisture when the nearby air or insulation materials are wet enough, and will lose moisture when the nearby air or insulation materials are dry enough. Simultaneously, the exterior surface of the sheathing will take on moisture when liquid water gets past the siding and dribbles down the sheathing, or when nighttime radiational cooling lowers the temperature of the siding below the temperature of the outdoor air, and will lose moisture when the sun comes out. Moisture flows in both directions from both surfaces of the sheathing, and the direction of the moisture flows can reverse several times a day. That's what we mean by a dynamic phenomenon.
Winter conditions will cause the moisture content of wood exposed to exterior conditions to rise, up to a point. John Straube said, "As things get cold, water vapor begins to store in the material. However, it will not rise to 25% MC unless you do something else -- unless, for example, there is condensation or rain, frost or dew."
Other factors that can increase the moisture content of sheathing are air leaks in the wall assembly and elevated indoor humidity levels. In some cases, even diffusion can make a difference -- that is, diffusion of interior water vapor through the drywall and insulation. That's why Kohta Ueno speculates that the 0.5-perm vapor-retarder paint on Lois Arena's wall might account for some fraction of her lower readings.
Response to Trevor Trainor
Trevor,
Thanks very much for the information you provided. I have edited the article to reflect that information.
excellent article
Thank you for this fantastic article, which comprehensively addresses a hot (cold?) issue with a very popular wall assembly. I contacted a prominent building science practitioner/author who said that virtually every wall they model fails ASHRAE criteria, that it is generally felt to be far too strict, and that they don't rely on that metric. I have two questions related to this article:
1. I'm not entirely understanding the science behind why that walls with exterior foam (or other exterior insulation) appear to be staying more dry - as they seemingly can't dry to the outside as effectively should they get wet. Is it the retention of heat keeping the sheathing warm, or the lack of penetration of exterior moisture, or some other factor?
2. I'm interested in thoughts about Zip system exterior sheathing, due to the fact that it is seemingly very popular and less permeable in various ways than OSB. I'm particularly interested in this in the context that the Bensonwood OBPlusWall seems to use this as exterior sheathing, with interior OSB sheathing as an air barrier and 9.5" dense packed cellulose. I suppose I don't understand how a wall system that uses exterior sheathing even less permeable than OSB could effectively prevent moisture issues in this context.
Response to Michael Pi
Micheal,
Q. "I'm not entirely understanding the science behind why that walls with exterior foam (or other exterior insulation) appear to be staying more dry - as they seemingly can't dry to the outside as effectively should they get wet. Is it the retention of heat keeping the sheathing warm, or the lack of penetration of exterior moisture, or some other factor?"
A. Physics (and the psychrometric chart) tell us that cold materials tend to be damp and warm materials tend to be dry, which is why our clothes dryers use heat to remove moisture from clothes, and why we put damp crackers in the oven to revive them and restore their crispness.
There is a temperature gradient through a wall assembly during the winter. The temperature of the siding is close to the outdoor air temperature, while the temperature of the gypsum drywall is close to the indoor air temperature. The temperature of the sheathing will be somewhere in between.
If you install foam insulation on the exterior side of the sheathing, the sheathing moves closer in temperature to the interior conditions. In winter, the interior conditions are warm and dry.
If you take the extreme case -- a PERSIST wall, with all of the insulation on the exterior side of the sheathing, and no insulation between the studs -- then the OSB sheathing will be at interior conditions -- as warm and dry as your coffee table.
If your wall has enough exterior foam to keep the sheathing above the dew point during the coldest weather of winter, then there is no need for the wall to dry to the exterior. It can dry to the interior if necessary. The sheathing really doesn't have to dry out in any case -- unless there is a catastrophic problem like a flashing defect that dumps rain into the wall assembly -- because the sheathing is always warm and dry. While the sheathing on an ordinary wall gets wet every February, and therefore needs to dry out seasonally, the sheathing on a wall with exterior foam never gets wet in the first place.
Second response to Michael Pi
Michael,
Q. "I'm interested in thoughts about Zip system exterior sheathing."
A. Zip System sheathing is a type of OSB. The manufacturer uses better resins (adhesives) to bond the wood flakes together than are used on ordinary OSB, so Zip System sheathing isn't as vulnerable to rot as ordinary OSB.
The vapor permeance of Zip System sheathing is about the same as other brands of OSB -- in other words, not very permeable. The numbers are a little loose, because the vapor permeance of ordinary OSB will vary depending on its moisture content.
Zip System sheathing has negligible R-value, so it doesn't solve the cold-OSB problem. However, it is probably more likely to stand up to repeated cycles of wetting than ordinary OSB.
Here's what John Straube wrote: "The vapor permeance of Huber Zip is in the same range as commodity OSBs. OSBs that we test have quite a range of wet-cup vapor permeances, and both roof and wall ZIP are essentially the same. But Zip, like all OSB, does not have the high permeability of a building paper, so one needs to design carefully. For best practice, that means insulation on the exterior to warm it up and to blunt thermal bridges at mudsills and floor joists. If you use a double-stud wall, then I am concerned, but I have a low risk threshold since I am a forensic consultant too. :)
The back of the Zip has a different texture because of the way the product is pressed, and you will notice that other OSB is like this too. There is no difference in performance as far as we have been able to see. Finally, sealing [tape] to plywood is very difficult without mastic; sealing to OSB works OK with primers and some care; and sealing tape to Zip is pretty darn easy. The smooth surface of Zip is its huge virtue."
Leigha's Question; Additional Writeups
I think Martin answered this question pretty well, but if you wanted to see the graphs yourself, you can download them from BSC's website. Unfortunately, we have only completed writeup reports for that first winter (when the interior RH was very low)--the second and third winters will be in an upcoming Building America Report! But the first winter definitely demonstrates that the 12" cellulose wall was the wettest, and the 12" and 5-1/2" open cell foam walls stayed drier in the winter.
Day 1.2b: Double-Stud Wall Field Monitoring (http://www.buildingscienceconsulting.com/presentations/documents/0102b_Double_Stud_Walls.pdf)--presentation I gave in summer 2012; might be a bit cryptic without explanatory text (see report below).
BA-1303: New England Net Zero New Construction Evaluations (http://www.buildingscience.com/documents/bareports/ba-1303-new-england-net-zero-new-construction-evaluations/) Big brick of a report on our Building America work with that builder--see Chapter 5 Moisture Monitoring of Twelve-Inch Double-Stud Walls. Again, only covers the first winter.
Reply to Martin Holladay
Martin,
I understand your points about moisture flow being a dynamic phenomenon. Here is what I am getting at. The cold sheathing problem seems to be based on the observation and scientific expectation that moisture content of sheathing rises during the winter.
So, if that is all there is to it, then the only remedy is to prevent the sheathing temperature from dropping to the extent that moisture gain in sheathing can cause damage. This requires placing insulation outside of the sheathing to keep it warmer.
However, if the extra wintertime moisture observed in sheathing is also being caused by other factors such as poor air sealing, flashing problems, or outward vapor diffusion, in addition to just the lower temperature of the sheathing; then it opens the possibility that the problem may be adequately remedied by fixing those other defects without adding exterior insulation.
Therefore, it seems important to find out where the excess wintertime sheathing moisture is coming from. It is important because if we just assume the entire problem is based on the falling sheathing temperature, it says that the thickening the wall insulation in a double stud system is basically a flawed concept once a certain thickness is reached. This is a fairly dramatic conclusion that calls for a sea change in the approach to superinsulated wall design. So, I think it is important to make sure the conclusion is accurate.
I would like to see moisture testing done on the sheathing of thickened double stud walls with a test house that is 100% air sealed on the interior living space side. Then further testing could be done after making the interior living space side 100% diffusion sealed, so that diffusion alone could be quantified.
Testing with zero air leaks and zero diffusion would show to what extent the cold sheathing problem is actually caused by cold sheathing alone.
This experiment could also be done with a variety of air-permeable insulation types for further study of the effect. Then, finally, I would like to see the experiment accompanied by a sample of the sheathing placed alone in the open near the wall system being tested, so the moisture gain in the sheathing from season change alone could be measured and compared to what is happening to sheathing built into the wall system.
So while the moisture flow is a complex, dynamic phenomenon, I would like to see testing that can break down the complex dynamics so we can see what is happening with the cause and effect of the individual components of the complex phenomenon.
Yeti
I think of cold sheathing as the Big Foot (or perhaps Abominable Snowman is the more apt metaphor) of building science - much discussed, rarely seen. I've been asking for years for someone to show me sheathing rotting away from this problem and it seems like every issue is ultimately caused by something else- water intrusion, lousy flashing, poor window or insulation installation, etc.
Obviously the more insulation we put in an assembly the less our heat loss can cover for sloppiness elsewhere. But if anyone has direct field evidence of a problem purely caused by cold sheathing I would love to see it.
Second response to Ron Keagle
Ron,
Just as "house as a system" thinking forces us to pay attention to the ways that a range hood fan can affect the operation of a water heater, so hydrothermal thinking forces us to pay attention to all of the factors that affect moisture flows through building assemblies.
The list of factors is very long, which is why WUFI is such an amazing accomplishment (as well as why it is, for the most part, a useless tool from a builder's perspective). Here are some of the factors that affect moisture flows through walls:
Interior relative humidity
Interior temperature
Air flow rates through the envelope
Location and size of the air leaks through the envelope
Vapor permeance of all layers of the assembly, including paint, gypsum wallboard, smart retarders, insulation, sheathing, and siding
Presence or absence of a rainscreen gap
Depth of the rainscreen gap
Presence or absence of ventilation openings at the bottom and top of the rainscreen gap
Orientation of the wall
Width of the roof overhangs
Exterior climate (temperature, relative humidity, rainfall, and insolation data)
Exterior wind speed and exposure
Exterior shading
Trust me, Ron -- I left a few factors out.
So of course the MC of the exterior sheathing is not a simple function of the outdoor temperature. No building scientist in the world believes that it is.
And of course all building scientists would love to see more monitoring studies that measure the effects of all of the factors on my list. Building scientists love data, and always advocate for more studies and more funding.
In the meantime, we all muddle along with the best available information. Hats off to the scientists working to research these issues.
Response to Dan Kolbert
Dan,
Thanks for your comments. I remain committed to the hunt for the elusive yeti, as you do, and will be happy to announce that the yeti is a mythical creature, if the search is eventually fruitless.
Even John Straube admits, "These double-stud walls are on the edge, not obvious failures. After all, where are the bodies?"
location matter?
2 story frame... Sheathed... 18' or so ... is the moisture the same or is it high or low on the wall?
A thought I have had is that using pressure treated plates for exterior walls might be worth doing. Moisture and rot loves wood joints. If half the joint was pressure treated the joint just might be quite well protected even with standard studs and sheathing. I have the same idea for tiled showers and baths. Use PT framing the under water safe sheathings. PT plywood under toilets...
Anyway... Is wall moisture evenly distributed?
Accounting for the moisture
Ron -
Maybe I can shed a little light on the origin of the moisture.
In our experiment, with no diffusion and no intentional air leakage, the double stud sheathing moisture content reached 17% when it plateaued- midway through the first winter. The standard 2 X 6 wall and the exterior insulated walls were about half of that (7-9%). I believe that the difference here was due to the built-in moisture in the cellulose migrating to the exterior sheathing during cold weather (ie. the ping pong effect).
We then started injecting air into the walls. The double stud increased to 32% M.C. during air injection, while the standard wall maxed out at 29% M.C. The difference here was that the standard wall seems to have plateaued, while the double stud was still climbing at the end of the air injection phase. The exterior insulated walls stayed at 7-9%. The changes we see here are due to air leakage condensation. The standard wall and the double stud wall both had cold sheathing and reacted in a similar way to air leakage.
Later in our experiment (in late spring) we wetted the OSB from the outside using wetting mats (to simulate rain leakage). The double stud wall reacted exactly like the standard wall, reaching 14-15% M.C., but drying quickly when the wetting phase ended. (more on the exterior insulated walls at a later date)
My take away from all of this is that the sheathing of double stud walls will get wet in the winter - how wet they get will depend mostly on built in moisture and air leakage. They will also dry in the summer. Whether they dry down to safe moisture contents depends on how wet they got in the winter (along with solar exposure, climatic considerations, cladding ventilation and material choices)
Hopefully this helps a little - there is still much to find out!
Trevor
Response to A.J. Builder
A.J.
Q. "Is the moisture the same, or is it high or low on the wall?... Is wall moisture evenly distributed?"
A. The answer can be determined by studying the graphs (Images 4 and 5). In general, orientation matters more than height. In other words, all of the sensors on the south wall recorded lower MC readings than any of the sensors on the north wall.
The graph reporting Kohta Ueno's data shows that the upper sensor on the north wall stayed dryer than the mid-level sensor or the lower sensor.
"the modeling results don’t pass the sniff test"
What a valuable in-depth analysis... thank you. I especially appreciate your calling out issues w/ hygrothermal modeling that's sometimes based on flawed reference standards or other assumptions. Too often, WUFI amateurs blindly rely on misleading results to direct their design decisions: whether those decisions end up being overly conservative or unconservative, this is very worrisome to me. I really appreciate posts like this that collate data from real experts and interpret WUFI analyses in the context of understanding how these things perform in the real world. Kudos!
Response to Kohta Ueno
Kohta,
Thanks very much for your comments, and for the useful links. I have edited the article to include links to the two published BSC documents.
Response to Katy Hollbacher
Katy,
You wrote, "Too often, WUFI amateurs blindly rely on misleading results to direct their design decisions."
I couldn't agree more, which is why we should all see WUFI as a reseracher's tool, not a builder's tool or designer's tool.
I've been worried about the problem of yahoo WUFI users for years, and my irritation is coming to a boil. I'm planning to write a blog on the topic titled "WUFI is Driving Me Crazy."
Where are all the bodies?
From what I've seen and read, it seems that it usually takes a "perfect storm" of building science screw-ups to actually generate "bodies"...(think Vancouver condos or EIFS in Wilmington, NC, for example). But then again, what if EVERY builder in zones 5-8 started building double-stud walls with osb? Would we see bodies then?
Response to Martin Holladay
Martin,
I understand your points about all the complexity of vapor movement, but to say that “a double stud wall is on the edge [of failure]” seems like it is speaking fundamentally about double stud walls. Specifically, it refers to such walls causing the coldest sheathing. But if we are talking about the whole variety of moisture threats, then any wall is on the edge of failure if it has a big hole where the rain comes in. The sheathing temperature may not be causing any problem at all.
You said: “So of course the MC of the exterior sheathing is not a simple function of the outdoor temperature. No building scientist in the world believes that it is.”
Yet the conclusion that a double stud wall is on the edge of failure rests only on the premise that the sheathing takes on more moisture solely because it is exposed to outdoor temperature.
In the most fundamental sense, I do not see how one can draw any conclusions about sheathing temperature causing a moisture problem simply by measuring the moisture content of sheathing.
Yeti as myth...
http://www.slate.com/articles/technology/future_tense/2013/10/yeti_dna_is_probably_just_a_plain_old_bear_sorry_internet.html
While rot & structural failure on the OSB from vapor-diffusion alone in real-world assemblies may hit Yeti status, mold on the OSB sheathing in from vapor diffusion through higher-permeance interiors in cold climates isn't as rare as some might believe, and mold in stud-bays with known interior side air-leaks such as unsealed electrical outlets could be considered common (though not as common as mold in assemblies with air-leaky poly vapor barriers.)
Of course every house has mold- it's only a matter of how much, and where (and in some instances, what type.)
I'm not sure anyone is really concerned about structural failures from cold sheathing, but anything that increases mold hazard developing over decades of use can't be simply shrugged off. Rome wasn't built in a day, but it didn't burn in a day either. Houses built with an air-tight interior today isn't likely to be fully air-tight in every stud bay 100 years hence.
Designing and building to be resilient to imperfections, either baked-in-the cake on day-1 or developed over time is a worthy goal, given that the perfect house has yet to be built. Building with less susceptible sheathing or with sufficient exterior insulation are worthwhile, to improve the resilience of the assembly.
Robert Riversong's approach of skipping the sheathing and nailing ship-lap siding directly to the studs and dense-packing deep trusses with cellulose feels pretty dubious over a century long time frame (and certainly unsuitable for designs that lack deep roof overhangs.) It would be interesting to see one of THOSE houses instrumented & tracked, given that there are now probably dozens of examples out there in New England.
Response to Ron Keagle
Ron,
Here's how science works: scientists formulate a hypothesis, and test their hypothesis by conducting an experiment or gathering data. If the data don't fit the hypothesis, they refine their hypothesis so that their new hypothesis fits the data. Lather, rinse, repeat, as it said on the old shampoo bottles.
No hypothesis is definitive, and every hypothesis can be overturned at any point by new data.
That said, the correlation between the temperature of exterior sheathing and its moisture content is very well established. I invite you to report data strong enough to overturn it.
Other factors matter too, as I pointed out and as you acknowledge. If you want to build a wall with cold sheathing, clearly the sheathing temperature is out of your control. So you have to manipulate the other factors to bring your wall assembly a little bit back from the edge of the cliff. You do this by choosing a moisture-tolerant sheathing, by including a ventilated rainscreen gap, and by paying strict attention to airtightness. In that way, the wall assembly becomes less risky.
Or, alternatively, you install an adequate thickness of exterior rigid foam or mineral wool, thereby making the sheathing warm and dry. Your choice.
Response to John Semmelhack
John,
"It takes a perfect storm," you wrote. But if the leaky condo crisis in Vancouver taught us anything, it's that ordinary everyday construction practices can lead to catastrophic failure on a massive scale. The Vancouver experience wasn't a perfect storm -- it was just business as usual (no sill pans under windows, no flashing where deck rails are attached to walls, and stucco over OSB without an air gap).
Sometimes it takes a few years before we recognize the problems caused by business as usual. And then POW -- the problems hit us all at once, like a lightning bolt.
Response to Dana Dorsett
Dana,
You wrote, "Mold on the OSB sheathing from vapor diffusion through higher-permeance interiors in cold climates isn't as rare as some might believe, and mold in stud-bays with known interior side air-leaks such as unsealed electrical outlets could be considered common."
True enough. We've all seen 2x6 walls with moldy OSB. The question we are all wondering is, "Are today's double-stud walls riskier or less risky than the (somewhat leaky) 2x6 walls we're all used to?"
Time will tell.
Vancouver...
Martin,
My understanding of the Vancouver situation is that the combination of using interior polyethylene and more insulation in the 1980's and early 90's were the two additional ingredients (the "special sauce", if you will) to the existing recipe ('no sill pans under windows, no flashing where deck rails are attached to walls, and stucco over OSB without an air gap') that likely pushed these buildings over the cliff.
Quotes
Martin,
While it can be nice to see your name in print, I have to think my quote above doesn't really add much to the discussion.
We're certainly not building scientists, forensic experts or even builders here at our office. I was just repeating the warnings you've quoted above from actual experts like the folks you've been talking to in detail here like Building Science, Building America and Stephen Winter Associates. Standing on the shoulders of giants and all that…
It's great to hear actual technical detail on the issue. We haven't been seeing any bodies in our area either, just the usual rotted out windows sills. Did have a good one recently with rotted wood sheathing and studs behind a brick veneer wall right on the coast with no weeps or drainage plane, but that's an obvious one.
Jesse Thompson
Reply to Trevor Trainor
Trevor,
Thanks for that explanation of your testing method. As I understand it, you eliminated air leaks or outward diffusion, so that no extra moisture taken on by the sheathing could have come from the interior living space. You conclude that extra moisture came from the original moisture content of the cellulose when it was installed. Do you have any idea what result you would have gotten had there not been excess moisture in the cellulose?
If there is no air leakage or diffusion from the interior living space; and if the wall starts out with dry insulation; and if there are no exterior water leaks; then how much increase in moisture would you expect to see in the sheathing as it moves from summer into winter?
thanks, and a couple more thoughts
Kohta: thanks for the additional links! I will look for the upcoming Building America report. Since the open cell foam walls in the study did stay somewhat drier...does that mean we able to see the distinct fingerprints of vapor diffusion vs air infiltration in comparing the spray foam walls to the cellulose walls? All other things being equal, the cellulose walls allow both air movement across the cavity and vapor diffusion, while open cell spray foam walls would only allow vapor diffusion. I'm not sure how the vapor permeability of cellulose and open cell foam compare, though.
Trevor: Your detailed explanation of the methodology you used is really enlightening, thanks. I'm curious about about one thing--"with no diffusion and no intentional air leakage..." how did you create an initial situation with no vapor diffusion? Just make sure the humidity conditions on both sides of the wall were identical?
Response to John Semmelhack
John,
Yes, the Vancouver builders were using interior polyethylene -- but so were a great many builders all over North America. As I said, it was business as usual.
And the disasters weren't happening just in superinsulated buildings, so the disasters weren't a function of high levels of insulation. Just ordinary 2x6 walls with fiberglass batts, for the most part. As I said, business as usual.
One thing Vancouver has that Las Vegas doesn't is wind-driven rain, and so climate was one factor that made the city's disaster world-famous.
Replies to Ron and Leigha
Ron - the built-in moisture that I am referring to is just what the cellulose would pick up from atmospheric humidity. Since cellulose is very hygroscopic it will never be dry - unless the atmosphere surrounding it is completely dry. The 'ping pong' effect results in a concentration of what ever moisture is in the cellulose at the sheathing side when it is cold out. The thicker the cellulose wall, the greater the effect.
Liegha - We used 6 mil poly sheeting to stop any vapour diffusion from the interior
great topic
The discussion is interesting. My take, at least for starters, is simpler. The average RH in January in central Illinois is between 85% and 90%. So any material outdoors and protected—say a garage wall—will come to equilibrium at the high MC that corresponds to that RH. Boards and plywood, we’ve discovered, will do fine and dry out with the spring. My jury is still out with regard to OSB, whether it will last 100 years. It should relax over time, and weaken as it relaxes. We can discuss rain wetting and diffusion and drying rates and drying directions and WUFI runs, but I just want to know if the sheathing product will survive in equilibrium with winter conditions, protected and not subject to excessive loads.
That’s all we can do with good building science—protect it, keep it from excess water loads from outdoors and in, and make sure we are using our energy dollar for purposes other than artificial drying of exterior structural building materials. I used to think I could make a building last forever. I was saddened with the appearance of OSB to think that, perhaps, despite bringing the best of my art to a project, the sheathing may still become unserviceable 100 years later.
Response to Dana Dorsett
Dana, rather than using Robert Riversong's preferred wall section which still has the cellulose in contact with the wood siding, what do you think of the more recent alternatives which substitute a WRB for the exterior sheathing and separate that from the siding by including rain screen strapping?
What is it about OSB?
The processing of the fibers? The glues holding them together?
Vapour Permeable Sheathing Panels
Just curious - would these "vapour permeable sheathing panels" that are apparently used in Europe solve these issues? I'm thinking that the sheathing would still get cold, but the vapour would pass through more easily ... or at least dry out faster?
Safe(r) moisture (M%) of OSB/plywood and wood sheathing
It is good to hear that expert comment that these double studs walls with exterior OSB/plywood are on the tipping point, but it might be worrisome that most find being on the edge an acceptable practice. I certainly agree these walls are risky, when seeing OSB M% that exceed 20% and even go into the low 30%'s (just like WUFIs run with 160p set up right, EN15026 or a simple sine curve predict).
In my opinion this means that all the safety reserves of the exterior sheathing moisture storage capacity are (more than) exhausted. The question is: do we want to build on the edge - or have a drying reserve/M% reserve that can give some play for some small amount of interior air and exterior water leaks...? I would argue one would, especially in these well insulated assemblies.
If you look at the European norm EN 353/EN636 (see APA-Europe's EN publications It has been defined there that solid lumber (EN 335 in (see Materials for Architects page 122) is regarded free from damage/decay below 20 M%, when not PT. The limit for Plywood/OSB is defined in EN335-2/EN636-2 at an equivalent of 18M% (see also page 130 of Materials for architects").
Consequently, using retarding sheathing (OSB, Plywood) or smart vapor retarders on the exterior of the interior stud wall, makes much more sense for well insulated assemblies. It allows the sheathing to be warm and when combined with fibrous insulation to it's exterior, gives much more drying reserves and creates a more forgiving enclosure (See image). A variety of these assemblies can also be found on our 475 blog:Construction details for Foam free construction
I volunteer my 12" cellulose/exterior roxul wall for study
If anyone wants to measure moisture content in our walls they are welcome. We are at low elevation on Vancouver Island, BC and have recently completed house and suite with staggered stud walls 9.25" thick blown with dense packed cellulose, plywood sheathing, typar, 2" roxul comfortboard IS, 1x4 strapping, and fiber cement board siding. On the inside of the main house is airtight drywall with latex paint and on the suite there is Certainteed Vapour retarder Membrain and then regular drywall with latex paint. The entryway is 2x6 wall with roxul batts and 2" exterior roxul. Both units have HRV's and an interior humidity of around 45% to 48% though I might increase the ventilation a bit to bring down to 35-40% - at least while construction moisture dissipates. The Venmar EKO 1.5 I have on 30 minutes out of every hour to get 40 CFM which is about right for the new Building Science ventilation rate (1700 square feet, 3 bedroom, well mixed) and the suite is 950 sqft with a Venmar Constructo 1.5 set on low (60 CFM which about twice what I calculate is needed, but will likely get the Lite Touch Controller which can be set for 20min out of every hour, or maybe just a hygrometer and fiddle with the outside temperature dial on the main controller, until the humidity is just right ). You can learn more and reach me at http://www.agreenhearth.com
Response to Bill Rose
Bill,
Thanks for your comments. Anyone who has spent much time at construction sites has seen OSB that has decided to "relax." Relaxed OSB is returning to its origin as a pile of wafers.
1x8 boards eventually relax as well, of course, becoming organic matter and topsoil -- but the OSB seems to be in more of a rush to relax than the boards. OSB is like young New York City office workers taking the train to the beaches of Southampton on Friday night, eager to have a few drinks on the way to the beach house. Can't wait to start relaxing.
Response to Dana Dorsett and Malcolm Taylor
Dana and Malcolm,
Plenty of builders (not just Robert Riversong) have built homes without exterior OSB or plywood sheathing, either using board sheathing, skip sheathing, fiberboard sheathing, or one of the European membranes (a strong WRB) to retain the insulation. As long as you can keep the membrane from bellying too much when the cellulose is installed, and as long as you have a way to brace the walls (for example, with interior plywood), all of these approaches are worth considering.
The bellying problem can be real, however. I have visited a job site with fiberboard sheathing with serious bellying problems.
Response to Dan Kolbert
Dan,
You asked, "What is it about OSB?"
Let's see. You take a perfectly good tree, with long fibers held together by lignin, and you chop it up into a pile of wafers. Then you say, "Hmm. I wonder if I can try to glue these wafers back together to make something strong enough to build a building out of?"
Engineers and scientists will eventually make a final ruling on whether OSB is durable enough to use as a building material. But farmers with old-fashioned Yankee skepticism just raise their eyebrows.
I built a garage this summer, using a method common in northern Vermont. My nearest neighbor cut down some spruce trees, all within 1/4 mile of the building site. A farmer in Sheffield with a bandsaw mill cut the trees into 2x4s, 2x6s, 2x10s, and boards. That approach makes more sense to me than trying to glue a pile of wafers together.
Response to Andrew Robinson
Andrew,
Q. "Would these vapour-permeable sheathing panels that are apparently used in Europe solve these issues?"
A. To build a more resilient wall, we need to take as many steps as possible away from the edge of the cliff. Choosing vapor-premeable sheathing (as John Straube has advised) is certainly one way to take a step away from the cliff.
Don't forget the other recommendations as well: paying attention to airtightness and including a ventilated rainscreen gap.
Response to Floris Keverling Buisman
Floris,
The approach advocated by your company -- interior sheathing panels that provide bracing and act as a vapor retarder, and vapor-permeable exterior sheathing or exterior WRB membranes -- is indeed less risky than a typical OSB-sheathed double-stud wall.
However, proponents of this typical European approach need to acknowledge that it is just one way to build walls. Installing insulation on the exterior side of the wall sheathing (rigid foam or mineral wool) also works well.
Response to Patrick Walshe
Patrick,
Your wall sounds resilient. If you have plywood sheathing covered by Roxul mineral wool and a ventilated rainscreen gap, you're good to go. While it's always interesting to monitor the MC of sheathing, I'm guessing that your plywood is in the safe zone.
Stuff Happens
Of Course "Stuff" happens with all assemblies ... even "Outsulated" walls.
Even with the "A-team" contractors and supervision by a well known Building Scientist.
All it takes is something like "reverse flashing".
Response to John Brooks
John,
You're right. The only reason we are talking about resilience and the need to step back from the cliff is because "stuff happens."
It's possible to get flashing errors on a double-stud wall or on a foam-sheathed wall. So we want to start with a robust wall assembly -- one that isn't near the cliff -- so that the wall is able to handle "stuff" without catastrophic failure.
Whether you have a double-stud wall or a foam-sheathed wall, including a ventilated rainscreen gap goes a long way towards lessening the impact of flashing errors.
Regarding the "ping pong" effect
One first gets the impression that the studies are arguing against the double stud wall as being more risky, from the "cold sheathing" problem. Just as adding insulation outside of the sheathing puts the sheathing temperature closer to the interior temperature (warmer), adding instead some insulation inside the sheathing puts its temperature closer to the outside temperature. Comparing a single stud wall with the double stud wall, but with no exterior insulation in either case, the sheathing temperature already will be close to the outside temperature. Going from an R20 wall to R40, and assuming that the sheathing and siding provide R1, the double wall sheathing gets colder by perhaps a degree and a half, not a whopping drop.
It's already been noted that without artificially introduced "air leakage" from the inside into the wall cavity the sheathing moisture rise was substantially lower, as Trevor noted in his post #17. From Trevor's post#33: "Since cellulose is very hygroscopic it will never be dry - unless the atmosphere surrounding it is completely dry. The 'ping pong' effect results in a concentration of what ever moisture is in the cellulose at the sheathing side when it is cold out. The thicker the cellulose wall, the greater the effect."
It would seem, then, that the"problem" with the double stud wall might not be so much the thickness of the wall, vs a standard wall with the same fill, as it is the nature of the insulation. As I've argued before (in an earlier but similar thread on "cold sheathing"), much of the concern seems greatly lessened if there is a well-detailed interior air barrier. Perhaps also we could add to the list of details for good wall construction (rain screen, etc.) a substitution of a non-water absorbing cavity fill, such as blown fiberglass, for cellulose, to reduce or eliminate the "ping pong" effect, at least in climates where exterior humidity is of more concern.
Breaking Down the Cold Sheathing Problem
In thinking about the cold sheathing problem, I have developed the following observations and line of reasoning:
The cold sheathing problem is a result of too much insulation on the warm side of the sheathing. It results in the sheathing taking on a higher moisture content from any one of three different mechanisms:
1) If sheathing is chilled, it becomes thirstier, so it will take on additional moisture from the ambient air on either side of the sheathing.
2) If sheathing is chilled, it becomes thirstier, so if it is exposed to bulk water leakage, it will take on and hold a higher moisture content.
3) If sheathing is chilled below the dewpoint of air on either side of it, moisture in that air will condense on the sheathing and be drawn into the sheathing.
I would consider mechanism #1 to be the core of the problem because its only remedy is to prevent the sheathing from getting cold by insulating outside of it. Because of that requirement, only mechanism #1 fundamentally requires a change in double stud wall design. Basically, that change moves toward the obsolescence of double stud wall design because the more insulation placed outside of the sheathing, the less is needed inside. So, moving in this direction leads to seizing the advantage of reverting to a single stud system by moving sufficient insulation outside of the sheathing.
Mechanism #2 is not part of the core of the problem because it can be prevented by proper workmanship and maintenance.
Mechanism #3 is not part of the core of the problem because it can be prevented by proper air sealing.
So that leaves only mechanism #1 raising the question of whether it increases the moisture content of sheathing during the winter high enough to grow mold. If it raises the moisture content that high, then a conventional double stud wall is a flawed concept.
Therefore, it is important to isolate mechanism #1 and find out what its affect is.
Response to Ron Keagle
Ron,
You concluded, "So that leaves only mechanism #1 raising the question of whether it increases the moisture content of sheathing during the winter high enough to grow mold. If it raises the moisture content that high, then a conventional double stud wall is a flawed concept. Therefore, it is important to isolate mechanism #1 and find out what its effect is."
That's exactly the topic of this article, and the subject of the monitoring studies I cite. All of the studies I cite measured moisture contents above 20%, and some above 30%. These walls were built with above average attention to airtightness, and represent real-world double-stud walls. Only Straube's study in Waterloo involves a deliberate air leak.
If you want to explain away the results by assuming the the walls measured by Andy Shapiro, Kohta Ueno, and Lois Arena weren't airtight enough, I think that conclusion shows an unrealistic expectation for airtightening that is unlikely to be achieved on most job sites.
Response to Dick Russell
Dick,
You wrote, "It's already been noted that without artificially introduced 'air leakage' from the inside into the wall cavity the sheathing moisture rise was substantially lower."
That's kind of true, but not the whole picture. Andy Shapiro measured sheathing moisture contents over 30% without artificially introduced air leakage. As I noted in Comment #3, Kohta Ueno told me, "In the wall insulated with cellulose, the peak moisture content was in the range of 25%, maybe up as high as the low 30s, on the north side" -- without artificially introduced air leakage.
Response to Florus Kleverling Buisman
That is an interesting assembly. One question though: Why is the rain screen strapping such robust material?
Reply to Martin Holladay
Martin,
In post #50, you said:
“Ron,
You concluded, "So that leaves only mechanism #1 raising the question of whether it increases the moisture content of sheathing during the winter high enough to grow mold. If it raises the moisture content that high, then a conventional double stud wall is a flawed concept. Therefore, it is important to isolate mechanism #1 and find out what its effect is."
That's exactly the topic of this article, and the subject of the monitoring studies I cite. All of the studies I cite measured moisture contents above 20%, and some above 30%. These walls were built with above average attention to airtightness, and represent real-world double-stud walls. Only Straube's study in Waterloo involves a deliberate air leak.
If you want to explain away the results by assuming the the walls measured by Andy Shapiro, Kohta Ueno, and Lois Arena weren't airtight enough, I think that conclusion shows an unrealistic expectation for airtightening that is unlikely to be achieved on most job sites.”
**************************************************************************
With all due respect, you are reading a lot more into my comments than I intended them to mean. I am not trying to explain away anything, and certainly find no fault with the testing done by Andy Shapiro, Kohta Ueno, and Lois Arena.
GBA building science is always stressing the importance of air sealing, and even recommending an airtight drywall technique with the intent of preventing living space moisture from entering the walls. So when experiments find excess moisture in walls, I think it is a fair question to ask where it is coming from.
In the test by Andy Shapiro, it was stated that the interior was finished with painted gypsum wallboard with no interior vapor barrier.
In the test by Lois Arena, it was stated that the interior was finished with gypsum wallboard painted with a 0.5-perm vapor-retarder primer.
In the test by Kohta Ueno, it was stated that the interior was finished with straight-up latex paint.
I have no way of knowing whether these walls were or were not “airtight enough,” as you say, but the tests indicate a moisture problem that is directly related to the airtightness of the walls.
In post #25, you explained how science works. It seems to me that the questions I am asking are an example of that spirit of questioning and a willingness to always reconsider any conclusions if there is a reason to do so. I just want to make sure the data fits the hypothesis.
Response to Ron Keagle
Ron,
I agree completely that it is essential for our hypothesis to fit the data. Like you, I would welcome any researchers to comment on the question, "Where is the moisture coming from?" and to propose new studies to sharpen our understanding of these issues.
I'm not yet convinced, however, that "the tests indicate a moisture problem that is directly related to the airtightness of the walls." We don't have any indication of that at this point. (We know that if researchers deliberately inject humid air into a wall assembly as part of a research project, there will be a spike in the MC of the sheathing. But we don't know that the high MC readings recorded by Shapiro, Ueno, and Arena are due to sloppy air sealing. If anything, I would argue the contrary position -- that the walls measured by these three researchers were tighter than average.)
Your summary seems to highlight a possible difference in outward vapor diffusion rates (not air leakage) between the different walls. Kohta Ueno raised the same question. It would be great for builders to know the optimum vapor permeance to aim for when selecting a paint for this type of wall. It's possible that (a) the vapor permeance of the interior paint is irrelevant, or that (b) the vapor-permeance of the paint can make a small difference to help bring us back from the edge of the cliff.
It's also possible that the optimum vapor permeance of the paint layer will differ by climate.
While I'd love to know the answers to these questions, and while I feel confident that these questions will eventually be answered, I don't think that we can hang the resilience of our wall system on the vapor permeance of the paint. Homeowners add layers of paint over the years, which complicates the issue, and it's hard to be sure that a layer of paint is perfectly applied. I hope no one chooses a wall system that is so close to the edge of the cliff that the resilience of the wall depends on the thickness of the primer rolled onto the drywall.
The knowledge required by designers and builders in order to move forward with a project is different from the knowledge sought by researchers. Ultimately, it may not matter to a builder why the MC of the sheathing on a double-stud wall hits 30% in February. If a builder knows that monitoring studies report 30% MC, it's important to take steps to address the problem, and frankly we know most of the things necessary to create safer walls. These steps include the following measures: include a ventilated rainscreen gap. Don't use OSB sheathing. Pay attention to airtightness.
And if you're the kind of builder who is made nervous by MC readings in the high 20s or low 30s, another approach is to build a 2x6 wall and install exterior rigid foam or mineral wool.
Vapor drive?
It is my understanding that the problems with vapor drive get worse as we move to colder climates. In the extreme cold, the vapor differential between indoors and outdoors becomes so high that an absolute vapor "barrier" becomes necessary. My question is, where do we cross this line in the climate regions?
Response to Garth Sproule
Garth,
My understanding: Joe Lstiburek says that we cross the line if we enter the "very cold climate" zone, defined as one with 9,000 heating-degree days or more. And if the building has air conditioning, a smart retarder is preferable to polyethylene.
Response to Malcolm Taylor - why 2x3 vent strapping in 475detail
Malcolm, the strapping on the suggested 475 wall is 2x3 strapping at vented rainscreen IF you use SOLITEX MENTO Plus as the reinforced vapor open WRB material with dense packed insulation blown behind it. To prevent the pillowing of the membrane by the insulation pressure to touch the siding, that additional space is needed for optimum venting and durability. Other solutions are possible (1x's with intermediate flying battens at see detail and image of a Chris Corson project)
If using insulation batts behind the unreinfroced SOLITEX MENTO 1000 membrane, you can opt to use 1x strapping to, as pillowing is no longer a concern.
Late response to Martin H's response to my earlier response
Martin,
Your 2x6 walls with airtight OSB and exterior insulation works if sufficient insulation is used on the ext. For a moisture perspective I would rather see mineral wool to facilitate outward drying in the winter, than foam.
However, since we are talking about double stud walls in this post, and if those assemblies' exterior OSB on gets damp? (ie are such double studs walls a good idea). I wanted to show that there is a different way to build a double stud wall, that keeps the OSB dry and actually makes it's vapor retardency an asset, not a liability. And yes, it is a proven European high performance approach which matches well with the ProClima airsealing tapes we supply at 475. Furthermore it takes the drywall out of the equation, it no longer has to be airtight or retarding and can become an interior sacrificial layer, free for the homeowner to punch holes in and frees you from having to seal each and every outlet.
Response to Floris Keverling Buisman
Floris,
The European approach moves the OSB sheathing from the exterior of the wall to the interior of the wall. This method works, although it's likely to cause head-scratching from U.S. building inspectors and some engineers, who will want to verify that bracing requirements are met (especially in earthquake zones).
In the Belgian Passivhaus fiasco -- the famous Belgian house with wet walls and inward solar vapor drive problems -- some researchers speculated that the OSB was getting so damp from the inward solar vapor drive that the OSB was off-gassing chemicals that affected the occupants' health. (Yes, I know that the Belgian house was not a certified Passivhaus, and that the house had an air leakage rate that was slightly above the Passivhaus threshold -- meaning that the house was "not really a Passivhaus.")
That case made me wonder whether I would really want to have OSB on the interior side of my wall. The case also emphasized the possible dangers of taking a European approach; designers must thoroughly understand inward solar vapor drive before proceeding, especially in U.S. climates where air conditioning is much more common than it is in Europe.
Except for these caveats, your points are well taken.
Where does the water come from?
In response to comment #54 above:
Imagine if all the outdoor water vapor molecules were painted one color--blue, and the inside water vapor molecules were painted red. The question becomes how red or blue is the water in the sheathing?
Try doing a simple steady-state conduction diffusion analysis using the ASHRAE profile method. It uses two vapor pressures, say, vpo for outdoors (blue) and vpi for indoors (red). The outcome is a vapor pressure profile through the wall. Now, do another profile, this time from vpo outdoors to zero indoors. Clearly all the water in the profile is outdoor water (blue). Now, another profile, this time from vpi indoors to zero outdoors. All the water in this profile is indoor water (red).
It turns out that our original profile from vpo to vpi is exactly the sum of the other two profiles, the blue-only profile and the red-only profile. Voila. At each location in the wall there will be a well-defined proportion of indoor vapor and outdoor vapor.
This profile method is fine for theory, but its use for design purposes is strongly discouraged by ASHRAE. So don't pretend that the answer given by this exercise is the real answer. But the exercise does support the idea of Mechanism #1--cold wood gets thirsty and it gets water from wherever it can.
Response to Bill Rose
Bill,
Thanks very much for you comments.
To clarify: Bill's observation, "the exercise does support the idea of Mechanism #1--cold wood gets thirsty and it gets water from wherever it can," refers to the three mechanisms proposed by Ron Keagle in Comment #49.
relaxing
Martin, you mentioned above how all materials relax. I figure that most materials, including OSB, have an elastic range and a plastic range. You get full recovery in the elastic range, so there's no shortening of service life by any activity--temperature or humidity--in that range. Most building materials operate in the elastic range. I've tried to get a sense of what OSB's hygric elastic range is, without much success. How much wafer-swelling does it take to break a resin bond? I wish I knew.
vapor pressure
For those who want to know a bit more about vapor pressure here are the two rules from building science that I found helpful
moisture goes from warm to cold
moisture goes from more to less
In their discussion they explain mold on FEMA trailers in New Orleans, and foam shedding on the space shuttle
http://www.buildingscience.com/documents/insights/bsi-021-thermodynamics-its-not-rocket-science
They also have a doc on external insulation
http://www.buildingscience.com/documents/bareports/ba-1204-external-insulation-masonry-walls-wood-framed-walls/view
Existing construction
I have have a 16 year old home near Concord, NH. It has 2 x 6 stud walls, 16" oc filled with fiberglass batt insulation. There is an approximately 10 mil poly moisture barrier that was reasonably well installed on the interior under the sheetrock, but I am sure there are some areas (where the 2nd floor joists sit on the first floor walls, outlets, and light switches) where water vapor is free to permeate the wall cavity. The cavities between the window and door frames and the rough openings were sealed with foam insulation. The siding is OSB with a Tyvek wrap and vinyl siding nailed on. I was considering installing one 2" layer of rigid foam insulation on the exterior (removing the siding, and installing new). After reading this article and blog, it appears to be a risky solution to get a higher R value, since there is a significant risk of trapping moisture in the wall. Is there a better solution? I have not checked the winter RH in the home, but I don't humidify for many of the reasons mentioned above. I assume the interior RH in the core of the winter is 25% or less.
Response to Bruce Merges
Bruce,
You raise three questions:
1. Does installing rigid foam on the exterior side of wall sheathing make the sheathing dryer or wetter?
2. What is the minimum thickness of exterior rigid foam for this type of wall?
3. Does the presence of interior polyethylene complicate the situation?
Concerning question 1: Adding exterior insulation tends to made your sheathing warmer in winter (and therefore dryer). That's a good thing. As long as the wall has adequate drying potential, adding exterior insulation makes the wall safer, not riskier.
Concerning question 2: Thick exterior foam is safer than thin exterior foam. To learn why, read Calculating the Minimum Thickness of Rigid Foam Sheathing. In your location (climate zone 6), the minimum R-value of rigid foam installed on a 2x6 wall is R-11.25. So if you want to install 2 inches of foam, it has to be polyiso -- not XPS or EPS.
Question #3 is the most complicated, and involves the most judgment. Here is my standard answer:
Many energy experts have worried whether it's a good idea to install exterior foam on a house with interior polyethylene. Although it would be better if the poly wasn't there, the fact is that tens of thousands of Canadian homes with interior poly have been retrofitted with exterior rigid foam, and there haven't been any reports of widespread problems. According to building scientist John Straube, all indications show that these retrofits are "not so risky as most people think. These homes will probably be fine."
That said, the installation of exterior foam is not advised on any home that has suffered wet-wall problems like leaking windows, condensation in stud cavities, or mold. If you plan to install exterior foam during a siding replacement job, keep an eye out for any signs of moisture problems when stripping the old siding from the walls. Investigate any water stains on housewrap or sheathing to determine whether the existing flashing was adequate.
If there is any sheathing rot, determine the cause -- the most common cause is a flashing problem, but condensation of interior moisture is not impossible -- and correct the problem if possible. If you are unsure of the source of the moisture, hire a home performance contractor to help you solve the mystery.
If your sheathing is dry and sound, I don't think you need to worry about adding exterior foam. Adding a rainscreen gap will certainly go a long way toward avoiding future moisture problems. Of course, it's important to be meticulous with your details when you are installing your new WRB and window flashing. It's also important to keep your interior relative humidity within reasonable levels during the winter. Never use a humidifier.
To summarize, here are four caveats:
1. Be sure that your foam is thick enough to keep the wall sheathing above the dew point in winter.
2. When the siding is being removed, inspect the existing sheathing carefully for any signs of water intrusion, and correct any flashing or housewrap problems.
3. Install rainscreen strapping so that there is a ventilated gap between the new exterior foam and the siding.
4. Keep your interior humidity under control during the winter; if the interior humidity gets too high, operate your ventilation fan more frequently.
Research Question
In earlier (perhaps the first) discussions about the cold sheathing problem, I understood the concern to be about the sheathing gaining moisture from the outdoor ambient air. There was a lot of focus on the summer/winter outdoor humidity and how wood takes on higher moisture level when its temperature falls. This would make sheathing wetter in winter; completely independent of outward vapor drive from the living space.
Now, at this point in the discussion, we are talking about the cold sheathing problem being caused by the condensation of outward vapor drive when it reaches the sheathing because the sheathing is below the dew point. However, the outward vapor drive is traditionally stopped short of the cold sheathing by proper air sealing, vapor diffusion retarders, or even vapor diffusion barriers.
Therefore, the cold sheathing problem arising from the condensation of outward vapor drive is due to a failure of these traditional measures to stop outward vapor drive before it reaches the sheathing. Under these terms, preventing cold sheathing is simply a backup safety measure in case of defects in the system of preventing outward vapor drive from reaching the sheathing.
However, if the outward vapor drive were properly prevented from reaching the sheathing or any portion of the wall interior that was below the dew point, would there still be a cold sheathing problem due only to the effect of sheathing getting wetter from absorbing outdoor humidity during the winter simply because the average temperature of the sheathing is colder during the winter? What experiments have been conducted to answer this question, and what was the result?
Response to Ron Keagle
Ron,
I have been trying to disabuse you of the fallacy that there is a single smoking gun that is responsible for the high moisture content of sheathing in February.
In Comment #6, I explained that moisture flows through wall assemblies are a dynamic phenomenon.
In Comment #14, I listed 13 factors that affect these moisture flows, and noted that the list is incomplete.
You wrote, "We are talking about the cold sheathing problem being caused by the condensation of outward vapor drive when it reaches the sheathing because the sheathing is below the dew point." No one besides you has said that. At best, there are some hints that outward vapor diffusion may be a small factor in the phenomenon, but by no means the only factor. Even when a wall has a 0.5-perm interior vapor retarder, as in the case of Lois Arena's walls, the sheathing MC rose to 20%.
The studies described here are trying to gather the data you seek. The studies are not yet complete, and the researchers' papers have not yet been written.
Further response to Ron Keagle
Ron,
I should also note that two researchers have posted comments suggesting where the moisture is coming from. Neither researcher mentioned outward vapor diffusion.
In Comment #17, Trevor Trainor wrote, "My takeaway from all of this is that the sheathing of double stud walls will get wet in the winter - how wet they get will depend mostly on built-in moisture and air leakage."
In Comment #60, Bill Rose answered the question "Where does the water come from?" this way: "Cold wood gets thirsty and it gets water from wherever it can." In other words, the sheathing takes on moisture from its immediate surroundings.
Reply to Martin Holladay
Martin:
You said this:
You [me] wrote, "We are talking about the cold sheathing problem being caused by the condensation of outward vapor drive when it reaches the sheathing because the sheathing is below the dew point."
[your reply to me] No one besides you has said that. At best, there are some hints that outward vapor diffusion may be a small factor in the phenomenon, but by no means the only factor.
*****************************************
When you replied that no one besides me has said that, are you referring to my reference to outward vapor flow condensing due to encountering a temperature below its dew point?
I did mention condensing, but the point of my post could be based just as readily only on outward vapor flow without any condensing. And certainly, I am not the only one who has made that point. Others have mentioned both air leaks and diffusion. Generally, the comments seem to indicate that the source of the moisture is the interior rather than the outdoors. And indeed, if outward vapor flow is part of the cold sheathing problem then I don’t see how you can exclude the possibility of condensation.
But, in any case, condensation or not, my only point was to distinguish the cold sheathing problem being caused by outward air leaks or diffusion, as opposed to being caused by outdoor ambient moisture contacting the sheathing.
I am not looking for a smoking gun. I am not sure why you think I am. I can fully accept that there may be multiple causes.
But when people conclude that double stud walls are risky, I want to know if the risk could be managed by air sealing and diffusion control; or if the only way to control the risk is by adding exterior insulation.
I ask because it may be preferable for a variety of reasons to control the risk by raising the quality of air sealing and diffusion control, as opposed to adding exterior insulation. But if the cold sheathing problem is rooted just as deeply or deeper in the presence of outdoor moisture, then solving it by better air sealing and diffusion control is futile.
More studies to study
Martin,
Excellent article and dialogue, I've learned much. Slightly off topic: I wonder if you and your colleagues have been following the OSB moisture content study under way by NAHB's Home Innovation Research Labs. While it is not focused on double stud walls, the 22 test homes under 24-7-365 electronic monitoring in a variety of climate zones in moisture zones A and C is revealing some excellent data. The homes were presumed to be built to the 2012 IECC and include a wide variety of wall assemblies.
The testing was undertaken by NAHB to test a hypothesis that "tight" homes result in wall cavity moisture problems, a hypothesis that was overly broad and "politically" fraught, in my opinion, with far too many variables to come to any consensus conclusion. But it is accurate moisture data nonetheless.
Numerous battery operated monitors are spread over all exterior walls and send signals every 15 minutes. Their life expectancy of the devices is presumed to be 3-5 years. The moisture/time charts are fascinating and can be cross-referenced to similar day-by day weather data that often explain anomalies, like what was the weather like when the OSB was first installed before dry-in.
I would welcome your analysis of the data in a future blog post.
Response to Kim Shanahan
Kim,
Is this the study you are referring to?
"Initial Study of the Moisture Performance of OSB-Sheathed Walls in Homes in Climate Zones 4 and 5."
As far as I can tell, this was a one-year study that looked only at the first winter after construction was complete. It's possible that the sheathing MC never reached the same levels in subsequent winters.
A few conclusions from the study:
"The OSB moisture content increased from less than 6% to 11% at the time of panel installation or jobsite visit to anywhere from 10% to 25% or higher at the time of first measurements taken after the beginning of occupancy that started in the months of January or February. In three of the homes, the OSB moisture content levels exceeded 15% for an extended period of time. In two of those three homes, moisture content peaked at over 25% with the highest measured moisture content at around 30% (one sensor only). ...
"In all homes, the moisture content started trending down in the spring months as the average daily temperature increased. A typical observed drop of OSB moisture content ranged from 5 to 10 percent. The largest drop in OSB moisture content in each of the homes occurred during the period of April 1 through April 9 and was associated with increased ambient temperatures and decreased ambient relative humidity leading to increased cavity temperatures and drying potential."
Two graphs showing high sheathing MC levels are reproduced below.
.
I feel sick...
I just built a double wall house in Maine using two 2x4 fiberglass batt filled walls on either side of 2" foil faced polyisocyanurate board. The house is sheathed with Zip system and sided with wood clapboards. How screwed am I?
I was always under the impression I just needed to eliminate condensation, I never knew about "thirsty wood". Why am I just hearing about this now?! Arg!!!
At this point, it would seem my only option would be to reduce the interior humidity, (?). Any thoughts on a safe RH?
Response to William Malcolm
William,
That's an unusual wall system, for sure. The "Malcolm wall" is unique.
If you have a layer of foil-faced polyiso dead-center in the middle of your double-stud wall, your interior humidity levels are irrelevant (at least as far as your sheathing moisture content is concerned), because the foil-faced polyiso, for good or ill, will stop any outward vapor drive from the interior to the sheathing.
Did you at least include a ventilated rainscreen gap between the siding and the sheathing, as GBA recommended in our 2010 article on the topic ("How Risky Is Cold OSB Wall Sheathing?")?
Misleading metaphors
Having read through this excellent article and the entire discussion up to this point, I'd like to quibble over one term that some have used. I think "thirsty" sheathing is a misleading metaphor, that is causing more confusion than clarification. Even the statement that OSB sheathing is "wetter" in the winter requires attention to the somewhat counterintuitive physics of temperature, relative humidity (RH), and moisture content (MC), to avoid drawing incorrect conclusions. Vermont sheathing is likely to contain more grains per sq ft/grams per square meter of water in September than it does in January, by a significant margin. If anyone wants to argue that it is wetter in September, they have this fact on their side. On the other hand, the figures for OSB sheathing moisture content percentage reported in this article are quite a bit higher in January than in September, so the sheathing is certainly wetter in that sense. The moisture content of the OSB is higher in January, even though the number of water molecules it contains is less.
While many readers will consider this obvious, my impression is that some haven't thought through all the relationships. If you were to take a January outdoor air sample at -10 degrees F and 95% RH, and warm it to a summery 95 degrees, the relative humidity would become 1.6%, even though the air had not lost a single molecule of moisture (according to an online calculator). Conversely, 95 degree air at 90% RH, if cooled to -10 degrees, would attain a relative humidity of 5380%, if the moisture didn't begin condensing at an RH of 100%. The excess moisture doesn't need to come from anywhere. It is contained in the sample at the start of the heating or cooling.
While OSB moisture content doesn't show such extreme variation, it does follow the same temperature trend. This is the foundation for Martin's opening statement, "wall sheathing gets cold and wet during the winter". At the risk of introducing another misleading metaphor, cooling air or sheathing doesn't make it thirsty, it makes it sweat. Excess moisture is released from the sheathing as it cools. Very small amounts of moisture at very low temperatures result in very high relative humidity and high moisture content for the OSB sheathing. This agrees with the data presented. Houses built to better than average tightness levels will still allow a small amount of moisture transmission. Combined with fairly normal New England winter conditions of precipitation, temperature, relative humidity, and wall construction, high moisture content in the sheathing is the result.
Reply to Derek Roff
Derek,
Regarding the metaphor, “thirsty,” in post # 49 on page one, I said this:
The cold sheathing problem is a result of too much insulation on the warm side of the sheathing. It results in the sheathing taking on a higher moisture content from any one of three different mechanisms:
1) If sheathing is chilled, it becomes thirstier, so it will take on additional moisture from the ambient air on either side of the sheathing.
2) If sheathing is chilled, it becomes thirstier, so if it is exposed to bulk water leakage, it will take on and hold a higher moisture content.
3) If sheathing is chilled below the dewpoint of air on either side of it, moisture in that air will condense on the sheathing and be drawn into the sheathing.
*******************************************************
I understand your point that air with a given amount of water vapor will rise in relative humidity as its temperature drops, and when the RH reaches 100%, it must give up water vapor by condensation into water. You seem to be applying that principle to the wood sheathing when you say that wood will sweat off excess moisture that it cannot hold as the temperature drops.
However, according to Bill Rose, wood will gain moisture as its temperature drops because it is able to hold more moisture at lower temperatures. It is not merely acting analogous to air gaining relative humidity. The wood is gaining molecules of water. In post #60 above, Mr. Rose said this in reference to my comment above using the “thirsty” metaphor:
“But the exercise does support the idea of Mechanism #1--cold wood gets thirsty and it gets water from wherever it can.”
However, I note that the cooling wood will gain moisture only if there is moisture available. Cooling wood will not gain moisture if no moisture is available. Therefore, considering that qualification, rather than assert that wood unequivocally will gain moisture if it cools, I only say that the wood will get thirstier as it cools.
Response to Derek Roff
Derek,
To speak frankly, you are wrong.
Perhaps it is easier to understand what's going on if we imagine a 2-foot length of 2x4 that is outdoors in the rain in September. Bring it indoors and weigh it. Let's say it weighs 2 pounds.
Put it in a 200 degree oven for two hours. Weigh it again. Now it weighs 1 pound.
When it is warm and dry, it weighs less. The missing weight was water.
It had more molecules of water when it was cold and wet than when it was warm and dry.
You are confusing the relative humidity (RH) of the air with the moisture content (MC) of wood.
response to Martin: re: I feel sick....
Thank you. I feel a little better that my, "interior humidity levels are irrelevant".
No, I don't have a ventilated rainscreen gap. I framed my house in 2009, so I hadn't read the 2010 cold OSB article yet...
Again, Arg!
My house stays at 45% RH whether I like it or not. I'd like to bring it down to 30%(?). I'm thinking differential enthalpy control of a HRV... Any other ideas?
Help! I can't sleep at night! What is going on in my walls?!
Response to William Malcolm
William,
First of all, try to get a good night's sleep. There is no reason to panic. I don't think you have any evidence of moisture problems.
If your interior RH is at 45% during warm weather, that's normal. In the middle of winter, that would be considered high. To lower your interior RH during the winter, run your ventilation system for more hours per day. Unless you have an unusual source of moisture in your house -- a wet basement, too many houseplants, 10 aquariums, or a cord of wood drying in your living room -- increased ventilation should do the trick.
Wrong report
Martin,
The study you cited from HIRL is not the one I mentioned. I was not aware of that older study. The one I'm referring to is ongoing. NAHB wants all 22 homes to go through a 5 quarter cycle of data before conclusions are drawn. That may take another year since the starts for the 22 homes has been staggered over 18 months.
The data on the first homes that have completed the 5 quarter measure are not causing alarm and in fact disprove the original hypothesis. A couple of those first homes were in Michigan in I believe climate zone 5 and 6.
HIRL's Vladimir Kotchkin is the keeper of the data. I would expect that he would share with GBA.
Question for Kim Shanahan
Kim,
You mentioned that the NAHB hypothesis was that “tight” homes result in wall cavity moisture problems. Do they offer any elaboration on that hypothesis? In their tests, do they define what they mean by “tight”?
A note to Andy Shapiro
If you use cedar that is NOT back primed over Tyvek, the wall will NOT be vapor permeable. I have opened many of walls, and can completely with confidence declare that the tannins in cedar breakdown Tyvek, clog the pores, and results in very moist or wet sheathing. People, please get with it and leave the OSB at the lumber yard. Use PVA primers, Kraft paper( such as Fortifiber AquaBar B), and otherwise high perm assemblies, proper flashing, and rain screens and bigfoot will not rear it's ugly head.
The belgium PH again - but 475 now has a solution
Martin,
The Belgium PH had a brick reservoir cladding, with no vents at top or bottom....so no way for the solar driven diffusion to go anywhere - but indeed the OSB on the inside was not a good idea in that wall (INTELLO might have been fine).
It would have indeed been possible to build such a brick veneer wall, vented, with a perm 6.5 WRB - we happen to have one of those available now SOLITEX FRONTA HUMIDA. This perm 6.5 WRB limits the solar vapor drive into the assembly, while still being sufficiently vapor open to allow outward drying in winter. Of course one should use INTELLO Plus on the inside to minimize vapor diffusion into the wall in the winter, as the outward drying capacity is limited because of the FRONTA HUMIDA perm rating.
Response to Floris Keverling Buisman
Floris,
I agree completely, of course, that the lack of weep holes at the bottom of the brick veneer wall in Belgium was a major mistake, and that it is a good idea to vent the top of the gap behind brick veneer. But the case provides important lessons. Many wall assembly designs -- whether typical American designs like a cellulose-insulated double stud wall, or a typical European design like a wall with vapor-open exterior sheathing and an interior vapor retarder consisting of OSB -- work well unless there are one or two construction errors that push the wall to failure.
In Belgium, the construction error that pushed the wall into failure was the failure to properly vent the cavity behind the brick veneer. (I can assure you, Floris, that that error is extremely common, all over the world).
In the U.S., a typical construction error that can push a wall into failure is the lack of good sill pan flashing under the windows.
On both sides of the Atlantic, we need to (a) understand the mechanisms that drive moisture through walls, and (b) develop wall assembly details that are so robust that they can survive the occasional construction error.
It's a big challenge.
Zip R sheathing solution equals a PDMTDSW
Martin, killing a few birds with one stone. I am thinking Zip R sheathing would work very well with double stud 12" plus thick exterior walls built in cold zones 6 etc. Why? Both sides of the panels are well protected from water. And if the inside joints were caulked with one of the two systems that are out there now then the OSB part of the panels would be very well protected from water. And to satisfy Dana the protective layers still have permeability. Air stopped but not vapor stopped. On the inside I would think the foam layer would not react to condensation events. Add my other idea of a pressure treated bottom plate and I would say a PDMTDSW (prettty d*mn moisture tolerant double stud wall).
Second point. The formula that says we need to add enough exterior insulation to keep our homes safe from moisture damage bugs me. Why? Well very most homes today near me if not all are built with no or just a very thin layer of exterior insulation. And the homes are not being built with air tight sealing of wall penetrations such as exterior electric box cut ins. And more foam is mo money. Any hoot, Zip R sheathing is expensive but... hey.... I say... if used... with a PT sill plate... boom... no moisture problemos... STraube... Rose... Joe... Huber... I will test the theory... Get me a grant and lets add meters probes and start thinking out of the box...
Martin, Dana? Poke some holes in my thoughts... glad to hear them.
Response to AJ Builder
AJ,
You understand the concept -- "Air stopped but not vapor stopped" -- but you have the wrong product.
The permeance of Zip sheathing is about 2 to 3 perms. John Straube has written, "The vapor permeance of Huber Zip is in the same range as commodity OSBs. OSBs that we test have quite a range of wet-cup vapor permeances, and both roof and wall Zip are essentially the same."
I agree with your goal for cold sheathing -- "Air stopped but not vapor stopped." That's why fiberboard, board sheathing, and plywood are all better choices than OSB or Zip sheathing.
Another response to AJ Builder
AJ,
I'm not quite sure what your point is when you write that most homes near you are built with no exterior rigid foam or very thin foam, and are built with no attention to airtightness.
Do you conclude that these are good methods? I suspect that the homes you describe aren't as comfortable as a well-built home, and that some of them may have damp wall sheathing.
Zip R sheathing keeping walls from moisture failure
Main point Martin is Zip R sheathing seals the OSB part of the sheathing on both sides from being soaked in moisture. I am thinking that if monitored it would not be going into the moisture percentage levels discussed that are going to start the OSB part of the panels to decay... expand and stay expanded etc.
My point about how most is built that I see is that if any foam is installed, it is done in a way that should be very much bad for the underlying OSB. To me foam on the Zip Sheathing sets up the sheathing to be safe from long term damage. Just guessing but makes sense to me anyway.
One other point... since growth loves a mild climate all the builders out there adding not enough foam to the exterior of OSB are in my thinking setting up the walls to rot much more often then if no foam was used. The sheathing is going to be in the rot mold temperatures for more time I am positing.
So I advocate... no foam... Zip R sheathing or two layers of 2" minimum over 24"oc framing in my zone 6A.
I'm really starting to like my idea of a cellulose double wall with Zip R sheathing sealed on both sides. The cost, the labor and the simplicity along with what may prove to be durable too,.
aj
Another response to AJ Builder
AJ,
Your tune seems to have changed. It sounds like you're no longer saying "Air stopped but not vapor stopped" -- now you are saying "Air and vapor both stopped," which means that water will dribble down to the bottom plate and puddle. So you want to install a pressure-treated bottom plate, so that the puddles don't cause too many problems.
Count me out.
Just say no to drip Martin: cellulose insulation
The point of this blog is that the OSB may be getting too high a moisture level.
My assembly I believe solves that for the OSB.
And puddles? Martin. Dense packed hygroscopic cellulose takes care of that.
As to permeability. The Zip R sheathing is in my mind bullet proof. It is protected by permeable materials on two sides. The tape and the caulk and the cellulose take further care of the moisture sensitive OSB.
It is time to test my assembly.
Aj
Comment
AJ,
I would not want puddles in my walls even if I did have treated floor plates.
NO PUDDLES SILLY BOYS
GBA for years and years has had posts about the value of dense packed cellulose and double stud walls. Years!!! No one ever has mentioned puddles forming. NO ONE. My idea of pressure treated plates is not to deal with puddles. NO puddles! Rediculous. I have used thousands of feet of pressure treated wood. The PT wood idea for me is to use it all throughout a home where at times not all the time there are elevated moisture concerns. Like near toilets and showers and bottom sill plates of walls and windows. I have taken enough homes apart to know often rot damage is limited to areas of a home like a band joist for example and the sill plates that we build first floor decks upon.
This Zip R sheathing idea for a dense packed double stud wall is to me the solution to what is being observed in the experiments mentioned in this blog. If my wall is built and monitored at the OSB layer and wherever else like the plates I bet the findings would be of value.
At this point to be negative is just rediculous. At this point we should build this wall and test it, period.
Martin, be positive aj... well this is my positive. Lets build this wall and monitor it.
list of monitors prices sellers
Martin, could you work on getting and sharing monitor device information from the participants of this blog and thread?
We need to build monitor share our assembly ideas and move forward to improved assemblies that are out of the high risk zone such as the assembly that is the topic of this blog.
Maybe those that have monitored can post and share more on their devices used.
Reply to AJ
AJ,
In post #84, you said this:
“Add my other idea of a pressure treated bottom plate and I would say a PDMTDSW (prettty d*mn moisture tolerant double stud wall).”
In post #88, Martin replied to you:
“now you are saying "Air and vapor both stopped," which means that water will dribble down to the bottom plate and puddle. So you want to install a pressure-treated bottom plate, so that the puddles don't cause too many problems.”
From this exchange, I interpreted you to be proposing a treated bottom plate in order to withstand the wetting that Martin refers to as “puddles.”
However, if there will be no puddles, as you say, why are you proposing that the bottom plate be treated? If you are just concerned about elevated moisture, why would the bottom plate be at risk and not the studs?
Source of monitors
AJ,
Many builders and researchers use Hobo data loggers from Onset. Here is the link:
http://www.onsetcomp.com/
Before you get too deep into the world of backyard moisture content monitoring, you should know that it's easy to get bad readings when you are measuring the moisture content of lumber or OSB. Read up on the topic to make sure that you avoid all the usual newbie errors.
If you don't know all the pitfalls, it's easy to get worthless data.
ZIP R-Sheathing
Martin,
Old thread resurrection time, I just figured out what AJ was trying to talk about somehow. He must be referring to a specific new Huber product, ZIP R-Sheathing which is OSB with 1/2" or 1" of poly-iso bonded to the inside face: http://www.huberwood.com/zipsystem/products/zip-system-rsheathing
We're seeing several local builders starting to use this product.
I think he was proposing that using this as exterior sheathing on a high-R thick wall instead of typical OSB would be similar to a closed cell spray foam flash & dense-pack wall, in that the thin layer of foam on the interior face of OSB would prevent moisture from accumulating on the OSB, even if it was below the dew point.
Response to Jesse Thompson
Jesse,
You may be right; however, it's hard to be sure what A.J. Builder was referring to.
There is a problem with thin foam on the exterior side of a thick insulated wall: if the foam is so thin that the interior surface of the foam is below the dew point, moisture can condense on the foam. In the case of Zip R sheathing, the moisture may not reach the OSB, but it can dribble down the foam and collect in puddles on the bottom plate. That is obviously undesirable.
Anyone contemplating flash-and-batt (or Zip R sheathing) has to make sure that the foam layer is thick enough to avoid condensation or moisture accumulation. For more information on this issue, see Calculating the Minimum Thickness of Rigid Foam Sheathing.
Insulation over and over
I have done quite a bit of reading around the Internet and I am still little confused. I own the house on uninsulated concrete slab (it will stay like this at least for some time, no basement). House had the second floor addition done in 2008, walls should be R15 and attic R30. Studs on both floors are 2x4,16" apart. The problem is the first floor. It is 60 years old, and has poor rock wool insulation inside. Outside it has probably asbestos siding covered with vinyl siding. I do not think there is any rigid foam in between. I am in NY, Long Island, which can be pretty dump in summer time, zone 4A.
I would like to remove sheet rock (walls and ceiling) and insulate the walls, and ceilingrim joists (only where they “meet” outside wall). I have the following concerns (let’s say that money is not a concern in a sense that I would like to stay frugal, but I am willing to pay more for better and long standing solutioncomfort of my house):
1. First, I wanted to fill all stud cavities with closed cell foam. However, I am afraid that I will create vapor barrier, and as far as I understand reading on the Internet I should not do it, as the moisture will transfer to studs and can create mold problems. Is that, correct? Plus it is expensive and supposedly not feasible in economic sense as I would lose its R value do to studs...
2. Second, I was thinking to spray 1” or so of closed cell and fill the rest with open cell. Is that good option? Should I be careful again not to create vapor barrier by closed cell foam, so keep it thin? I also read that closed cell needs to be sprayed in at least 2" layers...
3. Third option I was considering was this:
http://www.builditsolar.com/Projects/Conservation/MooneyWall/MooneyWall.htm
However, I would space horizontal 2x2 by 24” to further minimize thermal bridging. Can I, alike in #2, spray 1" or 2" closed cell foam first and then fill the rest with open cell foam (I think it would be to thick in this case and act as vapor barrier too), or rather cellulose (as pictured on the website)? My wall would be 2x6 only so I hope moisture problems related to extra thick walls will not apply? Do I need to calculate dew point here?
4. I am afraid of cellulose settling. Is it a big deal nowadays? It somehow makes me resistant to use it. Am I exaggerating? I saw cellulose being recommended here multiple times… so maybe I should just go with 2" of closed cell and then cellulose...
5. In all cases I would NOT add vapor barrier on inside.
6. If none of the above is good solution, what would be the best way to achieve the highest possible R value for given circumstances (also keeping economy in mind)? I would like to avoid removing exterior siding.
I greatly appreciate any help and suggestions!
Response to Sebastian Smith
Sebastian,
The fact that many types of spray foam and rigid foam are also vapor retarders or vapor barriers is not a problem. Because these insulation materials have a significant R-value and are air barriers, there are no opportunities for vapor to condense on cold surfaces.
The Mooney Wall solution is fine, if you want to go that route.
If you choose a competent insulation contractor who has experience with the dense-pack method, there is no reason to worry about cellulose settling.
If you want the R-value of your wall to be higher, make your wall thicker. Most people compromise on R-value, rather than aiming for the "highest" R-value. There is no "highest."
defending WUFI
I designed and built the first certified Passive House in the western United States in 2009 using a double stud wall with exterior OSB in a 6000 hdd climate. At that time, there was no government supported resarch and no monitored/measured data regarding long term moisture risk to these walls (at least that I knew of). So I turned to WUFI, and despite being an amateur, used it to see what I could see. And guess what I found out? Pretty much exactly what the monitored research is starting to show now, and what Martin is reporting in this blog--that a double stud wall is borderline in cold climates, but can work with good air-tight layer detailing and attention to the moisture profile of the wall. In my climate, which is very dry in summertime, I was satisfied that the risk was low enough to build.
Of course there were things I did not know in those early days: nothing about ASHRAE 160p. I also did not know at the time is that the model was assuming basically zero air leakage, so in that sense I was lucky to be building an air-tight house. I also did not know enough to model the north wall as the worst case scenario. But for an imperfect computer model, it allowed me to make some prudent design decisions. And when I submitted the building to PHIUS for certification, one of the first questions was: how do you know this wall is safe from moisture damage?
Even now that we have what amounts to a tiny bit of monitored data showing moisture in double stud walls, WUFI remains an invaluable tool for building designers and anybody else who is working with unconventional and/or super insulated walls. For instance: what happens when you add exterior foam to the double stud wall? How thick is the foam and can the wall still dry to the outside? one inch? two inches? XPS or EPS? What if there is a rain leak in the wall? do you need a ventilated rain screen then? Boston or Salt Lake City?
...as you can see, there will never be enough monitored data to answer these questions.And now, with Wufi Passive, practitioners can do Energy modeling and hygrothermal modeling with the very same program. It is the future. WUFI only drives you crazy if you don't know how to use it, and understand it' limitations.
Response to Dave Brach
Dave,
You listed several things about WUFI that you didn't really understand when you used the program -- for example, "I also did not know at the time is that the model was assuming basically zero air leakage, so in that sense I was lucky to be building an air-tight house." So you were lucky.
Other builders and designers who have tried to use WUFI have understood the program even less well than you did when you first used it. Do they know whether their buildings are exposed to the wind or sheltered? Have they guessed correctly at the air leakage rate? Do they know what the interior moisture loads are? Do they have enough experience to know whether the WUFI results pass the smell test?
I recently asked Achilles Karagiozis why the cold temperature performance of polyiso didn't show up when WUFI was validated. His answer was that WUFI hasn't been validated for buildings with polyiso on their walls. Well, that's an interesting hole in the validation procedure, isn't it?
You wrote, "WUFI only drives you crazy if you don't know how to use it, and understand its limitations." Fair enough. However, few builders and designers are likely to have the time required to learn how to use it, and to learn all its limitations.
As Sargent Joe Friday always
As Sargent Joe Friday always said-' The facts , just the facts' Thanks for the facts, or such as we have for now..
Will a better VB help?
I enjoyed and appreciated Martin Holladay's article "Monitoring Moisture Levels in Double-Stud Walls" and after pondering its conclusions I have two thoughts:
1. How much improvement would a well placed vapor barrier offer? One obvious problem with vapor barriers is that they are prone to puncture BUT if a good poly VB is placed on the backside of the inner stud wall (i.e. 3.5" into the wall) all of the tasks that typically cause punctures (running pipes, power, electrical boxes) would be inboard of the VB and likely not damage it. True, this implies cellulose being blown in two stages but wouldn't it have the potential to drastically reduce the amount of moisture that condenses on the sheathing?
2. The article notes that research has shown 6" walls performing better as far as condensation goes BUT that is because they are performing more poorly in terms of keeping heat inside the house - i.e. the sheathing is warm because heat is escaping. Since keep heat in is the goal of these walls it seems more research on sheathing materials that will not suffer from the condensation is an appropriate next step. The concluding suggestion of the article notes several alternate sheathing suggestions. Have any walls using DensGlas or similar non-organic materials been tested to determine the results? Thanks you.
Log in or become a member to post a comment.
Sign up Log in