Vapor retarders in walls and unvented cathedral ceiling in Climate Zone 6B
We are building a house in climate zone 6B (at 8,500 ft elevation, 8,000 heating degree days, with warm and dry summers). I have questions about the placement of vapor retarders.
The roof assembly, from exterior to interior, is this (essentially as in Fig. 14 of this Building Science report (https://www.buildingscience.com/sites/default/files/migrate/pdf/BA-1005_High%20R-Value_Walls_Case_Study.pdf):
* Metal roof
* Water drainage plane + sheathing
* Air gap to prevent ice dams
* Rigid foam insulation (R-27)
* Peel-and-stick air barrier
* Sheating
* Batted insulation R-38
* Class-II vapor retarder (required by code)
The wall assembly is similar, but with Rockwool in place of rigid foam:
* Siding
* Rockwool (R-12)
* Peel-and-stick WRB and air barrier
* Sheathing
* Batted insulation R-19
* Class-II vapor retarder (required by code)
My question is about the boldfaced items and the role of vapor retarders here. If the interior vapor retarders were not required, it’s clear (I think) that one would want the peel-and-stick membrane on the exterior of the structural sheathing to be a vapor retarder. But the vapor retarder on the interior is required. If one would make the peel-and-stick on the exterior vapor impermeable, vapor that penetrates into the interior of the wall/ceiling cavity has no easy place to go. So we thought we use a smart membrane on the interior (e.g., CertainTeed) to give vapor a way out when needed. That raises two questions:
1. Is a smart membrane on the interior (where we have to have a vapor retarder) worth it?
2. If we add a smart membrane, would you make the peel-and-stick membrane on the exterior sheathing (both for wall and roof) vapor permeable (e.g., Henry Blueskin VP100) or impermeable (e.g., Henry Blueskin SA)? (I can see arguments either way, preventing vapor diffusion into the cold exterior insulation or allowing some drying to the outside.)
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Andrea, your roof assembly meets the IRC prescriptive requirements for condensation control by putting R25 or greater on the exterior side, but in this case the code is not as clear as it should be. There is a ratio of exterior to interior insulation proposed by Joe Lstiburek that the code council decided was too tricky, so they used the R-25 for zone 6, but what is not written is that they assume nobody will go above code-minimum levels. Lstiburek had proposed at least 50% of the insulation on the exterior in zone 6. The ratio is the main factor governing where in the assembly condensation can occur. With R-38 fluffy insulation, you should have at least R-38 on the exterior as well. Or use R-30 on both the interior and exterior for roughly similar performance to what you propose. If you go with your proposed ratio you might be fine, especially if you use a variable permeance membrane on the interior and control your indoor relative humidity levels. But it's not a particularly safe assembly.
Another note: don't trust published R-values for foam insulation. Cold temperatures and time affect polyiso's performance, and it's over time and at cold temperatures where you need to worry about condensation resistance. I use R-5 to R-5.5/in for polyiso in most cases, though some use even more conservative values.
You can make it a safer assembly by using a vapor permeable membrane on the exterior instead of peel + stick. Pro Clima Mento and Siga Majcoat are good options for this. Rothoblaas also has some options. You may be able to get them as self-adhered, but tape also works for making an airtight assembly.
Your wall assembly has safe exterior-to-interior insulation ratios. A variable permeance interior membrane would be best, to allow drying to the interior when necessary. For various reasons I would not use an impermeable peel and stick on walls, but technically you could. Taped sheathing is usually the simplest way to get an airtight envelope.
As Michael says, your wall meets the minimum ratio of exterior to interior insulation for your climate zone. That alone will help keep the sheathing dry (by virtue of being warm) and— if I'm not mistaken— allows you to use a class-III vapor retarder on the inside (such as painted drywall). https://codes.iccsafe.org/content/IRC2018/chapter-7-wall-covering?site_type=public (table 702.7.1)
Is there a reason you've chosen rockwool for the wall exterior? One of it's possible benefits is that it allows vapor passage. So putting up an impermeable peel-and-stick would largely negate this 'benefit' (though, as Michael says, this benefit isn't technically 'needed' since the condensation control is baked into your R-value ratios). But it's frosting on the cake for less money if you just tape your sheathing.
Michael and Tyler, thank you for your replies.
— Why Rockwool on the walls? Because it is fire resistant, and this is in a wildfire zone. It also allows drying to the outside (especially in summer). (Peel-and-stick membrane is under Rockwool.)
— I understand about heat diffusion and why the ratio of exterior to interior insulation matter for where the dew point is. The challenge with foam on the roof in this case are the (large) snow load requirements, which made roof engineering challenging. We could reduce the interior insulation, but the ratio we have seemed ok in dew point calculations we did.
In any case, you both seem to favor vapor permeable membranes on the sheathing. I’ve heard about Siga and ProClima. Any reason not to use Henry Blueskin VP100 (which is also vapor permeable but more available around here)?
A few clarifying points/questions:
1) >"[Rockwool] also allows drying to the outside (especially in summer). (Peel-and-stick membrane is under Rockwool.)"
Drying of what? If you use an impermeable peel-and-stick on your sheathing, there wouldn't really be a drying path of your wall assembly through the rockwool to the exterior.
My understanding is that this is most important in winter, not summer. As far as I know, moisture drives in summer are typically from exterior to interior. (which could actually be a case in favor of an impermeable membrane, but not sure it's a net benefit, and depends on your siding/ climate, etc) (using class III or smart retarder—instead of class II—might be beneficial to limit moisture accumulation at the interior from any exterior vapor drives that do occur. I'm not familiar with your climate though, you say it's dry...)
2) >"the ratio we have seemed ok in dew point calculations we did."
How was this calculated? Was it calculated specifically with the class II inside vapor barrier? I can't really speak to the robustness of such an assembly, since it gets pretty technical and there's a variety of opinions on 'how risky' it is. It definitely would need a good air-barrier, and a smart retarder would be an advantage there. Others might have thoughts on that.
3) I don't see why the Blueskin wouldn't work, but maybe Micahel has thoughts on that.
4) If you're using relatively impermeable rigid foam above the roof sheathing, there is probably little advantage to having a vapor permeable underlayment there since the foam won't allow much through anyways. Using what's cheaper would probably make sense there.
Dewpoint calculations on roofs are tricky because of night sky radiation, which cools the roof surface below the air temperature. Airtightness also matters, and the building code authors have to assume code-minimum or worse; an airtight house will have somewhat less moist air accumulating at the ceiling. The codes are general for each zone, so local conditions may vary, especially when you get into unconventional assemblies. If you can do an accurate hygrothermal analysis, that will tell you how safe your assembly is--my guess is that it's likely ok. But without doing that, I defer to the building code as a reasonably safe ratio. But not too safe--they estimate the condensing surface to be 45°F, which may be a bit low for some projects.
I tend to spec materials I have used myself and can vouch for. Pro Clima Mento is tough stuff--I've installed a lot of it. I have not personally installed Siga's exterior products but I have used their interior products and spent enough time playing with their exterior products that I know they are also tough. I have not installed Henry Blueskin myself but by all accounts it's a good product. The adhesive is butyl; I generally trust acrylic adhesive to hold up better over time based on what I've heard from building scientists, but I don't have a study at hand to support that position. I have heard that Blueskin is somewhat tender; you have to be careful with ladders etc., but that shouldn't be a deal breaker. I don't believe it is intended for installation on roofs.
You can order Siga easily here: https://sigatapes.com/. You can order Pro Clima easily here: https://foursevenfive.com/.
Rothoblaas is new to the North American market but they have a range of high quality products, coming from Europe, who are generally ahead of us on these types of materials: https://www.rothoblaas.com/products/airtightness-and-waterproofing. You can contact them to find a sales rep in your area.
One thing that might help a bit. From energy use point of view, roof and walls are pretty close to being the same with a bit more loss through a roof through night time radiation. When you are building with expensive insulation like you are, there is very little benefit in going nuts on roof insulation.
Generally the best value is making your roof and wall R values about the same with the roof slightly higher.
You can also pass code on U value basis for your roof, which significantly reduces the amount of insulation you need. There is nothing wrong with not fully insulating the rafter cavity as long as the insulation is pushed up against the sheathing.
For example, in your case, you could go with R20 rigid insulation above the roof and R19 fluffy bellow.
Generally a full peel and stick for the walls is not worth it, you are better to tape the seams and install a standard house wrap. This preforms just as well and much cheaper.
If you do decide to go with the full peel and stick, the VP100 would work just as well. You can simplify the install significantly by going with vertical strips. There is no special overlap or details needed to do this and much quicker install.
Our code allows for the warms side vapor barrier to be at the rigid insulation boundary provided you have enough rigid for condensation control, don't know if there are such provisions in yours. Either way, the interior smart vapor retarder doesn't hurt. The facers on fiberglass batts can also work as a smart vapor retarder.
+1 on #1 and #5. With low, questionable or at-the-limit foam ratios, things to make it work better/well are:
1) interior side air sealing + testing
2) smart Class II vapor retarders
3) limit interior humidity when very cold
Some of this has been quantified for walls, but I'm not aware of the same for roofs.
Thanks all, very helpful!
Re drying of wall to outside through Rockwool: I was just thinking of rain; vapor drive is not an issue here in summer.
Dew point calculations were just that: given a reasonable relative humidity and interior temperature, where would the dew point lie in the wall/roof when it’s cold outside (O F), assuming a temperature profile linear in R between interior and the point in the wall? Nighttime radiative cooling is not an issue for us: 10’ of snow are not uncommon. Basically, when it’s cold, there’s snow on the roof.
One final question: to make the roof assembly safer, would it make sense to put, say, 1” of closed-cell spray foam under the roof sheathing, plus batting? That would provide some extra safety in moving the dew point to the exterior of any condensing surface.