Cathedral roof insulation zone 5
pmny
| Posted in Energy Efficiency and Durability on
Hi,
Working on a cathedral roof in upstate NY (zone 5) and have a question regarding insulation, and how it relates to moisture and venting requirements.
The plan is:
2×12 rafters at 16″ o.c.
This would be an R-54 Assembly
1″ Foil-faced iso-board continuous on the board roof deck R-6
1/2″ unfaced iso-board in between the rafters, R-3
7-1/4″ & 3-1/2″ mineral wool batts in between rafters, R-30 + R-15
Is there an issue with closed cell foam being used on both sides of roof deck?
Should I have venting on either side of the deck before it makes contact with the foam?
Should I use a smart membrane as a vapor barrier under the rafters?
Thanks,
Paul
Paul
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Paul,
- Practically and to meet code, if you are not venting the roof it needs a higher ratio of impermeable (foam) insulation to permeable (batts). See this link: https://www.greenbuildingadvisor.com/article/five-cathedral-ceilings-that-work
- If you are venting the roof, there is no point using any insulation above the sheathing.
- Rigid insulation over the roof sheathing adds complexity and expense. It's only worth doing if you are using a significant amount. I would either eliminate it, or bump it up from the 1" you are proposing.
Thanks Malcolm.
I'm not tied to any particular approach, but with 12" rafters, can I put enough insulation (without using closed cell spray foam) between them to reach R49 with underside venting?
When you say bump the exterior foam up, is it something like 4" of rigid on the exterior?
Is there a scenario where I could vent on the exterior of the roof only under the standing seam metal roof, and what breakdown would that be to reach R49+?
Thanks again, I appreciate your advise.
Paul
2x12 is actually 11.25"
If you have a 1.25" ventilation channel, plus a 1/2" thick home made baffle made out of XPS Foam (baffle itself is R3), you have 9.5 inches left to work with, which is exactly what you need for R38 rockwool. This results in a total of R41. You'd be hard pressed to really notice the difference between R41 and R49.
The other option is a cut and cobble solution with Polyiso foam. Same 1.25" vent channel and 10 inches of polyiso gets you to r60 on paper, although in the cold with age derating it could be considerably less in practice (https://www.greenbuildingadvisor.com/question/performance-of-polyiso-exterior-foam-at-cold-temperatures#:~:text=So%20at%20a%20mean%20temp,R12+%20for%20the%20seasonal%20average.)
The way I see it spray foam is an act of desperation and poor planning best avoided.
Walta
Walta,
There is no spray-foam in his proposed roof assembly.
Is this a new build or a retrofit?
Hi,
It's a new build, and I'm trying to get the best option to get to R49+.
Thanks,
Paul
What is below the rafters?
You need either (1) a lot more rigid foam ABOVE the sheathing, (2) closed cell spray foam UNDER the sheathing, or (3) a vented assembly.
I would go with (3), and build this as a vented assembly, which is the safest way to go. I would use 1x2 furring strips tacked to the inside edges of the rafters tight to the underside of the roof sheathing, then 1/4" waferboard over those. That would give you a 1.5" vent channel. This vent channel needs to be continuous from soffit to ridge, each of which needs to have vents, and you need a vent channel in every rafter bay. This leaves you with 9" in the rafter bays to fill with insulation. You can get a bit over R38 in there if using mineral wool, and if your target is R54, I'd use 2" foil faced polyiso under the rafters with the seams taped, which will add R13 to your assembly. This will get you up to around R51-R52, which is pretty close to your target while also being relatively easy to build. If you need more R value than that, then I'd use fiber faced polyiso under the 1x2s instead of 1/4" waferboard, which gets you a bit more R per inch over mineral wool, and mess around with the ratios of baffle polyiso to mineral wool space to arrive at the total R value you need. You would not put any insulation ABOVE the sheathing if building like this, since you don't want to insulate the exterior side of the vent channel.
Bill
Foam free R60 isn't difficult with a vented cathedral roof/ vaulted ceiling, r53 is, well, easy!
2x12 rafters and a bearing ridge. Mento 1000 fastened to the topside of the rafters, 2x furring strips above the Mento, roof deck applied to the furring strips. That'll leave a 1.5" deep vent channel above the rafters. Mento can be treated as the air control layer, or that layer can be inbound of the rafters (ex. Intello Plus, which you need anyway as a moisture control layer). A 2x12 will allow for R23 + R30 mineral wool = R53. Build the rafter depth out with a Bonfiglioli approach to allow for 2 layers of R30 mineral wool = r60 and minimal thermal bridging.
This assembly can also be dense packed. The mineral wool, however, you can install.
The tactic you initially described is a little spooky and poor performing.
You could just use "regular" house wrap instead of the Mento 1000 here -- it's not a very demanding application, since it's entirely enclosed within the larger assembly.
I'm not a fan of the Bonfiglioli method. What I'd do instead would be to run a layer of 1/2" or 3/4" foil faced polyiso under the rafters with seams taped. Now you have an interior side vapor barrier, to help keep moisture out of the ceiling/roof assembly, and you have some continuous insulation to cut down on thermal bridging. 1/2" or 3/4" isn't a problem to fasten through when attaching drywall, so it doesn't really complicate other parts of the construction process, and it's easy and cheap to put up.
Aside from those two points, I don't see a problem doing what you describe if this is a new build and not a retrofit or renovation.
Bill
Thank you Bill,
Taking your advise and mixing it with climbing_carpenter, hopefully i got it right.
I drew up a cross-section of the roof, if you don't mind taking a look and seeing if it's correct.
Much appreciated.
P.
Sorry, forgot to attach earlier!
Paul,
The house-wrap / foam baffles are really an either or choice. The advantage of using the house-wrap is to move the vent space above the rafters giving you more room for insulation, and avoid having to cut baffles in each rafter bay. Adding the 1/2" foam negates that advantage, and decreases the drying ability. The foam would also need backing for attachment.
The 2"x furring on top of the house-wrap is usually either 2"x3" or 2"x4" . Using 2"x2"s risk splitting and makes the sheathing more difficult. https://www.greenbuildingadvisor.com/article/building-a-vaulted-high-performance-and-foam-free-roof-assembly
Hi Malcom,
Can I ask you to take a look at this revised version (preferred by the GC) which has a cold roof approach with no internal venting.
Does it work?
Thanks again.
Paul
Paul,
You are missing a layer: you can't install metal roofing on vertically oriented strapping.
If you decide to go with venting above the sheathing, plywood would be safer than ZIP, as any moisture has to diffuse though the sheathing before it can be removed by venting, and plywood becomes more vapour-open if it gets damp.
Practically there is a case to be made that over-venting isn't much different than venting below the sheathing when using plywood for baffles. That's an argument that Steve Baczek has successfully argued to get a similar over-vented roof assembly approved - but be aware, it isn't code compliant, so you may have to make the case for it to your BI. The IRC doesn't allow permeable insulation to be in contact with the underside of the sheathing unless there is impermeable insulation above.
I have reservations about it over the long term. You are asking the sheathing to go though repeated wetting and drying cycles. That is the same with plywood baffles, but the consequences of them losing structural integrity are very different.
To add to Malcolm's concerns with 2x2s splitting, I find that 2x2s are especially bendy and twisty -- it's much harder to find straight and square 2x2s compared with the larger 2x3 and 2x4 boards. I'd probably try 2x3s here, unless 2x4s ended up being cheaper, which they may be.
The housewrap is a sort of zero dimensional barrier to define the vent channels. There is no need for polyiso below it -- the housewrap does not need any support here, and it doesn't matter if it bows a bit when the insulation is stuffed in. Code requires a minimum 1" depth for the vent channel, and the 2x framing gives you 1.5", so a little bowing down to 1-1/4" or 1" in the middle really doesn't matter. Vent channels don't have to be perfect. Note that the polyiso UNDER the rafters is still useful though. I especially like having some sort of vapor barrer (which is the foil facer on the polyiso here) in these assemblies to limit how much moisture can get up into the assembly, since cathedral ceilings always need more attention to detail than "regular" vented attics do.
BTW, if your GC tries horizontal strapping instead of vertical to address the issue Malcolm mentioned regarding the metal roof, keep in mind that horizontal vent channels DO NOT work reliably. You need vertical vent channels to allow for convection airflow. You could potentially do two layers of strapping at right angles to address this, but I share Malcolm's concerns with the moisture cycling plywood: I expect that over time you'd have delamination issues and warping of some of the plies, which would be bad for roof assembly structuraly.
Bill
Thanks C_C,
Much appreciated.
I'll post what I've drawn up as a cross-section of the roof and see if I'm on the right track.
P
Re: post 16. NO inbound ISO between rafters. Further, 1/2 ISO isn't worth the effort, even if applied to the inside face of the rafters. It's effectively a vapor barrier, that's it.
I'm a fan of Bonfiglioli configurations due to price point, this approach is extremely cost effective versus other high(er) performing assemblies. This is based on years in the field, building and rebuilding many houses with various approaches. I do agree with Bill: thick, inbound ISO is another option, albeit more expensive than building rafter bays out. You'll still need furring strips outbound of the house wrap to create vents. Inbound ISO, however, eliminates the ability of an assembly to dry to the interior. With careful exterior detailing, this shouldn't be a concern. Continuous inbound ISO will eliminate thermal bridging through the rafters- think 2 inches (r12). A Bonfiglioli configuration will help with bridging too. If you decide on inbound ISO, look for a reclaimer or distributor who deals in factory seconds. You'll save 50% or more. One more note, you'd apply strapping to the inside face of the ISO, if you go that route, to allow for drywall fastening. This airgap will add r1 to the assembly- with r53 mineral wool, r12 ISO, and an r1 air gap, you'd land at mid r60s.
Note: not all house wraps are created equal. I specified Mento because it's robust. Tyvek in this application isn't appropriate- whatever you use needs to survive the construction process, especially if you dense pack.
The primary purpose of the foil faced polyiso would be to establish a reliable air/vapor barrier. The vapor barrier here IS important though: I would argue that inward drying in CZ5 in a vented roof assembly isn't an issue, as it's much more important to keep moisture OUT of the assembly in the first place. With vapor drive being from indoors to outdoors over the heating season, that interior vapor barrier helps to prevent moisture from getting into the assembly in the first place, and the vent channels will draw any moisture out that does get in.
I agree 1/2" polyiso isn't a lot of additional R value, but it's better than nothing, and relatively cheap and easy to install. Even a traditional poly vapor barrier could be used here, but the foil faced polyiso is more robust.
Bill
Hi Bill,
Firstly, thank you, Malcolm, and climbing carpenter for all the advice and patience as I try and get this right, I really appreciate all the comments.
I redrew the cross section to (hopefully) reflect the advise you gave, so please let me know if this looks like it would work.
The Foil faced Polyiso under the rafters takes the place of a vapor barrier.
Thanks,
Paul
Paul,
That section looks good.
I understand your builder's reticence. When I first saw that roof assembly I was a bit worried thinking about how to build it, but both Josh Salinger and Michael Maines have reassured me it isn't as difficult installing the membrane and strapping as it seems. The finished product is a very well performing roof.
Thanks Malcolm, that is a relief to know it works.
You mentioned in an earlier response that "over-venting" could be an option, so I'm wondering, as a backup plan, what would make that scenario work.
Would it be the case of 6" or more of exterior rigid foam and just partially filling the rafters with mineral wool to keep the ratios in check for Z5, or is there a hibrid alternative closer to what I had proposed in my second option I sent yesterday?
Thanks,
Paul
There is no need to worry about insulation ratios when all the insulation is one one side of the sheathing, is of the same type, and is part of a ventilated roof assembly.
If you're looking for a way to save some money on mineral wool, I suggest looking into high density fiberglass batts, which will get you similar performance in terms of R value, but at lower cost.
Bill
Paul,
Over-venting was my poor way of describing venting above the sheathing. That's what your section in post #16 showed - and in my rely I described the code difficulties, and need for either two layers of strapping, or a second layer of sheathing to support the metal roofing.
Hybrid roofs with Insulation above and below the sheathing are complex and expensive. You can do it as you suggest and use the right ratios, but that's an assembly I'd describe as one of last resort in new construction.
Taking a step back. The problem you are facing - that the depth of the roof structure limits your insulation choices - is the most common one that crops up in the Q+A. It should never occur on new construction, and should be solved at the design stage by choosing your rafter or truss sizing based on the depth necessary to use a simple, reliable assembly that meets the required R-values - not picking them based on how far they can span.
If you begin with that premise, you don't end up having to compromise, or use a risky or expensive insulation strategy. If you want or need more batt insulation than you can get in a 2"x12", I'd back up and begin again from that starting point.
Thanks Malcolm,
I think the GC is good with the interior vented version, so thanks again for all your help.
Until the next time...
All the best,
Paul
Looks OK to me too. I would specify that the seams between sheets of polyiso are to be taped. I like to use Nashua 324A foil tape for this, because it sticks well and looks good on the foil facer, but any seaming/flashing tape should work fine too.
I also see no issue using a beefier housewrap material here, as stronger material will only help you in terms of making damage less likely. Just be sure whatever you use is VAPOR permable (no polyethylene sheet, etc.).
BTW, when I put polyiso under rafters/joists like this, I use a sharpie marker to mark the centerline of each joist on the edges of the polyiso as put each sheet of polyiso up. This lets you see the joist locations easily when you're putting up strapping later. Note that you usually can't get a chalk line to work well on the foil facer, so it's often better to just use straight edges and sharpie markers here.
Bill
Thanks Bill,
Happy youre good with the set up, I'll make sure your suggestions are implemented.
All very helpful!
Cheers,
Paul
Hi Bill,
It's not about the expense, it's that the GC is trying to keep the depth of roof interior to the 12" rafters (with mineral wool in between), some 3/4" strapping, and the 3/4" brown boards as the finished ceiling. Pushing the venting under the standing seam roof on the exterior.
Trying to make that work, that's acceptable for zone 5 is challenging, but is there a way without really deep rigid foam on top of the roof deck?
Thanks,
Paul
You can use thicker rigid foam UNDER the rafters too. You can put the R value in the cavities or under the rafter with this assembly. Mineral wool is about as good as you can do for R value unless you want to step up to spray foam (not really needed here), or cut'n'cobble loads of polyiso (which will have you hating life).
I would look into TJI joists as rafters here. You're sort of site building half of one already by tacking a 2x3 on top of a 2x12. The TJIs have several advantages in your proposed assembly:
1- You can get them sized for just about any reasonable span, and they're going to be a lot straighter and squarer than sawn dimensional lumber like 2x12s will be.
2- You can tack 1/4" waferboard under the lower edge of the top chord, which is a 2x piece of framing lumber like you were planning to add to the top of the 2x12s anyway. You probably wouldn't want to try housewrap for this due to the need to staple it a LOT to get it in there right.
3- You'll have less thermal bridging through the relatively thin OSB web in the TJI joist than you will with the 2x12.
You can size the depth of the TJIs to handle the insulation you need to get up to your target R value too. Dense pack cellulose becomes an option with this assembly too.
You can't put insulation above the roof deck with a vent channel below the deck. The vent channel effectively "cancels out" that above-deck insulation, because the vent channel is circulating outside air.
Bill
CC,
I agree - regular Tyvek is too fragile. I'd use Tyvek Commercial. It's hard to tear, and is 28 perms.
I just wanted to say this was a good discussion. Thanks.