2×6 rafter insulation
Hi,
I’m sorry if this question feels repetitive but in my research I haven’t found any conclusive information.
I’m working on a renovation project in Vermont (Zone 6) that has a cathedral ceiling second floor framed with 2×6 rafters. There’s very little headroom as is so furring to the interior wouldn’t work. The house also has a ~10 year old standing seam roof.
I’m trying to come up with a way to insulate the roof. The best plan that I’ve come up with is from sheathing to interior: 1″ furred air gap, 2″ EPS foam, R15 Rockwool with an interior intello air barrier. This wouldn’t come close to meeting code for insulation, but I’m wondering if someone could help me with how risky of an assembly this would be. My thought is that the danger would be condensation forming between the Rockwool and the foam, but that the intello would keep moisture out and allow drying to the inside.
Does that seem like a valid assessment, and is this a horrible idea?
The only alternatives I can see would be to life the roof to create more headroom, which would be a huge project, or to remove and replace the standing seam roof and insulate from the outside, which again would be a pretty large project.
Thank you,
Niels
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Vented roofs are much more forgiving of moisture, since the ventilation removes the vast majority of the moisture that manages to get up into the roof assembly. Due to that vent channel, you should have no problems with condensation in your assembly, at least not enough to cause any issues. You will get a small amount of moisture migration through the framing "around" the EPS, and with that interior side vapor retarder (intello), you'll have very little moisture to deal with in the asembly.
You can get a bit more R value in there using all polyiso, which, with 4.5" of space, (a 2x6 is 5.5" deep, and you'd lose 1" of that for your vent channel), you'd be able to get about R29 worth of R value in there. Your proposed 2" EPS + R15 rockwool won't fit, because it would be 5.5" deep -- leaving no room for your 1" vent channel -- and it would get you about R23 worth of R value.
What I would do would be the following:
Furr out the 1" vent channel as you originally planned. Normally I'd want to see a 1.5" vent channel, but you're so close on space here, I'd stick with the 1" vent channel. BE SURE you have soffit and ridge vents so that you get AIRFLOW through those vent channels! That's VERY IMPORTANT here! I'd rip strips of 1/2"-3/4" plywood to make the furring strips. Note that you want to use a good straight edge to cut those if using a circular saw, not just some framing lumber that might not be as straight as it appears. Banana-shaped furring strips are annoying. Ask me how I know...
Put in all polyiso for max R value (~R29), seal it in with canned foam. This is a lot of labor, but the only want to get max R value without resorting to spray foam. You could go with all EPS to save money, but that would only get you about R19. Stuffing fiberglass batts in there would be about the same, maybe a bit less.
Put as thick a layer of polyiso as you can over the underside of the rafters, then drywall over that. If you can only manage a 1/2" layer of polyiso, that adds just shy of R3 to the assembly, but it's *continuous* insulation, which helps with thermal bridges, and that's very helpful in a depth-limited assembly like this. If you can fit 1" of polyiso, that adds about R6 to the assembly. If you go with 1" polyiso under the rafters and all-polyiso between the rafters, your assembly is up around R35 or so, and a decent percentage of that is continuous insulation, resulting in a much better performing assembly.
Tape the seams of the polyiso under the rafters, which now gives you and air AND vapor BARRIER on the interior, so you can skip the intello to save some money.
Bill
I second Bill’s recommendations.
I dealt with a similar home in Vermont with 2x6 rafters. I added the 1” air space beneath the roof sheathing, and I furred out the rafters with 2x stock.
I wish I had used all polyiso in the rafter cavities. Instead I used an inch of rigid with high-density batts to fill the remainder of the cavities.
The house gets some pretty intense icicles, especially under sunny conditions with temps in the 20s or less. I did not find any damage related to ice damming, so that was a huge relief. Ice & water shield was installed when the roof was re-shingled.
In the future I will use polyiso to completely fill the cavities, and use a layer of rigid attached to the bottom face of the rafters (to deal, somewhat, with thermal bridging).
Thank you both, this is really helpful and takes a load off my mind for the next decade or two. I'm inspired by your detailed and complete answers!
I have one follow up question, more for my own knowledge––I've read that poly-iso drops in r-value at cold temperatures more than other rigid foam. Is that a consideration here, especially with the air gap? Or will heating the space generally keep the foam above the temperature at which that would matter?
The "polyiso loses R value in the cold" thing is overrated, and based on one paper from Dow years ago that they were using in part to help pitch their Thermax product (which is pretty expensive polyiso, although also very good). In reality, you don't see a lot of difference, and it's only in the outer layers anyway since the material effectively insulates itself. One of these days I'll get around to measuring this effect, since I have the equipment to build a test rig, but I don't have the time to build the test rig and do the testing right now.
Basically even if it's really cold outside, the temperature within the polyiso isn't that cold except near the outer surface. As you get deeper inside the material, the temperature gets warmer and the issue goes away.
You could save money using reclaimed or factory second polyiso here if that material is available in your area.
Bill
Just seeing this reply, thank you!
I just built the exact same thing in Zone 5, except I used 1" of EPS because, as already mentioned, we only have 5.5" to work with. I also used (faced) fiberglass R15 because of cost.
I did a ton of research and also posted my plans here a few months ago. Your proposed assembly should be safe, but I would STRONGLY recommend keeping the EPS thickness to a minimum to ensure vapor permeability.
Last, I don't like the idea of doing the whole thing with polyiso for three reasons:
1) You're likely going to have to use multiple layers, and since polyiso isn't permeable like EPS is, especially at thicknesses greater than 1", you could trap moisture
2) It's expensive
3) You're in a cold climate, and polyiso loses R value quickly in the cold. It's a great insulation during AC season, but not during heat. It also slowly loses r value over time.
You could also use GPS foam to make your 1" baffles instead of EPS - it'll get you R5 instead of R4 in that inch, but it's kinda hard to find. I was able to source EPS locally at $12/sheet, GPS was a long-lead special order item so I skipped it.
Rergarding your points:
1- You won't trap moisture between layers, since there is no air space between the layers for moisture to accumulate in. Also, it is the facer material that is impermeable with polyiso, if using foil-faced polyiso. If you use fiber or kraft faced polyiso, the polyiso insulation itself is actually fairly vapor open, I think even more so than EPS (but don't quote me on that last part :-)
2- It's not expensive if you can get reclaimed or factory second material. Polyiso is commonly used to insulate commercial roofs, and is often reclaimed during commercial re-roof projects. If you have a reclaimer in your area, you can usually get reclaimed polyiso for less than new
EPS.
3- This issue is very overrated. See my earlier post, above. Polyiso does NOT "quickly lose" R value when it's cold. It has to get very cold, and even then it doesn't lose all it's insulating value. The usual way to deal with this is to derate the R value of polyiso to around R5 per inch, which is still better than EPS (which is about R4.2 per inch).
Note that there will be enough of a path for moisture to escape going through the wood framing as long as the interior air/vapor barrier is good (which would certainly be the case with an interior side of foil faced polyiso with taped seams). Only very, very small amounts of moisture would have to be dealt with within the assembly this way.
I see no issue using stacked EPS or GPS aside from lower overall R value for the assembly constructed this way. I would NOT prefer fiber insulation (batts) over layers of rigid foam here though, because the batts are much more likely to be a problem with moisture than are layers of rigid foam materials.
Bill
I'm with MIzone5Woes. If you are going to do continuous baffles, those need to be from something permeable or thick enough for condensation control (about same R value as batts).
I would offer my alternative assembly for 2x6 that I've used. Continuous PERMEABLE baffles (think foam ones work, the plastic ones don't). Insulated with either R21 batts compressed, they will touch the roof deck beside the baffle which is fine as you still have the vent gap down the middle.
Underneath a layer of foil faced rigid. Your choice of thickness and type, they all work. Tape the seams of the rigid for air barrier, the rigid is also now your vapor barrier. Install drywall directly over the rigid or install strapping than drywall.
This gets you a decent R value assembly with minimal layers. Depending on the thickness of the rigid, you get a decent thermal break for the rafters.
Thank you!
Deleted
Deleted