Attic insulation advice/feedback
We recently moved into a 1952 Cape Cod home and will be remodeling the entire 2nd floor attic/living space so now is the time to get the insulation done right. As a result, I have been doing considerable reading on the topic of insulation and would greatly appreciate some advice on my current plan to “cathedralize” and insulate this space.
The plan:
Vent the soffits.
Create a 1 inch deep soffit to ridge ventilation channel in each rafter bay using .5 inch rigid XPS foam board sealed in place with caulk/spray foam.
Place R15 Roxul rock wool insulation batts in the remaining 3.5 inches of rafter bay space.
Install two 2 inch layers (4inches total) of polyiso foam board on the underside of the rafters- seams staggered and taped. Foil facing the living space.
Drywall spaced .5 from the polyiso with furring strips screwed through the polyiso into the rafters.
Install a ridge vent (looking for suggestions on brand/type)
Relevant information/considerations:
Total 2nd floor square footage equals 960- all conditioned space
2×6 roof rafters
All combustion appliances are in the basement
Current roof sheathing is 1×6 boards which may get sheathed over with OSB whenever the roof gets replaced. (I know not to insulate above AND below the sheathing)
12/10 roof pitch, no dormers on the front, single shed dormer spans nearly all the backside of the house.
Williamsburg VA- Climate Zone 4A
Professional (even DIY) application of spray foam exceeds our budget.
Specific Questions:
Should I increase the depth of the ventilation channels to 1.5 inch and compress the 3.5 inch rock wool batts down to 3 inch?
Am I using the appropriate materials in the appropriate location?
Do you have any suggestions to improve this design?
I have heard good things about the Roxul product, can anybody comment on whether pests (mice/bugs) like to nest in it like the pink stuff?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Chris,
Q. "Should I increase the depth of the ventilation channels to 1.5 inch and compress the 3.5 inch rock wool batts down to 3 inch?"
A. Yes, I think that would be a good idea.
Q. "Am I using the appropriate materials in the appropriate location?"
A. Your plan will work. The most appropriate approach would be different, however. The best way to insulate this type of roof is to install a thick layer (or layers) of rigid foam above the roof sheathing, followed by 2x4s on the flat to create vent channels, a layer of plywood or OSB, roofing underlayment, and new roofing.
Q. "Do you have any suggestions to improve this design?"
A. See my answer to your last question.
Q. "I have heard good things about the Roxul product. Can anybody comment on whether pests (mice/bugs) like to nest in it like the pink stuff?"
A. I have no direct experience on this matter.
The half-inch XPS in the stackup isn't really necessary. Using housewrap as the exterior air-barrier to the R-15 Roxul would leave you a 1.5" nominal vent channel an you wouldn't have to compress it. The difference in thermal performance would be negligible. That would give you R15 between the rafters, and R24-R25 for the polyiso, for a center-cavity R of about R40.
If you want to leave open the option of later putting some rigid foam above the roof deck and filling the vent channel with blown insulation, use unfaced EPS on the interior side, not polyiso. Polyiso normally comes with low permeance facers to bring the vapor permeance under 1 perm even for fiber-faced roofing foam, but foil facers are true vapor barriers, which you can't have on the interior side of an unvented roof assembly since it would create a moisture trap. At 4" thickness unfaced Type-II EPS (1.5lbs per cubic foot nominal density) will have a vapor permeance between 0.5-0.75 perms, a Class-II vapor retarder which is a sufficient seasonal drying path, but it is also a sufficient vapor retarder to protect the roof deck from winter time moisture drives in a 4A climate.
While the code is silent when it comes to Class-II interior side vapor retarders on unvented insulated roof assemblies, it's not very risky, especially if the shingles are dark. As long as they are not "cool roof" shingles (with pigments selected for high infra-red emissivity and moderate to high solar reflectance) it should be fine to install R23 Roxul in the rafters and 4" of Type-II EPS (R16.8) on the interior, as long as you take pains to make it air-tight, and you would again be at about R40 center-cavity.
If it's too difficult to air seal each layer, installing a thin membrane type smart vapor retarder such as Certainteed MemBrain between the rock wool & foam, between the foam layers or between the interior side foam & gypsum, as an air barrier. That will not reduce the drying capacity at all, but will reduce the wintertime moisture diffusion rates slightly more than what the EPS does on it's own, since MemBrain is a class-II vapor retarder when the interior side air is under 35% RH or so, which it likely would be during the coldest part of the winter. (With R23 rock wool on the exterior of R17 EPS you have plenty of dew point margin at the air barrier even if the air barrier is at the foam/fiber boundary.)
Then when you re-roof you can add 4" of EPS or 3" of polyiso above the roof deck and be slightly above the R49 code-min, coming in at R57-58, and compliant with the IRC chapter 8 prescriptive for R15+ on the exterior of the roof deck, which would otherwise have allowed you to use a Class-III vapor retarder (such as standard latex paint) instead of Class-II (4" of EPS,&/or MemBrain) on the interior.
http://publicecodes.cyberregs.com/icod/irc/2012/icod_irc_2012_8_sec006.htm
Mice will nest in rock wool as well as fiberglass, but it takes more work for them to chew through high density fiberglass or rock wool than with low density R11s & R19s.
If you're planning to re-roof any time soon, installing just the R23s and only 2" of interior EPS (R8.4) MemBrain now, and R15 foam above the roof deck later would get you to about R46-R47 at center cavity, but would be still be code-compliant based on U-factor, (sub-u0.026) since the thermally bridging rafters would then be thermally broken with R23 of rigid foam. That would give you more interior side head room, and you could get there with just 3" of roofing polyiso (or a 3.5" polyiso nailbase panel) on the exterior.
Wow! Tons of great info there. We do plan to re-roof in the next year or two so it makes sense rethink my approach and plan for insulation above the roof decking- great advice Dana.
I am being showered by critter droppings as I remove these old r11 pink fiberglass batts and would like to rplece them with something that will discourage future infestation. What are my best options?
Chris,
Either dense-packed cellulose or spray foam are better than fiberglass batts if you are trying to discourage rodents.
The borate fire retardents in cellulose are eye irritants, which limits but doesn't totally eliminate rodents from picking at it.
A few years ago I had a squirrel living in my garage for about 8 months, who chewed into a bag of cellulose stored there and made a bit of a mess. But he didn't pull out more than a baseball-sized volume, and didn't opt to nest in it. (Instead he chewed a 2" diameter hole through some ship-lap planking into the top of a cellulose insulated stud bay that had not fully filled due to a framing obstruction. I used the hole to dense-pack that void with cellulose, stapled some hardware cloth over the hole and sealed it with expanding can-foam. He wasn't fully evicted from the garage until a couple months later, after having re-nested in a narrow empty cavity in the garage door framing.)
R11s are very low density, barely more than half the density of R13s, which make it very easy for rodents to create pathways and nests. High density R15s are about 2x the density of R13s, and firm enough that they can't just push their way through, compressing the insulation as they go, which they are able do with R11s and R19. Instead they have to chew their way through, which they can do, but it makes it a bit less attractive. The same is true of dense-packed cellulose (3lbs per cubic foot or more.). How much less attractive is debatable, given that many rodents normally tunnel through topsoil to create predator-safe pathways & nests. But you don't have to make it super easy for them with low density easily compressible fiber.
I've read internet accounts of rodents creating insulated caves in 2lb polyurethane foam too. Nothing is really immune, except for foamed cemeticious products like AirKrete(?).
While you have the stud bays open caulk the studs & bottom/top plates to the sheathing, and don't forget to caulk the bottom plate to the subfloor too. (Acoustic sealant caulk is a pretty good option here.)
The clear consensus is that adding insulation above the sheathing is the preferred solution.
Since we are planning on a new roof in the next 12-18 months anyway, it seems to make sense to plan for insulation above the sheathing in the future. Unfortunately my quest for answers seems to have led to more questions...
If I am reading Dana's reply correctly it outlines 3 options for adding rigid foam above the sheathing at a later time. 1- vent and insulate as I had planned substituting type II EPS for the polyiso due to permeance and moisture concerns and then blow dense packed cellulose to fill the vent cavity when I am ready to add the foam above the sheathing. 2- Skip the venting and use r-23 roxul filling the entire 6 inch rafter bay while air sealing the 4 inches of interior EPS very well and/or adding a smart vapor barrier. 3- Similar to option 2 but only using 2 inches of interior EPS to allow a bit more headroom.
I like options 2&3 because my roof currently has gable vents and adding a ridge vent with corresponding soffit vents plus the vent channels represents wasted money and labor once the insulation is added above the sheathing.
In options 2 & 3, where is the best location in the stack to incorporate the smart vapor barrier? and does the performance of these vapor barriers degrade over time?
With the new roof being at least 12 months out, how risky is having an unvented roof assembly with no insulation above? (current shingles are medium gray in color) I know Dana addressed this question to some degree but Dr. Joe Lstiburek makes the case that it is pretty close to impossible to do an adequate enough job of interior air sealing to keep interior moisture out of the ceiling well enough to prevent sheathing rot. (http://www.buildingscience.com/documents/insights/bsi-043-dont-be-dense).
EPS has significantly lower R value per inch relative to XPS. I am assuming the recommendation to use EPS on the inside rather than XPS is based on the lower permeance and the possibility of creating a moisture trap. Am I understanding this correctly?
My current roof sheathing is 1x8 planking, does this have any implications relative to the insulation approaches outlined above?
Thanks so much for sharing your expertise! Chris
Chris,
Q. "Where is the best location in the stack to incorporate the smart vapor barrier? and does the performance of these vapor barriers degrade over time?"
A. Building codes in the U.S. do not require an interior vapor barrier, although some codes requires a vapor retarder (a less stringent barrier) on the interior.
If you need a vapor retarder, you can use the kraft facing on fiberglass batts or vapor-retarder paint. In all cases, ceiling assemblies need an air barrier. For more information, see Do I Need a Vapor Retarder?