Achieving R-49 with fluffy batts in 2×10 bays +R-17 above sheathing
The City of Kirkland, WA (Marine 4 / 4C) rejected my design (pdf attached) saying that the polyiso above roof deck “does not count” toward required R49 per IRC and that I must use closed cell spray foam for an unvented roof. It is my understanding that IRC 806.5.1.2 (included at end of post) does indeed allow fluffy insulation under roof sheathing with rigid foam above sheathing for condensation control. And I’ve read just about everything Martin has written on the subject, so I feel confident it CAN work.
Before I respond to the city I want to see if maybe I’m missing something else, and not just for code compliance but also proper building science. For example:
1) Am I misunderstanding the IRC? For example is the R15 above sheathing ONLY serving as condensation control without contributing to the R value of the overall assembly? I can’t imagine why that would be the case but thought it worth asking.
2) Am I missing an air barrier or vapor retarder detail that invalidates my design? Is unpainted gypsum ceiling with silicone sufficiently airtight?
3) Is my ratio of fluffy-to-rigid ok for condensation control? I have R32 below sheathing (value of R38 fiberglass batts compressed into 9.25″ rafter bays) and R17.4 above (3″ of polyiso). That appears to meet the table in IRC but doesn’t match Martin’s table precisely (can’t get R38 fluffy into my 2×10 joists).
4) Does my unvented soffit detail make sense, e.g. is “taped cdx” sufficient or do I need to carry tyvek down over cdx? Is sealant in right place (I borrowed the idea of intentional gap from floors overhanging unconditioned space).
Any problems in areas I forgot to ask about?
BTW in case you are scratching your head over the polyiso tapering from 3″ to 7″ … I designed the roof to have 3 slopes. The main monopitch is set by rafters at 3/4″. That monopitch slopes TOWARD the middle of the living room where it meets a knee wall/cricket created by 2 gabled roof planes (at 45 degrees to the monopitch) which diverts water to one side or another. Rather than depend solely on the diverter I added a secondary 1/4″ slope to the sides so that much of the water will fall into side gutters thus reducing volume dependent on the diverter. I’m open to critique of that design element as either being over-complicating or not sloped enough to be worthwhile.
Here’s the IRC section I’m referencing:
5.1.1. Where only air-impermeable insulation is provided, it shall be applied in direct contact with the underside of the structural roof sheathing.
5.1.2. Where air-permeable insulation is provided inside the building thermal envelope, it shall be installed in accordance with Section 5.1.1. In addition to the air-permeable insulation installed directly below the structural sheathing, rigid board or sheet insulation shall be installed directly above the structural roof sheathing in accordance with the R-values in Table R806.5 for condensation control.
TABLE R806.5 INSULATION FOR CONDENSATION CONTROL
| MINIMUM RIGID BOARD | |
| 2B and 3B tile roof only | (none required) |
| 1, 2A, 2B, 3A, 3B, 3C | R-5 |
| 4C | R-10 |
| 4A, 4B | R-15 |
| 5 | R-20 |
| 6 | R-25 |
| 7 | R-30 |
| 8 | R-35 |
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Dave,
You're right. The building inspector is wrong. It's a simple as that.
If you encounter a stubborn, ignorant official, you may end up in a tight spot. Sometimes there is a way to figure out whether the official who doesn't properly understand the code has a more knowledgeable supervisor. Good luck.
HI Dave -
Any building inspector can use the initial "101" section of the code where it states that the building official can use his or her own review to accept a different approach. Most building officials are loathe to employ this part of the code, but you might have a bit of a conversation with him or her about this?
The other thing is that Washington state has a "green building code officials" group or organization run by an amazing woman: Patti Southard. (see https://kingcounty.gov/depts/dnrp/solid-waste/programs/green-building/sustainable-cities/city-government/green-building-ordinances-policies.aspx). You might want to see if you can tap into this group for support?
Peter
Utter ignorance of the official aside, a high density fiberglass R38C "cathedral ceiling" batt hits R35 when compressed to 9.25" in a 2x10 bay, but with 1" foam board edge strips on the rafters, through-screwing the gypsum board to the rafters would perform at the full R38 at 10.25" of loft, and provide additional thermal break beyond the R17 above. If not foam board, ripping some 1" OSB or plywood into 1.5" wide strips ring-shank nailed to the rafter edges works too.
Alternatively, 1.8lb density blown fiberglass hits ~R39 in 9.25" of depth. eg- see page 4: https://www.buildsite.com/pdf/johnsmanville/Blow-In-Blanket-System-with-SpiderFiber-Glass-Summary-Brochure-B18666.pdf (Optima, L77, Spider et al have pretty much the same specs at 1.8lbs density.)
You’re correct, it should work for an unvented roof. The building inspector would only be correct if you put insulation on top of a vented roof where the venting would effectively cancel out the above-sheathing insulation.
It is not uncommon to have issues with local inspectors not understanding new or complex assemblies. Local inspectors tend to be most familiar with the most common construction methods in their area, and anything outside the norm is suspect. You may try contacting your state-level building people and ask if they can review your design and overrule your local inspectors. I’ve done this many times with commercial projects, but it’s unusual with residential. The other option you might try is to contact a contract engineering/architecture firm and have them approve AND SEAL the drawings for your design. The seal might smooth things over with your local inspectors and avoid the need to deal with the state, but the engineer/architect will charge for their service.
Bill
Thanks, all!
Aside from the code compliance discussion are there any other details about the roof design that strike you as risky or unnecessary?
For example:
1) Is a class III vapor retarder recommended over sheathing in addition to taped polyiso (which is also a vapor retarder)?
2) Is silicone between rafters and gypsum a useful detail (to avoid air-bound moisture from getting into the assembly from underneath)?
3) Am I right NOT to paint gypsum on ceiling (and under T&G) so the overall assembly can better dry to the inside? Can not painting but using silicone both make sense (one about air and the other moisture?)
4) Is sealant detail in the soffit design in the right place?
Dana - aside from the general rule of "the more R the better" is there a compelling reason to add rafter depth or change to blown fiberglass? Also any source could I cite, and batt detail to spec beyond "high density" that gets me to R35 compressed in 9.25"? I got R32 from a Washington State code table figuring that was easiest for the City to swallow.
Re 806.5 compliance I'm going to simply lay out the specific code provision (806.5.1.2) I am trying to comply with in the hopes that it speaks for itself. Perhaps if I were a lawyer, or inspector trying to protect my ego, I would say the language in 5.1.2 -- that the rigid insulation is "for condensation control" - means only that and not for preventing heat loss which happens only inside the thermal envelope. Not that it would be right, just saying this is one aspect of code that isn't 100% clearly supporting my design. If the City sticks to its original finding then I'll allow my ego a quiet private victory but go along with their request for the more expensive (in $ and to environment) closed cell foam underneath (aka method 5.1.3) since, sadly, this would be faster and less expensive than invoking more 3rd parties into a battle with a stubborn inspector.
Dave,
One part of the confusion on the inspector's side might be that they are used to dealing with vented roof assemblies. If you have a vented roof, putting exterior rigid insulation above doesn't do much for the R value of the whole roof, this might be their thinking behind the rejection.
Thanks, Akos. Yeah, that's what I took from Zephyr's reply as well, tho the City told me I needed to detail R38 closed cell foam under sheathing which would also be an unvented solution with rigid foam above sheathing. I'm far from an expert (not an architect or builder, just a homeowner who has spent way too much time learning barely enough to be dangerous) but my guess is an old IRC only specified the closed cell impermeable approach and that's the one he's familiar with. Maybe he'll be happy to learn something new ;)
If you want paint, look into a pure mineral paint. Expensive but very breathable.
>" Also any source could I cite, and batt detail to spec beyond "high density" that gets me to R35 compressed in 9.25"? I got R32 from a Washington State code table figuring that was easiest for the City to swallow."
The manufacturer's compression charts indicate R38 HDs hit R35@ 9.25". eg: From Owens Corning:
https://s3.amazonaws.com/greenbuildingadvisor.s3.tauntoncloud.com/app/uploads/2018/08/10015707/Compressing%20fiberglass-700x310.jpg
>"...is there a compelling reason to add rafter depth or change to blown fiberglass?"
That was to explain the means by which one could "...get R38 fluffy into my 2×10 joists)", which you had implied couldn't be done. It can.
Thanks, Dana. I'll use the R35 compressed rating from OC. And now that I read your first reply again I realize that, if the City rejects my second submission, I would rather use your method of foam board edge strips than resort to closed-cell spray foam.
An even easier/better batt solution would be to stack R23 rock wool batts (nominally 5.5" ) onto R15 rock wool batts (nominally 3.5") for R38 for 9.0" total stack depth with 1/4" of air gap.
It's fairly easy to fluff the extra 1/4" to the batt loft for a compression fit during intallaion (or insert a 1/4" foam spacer) and it'll probably perform at R39 or a bit higher, but certainly not less than R38.