Unvented attic/rafter spaces

[Sifu]

2018 IBC, type V building, climate zone 5

I find the language for unvented attics is making me re-think my assumptions. I have a proposal for all insulation to be placed in the roof framing directly under the roof sheathing. It is air-permeable, batt fiberglass. There is no "ceiling" applied directly to the underside of the roof framing, and there is no airsapce (baffle) with any eave or ridge openings. So the attic spaces are "warm", and the thermal barrier is at the roof sheathing. My first comment was to ask how they are ventilating the attic per 1202. I expected them to tell me how they are going to put the baffles and ventilation in. Instead, they responded by saying the attic is "warm" so no ventilation is required. So I now must assume they are going to go with 1202.3, which would require additional rigid or sheet insulation on the roof deck. Easy right? But a closer reading of 1202 contains this for both 1202.2.1 (ventilated rafter spaces) and 1202.3 (unvented enclosed rafter spaces): "ceilings are applied directly to the underside of roof framing members". In this case, there is no "ceiling". I don't think this matters, as the intent is to prevent moisture migration through the air-permeable insulation and condensing on the underside of the roof sheathing, but it makes me wonder if I misunderstand this. Commentary doesn't help. I think this is poor semantics, but I would like to hear other opinions.

I also see an issue with applying the required R20 rigid insulation on top of the structural sheathing because the proposal is for asphalt shingle roof coverings, and I am assuming this will not work through several inches of rigid foam.

 

[bill1952]

Is the proposed design both fibreglass between rafters AND 4" of foam on top of the roof sheathing? I agree just the fibreglass will likley result in condensation on the sheathing. With the continuous foam on top, it should keep the sheathing above dew point and avoid condensation.

I dont know how shingles could be attached to just foam. It would need some sort of sheathing on top of foam or some product for attaching shingles.

 

[Sifu]

Is the proposed design both fibreglass between rafters AND 4" of foam on top of the roof sheathing? I agree just the fibreglass will likley result in condensation on the sheathing. With the continuous foam on top, it should keep the sheathing above dew point and avoid condensation.

I dont know how shingles could be attached to just foam. It would need some sort of sheathing on top of foam or some product for attaching shingles.

No, the proposal is batt fiberglass only, which the IECC would allow purely for the thermal envelope, but if my understanding of the code is correct, the IBC would not due to condensation which is why I assumed they would realize and provide an air pathway and ventilation openings. And the shingles can't be "nailed" to the foam, or through it....which is a problem that should resolve itself if they see the light.

 

[bill1952]

The fibreglass only in zone 5 would be bad quick IMHO. With a fantastically perfect vapor barrier - the turn it upside down and fill with water and no leaks ever kind of perfect - it might work. But I just did a vapor barrier under a very well vented attic and 20" of cellulose with butyl tape continuous on every rafter so the t&g ceiling nails stood a good chance of being sealed - and that's not good enough for what they propose.

With the continuous insulation on exterior - two overlapping layers - it could work. We're I doing it I anchor 2x4 to rafter through foam and sheath. Planning something like that now but metal roofing so just need horizontal purlins every 3 ft or so. But want 10" of foam and that's a long timber-tek. Sort of surprised we don't see the c.i. on roofs in residential more.

Good luck.

 

[steveray]

Where/ what is the vapor barrier on the "warm" side of the fiberglass?

 

[Sifu]

Where/ what is the vapor barrier on the "warm" side of the fiberglass?

There isn't one.

Unless one considers the limited ability of the roof felt to act as a vapor retarder, which really doesn't count because there isn't any insulation above it.

 

[Inspector Gadget]

Based on the research I've done, from a building science point of view, fibreglass in roof systems should really be ventilated.

 

[bill1952]

Based on the research I've done, from a building science point of view, fibreglass in roof systems should really be ventilated.

Not strictly disagreeing but the flash and batt "hot roof" with several inches of closed cell foam against and under the roof sheathing with fibreglass (or cellulose or rockwool) under that and NO vapor barrier inside so it can dry inwardly can work.

It's a little bit of a difficult subject to research with the influence of the venting product manufacturers, vendors, and installers in the marketplace and in the code development.

 

[Inspector Gadget]

Not strictly disagreeing but the flash and batt "hot roof" with several inches of closed cell foam against and under the roof sheathing with fibreglass (or cellulose or rockwool) under that and NO vapor barrier inside so it can dry inwardly can work.

As soon as you have a certain volume of fibreglass on the inboard side of foam, you also need vapour barrier. The issue is that the dewpoint might form in the fibreglass part of the assembly. In our area, the trigger is when fibreglass forms ~20 per cent of the total thermal resistance.

There are some studies showing that closed-cell foam on the underside of roof decks is OK. We have accepted Alternative Solutions proposals for this approach on several builds.

 

[bill1952]

Understand. In zone 6 , 2" of closed cell is claimed to be enough to keep surface above dew point. The problem of adding an interior vapor barrier is the double vapor barrier and no way to dry either in or out. But not without risks of moisture problems. Exactly the reason I'm drawn to all foam - 10 to 12" - on top of sheathing.

 

[steveray]

Understand. In zone 6 , 2" of closed cell is claimed to be enough to keep surface above dew point. The problem of adding an interior vapor barrier is the double vapor barrier and no way to dry either in or out. But not without risks of moisture problems. Exactly the reason I'm drawn to all foam - 10 to 12" - on top of sheathing.

No problem....just have them submit the engineering...

R702.7 Vapor Retarders​

Vapor retarder materials shall be classified in accordance with Table R702.7(1). A vapor retarder shall be provided on the interior side of frame walls of the class indicated in Table R702.7(2), including compliance with Table R702.7(3) or R702.7(4) where applicable. An approved design using accepted engineering practice for hygrothermal analysis shall be permitted as an alternative. The climate zone shall be determined in accordance with Section N1101.7.

 

[Sifu]

IMO, no way what they propose, or their response can work. I really think this is the DP having a poor understanding of the condensation issue, and a poor practice of reading the code....even when I gave them the code section in my first comment.