Hi Everyone,

I've read most of the threads on here about Cathedral ceilings, as well as roof stack ups. I'm building an ICF home in Maryland and I've spoken with a few SPF contractors about how I should construct / insulate the roof. I've also read the "Moisture-Safe Unvented Wood Roof Systems" article that has been recommended a number of times.

http://buildingscience.com/file/5809/download?token=18Y6NJQ8


Here are the particulars of my project:

- 9:12 Roof Pitch
- 14 in TJI rafters 16 o.c.
- Climate Zone 4A Zip 21012
- The ridge beam of the house lays 324 degrees
- There will be no can lighting (*gasp*) installed in the roof cavity.  Hopefully I can even re-route the plumbing vents.

Proposed Stackup: Asphalt Shingles (dark) -> 30# Felt -> OSB -> 10+in ocSPF against Roof Deck. -> Sheetrock

In the article I reference above, the tabulated information recommends against a full fill of ocSPF for Zone 4 but later in the text it states:
 
   "...field experience has reported some examples of high moisture content roof sheathing with full depth ocSPF roofs in climates in Zone 6, with occasional reports in Zone 5 and Zone 4C. However, Zone 4A should be a safe location for such a system except for high interior humidity or low solar heating of the roofing."

The house will have an "ERV" on each level (more like an "ERV-lite"; Panasonic FV-04VE1)
Humidity will be monitored but I don't have any specific plans to control it unless I exceed 55% too often - then I'll add de-humidification.
Heat (cooling) sources: 27k BtU of Minisplits (although I won't need all of that for heating) 
Radiant floor heat in basement.

Not sure if this is relevant, but based on 3 different models, the cooling loads on the house are between 28k and 31k.

What do you all think of the stackup for my situation?

Should I be concerned with conductive heat loss through the rafters and add insulation between the rock and the rafters?

Should I increase the thickness of the proposed ocSPF to 12"+ or completely fill the cavity (beyond code)?

Should I install a vapor barrier (or smart vapor barrier) on the interior?



I appreciate your thoughts.

-Craig

The concern here would be the drying rate of the north facing pitches. There is effectively zero drying through a ~0.1 perm layup of #30 felt + asphalt shingle, a modest accumulation & drying rate through 10" of ocSPF.

In the BA-1001 document table summary you'll note that both of the zone 4A climates simulated showed potentially risky moisture levels in the OSB, independent of roofing type, even dark shingles.

A full cavity fill would bring it up to something like the current R49 IRC code (10" would only meet IRC 2009 code minimums), and putting a smart vapor retarder such as Certainteed MemBrain between the ceiling gypsum and foam/TJI edges would limit the moisture uptake without limiting the drying rate when the sun is driving moisture out of the roof deck.

Note, the High humidity simulation was only 40% interior RH when ever it was below 15F outside, and the Normal was 30%.(See page 7). Letting it get as high as 55% at an outdoor temp of 40F (probably your wintertime average) would definitely be on the high side. Tweak the ventilation rates to shoot for 40-45%RH or lower indoors until the daily average temps are 45F or higher.

Why not use an "over the rafters" vented roof with one layer of OSB and some rigid foam below the gap (no spray foam)? Spray foam contractors might not be the best ones to ask about such a design.

Posted By jonr on 14 Jan 2016 04:13 PM
Why not use an "over the rafters" vented roof with one layer of OSB and some rigid foam below the gap (no spray foam)? Spray foam contractors might not be the best ones to ask about such a design.

I'm not quite understanding your proposed stackup.

Venting above the roof deck can work in zone 4 if you have a fairly vapor permeable underlayment and soffit to ridge venting. The OSB itself runs about 1-perm when dry, 5 perms when damp enough to support mold, so a smart vapor retarder would still be necessary with this approach. In zone 5 it's risky and in zones 6+ it won't cut it.

It's not nearly as effective as under the roof deck venting, where moisture coming from the interior hits the convecting ventilation gap first, and doesn't need to pass through the OSB to get out.

The other option I was considering... although I've pretty much moved on from it... Is to install 1 in (or 2 in.) rigid foam against the inside edges of the top "2 by plates" of the TJI's. This would create a 1.5" cavity through which air could flow against the inside of the roof deck. Vent it at the ridge and sofit. Then spray my 10+ inches of ocSPF.

I prefer however the simplicity of the completely un-vented design.

Anyone have any opinions on Thermal Breaks from the inside of the TJI's to the sheetrock? If I add a smart vapor barrier, then I'd be inclined to use the EPS strips to hold the vapor barrier to the TJI's.

Thanks all.

Posted By Dana1 on 14 Jan 2016 02:43 PM
The concern here would be the drying rate of the north facing pitches. There is effectively zero drying through a ~0.1 perm layup of #30 felt + asphalt shingle, a modest accumulation & drying rate through 10" of ocSPF.

In the BA-1001 document table summary you'll note that both of the zone 4A climates simulated showed potentially risky moisture levels in the OSB, independent of roofing type, even dark shingles.

A full cavity fill would bring it up to something like the current R49 IRC code (10" would only meet IRC 2009 code minimums), and putting a smart vapor retarder such as Certainteed MemBrain between the ceiling gypsum and foam/TJI edges would limit the moisture uptake without limiting the drying rate when the sun is driving moisture out of the roof deck.

Note, the High humidity simulation was only 40% interior RH when ever it was below 15F outside, and the Normal was 30%.(See page 7). Letting it get as high as 55% at an outdoor temp of 40F (probably your wintertime average) would definitely be on the high side. Tweak the ventilation rates to shoot for 40-45%RH or lower indoors until the daily average temps are 45F or higher.

Good points... I'd never be 55% in winter anyway.. unless I was continuously using the indoor hot-tub I've planned.... :) Thanks for your insight...

Posted By craigtoo on 15 Jan 2016 11:18 AM
The other option I was considering... although I've pretty much moved on from it... Is to install 1 in (or 2 in.) rigid foam against the inside edges of the top "2 by plates" of the TJI's. This would create a 1.5" cavity through which air could flow against the inside of the roof deck. Vent it at the ridge and sofit. Then spray my 10+ inches of ocSPF.

I prefer however the simplicity of the completely un-vented design.

Anyone have any opinions on Thermal Breaks from the inside of the TJI's to the sheetrock? If I add a smart vapor barrier, then I'd be inclined to use the EPS strips to hold the vapor barrier to the TJI's.

Thanks all.

That solution for creating a ventilation space is actually a pretty commonly used one. For ocSPF and 1.5" nominal vent cavity you don't really need 1" foam board, and at 1" XPS (or any foam with facers) borders on being too vapor tight. Half-inch XPS or unfaced 3/4-1" EPS would be fine.

The thermal bridging of a TJI is only about a third that of 2x milled lumber rafters (0.5" of web instead of 1.5" solid wood.) Whether it's "worth" adding edge strips is debatable, but you'd get more thermal break out of foil-faced polyiso than you would EPS, and it's easier to cut into strips cleanly than EPS. The strips should be at least an inch wider than the TJI flanges to get the most out of them.