Hot roof underlayment: TigerPaw?
sleaton
| Posted in Energy Efficiency and Durability on
I will have an unvented roof assembly. I was looking for an underlayment for my roof deck. I noticed that sharkskin and others aren’t going to work, as they are very low perm.
It looks to me like TigerPaw would work, or would you just throw down a 30# felt, or some other product ?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Sleaton,
1. I'm not sure whether I agree with you that TigerPaw underlayment can be used with unvented roof assemblies. The TigerPaw installation instructions note, "Provisions must be made in the roof design and installation to provide proper ventilation to avoid high humidity, condensation, and mold growth problems."
This statement is somewhat ambiguous, so you might want to talk to a GAF representative to understand whether using TigerPaw on an unvented roof violates warranty requirements.
2. Why does the vapor permeance of the underlayment matter? Most types of roofing don't allow outward drying. Can you tell us what kind of roofing you will be installing?
This will just be Owens Corning Oakridge over an unvented roof. I verified with Owens that they have modified their warranties of late and do cover unvented roof assemblies. The roof will be sprayed with 9in of open cell underneath and covering the rafters. I have read on numerous underlayment documentation and other places on this site, not to use sharkskin and to be very picky about what underlayment to put over the osb under the shingles on an unvented assembly. I am starting to lean towards 30# felt. On the GAF site it claims "Help reduce damaging moisture that can become trapped on your roof deck, Helps remove nearly twice as much damaging moisture from your roof deck as the leading synthetic non-breathable underlayment" on both TigerPaw and Deck Armor underlayment. I will have whole house dehumidification (Austin Area), but even with that, I know there is much debate about open cell and OSB rotting.
That was pretty much my concern.
Sleaton,
You are right to be concerned about moisture accumulation in your roof sheathing, but you are mistaken if you think that the roof sheathing can dry to the exterior -- no matter what type of roofing underlayment you specify.
The only way to help the roof sheathing dry to the exterior would be to establish ventilation channels, soffits to ridge, above the roof sheathing.
I wasn't sure what "Owens Corning Oakridge" means, so I googled it. I guess you are talking about asphalt shingles. They don't allow any drying to the exterior.
If going with fiber insulation and an unvented roof, the shingle underlayment isn't going to make the difference. But there is reason to believe that a replacing foot of roof decking at both sides of the ridge with exterior grade fiberglass faced gypsum board covered by a mesh type ridge vent, and using dense packed cellulose as the insulation can have a substantial benefit. The highly permeable gypsum board at the ridge with the mesh vent above it is referred to as a "diffusion vent". An alternative is to See:
https://www1.eere.energy.gov/buildings/publications/pdfs/building_america/unvented-roofs-asphalt-shingles-cold-hothumid.pdf
The shortest summary of the benefit is in Table 7, p.56 (p69 in PDF pagination), but read the whole thing (tedious as that might be...)
Table 7 doesn't tell the entire story- the section (roof #4) with the lowest ASHRAE 160 failure hours was fiberglass insulation in a top-vented configuration, but dissecting the assembly after the fact showed more significant moisture damage than on the cellulose insulated bays. The moisture content of the diffusion vented cellulose bay peaked at higher than desirable levels, but dried more rapidly than the unvented dense-packed cellulose.
In warm humid climates a diffusion vent made by removing a just a few inches either side of the ridge vent, covering it with air tight permeable underlayment makes enough of a difference, even for fiberglass insulation with no interior side vapor retarder. See Figure, 95, 96, 97 & 98 starting on p.65 (p.78 in PDF) and Table 12, p.92 ( p.105 in PDF)
Asphalt shingle systems dry to the exterior at some low rate (say .1 to .65 perms depending - and note that dry cup measurement is the wrong test for this). That's not much, but I haven't seen evidence that this isn't significant when the wetting rate is less than this (ie, no air flow and low perms to the interior - possible with open cell + a vapor barrier/retarder).
Under the sheathing vent channels may be a well proven, code compliant option.
Dana,
Thank you for that information. It looks like there is still a long way to go in understanding the in's and outs. There was more to absorb there than I understood. But by doing the partial venting it really seems to make enough a dent into the gains of encapsulating, it makes me wonder if one should do it at all. Every situation is different, but it seems like doing the old fashioned vented roof, and using buried ducts, or spiral ducting to bring the ducts internal to the envelope. I haven't seen much about ducts in the crawlspace, that too I would think could be a solution for some.