Utilize an air gap under a metal roof above a conditioned attic/cathedral ceiling?
Looking for a sanity check here.
In zone 2A (Texas) ~35″ of rain a year, planning an unvented cathedral ceiling. Utilizing offset external poly-iso foam layers to warm the sheathing above the dew point (>R5) in my climate zone. On top of the insulation, putting a peel and stick high temperature water/vapor barrier. Then add 1×4 battens on a diagonal and gapped to provide a ventilating channel from the eaves to the ridge. The standing seam metal roof clips attach to the battens. In my mind, this allows for two things.
1) It provides a thermal break from the metal.
Due to radiation on clear nights the metal temperature can be 18F (10C) below the ambient in my area (See Fig. 1, http://www.sciencedirect.com/science/article/pii/S1876610212016219/pdf?md5=10ed7f9883cd792ee24c53d3088b1d81&pid=1-s2.0-S1876610212016219-main.pdf). The R5 external insulation calculations are based generally on the average monthly external temperatures and target dewpoints in the home on the condensing surface (inboard side of the sheathing). I’m hoping the air channel keeps the temperature closer to the ambient at the poly-iso surface.
On the high temperature side of things in summer, I always worry if high temperatures accelerate the degradation of the peel and stick water/vapor barrier.
2) If and when a small water leak develops in the roof, then there is a good path for it to dry and drain. I guess it also allows the overnight dew on the underside of the standing seamed metal to evacuate as well.
My roofing contractor is recommending not to use the battens and use a special tie that holds down the foam and allows for attachment of the roofing clips. I think the perceived advantage is the added support of the foam under all of the metal surface. Maybe reduces oil-canning? But to me it seems that with the metal in direct contact with the vapor barrier, liquid water movement will be limited due to surface tension and in my experience standing water always finds a way into places it is not wanted. Might even be driven as a vapor as the metal heats up during the day.
What do you think?
Thanks,
Eric Seaberg
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Matt Risinger believes that some airflow under the metal is beneficial in Texas. Either with battens or 3D mesh (which is more evenly supportive).
http://mattrisinger.com/venting-a-metal-roof-with-rigid-foam-underneith/
https://www.youtube.com/watch?v=ezzzeIesmhQ
Better would be actual measured data, not opinions. I've seen some for temperature reduction (ventilation is beneficial), but not sheathing moisture. Note that underlayment vapor permeability may also play a role (non permeable underlayment blocks a drying/wetting path).
Condensation on the bottom side of the metal needs to have a way out..
Eric,
Unlike shingles, I don't believe standing-seam metal roofs need to be designed for their inevitable leaks - especially in climates without the possibility of ice-dams. The roof should not leak until it deteriorates and needs replacement.
There may be some benefit to de-coupling the metal panels from the underlayment in terms of surface temperature, but not in dealing with condensation due to night time radiance. The moist air under the panels is only there because it has been introduced and occupies a gap you have created with the battens. No gap means no moist air to condense.
Unless you glue the metal to the sheathing, I suggest that "no gap" never happens.
> but not in dealing with condensation
Data?
My opinion - roofs are much like walls. A gap (even a tiny one like DrainWrap) is beneficial and in most climates, it's better if you don't seal it up with a low perm exterior.
Jon,
Data? Surely the onus is on the advocates of including something in an assembly to justify it, not those saying it isn't effective? The argument for inclusion always seems to boil down to: I'm worried about moisture, so I think I'll include a gap. That's not data.
We cover the same ground every time this comes up here - and in the absence of good data either way, for now I guess it does come down to opinions, however I'll make one last point:
A lot depends on what we are talking about when we refer to a gap. Ventilated chutes from soffit to ridge like those commonly placed under sheathing are a completely different animal that mesh underlayments, or battens placed horizontally or diagonally. A small or convoluted path for air to circulate, or moisture to drain doesn't offer much, and needs careful thought as to how they are detailed. I've yet to see a gable or eaves detail of these roofs that shows how the air or moisture management interacts with the flashings used to stop the intrusion of bulk water.
Hi Everyone,
Have been doing a little reading and ran across this paper with measured and simulated data quantifying the effectiveness of an air gap below a "cool metal" roof.
https://web.ornl.gov/sci/buildings/conf-archive/2013%20B12%20papers/162_Kriner.pdf
I got out of the paper:
1) Metal roofing mounted directly on the deck without Above Sheathing Ventillation (ASV) results in about the same heat flows into the attic as asphalt shingles.
2) The heat flow into the attic can be reduced by ~45% with ASV
3) The air gap can compensate for a good bit of coating aging
Now heat flow might be a benefit in some climates (winter), but in the cooling dominated area that I am in, It sounds like the battens are a good thing.
I haven't found data showing that night time radiation of the metal to the sky reduces the attic sheathing temperature, resulting in an increase in the probability of moisture condensation on the interior face of the sheathing (My roof layers are constructed to dry to the inside of the home). However, it seems common sense to me that an air gap would help this as well.
Hi Eric.
Peter Pfeiffer has been designing and building high-performance homes in your climate for decades and he has tried and monitored many roof assemblies. Today, he regularly uses a system for unvented, insulated rooflines with metal roofing similar to the one you are describing. If you haven't read it already, check out this article: Designing for a Hot, Humid City
Hi Brian,
Thanks for the reference. I missed this one while searching around. Lots of good information there.
I assume your selected standing seam roof is allowed to be installed on purlins? Even if it is, it makes installation and walking on the roof more difficult as the metal will kink if stood on next to a purlin.
The valleys in this assembly are generally solid 3/4 decking as most people are walking there. The ridge and eaves generally double the number of 1x4's to harden those areas to potential traffic.
I must say this is not my original thought. I started in the general direction based off a detail I saw in some plans, and worked my way back to Peter Pfieffer in the article Brian referenced above, educating myself along the way.
By the way, I contacted a local metal supplier and he was familiar with the assembly. His standing seam was compatible as long as it was fastened with the appropriate frequency and generally was the clip type. Evidently the mechanical seamers can get off when running over the nonuniformally supported panel.