Insulating a Standing-Seam Metal Roof on a Deck-Style House
Hi All —
Getting ready to do a total reroof. Deck House-type construction. I’ve seen a number of postings that address some of our issues, but nothing specifically.
Built in 1963 outside of Boston. Not by Deck, and not by TechBuilt, but has similar characteristics. Roof is very much like a Deck.
Ceiling is 3×8 T&G Fir. This is also the roof deck, originally covered with tar paper and shingles. Was retrofit poorly around 1990 with 2×6 sleepers, rigid foam panels, plywood sheathing, and architectural asphalt shingles.
We have a bad leak that needs to be fixed. Seems to be affecting a number of areas around skylights.
Idea #1
Rip off everything down to deck. Recover w/vapor barrier, 2 layers of overlapping and taped polyiso panels like Hunter or RMax, followed by 5/8″ ply or OSB glued and screwed through panels, ice and water shield, and architectural asphalt shingles.
Idea #2
Same as above, but forego sheathing and cover w/standing seam roof.
Hybrid #3
Rip off existing sheathing. Remove & discard existing rigid foam, retain sleepers and fill cavities with spray foam, ice and water shield, and standing seam roof. In this scenario, we would also look for spray foam to act as a vapor barrier, not sure open foam can do that.
I know the standing seam can go directly on the polyiso panels without the need for sheathing. But can we substitute spray foam for the panels?
I’ve seen postings here about the pros & cons of spray foam, and not sure which, if either, is appropriate for this install. Open Cell is about the same price as the polyiso panels, but it is a big labor saver. It seems to be a fire hazard, and attracts water like a sponge
I’ve also read here about combining rock wool with spray foam.
We’re trying to find a solution that’s cost effective and can provide good R-value…would Closed Cell work???
Looking forward to hearing the group’s ideas…thanks!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I did a Deck House roof several years back. Option #1 works, and is very similar to what we eventually did. The previous roofer didn't understand the concept of the roof and there were nasty condensation issues. If you do option #1, the air barrier on top of the decking is critical. We did use foil-faced polyiso panels and taped them too, but I'm not sure that's necessary. Our building owner didn't want to pay for yet another roof, so we did both. Absolutely use low-rise spray foam around the skylights. Air leakage around the skylights was a major point of failure in the first reroofing. You might consider a vapor permeable underlayment over the sheathing. This would, at least in theory, allow any trapped moisture to flow through the plywood and underlayment. Even if it condenses on the underside of the shingles, it will generally then flow down to exit at the eaves. We definitely saw the roof trying to dry to the exterior during the first failure.
Thanks, Peter...great info...I'm also getting conflicting info re: code and asking my builder for clarification re: how much R-value...I've asked him to check our local building dept. If this is a reroof we need to install 7" polyiso. If this is a repair, we can get by with less.
I understand the long-term savings potential with higher R-value, but we only run our furnace less than 3 hours a day on average during the coldest months, and it would take us a long time to realize the cost differential.
I helped a friend with a similar full cathedral reroof on his house, in CZ3. He has 2x6 T&G, so we installed a peel & stick VB on top of the deck, 2 layers of 2” iso and 1 layer of 1.5” iso, all with staggered taped seams. The 5.5” was to use 2x6 perimeter nailers. A second P&S VB encapsulated the assembly. 2x4 notched purlins at 24” o.c, allows for ventilation and drainage. Standing seam roof was his choice for durability, energy efficiency, safety and environmentally friendlier roof.
As you mentioned, you would need 7” iso in your climate zone, but if your budget allows it, I would choose Idea #2.
I forgot to mention that ocSPF is not allowed by code, without ccSPF or rigid foam on top of it. Options 1 & 2 are your best options.
If you wanted to use Rockwool, you would need to build a roof deck wit 16” TJIs, fill the cavities with 14” insulation, and create a vent channel on the top flange of the TJIs. Finally, you install the roof deck, WRB, and roofing system.
Thanks Armondo.
Leaning towards option number two. Plywood is so darn expensive. And then there's all the labor for hauling it, and installing it. If we can figure out a way to use the spray foam, it saves us additional labor on installing the foam panels. Those savings start bringing the standing seam roof into parity cost-wise with originally planned architectural asphalt shingles.
I can picture everything you describe, but don't understand how the purlins find their way into the puzzle. Also, so many sources advise against a vented roof on a Deck-style house. I think it has to do with the very low pitch.
We have a limited budget, and am not interested in cutting corners. Would rather have a proletarian roof done right, than a bourgeois roof that's iffy.
Again, thanks for your feedback.
Early in Covid I did a studio roof that is similar to what you are thinking about. It had T&G roof deck with asphalt shingles.
First sealed all the grooves of the deck where it goes over the wall top plates by drilling holes between the boards and injecting SPF into the hole. This is important as all those small gaps add up to a huge air leak.
The roof was covered with a budget ice and water peel and stick to seal it up.
I strapped out the roof with horizontal 2x3 on edge 25.5" OC with 2' wide 2.5" polyiso in between.
Covered the whole thing with a permeable underlayment and installed a new metal roof directly onto the 2x3s.
Not the highest R value roof but much better than the leaky uninsulated roof it replaced.
In your case, I would do something similar but instead of 2x on edge I would go with 1x4s on flat over the rigid. This would eliminate the thermal bridging of the wood purlins and would allow you to pass code on U factor basis, which reduces the amount of foam you need.
Assuming your 3x8 deck is 2.5" thick, this would mean around 5.5" of polyiso above to get an R38 assembly. There is no ROI in going up to R49 of foam, the extra cost of the rigid plus dealing with the huge screws will never pay in energy savings.
"First sealed all the grooves of the deck where it goes over the wall top plates by drilling holes between the boards and injecting SPF into the hole."
That is a great idea. I could never figure out how to deal with the gaps.
That works, but you've got to be careful - it's easy to make a mess of things. I also wonder how long the spray foam is going to stick. The wood in these decks moves a lot. That's still got to be better than nothing, though.
Peter,
Maybe something like Big Stretch instead?
"In your case, I would do something similar but instead of 2x on edge I would go with 1x4s on flat over the rigid. This would eliminate the thermal bridging of the wood purlins and would allow you to pass code on U factor basis, which reduces the amount of foam you need."
Thanks, Akos...our roofer suggested the 1x4s just like you mention...we're checking w/local building dept. to determine if this is considered a repair or a re-roof...if re-roof, I believe we have to go whole hog up to 49...if repair, we can rebuild to current specs...
"sealed all the grooves of the deck where it goes over the wall top plates"...have to agree w/Peter E, and also wonder about the ROI on time involved...I'd think a bead of caulk on both sides of wall, while not doing much for insulation, would help mitigate air migration. Might be a bit more flexible, too, for when the deck shifts.
Even if you need R49 (prescriptive option) there is usually a U factor based performance compliance option. For an assembly with all rigid on the outside, this means you can use less insulation. All building codes have a performance option, even if you building department is not familiar with it.
Sealing the grooves was one of the easiest steps on the whole re-roof. For a 600sqft roof it took no more than half an hour. Caulking the inside won't work, you need to block a pretty convoluted 3d airflow path.
The gaps between these boards can be quite big, say even a very small 1/8" gap on a 2.5" board, that works out to about a 4" hole per 20 boards. That is not a small leak. The SPF might not seal it 100% and there will be some leaks down the road, but they will still be way better than that.
"All building codes have a performance option"...thanks, Akos...we'll check on that...also, any chance you can send a small sketch illustrating the "drilling holes between the boards" technique...I'm having a hard time visualizing it, and it sounds like a path worth pursuing
Deleted
Here is a shot of the roof deck after the spray foaming. I had pretty large gaps to fill.
I drilled down from above the top plate with a 3/8" drill between each board and filled the gap with gun foam. You want to make sure that enough SPF makes it in there to fill the bottom side of the T&G as well. Because of the big gap, I had some overflow and excess that had to be trimmed off from the inside as well.
In my case the gaps were big enough and there was enough air leaks there that I had spider webs in most of them. Good indication of how much air was leaking out of the place.
Deleted