Insulating Walls Without Sheathing from Interior
I am renovating a 100-year old house near Seattle (climate zone 4 Marine). The house has wood siding nailed directly to 2×4 stud walls (no sheathing, no insulation, no tarpaper or other membranes). There are no obvious signs of rot or other water issues. The owner wants to gut the interior and insulate the exterior walls from the inside while leaving the siding in place. Of course, when I insulate these walls I want to avoid creating new moisture problems. I’ve been rereading all of relevant GBA articles I can find, primarily, Insulating Walls in an Old House With No Sheathing. https://www.greenbuildingadvisor.com/article/insulating-walls-in-an-old-house-with-no-sheathing
I’m tentatively planning to install Dorken Delta-Dry rainscreen against the inside of the siding. https://www.dorken.com/media/docs/products/Product-Details/DELTA-DRY-Brochure.pdf The rest of the stud bays are then filled with closed-cell spray foam, as shown in the attached detail. The foam is applied directly to the inside of the rainscreen mat. Dorken’s technical consultant recommends this approach. It seems to meet all of the criteria described in Martin’s article. It’s similar to the approach demonstrated in this Matt Risinger video. https://www.youtube.com/watch?v=jhoQ_f9ZnzQ Risinger drills vent holes in the top & bottom plates to increase airflow behind the siding, but I’m not planning to do this because Martin says not to!
Has anyone actually used the Delta-Dry product in this way? Any concerns or feedback to share? I’m slightly wary because even though Dorken’s technical staff recommend this approach, it seems that Dorken doesn’t officially endorse this use of their product. (E.G. I can’t find any documentation or details from Dorken showing the Delta-Dry product used in this kind of retrofit application.)
Another thing that bugs me is I don’t fully understand how water exits the wall after it trickles down the rainscreen and reaches the bottom plate. (Unlike a typical rainscreen in new construction, there are no weep holes.) I think the answer is “the same way it always has” — I.E. it collects on the plate until it evaporates or leaks out through gaps in the siding. Does that sound right?
P.S. The applicable codes are 2018 Washington State Building Code (IRC w/ state amendments) and 2018 Washington State Energy Code – Residential (IECC w/ state amendments).
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
I'm not sure what the Delta Dry buys you in this case except for more cost.
The original suggestions from the article (1x1 strips in the corner, foam/housewrap stapled on) is about the same work, lower cost and since it is permeable it lets you insulate with open cell foam or fluffy insulation. This also allows for the cavity to dry to both inside and outside as well.
One trick I've found that works great for building roof vent channels which would also work here is to grab a roll of house wrap and cut it on the miter saw to a bit less than the width of the stud bay. You now have precut rolls you can unroll and staple as you go making for a very quick install.
Thanks Akos. I'm sure you're correct about the "strips" method being more affordable.
I agree with Akos here -- adding a "rain screen" on the interior of a wall isn't really gaining you anything. A rain screen is supposed to help keep water out of the interior of the wall assembly, so putting the rain screen inside the wall isn't helpful. You're also correct about the "where does the water go?" problem here. I suppose such an installation would help ensure that only the lower part of your wall would rot out...
I would double check the exterior flashing details and make sure you've done the best job you can to keep water from getting behind that sheathing in the first place. Insulate with mineral wool batts, which are less of an issue than fiberglass if any water does make it into the wall. Do air sealing the old fashioned way with caulk and canned foam. This way you don't need any spray foam, which will save you more money. Put some of that savings into a basic interior side smart vapor retarder (like Membrain) for some extra insurance. That's what I did on parts of my own home here in CZ5 prior to replacing the exterior T1-11 for rigid foam and new fiber cement siding.
Hopefully Malcolm will see your post and have some ideas for you too. He's done a lot of work in your climate zone and knows about walls that work in damp enviornments.
The gap you create is primarily a capillary break, and to allow more drying. If the siding is so ineffective that the gap is necessary to drain bulk water then you should probably deal with that at the cladding layer, and not rely on the inverted rain-screen to try and solve the problem.
I agree with Akos and Bill. A simple gap with as little bridging it as possible, maintained by permeable baffles, would be my preference too.
Thanks Malcolm and Bill.
I shouldn’t have referred to the Dorken mat as a “rainscreen”. I recognize that in this case it mainly functions to ventilate the back of the old siding.
It sounds like the spacers+rigid+batt approach that Martin’s article describes is preferable to the drainmat+sprayfoam approach. But, it seems like both approaches have the same “where does the water go” problem of water trickling down and collecting at the plate. Following Malcolm’s comment about ineffective siding, it seems that if significant amounts of rain are getting through, no insulate-from-the-interior retrofit is going to work unless it fixes the cladding problem, right? Since there isn’t currently any obvious rot, it seems reasonable to assume that the existing siding is reasonably effective (I hope).
I should clarify: You may well get wetting of the back of the siding and even some on the sills, but as long as there is a good capacity to dry that will probably not pose much risk. If you don't see any rot now, you are likely to be just fine with what you plan.
What type of rigid insulation would you recommend for the "permeable baffle"? I'd prefer to use XPS for the higher R-value, unless you think the permeance is too low.
Thinking about it again, if you are using foam I'm not sure the permeability of the baffle matters. The wall has good drying capacity in both directions from the foam layer as long as you don't include an interior vapour-barrier.
Thin (1/2" or so) unfaced XPS is probably OK here. EPS would be more permeable. I would avoid polyiso in this particular application though -- foil faced polyiso is a vapor barrier because of the foil, and while the fiberglass mat faced polyiso is permeable, it's also not happy if it gets wet.
I don't think the choice of rigid foam is super critical here as long as it's not foil or poly faced, assuming it's not particularly thick.
Some of it depends on who is doing the work and how much that labour costs.
If you are paying for trades, I would go for 1x1 stick with stapled house wrap and fill the cavity with open cell spray foam. This seals up the assembly, uses a reasonable amount of polymer and gets you a decent R value assembly.
If you are looking at DIY, the 1x1+rigid+batts is probably the cheapest. Much more work to cut and cobble the foam and to seal it to the studs.
Since the rigid is between studs, you get the thermal bridging from the wood studs so the R value of the rigid doesn't matter much. You get a higher center center of cavity R value with polyiso but the assembly R value barely budges. The best is to select a foam that is permeable as suggested bellow or one that is thick enough to provide condensation control in your climate (~R3.5 in zone 4 for a 2x4 wall).
Thanks again for the responses. It sounds like there’s no consensus about the importance of foam insulation permeance. The options seem to be:
• Use permeable foam to facilitate drying to both sides, or
• Use impermeable foam that is thick enough to prevent condensation, or perhaps
• Trust that the assembly will always dry to at least one side without worrying too much about the foam permeance.
Good points, Akos. In your spray foam example, is the stapled housewrap necessary to protect the foam, or is it just there to create a gap behind the siding? My understanding is that it’s unnecessary to have a protective layer between the foam insulation (regardless of type) and the air gap behind the siding.
The house wrap serves a number of purposes. It works as a WRB behind the siding, this is needed as open cell spray foam can absorb water. It provides a backing to spray foam against plus it keeps the vent gap to the siding.
The house wrap is only needed for a spray foam install. If you go with cut and cobble rigid insulation you can skip at.
Polyiso does absorb some water, but not as easily as many seem to think. I've used it over many years, and have observed many situations where the exposure to water varied. My experience is that unless it is literally trapped underwater the penetration of water is all but negligible. So it makes good sense not to use it under a slab, for instance, but even leaving it in the rain is no worse than exposing any wood to rain. It will take on some moisture, but it will dry at about the same rate, once protected, as lumber. I have seen that EPS will also get waterlogged underwater unless it is a dense formulation.
Adding another detail drawing of the "indoor rainscreen" (with fire blocking) from this post: Managing Water and Insulating Walls Without Sheathing.
Also in the Northwest with the same project ahead of me. Can one combine this with interior rigid insulation, or would that be creating a problem?