GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Creating a Thermal Break

soulincode | Posted in Pretty Good House on

Hello again! After some time and lots of work since my last post, I’m writing again to request input from the GBA community.

This is for a PGH building envelope design on a renovation in the Catskills, NY (I’ll paste link to last post at end of this one).

The new roof is going on next week, along with new hemlock soffits and cut-outs in them for continuous soffit vents that will be installed.

A crucial and timely decision needs to occur ASAP related to this work: whether to add layer of exterior insulation and rain screen, to thermal break the studs in framing.

I have a quote from the builder who will do all the air sealing and building envelope execution, for the design in the attached PDF. You can see it has Steico wood fiber insulation at R7.5, also the rain screen, which would bump out the current exterior wall and affect location of soffit vent location.

This option provides a thermal break for the wall studs, but will add significant cost to the project. An alternative option is outlined below, achieving good R value, but no thermal break, as this one insulates from current walls inward.

Please note the floor joists will lack a thermal break. With the pier and beam crawl space, I could not find way to affordably condition the space or thermal break joists, so the resulting plan is to use 4″ closed cell HFO spray foam to R30

For the walls, here are the two option in mind, the first one being what is shown in the PDF drawing, the second one not shown in drawing, but quite similar, where rain screen and wood fiber are removed, adding a Tyvek or other material for some protection:

R39.13 w/ Thermal Break

From exterior: Board and Batten Cladding, Rain Screen, Steico Wood Fiber Insulation, Old Plywood Siding, Rockwool Bat, Air Barrier, Service Cavity with Mineral Wool Insulation, Drywall

R32.13 w/o Thermal Break

From exterior: Board and Batten Cladding, Tyvek (or equivalent), Old Plywood Siding, Rockwool Bat, Air Barrier, Service Cavity with Mineral Wool Insulation, Drywall

At the moment, I’m leaning towards “R32.13 w/o Thermal Break” option for two reasons:

1) The cost to thermal break walls is high, but still leaves the floors without thermal break. So it’s a lot of cost that still leaves inefficiency in floors. I can’t afford to thermal break both of them.

2) The cost savings from not making thermal break in walls can be applied towards solar panels, which ideally help make up or at least reduce the higher heating costs.

Here are some relevant numbers:

* Total cost for rain screen and R7.5 wood fiber insulation is $16K: rain screen $5K, Steico $6K, Steico labor $5K.
* Cost for solar panel installation (when including federal and state rebates): $20K
* Cost for labor and R23 Rockwool in 2X6 insulation: $5000
* Air sealing walls and ceiling: $6000 (foam in joists would seal floor)
* Cost for service cavity: $5000
* No cost yet for R9 mineral wool insulation in service cavity. A guess is $2500-4000 range.

I would appreciate input and advice! I’ve got a feeling, but also feel stumped, where maybe more experienced hands in this kind of situation can help.

There is the obvious, where not paying costs now to stop thermal bridging creates more cost over lifetime to heat the house. Here, I wonder if the plan to put money into solar is a smart one. I’m also wondering if  the “R32.13 w/o Thermal Break” option would lead to moisture issues in Climate Zone 6, or would I be ok? The current house exterior siding seems pretty ok.

And fyi, here is my first post made on this project:

Building Envelope Design for Pretty Good House

Thanks in advance for any thoughts and feedback!

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member
    Michael Maines | | #1

    Unfortunately you are not likely to see a reasonable ROI with exterior wood fiber insulation in an assembly that also has an insulated service cavity. The only way to know would be to do an energy model that includes the thermal bridging of the studs, the cost of the insulation and the energy performance. But having done a lot of those, I wish I could make a financial argument but I'd guess you're in the low single digit ROI range.

    There are other benefits to exterior insulation, and to using wood fiber, such as keeping the sheathing warmer and thus drier.

    Insulating outside the rim joist in your design would have little impact.

    1. soulincode | | #4

      Hi Michael,

      Thanks for your input!

      If going with exterior wood fiber insulation, we could remove insulation inside the service cavity, saving some cost and still getting ok R value. However, it seems we'd still need the service cavity, in order to get plumbing and electric inside and sealed envelope and not puncture. So there would still be that cost.

      If not doing the exterior wood fiber insulation, I do have some concern for how well the sheathing will last, and keeping it dry.

      I see what you mean about the rim mean because insulation of 4" under and across all the floor joists would keep whatever cold moves through rim joists from going up into house? I guess some could then go up through thermal bridge of exposed floor joists, but a guess is it would not move that needle too much.

      1. Expert Member
        Michael Maines | | #8

        I almost always make the sheathing layer airtight and don't worry about getting the interior layer (drywall and usually a variable permeance vapor retarder) airtight, so the structural layer doubles as its own service cavity. Having an additional service cavity can be nice but on most projects I don't think it's worth the effort to create a fully sealed service cavity. You only need one nearly-perfect air control layer, though more than one airtight layer increases performance and resilience, if vapor is controlled.

        If you have a vented rainscreen and use plywood sheathing, and a variable permeance vapor retarder on the interior, the tiny amount of moisture that makes its way to your sheathing will dry quickly.

        As for your rim joists, correct--if you had a sealed, insulated crawlspace or basement, insulating the exterior of the rim joist might have a thermal benefit, but with spray-foamed floor cavities and relatively thick main-level walls, the temperature at the floor/wall inside corner would be the same either way. That's a vulnerable location where condensation and mold often form. You will have slightly more heat loss through the joists but if the crawlspace is near the outdoor temperature it won't make a noticeable difference. Have you considered insulating the floor cavities with fluffy insulation and adding a thin layer of foam below the joists?

        1. soulincode | | #10

          Hi Michael,

          Thanks...I wonder how you make your sheathing air tight or how it might work best to get air tight on exterior in this setup? See attached here, a revision for what I think you're suggesting...I'm not sure where the air tight aspect would occur, would it be tape sealing the Steico? Or something else?

          I don't have a variable permeance vapor retarder in this spec either. What products do you usually use?

          By saying structural layer is the service cavity, I'm guessing you mean within the 2X6s in this situation, and the electric at least, and possibly the plumbing could be placed inside and drilled through holes? I think plumbers like to avoid water moving along outside walls, so am not sure this R30 setup with the thermal break is enough. The alternative is running water across top of floor joists, but they would then need to be spray foamed in place more or less permanently, with no access.

          Thanks on suggestion for fluffy insulation in floor joists, then foam below joists. The challenge I see there is we'd then need to get layer of plywood across whole underside I think, in order to keep critters from nesting in the insulation, correct? I think that will be difficult and costly with the crawl space pretty cramped in some locations.

          1. Expert Member
            Michael Maines | | #11

            I spec high-quality tape over CDX sheathing, or sometimes Zip sheathing. CDX will likely handle occasional wetting better than OSB, including Zip, and it's airtight as long as you tape the seams. Voids along the edges make it somewhat challenging, so if the exterior insulation has enough R-value to keep the sheathing above the dewpoint temperature, I prefer the ease of Zip sheathing.

            This article goes into more depth on how to keep the sheathing above the dewpoint temperature: The quick version is that in climate zone 6, at least 30% of the R-value of the wall assembly needs to be on the exterior, or about R-12. If you don't have that much R-value on the exterior, as in your current design, it's important for the exterior insulation to be vapor-open, which Steico is so you're safe. But you'd be safer if you used more exterior insulation.

            Another option for making the sheathing layer airtight is to use a fully self-adhered WRB such as Henry Blueskin VP100, Siga Majvest SA or Pro Clima Adhero or Rothoblaas Transpir. It would go directly on the sheathing, and the Steico would go outside of that.

            Your previous design had exposed spray foam in the floor system so I'm not sure why you're now concerned about exposed rigid foam. In any case, polyiso and EPS are both available with a foil facing that is pest-resistant. Or you could use Zip-R, or install plywood or OSB under the rigid foam.

            When building a high-performance home there are lots of things that subcontractors will not be used to. It is generally a good idea to keep plumbing out of exterior walls but with continuous exterior insulation it can be done safely.

  2. DavidDrake | | #2

    Hi Soulincode,
    A couple questions: How are you enclosing the crawlspace between concrete piers? (It's hard for me to tell from your section and photos.) Will it just be the coil stock skirting? How is the poly vapor barrier integrated with the crawlspace enclosure between piers?

    Is the reason for using wood fiber insulation rather than something like EPS concern over EPS global warming potential? If so, it seems your costs might be lower and carbon footprint unchanged if you swap wood fiber for EPS, and swap HFO spray for a carbon neutral floor joist insulation. Maybe flash-and-batt if spray foam is key to you air-sealing strategy.

    Does it make sense in your climate to consider the rainscreen siding detail a premium cost that can be cut? I'd guess a rainscreen is cheap insurance to protect your siding and your walls, and should be included regardless of other cost-cutting measures.


    1. Deleted | | #6


    2. soulincode | | #7

      Thanks David…

      We have not finalized how to handle skirting between the piers and around bottom of house. The rough plan so far is getting board and batten pretty low, then it seems skirting needs to be metal or stone. This Duraskirt product caught my eye:

      It looks like input is to get it 6" into ground, then backfill. We'd need to add some 2X4 framing as shown in that link. I'm thinking something like this would keep out critters, and the poly vapor barrier could be attached to the 2x4 framing. I wonder if anyone has experience with Duraskirt or other options? It seems either way, the crawl space might need vents in this case? To keep air flow, move radon if present, etc?

    3. soulincode | | #9

      For wood fiber rather than EPS, yes, it's global warming mainly, also vapor permeability, where we would not need to figure exactly on the dew point situation. On flash and batt option, I think builder handling the 2X6 insulation was hoping to DIY with R23 mineral wool, without dealing with spray contractor too, with air sealing layer Siga Majrex or Intello because of variable vapor control movement. This said, I'm not sure if we nix the exterior insulation, where this setup leaves the dew point and vapor movement in and out.

      For the floor joists and HFO spray, the challenge has been how to air seal underneath with alternative insulation methods, because it will be super labor and cost intensive, and maybe not a great air seal in the end, with difficulty to crawl under there and air seal. I suppose we could try to air seal from above on sub-floor. So far, most everyone suggested the closed cell foam solution is best for this kind of difficult crawl space situation. Maybe you have some ideas...

      With rain screen, the builder is passive and almost always does rain screens, but said he's noticed so many critters in this area, it may be be better, considering use of board and batten somewhat affordable hemlock for siding on this project, to skip rain screen so the cladding is tight and reduces crawl space for critters.

      1. DavidDrake | | #12

        As a clarification, I was suggesting flash-and-batt for the floor joists bays, NOT the wall cavities.

        Idea would be to save some of the cost and GHG emissions associated with HFO spray foam, by using less of it.

        Now, it may be that labor on spray foam is high enough that the cost difference between a thin layer of foam to air seal versus a thick layer that takes care of all the underfloor insulation is insignificant, esp. with the added labor of installing fluffy insulation after the thin layer of foam if you did flash-and-batt.

        1. soulincode | | #14

          Thanks for clarifying David. I think the challenge of flash and batt for floor joists is the fluffy stuff would probably go out to edge of floor joists, then need plywood layer to keep critters from nesting in it. So there are three layers of labor in quite limited space. I would much prefer something like this, also continuous insulation layer over the floor joists, but the cost would be so much higher. Renovating for high performance and eco friendliness is tough.

          Am meanwhile curious if you have thoughts about that Duraskirt above, in answer to your question...

          1. DavidDrake | | #15

            You don't say how many LF the perimeter of your building is, but I'm guessing about 16o LF (based on your drawing and earlier post indicating 1500 SF). Looks like 24" Duraskirt for that length is about $5120 @ $32/LF, before shipping from WA State and taxes. Figure $5500-6000 all in. Not cheap.

            If you cut 1/2" x 3' x 5' Durock or similar in half, it's $1.50/LF, or about $250 for your project, and presumably available from the local home store without shipping.

            Is it rated for below grade use? No, but as others have said: if you've got drain rock all around and below, you should be fine. Figure some extra labor for cutting the panels in half, and materials and labor for a skim coat of stucco to make it look pretty, maybe you're in for $1000 or so. At that cost, it's not the end of the world if it isn't perfect or under warranty.

            If it were me, and I had all the old siding torn off and the crawlspace exposed, I'd use the opportunity to make sure the crawlspace was at least 18" from grade to joist, and probably deeper. Trying to get a good seal against moisture from the ground by wrapping poly around piers and then sealing it to the skirting doesn't sound easy, but since it's not air sealing, perhaps good enough is fine.

            Also, if it were me, I'd probably go with Michael's suggestion to use fluffy insulation in the joist bays and a thin layer of foil-faced polyiso below the joists. In my areas, materials cost for that would be about $2600, assuming 1500 SF floor area. If you're still worried about 'critters' chewing through the foil-faced polyiso, tack a layer of hardware cloth or chicken wire over the foam. You don't mention cost for 4" installed HFO foam, but I can't imagine it's cheap.

            Now that you've got the GHG emitters (mostly) out of the floors, maybe you won't feel so bad about using EPS on the exterior. Plus, its probably manufactured more or less locally—not sure that's the case with wood fiber. I calculate the$/R-BFof Type 1 EPS available in my area as ~$0.05—only cellulose is cheaper (material costs only, not installed).

            Or (again, Michael's suggestion): forget the exterior insulation and rely on the service cavity insulation for your thermal break. Or use Zip R. Or whatever. If you eliminate the wood fiber and the spray foam through carefully considered alternatives, I'll bet you'll save at least the $20K you've budgeted for PV, without decreasing performance or increasing your carbon footprint.

  3. Expert Member
    Akos | | #3

    I would look at Zip R or similar nailbase panel. This would give you a thermal break plus a solid surface to attach your rain screen/siding. It would also eliminate lot of the cost in your option 1. There is also the benefit of exterior insulation is that you don't loose interior floor space.

    Taped Zip can also become your main air barrier which is much easier to detail properly than an interior one.

    1. soulincode | | #5

      Hi Akos,

      Thanks, I have a feeling the builder would say the Zip R stops drying options outward, versus wood fiber. Also unsure about the OSB and longevity. I do see how it could reduce costs though.

      With losing interior floor space, I feel like builder would still want to have air seal layer on inside and service cavity, as just relying on the zip tape as only air seal may fail. This said, we were only talking about the one layer inside as primary air seal. It's a bit of a balancing act.

      1. Expert Member
        Michael Maines | | #13

        Zip-R actually doesn't block outward drying; the least-permeable part of the assembly is the OSB so it's not much different from plain Zip in that regard. But in climate zone 6, with 2x6 framing, you need to use at least 2" of exterior polyiso to keep the interior face above the dewpoint temperature. That means using their Zip R-12 product, which requires 4" framing nails and a special nail gun that can shoot nails that long, and even then it's hard to get the nails set at the right depth.

  4. soulincode | | #16

    Hi all,

    Thanks SO MUCH for the replies. This past week, I talked with a good Steico distributer my builder uses, and they thought the exterior insulation assembly drawing I shared (same one as included here on this post) looked good and would work well.

    However, the builder came by the site yesterday to discuss the two options, with thermal break and without it, and noted he feels better going with the "R32.13 w/o Thermal Break" option is best for this retrofit. Here are the reasons:

    * The inside of the house is gutted and he can tell for sure that this option will work ok for the air seal. He knows how he can connect the air seal between wall and ceiling, along with sealing the top plates.

    * The outside of the house still has T111 siding and old soffits, with the roofer coming this Monday to redo the roof, along with soffits and fascia. This will make it so he can't do an exterior WRB air seal layer that goes up the wall, past soffit area, and connects at least to a top plate liquid seal on each side, ie between exterior wall and interior ceiling air seal.

    * Overall, he feels there are too many unknowns with the exterior option, which could became issues and more cost, and while achieving a thermal break, may result in a worse air seal.

    As far as dew point, condensation, and vapor drive, we talked about it, how we're mainly dealing with outward vapor drive in Climate Zone 6 with hot air inside, cold outside, and he feels as long as the air barrier inside holds up, this will stop that vapor from getting into the walls. He's a passive house builder with a good amount of experience, so I think the job will be done well. It does leave me with some concern, but it also doesn't really seem feasible to do exterior insulation with this renovation.

    Having the air barrier inside the 2X6 wall will require this service cavity for the electric. We'll also add R value there as well. See the attached assembly for latest. Any input is welcome, in terms of some things we may do to improve longevity of this assembly.

    Fyi, with regard to the insulation in the floor, thanks. I'm thinking we'll leave the floor uninsulated over this winter and tackle the floor in the spring, to see if there's a more affordable, working solution. There are good ideas shared here and I need to compare, as the current cost for 4" HFO closed cell R30 is $11,200.00. I'd like to reduce cost and emissions / off gassing if possible down there, and will post separately on this. I'm also going to look into DIY loose fill cellulose in attic, as it seems like a place to save on cost.

    Thanks again, I appreciate the people on this website for their thoughts, knowledge and caring for building science. I agree 100% with Michael there is a need in the end for the knowledgable local builder on the ground. At same time, I find that bouncing ideas here is very helpful to educate myself and help in that conversation, also the resulting decisions.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |