Exterior foam justification
I’m trying to make a determination about diminishing returns…
We’re building a house in zone 5 (near Colorado Springs at about 7400 ft, some wind exposure). It’s about 2800 sqft, ranch with basement, 2×6 framing. Heating will be hydronic radiant floors – staple up (radiant for comfort, staple up is our builder’s preference). We’ll have 3 ft. overhangs for architectural stying, but hope the shade/shelter will help, too. Our builder is an eager learner and was already working hard on airtightness, but I’ve been educating him on ways to take it up a step to Pretty Good or PassiveHouse.
Current insulation plan is 5.5in. open cell spray foam on the main floor, 2in. closed cell on the attic floor with R-38 (possibly R-49) blown fiberglass on top, and 2in. closed-cell in the basement.
Our builder hadn’t done a double-stud wall before, so we decided to work with foam instead. He recently finished his own house with just 5.5in open cell; the insulation subcontractor thinks that’s enough for our climate and any more is overkill, so I’m getting some raised eyebrows about whether or not external foam is necessary/worthwhile.
I would really like to do 2in. rigid foam (XPS) on the exterior, but other considerations could be to incorrporate Zip or Zip-R3. Cost-wise, the 2in. XPS ends up very close to Zip with 1in. XPS, or just Zip-R3. (I’ve seen that Huber doesn’t recommended foam over the Zip-R and that would be the most costly option anyway. We didn’t get pricing for Zip-R6.)
I’ve seen the 2012 IRC requirements for >R7.5 for 2×6 walls, so realize that our safest options are probably either the basic Zip (no exterior foam) or the 2in. XPS.
I realize this isn’t a new issue, but any thoughts on the above plan and reassurance/comments on the benefit of the exterior foam added cost for our situation would be appreciated. Thanks in advance, Alex
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Alex,
First of all, your plan to insulate your basement walls to R-12 doesn't meet minimum code requirements. The 2012 IRC calls for a minimum of R-15 continuous insulation for basement walls in your climate zone.
Second, your instinct that exterior rigid foam might be a good way to insulate your above-grade walls is a good one. In your climate zone, the minimum acceptable R-value for the rigid foam layer on a 2x6 wall is R-7.5. For more information on this topic, see Calculating the Minimum Thickness of Rigid Foam Sheathing.
Green builders avoid the use of XPS, which is manufactured with a blowing agent that has a high global warming potential. Either EPS or polyiso can work; if you choose to install polyiso, you should de-rate the R-value to account for the poor performance of polyiso at cold temperatures. Assume that the polyiso performs at R-4.5 per inch or R-5 per inch, not R-6.
Once your walls are insulated on the exterior with rigid foam, there is little justification for the use of spray foam between the studs. I would insulate between the studs with cellulose if I were you.
AW,
I have reasoned through the same issues for my house in Michigan, also zone 5. The majority of what I know about this subject comes from GBA. We are just now beginning actual construction, so I am not yet in a position to discuss the lessons learned.
After my research, I decided that the best way to build walls was 2x6 framing with 2" of exterior EPS and cellulose in the cavities. I got quite a lot of resistance from my builder. He has used exterior foam once before and it wasn't a pleasant experience for him. On hat project, there was a lot of standing around the job site figuring out how to do stuff and extra cost due to unanticipated issues. Customary construction in my region is 2x6 framing with cellulose or fiberglass. The few builders that want to claim superior insulation use flash and fill. I insisted so we are proceeding with 2" of Insulfoam Platinum. That has about the same R-value as XPS and much less global warming potential. We have to order it, though. It is not available locally. We will use spray foam in strategic locations mostly for the air sealing benefits. The upcharge I am paying is higher than I expected it to be. Early adopters get stuck paying for the learning curve. We have tried to ward off some of the issues he had before by having details drawn and doing the debating during the design stage. Time will tell whether we succeeded.
Since it is my money, nobody has forced me to produce a well-reasoned justification for the extra cost. At current energy prices, I don't think I could justify the cost purely based on the money it will save for heating and cooling. I believe that addressing thermal bridging will improve the comfort of the home. That is hard to quantify and put a price tag on. I believe that my wall will have better durability than the customary wall, but I don't have any evidence that the durability of customary walls is a problem in my area. I get enjoyment out of learning and experimenting, so that aspect has some direct benefit for me. For others, that would be a downside. I feel like helping my builder and his subcontractors climb the learning curve is sort of a public service that has some value. These factors in combination make me feel good about my decision to insist on exterior foam.
"....2in. closed cell on the attic floor with R-38 (possibly R-49) blown fiberglass on top..."
First, there's no good rationale for the 2" closed cell foam with a vented attic in a zone 5 climate, and like XPS it's blown with a very high global warming potential blowing agent. The attic floor can be air sealed in any number of ways at lower cost an lower environmental impact than 2" of closed cell foam.
Second, fiberglass is slightly translucent to infra-red radiation and will underperform it's rated-R during the cooling season, since the hottest point in the insulation will be an inch or three below the top surface, and hotter than the attic air. (Insulating against a higher temp with less insulation.) If you're only going for R50-R75 using16-25" (initial depth) cellulose would be a better option. (Much higher R than that you have weight issues to factor in given the higher density of cellulose vs. fiberglass.)
There is more value to exterior foam than a mere net-present-value of future energy cost savings. In particular it makes the assembly more moisture resilient, since it can be set up to dry easily toward the conditioned interior.
Staple-up radiant has different meanings to different people, but it requires higher water temperatures than methods using aluminum heat spreader plates. Even the cheapest sheet metal heat spreaders will lower the temperature requirements (good for efficiency) and limit the amount of obvious temperature striping on the floor. Staple-up EPDM tubing goes up very easily & quickly and has better heat transfer than simple staple-up PEX, but EPDM tubing has an abyssmal track record in the industry, and should be considered a higher risk to the boiler, pumps and other components than PEX.
In new construction it's often cheaper to pour the foundation in a minimalist 2" + 2" EPS insulated concrete form than to spray 2" of close-cell foam, and unlike the 2" of closed cell foam, it would meet code min. With a 2" + 2" EPS ICF and 2" of exterior foam on the framed walls it's easy to park the foundation sill directly at the edge of the concrete to create a continuous foam thermal break, with a 1/2" cantilever (the thickness of the sheathing), which allows some depth for finishing the exterior of the ICF foam without creating drainage issue. This works best with "outie" windows, where the drain-plane is located on the exterior side of the foam, otherwise there would have to be Z-flashing and drainage space allowance where the above-grade foam meets the ICF foam, interrupting the continuity of the foam thermal break.
"Second, fiberglass is slightly translucent to infra-red radiation and will underperform it's rated-R during the cooling season, since the hottest point in the insulation will be an inch or three below the top surface, and hotter than the attic air. (Insulating against a higher temp with less insulation."
Eek, i have never heard of this, is there more information on this phenomenon?
Heat transfer inside insulation is pretty complicated. It includes conduction through the solid portions and through the gas, convection within cells or in larger loops if the insulation allows that, radiation across cells, and radiation through some of the cells walls if they are thin enough and translucent enough to allow that. The translucency issue is not just an issue for fiberglass. It's also an issue for foam. Graphite infused EPS (e.g., Neopor brand) works by being less translucent to IR. EPS is translucent to IR only on a very short length scale (a few mm IIRC), but that's big compared to the pore size, so it does have a significant effect on overall R-value.
However, a building designer doesn't normally need to worry about those details, because a standard R-value measurement includes all of those modes of heat transfer through the insulation. The standard R-value test has the insulation in contact with a plate on each side. Ordinarily we say that radiation doesn't matter if the insulation is in contact with a surface. But in fact, for insulation that is translucent to IR, some of the heat flow through the insulation is radiation from the plate surface through the first "inch or three" (I'm assuming Dana has the numbers right on that distance since he usually has numbers right.) of the insulation, that is then absorbed and passed on by some combination of heat flow mechanisms.
So although that's a real effect, it's already captured to some extent in the R-value test (assuming the test is done with a high-emissivity plate, which I believe is standard). Dana's correct that in the scenario he decribes, it is worse than the R-value test indicates, because the radiant temperature at the surface of the insulation is higher than the air temperature, whereas in the R-value test, the two are the same. But it's not like it's a totally new effect that is completely missed by the R-value test.
The resistance to wind washing is another reason to prefer cellulose to fiberglass for attics, at least for the top layer.
Reid,
What a sensible and pragmatic approach. I bet that type of thinking will stand you in good stead during your build.
Thank you for all of the comments. I'm fascinated by how challenging these decisions can be by the time you juggle all of the variables, so it's really helpful to hear other perspectives. I also enjoy the researching process but it's frustrating when you don't have easy access to local costs.
There seems to be good consensus that the exterior foam really is a good way to go. I should have mentioned that the source of my 2012 IRC reference was that article and also the follow-up after the revision last year. I missed the basement requirements.
I'm hoping for the same comfort benefits, and hopefully acoustic reduction, too. I'll have to look into the Insulfoam products. How does their cost generally compare to XPS? Is the durability of EPS in this use adequate? I wasn't aware of the graphite versions of EPS, so more research... I also just found another one of Dana's replies about considering recycled foam. I'll have to look into that more, too. If it helps, we're doing stucco and manufactured stone veneer cladding, which are the most common finishes in this area.
The insulation sub said he had seen too many cases of cellulose settling to be comfortable with it. Of course, he is now very heavily involved with spray foam so I understand the potential for bias. I may have to revisit that as I realize that cellulose is probably better acoustically than spray foam.
Thanks for catching the concern in the basement. I had been referencing several of the BSC articles for Foundation options and had overlooked the insulator's default to 2in closed-cell instead of the 3.5in. recommended. The standard around here seems to be R13 unfaced batt for finished areas and R11 MBI for unfinished. Our current insulation bid actually still includes those, though I'm not sure that's the intent. He simply added the 2in closed cell to the end of it. Flash-and-batt is also acceptable in the basement, correct? What other options should I consider that would still be relatively easy to implement? We poured the footings this week, so too late for ICF. I considered it but there aren't that many local installers around here.
With the radiant floor, we will have aluminum plates. I think we're using PEX tubing but will double-check that. Fortunately, the 2in. sub-slab foam (polyiso) was an easy sell, though it sounds like blankets are still more common here. They're supposed to be R-3 but I have a hard time believing that after they're compressed. I've seen articles on the slightly lower energy transfer efficiency with staple-up but it seems that the easier and lighter installation makes it nearly a wash. (About 35 vs 40 BTU/sqft as I recall.)
I struggled with the attic options, too. Our builder routinely uses canned spray foam for air sealing, followed by blown fiberglass, so I thought the spray foam would be a good way to stretch his comfort zone. We're planning a metal roof, so I was also hoping for some acoustic benefit there, too. I'm certainly open to other suggestions.
Somewhat related to that, it seems that good top plate sealing can also be a challenge. Any suggestions on how to help my builder with that? I'm also hoping to use EPDM mudsill gaskets (Conservation Technology is currently out of stock of the 2x6 version). EPDM is a lot more affordable than the ProtectoWrap.
Thanks again for the helpful discussion, Alex
"I struggled with the attic options, too. Our builder routinely uses canned spray foam for air sealing, followed by blown fiberglass, so I thought the spray foam would be a good way to stretch his comfort zone"
1-part canned spray foam doesn't have the blowing agent environmental liability that 2-part spray foam (as would be used if you did the whole surface) does. In addition, it's possible for 2-part spray foaming jobs to go wrong. I'm not sure whether you'd be having your builder do that himself or hire it out, but I would not want my house to be someone's first experiment with 2-part spray foam. If the mix is off, you end up with the smells and health problems of a spray foam job gone bad. For both of those reasons, I'd let the builder stay in his comfort zone. Especially since you'd get the same performance at lower cost with his approach.
The Insulfoam Platinum that Reid mentions is the graphite infused EPS that I referred to. BASF Neopor is another brand name. Its main advantage relative to EPS is that you get a little more R-value per inch. I doubt you get more R-value per dollar, but if you are space constrained, or constrained by your builder's comfort level with thick insulation, the R-value per inch can matter.
"I had been referencing several of the BSC articles for Foundation options and had overlooked the insulator's default to 2in closed-cell instead of the 3.5in. recommended."
As a retrofit it sometimes makes sense to go with closed cell foam on the interior side of foundations particularly on uneven stone, CMU, or brick foundations. But for new construction it's not the best bang/buck. They usually limit the closed cell spray foam to 2" because that's the deepest that can be safely sprayed in a single lift. Being the most expensive type of commonly used insulation, the economics of another lift rarely work, even though at 2" it's below current code-min.
"With the radiant floor, we will have aluminum plates. I think we're using PEX tubing but will double-check that. Fortunately, the 2in. sub-slab foam (polyiso) was an easy sell, though it sounds like blankets are still more common here."
????
You can't use polyiso under slabs or in contact with soil since it wicks and retains water in those environments.
With under-the subfloor radiant floors it's better to use cheap R11-R19 "contractor roll" batts snugged right up to the tubing & plates for zone isolation, no need for the more expensive rigid foam.
"I struggled with the attic options, too. Our builder routinely uses canned spray foam for air sealing, followed by blown fiberglass, so I thought the spray foam would be a good way to stretch his comfort zone. We're planning a metal roof, so I was also hoping for some acoustic benefit there, too."
The biggest factor for acoustic transfer is making it fully air tight, which can be readily achieved with spot-foaming, and verifying the air tightness with a blower door test before insulating. There's very little acoustic advantage to 2" of closed cell foam buried in fiberglass. There may be some acoustic advantage to using denser, more air-retardent cellulose rather than blown fiberglass though. (I haven't seen good test data on the differences in acoustic properties of open-blown fiber insulation though.)
AW,
You mention stone veneer cladding. Make sure you have a plan for how the cladding will be attached to the house.
Insulfoam Platinum is about R5/inch as opposed to about R4/inch for regular EPS. The price quotes I got for the Platinum were less than XPS or PIS but more per R than regular EPS. My builder was planning on using PIS, so he was happy with the price. (He was unaware of the cold temperature issues with PIS.) We are close to the border of zone 5 and zone 6, so upping the foam R-value to the zone 6 recommendation made me more comfortable. Gaining that additional R value with thicker standard EPS would have been cheaper but it would have pushed the builder farther from his comfort zone and pushed the bricks 1/2 inch off the brick ledge.
Part of my goal was to have the R values of various parts more or less in balance. Going with extra R value in the walls doesn't make sense unless you are exceeding code requirements in the attic and foundation and using better than average windows.
My mistake... Subslab is XPS. The blankets I was referencing are used in place of the subslab rigid foam. Yes, we are planning subfloor batts tucked under the tubing, too. Thank you for the reminder.
I'll definitely need to get a quote on the Insulfoam and Neopor. Our builder says he has asked about installing the stone over the foam and that the installer said he can do it. Again, thank you.
Another difference between XPS and EPS (graphite loaded or not) is that over the first 50 years the XPS will lose12-15% of it's R-value as it's blowing agent leaks out doing it's environmental damage, whereas EPS would be performing about as well as it did on day-1.
Under slabs you the extra thickness of R4.2/inch Type-II EPS relative to the R5/inch Neopor doesn't much matter. It's cheaper to excavate an "extra" quarter inch or so of depth than it is to pay the upcharge for Neopor. On walls the thickness of the foam matters a bit more.