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

GBA blog dated 2015: “How to Insulate a Foundation”

user-6759891 | Posted in General Questions on

I’ve read many articles from various reputable sources as well as this one from GBA http://bit.ly/2lAt69b and I have a follow up question…

Martin, you’ve recently replied to my retrofitting basement questions regarding insulating a 1974 bungalow in climate zone 7a, so thank you.

Based on all of the research I’ve done over the past few months, I’ve been wondering something for a while… what’s the feasibility of insulating an internal basement foundation with a continuous thermal bridge in 2 different layers… the upper half portion, say 4 feet, with a higher density and greater air & vapor impermeable product such as XPS or polyiso, and insulating the bottom half (4′) portion with a slightly lower R-value and more air & vapor permeable product like EPS or Roxul ComfortBoard?

Then the usual 2×4 or 2×6 stud wall with batt insulation and appropriate vapor retarder?

Just curious if there have been studies on this and what the effects or advice is?

Cheers
Todd

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.

Replies

  1. GBA Editor
    Martin Holladay | | #1

    Todd,
    The article you linked to is a Q&A Spotlight focusing on a single case. It's a useful article, but you may also want to read this overview article: How to Insulate a Basement Wall.

    Your idea of insulating a wall with XPS or polyiso near the top, and EPS near the bottom, will work -- but it isn't necessary. A basement wall doesn't need to dry toward the interior. (For more information on that topic, see Joe Lstiburek Discusses Basement Insulation and Vapor Retarders.)

    Your idea of using Roxul mineral wool insulation near the bottom of the wall is definitely a bad idea. Mineral wool insulation is air-permeable, so it will allow humid interior air to contact the cold concrete, resulting in condensation. (Remember that in spring and summer, the bottom of a basement wall is the coldest section of the wall -- the section most at risk for condensation.)

    If you decide to install a 2x4 or 2x6 wall on the interior side of the rigid foam, that wall should not include any polyethylene. The rigid foam acts as a vapor retarder.

    -- Martin Holladay

  2. user-6759891 | | #2

    Thanks Martin, yes, I have read that before and found myself going back to it several times after reading every relevant article on Building Science Corp's site, plus other sources including a variety of articles and QA's on the GBA website.

    If I've interpreted the info on your site and that on BSC's site, incl your article, "Calculating the Minimum Thickness of Rigid Foam Sheathing (http://bit.ly/2l9kb1b) in climate zone 7a I would (1) need rigid foam with a min R10 against the internal foundation wall to create the thermal barrier required to prevent condensation from being able to form, and (2) the product's min permeance rate has to be accounted for in order to pass as a compliment vapor "barrier." Based on this, 2" of XPS would meet these requirements, but more than 2" of EPS would be required to achieve the same results due to its lower R-value and greater perm rates. Am I correct?

    As previously discussed, this wall assembly, coupled with the 2x4" wood @ 24" OC with 4" Roxul CavityRock + 1/2" drywall should hopefully eliminate the additional vapor barrier requirement. I just spoke with my local building inspector about this assembly and my objective to eliminate the vapor barrier, and his response was this:

    "Min code for basement right now is R16, and vapor barrier that meets CAN-CGSB 51.33M and this is almost always 6 mil poly. My assembly may potentially be correct, but code as it stands requires that the vapor barrier be within 1/3 of the total R-value and on the warm side. So based on this, that wall would not pass code."

    For this to pass, I'd need an architect sign off on it, or I could submit a diagram of the assembly with all parts and their tech specs for review, but no guaranty it would be approved.

    So despite better building science, many Canadians' hands are tied due to our strict code requirements for the vapor barrier. I will look further into MemBrain's compliance with our code, but so far nowhere does it specifically state that it complies with CAN-CGSB 51.33M.

    If I can't get around having to add the vapor barrier, from a cost-benefit perspective would you opt to go with 2" type 2 EPS with R8.08 or 1.5" foil-faced polyiso for a similar price per square foot? XPS is just too expensive and I don't want to get too thick or I'll lose too much internal space.

    Cheers
    Todd

  3. Expert Member
    Malcolm Taylor | | #3

    Todd,
    Your inspector was right and wrong. He was right in that 6 mil poly is the most common vapour barrier used, but the code anticipates and lists various other alternatives. He was wrong about the 1/3rd rule for the location. That appears nowhere in the code I've ever been able to find.
    https://www.greenbuildingadvisor.com/community/forum/building-code-questions/59781/poly-and-canadian-building-codes

  4. GBA Editor
    Martin Holladay | | #4

    Canada has the world's best building scientists, but (perhaps) the world's worst code officials.

    (Unfair -- I know. They're just interpreting the code as written, for the most part. But there are still way too many cases where unnecessary polyethylene is required by a half-educated code official.)

  5. Expert Member
    Malcolm Taylor | | #5

    I've championed our system here a few times but recently have realized how dependant it is on the quality of those administering it.

    Our long-time inspector retired a couple of years ago. He was meticulous, checking nail length and spacing, integrity of air barriers - all sorts of other small things. I hardly ever had an inspection when I didn't get pink-slipped, but never resented it - they were corrections that made the building better.

    Now I often have to point out code requirements to the new guy. The general level of construction in the community has still improved a lot due to code changes, but the enforcement has meant many small but significant things get missed.

    What really annoys me is the "do it like we always have" approach inspectors take - especially when the code has specifically changed to include new knowledge of how building assemblies work.

  6. user-6759891 | | #6

    Thanks, Malcolm. I want to upgrade this house better than what min code requires for obvious reasons, but it's not a new home so I have to draw the line somewhere with respect to cost.

    "code anticipates and lists various other alternatives"... do you have a link to these? I have done extensive research on this and the only link that comes up is nrcan and I don't get access. You'd think it would be easy info to find??

    I was going to go the XPS route to reach that R10 thermal barrier and for its lower perm rate, but it's expensive, etc. I spoke with Beaver Plastics/ AMC Foam and they recommended their Halo Interra EPS foam (http://bit.ly/2l9wY3A), so this looks to be a promising alternative both spec and price-wise.

    I then accidentally just came across this older GBA post "What to do when the ends of the floor joists are embedded in concrete?" (http://bit.ly/2l9DoQk) and realized that I have this bloody issue. I also have the 2 mainfloor wooden horizontal support beams also imbedded in the poured concrete foundation. These extend all the way to the other (external) side of the foundation and their ends are covered with stucco. The attached pic isn't my basement but this looks very much like how mine is built.

    I read the articles Martin mentioned and I did other research on this but everywhere it just says to seal it up (as per normal). Based on my intended insulated basement wall assembly design, are you guys able to offer any recommendations on how I should go about insulating the joist spaces and around the beam?

    It's just one thing after another!

    Cheers
    Todd

  7. Expert Member
    Malcolm Taylor | | #7

    Todd,
    The general requirement is that any material under 60ng(pa.s.m2) can serve as a vb.
    9.25.4.2.(3) gives the requirements if poly is used.
    9.25.4.2(4) for other membrane-type vbs.
    9.25.4.2(5) for coatings.
    9.25.4.2(6) for foam.

    I'm sorry, I don't have enough experience to offer any useful advice about wood embedded in concrete. Hopefully someone else will chime in.

  8. Charlie Sullivan | | #8

    +1 for Halo foam. It's graphite infused EPS which has a higher R-value per inch than regular EPS, and higher than XPS has left after its severely climate-damaging gas has escaped. It also goes by BASF's trademark name for it, "Neopor" and by the confusing acronym GPS (graphite polystyrene). If you search under neopor on GBA you'll hear it praised regularly, and not just by me.

    Sorry I don't know what to do about wood in concrete.

  9. user-6759891 | | #9

    Thanks, Malcolm and Charlie.

    Plasti-Fab has their EnerSpan M-24 Insulation (http://bit.ly/2lKfJn8), which sounds fairly similar, but it's more expensive.

    I'll continue my research into the wood joists in concrete. There's this GBA thread http://bit.ly/2lK99Nk I'll read and keep looking.

    Fyi, I live in Edmonton, AB and was reading some articles/ studies by Andrew Stiffman and he talks about the unique climate we have here relative to much of Canada. They didn't have anything to do with this but the point was our temp diffs and humidity levels don't cause buildings here to have a lot of issues faced by those in other Canadian cities. No idea if this is relevant at all to this situation.

    But the other consideration is the building construction of this home consisting of stucco allows for plenty of outward drying. Other than the old bitumen paper on top of the sheathing there's nothing limiting drying to the outside.

    Thanks again, folks!
    Todd

  10. Expert Member
    Malcolm Taylor | | #10

    Todd,
    I don't know if it is still true but in Alberta, rather than using anchor bolts, the standard way of attaching the floor framing to the foundation was by embedding SPF 2"x4"s on edge in both sides of the concrete wall. Being from moisture-obsessed BC I was astonished, but was reassured by all the builders I know there that the sills didn't suffer any damage over time. You may well be fine.

  11. GBA Editor
    Martin Holladay | | #11

    Todd,
    There are lots of factors affecting the moisture content of joists or beams embedded in concrete. In some cases, embedded joists or beams can stay sound for years. In other cases, they can rot quickly.

    1. The risk is greater if moisture levels are high in the basement air (or crawl space air) and if exterior moisture levels are high.

    2. The risk is greater if the concrete is cold. Clearly, this risk can be addressed by insulating the foundation on the exterior with rigid foam.

    3. The risk is greater if the exterior grade is high and if the exterior of the foundation is surrounded by bushes that prevent sunshine from drying the foundation.

    One solution is to cut off the embedded beam end and to support the beam on a new post on a new footing.

    -- Martin Holladay

  12. Charlie Sullivan | | #12

    One addition to Martin's list note, "The risk is greater if the concrete is cold. Clearly, this risk can be addressed by insulating the foundation on the exterior with rigid foam."

    Exterior mineral wool board could be even better: It keeps the concrete warm and allows faster/easier drying through the insulation.

  13. GBA Editor
    Martin Holladay | | #13

    Charles,
    Good point. Thanks.

  14. user-6759891 | | #14

    Thanks Martin and Charles. This side of the house faces south west so always gets lots of sun throughout the day. I still have some grading to do in the back yard, but as for the soil around house, it's dry and the grading is not too high. Of course, the uninsulated basement has helped keep the ground around the foundation dry.

    I inadvertently cut into the beams coming out of the exterior side when I was breaking off some of the old parging that had been coming off the foundation wall. I soaked the ends of those with some pressure treated solution and let dry for a bit, then recovered with parging. I still have to fix the parging around that whole side of the house so maybe it's a blessing in disguise that I never did break off all of the old stuff and re-parge!

    I could put some ComfortBoard there, which is what I was thinking yesterday when I read Martin and Dana suggest external insulation. So will just evaluate in the spring and do some research on how to best cover that up and then add some flashing. I'd ever consider possibly adding some external insulation and a rain/wind barrier around the house, but that will likely require me having to pull the windows out, etc (I'm assuming) and not sure if this old house would give an ROI.

    So in closing, if I'm going to continue with my basement renos and insulate the walls very shortly, should I still go ahead and seal the floor joist cavities as per normal? My original plan was to add a layer of 1 or 2" XPS and seal and then put 5.5" of Roxul. Stick with that plan or do you think using EPS would be better in this case? I know I could use 2-component spray foam but in case there are any issues that stuff would be a bugger to get off.

    Thanks again for everyone's insight and advice; much appreciated!

    Cheers
    Todd

  15. user-6759891 | | #15

    I'm sorry to do this, but I just realized something... my proposed interior foundation wall assembly is:

    2" Halo Interra GPS foam + 2x4 wood stud @ 24" OC + R14 Roxul + 1/2" drywall + layer of latex primer/sealer + 2 layers of latex paint

    However, Halo Interra is foil-faced and it states that it has a built-in vapor barrier, and just saw its vapor perm rating max per 1" thickness is 0.03, making it a Class I vapor barrier.

    On building Science Corp's website (http://bit.ly/2lUlsXV), it states that this layer should be more permeable than this, but still vapor semi-impermeable in a cold climate like 7a.

    In 2002-2003, CARB performed side-by-side tests of several insulation systems in the basement of a new home in northern Illinois (Figure 6, Aldrich et al. 2006, Zuluaga et al. 2004). Researchers at BSC performed similar evaluations (Ueno 2006, Ueno 2007) and concluded, "...insulation systems were dryer when water vapor was allowed to move between the basement and the concrete wall. Several systems with continuous Class I vapor barriers were more likely to experience higher moisture levels (even liquid water) against the concrete."

    Alternatively, their Exterra GPS foam has a max perm rating of 1.34, making it a Class III, so that may be too much... but it does achieve R ≥10 over 2x4 wall. But this is Canada, not the US.

    The whole point of my design and the choice of building products is to hopefully get around our building code requiring the additional vap bar on the warm side. Fyi, if you do see this, Malcolm, regarding my earlier comment (above) about the vap bar having to be on the warm side within 1/3 of the walls total R-value is, in fact, correct... so that is a code req here. It might not be worded that way and more have to do with initial and latter perm rates. I still hope to get around it IF I can prove my design concept.

    Am I back to square one again and have to pick a new rigid foam to go against the interior foundation???

    Where's my rope...

    Todd

  16. Expert Member
    Malcolm Taylor | | #16

    Todd,

    Thanks for the update. Can you point me to the code reference for the 1/3rd rule? All I can find in ours is the table at 9.25.5.2 Appendix A. It may work out to 1/3rd of the insulation in some circumstances, but not always.

  17. user-6759891 | | #17

    Malcom

    Source: NRCAN, 2017

    "Add a vapour barrier on the warm side of the wall within the first third of the thermal resistance value of the overall wall insulation. (See the 1/3 – 2/3 rule in Section 2.4, Control of moisture flow.)"

    "This rule requires that at least two thirds of the insulation value of the wall is on the cold side of the vapour barrier (see Figure 2-12)."

    This amount of insulating value should be reduced to one quarter or less in very cold climates or in buildings with high moisture sources such as swimming pools."

    Now I'm not certain if this applies only to above grade walls, or if it also applies to below grade wall assemblies. Because the same source also states the following regarding interior basement wall assemblies comprised of rigid foam + stud wall with batt insulation:

    "6.2.4 Frame wall with batt insulation and rigid board insulation
    This method involves gluing rigid board insulation to the foundation wall and then covering it with a frame wall incorporating batt insulation. The result is higher insulation values with less loss of interior space, a very good reduction of thermal bridging and no need for an additional moisture barrier.

    Use only moisture-resistant board insulation such as extruded polystyrene or Type IV expanded polystyrene. Using rigid board insulation with at least RSI 1.76 (R-10).

    Finishing
    Do not use a polyethylene air and vapour barrier with this approach, as there is a risk of creating a double vapour barrier with the foam board. Instead, use either the smart barrier or the ADA method."

    So if I interpret this correctly, in Canada we still have to use at least Polyamide sheeting/ Nylon-6 vapour retarder film or smart barrier... Certainteed's MemBrain would classify as this. Without looking it up, perhaps certain latex paints would also fall under this.

    Now going back to basement interior retrofit, I'd really like to know if the 2" of Halo Interra I was planning to use is acceptable? It's a Class I vapor barrier and will provide an R10. But I have read on Building Science Corp's website (http://bit.ly/2lUlsXV) that a Class I should ideally not be used but, rather, a Class II or III should be used. But in cold climates you don't want to much permeability in this layer. Their Exterra foam has a perm of 1.34 so this may be too much, and would be better if it was <1.

    Halo foam products info: http://bit.ly/2lYYdf7 and tech specs: http://bit.ly/2lZcQz5

    After learning so much over the past month I now feel like I'm right back where I started! If anybody knowledgeable on the subject could please assist, it would be much appreciated!

    Todd

  18. GBA Editor
    Martin Holladay | | #18

    Todd,
    You and Malcolm should put your heads together to interpret Canadian code issues. I'm not an expert on Canadian codes.

    Concerning the building science: You have chosen to specify a foil-faced graphite-infused EPS foam. (It's Neopor.) Don't worry about the vapor permeance of this product. Foil-faced foam works fine in this location.

    While it's true that the concrete wall may be damp, that doesn't matter. Concrete is often damp. Concrete doesn't mind being damp.

    In the past, Joe Lstiburek promoted the idea that rigid foam installed on the interior of a basement wall needed to be vapor-permeable. I disagreed with him on this issue. He eventually came around to my viewpoint. To learn more about this issue, see this article: Joe Lstiburek Discusses Basement Insulation and Vapor Retarders.

    -- Martin Holladay

  19. Expert Member
    Malcolm Taylor | | #19

    Todd,
    I'm not trying to belabour this but do you have a code reference? Your quote comes from NRCAN research. Their work is generally very valuable and represents what they see as best practices, but if it hasn't been incorporated into your building code, you don't have to follow it.

  20. user-6759891 | | #20

    Thanks Martin. One reads this, then that, get's recommendations from this person, gets opinions from others, speaks to manufacturers, etc and it's frustrating to figure out truth from fact when there's so much conflicting info out there and little consensus.

    Malcolm, I can't argue... I've been through all of the code, I've looked up definitions, have spoken to some builders, build prod mfrs, etc. All I can say is my building inspector made reference to this a while ago, and so too did the tech support for Halo. They know what I'm trying to achieve and don't argue with the science behind it, per se, only that I will very likely face resistance from Alberta code officials.

    At the end of the day, current code requires a vapor barrier/retarder that meets CAN-CGSB 51.33M and the other applicable codes are 9.25.4.2.(3-6). Alberta code stipulates that this must be on the warm side of the wall, etc and what I'm seeing is the general consensus of interpretation being towards the "1/3 - 2/3 rule."

    You can see this here: http://bit.ly/2m0zs29... page 25.

    If one looks at the evolution over the Can building codes over the past couple of decades and the recommendations by nrcan and others, I'm sure that one day soon there will be some additional changes with respect to all of this where a min thermal break coupled with max perm rates can negate the req't for an additional vap barrier/ retarder in certain situations.

    All I can do is make a case and submit this to local authorities for their review that includes a drawing of my planned wall assembly, list the details of the layers, provide tech specs for each, list the RSI for each, outline the thermal breaking capacity of my continuous foam layer, and then include supporting evidence from studies done by NRCAN, BSC, GBA, and others.

    Todd

  21. Expert Member
    Malcolm Taylor | | #21

    Todd,
    Good luck. It's never fun to have additional issues you have to fight inspectors on when any construction project puts enough problems on your plate to deal with. Hope things turn out well.

  22. user-6759891 | | #22

    Thanks, and apologies for a being a pain in the arse... just want to do things correctly and that exceed min code, while battling it at the same time!

    Btw, I did previously read your earlier thread http://bit.ly/2l1VS5i.

    I believe I may have misspoken when used the term "fact" and you're more correct when you stated, "The 1/3rd -2/3rd ratio seems to be folk wisdom and certainly isn't in the code."

    Cheers
    Todd

  23. user-6759891 | | #23

    What would you guys do about these things (pic attached) if I intend to install rigid foam on foundation wall?

  24. GBA Editor
    Martin Holladay | | #24

    Are you worried about the rectangular steel washer and nut, or the embedded joists?

    -- Martin Holladay

  25. user-6759891 | | #25

    I was specifically asking about the steel nut... keep it in or cut off and seal? I'm not sure if this one is tying the garage to the house, but I also have a couple of others elsewhere that don't tie into anything.

    I'm about to go around and apply hydraulic cement in all the old cam-lock/ steel tie holes and anywhere there is a little honeycombing, etc before installing foam. They don't leak any water, but I'm just going the extra-mile before insulating.

    Re: embedded joists, I have researched this as well as made an earlier post about this but I don't believe there was any definitive answer. So I was planning to just install the 2" XPS/ GPS foam in the spaces and seal edges with 1-component spray foam.

    1. WinnipegSteph | | #27

      Did you end up installing the 2" XPS/GPS foam in the joist spaces and sealing the edges with spray foam? Trying to figure out what to do with our embedded joist spaces. We are planning to spray foam half our basement this fall with 3 inches, but are unsure of what to do in the joist space. Any advice would be appreciated.

  26. GBA Editor
    Martin Holladay | | #26

    Todd,
    Create a recess in the foam using a spoon or a chisel. Test fit. Then spray a little canned spray foam on the washer and nut, install the rigid foam, and don't worry about it.

    -- Martin Holladay

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |