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

Do We Really Need 12 Inches of Foam Under Our Slab?

A last-minute attempt to apply value engineering to our Passivhaus design

What's the optimal thickness of EPS foam under a basement slab in Maine? Roger Normand is trying to decide whether to install 8 inches or 12 inches of EPS foam under his basement slab. The photo shows EPS installed at a construction project in Portland, Oregon.
Image Credit: Matthew Bradburn - Brooklyn House blog

[Editor’s note: Roger and Lynn Normand are building a Passivhaus in Maine. This is the 21st article in a series that will follow their project from planning through construction.]

After kicking the tires on the Passive House Planning Package (PHPP) results obtained by the Passive House Academy (PHA) on EdgewaterHaus, we have decided to make one design change that, if acceptable to PHA, will save us money and still allow us to comfortably meet the annual heat demand limit set by PHPP.

In a northern climate like Maine, the challenge in meeting the Passivhaus standard lies in achieving the exceptionally low annual heat demand (AHD) of 4.75 kBTU/(ft2•yr). As my previous blog noted, EdgewaterHaus design had an annual heat demand of 3.97 kBTU/(ft2•yr) and earned a Passive House Design Stage Assurance.

Maybe we can install thinner foam

The first place to look at reducing the cost of the building envelope was the 12 inches of EPS foam insulation below the basement slab, or perhaps some of the Roxul Drainboard insulation outside of the ICF foundation. Both are more expensive than the cellulose insulation used to fill the above-grade wall cavities and blown into the attic, and provide less real insulating benefit.

The Roxul and EPS are installed below grade, where the soil moderates temperature extremes, and wind is not an added burden. Because both are installed as the outermost layer of the building envelope, either could be reduced without significantly affecting the design of the house. And since heat rises and cold falls, common sense suggests that any reduction in insulation happen below ground.

The 12 inches of EPS foam was to be installed in 3 layers of 4-inch-thick foam, with taped seams. The foam comes in 4 foot by 8 foot sheets. The EPS costs about $60 per sheet plus tax, tape, and labor. There are some 70 sheets per layer.

Our energy analyst Marc Rosenbaum calculated that going from 12 inches of EPS foam down to 8 inches would raise our AHD from 3.97 to 4.35 kBTU/(ft2•yr), still comfortably below the 4.75 kBTU/(ft2•yr) ceiling.

So, let’s eliminate one of the three layers of EPS.

Oops! The EPS foam has already been ordered

And then a complication. Four-inch-thick EPS foam is not a commonly stocked item. To prepare for the start of construction, our builder Caleb Johnson Architects had already ordered the material and its delivery was imminent; according to the supplier, it was too late to modify the order. They would not accept a return. We owned the EPS.

How disappointing: after over 1 1/2 years in the design phase of our project, we still could not make all the necessary material decisions in time!

Perhaps we could use the 4-inch-thick EPS to substitute for some of the Roxul. From a thermal perspective, that would work. Marc cautioned us to keep the EPS well below grade, because it tends to harbor ants. (Our EPS destined for use well below grade is not treated with an insecticide.)

If you want that type of Roxul, you have to buy a pallet

Another complication: the planned 2 3/8-inch-thick Roxul Drainboard is also not commonly stocked, and unlike the 1-inch-thick version, only comes in pallet-sized quantities. We would also have to match exterior material thicknesses.

Then our architect Chris Briley learned that we may be able to purchase some surplus 2 3/8 inch Roxul retained by another nearby contractor — price to be determined. Chris has drafted a revised drawing showing the reduced amount of Roxul, the EPS topped with flashing to prevent water from the above drainboard to flow between the EPS and the ICF, and the change to only 8 inches of EPS beneath the slab.

Before we implement this design change, Chris has submitted the revised drawing to PHA. We want PHA to confirm that we still meet the Passivhaus annual heat demand limit, and still retain the Passive House Design Stage Assurance. We expect a quick response from PHA, as excavation is imminent.

The first article in this series was Kicking the Tires on a Passivhaus Project. Roger Normand’s construction blog is called EdgewaterHaus.


  1. user-723121 | | #1

    How does PHPP address climate change?
    The 30 year average for heating degree days dropped in Minneapolis from 7876 to 7580 when the average was recalculated in 2010. The decade of the 2000's was nearly 900 hdd warmer than the 30 year running average.

    The average soil temperature has surely risen in the last 10 years so calculations for below grade insulation may have to be adjusted accordingly.

  2. user-757117 | | #2

    Maybe it's too late already...
    Another option for that extra EPS that you own:
    Put it up for sale.

    I haven't found it difficult to get rid of surplus building supplies on the local buy-and-sell.

  3. jinmtvt | | #3

    Passivehaus cost effectivness
    I'd like to see the payback term on the 10th and 12th inches of underslab EPS on this house
    Probably in the 150-200 years +

    let's hope it will still be functional and used by then :)

  4. watercop | | #4

    Another alternative occurs to me...
    Rename it EdgeWater Pretty Good House (yes, I mean "House" not "Haus") select reasonable insulation levels and dispense with all this capricious Teutonic nonsense.

    The question of cost payback has been duly raised. Let me ask also when the planet gets back the energy and CO2 emissions embodied within inches 9,10,11,and 12 of subslab foam placed anywhere south of the Arctic Circle. While I'm at it, Roxul Mineral Wool requires 1500*C (2732*F) during manufacture per Roxul's website, so the odd extra million Btus has gone up in smoke to make that stuff as well.

    This whole topic seems quite asinine, or should I spell that "ausinine" in keeping with the spirit of the subject?

  5. jklingel | | #5

    curt: you are certainly welcome to feel as you wish, but i don't agree w/ your assessment. ok, roxul takes energy to make; are we to use dirt instead? fiberglass? cotton candy? btw: we are south of the arctic circle, and i can justify 10" under a slab; others do so for 12". are you basing your conclusions on the cost of fuel now, or your predicted cost in the future? no one knows what energy will cost down the road, so i think we ought to insulate what we assess as appropriate. finally, as bill clinton might say, define "reasonable".

  6. watercop | | #6

    I'd like to see the load calcs
    revealing energy saved vs energy embodied for these highly elevated levels of insulation.

    This is not about "feelings" but instead cold hard facts. How much extra energy would Pretty Good Edgewater House with 8" of subslab insulation burn over EdgeWaterHaus boasting 12"? How long would that almost-certainly minuscule savings take to make up the embodied energy in 4 more inches of foam?

    Why is it that the arbitrary and capricious PH standard credits solar thermal, with all its attendant cost, complexity, moving parts, roof penetrations and freeze vulnerability but rejects solar photovoltaic? Can you think of a better descriptor than "ausinine" to describe that stance?

    I agree that we all should choose insulation levels and make other energy conservation investments as we individually deem reasonable and appropriate...that's the American way....OTOH PH does NOT allow such choices, rather it dictates arbitrary performance levels from afar, regardless of latitude and climate.

    BTW, I live in the deep south - subslab insulation here is R0.0 since our soil temperature of 70* is quite welcome year round. Note that PH has made zero inroads here since it virtually totally ignores summer humidity, which we must mitigate at whatever cost.

    I simply can't, nor will I try, to base a building science business upon discomfort and deprivation.

  7. user-1030561 | | #7

    Reply to Curt Kinder
    You don't even have to leave this site to find examples of PH in the deep south, here's one in Louisiana, with R16.5 in the slab... a bit more than your R0.0 suggestion, but certainly not unreasonable:

    But PHPP didn't make the Edgewater folks use 12" of foam – that was their design choice, which the PH software allows. Clearly the author states that they could easily meet the PH demand requirements with just 8" of EPS, and I suspect they could even go smaller with other design adjustments.

  8. Mike Eliason | | #8

    wow. lots of misconceptions
    wow. lots of misconceptions there, curt... PH doesn't 'totally ignore summer humidity' at all, and there are definitely 'in-roads' to PH in the south, albeit it much slower than the north. this report recently prepared by the PHI in darmstadt specifically addresses the humidity issue for PHs in mexico.

    regarding the embodied energy payback on that add'l (non-necessary) 4" of subslab EPS? ~ 20 years, the embodied carbon payback is ~3.2.

    and mr. barnes nailed it. 12" sub-slab foam isn't necessary for smart designs or buildings huddled together (duplex/triplex/MFH). even roger's project could have hit the space heating demand w/ 8" AND the form is rather 'loose' for a PH, so an even more compact one might be able to get away w/ even less than 8".

    regarding the 'payback'... economic, MJ and CO2 aren't ‘immediate’ in new construction (or retrofits), but can be recoup'd quickly, even faster in a retrofit situation... in some cases under 3 years for all, depending on climate and electrical costs.

    regarding new construction, we've just calc'd the numbers on an ubersmall PH (reference area = 577 ft2) we're working on...

    6” EPS: SSHD = 4.40 kBTU/ft²a x 577 sf (TFA) = 2540 kBTU/a (744 kWh/a)
    4” EPS: SSHD = 4.88 kBTU/ft²a x 577 sf (TFA) = 2816 kBTU/a (825 kWh/a)
    kWh saved w/ last 2" EPS: 81 kWh/a
    Electricity cost/kWh: $0.09
    Annual savings $7.29/a
    Cost for 2" add'l sub slab foam: $325
    Equivalent PV production: 70 kWh/a
    Embodied carbon of add'l 2" EPS: 30.4kg x 2.5 kgCO2/kg = 76 kg CO2
    Embodied energy of add'l 2" EPS: 30.4kg x 89 MJ/kg = 2,706 MJ

    simple payback of embodied carbon: 81kWh/a x 0.68 kgCO2/kWh = 55 kg CO2 avoided per year, payback = 1 year, 4.6 months (75 kg CO2/55 kg CO2)

    embodied energy payback: 2,706 MJ avoided per year. payback = 9.27 years (2,706 MJ / 292 MJ/a)

    economics simple payback = 44.5 years ($325/$7.29)

    in our case, the last 2" of insulation save more kWh/a than equivalent cost PV can produce (this was actually true everywhere we migrated the model). granted, the economic payback is longer than we'd like, but our electrical costs are pretty low and we won’t be moving the building anytime soon. when we 'migrated' the project to other heating-dominated climates (boston, minneapolis, omaha, maine) we found that the economic payback for the 'last 2 inches of sub-slab foam' was under 20 years. the embodied CO2 and energy paybacks were much quicker in those climates as well.

    however, that's subject to change - we're in for some nice rate hikes over the next few years.

    however, i'd submit that only looking at PH through a narrow lens (e.g. focusing solely on sub-slab foam, especially of non-optimized projects) will always distort the numbers.

  9. jklingel | | #9

    mike: that is quite
    mike: that is quite interesting. what, in general, does one spend on a complete investigation like this, and what all does it entail? not necessarily your company, just a general question. i don't mean to highjack this post, so feel free to email me at [email protected] if you care to. thanks. john

  10. GBA Editor
    Martin Holladay | | #10

    Response to Mike Eliason
    Thanks for your comment, and thanks for your further analysis on this issue, which you have posted on your blog.

    I posted a response on your blog; I am also posting my comments below.

    Your analysis shows that the optimum thickness of EPS foam (which I'm assuming is about R-4 per inch) is 4 inches in Portland (R-16) and 6 inches in most other locations (R-24). That's entirely consistent with John Straube's calculations, and is a conclusion which matches what I have been reporting for years.

    In my August 2009 article on the topic ("Can Foam Insulation Be Too Thick?"), I quoted John Straube, who said, “The cost of insulation becomes more than the cost of generating energy for ... slabs [on grade] at about R-20 to R-25, depending the cost of placing EPS (which costs around 10 cents per R per square foot)."

    So it looks like we are in agreement.

    The projects that I have been questioning are those with between 10 inches and 16 inches of rigid foam under the slab.

  11. Kopper37 | | #11

    Passivhaus in Hot Climates
    A friend sent me this link. Looks like Passivhaus is being investigated in the oil rich Middle East:

    Looks like a thoughtful project. Reminds me of some FSEC (Florida Solar Energy Center) studies.

  12. watercop | | #12

    Hot Humid PH
    The Mexican PH modeling document is interesting - I only skimmed it, but for the first time I see PH accommodating the concept that separate recirculating cooling systems are needed (not just ERV) to manage both sensible and latent loads, at least in the case of the Cancun climate, which Florida closely emulates. That's a refreshing change.

    I know of no PH projects in Florida. My limited understanding of the one in New Orleans is that it, at least initially, failed to provide relief from summer humidity.

    I'll stipulate the embodied energy data posted above, but stand by my stance that given finite budgets, investing in PV makes more sense than extremely high performance building assemblies such as feet of sub slab insulation or windows available only from Germany.

    While I'm all about investment in conservation, and understand that some hardy folk are willing to tolerate oppressive humidity in the name of extreme energy performance, I cannot expect clients to endure discomfort and deprivation.

  13. KHWillets | | #13

    R-value of soil
    I'm still surprised that the R-value of soil is not accounted for. Dry soil can reach R-10 per foot, and even a conservative value of R-1 to R-4 would mean that foam should taper towards the interior.

  14. Mike Eliason | | #14

    just because you don't


    just because you don't know of any doesn't mean they don't exist. 5 minutes on teh googles and i found 3 projects in planning. and in that climate, you won't need 'feet of sub slab insulation' nor windows from germany.

  15. watercop | | #15

    The best laid plans...
    ...get planned to lay.

    I was in on one a couple years ago, and it foundered on fear of not being able to comfortably control humidity within the tight (and arbitrary) confines of PH.

    There seem to me other faster cheaper paths to net zero than PH in hot humid climates.

  16. Mike Eliason | | #16

    K Willets,
    The R-value of the

    K Willets,

    The R-value of the soil is accounted for in PHPP. customizable per soil type, as is heat capacity, and if you want to be hyper nerdy - groundwater depth & flow rates, etc.

  17. KHWillets | | #17

    thanks mike
    I had tried to find that out, but failed. I wish PHPP was open-source.

    Offhand, do you know what values it uses? I could not find a standard reference.

  18. Mike Eliason | | #18

    believe me, you aren't the
    believe me, you aren't the only one - however it's not uber expensive. the hard part is learning how to use it.

    for defaults, you mean? i believe really low, ~R-1/foot for thermal conductivity. but have ability to model wide variety.

  19. Roy devries | | #19

    Damn near Passive House in Canada
    We are currently working on a project in Canada and truly appreciate the wisdom, insight and experiences of those who have done this before us. The information is invaluable.

    If interested our website is and all comments, insights and experiences are welcomed. Thanks

  20. user-2405374 | | #20

    Heat Doesn't Rise
    "And since heat rises and cold falls, common sense suggests that any reduction in insulation happen below ground." I think it's important to note that heat doesn't rise, warm air does. With an airtight building the stack effect of warm air rising is downplayed significantly. The difference between the ground temp and the ambient air is the reason you can move insulation from the ground and attach to the walls or roof.

  21. ErgoDesk | | #21

    Re. Passivhaus in Hot Climates by Daniel Ernst
    This is an EPS Composite type Envelope that is basically using the strength of the EPS as the framing and then rendered with Thin-Shell Concrete. This combination is ideal for speed, and lack of any thermal breaks. Air tightness is achieved without the usual tricks and blower door test are always achieving high results. Construction cost can be easily reduced by 25%.

Log in or create an account to post a comment.


Recent Questions and Replies

  • |
  • |
  • |
  • |