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
Musings of an Energy Nerd

Can Foam Insulation Be Too Thick?

Determining the best thickness for sub-slab foam

That’s a lot of foam. The designers of the first certified Passivhaus building in the U.S., a language institute in Bemidji, Minnesota, specified 16 inches of expanded polystyrene (EPS) foam under the foundation slab.
Image Credit: Edwin Dehler-Seter

In the U.S., designers of cutting-edge superinsulated homes generally recommend 2 to 6 inches of rigid foam insulation under residential slabs. For builders who use extruded polystyrene (XPS), the most commonly used sub-slab insulation, that amounts to R-10 to R-30.

As Alex Wilson recently reported, “Building science expert Joe Lstiburek … argues that for any house north of the Mason-Dixon Line we should follow the ‘10-20-40-60 rule’ for R-values: R-10 under foundation floor slabs; R-20 foundation walls; R-40 house walls, and R-60 ceilings or roofs.”

For reasons that are somewhat murky, however, Passivhaus builders install much thicker layers of sub-slab insulation than most superinsulation nerds.

Passivhaus buildings have very thick sub-slab foam

Passivhaus designers use an oft-praised software package developed with the help of German physicists — the Passive House Planning Package (PHPP). The PHPP software helps designers determine how thick insulation needs to be for a house to achieve the Passivhaus standard. Among the key requirements of the standard: the house must have a maximum annual heating energy use of 15 kWh per square meter (4,755 Btu per square foot) and maximum source energy use for all purposes of 120 kWh per square meter (11.1 kWh per square foot).

To meet the standard, Katrin Klingenberg, the founder of Passive House Institute U.S., installed 14 inches of expanded polystyrene (EPS) insulation — 7 layers of 2-inch foam (a total of R-56) — under the slab of her home in Urbana, Illinois. The Waldsee Biohaus, a Passivhaus language institute in Bemidji, Minnesota, has 16 inches of EPS under its foundation slab.

What’s the explanation for these differing recommendations?

I recently approached engineer John Straube in hopes of satisfying my curiosity on the surprising disparity between the sub-slab insulation recommendations of North American physicists and Passivhaus advocates. John Straube is a colleague of Joe Lstiburek at the Building Science Corporation,…

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. Katrin | | #1

    Hi everybody,

    I have posted a reply to this article on the Passive House Institute bulletin board.

  2. Jon Vara | | #2

    Can Foam Insulation Be Too Thick?
    Great piece, Martin. One question: in the first part of the story, you're having a conversation with John Straube, then partway through you shift to saying "Straube wrote." It's not clear where he's writing. An article or paper on the subject? Personal emails to you? If the former, it would be useful to know the source.

  3. GBA Editor
    Martin Holladay | | #3

    He sent me an e-mail
    After I spoke with John Straube on this subject at the Westford Building Science Symposium — a.k.a. "Summer Camp" — we had an exchange of e-mails on this topic.

    Thanks for your comment. I'll clarify the ambiguity in the text.

  4. homedesign | | #4

    You are writing about the things that I am struggling with.
    The problem that I have with PV is that it is ugly and it usually involves holes in the roof.
    And what about when it is time to replace the roof cladding?
    I prefer the German approach.
    My clients do not want to see solar acne on their homes and neither do I.

  5. Jon Vara | | #5

    "Solar acne?" The horror! And get rid of those damned outdoor clotheslines! We're trying to look at nice houses here!

  6. GBA Editor
    MIKE GUERTIN | | #6

    One Comment - Two Questions
    I struggled to speculate why Passivhaus case studies typically used such thick slab insulation. Now I get it; great investigation of the rational and sorting out alternative positions.

    Two questions for you Martin: In one calculation you wrote ” …the cost of PV-powered heat is no more than 60/2.5 = 24 cents per kWh.” I’m presuming the ‘60’ is from the “50 to 60 cents per kWh” you noted earlier?
    And what is the best / most common way to express cost per R of various insulations? You used “…EPS (which costs around 10 cents per R per board foot).” Is it appropriate to use ‘Cost / R / board foot’ or ‘Cost / R / square foot’? Since different insulation materials have different R values per inch thickness and a board foot measures volume (12”x12”x1”) I’ve used the Cost/R/ square foot (area (12”x12”) but that may be wrong. I get confused because the resulting number (144) is the same but the unit changes.
    [Editor's note: In response to Mike Guertin's perceptive question, the reference to "cost / R / board foot" in the original article has been edited to read "cost / R / square foot."]

  7. homedesign | | #7

    warped perception of science
    Martin & Jon Straube,
    Forgive me....I am an Architect ... not a Scientist.
    It is difficult for me to imagine that heat transfer is exactly the same from House to Air as from House to Earth when DeltaT is the same.
    I visualize the Earth as an easier "path" than the air....
    Maybe the difference is not very great?
    Are they really the same?

  8. homedesign | | #8

    Sorry Jon,
    I have seen homes with pv integrated into the design ....and it can be attractive.
    Or at least not offensive.
    More times than not it is less than attractive..

  9. GBA Editor
    Martin Holladay | | #9

    Cost of insulation
    John Straube wrote "10 cents per R per board foot." I agree with you that the units we should be discussing are the cost per R per SQUARE foot. If an insulation product costs 10 cents per R per SQUARE foot, then R-10 of insulation costs $1 per square foot, or $100 for 100 square feet. Or R-20 insulation would cost $2 per square foot or $200 for 100 square feet.

    Before I change a direct quote from John Straube, however, I'll give him a chance to respond.

  10. GBA Editor
    Martin Holladay | | #10

    50 to 60 cents per kWh
    You're right that the number 60 in the calculation 60/2.5 = 24 cents per kWh comes from the high end of the range of PV prices discussed earlier by John Straube. He figures that PV electricity costs between 50 and 60 cents per kWh.

  11. AndrewInChelseaQC | | #11

    PV means you have one more trade to schedule
    Hi Martin,

    One thing that got left out in your PV versus more sub slab insulation discussion is that if the designer chooses the PV route, then the project manager has to fit the electrician/ PV panel installers in to the schedule. These are highly skilled trades, and electricians that know how to install a PV system are still not all that common, which may make the schedule less flexible. In the end it's not simplifying things.

    Whereas the greater sub slab insulation option of Passive House shouldn't necessarily complicate the schedule, either way sub slab insulation will be placed down, it's on the schedule and it doesn't require a highly skilled electrician.

    That said if the designer is planning to design to Passive House and also to install a PV system for the small energy load remaining then my above argument may, or may not, be valid. Still, I have to say that for the small remaining energy load, grid supplied electricity should be cheaper (at the moment).

    The tech.view columnist at wrote a column
    last week reminding us of the virtue of simplicity which is worth reading with respect to this posting.

    Finally, is the homeowner going to have the necessary knowledge to maintain the PV system. That's not an issue with insulation. Let's not forget about the occupants!



    P.S. Martin is there a way we can suggest story/ blog post ideas to you?

  12. GBA Editor
    Martin Holladay | | #12

    Simplicity, maintenance, and blog ideas
    1. Your point about simplicity is absolutely right. Many designers and homeowners will prefer the simplicity (in construction scheduling as well as equipment longevity) of thick sub-slab foam to PV.

    2. There really isn't any maintenance required for PV. My oldest PV module is 29 years old, and has had absolutely zero maintenance. It just sits on the roof, exposed to the weather. While it's true that I regularly remove the snow from my PV array in winter, most people don't have to — because they live somewhere where the snow is less frequent, and because they don't depend on every kWh of electricity from the sun in January and February like I do. (I'm off-grid). Since the days are very short and the sun is very weak during the winter, you don't lose much electricity if you don't clear the snow from your array. Finally, inverters are the weak link in a PV system. But I bought my inverter in 1985, and it's still working fine.

    3. Yes, I'd be very happy to entertain any and all suggestions for future blog topics.

  13. AndrewInChelseaQC | | #13

    Blog topics
    Hi Martin,

    It's ok to put the blog topic into a comment on one of your posts? I don't want to clutter up the conversation thread.



  14. GBA Editor
    Martin Holladay | | #14

    Or you can e-mail me
    Post it here if you want, or e-mail me directly at
    [email protected]

  15. user-626934 | | #15

    Ground temperature
    A heating season ground temperature of 55F in Finland sounds pretty outrageous to me. I calculated an average ground temperature for the heating season (approx. Oct. to April) for the Helsinki, FInland climate data set from the Passive House Planning Package (PHPP) and came up with 41.3F. I also used the classic rule of thumb for average ground temperatures: "average annual ground temperature = average annual ambient air temperature" to get 40.2F (again using the PHPP data set for Helsinki, Finland).

    Where is John Straube's data from?

  16. GBA Editor
    Martin Holladay | | #16

    The temperature difference was measured
    According to John Straube's e-mail to me, he based his calculation on measured data. Here's what he wrote: "The prediction of heat loss through slabs is notoriously inaccurate. I recently did a literature survey of measured temperatures and heat loss of slabs and basements, looking for real measurements of insulated cases. As an example of one of the few results I found, a slab on grade insulated to R-32 in Finland had an average heating seasons soil temperature of 12.5 C (55 F)."

    I'll try to get in touch with him to track down a reference for the study.

  17. user-626934 | | #17

    Ground temperature
    If the information from John Straube is accurate, Martin, then you should have written a different article! Instead of "Hey! You're using too much insulation below your slabs!" you should have written "Hey! The ground temperature in your climate data is dead wrong!"

    That's mostly to say: I'd love to see an in-depth follow-up on the ground temperature issue.

  18. GBA Editor
    Martin Holladay | | #18

    More data
    Although you imply that John Straube's reference to 12.5°C soil under an insulated slab in Finland was surprisingly warm, I found a reference to a study that refers to soil temperatures that are even higher -- between 15°C and 17°C -- under insulated slabs in Finland.

    I haven't read the study; just the abstract. I don't know the R-value of the insulation, so the data may not be relevant. If anyone is interested, here's the link:

  19. Graham Irwin | | #19

    This is complete nonsense. I am torn between ignoring this stuff and spending time addressing inaccurate and poorly researched conclusions, but here we go:

    1) Katrin's house exceeded the PH heating limit by almost 50% because she opted to use electric heat rather than fossil fuels and faced a source energy penalty due to inefficient generation.

    2) Mr. Straube attempts to justify his hypothesis by assembling a hypothetical PV/heat pump assembly to supply heat that might otherwise be preserved with insulation. He does acknowledge that the COP of the heat pump goes down with temperature, but fails to provide analysis on the seasonal variation in PV output. Presumably, the PV output is far lower during heating season than other times of the year, or it wouldn't be heating season!

    One might then turn the argument toward "net metering," which really means that the PV system overproduces in summer (helping to cool inefficient commercial buildings) and in winter, this "good deed" is magically repaid with fossil fuel produced electricity.

    If one wanted to truly operate at an equivalent level of sustainability by supplanting insulation with a heat pump, what is required is clean power in winter. The PV array that powers the heat pump should be installed in Arizona (Chile even better) with dedicated transmission lines to the heat pump. Even if the PV panels are cheaper than foam, things ain't so rosey now...

  20. GBA Editor
    Martin Holladay | | #20

    Not really nonsense
    As I pointed out in the last paragraph of my article, choosing 14 inches of foam over PV is "a defensible position," so I have already agreed with you.

    That said, it's important to note the point at which thicker and thicker insulation becomes more expensive than PV. To do so is not nonsense. Relying on net metering is not nonsense; it is a routine fact for tens of thousands of homeowners in Germany, the U.S., and many countries around the world.

    By ridiculing net-metering, Graham, you seem to be implying that using the grid to distribute PV electricity is somehow inappropriate. But all Passivhaus buildings use electricity; all Passivhaus buildings use the grid. Since they are already grid-dependent buildings, what's wrong with PV-equipped buildings using the grid to distribute excess electricity in the summer, and deliver needed electricity in the winter?

    If it's OK for a Passivhaus building to buy electricity from the grid whenever the building needs it, then we should all be willing to consider a distributed-generation model that includes the use of the grid to share PV-generated electricity. Net metering is the way that happens.

  21. GBA Editor
    Martin Holladay | | #21

    #1 of three e-mails from John Straube
    I have received three e-mails from John Straube responding to questions posted here.

    #1 concerns Mike Guertin's contention that Straube intended to refer to "cost / R / square foot" when he mistakenly referred to "cost / R / board foot." I responded to Mike by saying I thought he (Mike) was right.

    John Straube wrote, "I agree, Martin."

    In response to Mike's perceptive question, I have edited the article to reflect the intended units (cost / R / square foot).

  22. GBA Editor
    Martin Holladay | | #22

    #2 of three e-mails from John Straube
    E-mail #2 concerns John Brooks' question. Brooks wrote, "It is difficult for me to imagine that heat transfer is exactly the same from House to Air as from House to Earth when Delta-T is the same. I visualize the Earth as an easier 'path' than the air."

    John Straube wrote, "The difference between earth-slab and wall-air heat transfer is encompassed in the 'surface film' coefficient. This coefficient (fudge factor) accounts for the insulating effect that a still layer of air near the surface provides to a wall. It is not that important for well insulated walls, as its values is less than R-1. Another possible difference is the dynamic nature of the above-ground air temperature variations, but for lightweight structures this mass is not that important. Otherwise, the physics of heat flow above and below grade ARE the same."

  23. GBA Editor
    Martin Holladay | | #23

    #3 of three e-mails from John Straube
    E-mail #3 concerns John Semmelhack's challenge of Straube's sub-slab soil temperatures.

    John Straube wrote, "Mr Semmelhack's question points to the root of the confusion: the lack of good models because of a lack of measured field data. I agree that PHPP gives too low a temperature, and that the 'average annual ground temperature = average annual ambient air temperature' also misses the target (mostly because it ignores the fact that there is a heat source above the soil — the house — providing heat for 12 months of every year). This is the reason that I was basing my numbers on measured data, even though it is sparse.

    "The closest example I have is the Finnish paper 'Thermal and Moisture Conditions of Coarse-grained Fill Layer Under a Slab-on-ground Structure in Cold Climate' from the Journal of Thermal Envelopes and Building Science, Vol 28, no. 1, July 2004. The coldest temperature measured was 45°F, rising to 68°F by the end of the summer, before falling again.

    "A more recent paper, 'Heat, Air, and Moisture Control in Slab-on-ground Structures,' Journal of Building Physics, Vol. 32, No. 4, April 2009, shows even higher temperatures, with averages of 15°C/60°F in the winter, except near the edge where they dropped to 10°C/50°F (winter average over 15°C though).

    "In Norway, another cold climate that builds lots of slab on grade, a paper in the 7th Nordic Building Physics Conference on slab heat loss reported, 'In Norwegian climatic conditions, with a yearly mean soil temperature varying from 2° ~ 7°C, we can use 12°±1°C as a default value for the inner zone reference soil temperature". 12°C is 53.6° Fahrenheit.

    "In Britain, a study of a commercial building is reported in 'Temperatures in and under a slab-on-ground: two- and three-dimensional numerical simulations and comparison with experimental data,' in the journal Building and Environment, Vol. 35, 2000, found temperatures (in a milder climate near Cardiff, Wales) in the 14°-16°C (57°-61°F) range during winter periods. I also have some papers from Sweden, and basement measurements from Canada.

    "Suffice it to say, if I base my design on real measured results, rather than someone's model, I repeatedly find that the temperatures of the sub slab range from the 10°C to 15°C range (50° to 60°F).

    "I would be interested in hearing about any published measured data to add to my small but growing collection."

  24. Stephen Thwaites | | #24

    Sub-Slab Temperatures

    I like John Straube's comment about slabs warming up the soil below the slab. It makes sense.

    If that is the case, then aren't increasing amounts of slab insulation somewhat self-justifying?

    John Straube's comment would also suggest there is a region of soil that is acting like a buffer, dare I say insulator, between the slab and the 'ambient' soil temperature.

    As a window person I can see an analogy between window analysis, which looks at Frame, Edge of lass (2 1/2”) and Centre of Glass regions. In this case we might have a Slab, Soil near Slab and Ambient Soil regions. I suspect the distinctions between these regions isn't nearly as distinct as with windows. Clearly there is no foundation analogy to the highly conductive metal spacers found in most windows. Nevertheless, I think there's something similar going on.

  25. reese mcclure | | #25

    thick insulation
    Just like most green technologies out there foam is over priced. Build with straw bale and you dont have these kind of issues. If your structure works in harmony with the its environment the costs are insignificant. A typical straw wall will have an R-value of R-55. I have built a home that uses straw bale insulation,solar/wind power,methane harvester,radiantant heating,solar hot water,grey water collection,rain water collection for flushing toilets,responsible watershed,reclaimed building materials,and many other obviouse green technologies. And it will come in under $125sqft. I would love to know what they spent on that foam insulation, and will they ever recoup the costs. The costs need to need to make sense. By the way solar can be purchased for under $4 bucks a watt and by the time the rebates kick in its under $2 bucks a watt, so why are people still trying to demonize solar. If they just did a little shopping around. If you can find it cheap call me or if you need a straw bale home built , I will help. reese 951-453-3899, And for all you skeptics out there call me and we can talk about it.

  26. GBA Editor
    Martin Holladay | | #26

    Straw bale walls have an R-value of R-27.5
    The most thorough R-value testing of straw-bale performance was done at Oak Ridge National Labs. According to these tests, a wall built from 19-inch straw bales has an R-value of 27.5 — not 55.

    To read more about the testing, see "Refining Straw Bale R-Values," from the March/April 1999 issue of Home Energy magazine.

    Needless to say, straw-bale homes still need foundation insulation. If the straw-bale home has a slab-on-grade foundation, it should certainly include a continuous layer of XPS or EPS foam under the slab. Foam under slabs makes sense.

  27. reese mcclure | | #27

    Testing info out dated (straw rocks)
    Hey Martin thanks for the info, if you google straw bale r values you will get several sources for r values. 99% are pretty constant on the r value. Only the one you sent. So I would wonder who funded the test, maybe a foam manufacturer or a lumber company. I would take the info with a grain of salt and put an actual home constructed of both materials and see the results. I have always believed that the foam is over rated. As far as need insulation under the slabs, I would look to the old adobes and say wheres the foam. I think this foam craze we are on is just that. Man once again trying to out do mother nature. I am a green builder in southern california that loves new technology, I have looked at spray in foam for a long time and I think its over priced and will fall to the side in the next 5 years. We will always have a use for it ,just not as big as those invested in that industry thinks. Hope your not a foam guy. Any how thanks for the info feel free to call and share any other info you might have. reese 951-453-3899

  28. homedesign | | #28

    Straw Bale Discussion
    Hey Reese,
    You should probably try to start up a straw bale discussion in the Question Section.
    I think that you caught Martin wearing his cold climate glasses ;-)
    He forgot to ask you where you live.
    Perhaps most climates should be thinking about slab edge and maybe slab perimeter insulation.

  29. homedesign | | #29

    Heat Store or Heat Sink? Storing or Bleeding?
    I remember a discussion between Robert Riversong and Thorsten Chlupp.
    Two very wise builders.
    Robert said that we are storing energy below or slabs and Thorsten said that we are "bleeding" energy.
    Jon Straube and Wolfgang Feist are a couple of very wise Scientists.
    Why is there such a major difference here?
    Martin... have you tried to get a comment from Wolfgang Feist?

  30. homedesign | | #30

    Who is Right?
    The Physics here is over my head.
    I hope that the two teams will duke it out for everyone's sake.
    We need good answers and "Control Panel Design" sounds like a good concept to me.

    What I see is a difference in priorities.
    We (N. Americans) are making our decisions based on dollar costs and Return on Investment.
    The Germans have set a High Bar for their Enclosure Performance and they seem to be achieving their goals.
    By the time 2030 gets here.....Who will be ahead?
    Who will have the most Low Energy Homes Built?
    What would be wrong with damn good enclosures with Design integrated PV on top if you like.
    Why is it one or the other?

  31. homedesign | | #31

    Sorry John Straube... I knew it was John but typed Jon (twice)
    Jon Brooks

  32. GBA Editor
    Martin Holladay | | #32

    Who is right?
    John Brooks,
    I don't think this is a situation where we have to say that either Wolfgang Feist or John Straube is right, and the other guy is wrong. Both are right.

    To sum up:
    1. The Passivhaus people, who I respect very much, have a stringent energy goal that they have decided must be reached WITHOUT photovoltaics. As I have already noted, this is a defensible position, since envelope improvements are more foolproof and long-lasting than PV modules.

    2. To reach this goal in cold North American climates appears to require very thick insulation levels. When it comes to foam beneath slabs, the last 6 or 10 inches of Passivhaus foam are quite expensive in terms of energy saved per dollar invested -- more expensive than PV. But to Passivhaus proponents, the money is well spent, because of whole-house benefits. They're willing to make the investment.

    3. The PHPP software APPEARS to assume that subslab soil temperatures are colder than shown by monitoring studies using subslab thermocouples. However, I stand ready to amend this statement if anyone provides data to show me to be wrong.

    4. PV is expensive, but relatively long-lived and maintenance-free. Some U.S. homeowners with access to net-metering agreements may prefer to invest $4,000 in rooftop PV, and save $4,000 on sub-slab foam by stopping at R-20 per Straube's recommendations.

  33. homedesign | | #33

    Investing in the enclosure first
    I don't think that anyone has mentioned that it might be easier to add PV in the future than it would be to add insulation below the slab.

  34. Dark Lad Slim | | #34

    Data confusion?
    I'm not certain, I would need to consult EN standards and the calculation method used by to John Straube, but John Straube’s suggestion that the ground temps in PHPP are to low could be flawed. The reason for this relates to standards and methodologies and the way in which they use data sets.

    *John Straube is using a seasonal average measured *beneath* a slab conditions (in accordance with some standard or method that is unclear).
    *PHPP is using the seasonal average based upon ambient ground conditions (in accordance with the EN standard) i.e. not beneath the slab.

    Now consider that:
    * The Cardiff paper that John Straube cites finds that the calculation method in the EN standard shows a good correlation with real heat losses.
    * The PassivHaus Institute studies determined that for super insulated buildings the EN standard did not consider basements and the proximity of ground water adequately. The building physics has been written up in a Protokollband. Such refinements to the EN standard are contained in PHPP. Also consider that PHPP can accept input based upon dynamic calculations.

    So in conclusion to this section the EN standard is generally good for insulated building but could be refined.

    Data and stanards confusion:
    I fear that John Straube may not have considered that the methodology of the EN standard may differ from the one that he is using. So his warmer ground temps, which relative to external ambient conditions would reduce the rate at which heat is lost, is considered by the EN calculation used by PHPP but not in the same manner. (Also it is wise to remember that as the U-value is improved the heat losses are reduced and therefore the under slab ground temperature becomes closer to ambient ground conditions). To account for the fact that the under slab temp relates to the U-value the EN standard calculates a weighting factor that is applied during the calculation.

    Perhaps this would adequately explain the difference between the figures in PHPP and the measured ground temps that he cites. If John Straube is using the measure below slab temps without considering the fact that the temp will fall as the insulation is improved then this may also suggest that the point at which the cost effective limit for under slab insulation could be beyond that which is currently being considered.

    Any thoughts on this issue?

  35. Graham Irwin | | #35

    Actually, what you said was "For reasons that are somewhat murky, however, Passivhaus builders install much thicker layers of sub-slab insulation than most superinsulation nerds."

    The reasons are not murky in the slightest. They are the result of precise calculation of the net and incremental benefit of a given thickness of insulation, balanced with solar heat gain, etc., and they are figured quite precisely on a project-by-project basis.

    Further, I am not ridiculing net-metering, I am speaking to the ridiculous notion that net-metered PV power is carbon-free power. AGAIN, while one may produce a surplus of power in summer, that gets sucked up by inefficient commercial buildings. In winter, there is no surplus of PV electricity for heating because the sun isn't out much, or you wouldn't need heating, because lack of sun is what causes winter. The power we get in winter comes mostly from fossil fuels, unless we're in an area of the country that has a surplus of hydro, wind, nuclear or geothermal.

    With me so far? If this is the case, then the calculations suggested have to be rejiggered to account for the low output of PV during heating season - you can't assume the average daily output in the middle of winter. Contrast this with a proper level of insulation - it is effective at precisely the times it is needed most because it is designed for precisely the times it's needed most.

    Do I think PV is bad? No, but it is not the panacea it is held up to be. Yes, we need power for buildings, and Passive Houses need power too, because it is recognized that insulating to zero energy is not cost effective. That said, it is recognized by PH that grid electricity produced with fossil fuels represents consumption of fossil fuels. The net metering argument is not much different than buying carbon credits (i.e., paying to plant trees in Sri Lanka or whatever.) Perhaps your next article could discuss how buying carbon credits is cheaper than insulation... ;-)

    My point is that putting PV on a roof does not make the building more efficient, and that PH is the most cost effective point of efficiency, because it is calculated precisely to reward the builder with mechanical system reductions in exchange for increased efficiency. This precise and project-specific calculation enables one to determine exactly how much energy is being saved by one thickness of insulation over another - how is that "murky?"

    To suggest that "any house north of the Mason-Dixon Line ...should follow the ‘10-20-40-60 rule’ for R-values: R-10 under foundation floor slabs; R-20 foundation walls; R-40 house walls, and R-60 ceilings or roofs." because we can make up the difference with PV assumes that there are no differences in orientation or solar exposure among ALL houses north of the Mason-Dixon line. Yet you call the site-specific calculations involved in Passive House "murky." Don't you see how COMPLETELY ASININE this assertion is?

  36. GBA Editor
    Martin Holladay | | #36

    Asininity is hard to assess
    I prefer to discuss the technical issues on their merits; I'll leave it to others to determine my asininity. I can assure you, however, that I'm doing my best to avoid asininity.

    You assert that "Passivhaus is the most cost effective point of efficiency," but your statement is only true if you assume that it makes more sense to invest in sub-slab foam that is expensive (considered on a basis of energy saved versus money invested) compared to PV, which yields more energy for the money invested.

    You disdain PV because PV systems produce most of their energy during the summer, when heating is not needed, rather than during the winter. Since I have lived off-grid for 34 years, and since I have been watching the meters on my PV system every day for 29 years, I can assure you I am very familiar with this fact.

    I understand your disdain for the concept of net metering, because you have explained it twice. I nevertheless believe that I would rather live in a house that obtains much of its electricity from PV -- even if the electricity is seasonally unbalanced -- than to live in a house (like most Passivhaus buildings) that obtains all of its electricity from the grid. PV is expensive, but it is part of the solution we need as we make the transition from fossil-fuel-generated electricity to renewably generated electricity.

    This morning I spent several hours stacking firewood gathered from a clearing on a ridge near my house. Once I've completed work on the clearing, I hope to erect a wind turbine there. My hope is that the wind turbine will generate electricity during the dark days of winter.

    As I wrote in my essay, the decision to include very thick sub-slab foam under Passivhaus buildings is entirely defensible. I'm not ridiculing the position, and I certainly believe, Graham, that your position is not asinine.

  37. jbmoyer | | #37

    Holy Lord of the Foam
    I may sound like a building science neanderthal, but isn't there a point at which one has to step back and say.... "MAN, THATS A #$@% LOAD OF FOAM!!!!"
    Furthermore, shouldn't we be directing our efforts towards changing the way the american home is built?
    Shameful, horrific homes are being built this very moment throughout the country that are far from energy efficient. Lets address these homes first, and then worry about the validity of PHPPs and yearly mean soil temperature calculations. I can appreciate a healthy debate, and I understand the need for progressive thinking, but come on...
    AAAAAND to Graham Irwin,
    You sir, are way out of line calling Martin asinine!! (sorry Martin, I'm sure you can defend yourself, but I had to say something)

  38. GBA Editor
    Martin Holladay | | #38

    #4 of four e-mails from John Straube
    In response to Dark Lad Slim's comment that "John Straube’s suggestion that the ground temperatures in PHPP are too low could be flawed," Straube writes the following:

    "There must be some confusion. I am not 'following a standard' other than the stated laws of physics. I explicitly DID NOT consider the PHPP/ EN standard as this approach appears to give the wrong answers. The whole point of my investigation was that the 'standard' approaches seem not be getting us to numbers that match real measured data.

    "By measuring the soil temperature under a slab, the impact of insulation on reducing heat flow and the insulation and thermal storage capacity of the soil are directly considered. No estimates or fudge factors needed. All but the Cardiff slab were measured in a cold climates with lots of insulation (4 to 8 inches), so the impact of heat loss should be mostly accounted for.

    "Heat flow through a layer of insulation, whether under slab, roof, or wall, is driven by the temperature difference across it. For roofs, the challenge is to account for the solar impact on the surface temperature. For slabs, the challenge is to estimate the soil temperature.

    "So, I repeat, regardless of the fudge factors, standards, estimates, and computer models, heat flow across an insulated slab is due to the temperature difference across it, and the limited measured data provides consistent information about the size of this temperature difference. The real-world measurements simply do not match the standard approaches and assumptions."

  39. homedesign | | #39

    I think that Martin Likes to see "The Fur Fly"
    He likes to stir things up.
    He brings up some darn good topics.
    This has been a great discussion ... but Not-So-Friendly..

    Maybe there should be a Beer Summit .. sometime around October 16-18
    John Straube,Wolfgang Feist,Katrin and Martin
    German and Canadian Beer......not sure about Vermont beer?
    I think that PH & BSC would make a good team and should be sharing knowledge.
    Let's lighten up a little.

  40. GBA Editor
    Martin Holladay | | #40

    Trying to be friendly
    I sent an e-mail to Katrin Klingenberg this morning. I said, in part, "I would be happy to interview you for an upcoming blog; or, if you prefer, to provide you the opportunity to write a guest blog on the GBA Web site. As you probably know, you are always free to post a comment on our Web site. I have always valued your input. ... I hope that we can continue to work together and keep the channels of dialog open. I'm not sure you realize how often, in my writings and conversations, I have spoken highly of the Passive House standard and its accomplishments."

  41. Graham Irwin | | #41


    If I led you, or others, to believe that I was calling you "asinine" I apologize, but also point that my writing has been misread. Specifically, I asserted that a single prescriptive level of insulation for every home from Maryland to Minnesota is absurd if one is concerned with cost effectiveness, as is the assumption that PV output and cost is uniform across this area. Further, I do not disdain PV, but the notion that net-metering = carbon neutrality = PV is dollar for dollar as sustainable as insulation does not hold.

    Of course it is better to use PV than fossil fuel electricity, but since PV output is minimal when heating load is maximal, the solution is not to assume that you're getting an average daily output of electricity at the depth of winter and compare that cost to insulation, the solution is to reach the most cost effective point for efficiency, then add renewables after (ENDING with PV.) Passive House is designed to do this, and the insulation benefit is carefully evaluated and traded off with other measures. The building in Bemidji is heavily insulated because it is REALLY cold there in winter, but the thickness was carefully calculated and justified.

    I tend to lose my patience with this forum because I believe that you know these things and are attempting to stir up controversy, conversation, web traffic with sensationalist statements. Trouble is, as much as they may draw traffic to your site, they do not result in any greater insight, nor do they help with the fledgling Passive House movement in the US - they just suck up and waste time from people who already have to much to do. We either have to let these ridiculous assertions stand or work to dispel them, and we have enough to do already. If you want to explore these issues, please do so objectively and responsibly, avoid subjective words like "murky," and devote equal effort and time to researching and presenting both sides of an argument. If you are unclear about how the Passive House standard is derived or justified, please consult with those who know BEFORE your write these articles. This is the road to responsible journalism vs muckraking, IM(H)O.

  42. GBA Editor
    Martin Holladay | | #42

    What's murky and what isn't
    I'll try to address the issues in the order that you raised them.

    1. "A single prescriptive level of insulation for every home from Maryland to Minnesota is absurd." You are referring to my quotation of an Alex Wilson blog that reports on Alex's understanding of Joe Lstiburek's recommendations. I used the quote as a short-hand reference to U.S. recommendations for cold-climate insulation levels. It's fair to say that Lsitburek's understanding of the cost-effectiveness of insulation is more subtle than your parody implies. By focusing on the Alex Wilson quote, which was never intended as a location-specific recommendation, you distort Lstiburek's sophistication as well as my own.

    The intent of the quote was to illustrate the fact that U.S. builders following superinsulation principles generally use less sub-slab foam than do cold-climate Passivhaus designers. That fact remains — whether or not the sentence from Alex Wilson was well crafted or poorly crafted.

    2. I never claimed that "net-metering = carbon neutrality = PV is dollar for dollar as sustainable as insulation." I'm not sure if anyone knows whether the manufacture or use of either polystyrene or photovoltaic modules is sustainable. Probably, neither practice is sustainable in the long run. That's why I try to avoid the use of the word "sustainable."

    3. You assail the assumption that "you're getting an average daily output of electricity at the depth of winter and compare that cost to insulation." I never made that assumption; as I pointed out, I'm well aware that the daily output of a PV array in winter is less than the daily average output on an annual basis. However, in the U.S., the calculation of the return on investment for a PV array takes the value of the electrical output of the PV array over its lifetime, and compares that value to the cost of the investment. I'm not the only one to analyze PV investments this way; it's the standard way it's done.

    4. You write that "the solution is to reach the most cost-effective point for efficiency ... Passive House is designed to do this." But Straube shows that Passivhaus designers end up overinvesting in sub-slab insulation -- that is, going beyond the point of cost-effectiveness. You reject that by challenging the standard way that PV investments are assessed. I'm afraid your position is rather lonely.

    5. "I believe that you know these things and are attempting to stir up controversy." I can assure you that I approached John Straube with a technical question because of a single motivation: curiosity. When Straube answered my question, I learned something new. I am attempting to share my new knowledge with any interested readers who care to follow my blog. There was no controversy until various people posted fairly vociferous responses on this page — your post that "this is complete nonsense," Katrin's assertion on her blog that I (Martin) "prefer to talk to other people, 'experts' and smart guys, who I bet have little idea about the PHPP," your memorable line, "Don't you see how COMPLETELY ASININE this assertion is?," and your latest characterization of my words as "ridiculous assertions." To the extent that there is a controversy, I haven't been fanning the flames. I'm trying to focus on technical issues.

    6. Finally, you advise me to "avoid subjective words like 'murky.' " Your advice on vocabulary selection is interesting, coming as it does from a writer I might charitably call "unrestrained." Evidently you are unhappy that I wrote, "For reasons that are somewhat murky, however, Passivhaus builders install much thicker layers of sub-slab insulation than most superinsulation nerds." It is a simple fact that the reason for the disparity discussed in my essay was murky to me. The reason was evidently also murky to Jon Vara and Mike Guertin, who were enlightened by my investigation into the topic. Clearly, the reason was not murky to you. I don't doubt you're a step ahead of me. But my intention as a journalist has always been to try to shine light on murky areas — including areas that are murky to me — and to share what I learn with my readers.

  43. Dark Lad Slim | | #43

    The conclusion of the Cardiff
    The conclusion of the Cardiff paper -
    Thermal transmittance values have been calculated from the measured data and compared with the values determined from current design guides. ….. the U-values calculated using the CEN draft document were found to be in reasonable agreement with the measured value for normal weight concrete and in excellent agreement with the light weight concrete.

    ....hmmmm - Does this sound as though the real-world measurements simply do not match the standard approaches and assumptions? Okay, maybe not 100% correlation but certainly within realistic parameters? Yes, very much so.

    "…heat flow across an insulated slab is due to the temperature difference across it" This simple fact is obvious to anyone that has spent a few minutes considering heat flow. So where is the surprise? Where is the physics?

    A singular measurement is just a snap shot at a given geometric location - even it this is a seasonal average it is still specific to a location. In so far as I appreciate John Straube's average under slab temp accommodates the average of seasonal fluctuations and so it would seem that he is using a steady state model rather than full three dimensional dynamic modeling. Fine. A steady state model is what's used in a number of energy design tools.

    I wonder, has he has applied this average to the whole of the floor slab? And, what was the geometry of the buildings in question? Was the average at the center of the slab, across the whole are of the slab?

    The reason for these queries stem from the fact that that the temperature under the slab is not consistent i.e. an average fails to consider geometry and the fact that heat loss is greater at the perimeter (there are greater temperature differences at the edge of the slab than at the center of the slab due to greater heat losses to ambient air.) The zone affected by this thermal bridging could be up to 1-2m from the perimeter edge - for a house that is 10m deep and 15m long (is that okay for McMansion?) that is accounts for a substantial proportion of the floor area (about 30% or so) This fractional area will only increase as houses get smaller.…If he has just used one temperature, and has applied it to the whole slab, then this could result in an error. On this basis further clarification on the calculation method for "following ... the stated laws of physics" would be appreciated.


  44. Jesse | | #44

    The 10-20-40-60 rule is fun! You have your different shaped, colored blocks. Flat blue blocks for the floor, big red blocks make the roof. But Passive House is not so fun. Germans in lab coats, beakers, mysterious humming machinery, holistic systems. Too serious!
    And you know what's cooler than a really low electric bill in winter (boring!), is watching the meter turn backwards in summer ( Awesome!)

  45. Elana | | #45

    Passivhaus slab
    I can't believe the grief Martin is getting for arguing the technical merits of insulation versus PV! It seems that some of the people reacting here have "drunk the Kool-Aid" (an American reference) and are uncomfortable with scientific analysis of energy questions. Keep up the awesome work, Martin.

  46. GBA Editor
    Martin Holladay | | #46

    #5 of five e-mails from John Straube
    In response to Dark Lad Slim's questions about "the conclusions of the Cardiff paper," John Straube responded:

    "Why is Dark Lad Slim hiding behind a fake name?

    "The standard approach worked in Cardiff because the slab was almost uninsulated in a mild climate, exactly the type of scenario the standards were developed for and where they work. I did not reference it for that reason. I referenced it for the soil temperatures. Read all of the papers, especially the ones in cold climates with 4" to 8" of insulation to get the whole picture, e.g., the one that I was talking about. There are several reasons why the standard approach works for such slabs, but that is a whole article in itself.

    " [Quoting DLS] ' "…heat flow across an insulated slab is due to the temperature difference across it." This simple fact is obvious to anyone that has spent a few minutes considering heat flow. So where is the surprise? Where is the physics?'

    "This was a response to the query as to why I would use temperatures below the slab. Obvious to me. Confusing to many apparently.

    "[Quoting DLS] 'A singular measurement is just a snapshot at a given geometric location — even if this is a seasonal average it is still specific to a location. In so far as I appreciate John Straube's average under slab temp accommodates the average of seasonal fluctuations and so it would seem that he is using a steady state model rather than full three dimensional dynamic modeling. Fine. A steady state model is what's used in a number of energy design tools.'

    "Read the papers. I did not use a snapshot. I use an average of thousands of measurements (hourly) over the year. As you must know, the thermal mass of the soil under the slab means that the temperature varies incredibly slowly. I did not use a steady state model. I am using measured boundary conditions that are far from steady state. I have also done these types of calculations with Heat 2D, a dynamic 2D program, and was brought up on Mitalas's brilliant and still relevant basement heat loss models, which include 3D effects, dynamics, etc but unlike all other models, was carefully benchmarked against multi-year heat loss studies of DOZENS of REAL basements (the reason I trust his results more than most other paper studies).

    "[Quoting DLS] 'I wonder, has he has applied this average to the whole of the floor slab? And, what was the geometry of the buildings in question? Was the average at the center of the slab, across the whole are of the slab?'

    "Read the papers. Inform yourself.

    "[Quoting DLS] 'The reason for these queries stem from the fact that that the temperature under the slab is not consistent i.e. an average fails to consider geometry and the fact that heat loss is greater at the perimeter (there are greater temperature differences at the edge of the slab than at the center of the slab due to greater heat losses to ambient air.) The zone affected by this thermal bridging could be up to 1-2m from the perimeter edge - for a house that is 10m deep and 15m long (is that okay for McMansion?) that is accounts for a substantial proportion of the floor area (about 30% or so) This fractional area will only increase as houses get smaller.…If he has just used one temperature, and has applied it to the whole slab, then this could result in an error. On this basis further clarification on the calculation method for "following ... the stated laws of physics" would be appreciated.'

    "You act as if you know something about the topic, yet your questions imply that you have little understanding of the basics. Of course the perimeter is different than the edges. Of course geometry has an impact. But both of these do not affect the answer to the basic question of how much insulation should be under a slab. I hope it is also obvious that the perimeter should have more insulation if it is easy to do so, and the center less. My calculations are normally based on a 7.5x12 m floor plan, which may be considered too large by DSL but reflects the lower 50% of the housing market in North America, and probably 70% in Europe. A 10x15 plan is OK for a ranch house, and will mean the perimeter zone is about 25% of the total slab area, depending on your definition of a zone.

    "Physics of heat flow across a slab much wider than it is thick exposed to slowly varying temperatures: Q = U A Delta T. The only issue is the Delta T. Back to the reason for my looking for Delta T.

    "It would be nice to give a whole course on basic sub-grade heat flow, but I am not going to do this by e-mail."

  47. furniturefarmer | | #47

    interesting idea, but I'll stick with Wolfgang & Katrin
    Martin, thanks for bringing up this idea. This discussion made me re-evaluate the merits of large quantities of insulation, but I think Graham ultimately make some very good points about comparing insulation with electricity generation. Insulation saves heat in the winter, when it is most precious. If I'm paying a bit more to save a Btu in winter than I would to produce it insummer, the simplicity and logic of the more pasive technology make this a more attractive choice. The PHPP is a design tool for seriously reducing loads through a systems approach, not for calculating energy production, and if it turns out that eschewing PV in lieu of thicker insulation is not 100% cost effective, the simplicity, rigor, technical elegance, and usefuleness of the tool itself far outweighs this consideration.

  48. Mark | | #48

    Not all HDD climates are made the same
    Given the low internal gains within a Passive House, and the assumed 20C internal temp, it is worth remembering that not all HDD climates are made the same. HDDs are dependent upon internal temperature and internal gains (and solar gains). On this basis one mans 7,200 HDD climate may be another mans - lets say - 8,900 HDD climate (this is a rhetorical number rather than anything founded on calculation, but I'm sure that you get the point.) Unless the same HDD convention is used the for Passive House buildings and Building Science buildings are the same then the cost per kWh saved could be distorted.

  49. Dan Whitmore | | #49

    Hmmm... Murky
    Initially I was intrigued by tyour article, especially since I'm currently turning the dials on the control board for insulation on a project using PHPP. Mr Straube has some interesting scientific analysis. I can generally understand the possibilities of 'banking' a small amount of heat sub-slab so one's not dealing with the same Delta T below the structure as one is sub-soil 15' away. Attending a presentation by him would most likely be an excellent learning experience. Unfortunately the way the article presents his conclusions and conclusions others have drawn from his analysis I can only think to describe as 'murky'.

    Dr Straube is 'a very smart guy.' OK, I can accept that. However when he's presented in a sentence such as: "...the surprising disparity between the sub-slab insulation recommendations of North American physicists and Passivhaus advocates..." there seems to be a good bit of murky subjective journalism going on. Dr Feist, Ms Klingenberg, are merely 'advocates?' Please.

    PHPP is founded on years of scientific analysis and site studies. Certainly it needs more. I've never heard anyone who's got a genuine interest in it's application say otherwise.

    As for the question of utilizing PV instead of insulation? I wish this were separated from the discussion about sub-slab insulation needs because i see too many holes in the argument.

    - One of the large benefits of a Passive House is limiting the heat load to a point where an expensive heat source isn't needed. (If you want PV, scrap the pricey heat pump, put up a few panels & some foolishly inefficient baseboard heaters. Oh, right the sun doesn't shine in winter... back to Carbon trading.)

    - As for the assertion that PHPP has no balance for site-generated electricity. Check near the back for the Primary Energy sheet. At the bottom where it says:

    Solar Electricity
    Planned Annual Electricity Generation

    Specific Demand
    PE Value: Conservation by Solar Electricity
    CO2-Emissions Avoided Due to Solar Electricity

    -Yes 14" of foam is a lot, but as someone earlier pointed to, it's rather difficult to retrofit more. It's installation costs are very low and can replace some of the expense of groundwork/backfilling. (Back of the envelope calculations for the Smith house: using 5" of foam instead of 14" for the interior area of the slab (not incl. the perimeter thickened edge which had 4"(?))
    600s.f. x 56/s.f. x $.10=$3,360
    600s.f. x 25/s.f. x $.10=$1,500

    Please tell me where I can get a PV system of effective size for $2000.

    -PV in nearly all of it's phases is a complete add-on using delicate equipment and skilled technicians (yes I've had a few costly interactions with delicate PV systems). Retrofitting PV is a huge part of that industry and will likely just decrease in cost as huge amounts of public $ are thrown at it. In the cost comparison game, PV may win down the road but in my calculations it's not there currently. Sadly/ disturbingly the idea of conservation has long been counter to the capitalist growth model which is captaining the ship we all inhabit.

    "the calculation of the return on investment for a PV array takes the value of the electrical output of the PV array over its lifetime, and compares that value to the cost of the investment. I'm not the only one to analyze PV investments this way; it's the standard way it's done."

    Hmmm... Thought the idea of Green Building was to attempt to alleviate the the problems created by the building industry/built infrastructure not just analyze costs. I guess that's just me & my rosy glasses.

    New building methods and techniques definitely do need to be analyzed & critiqued and it's to be expected that one that challenges so many traditional methods will take a lot of shots. Passive House is new on this side of the Atlantic so it's getting it's share. Martin, I wish that this article & the one regarding the Boston seat-of-the-pants superinsulated retrofit (clearly not a Passive House) were more interested in analysis than what seems to be just poking holes at the parts you haven't grasped. The science used isn't murky, it's just involved.

    Dan Whitmore

  50. GBA Editor
    Martin Holladay | | #50

    Response to Dan Whitmore
    Thanks for your thoughtful comments. I agree with many of your points.

    1. Straube is not advocating that anyone should plan to "bank" heat in the soil. Rather, he has calculated the heat loss through an insulated slab, and determined at what point further insulation thickness becomes uneconomical.

    2. I meant no offense by the use of the word "advocates." The word was intended to refer not to any specific individuals, but to people who advocate in favor of building to the Passivhaus standard. I don't think there are any negative connotations to the word "advocates." I'd be happy if anyone wanted to refer to me as an "energy-efficiency advocate" or an "energy conservation advocate."

    3. Every Passivhaus building I have read about includes a heat source. If there are some that have no heat source, I'd be interested to know the climate where they are located and their internal loads.

    4. I don't think the cost of the HVAC equipment installed in the typical Passivhaus building -- usually including a HRV, an air-source heat pump, and a water storage tank connected to the air-source heat pump, which scavenges heat from outgoing exhaust air -- is less expensive than the HVAC equipment in an American superinsulated house, which can be as simple as a Panasonic exhaust fan with a timer (or an HRV if you prefer) and a small gas space heater with through-the-wall venting. Another option is a Mitsubishi Mr. Slim ductless minisplit for heating, as was used by Carter Scott in Mass. In other words, I'm not convinced that Passivhaus HVAC equipment is cheap.

    5. Dr. Feist told me, "From what I have seen, most builders I have talked with in North America still think that increasing insulation is an expensive thing. … I’m surprised, because insulation is the cheapest thing you can do." He's right — up to a point. But as Straube shows, installing more than about 4 or 5 inches of polystyrene is very expensive. The insulation ends up costing more than 60 cents a kWh. That's more than most of us can afford.

    6. You ask, "Please tell me where I can get a PV system of effective size for $2,000." Well, my first PV system was a $275 PV module hooked up to a $35 car battery. The system was effective. I powered a radio and one small fluorescent light. Of course, most people want more energy than I was satisfied with back in those days. Here's the point: I can easily design a PV system for your roof that will cost $2,000. The percentage yield for the dollars invested will be very close to the same yield that one would get for buying a $20,000 system (with a few minor adjustments due to economies of scale for the larger system). You invest one-tenth of the capital, and you get one-tenth as much electricity. But in either case, your meter turns backwards on sunny days. And the yield per dollar invested is identical.

    But Straube isn't saying that you have to buy the $2,000 PV system. He's simply saying that 60 cents per kWh is probably a good place to stop when it comes to insulation investments. Once these investments are made, you're done. You end up with a cheaper house -- or perhaps with $2,000 more to contribute to your window budget.

  51. GBA Editor
    Martin Holladay | | #51

    Response to Dave Brach
    Thanks very much for your thoughtful response. I agree with you completely when you wrote, "If it turns out that eschewing PV in lieu of thicker insulation is not 100% cost-effective, the simplicity, rigor, technical elegance, and usefulness of the tool itself far outweighs this consideration."

    Right! That's what I tried to get across in the last paragraph of my essay: choosing thicker insulation over PV is a defensible position, even if the insulation is rather expensive. That's why I'm surprised at the vociferous objection to what I wrote.

  52. m_sev | | #52

    Glad to see someone else working the insulation / PV issue
    Martin / John -

    It's amazing to me how few people are willing to run a few simple hand calculations to understand the diminishing value of increased insulation versus the somewhat diminishing costs of PV.

    I made the observation about diminishing value of insulation while working at BSC; the difference in yearly utility cost on one project taken from a R-40 roof to a R-60 roof was ~$20/yr in the Boston area. If insulation is free and easy to install, have at it, but if it costs any significant money, check your assumptions, and ignore emotional appeals to the 'some is good, more is better' philosophy. Plotting U-value vs R-value yields a log function, and highlights that addressing the lowest R-values like air leakage and windows tend to be higher priorities than taking a slab from R-20 to R-40 or a roof from R-40 to R-60 Changing from R-1 to R-2 cuts heat flow by 50%, R-2 to R-3 cuts heat flow an additional 17%, R-3 to R-4 an additional 8%, changing from R-40 to R-60 reduces heat flow by ~1%. What's the payback on that and when did this stop making sense?

    My view is that cost effectiveness of insulation probably crosses the cost effectiveness of PV somewhere in the R-20-40 range (walls/roofs), but doubtful many people would agree. It's very easy to quantify the cost of PV (~$8/watt), but quantifying costs of increasing insulation thickness in new buildings gets lost and distributed through a whole series of decisions and trades from the foundation to framing to insulation to siding to finish carpentry, and therefore is not as easy to compare. In retrofits, however, it's much easier, and I think PV is probably cost competitive down in the R-10-20 range, since the costs of retrofit tend to be so much higher.

    On the other hand, the value of PV is relatively constant, since every next PV watt purchased yields 1-1.25 kWh/yr if it's on a roof (flat or facing between SE and SW), not in the shade in most parts of the US. In fact, the cost per watt of PV goes down slightly as the system gets larger, since overhead and mobilization costs are distributed by a larger number of watts. Oh, and PV installations aren't usually complicated holistic affairs like insulation upgrades, and have very precitable results, so after the basic air-sealing and retro-insulating techniques are likely to be a more reliable performer than way-thick insulation.

    My question for the PH people is: How did someone arrive at 15 kWh/m2? Why not 22, or 7? If there is no consideration for economics, how about 0 kWh/m2?, especially since the whole of the PH standard allows 105 kWh/m2 for non-heating loads for a total of 120 kWh/m2/yr. If there are ~6 months of heating, essentially, PH allows 67 kWh/m2 for heating (105/2 +15), since almost all of that energy input is keeping the place warm (excepting things like clothes dryer exhaust that cross the building enclosure). And what about different climates - it doesn't really make sense to have the same standard for sunny southern California and for northern Minnesota. I have to say I think the PH standard could use some work - I think the media attention it has gotten is great, but technically, it seems like a 'one size fits all' approach with enough complications that the media doesn't really understand what they are essentially promoting. I'm all for energy efficient buildings, but how about a more holistic standard like the USDOE Building America Benchmark, why can't that get more headlines?

    On the durability of PV or heat pumps versus insulation, I wouldn't assume that foam insulation is a forever product, since most of them have no critter repellent, and the thermoplastic insulations (EPS, XPS) seem to shrink over time (2% is allowed, ~2" on an 8' board). Although Straube wasn't concerned about insects and foam at my last interchange with him, I know of enough problems in surprising places that I'm certainly not convinced - in fact, my point is from the other side - why wouldn't insects move right in? - foams make some good livin' for insects. I don't think foam's problems are difficult to address, there just needs to be motivation to do so from industry experts...ideally before those long term payback foams are being torn off buildings due to infestations of insects. Pennies to the manufacturers now, thousands to homeowners and insurers later. That would change the PV cost question, now wouldn't it?

    Anyway, thanks for getting people to think harder about the relative value of options for reaching lower energy consumption. The math says that eventually this will become common knowledge, but it's a surprisingly emotional issue at this point and few people seem wiling to pick up a pencil or calculator to double check their math.


  53. GBA Editor
    Martin Holladay | | #53

    On the origin of 15 kWh per square meter
    Thanks for your detailed and thoughtful comments. One of your questions — "How did someone arrive at 15 kWh/m2? Why not 22, or 7?" — gets to the heart of the Passivhaus discussion. It is a question often raised in the U.S.

    I raised the same question when I interviewed Wolfgang Feist in 2007. His answer was vague and unsatisfying, but it did offer a clue. Here is the exchange:

    MH: "What do you say to critics who question the 15 kWh per square meter goal, calling it arbitrary?"

    Feist: "The definition of a Passivhaus doesn’t need any number. As long as you build a house in a way that you can use the heat-recovery ventilation system — a system that you need anyway for indoor air requirements — to provide the heat and cooling, it can be considered a Passivhaus."

    Of course, Passivhaus designers sweat buckets to achieve 15 kWh per square meter, so it's a little flippant to say that "the definition of a Passivhaus doesn’t need any number."

    Feist's insistence on the centrality of an HRV provides an important clue to the origin of the "Magic 15" number. Here's what I wrote in the introduction to my interview (Energy Design Update, January 2008):

    "In Central Europe, the vast majority of Passivhaus designers choose to deliver space heat through a home’s ventilation system. This method imposes certain limitations; in Passivhaus buildings, ventilation airflow is usually in the range of 0.3 to 0.4 air changes per hour. Obviously, ventilation air cannot be delivered at unacceptably high temperatures; Feists’s Passivhaus Institut advises that ventilation air should be no hotter than 122°F. These criteria establish limits to the amount of heat that can be delivered by a Passivhaus ventilation system."

    So, it APPEARS that this was the reasoning behind establishing the number 15:

    1. Space heat must be delivered through ventilation ducts.
    2. The ventilation rate shall be 0.3 to 0.4 air changes per hour.
    3. Air temperatures of delivered air shall be no higher than 122°F.
    4. The best windows in Europe are U-0.14 windows; the best achievable air tightness is 0.6 ACH @ 50 Pa; with these limits specified, the best houses in a central European climate will need 15 kWh per square meter for heating.

    Once established, the 15 kWh per square meter is now being applied in Minnesota, where it is much more difficult to achieve than in Germany, which has a milder climate.

  54. GBA Editor
    Martin Holladay | | #54

    #6 of six e-mails from John Straube
    John Straube e-mailed:
    "The conversation on slabs was useful. Got me thinking, looking up papers, etc., etc.

    "All that I have uncovered has strengthened my argument. People have sent me their 2-D and TRNSYS calculations off line. I found several new papers, I ran the BaseSim model and reviewed all the old Canadian basement research. And over R-20 is simply impossible to justify. R-10 under basement slabs seems like a safe number for today's understanding of the future, but someone who is really paranoid of the future might consider R-20. Slab-on-grade not much different, assuming you insulate the vertical frost walls to protect against frost getting under the footing."

  55. Marc Rosenbaum | | #55

    everyone is right :-)
    WARNING: Content is potentially asinine, read at your own risk!

    When I first learned about PH over 10 years ago, one of the most baffling things was the amount of sub-slab insulation. Now that I have been through the PHC training and have worked with PHPP, I think I understand the answer a bit better.

    1 - As Martin quotes Straube quoting me, there are only so many "dials" one can adjust to meet the PH heating criterion of 15 kWh/m2/year. In North America, we have less advanced windows and doors (esp. doors) and ERVs/HRVs, so those dials get maxed out quickly in northern climates. You may be able to hit a lower ACH50 rate (easier if the house is bigger) and pick up some more. You've optimized the south facing glazing. What's left is the opaque surface thermal conductances. The lowest marginal cost in most situations is below the slab.

    2 - I don't necessarily think PHPP is far off in calculating the heat loss to the ground. Please recall that this is a 'simple' spreadsheet but it has been based on much more sophisticated simulations and well vetted with energy measurements - i doubt we have anything even close in North America in terms of vetting predicted vs. actual building energy use. PHPP calculates a Ground Reduction Factor to reduce the degree days used for losses to the ground. It uses a single factor for below grade walls and slabs. I typically see factors in the range of 0.5 - 0.55 in my calcs for upper New England. Using Burlington VT climate data in PHPP I got a GRF of 0.55 for a slab on grade house with an R-55 slab. The average ground temperature used (data is monthly) is between 52-53F. I calculate that the modified HDDs across the slab is 205 days x 15.5F, or 3178 HDDs (205 days is the predicted length of the heating season). That's about 41-44% of the annual HDDs for Burlington. Recognize that the losses near the edge are higher, and the GRF of 55% seems fine. I have done similar calcs for a heated basement building in Boston climate.

    Finally, as to whether PV or sub-slab foam is more cost effective - there is certainly a point where load reduction should hand the baton over to renewable generation. I don't think that the costs we use for PV include depreciated output over time, maintenance (small), and replacement of failed components (like my inverter that died). Same is true for heat pumps. The cross-over point between the two needs to take this data into account. And if you think that in the future our ability to manufacture large amounts of hardware might be affected by peak energy, then all the more reason to do something once that never needs re-doing.

    I look forward to the visit in October of Dr. Feist and hope we can discuss some of these issues with him directly. In the meantime, the reason I have gotten involved with PH is that I think it's rigorous targets are in the ballpark of what I think is needed for our building stock. It has some flaws when imported into North America IMO, and discussions such as this can help push the process forward if we keep our wits about us. In full disclosure i need to say I'm on the Board of PHIUS, I've taken the PHC training, and have also begun to teach in PHC trainings, so I'm in pretty deep, yet not without a critical view of the process.

  56. GBA Editor
    Martin Holladay | | #56

    A question for Marc
    Thanks for posting. Your voice of reason is appreciated; like you, I tried, in my August 23 post, to affirm (too late, evidently) that both Feist and Straube are right.

    For me, the most interesting sentence in your post was this one: "There is certainly a point where load reduction should hand the baton over to renewable generation." Yet (perhaps wisely) you remain silent about pin-pointing when that baton passing should occur.

    Am I correct that the PHPP software (unlike the BEOpt spoftware) never considers the question of whether a proposed building envelope measure is more expensive than PV? In other words, there is no warning that pops up when a Passivhaus designer goes too far in the direction of non-cost-effective insulation?

  57. Marc Rosenbaum | | #57

    Martin's question:
    Am I correct that the PHPP software (unlike the BEOpt spoftware) never considers the question of whether a proposed building envelope measure is more expensive than PV? In other words, there is no warning that pops up when a Passivhaus designer goes too far in the direction of non-cost-effective insulation?

    Yes - there is no cost-effectiveness analysis going on in PHPP. I'd hate this if it didn't let me toggle the assumptions beneath the calcs.

  58. homedesign | | #58

    Compared to what?
    Non-cost-effective insulation?

    Cost effective compared to what?
    Compared to the way we have always done things?
    Compared to doing it wrong?

    I can understand comparing to other less costly measures that we could take to improve the enclosure.....

    I do not understand comparing to the cost of adding PV.
    Most of the neighborhoods in North Texas do not allow PV... Or solar clothes dryers ............
    It is prohibited in the Deed Restrictions.

  59. GBA Editor
    Martin Holladay | | #59

    Yes, John — compared to PV
    I'm surprised that comparing the cost of insulation to the cost of PV seems such a stretch to so many people, especially those following the Passivhaus guidelines.

    For those of us who have been following (over the past 7 or 8 years) the discussions surrounding the design of net-zero-energy houses, this is not a new concept. It has been standard operating procedure.

    If anything, PV has been criticized as too expensive. PV-generated electricity is so costly that many designers say, "Wo! Why go that far? Why burden the homeowners with expensive electricity? I'll just stop short of adding PV, and they'll get the least-cost house."

    Now Passivhaus advocates are strongly defending an even more expensive option — more expensive than PV. As I've said repeatedly, it's perhaps defensible. But it's surprising.

  60. Mark Siddall | | #60

    Whole systems design?
    Apologies for the length of this one – should have read all the posts before putting finger to keyboard.

    Mark (M_Sev),
    As Martin indicated the 15 kWh/m2.yr ties in with the ventilation system but in truth with derived from two factors (Dr. Fiest confirmed this to me in person earlier this year). These factors are:

    1) the ability to heat a home using an full fresh air heating system i.e. it does not re-circulate the air, and
    2) passive solar gains.

    Item 1, along with avoiding the pyrolosis of the dust in the air (occurs at 50C) gives rise to a peak load requirement of 10W/m2 (30m3 per person per hour on the basis of 30sqm per person). Once this target is achieve this allows central heating systems to be removed and the ventilation system to be to supply both fresh air and any heating requirement. The capital saving from omitting the central heating system allow this same money to be reinvested in the building envelope i.e. with good design you are able to improve the envelope at no additional capital cost.

    The target of 15 kWh/m2.yr has been developed based upon whole systems design and life cycle analysis. It is fair to say that the 15 kWh/m2.yr was developed for a central European climate. More challenging climates can, and have, been able to satisfy the standard a little or no marginal cost (in Sweden PassivHaus schemes are being built at no marginal cost.)

    Depending upon the size and geometry of the building the 10W/m2 may result in energy consumption of
    Achieving the 15 kWh/m2.yr target is a careful balancing act so it is worth recognising that there is a point in some climates, say northern climates where daylight is not always in abundance it may not be possible to exploit the solar gains during the winter, the 15 kWh/m2.yr target can push you towards increasing the glazing area beyond the point at which the 10 W/m2 can be achieved - this is why in Sweden attention is paid to daylighting and orientation rather than full blown passive solar and why the PassivHaus Institut also allows certification using the 10W/m2 peak load ( in some cases this means an annual energy consumtpion of about 20 kWh/m2.yr).

    0 kWh/m2.yr has been explored in Germany this was found to require technologies that resulted in 1) significant capital costs 2) many new products 3) resulted in a whole life energy demand (including embodied energy) that was greater than that of a Passivhaus. This included PV and SHW etc.

    I fail to understand where you get you 67 kWh/m2.yr from. 15 kWh/m2/yr is for space heating. 120 kWh/m2.yr primary energy is about 50 kWh/m2.yr delivered energy for the whole home much of which will at the end of the day result in useful internal gains (I understand that best practice Passivhaus design achieves about 70 kWh/m2.yr primary energy).

    Given the whole systems analysis and life cycle costs it does make sense to have one energy standard. Why? Because the MVHR and the airtightness are the backbone of the concept as a consequence when you design in different climate the building envelope - U-values/ R-value and the efficiency of the MVHR for that matter - that adapts to the climate.

    It strikes me that focusing upon the elemental approach - a wall, a roof, a floor, or a window - squanders the any ability to engage with whole systems engineering - by taking this elemental approach you can 'pessimise' the whole.

    The whole systems approach that is in many ways inherent to PassivHaus has been shown to be cost effective in Europe, and recently there was an interesting study from the USA in the journal "Energy and Buildings" that looks at Net-Zero Energy Homes and Katrin's PassivHaus - with a theoretical PV array to achieve "zero". The study found that PassivHaus was one of the cheapest ways of achieving net-zero energy (though it did recognise that at this time there may be climatic limits to the economy).

    Obviously there will be a cost limit to what is affordable but this has to be viewed on a project by project basis over a realistic time frame, say 25-30 years, and you have to choose the correct datum - i.e. say an uninsulated building or the regional/national average stock rather than current regulations (if you don't do this you can't compare really apples with apples). With this in mind I recently tweaked a copy of PHPP so that I can calculate the net present value of the energy saved - and then the cost /kWh saved (as opposed to bought) for that period. As long as the NVP for the whole building remains within acceptable cost boundaries then you have an affordable solution. [ I've not managed to use my PHPP bolt-on for a project yet but one of the great things about PHPP is that you can make these tweaks and alterations (though I should note that if you want a Certified home you can't use a copy that you've tampered with!) ]

    So all in all I agree with Marc R that there is a point a renewables may prove to be more cost effective than insulation and I would add that this point would need to be climatically considered (there is now one size fits all) over the whole life cycle cost of the building.

    So where does this leave EPS? If you made a whole house, PassivHaus or not, using EPS insulation it may turn out with a NPV that is to high - to me this just suggests that the type of insulation should be refined to address whole life costs (this may be a result of climatic considerations requiring a certain resistance, due to cost/supply issues or some other factor). Here you could optmise sub-components, such as any EPS forming a slab on grade, by improving the U-values elsewhere and/or using cheaper insulation and by reducing that of that used under the slab. To my mind this is just good value engineering practice and has nothing to do with PassivHaus.

    I hope this contribution is recieved as constructively as it is intended.

    Kind regards,
    Mark S

    P.S. Martin, the average pressure test result in a European PassivHaus is about 0.37 ach/hr @50pa. The world record - as I understand it - was set in Canada 0.15 ach/h2 @50pa way back in 1981.

  61. GBA Editor
    Martin Holladay | | #61

    Mark Siddall,
    Thanks very much for your informative post.

    A couple of points that Passivhaus advocates often mention seem surprising to U.S. ears. One of these is the idea that if you deliver all of a building's space heat via ductwork, then you have "omitted the central heating system." You haven't, really -- you're just delivering the heat through ductwork instead of hydronic pipes.

    It also seems a little odd to stress the advantages of a heating system that "does not re-circulate the air." It's all fine and good to use 100% fresh air in your ventilation ducts, but to U.S. ears it doesn't seem such a crime if an HVAC system includes some recirculation. Most of the arguments against recirculated air (or "scorched" air) are unscientific.

    What I admire about the Passivhaus standard is the fact that Passivhaus buildings have very low energy requirements. However, Passivhaus designers sometimes exaggerate the negative consequences of building a building that uses a little more energy than a Passivhaus building, or delivers the heat through a method that is different (for example, not via ventilation ducts). It's good to have a goal. But U.S. designers who design buildings that require 22 or 27 kWh per square meter for heating in very cold climates have not designed a bad building. They just haven't designed a Passivhaus building.

  62. Mark Siddall | | #62

    I don't know how it

    I don't know how it is in the USA but here in Europe hydronic heating is the principle means of heat delivery. In houses with poor airtightness heat recovery and heating via air is very inefficient. For heating the specific heat capacity of water being that much greater than air, and for heat recovery a leaky building severely impairs the efficiency of the heat recovery.
    On this basis if an incremental approach is taken to a traditionally conceived building one improves insulation standards and airtightness. When the airtightness is at about 3-5 ach @50pa whole house mechanical ventilation becomes necessary for indoor air quality (VOCs, moisture, condensation, mould etc. i.e. the usual IAQ issues). Once you get to about 1.5 ach @50pa heat recovery starts to become economically viable - the cost effectiveness increases as the airtightness improves even further. In a low energy building, or even moderate or poorly insulated building, a decent airtightness leads to the inclusion of both hydronic heating and a ventilation system (examples include Swedish homes and the German Low Energy standard - say 55-65 kWh/m2.yr.) By pushing the envelope further, to PassivHaus standards of performance, the need for hydronic heating systems can be removed.

    I appreciate that the recirculation of air for heating is much more common in the USA. Much of the theory behind the PassivHaus standard is based addressing comfort standards by low energy or passive means. With regard to this technique there are issues relating to comfort that arguably need to be considered (drafts caused by air movement at to great a velocity and acoustics i.e. noise from fans etc.)

    With regard to the energy consumption - you say that "U.S. designers who design buildings that require 22 or 27 kWh per square meter for heating in very cold climates have not designed a bad building." I have to say that I agree. As noted above if energy performance is actually in this territory (when assessed using PHPP) then the designs are almost to PassivHaus standards if the all air heating approach is utilised - see the 10W/m2 element of my post above (PassivHaus buildings in Nordic countries above 60 latitude have heating energy demand 20 - 30 kWh/m2 according to location,

    Personally I have no interest in exaggerating claims about "lesser" buildings. I am interested in discussion of comfort criteria particularly with regard to ISO-7730 - the comfort standard that underpins much of the PassivHaus concepts. If other low energy performance standards can demonstrate comparable performance, and cost efficiency, then I would be interested in hearing about them.


  63. GBA Editor
    Martin Holladay | | #63

    Most U.S. houses deliver space heat by ducts
    In the U.S., most houses get their space heat via ductwork. In that way at least, they resemble Passivhaus buildings. Only a small minority of U.S. houses have hydronic heating systems. In the U.S., a house that delivers heat through ductwork is considered to be a house with a central heating system; whereas in Europe, many Passivhaus builders insist that their homes (which deliver heat through ductwork) have no central heating system.

    Those of us who advocate superinsulated construction have, since the early 1980s, been striving to reduce the air leakage rates of our buildings to an absolute minimum. Heat-recovery ventilators have been sold and installed in Vermont homes since the early 1980s. The superinsulated approach is not incremental. It advocates taking extreme measures to air seal houses.

    However, there is much more diversity in heating systems in the U.S. than among Passivhaus builders. Some superinsulated houses are heated with small sealed-combustion gas space heaters with through-the-wall venting; some are heated by a single hydronic baseboard on a loop from the water heater; some are heated by small electric resistance baseboard heaters; some have a woodstove. We aren't as fixated on delivering heat only through ventilation ducts that deliver only 100% outdoor air; this rule seems somewhat arbitrary and unnecessarily limiting.

  64. Mark Siddall | | #64

    Space heating strategies
    PassivHaus does restrict the technical means of providing space heating, all the means of heating that you mention can be used. The all air heating offers the most cost effective strategy (avoided labour, materials and maintenance) this is the only reason for a preferance - indeed in the UK some of those that support the PassivHaus approach like the idea of retaining some hydronic heating to enable greater zone and user control (personally I must confess that I am skeptical about this theory/ approach as the studies that I've read suggest that, over time, any the internal temperature differentials will tend to iron themselves out).


  65. homedesign | | #65

    HRV payback?
    Quoting Mark Siddall (from this blog):
    "Once you get to about 1.5 ach @50pa heat recovery starts to become economically viable - the cost effectiveness increases as the airtightness improves even further."

    Quoting Martin:
    (from a GBA question)
    "2. Heat-recovery ventilators are NOT cost-effective. In other words, the heat that is recovered is not enough to justify their high purchase price. However, they are the most effective available ventilation systems, and have the lowest operating cost."

    My question.....
    Is the equipment that much better in Europe...or is Martin being too negative?
    Is Heat Recovery really a folly with no economical payback in the US?

  66. Mark Siddall | | #66

    Heat Recovery a folly in the USA? Not as I see it.
    I'd say that there are two issues that need to be considered. The efficiency of the HRV and the specific fan power. The efficiency should be >75% which is, I think, available in the USA - though testing methods may differ!

    The specific fan power (SPF) may be more tricky to address. The SPF is determined by the fan motor and the design (small ducts make the fans work harder due to friction). If the fan power is too high, i.e. consumes to much energy, then the efficiency of the whole system is degraded - look for electrically commutated fans as these are more efficient. As I see it there is no physical or technical reason why HRV should not be an option.

    I should confess that my suggestion for an ach @50pa of less than 1.5 is based upon my assessment of the mild climate UK and UK costs for energy and MVHR systems (i.e. not those of the USA). In a more challenging climate the whole life cost of HRV may be affordable/ suitable at less stringent levels of airtightness (but why would you allow less airtight construction? you'd only end up with drafts and discomfort.) Ultimately you'll have to do your own NPV calcs if you are solely concerned with the fiscal benefits of MVHR.

    In terms of airtigthness a starter for ten would be Canada's R2000 scheme, developed in 1984, requires HRV and has an airtightness requirement of 1.5 ach/hr - but why not employ best practice and target 0.6? (It is simply a matter of skills (design and construction) and does not require more technology.)

    Smart shopping is what is required, I'm sure that suitable products are out there if you look hard enough (even if they do not quite achieve the same standards of performance as PassivHaus units I'm sure that you can source some half decent products.)

  67. homedesign | | #67

    I Believe
    I believe that we (North Americans) can build airtight
    thermal bridge free
    Uber-Insulated and well ventilated Low Energy Homes
    I Believe...we can do it
    But we have to believe we can do it.

    We should not wait until 2030

  68. GBA Editor
    Martin Holladay | | #68

    HRV efficiency
    According to a 1998 study by researchers at Lawrence Berkeley National Laboratory (Roberson, Brown, Koomey, & Greenberg), the cost of operating an exhaust-only ventilation system averages 56 cents per day, while the cost of operating an HRV averages 49 cents per day, in a hypothetical average house. (These numbers represent the averages for several climates.)

    Let's say that a good Panasonic exhaust fan on a timer costs $400 to install, and the HRV costs $2,000 to install. A simple payback for the $1,600 incremental cost for the HRV would be 62 years.

    Of course, I'm sure readers will rush to challenge these assumptions. I'll be pre-emptive and do it myself: Those cost estimates are outdated -- they're from 1998! A good exhaust-only ventilation system costs more than $400 to install! I can install an HRV for less than $2,000!

    Okay, okay. You get the idea. One's conclusions depend on one's assumptions. These are some figures -- and they are fairly reasonable figures, but they are open to challenge. Here's the point: one chooses an HRV because it is a great ventilation system -- it provides good fresh air delivery at a low operating cost. One doesn't install it because one hopes that the $2,000 investment will yield a fast payback.

  69. Mark Siddall | | #69

    Back to a whole systems approach
    I agree of course that cone's conclusions depend upon the assumptions: Regarding the "cost of operating an HRV averages 49 cents per day" - is this based upon the cost of electricity alone or does this include the energy benefit from heat recovery? What was the efficiency of the heat recovery? What was the specific fan power? Was this whole house MEV or intermitant localised MEV?
    Also, and I appreciate that this muddies the discussion regarding be building fabric, but if heating is via an all air system the cost of the heat distribution system can be discounted. (This is why whole systems design is preferable to the discussion of individual components.)


  70. GBA Editor
    Martin Holladay | | #70

    Ventilation study assumptions
    The referenced ventilation study notes, "Ventilation operation costs include ventilation fan energy, the cost of tempering ventilation air, and the cost of tempering infiltration attributable to mechanical ventilation." The study is available online:

    "if heating is via an all air system the cost of the heat distribution system can be discounted." I assume that you mean that the blower energy degrades to heat. If the blower is within the conditioned envelope, as it should be, the heat from the blower motor is useful in the winter and detrimental in the summer. The same can be said for a hydronic pump motor, of course.

  71. Mark Siddall | | #71

    Whole systems
    What I mean is that you gain multiple benefits from single expenditures - the all air system supplies heat and fresh air, two functions rather than one. This avoids capital cost and reduces payback periods.


  72. Mark Siddall | | #72

    An Interesting Conclusion
    I’ve not read this in detail but the conclusion is interesting. Compared to an uninsulated concrete basement (case G47; 35,457 kWh), the largest reduction in the annual sensible heating load (30,667.87 kWh or 86.5%) was achieved using R40 sub-slab insulation, and R50 interior wall insulation (case A15; 4,789.13 kWh). As expected, A15 also has the longest payback period. Although this alternative has the longest payback period (2.3 years), it does show that at heating and construction costs current at the time of the study, even installing a high level of insulation has a short payback when compared to uninsulated basement walls.

    R40 = U-value of 0.14 W/m2K
    R50 = U-value of 0.11 W/m2K

    A payback of 2.3 years is not long, furthermore it is a ROI of 43% per annum – that’s damn good business. There are of course faster paybacks using lesser specifications but this would also have to be balanced against life cycle costs – you pay energy bills for the life of the building.


  73. GBA Editor
    Martin Holladay | | #73

    Interesting study
    Thanks for pointing out an interesting study. (If any readers are confused about what Mark is talking about — as I was at first reading — his comments refer to a study, "Analysis of Basement Insulation Alternatives," that is available by clicking the link at the bottom of his post.)

    A couple of points:

    1. The study concerns insulation levels in Yukon, Northwest Territories — an extreme climate that is considerably colder than Minnesota.

    2. Fuel oil cost is assumed to be $1 per liter ($3.79 per gallon).

    Although the researchers calculated the payback period of thick insulation COMPARED TO AN UNINSULATED BASEMENT, they didn't calculate the payback period for adding R-10 additional insulation to a basement that was already insulated to R-30, compared to a basement insulated to R-30. These calculations will, of course, yield very different results.

    Here's what I found interesting: "Compared to an uninsulated concrete basement (case G47; 35,457 kWh), the largest reduction in the annual sensible heating load (30,667.87 kWh or 86.5%) was achieved using R40 sub-slab insulation, and R50 interior wall insulation (case A15; 4,789.13 kWh). As expected, A15 also has the longest payback period. Although this alternative has the longest payback period (2.3 years), it does show that at heating and construction costs current at the time of the study, even installing a high level of insulation has a short payback when compared to uninsulated basement walls. Some of the wall and sub-slab insulation options have very similar energy savings. For example, R50 wall insulation and R20 sub-slab insulation (A9) result in an annual heat loss that is only 566 kWh (or 2.7%) higher than the best case A15 with R50 wall insulation and R40 sub-slab insulation."

  74. Mark Siddall | | #74

    Martin, Since when was the
    Since when was the discussion about retrofitting insulation to existing floors? We are talking about new buildings (and at best retrofits to poor building stock.) Upgrading an R30 to R40 will indeed result in a very different situation but that was not the central argument of your article. In my view "retrofitting" a theoretical insulated building that had not been built and comparing this to a better insulated one is a very perverse logic indeed. For energy efficiency to be calculated properly, and to avoid developing a false impression of diminishing returns, you have to compare all insulation strategies to an uninsulated building. [Editor's note: text corrected per request of Mark Siddall.] It is the only appropriate datum when looking at whole life costs - or payback periods and ROI if that is the interest.

    In my view, with regard to energy efficiency, payback periods are a fools game and ROI is not much better. A building lasts for 50+ years (over 100 in the UK), mortgages last for 25 years (in the UK at least). Provided that the cost of the energy efficiency measures is less than the cost of the fuel over the life span (I choose 25 years) then the efficiency measure is affordable. The best way to think of this is NPV - to get a better grip on the numbers this can be converted into the cost per kWh saved, rather than the cost per kWh bought. (I prefer to use cost per kWh saved rather than ROI as this a more immediate, and relevant, unit that also serves to avoid certain distortions resulting from whole life cost analysis using the NPV alone.)

    Indeed other sections of the conclusion are interesting, however, this goes back to the central thesis of payback periods and the nonsensical nature of them with regard to buildings.

    Thanks for pointing out the climatic differences, being a foreigner I had little appreciation for the climate mentioned in this report.


  75. Mark Siddall | | #75

    Erratum: "you have to compare all insulation strategies to an uninsulated floor slab" should read "you have to compare all insulation strategies to an uninsulated building"

    [Editor's note: correction made.]

  76. GBA Editor
    Martin Holladay | | #76

    The topic is being discussed elsewhere
    This interesting dialog continues ... on a different page:

  77. GBA Editor
    Martin Holladay | | #77

    British reflections on these issues
    BuildingGreen's Mark Piepkorn recently posted a link on his BuildingGreen blog to a British Web site that discusses some of the issues addressed here.

    To read Mark Brinkley's thoughts after he toured some Passivhaus buildings in Hanover, German, click this link:

  78. Shawn Busse | | #78

    A more simple insulation case study in action
    Here's a link to an article on our project where we attempted to quantify the cost of adding additional layers of insulation.
    We're on a limited budget so going from 2" to 4" was a consideration. Achieving the R60 or R80 of the passive homes is way beyond the capacity of our pocketbook, so I wrote an article on how this relates to the "ordinary" homeowner.

  79. GBA Editor
    Martin Holladay | | #79

    Thanks for showing us your calculations and procedure. Performing such calculations is always fruitful and stimulates thought.

  80. Riversong | | #80

    It's a shame that this "discussion" is generating more heat than light. If only we could harness this output to heat our homes, we would not have to worry about insulation at all.

    Ironically, both sides of the dispute use the same premise to arrive at completely opposite conclusions – and the premise is fundamentally flawed.

    m_sev ( a pseudonym who says he worked at BSC) argues that "plotting U-value vs R-value yields a log function" when, in fact, it's a hyperbolic function (the one is the inverse of the other). And that "changing from R-1 to R-2 cuts heat flow by 50%, R-2 to R-3 cuts heat flow an additional 17%, R-3 to R-4 an additional 8%, changing from R-40 to R-60 reduces heat flow by ~1%."

    This sophistic mathematical exercise attempts to undermine the argument for high insulation levels. But the analysis compares the first incremental reduction in heat loss to the absurd base condition of an R-1 thermal envelope and then measures each successive incremental reduction as a percentage of the initial incremental savings rather than as a percentage of either the base case or the starting point of that incremental change. In other words, while going from R-1 to R-2 offers a 50% reduction in heat loss, stepping up to R-3 offers a further 33% improvement (not 33% of 50% = 17%), and stepping to R-4 offers another incremental improvement of 25% (not 8%). Thus, while there is a diminishing incremental return, it is not nearly as low as m_sev suggests. And making an incremental shift from R-40 to R-60 offers a 33% improvement in thermal performance, not the absurd ~1% of m_sev's faulty math.

    It is, of course, true that the last 33% improvement is upon an already very low annual heat load, so the incremental dollar savings is smaller with each equivalent incremental thermal envelope improvement. If one were to compare the energy or dollar savings for each increment to a base case (for new construction), then that base should be a current energy-code minimum home not an R-1 home.

    And then PH advocate Mark Siddall, in order to attempt to justify extreme insulation levels, says "you have to compare all insulation strategies to an uninsulated building". If you're considering the benefits of retrofitting a currently un-insulated building, that statement might make some sense. But when comparing the incremental cost and advantage of adding additional insulation to either a new or existing envelope, one must compare each additional incremental improvement to the baseline of that particular increment and to some measure of energy or financial return on investment (which is what Martin has attempted to do in his blog).

    This entire "debate" seems more like theater of the absurd than a rational argument. And I continue to find it fascinating that PH advocates (Marc Rosenbaum excepted) get so upset about any challenge that they become verbally abusive. This is the response one would expect from a "true believer" rather than a rational advocate. Is PH some kind of cult?

  81. Riversong | | #81

    Back to the discussion...
    I agree that it's difficult, and perhaps misleading, to compare permanent envelope improvements such as slab insulation to shorter-lived mechanical enhancements such as PV. Financial return on initial investment is only one part of the equation. Financial and ecological life-cycle costs have to be factored in (as some here have proposed) - PV production and end-of-life disposal has its own enviromental costs, as does petrochemical foam insulation. And we would do well to follow Amory Lovin's dictum that the cheapest (to society and the world) megawatt is a negawatt - conservation is always to be prefered over production. But that means consuming less by our lifestyles, not using non-renewable materials to reduce the energy impact of a profligate and unsustainable lifestyle.

    The whole debate over thermal losses to the ground ignores the thermal mass benefits of a warmed layer of earth surrounding our foundations and underlying our slabs. I've yet to see an analysis of the relative heat loss downward with lower insulation levels and warmer earth compared to higher insulation levels and cooler earth. This is the strategy that Passive Annual Heat Storage and Annualized Geo-Solar systems employ. In most cold climates, there is an insulating snow layer on the ground for much of the winter, and heat loss into the ground does not blow away as it does to the air or radiate to the night sky, but creates a dynamic mass benefit which further undermines the incremental benefit of additional sub-slab insulation.

  82. David Whte | | #82

    two points
    I may be too late in commenting here, but looking over the dialog I see a couple of things missing:

    1. Dr. Straub's numbers for sub-slab temperatures seem in at least two cases to refer to a central zone of the slab. However, the PHPP analysis effectively applies a UA-averaged soil temperature over the entire underside of the slab, and this is strongly weighted by the colder temperatures near the edge. With adjustment for this, the presumedly too-cold PHPP outputs would match more closely with the research.

    2. Is the issue to find minimum life cycle cost of heating the building or maximum cost effectiveness of saving the planet? In the former, the cost must include maintenance, not as a second argument but as part of the basic cost calculation. In the latter, one must include the embodied energy of the PV, which is often left out of discussions because we assume that renewable energy is immune to this. I'm not sure what the embodied energy of PV really is, but note that the PHI uses a PE factor of 0.7 for PV. That's much lower than typical grid electricity, but much higher than zero. This may be one reason that the PHPP does not consider PV a fair strategy in reducing primary energy demand - it does not have equivalent environmental benefit as energy savings using a lower embodied energy strategy.

  83. TC Feick | | #83

    comment and simpleton question
    I believe in a pragmatic cost/benefit analysis to any strategy towards sustainable building. In other words, any strategy has to make economic sense over a timeline relevant to the primary user/decision maker to gain widespread acceptance in the marketplace. As one early post pointed out, there is a vast ocean of difference between energy efficiency of stock housing and that of BSC's recommendations and Passivhaus. The latter two approaches will provide very good performance for energy efficiency, agreed? So, how about the delta of the cost between these implementations and the cost savings to take BSC's approach? Seems like there is a good argument to be made for saving some cash in the name of broader acceptance of something approaching an energy efficient home.
    Now my question- Why EPS below a slab? I would think that EPS would absorb moisture and affect its R-value?

  84. GBA Editor
    Martin Holladay | | #84

    Cost versus benefit
    T.C. Feick,
    Concerning cost/benefit: many people have run cost/benefit analyses for insulation. There are several variables, of course, the most important of which are the climate where the house is located, the presumed future cost of energy, and the expected lifespan of the building.

    The colder the climate, the higher one's estimate of future energy costs, and the longer a building is expected to last, the easier it is to justify thick insulation. Most such cost/benefit analyses end up with sub-slab insulation thicknesses in the ranges recommended by the Building Science Corp. The extreme thicknesses advocated by North American Passivhaus proponents fall outside of these analyses.

    EPS does not generally absorb moisture; that's why it is commonly used to make coffee cups and dock floats. Of course, you have to specify the right type of EPS — the denser the better. EPS comes in a range of densities; if you want to install it below a slab, don't use the cheap stuff.

  85. Ted Clifton | | #85

    Under-slab insulation
    I appreciate all the good comments on this blog, there is a lot of good, even if seemingly conflicting information here. One thing missing so far, a few people got close....

    It is not enough to just insulate under the slab. You must also insulate around the perimeter of the slab to a depth that exceeds the depth of major seasonal fluctuations in the soil temperature, at least to the extent that is cost effective. By doing this, you keep the temperature immediately below the slab insulation closer to 65 degrees, the soil below that 60, etc., and that soil is much less affected by the adjacent soil temperatures that follow the seasonal air temperatures. To my knowledge, nobody has aver done a detailed study to quantify this effect, but we have real-time results that indicate the effectiveness of this strategy.

    There are many variables, as each climate zone will have very different soil temperature seasonal variations, and different moisture conditions in the soils will also have a great effect on the transfer rate of energy. It would follow that if you can eliminate the transfer of water beneath your slab, you can also limit the transfer of energy. A closed-cell foam insulation barrier, placed in a vertical position around the slab, would also help with that issue. Our experience suggests that if you are using four inches of foam under the slab, you will receive many times the benefit of an additional two inches of foam under the slab by placing the same two inches vertically around the perimeter of the slab. The amount of foam required is just a small fraction of that required to cover the entire slab area. Since the savings are so great, just go with 4" of foam around the perimeter (in addition to the 4" underneath), and wait for the experts to finish arguing the point as you are saving money in your warm comfortable house, with your PV system supplying energy to your far more cost-effective ductless mini-split heat pump.....

    By the way, in many locations around the country, January and February are among the months with the most hours of sunshine! Check it out on Climate Consultant 5.

  86. johnsave | | #86

    Insulation and delta T
    I am embarrassed to ask this question as it appears so dumb.I understand that delta T is the energy difference across a partition.If
    I put 2" of EXP under my slab then the delta T is still the same. Given enough time the slab and foam will come into equilibrium. I will then be in the same position as I was without the foam, but a time interval later.As I see it the insulation delays getting to equilibrium but doesn't prevent it.Is there a factor which denotes the time delay? All vacuum bottles cool with time why not subslab insulation? Cheers and thanks Mike Legge

  87. GBA Editor
    Martin Holladay | | #87

    Response to Mike
    If the soil temperature under the slab is the same as the temperature of the concrete slab, then the foam insulation will also be at the same temperature.

    However, in winter, a more typical condition is that the soil temperature is at 50 degrees and the slab temperature is at 65 degrees. In this situation, the foam slows the heat loss from the slab to the soil.

    The soil is an enormous heat sink, so the heat loss from the slab is unlikely to raise the soil temperature to 65 degrees -- therefore the delta-T is likely to be greater than zero all winter long. In other words, the slab and the soil under the slab never reach thermal equilibrium.

  88. Perry525 | | #88

    There is a difference in perception.
    In Europe a Passive House is heavyweight and designed with a 400 year life span in mind.
    Most homes in Europe are over a hundred years old.
    Passive Homes are designed to be heated by the human body and the electrical equipment in the home, freezer, fridge, TV, computer etc.
    Fresh air is routed under the ground, and home to benefit from the steady (ish) 12C temperature and is merely topped up with a small in-line heater.
    The south facing windows triple glazed probably loose more heat over a typical year than they gain for the home, however they are popular.
    From the above you can appreciate that those extra inches of insulation pay for themselves many times over.
    In one to four hundred years your typical solar panels will require changing some four to sixteen times.....not at all economic.
    Dow brought Styrofoam to market 51 years ago, I started using it 41 years ago, it is in perfect condition, as new, and is still doing the job it was designed for.

  89. GBA Editor
    Martin Holladay | | #89

    Response to Roger Anthony
    I agree with your basic proposition, which is why my concluding paragraph states, "PV equipment and heat-pumps have a shorter life, and require more maintenance, than sub-slab insulation. In fact, this point may be enough to convince some builders to choose 14 inches of foam over a PV array. It’s a defensible position."

    A few points about your Passivhaus comments:
    1. You wrote, "Passive Homes are designed to be heated by the human body and the electrical equipment in the home, freezer, fridge, TV, computer etc." Actually, all Passivhaus buildings require active heating systems, just like all of the superinsulated homes built in the U.S. over the last 30 years.

    2. You wrote, "Fresh air is routed under the ground." Actually, Dr. Feist no longer recommends the use of earth tubes. When I interviewed Dr. Feist in December 2007, he told me, "There were problems [with earth tubes] in northern Europe, especially in Scandinavia. In Central Europe we haven’t
    had any hygienic problems so far. Actually, I’m not sure why we don’t have these problems in Central Europe. But I don’t advertise these systems any more, mainly because they are too expensive. If you have a good heat-recovery ventilator, you don’t need it."

    3. You wrote, "The south-facing windows triple glazed probably lose more heat over a typical year than they gain for the home." That may have been true in the past, but the best south-facing windows available today can easily gather more heat than they use. To learn more, see Windows That Perform Better Than Walls.

  90. user-1006269 | | #90

    Thick foam
    The diminishing returns on foam is a moving target. Cost of PV's, heating devices and backup fuel are all going to be factors.
    My 2 cents worth of comment is that foam is passive. It always works (assuming critters stay out of it!)

    PV systems are mechanical systems of a sort and are apt to be down sometime. Storm damage, component failure and other acts of God will leave you with a thermally compromised building, which is antithetical to the Passiv Haus concept.

    That being said, as one who has made a living working with foam and foam products, the one place that I would be skeptical of too much foam is in the foundation where insects are most likely to intrude and compromise that system.
    Time will sort these system configurations out.

    Tom Gocze

  91. user-1119494 | | #91

    Of course you can have too much foam...
    When the last cm of foam takes 50 or a hundred or more years for the embodied energy to be saved, let alone the time and money invested, we have left the zone of reason and entered that of religion. There may be other reasons for using that foam: perhaps it allows one to save on mechanical systems or gives extra security and peace of mind for blackouts, but saying that it makes sense from a standpoint of saving energy or money is hard to support.

  92. lutro | | #92

    And where should the foam be?
    It's interesting to read carefully through a three-year old article and comments. I didn't see anyone comparing the effectiveness of addition of more insulation to the roof, versus more under the slab. One example cited R-80, another R-60, under the slab, as a way of meeting the Passive House energy use goals. Roof insulation levels aren't mentioned, but I would think increasing insulation in the roof and walls would be much more effective than the second R-40 under the slab. Why are Passive House designers distributing their insulation in the way discussed in this article?

    I enjoyed the many interventions from the wise and witty John Straube. I wish he would comment more frequently in Green Building Advisor.

  93. GBA Editor
    Martin Holladay | | #93

    Response to Derek Roff
    One reason that more insulation wasn't installed in the roofs of these Passivhaus buildings is that many of them already have R-80 or R-100 insulation specified for the roof -- making the addition of more insulation up there physically difficult, as well as unlikely to return much energy benefit.

    I certainly agree with you about John Straube -- it would be great to have him comment on these pages more frequently.

  94. lutro | | #94

    Too much of a good thing.
    Thank you, Martin. It's amazing that the energy evaluation software gives any useful credit for such extreme insulation under the slab.

  95. user-1005777 | | #95

    Storing heat under the structure
    If I had the money to build, I would build a slab house on a 5 foot frost wall. Insulate the bottom of the 5 feet to R20, the inside walls of the frost wall to R20,backfill and insulate the top(under the slab to R20). Now I have a heat storage capacity of 6000 btu per square foot of storage. A well insulated house of 32 X 60 would require 10M btu for our 8ooo DHD area. 140 sq ft of thermal panel would supply all the heat needed for the house Heat loss to the house would handle the load on the shoulder months and mechanical heat extraction would only be needed in extreme temperatures.

    Summer cooling can be handled with nocturnal cooling. PV could supply the rest of the power requirements. Just my 2 cents.

  96. GBA Editor
    Martin Holladay | | #96

    Response to Roger Williams
    Your idea has been tried many times, and each experiment has led to failure.

    As I noted the last time this question came up, "To get a useful amount of heat from the sand during the coldest months of the year, the sand has to be hot enough to get water in a hydronic heat distribution system to at least 100°F. And that just isn't going to happen. The sand doesn't get that hot — or if it does, it doesn't stay that hot from early September (when it is likely to be hottest) until mid-November (when you begin to need it). Moreover, the pumping energy is a big energy penalty — parasitic energy that needs to be considered when analyzing possible benefits. Finally, the capital costs of all those extra solar collectors is high — an investment without a significant payback."

    Here are two GBA articles on the topic:

    From April 2011: Using Sand to Store Solar Energy

    From August 2010: Can Heat Be Stored in a Sand Bed Beneath the House?

  97. _Stephen_ | | #97

    Wow. That comment section was a wild ride.

    I'd be curious to know what people think on this topic at this point.

    I'm in the middle of building a net zero house, and my cost per watt of solar installed was $3 / watt. We have a 10.8 kW array that provides all of the energy for the house, and additionally, an extra 6 GJ worth of energy to feed into the electric car.

    This means that solar is only about 35% of the cost it was in 2009. Meanwhile, the cost of insulation has increased.

    My suspicion is that Passivhaus building is now much more expensive to achieve an objectively worse result. It's hard to beat net zero, especially when it's 10's of thousands of dollars cheaper.

  98. Expert Member
    Dana Dorsett | | #98

    The R-values in Table 2 of this document from 2009 are still about right from a cost-effectiveness point of view:

    While that was primarily about lifecycle economics, it's also pretty close to what it takes for a Net Zero house to have a PV array that fits easily on the roof (for most climate zones, anyway.)

    When that document was written rooftop residential PV was typically 12% panel efficiency, whereas now 15-20% is more common, and better class cold-climate heat pumps typically tested at an HSPF of 9-10 compared to 11-14 now. So if you fully optimized it all you'd probably be able to hit Net Zero at whole-assembly R values a full climate zone warmer than in Table 2, but sticking with the table still isn't insane, especially if using reclaimed foam board rather than virgin stock.

    For sub-slab foam, using reclaimed roofing EPS or reclaimed XPS can be quite a bit cheaper inch for inch than virgin-stock EPS was in 2009, and a heluva lot greener too, since no new polymer or blowing agents are being made. In my area reclaimed foam is usually less than 1/3 the cost of virgin-stock foam, sometimes less than 1/4 the cost, making PassivHaus levels of sub-slab-R less of a wallet-sting, closer to financial rationality no matter how one models the sub-soil heat transfer.

    It's amusing to re-read Graham's protestations about "...precise calculation of the net and incremental benefit of a given thickness of insulation..." . Precision isn't the same thing as accuracy, and the error bars on the benefits & performance levels are wide enough, let alone the CRAZY wide error bars on the future cost of energy (or carbon.)

    The cost of PV is still on a double-digit learning curve, with plenty of room left in the cost reduction ramp to go. The folks at GTM believe the bottom may be coming into view, but we're not there yet. At the utility scale they expect MAYBE $14/MWH (1.4 cents/kwh) for the year 2022, but under $20/MWh (2 cents/kwh) is all but certain:

    At the residential scale both Germany and Australia are hitting well under two bucks a watt (DC), all-in installed cost, before any subsidies, and that's likely to be happening in the US too before the tax credit subsidies evaporate. Leveraged against an HSPF 12 heat pump it's pretty cheap space heating, and even over-building the PV to the extent wintertime use is pretty much covered by wintertime production is probably cheaper than PassiveHouse even at $3/watt.

    Batteries are seeing a similar price curve for moving that daytime sun into nighttime heat, but that's still pretty pricey power even compared to offshore wind, which is now in the 7-8 cents/kwh power purchase agreement contracts, but that too is expected to be cheaper over time, even after the subsidies go away:

  99. qofmiwok | | #99

    Interesting (albeit old) read. I wish there were less about the economics compared to PV and more about the effect on condensation and mold.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |