Posted By jonr on 13 Mar 2011 10:34 AM
Good question - how does one insulate a slab with beams?

My cement finisher told me he finished a concrete slab over spray foam two weeks ago in Calgary. Its the first time I have heard of spraying 2" of foam on drainage rock (1.5") but don't see any reason it wouldn't work. If you made the ditches for the beams wider to accommodate the foam it might work for beam/slabs. Bob

At at trade show in Birmingham this past weekend I was told by David Vaughan, Director of Engineering/Product Development for Burtin Polymer Laboratories that the type of cell foam to use below grade was 2.8 pound per cubic foot closed cell.  He also mentioned that a variance from the code may be required for heavily termite infested areas such as the Southeast before foam can be used below grade in contact with the earth.  We were primarily talking about using foam below grade on the exterior of concrete walls but the same may apply for slabs.  At the Nudura booth I was told that Colphene waterproofing membrane that contained a termiticide could be used without having to get a variance to use foam below grade.  I was also told that adding borates to the ICF foam made it more brittle.  Has anyone else heard about this and are these two gentlemen correct in what they told me about using foam below grade?

Posted By toddm on 12 Mar 2011 08:28 AM
For those who struggle with the English language, as Snoslidr74 does, when you read "anecdotal evidence," it means don't generalize from this data.
My point was that Dana wasn't grasping the nuances of sunbelt construction. Neither is Snoslidr if he thinks the operative depth is 40 inches, and he doesn't put an asterisk after the third of building envelope business.*
My bad on the xps. It's a $1/sf around here, or $2.5k for the rancher in question.
Town Creek, you won't find a satisfactory answer to this question online, thanks in part to the Passivehaus purists. If you are going to rule out any fossil heat contribution, then you have also ruled out trade offs, even the net result is a wash, or close to it, after the heat pump cranks longer in some months and shorter in others.
If slabs are common in north Ala, your neighbors may be more help than Yankees.
*Subject to conductive heat loss rather than radiant or convective heat loss, which makes it an entirely different question.

I may not be the most-nuanced, but I usually start by taking a look as subsoil temps and seasonal averages for the specific location when looking at whether foundation insulation rather than comparing it to the utility bills of a house 500 miles away with a different HDD/CDD climate & subsoil temps with difference seasonal peaks & averages.  Perhaps pointing out that the LA PassiveHouse  had substantial underslab insulation was a bit of a red herring though, since Town Creek obviously isn't going PassiveHouse on this, but Lafayette LA (where they built that house) has slightly warmer subsoil, and warmer winter-averages, but  a similar cooling season profile to Auburn (unlike Dallas which has warmer weather averages year-round, with a similar winter profile to Lafayette.)

Town Creek was asking whether slab or perimeter insulation was going to be a net loss on primary cooling & heating energy given Auburn's climate. The quick math says no ( L'aritmetica non è un'opinione ) and the fact that even R16 under the slab in a home in a similar but slightly warmer climate was important to meeting their low-energy design goal is only the existence-proof.

But the simple math doesn't say whether slab insulation will be cost-effective in 10 year NPV, or whether it's the next-best place to spend the money on energy efficiency in this (barely described) house.  There's likely to be a 25 year break-even on a couple or three inches of EPS (at least around the edges),  but there are too many variables to say with any certainty that it would ever "pay", even in 100 years.  It's not huge money to install in advance, yet cost-prohibitive to install it after the fact.  The fact that Town Creek is springing for spray foam everywhere implies that it's NPV on utility savings isn't the primary motivator for wanting to know whether insulating the slab was working against the energy efficiency goals.  While answer for Auburn AL is "in general, no", without modeling the entire design with respect to the site there isn't a number you can assign to it, and if other aspects of the design aren't attended to it's clearly possible to build something that might make skipping the insulation more energy efficient.  (But that would be the exception, not the rule, and any potential down side would be very small.)  Unless the whole-wall R and attic insulation is at least ~1.5x code-min there may be a better financial argument for putting the money elsewhere.  (Open cell foam in 2x4 16" o.c. construction runs ~R10-11, but it's still comparatively cheap to bump that to ~R20+ whole wall values with 2x6 24" o.c. construction with spray cellulose cavity fill and an inch of XPS or iso on the exterior.  That would likely have a larger effect on utility bills & comfort than under-slab insulation in most 2x4 foamed-wall houses in Auburn, AL.)

FWIW: I may live here now, but I am by no means a "Yankee" by birth, attitude or life-history, nor am I a PassiveHouse purist. I've never set up residence in TX or AL, but I've worked in both states. ( I have lived in Tucson, AZ, which IIRC has a very non-New England type climate, with different thermal & moisture issues than where I currently live. ;-) ) But I agree that local expertise, particularly from somebody who can actually review your actual design & site plan might provide better, more detailed answers than can be delivered on a web-forum.

Regarding insulating a slab with beams, I suppose one could pour the beams, lay foam sheets on top and then pour the slab. I think the psi ratings work out.

Dana1,

Thank you for stating your in-depth reasons for your answer.  I find it very helpful and agree with you.  This is from someone who is originally from KY but has lived, taught, designed and built in the Auburn and Southeast area since 1983.  Most of us by now should know there are no simple answers when it comes to what is best or the most economical when dealing with energy consumption or savings.

Alton, Dana1 and any others thank you very much for your careful answers & thoughts. I had never used a forum such as this. i asked the question & then shortly my computer crashed, then i had an avalance of responses as i was now able to review.
I will be in touch with Alton for his counsel on this as he lives here in Auburn.

Sincerely,
Town Creek

Not to beat a dead horse, but I did stumble across what was the most direct guidance/recommendations I've found on slab insulation. The Building Science Corp. publication RR-1005 High-R Enclosures for All Climates, Table 0.2 on page 10 lists recommended insulation values for various portions of residential structures, and is delineated by ASHRAE climate zones.

http://www.buildingscience.com/documents/reports/rr-1005-building-america-high-r-value-high-performance-residential-buildings-all-climate-zones/view

Auburn appears to be in Zone 3 which, according to the Table, has a recommended sub-slab insulation R-value of R-5. Slab edge insulation is recommended at R-7.5.

Again, take this info in the context of the document.

Thanks!

snoslidr74,

Thanks for posting that reference and info.  Maybe it is time we started insulating our slabs in mild climates.  I have never known anyone using insulation under or around the edge of a slab in the Auburn area.  Maybe I can spread the word. 

Thanks for the BSC document, soslider74!

Going with 2" of Type-II EPS (R8) would only be a slight uptick on installed cost from BSC's minimum recommended R5 using 1" XPS, and probably worth it for "...clients who value environmental impact highly", the end of the qualifying statement BSC uses for their recommended R-value table (Table 0.2 on P10 of the RR-1005 document), and would most likely apply to Town Creek. The cost adder of bumping that all the way to R12 is also pretty small if using Type-I EPS rather than XPS or Type-II EPS.

Their Zone 3 recommendation in table 0.2 for R20 walls and R50 attic is also in-line with what I'd suggested in my rambling discussion too (I'd suggested lower-cost methodology R20 whole wall construction and a R40+ attic.) So I guess my off the cuff WAG wasn't too far off. That table is their recommended MINIMUMS for whole-wall (thermal bridging included), which means that in their economic analysis it'll at least pay off eventually (if not for somebody who is only living there for 6-7 years), using fairly conservative energy pricing models and standard construction methods & materials. Going higher than the minimums using cheaper material or methods (like celluose cavity fill + rigid board vs. an all spray-foam solution, or using EPS rather than XPS under the slab) is likely to shorten payback times even at a slightly elevated day-1 cost, but you'd have to run the NPV calc on the particulars to figure the break points, if you're sharpening your pencil that finely.

The discussion about the economic aspects of higher-R enclosures (p9 of the document) points out some of the ways it's possible to under or over estimate the present-value of high-efficiency building envelopes vs. high efficiency mechanical systems, etc. I expect most people don't really run the numbers, but refer to DOE, local code, or other generic tables such as the one BSC put out there.

For the true energy nerds & penny-accountants,the NREL BEopt is probably the right tool for fine-tuning the options:

http://www.nrel.gov/buildings/pdfs/39929.pdf

...but SFAIK it's not yet available to the general public. The PassiveHouse tools could be utilized in a similar fashion though, with a bit of off-screen analysis of the results.

Buildings & utility rate particulars will differ, but most of the time going 1.5-2x code-min on the R-values on new construction is still economic on a monthly cash-flow basis using average energy pricing & mortgage rates. But beyond the pure financials, even where BSC doesn't recommend going that high, the upgrade to R20 walls (2x6 fiber-fill + 1" iso) vs. R10 (2x4with batts or cellulose) and R40+ attics makes a noticeable difference in comfort at the temperature extremes. If energy were free there wouldn't be an economic rationale for ANY insulation, but the comfort factors of walls/floors/ceilings that stay close to the room temp are real. I wouldn't build anything less than R20-whole-wall for myself even in Zone-1. (YMMV) It's a relatively small premium to pay for some additional comfort. (And I wish my Zone-5 New England antique were easy & affordable to upgrade to R30+ walls the way new construction can be.)

Dana, a nuanced answer would not plant footings and stem wall in floating slab country.

Town Creek asked if underslab insulation would work against the earth's cooling effect, and the answer is yes in some months and no in others. If slab insulation is the gimme it's made out to be here, you'd think Alton would have come across at least one by now.

Nor have you answered how one insulates a floating slab. I suppose you could put wings around it like frost protected shallow foundation but then you'd need a second round of excavation. While spray foam could work well, I am guessing it is a bit more than $1/sf.

Snoslidr, in radiant heat loss from walls and windows, delta T is relentless, but not so in conductive heat loss in well drained soil. The heat goes in but it dissipates slowly, which is why soil temps are constant at about 15 feet or so, the frost depth is predictable, and people build earth berm houses. So the molecule of concrete heating the molecule of dirt below it in the exact middle of your slab might have a delta T of 1, being a degree warmer. From there it gets complicated, requiring a 3D heat loss model and a good understanding of local soil characteristics. Happily the Canadian government offers both here:
http://canmetenergy-canmetenergie.nrcan-rncan.gc.ca/eng/software_tools/basecalc.html

If you modeled your slab using Denver weather data, I think you'll find that the slab's heat loss isn't even in shouting distance of heat loss from windows and walls. I am assuming perimeter insulation because heat loss from the slab above grade is in fact radiant heat loss.

A caveat: Basecalc is for basements but it can tweaked enough to guesstimate slabs sitting on stem walls and footers. It's useless for floating slabs, which have a whole lot more angles and surfaces.

licensed architect in CA, CO, KS, MO, NC, and NM here.

I agree with above statement that the earth cooling effect in summer could be advantageous.
I would like to see discussion here about moisture proofing in the south.......

our only project in south was in Ft. Worth, TX where we did an insulated perimeter slab on grade
waffle slab over pier footings in sand to bedrock. with moisture barrier beneath slab.

should not Auburn have a moisture barrier beneath slab on grade?

I think most builders in Auburn place thin vinyl under slabs.