Insulating a raised slab
Alright, so here’s one that I can’t find in the great treasure troves of this site: how to handle insulating my raised slab on our new house. Quick details on the project: climate zone 6, 3500sqft, two story, zip-r external insulation/sheeting (R6), fiberglass batts in 2×6 walls, combination vented roof w/ R60 cellulose and unvented roof with R50 foam, low U 4th surface fiberglass windows, and a combination of slab on grade and vented crawl w/ R48 foam. The concrete will be polished as finish floor in slab areas. Heating is entirely mini-split, no in floor radiant.
Project is built over a very steep hill, with the steepest portion the vented crawl, everywhere else we built up concrete walls and back filled for slab sloping back to grade. As such the slab on grade portion ranges from 6” above grade to 10’ above grade… currently we have R10 under slab with edges turned up and cut at 45° around perimeter. The engineer required integrated rebar ties to walls around entire slab, 3’ rebar stingers from wall are bent at 90° to tie into slab. No vertical insulation was installed on inside of concrete walls below bottom of slab.
My concern is thermal transfer from external wall system to slab… is the only option insulation on the exterior? If so, how far down the walls would need to be insulated to keep the slab warm? I can find little information on thermal transfer properties of concrete walls, and am curious how much geo warmth we’ll get from ground temp up through the back filled areas and if that will help combat cold transfer from external walls.
Appreciate the help!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
IRC 2015 code requires R15 continuous insulation between the conditioned space and the soil on the walls. With 4" of EPS from the foundation sill down to the footing you'd get there, as would pouring those walls in an insulated concrete form. There will need to be Z-flashing to redirect bulk water from the ZIP-R to the exterior of the concrete wall's foam, and some sort if EIFS finish on the exposted above grade EPS.
User-6804212,
First of all, can you tell us your name?
I'm not sure I understand your question. But it sounds as if your detail is similar to the detail shown in the illustration below. Are you simply worried about heat flow at the slab/wall intersection, where the foam has a bevel cut?
.
Martin, name is Alex, fixed my profile after the first post. Been a subscriber since I started designing the house last year but just haven’t had to post because I was able to glean what I needed from existing resources. You nailed the detail, the exterior foam on stud walls (R6) and bevel cut on foam (R10) at edge of slab is exactly how we had to do it. The exception is the continuous rebar tie between wall and slab which penetrates the vertical foam every 2’, and our slab is up to 10’ above grade with stem walls exposed to air on the exterior (sand/gravel compact backfill on the interior).
Worries about heat loss in general of the slab... it’s primarily a comfort concern and secondarily a whole house energy loss concern (which is why I noted the rest of the stats in my post for comparison): without radiant heat in slab (which we don’t have enough insulation for anyways), will this slab warm up and be at least an acceptable level of cool in the winter? Or will we have freezing outside edge temps in the winter when it’s 0°F and warm edges in the summer when the south facing stem walls are exposed to 100°+ sun?
Thoughts?
Alex, the heat loss through the rebar is certainly measureable; concrete is thermally very conductive, and steel even more so. A three-dimensional finite element model would be needed to calculate it accurately. As an estimate, some quick math says you might have about 90 of these bars? If they are #4 (1/2") bars, that would be a total of about 18 in² of steel essentially connected directly to the outdoors, or a steel plate about 4 1/4" x 4 1/4". While that is certainly a bummer, with that amount of fill, I would make sure my structural engineer was happy. You're probably talking "tens of dollars" a year at most in energy loss at most, not hundreds.
I would not be surprised if the surface of the concrete near the steel bars got below 50° or so some of the time, but the mass and conductivity of the concrete should spread out heat loss enough that I doubt you would get frost. The temperature will almost certainly be low enough for there to be condensation some of the time, but it's hard to say if it would be enough to cause problems. In the summer the energy penalty would be very low, with no moisture issues.
If you want to reduce the heat loss, you could try thermally isolated reinforcing systems, such as those available here: https://www.schock-na.com/en-us/products.
With your detail, the sub-slab fill at the perimeter of the building will be near the outdoor temperature. This soil should be well-drained so there is not really any danger of frost heaving. You could insulate the interior of the wall down to the footings, which would keep the entire sub-slab zone much warmer. Or you could add more horizontal insulation below the slab. Which of those would be more effective for the money spent is a tricky calculation. With your tall walls, the sub-slab foam is probably a better investment. Or, if the slab is not yet poured, you could do a double slab, with the lower one tied into the wall, with foam and another slab (or other finished floor) above. Your engineer might even allow just a perimeter slab tied into the wall--something that looked like a strip footing, cast against the interior of the concrete wall, which would serve the purposes of bracing the concrete wall horizontally and would also prevent the finished slab above from settling.
How effective would adding exterior foam board insulation on the outside of the stem wall? How much would it take to make a difference and how far down from slab would it have to extend to be effective? I know some may say all the way to footing or below frost line, but I wonder what the real transfer would be of cold air to slab if we only went 4’ down with, say R10, on the outside of the stem walls? Would it be enough to mitigate or...? I’ve been musing about the calcs for FPSF, at some point the thermal transfer of air temp is mitigated by ground temp, in a raised slab situation I wonder where one starts receiving diminishing returns on depth of vertical insulation below the slab.
Alex,
The answers:
1. Yes, exterior rigid foam will help.
2. Thick foam is better than thin foam, but even 1 inch of rigid foam will reduce the chance of interior condensation.
3. I would say that the minimum width you should consider installing is 2-ft. wide foam. If the exterior rigid foam is proud of the siding, remember to install Z-flashing at the horizontal joint between the bottom course of siding and the rigid foam, and remember to protect the above-grade foam with metal flashing, pressure-treated plywood, or stucco.
Stopping R10 exterior foam at 4' below below the slab, or 4' below grade is good enough where the slab is at or near grade. On taller walls that are more than 2' below grade, bring it 2' feet below the slab, or down to the footing, whichever is shallower.
Deep subsoil temps in zone 6 aren't exactly "warm", and even though soil has some R-value (and a lot of thermal mass), leaving any portion of the below grade wall uninsulated would be a mistake, particularly in zone 6A, where summertime outdoor dew point temperature averages are in the 60s F well above the 40s Fdeep subsoil temperatures, creating a mold hazard if left uninsulated. (Do you want it to stink like a musty basement in summer?)
For an idea of what your local deep subsoil temperatures are, see:
https://www.builditsolar.com/Projects/Cooling/US-ground-temps.gif
For local shallow soil temperature averages see:
http://waterheatertimer.org/images/average-groundwater-tempera.jpg (<< note, these are the annual averages, with winter & summer variations dependent upon actual depth and seasonal weather history.)