Hello

I have been reading this forum for some time but have a question I haven't yet seen adressed so am I posting.  I am building a new house- northern state lots of clay and deep frost lines.  I wish to insulate the basement floor (30x60) for radiant heat.  Basement slab will be way below frost line.  My plan is to put 2' of XPS under the floor and 1-2 inches of XPS surrounding the perimeter of the basement floor slab.  There is currently a ledge of concrete that sticks out 5" in some places and only 1-1.5  in others at the bottom of the vertical wall.  When I put the 2" of XPS on the floor it will be level with ledge and the basement concrete will cover the ledge with 4" of concrete.  I believe I can cover the 5 inch ledge with 2" of xps both vertically and horizontally and be OK with the slab sitting on the ledge.  However, if I put even 1" of insulation on the 1 " ledge there is no lip for the basement slab to sit on the ledge.  My concrete guy will do what I want but doesn't th ink the peremeter insulation is a good idea because it will effect the integrety of the walls.  I'm confused as I was under the impression that a non-structual slab does't tie into the walls without rebar.  It seems to me that if the having the concrete laying over the ledge was important they would have poured the concrete with a 5 inch ledge all the way aound.  I want to insulate put don;t want to screw up the structure of my house foundation.  My questions are: will the 2" of foam (25 psi)between the slab and the wall be crushed by pressure and allow the vertical walls to move and does the slab need to sit on this ledge.  If the slab needs to sit on the ledge do they need to come out and repour something and get the ledge closer to 5"?  I h ave attatched photos of the wall with scraps of insulation in place to help in case my description if bad.  Thank you very much, N

Try again on the photos

It is not necessary for the slab to sit on the footing. It is critical that the slab be completely isolated from the walls, from the ground and from the ledge.

Sounds like the concrete contractor is assuming the walls need the floor to keep from collapsing into the basement, but thats pretty far fetched unless you have a huge problem with earthquakes. The walls should be stable on their own, and the foam has some structural value. Bottom line is you should be fine. My one concern is that you should have a thicker layer of foam - I'd recommend a minimum of 4", and 2" minimum between the wall & the slab. And if height is an issue you can get by with 3" of concrete.

Northern writes: My plan is to put 2' of XPS under the floor
Bob writes: My one concern is that you should have a thicker layer of foam

I don't know, I think that two feet should be plenty :-).

Thanks Bob for your answer. Jon - yes, 2 feet of foam should be adequate! Couldn't seem to get the pictures posted- sorry.

But it's nicer to the planet if you use EPS instead of XPS, since XPS is blown with stuff that has a global warming potential more than 1000x that of CO2, whereas EPS is blown with stuff that's only 7x CO2. It's the same polymer, comparable moisture tolerances (EPS edges out XPS slightly in the long term pond-soak, XPS edges out EPS in the short-term spritz.)

If think you really need 25psi goods (experience says you don't) you can get it with 2lb density "Type-IX" EPS.

PassiveHouse builders float 4" slabs on a foot or more of 15psi EPS (1.5lbs density aka "Type-II") all the time, unless local code spells out something different. As well as making an efficiency robbing thermal bridge, resting the edge of the slab on the footing give it a fulcrum point, and is more likely to cause cracking than floating the slab on a cushion of EPS.

For a radiant floor in US climate zone 6, R15 (4" of EPS) would be the minimum, and depending on the heating fuel R20 wouldn't be insane. For a slab that isn't the heating source, R12 (3" of EPS) would be fine.

What zone are you in? http://tboake.com/carbon-aia/images/climate_zones.jpg

I thought the XPS was the better stuff, thanks for more info Dana.  Not sure what zone it's considered but I'm in SD.  Our builder thinks we're over doing it with putting insulation below our radiant slab so it's good to hear people actually use more.  Thanks for the info on the concrete!  N

First, the ledge is the spread footing under the foundation wall. Its job is to spread the weight of the building over a larger area of soil then just the 8" or so of wall will provide. Sometimes the footing is not cribbed quite as accurately as the walls which allows for the variation of 1" to 5". Probably if you looked on the outside where the "ledge" is showing only 1" on the inside, you would see 5" on the outside and vis versa.

I understand where your concrete guy is coming from. In high frost areas wet clay soils tend to provide significant frost jacking. This is when the waters trapped in the clay soils expand on freezing. This will provide enough force that an unrestrained foundation wall will move inward. Each frost cycle (each year) will move it a bit more. In extreme situations such as unheated houses with no lower restraint, it is not unusual to see as much as 1/2 to 1" of movement in a single cycle. The depth of backfill and length of walls also play a part in this equation.

In a heated basement, there is little chance of wall movement as there is sufficient heat loss to the outside to prevent the soils next to the foundation from freezing. As you increase the insulation values, you let less heat escape, therefore the frost conditions get stronger closer to the wall. If you get the foundation wall insulated to around R20 there will not be enough heat loss to prevent frost jacking on the colder years so your contractor is concerned that the walls need to be restrained at the bottom as well as the top. This is normally done by pouring the floor slab right against the foundation wall. Now any frost jacking from one side pushes against the floor slab and transfers the push action to the opposite wall where it prevent the push from that side. The contractor is concerned that the foam you insert between the floor slab and the foundation wall will crush and allow the walls to move inward. If you stick with your 25 psi foam and it is consistent along the whole wall and the floor slab is 4" thick, I don't believe you should have a problem. You can google the psi for frost jacking for your area and compare that to the 4" per foot at 25 psi you are getting from your floor /foam interaction if you want to be sure.

There are also other things that prevent the foundation wall from moving. The friction of the bottom of the footer on the soils. The drainage material that you have place over the clay base of the house and against the side of the footer are some of them, however these are dependant on the connection between the footer and the wall. If the builder put a keyway or dowels at that point they will provide quite a lot of resistance to the push. If on the other hand, there has been a membrane inserted to prevent damp rising, then these conditions will do little to prevent the wall from moving as the wall will slide on the top of the footing.

There is no need for the floor slab to sit on the footing. All the floor is carrying other than its own dead weight is the occupant load that inhabits the basement and what ever furniture or materials you store down there. Usually an insignificant load. You might expect to see some cracking at the point where you have covered the footer with two inches of foam as that will reduce the slab thickness to 2". Concrete like to crack at thickness changes.

With a 30 by 60 slab you will have to install some form of crack inducer. It is normally done by cutting a 1/4 of the way through the slab from the top. As the concrete shrinks, these cut lines will induce the slab to crack at there location as they will be the weakest point in the slab. If you don't induce cracking, the concrete will crack anyway but were ever it likes.
As you are installing radiant heat, saw cutting can be problematic unless you are bottom loading the pex. Another way to induce the cracks is to install 1" x 1" drywall angle on the foam with one leg pointing up. As this cuts the slab by 1/4 from the bottom and the slab will crack over these angles in a controlled manner. If you can place these under walls so much better as you will not see the cracks. The slab should be cut once down the middle and at least twice across. Concrete if left to its own devises will usually crack into 20 foot by twenty foot squares and at any change in width, length or thickness. Always induce cracks at these points.

Posted By Northern on 08 Aug 2013 10:37 PM
I thought the XPS was the better stuff, thanks for more info Dana.  Not sure what zone it's considered but I'm in SD.  Our builder thinks we're over doing it with putting insulation below our radiant slab so it's good to hear people actually use more.  Thanks for the info on the concrete!  N

That's why I included the link:  http://tboake.com/carbon-aia/images/climate_zones.jpg

Most of SD is in US climate zone 6.   In climate zone 6 the long term economics of R10 (even using the more expsesive XPS) is arguable even without the radiant floor, even if you're heating with natural gas.  If you're heating with propane/electricity/oil R15 can even be rational on a lifecycle basis. WITH a radiant floor R15 would be the minimum, and R20 isn't insane if heating with a more-expensive energy source. That's 4-5" of EPS, 3-4" of XPS.  EPS is typically ~9-10 cents per R per square foot, XPS is usually 12-13 cents. So for rough budgeting figure on $1.50-2.00 per square foot.

For a mid-point on what's economically rational long-term for R-values by climate zone, see table 2 p10 of this document:

www.buildingscience.com/documents/r...mate-zones

Note, those are "whole-assembly" R-values, with the thermal bridging of the structural elements factored in. (eg. A 2x6 studwall with R20 cavity insulation comes in at about R14, not R20, once the R1.2 framing is factored in, and the R of the wallboard, sheathing & siding are added.)

Deep subsoil temps in SD run in the mid-40s F, and while the soil itself has an R value, that value varies a lot by the local soil type and it's moisture content.  In addition to saving on heating BTUs, by insulating the slab you will avoid moisture accumulation into the floor from the more humid summertime air, since the slab temp would then track the room temp, not being pulled cooler by losses to the subsoil.  The amount it takes for mere moisture protection isn't much- R5 would do it, but it's worth pointing out that added benefit.