I don't know if I thinking through this too much or if I should really be concerned. We have our old house up on jacks ready to move off the foundation and all the concrete is cut, should be an exciting week. It's almost time to start improving our existing foundation and getting it ready to build and now I need some advice. With the new design I have to move the footings for three teleposts as well I have to trench out the sewer about 15' to the new drain location, fill it all in, run the pex lines for the infloor heat and then pour Gypcrete. This will give us about 88" of headspace in the basement. Another thing we have to consider is type of flooring. We had planned carpet but with the infloor heat that won't work so well, any suggestions??? I keep looking at this and thinking maybe I should dig up the entire slab, insulate, lay pex tubing right in the new slab, and pour entire new slab. This would gain an extra 1 1/2" but cost a lot more. My thinking is if I just do the needed sections I can pour the footings and blend them into the existing floor smooth enough for the Gypcrete to go on top. As far as I can tell the slab is poured directly on the gravel, there is no moisture since the house was built in 1965 (my Mom owned it before me), and the basement didn't have any sort of weeping or waterproofing system besides some old tar. I'm trenching out the exterior to add weeping drains, water membrane, insulation, and dimple sheeting. I guess I just need to know that I'm doing it as best I can, any input???? Thanks, Sean

Not positive about this but i think if you are going to cover the floor with something you could save money by using concrete instead of gypecrete. I would repost this on the radiant board as well. Those guys are reat with this slab and radiant stuff. good luck!

Uh, yeah... no matter WHAT you put on the floor for a covering, if the floor is the heating system radiation you will be spending an inordinant fraction of your heating bill warming up the subsoil rather than your house without insulation under the slab.

If the slab is below grade and well below the frost line you might get away with as little as R5.  But putting R2 of rug on top is moving in the wrong direction, meaning you'd want at least R10 anyway and so on...

But there is a low-temp radiant solution that lets you have your cake & eat it too, getting your carpet to work FOR you rather than against you:

Putting a vapor-barrier on top of the existing slab, then R5 XPS between sleepers, then a 1/2"OSB/plywood sub-floor with carpet/wood/tile anything on top  of that will keep the floor from ever feeling cold.  Then a between-the-joist radiant-ceiling will warm the floor surface where your feet hit it, and the comfort level would be similar to a radiant slab floor. (Actually, it would have a bit more "give", less hardness than a slab floor, which is also a comfort factor adder.)  The radiant ceiling also heats the top surface of the floor whereever it's not obstructed by furniture- it sometimes FEELS like a radiant floor (!).  And the heat loss out of the floor will be less than or or comparable to a slab-done right, since the warmest part of the floor is now at the surface- inside conditioned space rather than in the rug/slab/insulation stackup where the warmest layer is the slab, which will be much warmer than conditioned space, for a bigger delta-T. 

The combined R-value of rug/sub-floor/1" XPS is about R7-R8, similar to fiberglass insulated 2x8 walls per square foot, but against a room-temp<->subsoil delta-T, not room-temp<-> freezing outdoors delta-T- it'll be less lossy than your walls upstairs.

You'd lose at most 2" of headroom from the existing slab (about the same as your gypcrete plan, maybe better.)  Insulating the foundation wall with either 2# EPS (the borate-loaded termite resistant type, if you're in a termite region) or XPS to at least R10 will also minimize the clammy-crypt basement feel.  Since you're already dug down to the footings, unless there are other factors that preclude it, insulating the outside to an even higher R-value using EPS is preferable & cheaper (it's cheaper per unit R-value, and won't eat up valuable floor area inside.)  By lowering the heat load with more insulation, it further reduces the radiant water temperature requirements, increasing system efficiency (particularly if you're using a condensing boiler.)  Insulating from the outside also puts the thermal mass of the foundation walls inside of the thermal envelope, increasing both comfort and seasonal heating & cooling efficiency.  But if you can't , R10 of XPS on the inside adds up to less than 3" of interior floor dimension lost, with higher performance than a 2x4 interior studwall (which has other issues in basements anyway.)

Wow Dana thanks, I guess I should add a little. We are using SIPs for everything above grade, 2lb spray in the ceilings (r40), joist ends etc., hydronic infloor heat on all three floors, triple low-e argon windows. I'm gonna trench out the foundation, waterproof and insulate the exterior and add a weeping system. The basement is 6' below grade. If I understand correctly the heat would come from above, that is, the main floor joist area heating downwards?? Thanks, Sean

Yep, a radiant ceiling is just what it sounds like: Radiant tubing (preferably with aluminum heat spreaders for lowest temps, highest efficiency) just above the finished ceiling gypsum, radiating downward.

Staple-up radiant ceiling without heat spreaders


with heat aluminum heat spreaders

With R40 SIPs above it's probably worth going R25-ish (6" of EPS) on the basement insulation ore even R30, at least down to the frost line, and maybe R16 (4" of EPS) below that.  The floor will still be pretty low-loss, but if you can take the hit in headroom you might consider R7.5 (1.5" XPS) with 2x2" sleepers 24" on center on the the sub-floor to minimize bridging. 

An uninsulated basement wall would dominate the heat loss of in an R40 SIP structure, particurly the portion sticking above grade- you absolutely need SOMETHING.

Hmmm, thanks again. Now that gives me two more questions. If I go with the heat spreaders I don't want to block the radiating heat coming off them so what's the best ceiling to take advantage of this. I like the idea, it'll keep me from putting a nail or screw through the PEX. Second, if I go 6" EPS on the exterior the sides of my foundation will stick out about 5" past my siding which starts 18" up from grade? I took some pictures of the house move today so I'll try to show what I have. Sean

Standard gypsum board (sheet rock or blueboard, etc) works fine for the ceiling- be sure to not use low-E paints though- you need the infra-red emissivity. NRT.Rob on the radiant heat forum has some experience with ceiling radiant heating design- I'd defer to his input on limits to water temps for radiant behind gypsum, maximum heat flux etc. (You might post a radiant ceiling question there, or search the forum for more info.)

For the above-grade portion of the foundation wall you may be able to chamfer the top edge to allow for proper flashing or design it a low slope toe/shelf top to it. Then, spray-foam sealing the sill & rim joist from the interior reduces the thermal bridging, halts the air infiltraion and restores some R-value to the transition to keep the thermal envelope continuousl. (2" thick half-pound foam is fine here, but if you're doing anything else in 2lb foam already, a 1" thick sealer layer of 2# stuff will give you a solid R6, won't break the bank, and it won't be thick enough to create a vapor trap with whatever flashing etc. on the exterior at that point.)

There are variations on stucco finishes, polymers, or cement-based parging that work well over exterior exposed EPS. (See the ICF forum here for ideas. Almost all insulated concrete forms are made of high-density EPS.) Worst case you might end up with a combination of exterior + interior insulation to make it work reasonably with the existing structure, but if you can, splitting the R-value with the higher insulation value on the exterior gives you some benefit to the thermal mass of the wall. If its in the budget, 2lb-foam on the above grade portion can get you R12-14 in about 2" of depth, and you can do the below grade portion in far cheaper high-density EPS. Then 2" of XPS or 2lb foam on the interior down to the frost line brings up the most-critical portion to R20+. There are several ways to attack it, but getting the clear-wall R-value of the above-frost-line portion up to some thing similar to your first-floor wall R-values is important. A 6" thick concrete wall alone has some thermal mass, but contributes a steady-state R-value of only ~R0.5. (Even a two foot thick wall has an R value of ~R2, but the substantial thermal mass saves you from the peak loading of daily temperature swings, so the heating/cooling systems only "see" the daily average delta-T, not the peaks.) Figure you'll get ~R4/inch of thickness for EPS, ~R5/inch for XPS,~ R6-7/inch for 2lb spray foam.

Visually it might look odd if it sticks out 3" beyond the siding or something and it would need sufficient slope & flashing to drain well. Some approaches to the step-out flashing/parging:

http://irc.nrc-cnrc.gc.ca/images/rr/rr199/Fig6-9.jpg

http://www.greenbuildingtalk.com/Forums/tabid/53/forumid/4/postid/44303/view/topic/Default.aspx

http://www.appliedtechnologies.com/home/thru-wall_flashing.html

http://www.enertia.com/DesktopModules/TekDesign-LogCabin/Thumbnail.aspx?IP=/Portals/0/gbwatershed1.jpg&IW=300&IH=500

http://www.rambuilders.com/ut_pubdrawing_files/att_t1.gif

FWIW: When I insulated my foundation to ~R20 (all interior in my case) it cut my heating fuel use per degree-day by fully 20%. The basement had never seemed particularly cold, but the heat loss was clearly there. Doing the math on a simplified thermal model the magnitude was obvious. Since you're actively heating the space with low-temp radiant, it's even more critical for YOU to get it under control than it was for me. (I wouldn't care if the basement stayed 60F all winter, but you might. In practice it stayed in the low-mid 60s post-insulation, but I haven't insulated the slab. My subsoil temp is ~53-55F, so R5-R8 on the floor makes sense for me too, but it's still a bit down the project list.)

I have to thank you again Dana, you've done a great job of explaining this as well as giving me examples and alternatives, I don't get that very often. Those links led me along for about two hours, I gotta learn to focus. Sean

Dana, you the man, lol. Hope you visit my posts as well, seems like sgo is already quite a bit more knowledgeable than I am, and I have a long way to go on this house. I wanted to hijack this thread with my own questions, but I think I've done that too many times already, ha.

Hey, ask away sometimes it brings up some good discussions. Sean

I finally finished up my post, lol, grab a coffee and enjoy some night time reading... lol.

Considering you are spending a lot of money on this project, why would you do this and then end up with 88" oc ceiling hgt? 6 more inches would make a relatively normal hgt room. and you could add some awning or casement windows on one wall!

I wanted to use the existing foundation to save myself about $35-40000, nothing comes cheap in Calgary, so that was our base. The max height we can go is 28'2" and no matter how much I begged for another couple of feet there was no way I was getting it, so I had to change my main roof line from a 7/12 to a 4/12. It still looks good but we are within about an inch of our max height, I'm still trying to sneak out another few inches, maybe double the sill plates. I don't think we will use the basement much anyways but it's nice to have the option, also I'm only 5'9" and my wife is 5'5" so we're safe. Sean

Heh, i almost raised my house, that was what I had planned at the start since i was going to redo all the electrical and plumbing anyway, i didn't think it would be a big deal, well my mason said there was no way he was going to block it back up to height, so i had to skip that avenue... Cool to actually see someone else doing it though, i had to just go up and out from where it was. Yeah, foundations are not cheap...