Dana et al. Been watching this discussion since I have a design on the way for a house in a location (zone 4/5 line in a marine environment) that has notoriously high dew points (in the low to mid 70's). I will have radiant in a basement slab so that changes the discussion but I will be putting R 20 under the whole slab. I have a wife with mold sensitivity and I want to make certain our basement is mold free. The groundwater is way down so that is not an issue - it's condensation that brings on the mold. I will have solar DHW and may oversize that a bit and dump excess hit into the basement slab to keep the surface above the dew point. This is not a question of energy payback however in this area it cost about $300/SF to build or more so having an 1,100 SF basement that is usable is definitely worth $2K worth of insulation!! I agree with Dana in that it's not always about the end formula on energy cost and savings that drives the decision....

What window did you end up going with if you don't mind me asking?

Pella Windows. There were about 70 windows and doors in the package and most of them are Architect Series aluminum-clad wood. Very pleased with them. Nice craftsmanship there in Pella, IA. I have some big exposures under constant wind and the windows are bombproof. A good fit for an ICF home because when they close, they seem to lock out the outside wind and noise, etc.

Just to reiterate, Marvins were nearly identical and architects like them because they say they can do just about anything, but the price was at least 20% -30% higher and then there is the challenge of getting them to do what they say they will.....

Posted By toddm on 29 Aug 2012 07:54 PM
So, in summary, your comparison of R5 full underslab insulation coverage in your addition vs zero sub slab in the original house tells us nothing about the relative comfort merits of full slab vs perimeter (and slab edge which icf gives you in the addition). The perimeter counts; the middle doesn't. In normal soil anyway. IIRC, you have water table issues as well.

You don't dispute that full slab coverage saves very little money. You don't dispute Basecalc's estimate that a Minneapolis basement in January loses heat from the middle of the slab at the miniscule rate of 0.67btu/sq ft/hr. Yet, somehow, sloooow changes in midslab make a major difference in comfort. How? And where is the proof?

Your summary misconstrues and assumes a lot.

BaseCalc is (and always was) by definition crude estimate and doesn't account for the radically different thermal characteristics of different soil types & moisture levels, etc., and focuses entirely on heating costs, ignoring latent load AC or mold risk factors.  BaseCalc is better than a WAG, but it's numbers should always be viewed with large 1-sigma error bars. (IIRC it tends to shoot to the high side of measured heat loss.)

Did somebody state that mid-slab insulation made ANY difference in comfort? (I sure didn't.)

My understanding is that PassiveHouse modeling is much better on precision end, but the cost of the soil testing & groundwater surveys required to narrow down whether you actually NEED that extra R15 or whatever to meet the Passivehouse energy use requirements would be higher than the cost of R15 of EPS.  (I've never read of a PassiveHouse design where all factors were actually measured.) In some ways the BSC recommendations may suffer from a similar fudge-factor- at 10cent/R/square-foot it's cheaper to install the recommended values than to analyze whether it could get by at half those values for any particular site.

You recall correctly that the average water table isn't very far below the slab at my house. If anything  the in-situ observations between the insulated and uninsulated slabs regarding slab temp & mold risk would tip the balance in favor of full-slab insulation as relatively cheap insurance. 

The notion that slab-edge insulation several feet below grade on walls with R18 insulation that extends to a few inches above the slab (and below the 100 year design frost depth) would somehow raise the center-slab temperature above the summertime dew point is absurd on the face of it.  But it's true that my older slab has no slab-edge insulation.

The situation Marc Rosenbaum ran into in his Martha's Vineyard house where the house was built with an insulated slab but without foundation-wall insulation, and a batt-insulated basement ceiling, it would be a structural hazard in US climate zone 7, (or even the colder parts of zone 6) and much of Canada.  By insulating the slab and the basement ceiling yet not the walls the direction of heat loss at the slab is reversed in winter, since the basement falls below the ground temperature.  Frost heaves grow in the direction of heat loss, and with a sub-freezing basement at the 100 year frost-depth design extremes it could buckle the foundation wall or the slab.  With insulated foundation walls on a heated house, even if the heat loss is low, the direction of the heat flow is always from the basement to the ground, and will never be going the other direction in frozen soil.  (In US zone 5 that risk is pretty tiny even for slab-on grade, and easily avoided with code-minimum slab edge insulation.)

I have much better things to do than play I-am-Dana-so-I-can't-be-wrong, but there is a danger here that readers might follow our doughty expert as he wades deeper into the weeds. This is my last post. My apologies to Cards66.

Call me silly but I see three broad criteria for accepting Dana's two-part slab as an answer to my question. (The question: Is there a comfort penalty in insulating only under the outer three feet of the basement floor vs insulating under all of it?) The first is that he compare a slab with perimeter insulation to one with full coverage. (Duh) He does not. The second is that extraneous factors be more or less equal. They are not because one has slab edge insulation and one does not. The third is that his example be broadly applicable. Again, thanks to his water table issues, it is not. At least in baseball, which is equally tedious in my view, you're obliged to sit down after three strikes.

But on to the errata. Do not read Dana's "absurd" and "silly" comments as dismissive of slab edge insulation. Heat loss underground is different because well drained soil cannot carry off btus fast enough. As the soil adjacent to the slab warms, the temperature difference (delta T) shrinks and heat loss slows. To complicate matters, where this process occurs in the foundation results in very different delta Ts: high ones on basement walls near the surface; lower ones deeper down the wall; and lowest under the slab. In fact, Themost important fix to the slab is to break the thermal bridges from floor to wall and floor to footer in order to keep it in the "lowest" camp. The first part of that job is done by slab edge insulation. The second can be done by perimeter insulation. Dana's original basement floor has neither.

As for basecalc, Dana apparently has never used it or he would know that the model has built-in soil and weather data for some U.S. cities, including Minneapolis. Adapting it to your site is super simple. Print out the user-defined variables. Take it to your county ag extension office and ask the soil experts there to fill in the blanks. As Dana says, it's not gospel but it's better than a WAG, yours, his or Building Science's.

Yeah, whatever, I can find at least a dozen different soil types within 5 miles of my house (and you probably can too.) Site specifics matter.

Well I got another bid on insulation today. This was for a system called Air Tite System. Here is a link to the brochure. http://www.starcompanieskc.com/ATS.pdf http://www.insulate123.com/pdfs/ATSbrochure.pdf Wanted to get opinions on this hybrid system and then put the EPS foam on the OSB? I was steering away from Fiberglass but this system looks like it might work.

Posted By cards66 on 02 Sep 2012 04:21 PM
Well I got another bid on insulation today. This was for a system called Air Tite System. Here is a link to the brochure. http://www.starcompanieskc.com/ATS.pdf http://www.insulate123.com/pdfs/ATSbrochure.pdf Wanted to get opinions on this hybrid system and then put the EPS foam on the OSB? I was steering away from Fiberglass but this system looks like it might work.

So it's basically a 1/3 spray foam design with 2/3 fiberglass batts. Therefore it is not as good as a complete spray foam or cellulose install but still slightly better than a complete fiberglass batt install. I am not a fan of fiberglass batts. Blown in cellulose is way better than fiberglass.

Their R-Values are completely off in their brochure because they completely ignore the thermal bridging of wood. You can knock that R-23 down to R-15 when you factor in all the thermal bridging. In addition you knock off some points because fiberglass batts never perform as well in the real world as their packaging claims.

IMHO, either do it 100% spray foam or go blown in cellulose, don't mickey mouse this stuff. Yes, do the EPS foam on the exterior, that will help with thermal bridging. How much EPS on the outside are you looking to do?

wow, didn't know subslab insulation was such a passionate discussion! On the window front, Pella has some decent windows, however there are a lot of issues floating around with them. Most seem to stem from install related, or at least that is what Pella blames it on. They do have some good glass packages, however Marvin's Integrity has caught up with low u values, high solar gain option, and finally added a triple pane glass package to their fiberglass line. Their wood ultrex line also trumps Pella's values barely. My opinion is fiberglass trumps roll formed thin aluminum on the Pellas (to be fair, my current home has Pellas in it without issue). Other brands such as Marvin and Kolbe use thick extruded aluminum on the exterior rather than roll formed. The new home starting construction in 2 weeks will have Integrity fiberglass, a combination of triple panes and high solar gain windows (their 180 glass package).

Posted By cards66 on 02 Sep 2012 04:21 PM
Well I got another bid on insulation today. This was for a system called Air Tite System. Here is a link to the brochure. http://www.starcompanieskc.com/ATS.pdf http://www.insulate123.com/pdfs/ATSbrochure.pdf Wanted to get opinions on this hybrid system and then put the EPS foam on the OSB? I was steering away from Fiberglass but this system looks like it might work.

That's called "flash'n'fill" or "flash'n'batt". It air-seals better than batts and puts a non-wicking condensing surface between the fiber insulation and the sheathing,m protecting the sheathing a bit better from wintertime moisure drives.   It's whole-wall thermal performance is only ~R1 over a dense-packed cellulose solution, and only provided the batts are installed PERFECTLY (which in practice only rarely comes close, even with some attention to detail).

It is nowhere NEAR what insulating sheathing on the exterior and detailing the structural sheathing as a primary air barrier buys you.