insulating floor over exterior

I am planning an update to my house this summer, by adding radiant heating and insulation to a 2×10 floor assembly that is over a carport.
We have received an estimate for foam insulation $4,900 for approximately 1,200 s.f., 2 1/2″ thick. The salesman reluctantly gave me an option for 2 coats (5″ total)saying that the single coat would be more than sufficient.This seems pretty high, but we want to make our house as energy efficient as possible.
Any thoughts and suggestions are welcomed. I can send a PDF of the proposed assembly. Thanks, Frank
Replies
Where are you?
I would typically want to see an R30 between a heated floor and the outdoors.
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
We are just south of the Mason-Dixon line in MD.
I agree about the R-30.
The Incylthane foam salesman claims that the R-value of fiberglass declines to 50% when relative humidity rises to 50%, while the foam is unaffected.2 layers of foam (5" thick) should yield R-35 @ R-7 per inch. It will cost us $8,100 however for approximately 1200 S.F. Over plumber wants to suspend the new radiant heating on 2" hangers to create a plenum of warm air under the subfloor. He does not care for the metal plate systems I have seen. This leaves us 7 1/2" +/- in our 2x10 joist bays. I was thinking of using 1/2" aluminized R-Max under the heat tubes, fastened to 1x2's on each joist side. I think this will keep the foam from pushing up into the plenum. With the R-Max + 2 coats of foam, it looks like we would be up to about R-38.I'm a little concerned about creating too many vapor barriers in the assembly.Is this idea crazy?Thanks, Frank
Not using real plates is a risky proposition, and uses a good 20-35 degree hotter water than a good extruded plate system. If you're insulating very well and have modest glass/floor coverings, it can be ok, but it's generally not a very good installation method. Under a suspended tube you have the hottest possible joist cavity as well, so the R30 is a minimum and your R38 is a better plan.I'm not sure about the vapor barrier issue here though.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Many thanks,
I will look more closely into his proposal.
there are cases where suspended tube makes sense, I'll note; if this is the only room of radiant, and the rest of the house is baseboard, for instance, then this could allow you to skip mixing.But I would make sure that a load calc has been done for this at least. suspended tube is weak and tops out at loads that many homes see, especially if there is a lot of glass, carpeting, etc.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Thanks again Rob,We do have a lot of glass. Its a 53 year old International Style house with approx. 80% of the exterior glazed.
Its on the National Register.
http://www.marylandhistoricaltrust.net/nr/NRDetail.aspx?HDID=1468&COUNTY=Carroll&FROM=NRCountyList.aspx?COUNTY=CarrollWe have a boiler with 2- 90,000 BTU AquaTherm a/h's, and stubs to add additional zones. We think the radiant will be more efficient in milder weather, and more comfortable, especially in the baths.
Controlling the system may be a bit of a challenge, but we think we will be able to operate the radiant and the forced hot air at the same time.Sound crazy?
Frank
if it's only floor conditioning (not primary heat) then suspended tube may again make sense.You might use two stage thermostats to run the radiant until you need the FHA. Or you could use floor sensors to condition the floor and prevent the radiant from being "forced out" by the FHA system.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Rob,
I really appreciate all the advice.
Frank
knarf...
in your climate i would use staple up plates and back them with 1" EPS Performguard foamed in place with gun foam
then i would staple and glue Insulmesh to the bottom of the joists ( if it's a crawlspace ----- if it's a basement , i would stapel the Insulmesh and furr the bottom of the joists )
then i would blow dens-pak cells into the resulting bay...
so your sandwich would look like this from the top down :
finish flooring
subfloor
staple up radiant heat
1" EPS PerformGuard
7" of Dens-Pak cellulose
Insulmesh
... you would wind up with a true R-29.... R-30 at a very reasonable cost
Mike Smith Rhode Island : Design / Build / Repair / Restore
Thanks Mike,
2.5" of urethane foam is ~R20. Or since 1/R=U, U=.05. That means heat loss is 5 out of 100, or your 95% efficient. Double that and you get 97.5% efficient.
The first 2.5" will do 95% of the work then the rest will do 2.5%. No payback for the overkill amount. The salesman is an honest person, trust him.
All the codes and R-value nonsence is based on FG which does not work. Foam stays the same R no matter how hard the wind is blowing or the temperature differential. Search "R-value myth" and read from 10 different sources how FG does not work and the whole R-value system is bogus from the word go.
Stu
Thanks Stu,
I will
Just to be clear, you've got a lot of assumptions built into that statement.
2.5% of what volume of heat loss, for instance? The 95% vs 2.5% sure sounds out of whack.. but 2.5% of a completely uninsulated heat load can be a pretty significant number. 5% is a standard amount of heat loss for a home using your measure, and cutting that in half can have pretty significant effects.
I know foam isn't cheap, and maybe with foam and your experience there is no payback in most homes. Frankly, starting with windows is probably the best bet before going to an R40 wall (and doing both, even better). But there is more to think about than the percentage of heat loss you are blocking.. the quantity of heat loss, not the percentage, is what calculates payback, along with fuel cost. Percentage is meaningless unless you know the answer to the question, "percentage of what?"
I'm not trying to call you out... just concerned for the people reading this post in other circumstances that may let those numbers stick in their head without really understanding the implications. There are cases where R40 walls make sense. Maybe using a combination of insulations, if doing it all with foam is too expensive, for example.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
I understand your concerns. I'm just trying to keep it simple with the percentages. The units would be BTU/wall-area/unit-time or some other mostly not comprehensible term. My point is R value is not linear. The progression is geometric since the formula is R=1/U and vice-versa, U=1/R.Every window has U value on it. A decent window is .3 or so. That's R=3.3.So to keep everything in perspective: Window, 30% efficient, R20 wall 95%, R40 wall 97.5%.Stu
absolutely it is not linear. I just would not discourage anyone willing to pay more to superinsulate.. neither you, nor I, nor them, can guess at energy prices in the future. With a superinsulated shell, the need to guess can be almost eliminated. And there is a big difference between R20 and R40. Not payback-grade differences in today's energy market, but it's still a very significant difference, even though the percentage numbers makes it appear insignificant, the quantity calculations reveal that it's not.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
To answer the question: "percent of what," figure that a wall section at R20 (U= 0.05 BTU/hour/sq.ft/degree F), loses only 5% of the heat lost through a wall with R=1, which would be about an inch of wood (typically R 1.0 to 1.2 per inch, depending on species.
That 1200 sq.ft, with an inside-outside temperature difference of say 70-10 = 60 degrees on a cold night, and an R20 rating will give 1200*60/20 = 3600 BTU/hr, or 1.05 kilowatt. The 2.5" of closed cell foam will have an aged R value closer to 6.0 per inch than 8.0, for an R15 floor. That gives 33% more heat loss, or about 4800 BTU/hr (1.41 kwatt). Maybe that still is small potatoes compared to the huge size of the heating system, but still a significant loss if this were for a really energy efficient house. Now, if the R15 were doubled to R30, the savings in heat would be half the 4800, or 2400 BTU/hr (0.7 kw). If it costs another $4,900 to add the second layer of closed cell insulation, hmmmm, that would take a while to recoup. Considering what the rest of the house is, perhaps not worth the effort; there are lots bigger fish to catch in that place. As for the floor, I would think Mike's suggestion about the densepack cellulose would be more cost-effective. At $1.63/board foot for the CC foam, that is expensive insulation. Open-cell is less expensive per R value, and cellulose even more so. The space is there, so one of the advantages of CC foam isn't applicable (high R in limited space).
Edit: I forgot that business about the radiant heating. The temperature above the insulation will be much hotter than 70 degree room temperature, so the heat losses will be more than doubled what I showed above. Scale up by temperature difference, and allow for thermal bridging by those joists, an easy calc.
Edited 3/21/2007 11:56 am ET by DickRussell
excellent points dick. the joist cavity temperature is why joist insulation levels are so important with radiant floor. and part of the reason suspended tube systems are not good primary heat methods, especially over cold spaces.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Just curious, Rob, as to the reason a suspended tube would be used. All I can see is that it would eliminate the direct conduction between tube and floor and add yet another layer of insulation (air) between the heat and where it is intended to go.
You mean instead of a staple up direct to the subfloor, but still not using plates?Two reasons:stapling direct to subfloor promotes heat stripes on the surface of most normal floors, and the staples used chew through the pipe over time.Suspended tube systems perform very similarly in a 2" plenum by utilizing a convective current, so the air isn't really insulative because it's not dead air. and you're close enough to radiate some heat directly to the subfloor as well.That said, only in very low load systems or floor conditioning systems are plateless joist systems a decent idea, IMHO.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com