Radiant Floor ** How do I insulate slab?
I am having hydronic radiant heating installed in both floors. I have questions on how to insulate and maximize the heat reflection back to the home:
– At the bottom floor, the PEX tubes will be placed in the 4″ concrete slab with foundation walls all around the 4 walls. I have seen discussions of 2″ foam on compacted gravel layer first before the pour, or thermal break around the slab perimeter against the foundation, and even reflective aluminum material. How do I REALLY install these (other than saying they are need)? Do they really work as insulators for radiant heating? If I combine foam and reflective aluminum sheets, would it increase efficiency or waste money?
– At the top floor, the PEX tubes will be stapled on to the plywood subfloor. Gypcrete would be poured over that. Should I place aluminum pad like TempShield on the subfloor before the PEX comes in? Does it work? What products do you find helpful for aluminum reflective padding?
Thx,
Tung.
Edited 7/16/2006 6:52 pm ET by Streamline
Replies
Keep away from the reflective products as they are way overrated by their promoters!!! They do next to nothing under a slab. In the upper floor joist cavities, install R12 fiberglass fully up against the subfloor .
Is this a slab-on-grade house or a full depth foundation?
Do you suggest something like stapling Owens Corning roll R12 pink fiberglass on the ceiling of the lower floor? The slab at first floor is sitting on in a deeper foundation. Thanks.
the PEX tubes will be placed in the 2" concrete slab Misprint?
I placed 1-1/2'' foam under our slab (continuous). Some advise only 4 ft in off the perimeter.
I also placed 3/4'' between the block foundation and slab. Important to separate the two so the heat or cold doesn't transfer through.
A great place for Information, Comraderie, and a sucker punch.
Remodeling Contractor just outside the Glass City.
Quittin' Time
i would put the 2" foam under the entire slab... but the minimum thickness i would pour would be 4" and i'd spec 4000 #Mike Smith Rhode Island : Design / Build / Repair / Restore
Calvin, should the 3/4" rigid foam insulation thermal break be leveled with the slab final grade when placed against the foundation? That means the insulation will be exposed to finish grade against the floor perimeter? Thanks.
PS it is a misprint on the 2" when it should be 4".
Edited 7/16/2006 7:07 pm ET by Streamline
I'm built into a hill. The "walkout-basement" is our first floor. I stepped the foundation down the hill. On the block A.F.F. I furred it out to accept drywall. So, the wall finish comes out over the 3/4 foamboard. It's your call on how to deal with that. I ran my foam high and chalked a line on it to pour to. If you cut the foam accurately, you could use the top as your screed line. What kind of finish are you having on the foundation above the slab? or does it not come above the finish grade of concrete floor?A great place for Information, Comraderie, and a sucker punch.
Remodeling Contractor just outside the Glass City.
Quittin' Time
I am with Calvin here.
2" foam over compacted gravel and 3/4 or 1" foam around the four perimeter foundation walls as a thermal break.
My foundation footings are 8" thick with 9' ICF walls on top of them. I am puting in 8" of compacted stone and the 2" of foam will install on top of it and but to the ICF walls. The foam will be on top of the footing and serve as a thermal break to isolate the slab from the footing. I will end up with a 8' 7 1/2" high finished basement height.
My slab will not be radiant heated, but is will serve as mass for thermal storage in the passive solar plan. Both of us are trying to accomplish the same thing. Keep as much of the heat where it is needed most, in the house, not the ground.
Dave
I think that what experienced said is right--my understanding is that reflective insulation must have an air space to work. Most radiant heat setups I've read about used rigid insulation (usually Styrofoam) under the slab.
Just a note- make sure the gypcrete is level on the 2nd floor before you let anyone leave the site. Ours isn't and is causing a major headache. If the installer and the GC had checked it, we wouldn't be unhappy customers.
http://www.buildingscience.com has some great illustrations of "buildings that work". search for "slab insulation" on their site and you'll get some great drawings showing the proper insulation details.
Reflective panels under concrete do a very poor job of insulating, and are a 100% total waste of money. 2" of rigid foam is the only way to go, as everyone else is saying, and they are all completely correct.
Anyone claiming reflective insulation under a slab is worth the money is wrong, period.
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
All this is very useful information. Thanks to all.
- At the top floor, the PEX tubes will be stapled on to the plywood subfloor. Gypcrete would be poured over that. Should I place aluminum pad like TempShield on the subfloor before the PEX comes in? Does it work?
Insulation between two conditioned spaces can be minimised or excluded.
You will get more efficiency out of your system if you use conductive fins on the PEX.
View Image
Minimized, yes.Excluded, no.You do want the heat to go upward. Radiant will conduct through whatever is touching it (subfloor, for example) and radiate in all directions (joist bay, ceiling below).This means your performance will be hurt as you are "losing" heat off the backside.. this is called back loss.Now, overall, this isn't a HUGE deal because you are "losing" heat to another floor. However, it does mean you need higher water temperatures to meet the same load on that upper floor. And I've seen plenty of 80 degree heated basements as well which, strangely, never turn on.You want at least an R10 between floors, typically. But not reflective, real insulation is what is needed.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Rob - Does it matter which insulation type I use? Styrofoam jammed from ceiling below between joists, or pink insulation rolls from the ceiling below?
as long as it's installed well and is REAL R-value, not "r-value equivalent", not reflective, then no, I'm not particularly concerned with the insulation type between heated floors. Personally I hate fiberglass and would do anything not to work with the stuff, but it would be fine. Batts, blown, rigid foam.. it's all good for that part of your application.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Now, overall, this isn't a HUGE deal because you are "losing" heat to another floor. However, it does mean you need higher water temperatures to meet the same load on that upper floor
Hmmn, thanks, I hadn't considered that. I was thinking in terms of the whole house envelope. And money. And thinking that what btu$ didn't go in one space (lower floor) would be spent in the upper. I can see where keeping temps lo would be beneficial to the system.SamT
Isn't that for an under subfloor installation?
If he is pouring gypcrete over pex stapled on top of the subfloor does he need a radiant reflector under the tubing?
Dave
does he need a radiant reflector under the tubing?
No.
But it will improve the transfer of heat from the PEX.
NRTRob can run the numbers, that's his job, to determine how much more efficient and if the ROI is worth it.
I can't. It's been waaaay too many years since I did Solar Heat and heat transfer math and for all I know all those books and manuals I once owned have turned to bookworm castings.SamT
You wouldn't use plates with gypcrete like this fellow is installing, the gyp is the transfer medium. If this were a "dry" install then plates nearly always make plenty of sense.If the other fellow is asking about a radiant barrier in the joists under the gyp, then I would skip it and use real insulation. Radiant barriers *may* work at first, but unless you can guarantee the upward facing reflective surface will stay clean, I personally choose not to trust it for long term usage.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Like I said, it's been over 20 years since I worked the numbers, and there is a lot of new technology I never saw then.
Have you actually ran the numbers to see what the difference was in heat transfer to surface of gypcrete with and without the aluminum plates?
Just curious what the theoretical difference is and what the co$t difference might be.
BTW, what is the U of gypcrete?SamT
It's a moot point, I believe the gypcrete would destroy the plates over time. and water temps are low enough with 'crete systems already, it most certainly wouldn't be worth adding aluminum even if it would survive. If you use aluminum, skip the 'crete (do a sandwich) and vice versa.Gypsum has a .25 k-value and concrete about a .50 k-value, so I would expect most gypcrete to be somewhere in between.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Gypsum has a .25 k-value and concrete about a .50 k-value, so I would expect most gypcrete to be somewhere in between
For the lost souls; R is resistance to heat flow; U = 1/R, the reciprical of R; K-value is the number of BTUs that will flow across a 1" thick by 1' square of stuff in 1 hour with a 1* difference in temp from side to side.
Normal concrete K = 9 -12Light weight concrete K = 2.5 - 3.6Gypcrete K = 2.45Drywall K = .546061 Aluminum K = 11605052 Aluminum K = 960Ref: 1, 2, 3.
Specific Heat, c. How many calories of heat to raise 1 gram 1*Celsius. Or Kilo Joules /Kilo Grams
Aluminum = 0.89 KJ/KgConcrete = 3.3 KJ/KgGypcrete = 0.94 KJ/KgRef: 1, 2,
Density lbs/1'x1'x1"
Aluminum = 14#concrete = 12.9#Gypcrete = 9.6#Ref: 1, 2,
So far, just an exercise in my ongoing education.
Just looking at Orders of magnitude, it appears that 0.1" of aluminum will conduct heat latteraly away from Pex about 40 times faster than 1" of gypcrete, (10 x concrete.) Of course, it would then have to travel upwards thru the gypcrete to enter the living space. It's going to take some differential equations to accurately know what's going on.
Rob, if those internet numbers are right, in that concrete has a k-value 3-4 times that of gypcrete and only weighs 3lbs more per sqft, why not use concrete over PEX? What's the big price difference in 59.6lb/ft load and 62.9lb/ft load.SamT
Well, gypcrete is much easier to install, for one; self-levelling and all that. I *believe* it is also supposed to stay together better at thin thicknesses such as this, but I can't confirm that with real knowledge. Lots of gyp out there is "specifically formulated" for radiant as well; again, whether it makes a difference or not, I'm not sure, I haven't investigated them directly to see how they stack up.For two, I'm not sure how weight stacks up. I know when concrete is used on suspended floors, people generally try to use "lightweight concrete" which is more like gypcrete than regular concrete in performance I believe.. so I guess I have no idea if your weight differential is correct, or whether such an apparently small weight differential is significant, or whether it has little to do with the decision ;) It's a good questions though for sure. remember the pour is 1.5" thickness minimum, so you're talking about a roughly 5lb difference per square foot.. if that's big or not, I'm not sure.
-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
It's time for me to get back out in the heat (any radient cooling around?) so I'm not gonna do any more research right now, but. . .
That takes the floor load from 50# normal to 65# with gypcrete and 70# with concrete. Not much difference between the two. 6.5%?
I'm really interested in Radient math, 'cuz I see Radient really taking off here, and I wanna get me some o' that money. With a background in solar engineering, it's a natural fer me.SamT
definitely, sounds like you'd do quite well at it. cooling is out there but not yet common, partly because it's fairly limited and you still usually need to address dehumidification, so it's not a stand alone solution.However if you wanna rock with heating and cooling, start looking into ceilings ;)-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
Rob,
What's the comfort perception on ceiling vs floor vs wall radient.
Also is there a difference in perception between top half of wall and bottom half?SamT
Very close, but you do lose the "warm foot" feeling in buildings with higher loads where that would be noticeable. However the lower the load, the cooler the floor, the more it makes sense to use ceiling as well. And of course we're all investing in the envelope first, right? :DI avoid wall typically; too much puncture risk for my taste. Too much furniture blockage too.You do have to watch proximity to the head and surface temps though for ceilings.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
SamT
NRTRob,
Have you ever used fiberglass insulation with a reflective covering in the floor joists between floors?
when I installed my floor the local radiant guys recommended using this material with an air gap between the tubes and the upper subfloor.
I have had great results with my system, but I am in a tempurate climate near Seattle.
JasonIf it wasn't for bad luck I wouldn't have Any!
I would regard the reflective covering as unnecessary, but I don't think it hurts you nor does it really affect the price very much, so no harm no foul.Though the 2" airspace is generally only for "suspended tube" systems... no plates.. in joist bays. If you have plates, you don't need the airspace. 2" won't hurt anything either way though. In a temperature climate, suspended tube can be just fine as well, even though it's not a method I generally recommend.-------------------------------------
-=Northeast Radiant Technology=-
Radiant Design, Consultation, Parts Supply
http://www.NRTradiant.com
On my project, I used 1" foil faced polyiso board glued to the CMU walls. I ran that down to the top of the footing. The polyiso board has a much better R-value per inch, so it seemed like a better choice for the walls.
Under the slab, I used 2 layers of 1" blue foam board. Coulda' used 1 layer of 2", but my supplier didn't have it in stock, so......
Anyway, all of this was done on the recommendation of my heating sub, so I guess he knows what he's talking about. We'll see come winter!
Hope this helps.