Closed cell insulation & fiberglass
![](https://secure.gravatar.com/avatar/9e2a55664654414d06633378b996dd4b?s=52&d=https%3A%2F%2Fimages.finehomebuilding.com%2Fapp%2Favatars%2F200x200%2Fp.png&r=g)
We are building a new house and have had quotes for three types of insulation. first one is icynene (open cell) using 9 1/2 inches for roof rafters and 5 1/2 for the walls. Second one is 5″ closed cell in rafters and 3″ in walls. Third one is 2″ closed cell plus R-19 fiberglass in rafters and 2″ closed cell plus R-13 fiberglass in walls. The price differential between option 3 and option 1 is $9k on a 4500 sq. ft. house. Does anyone have any advice about using fiberglass and closed cell together – I have a concern about possible moisture issues?
Replies
Issues with moisture and answers depend entirely on your location and dominant weather. Fill out your profile if you want advice that will mean something to your particular situation.
The 2" of closed cell will provide about R12 of insulation (aged R value 6/inch). It also will constitute a vapor retarder.
If you are in a humid, cooling-dominated climate, the foam-FG combination is fine, as the vapor retarder is on the outside. However, you wouldn't want a vapor barrier, such as polyethylene film or vinyl wallpaper, on the inside.
If you are in a heating-dominated climate, you have to address movement of water vapor from inside into the FG-filled cavity and whether or not it can condense under some conditions.
For purely discussion purposes, allow another R1 for sheathing, and ignore minor contributions such as film coefficients and the drywall.
Adding R19 of a porous batt insulation inside of R12 will put the temperature at the interface between foam and FG at about 19/32 or 59% of the way from inside to outside temperature. If you have, say, 70 degrees inside and 0 outside, the interface will be at 70 -0.59*(70-0) or 29 degrees. Similarly, for the wall, the temperature-drop fraction would be about 13/26 or 50%, and the interface temperature for 70 - 0 would be 35 degrees.
Since comfortable inside air will have a dew point of mid 40s, you could have a possible condensing situation. It would depend on how well you have kept inside air from entering the cavity.
If your cavity is "open" to the inside air with respect to convective transport of water vapor, through leaks around can lights, openings at wall outlets and under the drywall where wall meets floor, then you have a definite problem for either ceiling or wall for the temperatures described above. For a milder climate you may be fine.
If you tighten up the drywall boundary inside and use either VR latex paint or something like Certainteed's MemBrain vapor retarder film, then you may be able to make the system work. Keeping water vapor from entering the cavity by convective flow limits its entry to diffusion, a relatively slow process.
My own thoughts on this are, for a cold climate:
1. Go all closed cell foam, to the thickness for required total R.
2. Increase thickness of the CC foam so that the temperature at the foam-FG interface won't get below the dew point for any substantial length of time.
3. Use open-cell foam if you are using a porous insulation inside, with a vapor retarder inside and good detailing of the drywall boundary to eliminate convective flow. I'd prefer dense-packed cellulose over FG, because installation of the latter is difficult to detail well. The cellulose fills the cavity far better.
I'll defer to others on the issue of CC vs. OC foam under a roof deck.
Edited 10/19/2008 11:23 am ET by DickRussell
I agree, no way to answer this question with out knowing the climate he's building in.------------------
"You cannot work hard enough to make up for a sloppy estimate."
When there is both fiberglass and foam you are much better off than if it was just the fiberglass and I'm 98% sure that holds true for all climates.
If you are in a flood prone area I hear some builders using 100% foam so if something big and wet happens the insulation doesn't go to pot.
Beer was created so carpenters wouldn't rule the world.