Replies

1. GBA Editor
   Martin Holladay | Dec 11, 2009 04:43pm | #1

   Mariela,
   Either Icynene or dense-packed cellulose will work in this application, without ventilation, as long as the rafter bays are thick enough to provide room for code-minimum (or thicker) insulation. Don't be talked into skimping on R-value.

   If you use cellulose, pay particular attention to your ceiling air barrier. You want it to be impeccable.

2. 
   Robert Riversong | Dec 11, 2009 05:17pm | #2

   Mariela,

   You don't indicate where you are - what climate zone you're in.

   Neither cellulose nor Icynene are "moisture barriers", since they both have high vapor permeability. Properly-installed spray foams are better at being an air barrier, though dense-pack cellulose is more restrictive to air movement than any other fibrous insulation. In either case, you need an uninterrupted air barrier at the ceiling plane.

   While there is much debate about the merits of roof ventilation, the consensus of experts still leans in their favor for cold climates. See my posts in https://www.greenbuildingadvisor.com/community/forum/gba-pro-help/15535/when-continuous-self-adhered-membrane-required-roof.

3. 
   Robert Riversong | Dec 11, 2009 11:24pm | #3

   This is mostly for Martin (and anyone else who advocates unvented cathedral ceilings).

   I just came across a recently published article in the ASTM Journal, "Compact Asphalt Shingle Roof Systems: Should They be Vented?". http://www.astm.org/DIGITAL_LIBRARY/JOURNALS/JAI/PAGES/JAI102057.htm

   The abstract concludes:

   Our study showed better durability of vented roof assemblies with permeable insulation in cold climates due to redundancies that can tolerate incidental moisture and provide visual indicators of roof leakage; roof sheathing typically dries in 1-1/2 to 2 months. All of the unvented roof assemblies are intolerant of incidental water leakage and the moisture-sensitive layers (i.e., sheathing and gypsum wallboard (for open-cell polyurethane insulation)) exceed the threshold for decay. In hot, humid climates, the most durable roof assemblies are the vented, open-cell polyurethane systems with shorter drying time of the interior gypsum wallboard when compared to the unvented roof assembly; both the sheathing and gypsum wallboard dry out within 2-1/2 months. In an unvented assembly, the drying time for sheathing is similar but the drying time for gypsum wallboard increases to 6.5 months on average. Alternatively, unvented permeable shingled roofs are a viable option in hot, humid climates, although they are slightly less durable. The least tolerant roof assemblies in either climate are the unvented closed-cell polyurethane roof assembly due to trapped moisture and slow drying of the roof sheathing (up to 12 months in Miami, FL and 27 months in Boston, MA).

4. 
   Derek Vander Hoop | Dec 14, 2009 03:34pm | #4

   Mariela,
   I’ve been re-reading this post for days, trying to get my arms around what is being said by Martin, Robert and the authors of the study that Robert cites.
   Regarding your contractor’s belief that dense pack cellulose is a moisture barrier, he is incorrect. It does, however, restrict air movement into the assembly due to its density and its ability to fill all the nooks and crannies. As Martin and Robert state, it is important to have an air barrier at the ceiling to minimize stack effect, and in a cold climate, you would also want a vapor barrier at the bottom of the rafters. You probably already know that recessed can lights are a no-no in cathedral ceilings.
   As a homeowner, if my roof does develop a leak, I will want to know about it right away so that I can get it repaired or replaced in a timely manner. As a builder, I need to think of not only how to build the roof now, I need to consider what happens to the roof assembly when it does leak at some date in the future.
   If I am building a cathedral roof, I would be reluctant to use cellulose because when it gets wet, I would guess it takes a fair amount of time to dry out, even in a vented roof. In an unvented roof, with Visquine at the bottom and the sheathing at the top, it would take even longer. Secondly, cellulose can settle when it gets saturated., how much depends on how dense it was to begin with. How does that affect R-Value? In a pitched cathedral roof, does the cellulose not only settle, but slide just a little bit down along the visquine vapor barrier below (in a cold climate), causing a cold spot in the assembly? If so, how do you repair it short of opening up the roof to see just how much or if the insulation is compromised?
   I am an advocate for closed cell spray insulation for almost every assembly in a home, except for cathedral ceilings. Years ago, I did a test and hollowed out a small depression in a one square foot x 4” thick sample of Icynene (open cell) and also one of Corbond (closed cell) in my office and filled the depressions with water. In the next day or two, the water had penetrated the 4” of Icynene but the Corbond was impervious. If I acknowledge that all roofs leak eventually (and they all do), then if a roof insulated with Icynene leaks, eventually it should leak down to the Visquine below and rear its ugly head at the first screw or nail it encounters. When the homeowner sees this, twenty, thirty or forty years down the road, he calls in a roofer before any significant decay can occur. The advantage with Icynene is that it dries out without being compromised. It doesn’t settle, it doesn’t rot, it doesn’t move.
   But, in a cathedral ceiling sprayed with a closed cell foam insulation, I can envision a leak laying hidden for months, perhaps years, slowly rotting the sheathing and rafters without the homeowner having a clue that it is occurring.
   Regarding ventilation, everything I have read indicates that ventilating a cathedral roof is the way to go. It cuts down on the slight amount of heat conducted to the conditioned space below, dries out/ventilates the underside of the sheathing above and the fiberglas, cellulose or Icynene below, and lengthens the service life of the shingles.
   Concerning the study that Robert cites, I was surprised by how long it took for sheathing (2-1/2 months) and drywall (6-1/2 months) to dry out in an unvented assembly insulated with open cell spray-in foam, and how long it took for sheathing to dry in an unvented closed cell spray-in foam assembly. 27 months in Boston?!!!! In the middle of a Wisconsin summer, with the sun beating down, it is hard to believe that it can take that long. But that is why we do studies, to learn these things, (assuming the test conclusions are valid☺)

5. 
   Mariela | Dec 14, 2009 03:46pm | #5

   Thanks everybody for the very helpful answers. We have decided to go with Icynene. We live in Boston and therefore the study cited is very relevant to us!
   Thanks again,
   Mariela

6. 
   Robert Riversong | Dec 14, 2009 05:32pm | #6

   Mariela, I hope you also decided to ventilate the roof and air-seal it very well on the interior.

   Derek, the flaw in the examples your propose is the interior polyethylene vapor barrier. Almost all manufacturers of cellulose insulation strongly advise against using a VB, and some will not warrantee the product if a VB is installed. An air barrier is what's required. And, since any hygroscopic insulation (including open-cell foam) in a vented cathedral ceiling/roof should be separated from the vent channel, a self-draining material (such as breatheable housewrap) as a vent baffle will also drain any roof leakage down to the soffit vents and prevent excessive wetting of the insulation. Cellulose can safely absorb and release 30% of its weight in water without deformation or settling.

   As the ASTM study indicates, the appropriate VB to use with open-cell foam (Icynene) is a variable perm material that allows drying to the inside during the summer or in the event of a leak.

   There are almost no circumstances in which a polyethylene vapor barrier improves the performance of a wall or roof/ceiling assembly, if WRB and air barrier is properly detailed.

7. 
   Derek Vander Hoop | Dec 14, 2009 06:20pm | #7

   Hi Robert,

   The state of Wisconsin requires a vapor barrier, typically visquine, on inside wall surfaces for all insulations, except for closed cell foams, like Corbond. (Some building inspection departments will allow me to use a vapor barrier primer if I provide a receipt for quantity of primer purchased, a label stating the product is a vapor barrier, and an affidavit from the drywall contractor spraying the primer that he did use the product in question, but this doesn't address your concern that Icynene and cellulose shouldn't have a vapor barrier)

   I have used Tyvek, stapled, glued and caulked in retrofit applications to create a vent channel with some success, precisely to channel water from an existing shake roof installed without felt (you could see daylight when looking up at the 1x6 roof sheathing through the shakes!)

   I recognize that cellulose can safely hold a fair amount of moisture and safely release it, but are you not concerned that in some applications, water will condense of the inside of the exterior sheathing or if the moisture percentage is high enough, freeze at some point in the wall cavity? Sure, the cellulose gradually releases the moisture back into the building, but that doesn't do me any good if I have condensation worries during our winters? It makes sense to me to reduce vapor drive into the wall cavity, but I'll admit that a properly constructed wall will go a long way towards reducing the problem.

   For Mariela, living in Boston, their winters are not near as cold or severe as ours, so this is less of an issue for her.

8. GBA Editor
   Martin Holladay | Dec 15, 2009 08:20am | #8

   Derek,
   I'm sorry to hear that some building inspectors in Wisconsin are still misinterpreting the building code and forcing the inappropriate use of interior polyethylene. I'm glad that some inspectors understand that vapor retarding paint fulfills the requirement of the building code. I urge all builders who encounter such building inspectors to do their best to educate them so that the inappropriate use of polyethylene is gradually reduced.

   In a pinch, builders forced into a corner by stubborn and ill-educated building inspectors can resort to the installation of MemBrain -- an expensive solution, but a solution that is sometimes easier than beating your head against a brick wall.