insulating cathedral ceiling in Houston - GreenBuildingTalk - GreenBuildingTalk - Green Building Forums on Insulating Concrete Forms (ICF), Structural Insulated Panels (SIP), Radiant Heating, Geothermal Heat Pumps, Solar Power, Green Construction Projects - Green Building Forums - General Forum

Forums

Unanswered Active Topics

Forums Search Members

Forums > Green Building Forums > General Forum - Residential

insulating cathedral ceiling in Houston

Last Post 19 May 2010 09:57 AM by Dana1. 9 Replies.

Sort:

	Prev Next
You are not authorized to post a reply.

Author

Messages

Tim-Bob

New Member

Posts:2

27 Apr 2010 12:50 AM
The sheetrock and insulation in my living room cathedral ceiling was removed after Ike. I have a new roof that includes a ridge vent along the peak of the cathedral ceiling. 3/4 of the rafters in this ceiling connect to adjacent attic space, which is above bedrooms on one side and the kitchen on the other side. The house has soffit vents, so in principle air could be drawn in through the soffit vents and pass above the cathedral ceiling and out the ridge vent. I'm planning to staple aluminum screed radiant barrier about an inch below the decking, to the sides of the rafters. Below that, I plan to install fibreglass insulation, with a gap of about an inch between it and the radiant barrier above. At the low ends of the rafters, I will make sure there is clear access from the adjacent attic space over the bedrooms to the top side of the radiant barrier, but not below, which I will seal with foam. This way, I hope to get the benefit of cooler air (cooler than the decking anyway) flowing from the attic space above the radiant barrier and out the ridge vent, and have stagnant air between the radiant barrier and the fibreglass. The rafters are only 7 inches deep, so I can't get a very high R-value for the insulation with the 1-inch gap on both the top and bottom sides of the radiant barrier. I've read that bubble pack, coated with aluminum on both sides, is a good radiant barrier, and the bubbles inside can give it an R-value of as much as 18. My thought is that this will give me an overall higher R-value for the cathedral ceiling. My concern with the bubble pack is that it is also considered a vapor barrier. Am I risking a moisture problem if I use the aluminized bubble pack as a radiant barrier in this application? Or other problems? Any comments and ideas on insulating cathedral ceilings in the hot, humid Houston environment are welcome! thanks, Tim-Bob (southeast Houston)

Eric Anderson User is Offline

Basic Member

Posts:441

27 Apr 2010 08:33 AM
I think I would skip the bubble foil. First I would install High density fiberglass bats R21 (for 2X6). This should be 5.5" deep. This gives you a 1.5" air channel. Next I would add continuous ridged insulation to the underside of the cathedral ceiling and tape all the seams carefully. You probably don't want a vapor barrier here so I would use 1-1.5" XPS. I would then strap it with 1 x4 perpendicular to the rafters using long screws into the rafters. I would then sheetrock and paint it. This is relatively cheap and fast I will admit I have zero experience building in this climate so my advice is somewhat suspect Cheers Eric

Think Energy CT, LLC Comprehensive Home Performance Energy Auditing

Dana1

Senior Member

Posts:6991

27 Apr 2010 05:56 PM
If the pitch of the roof is steeper than 2:12 and the roofing is something lighter than jet-black, a radiant barrier with 1" gaps on both sides isn't going to buy you much. In Houston a vapor retarder on the interior of the roof deck rejects high vapor drives of sun on a wet roof though, and the bubble pack might serve this function, but unless that soffit/ridge ventilation scheme is working VERY well you could end up with the water vapor condensing and running down the bubble-pack, dripping into the wall. (A high roof pitch reduces that likelihood since more air will move through there). Foil radiant barrier, same story. Giving up two full inches of insulation in a 7" deep cavity isn't worth it- you're knocking it from a potential R25+ to something less than R20- the cut in R-value would be equal to what was gained during peak full-sun cooling conditions, and the net effect during non-peak hours would be lowered. R18 is not a realistic estimate of the AVERAGE R-value equivalent for bubble-pack, only the peak, and then only under the orientation & conditions it was tested, which would usually be a worst-case summertime delta-T cooked up by the manufacturer's marketing engine for number-inflation purposes. The steady-state R-value under more nighttime or average conditions with less-intense roof deck temps is under R5, probably closer to R2. Radiant barriers only make sense in an open attic situation, and are not a good subsitute for conductive R-value. There's a REASON the FTC doesn't allow R-value labeling for reflective insulation, but that doesn't seem to stop the marketers from implying a high R. The relative-R value isn't constant, not even close, and is only buying you a lot when the roof decks are VERY hot. You have to read the fine print for the test procedure and understand the test conditions to know what it's real value is an any given application, but R-18 it is not (except when it's under a flat roof in direct sun, with a 150F+ roof deck.) A full cavity fill of open cell foam (sealing up the venting) and vapor retardent latex on the interior of the roof deck could work, and it would dry to the air-conditioned interior. Alternatively, vapor retardent latex sprayed on the interior of the roof deck & dense-packing the cavity with cellulose (also sealing off the venting would give you similar but slightly higher R, but also a decent amount of thermal mass to delay and lower the peak AC load, which would be more comfortable than a open-cell foam or fiberglass solution. If you use fiberglass, paint the deck first with vapor retardent latex, then a layer of radiant-barrier paint, since fiberglass is somewhat translucent to radiated heat, but see if you can't find permeabilty numbers for the radiant barrier paint- if it's less than 0.25perms it may be too vapor retardent for the roof deck to dry toward the interior, and you may have periods where the roof deck stays wet. The Building Science sample design for Houston shows an unvented roof deck cathedral ceiling with no radiant barrier, foam sealed at the soffits, and a strong vapor retarder above the roof deck to limit inward vapor drives from hot wet roofs from raising the humidity of the roof deck. Since ripping up the roofing to apply a poly or foil vapor retarder isn't an option, putting a weaker vapor retarder like vapor retardent latex on the interior give the roof deck some inward drying capacity without letting peak vapor drives to move a lot of water in a short amount of time. With cellulose fill, water that gets in is wicked and distributed, so as long as the interior side is reasonably dry, it'll dry toward the interior just fine. With fiberglass fill there is a small but real chance of vapor condensation under some circumstances. Open cell foam fill will pass water as vapor a bit more readily than cellulose, but vapor retardent paint on the interior of the roof deck is still a good idea to limit the potential for condensing & dripping. The ceiling finish needs to be vapor-permeable no matter which way you go- yours or mine. Most latex paints are sufficiently permeable to pass vapor quickly enough to keep the cavity dry. (Vapor retardent latex is special stuff, designed for that purpose.)

Dana1

Senior Member

Posts:6991

28 Apr 2010 01:45 PM
In re-thinking this a bit, vapor-retardent paint on the interior of the roof deck may be TOO vapor-impermeable with a risk of trapping moisture in the roof deck. But a standard latex primer would limit vapor-transfer of moisture into the insulation from the roof deck somewhat, but still have ample drying capacity (at ~2-3 perms standard latex is semi-permeable, cutting the vapor-transport of moisture to the interior through the roof deck by about half rather than by 90% with vapor retardent paint.) At a later date (when reroofing) an exterior vapor retarder would be even more effective.

Tim-Bob

New Member

Posts:2

30 Apr 2010 08:46 PM
Thanks for the suggestions. For the cathedral ceiling, I could close off the openings to the ridge vent pretty easily, and could also block any air flow from the adjacent attic spaces. What remains to decide with this approach is whether or not to include the radiant barrier. The 1-inch (or less) gaps above and below it should be pretty well stagnant air, drying to the inside. Humidity is also a concern in this 1950's house (smells a little musty at times), so I am now inclined to seal the attic space as much as possible (close in soffit vents, plug gaps in the brick veneer, foam visible holes and cracks, etc), trying to follow the contemporary practice of not ventilating the attic. The roof pretty much has a vapor barrier in that the whole roof is covered with Stormguard ice and water shield due to the low slope (it's just a little steeper than 2:12). In the attic spaces, there is nothing between the brick veneer and the interior space. It's drafty. The insulation is about 6 inches of rather old fiberglass. I am thinking of stapling Tyvek to the studs on these walls, and taping the seams to form a (permeable) vapor barrier, to try to slow down the movement of humidity into the attic space. Would putting two layers of Tyvek help keep the interior attic space dryer? On the underside of the roof in these areas, I am planning to install a foil radiant barrier, stapling it to the bottoms of the rafters. This radiant barrier has lots of pinholes in it, so vapor shouldn't get trapped between it and the decking (which has the Stormguard providing a vapor barrier on the exterior). Since the fiberglass insulation isn't in great shape, I'm thinking of either adding new fiberglass or blowing in cellulose. Further comments more than welcome! Thanks again for your advice.

Dana1

Senior Member

Posts:6991

03 May 2010 03:21 PM
Posted By Tim-Bob on 30 Apr 2010 08:46 PM Thanks for the suggestions. For the cathedral ceiling, I could close off the openings to the ridge vent pretty easily, and could also block any air flow from the adjacent attic spaces. What remains to decide with this approach is whether or not to include the radiant barrier. The 1-inch (or less) gaps above and below it should be pretty well stagnant air, drying to the inside. Humidity is also a concern in this 1950's house (smells a little musty at times), so I am now inclined to seal the attic space as much as possible (close in soffit vents, plug gaps in the brick veneer, foam visible holes and cracks, etc), trying to follow the contemporary practice of not ventilating the attic. The roof pretty much has a vapor barrier in that the whole roof is covered with Stormguard ice and water shield due to the low slope (it's just a little steeper than 2:12). In the attic spaces, there is nothing between the brick veneer and the interior space. It's drafty. The insulation is about 6 inches of rather old fiberglass. I am thinking of stapling Tyvek to the studs on these walls, and taping the seams to form a (permeable) vapor barrier, to try to slow down the movement of humidity into the attic space. Would putting two layers of Tyvek help keep the interior attic space dryer? On the underside of the roof in these areas, I am planning to install a foil radiant barrier, stapling it to the bottoms of the rafters. This radiant barrier has lots of pinholes in it, so vapor shouldn't get trapped between it and the decking (which has the Stormguard providing a vapor barrier on the exterior). Since the fiberglass insulation isn't in great shape, I'm thinking of either adding new fiberglass or blowing in cellulose. Further comments more than welcome! Thanks again for your advice. I think you can skip the radiant barrier. In order to be effective against radiant heat you need at least 1/2" air gap on one side (but preferably both sides. You'll get as-much or more effectiveness out of filling that gap with cellulose (preferabaly at 3lbs+ density for highest thermal mass and lowest air infiltration. Since it's inside the roof deck you don't want a vapor BARRIER on an unvented roof or moisture can build up in the roof deck. (Above the roof deck would be preferred in your climate zone.) A vapor retarder of between 1.0-2.0 perms allows the roof deck to try toward the interior without saturating it. Dense-packed (over 3lbs/ft^3) cellulose WILL protect the roof deck from moisture & rot in Houston (but not necessarily in all climate zones.) Read this document. Stopping the air-infiltration is the most important method of managing moisture, but the hygric buffering of cellulose counts too. Either use wet-sprayed cellulose or specify "borate only" fire redardents, since some sulfated fire retardents will corrode metals if the moisture level gets too high. The borates also protect the roof somewhat from ant/termite damage, since they're toxic to insects (but safe enough to be used in skin soaps & laundry detergents etc for humans.) If the rafters are 16" o.c. you can dry-blow dense-pack it behind 5/8" gypsum, but if they're wider you may have to go to heavier stuff. Wet-sprayed "stabilized" cellulose can be installed at any density and would be pretty much self-supporting once the adhesives set up. Even low-density cellulose allows only a fraction of the air movement of the highest-density fiberglass batting which will pretty much block any convective infiltration drive, and will likely be cheaper & perform than open cell foam in this situation.

Roberth

New Member

Posts:97

17 May 2010 10:03 PM
What about dumping the vents and going with a dense pack cellulose. Does the area have too much vapor drive from the outside for this? Applegate Insulation has information on thier website about doing unvented cathedral cieling with dense pack cellulose. Fiberglass has to much air space that allows radiant heat to pass through. So between convection and radiant transfer the cieling would have a very low effective R value. I dont know how a spray foam systems would work with an air gap between the inside of the roof deck and the insualtion. I would think that you would not be hitting the dew point so any water vapor could escape from the cavity. The more insualtion and air sealing the less the need there is for a radiant barrier. What I have read on the ORNL site is that convection negates much of the value of a radiant barrier when it is in contact with a surface on 1 side. Thus laying the barrier on the insulation will get almost no benefit and taking up space that could be used for insualtion.

Dana1

Senior Member

Posts:6991

18 May 2010 04:27 PM
Posted By Roberth on 17 May 2010 10:03 PM What about dumping the vents and going with a dense pack cellulose. Does the area have too much vapor drive from the outside for this? Applegate Insulation has information on thier website about doing unvented cathedral cieling with dense pack cellulose. Fiberglass has to much air space that allows radiant heat to pass through. So between convection and radiant transfer the cieling would have a very low effective R value. I dont know how a spray foam systems would work with an air gap between the inside of the roof deck and the insualtion. I would think that you would not be hitting the dew point so any water vapor could escape from the cavity. The more insualtion and air sealing the less the need there is for a radiant barrier. What I have read on the ORNL site is that convection negates much of the value of a radiant barrier when it is in contact with a surface on 1 side. Thus laying the barrier on the insulation will get almost no benefit and taking up space that could be used for insualtion. If you click on the link in my prior post and read the document you'd find that dense-pack cellulose works perfectly in Houston under all circumstances, even without vapor retarders, but can become an issue in colder climates. (Applegate & National Fiber's documentation notwithstanding, a roof is NOT just a tilted wall, and dense packing cellulose in a cathedralized ceiling can create deck rot conditions in some circumstances, but not in Houston.) SPF could work with the air gap, but the air gap will not always be sufficient to purge the moisture after high vapor drive conditions, and condensation at night is a real likelihood if the ventilation rate through the gap is limited. Without an air gap SPF would work OK, and no liquid water could ever dribble down-slope between the SPF & roof deck to create a wet spot. Your read of ORNL stuff in-re radiant barrier is dead-on. Wasting valuable insulation space for sufficient air gaps would be a net-negative here. (But "cool-roof" roofing materials wouldn't be.)

Roberth

New Member

Posts:97

18 May 2010 09:33 PM
Dana1 I have a ? With Closed SPF is there any problem with the deck not being able to dry to the inside in a climate like Houston's. It would seem like putting rigid foam on outside of the roof deck would go a long way to prevent vapor drive with added benefits of preventing thermal bridging and moving the condensing plane away from the wood.

Dana1

Senior Member

Posts:6991

19 May 2010 09:57 AM
Posted By Roberth on 18 May 2010 09:33 PM Dana1 I have a ? With Closed SPF is there any problem with the deck not being able to dry to the inside in a climate like Houston's. It would seem like putting rigid foam on outside of the roof deck would go a long way to prevent vapor drive with added benefits of preventing thermal bridging and moving the condensing plane away from the wood. It depends on how thick, and what type of SPF. Open cell foam is fairly permeable, closed cell is semi-impermeable at thicknesses of 2" or more. At R30 (5") closed cell SPF has a permeance of something like 0.3 perms, which still allows some drying toward the interior (by contrast, 6" poly b ~0.05 perms), enough to handle vapor drive issues, but intermittent roof leaks could take a very long time to dry out. The RR-1001 paper (linked in a prior post) from Building Science Corp outlining the results of WUFI modeling of different cathedralized roof insulation under various climates & interior humidity shows essentially zero hazard for Houston from vapor drives using solely closed cell SPF to code-minimum R-values. (See the tables on p.11 of that document.) Putting foam on the exterior will always be a safer bet, since it places all of the structural wood inside the thermal & hygric envelope. And as you rightly point out, exterior foam eliminates thermal bridging by the rafters, and will outperform interior SPF of equivalent R-value. Relying on cavity-fills alone will eventually go the way of the dinosoars in many climate zones, since it's easier to get a high clear-wall R (with thinner walls) using exterior insulation as a thermal breaks on studs & rafters. Even with code-compliant cavity fill already in place, adding another R12-R30 above the roof deck when it's necessary to re-roof can be a decent long-term investment. Panelized iso or eps goods with pre-applied nailer decks aren't cheap, but they're not outrageous either, and they not only save utility costs, they lower the risk of moisture damage to the house considerably.