I am working on a new home in Long Island NY. I want opinions on the benefits of insulating the attic rafters and creating attic that is inside the building envelope verses insulating the floor and having the attic hot/cold. The air conditioning air handler is in the attic along with all the ductwork. My thoughts are as follows. If the rafters are insulated well (i’m thinking of using a spray in place foam), and the gable is insulated, then the heat of the summer sun will be insulated from the inside of the house letting the AC work less. this is the way it would be if the ceilings were catherial style. Additionally what are opinions on the ventialation of the attic. all roofing mfgrs show a roof with soffit vents admitting air and a ridge vent to let it excape. we all grew up believing this is the “correct” way. But is it? I see published articles recommending that the rafter bays be insulated without benefit of ventilation “chutes”. They cite studies where identical homes were built and then years later opened for evaluation and found no determential effects where seen when the roof did not have any ventialtion. It seems to make sense that having a flow of air on the underside of the roof would help keep it cool. But what about homes in the sun belt. their attics are really, really hot even if they ventilate. Am I missing something?
Discussion Forum
Discussion Forum
Up Next
Video Shorts
Featured Story
The longer length of these 3-in. stair gauges makes it easier to ensure an accurate layout when working with waned lumber.
Featured Video
Highlights
"I have learned so much thanks to the searchable articles on the FHB website. I can confidently say that I expect to be a life-long subscriber." - M.K.
Fine Homebuilding Magazine
Fine Home Building
Follow
-
Fine Homebuilding
Dig into cutting-edge approaches and decades of proven solutions with total access to our experts and tradespeople.
Start Free Trial Now -
GBA Prime
Get instant access to the latest developments in green building, research, and reports from the field.
Start Free Trial Now
Replies
You have done your research!
The approach you laid out is the best route. Do not comprimise and try to invent your own system. This has all been thought out by building scientists and proven by the methods you mention.
The biggest arguments I hear are these;
Q: Won't the roof get too hot and the shingles explode?
A: No. Roofing manufactures do warrenty this installation method. And like you mention, compare NY to texas or FL roof and see whats hotter!
Q: What if it gets wet?
A: If can dry and still perform unlike FG and cellulose. --You are using foam right?! Don't even consider anything else in that environment because I don't believe anything compares.
You are not missing anything. Just make sure you have proper mechanical ventilation in place to ensure a healty amount of fresh air is brought into the house. The only way to achieve real energy conservation is to control the air leakage and the foam in the attic is a great place to start.
thanks for your input, it sounds like you have experience with this. can you tell me more?
Get input from 2 people. The building inspector is first. Before you go see him print out the recomendation method described on either the Corbond or Icynene web site.
Then go see your HVAC guy. An educated HVAC person will recommend perhaps a 30-40% reduction in capacity when the unit and all the ducting is in the conditioned space. This is a huge up front savings. Good luck. Yes I do install spray foam BTW. These are the unbiased facts though. Check them out for yourself. Stu
Could one foam over baffles, even the lightweight baffles available in stores? Then one could bring the attic into the envelope without violating code or voiding roof warranty. If you foam the rafters, do you then have to hang DW on the rafters as a thermal barrier? Are there lighter weight alternatives?If the attic is now in the conditioned space, I assume some kind of attic ventilation is now in order to remove moisture accumulation.There is the added risk of unnoticed roof leaks....And Imerc for one had to spray with open cell foam since, in his climate (NC), roof shingles had to be driable to the inside. Not an issue if you can keep the roof venting.
Yes you could foam over baffles. This is not the preferred method but could be done. Usually a thermal barrier is not reqd where the access is only to service equipment like an attic. There is thermal barrier paint available as an alternative.
Most/many foams may require drywall in contact w/ the foam for fire protection (or some equivalent). Contact your local BO to find out what would be required. This may be required regardless of the use/occupancy of the space as implied by the other reply. My experience is that e.g. polystyrene by covered by drywall in direct contact w/ the foam regardless of other conditions ... even if in a concealed space. ... But again ... talk w/ the BO ... don't take my ... or any one else's word for it (except maybe an architect or fire protection engineer).
Foam is expensive, but yes ... a possible alternative.
Roof leaks? Difficult no matter what. They always show up in areas that they aren't (i.e. leak occurs here ... but shows up over there on the inside). The foam can make this worse, but good roof maintenance should always be there and preclude the need to worry about it.
When houses are built with foam core stress skin pannels there is no air space. And when the same pannels are used for roofs the shingles are applied right to the otter skin. Also the problem is not as much heat but heat and moisture. With fiberglass moisture can get in, with foam if it is done right moisture can not get in at all. But it is up to the inspector or you covincing him/her otherwise with evidence ( the foam supplier should be able to help you with this). They did this not to long ago on This Old House.
Also putting the HVAC in the conditioned space is part of the Engery Wise House building program.
Wally O
"But is it? I see published articles recommending that the rafter bays be insulated without benefit of ventilation "chutes". They cite studies where identical homes were built and then years later opened for evaluation and found no determential effects where seen when the roof did not have any ventialtion. It seems to make sense that having a flow of air on the underside of the roof would help keep it cool. But what about homes in the sun belt. their attics are really, really hot even if they ventilate. Am I missing something? "
It doesn't really matter what's best. NY code requires that you ventilate. It's much easier to insulate the floor, but I think it's marginally better for your AC system to do the rafters. But you'll have to have soffit and ridge vents.
You are on the cutting edge. Sealed unvented attic spaces will become common when the codes and building practices catch up with science. Changes in building methods develop very slowly. Difficult to get folks to think differently about what they have learned and come to accept as gospel.
The New Buildings Institute did a reasearch project for the California Energy Commission's PIER research projects on just this topic ... to insulate ceilings in attics or keep the A/C units in the unconditioned attic. Try the New Building Institute's web site or the California Energy Commission web sites and look up PIER research or residential project on this topic. This research was completed just a year or so ago.
The biggest issue w/ non ventilated roofs, I understand is to not have penetrations that would allow moisture into the roof structure from inside the house (most typically through e.g. recessed lights, but through really any penetration (e.g. J-boxes). Penetrations must be carefully sealed. You may not be able to seal a recessed light fixture or many electrical items w/out voiding the U-L rating designed to protect you from fire assume the installation of the device is in accordance w/ the product as tested.
"The biggest issue w/ non ventilated roofs, I understand is to not have penetrations that would allow moisture into the roof structure from inside the house (most typically through e.g. recessed lights, but through really any penetration (e.g. J-boxes). Penetrations must be carefully sealed. You may not be able to seal a recessed light fixture or many electrical items w/out voiding the U-L rating designed to protect you from fire assume the installation of the device is in accordance w/ the product as tested."
Ah, but by sealing at the roof deck instead of the attic plane, air leaks through the attic plane become a non-issue.
I don't think that is correct. Air sealing at the plane of the roofing (i.e. exterior surface of the construction could be asking for trouble. Vapor pressure occurs regardless of air leakage. If you have fiberglass roof insulation and penetrations into the space, I think moisture has an opportunity to enter that cavity and potentially condense. Joe Lstibrek (sp) is the expert in this field (I think he's in the Chicago area??). He might be able to shed some more light on the subject.
"If you have fiberglass roof insulation and penetrations into the space, I think moisture has an opportunity to enter that cavity and potentially condense. Joe Lstibrek (sp) is the expert in this field (I think he's in the Chicago area??). He might be able to shed some more light on the subject."
Fiberglass would not work well for this. As I recall, the poster was taking about using foam. The idea here is that you are moving the weather barrier up to the roof deck, so the attic plane is just decorative. Think of it as a cathedral ceiling.
The vast majority of moisture movement in and out of the house occurs because of air leakage; not because of vapour diffusion. Check out the two papers I read yesterday on air leakage and moisture transport.http://www.civil.uwaterloo.ca/beg/Downloads/8thBSTC%20Air%20Flow%20Control.pdfhttp://www.civil.uwaterloo.ca/beg/Downloads/ASHRAE%20Journal%20Jan%202002%20Moisture.pdfThese particular papers discuss, among other things, the idea that a vapour barrier is often unnecessary, even in a cold climate. Of course, this assumes that air sealing is done well.The idea is to allow materials to dry inside during the summer.The author of the papers Dr. John Straube is at the University of Waterloo (Canada) and is an authority on the subject (as is Dr. Joseph Lstiburek, i got the spelling on that right by copying from the building science website).Currently I am trying to decide whether my cathedral ceiling on my 100 year old house is going to have roof vents or not. And this with cellulose blown in at a high density (3.5lbs per cu. foot). Owen Sound, Ontario, Canada has about 8500 heating degree days (Fahrenheit). It is an interesting design challenge.
"Currently I am trying to decide whether my cathedral ceiling on my 100 year old house is going to have roof vents or not. And this with cellulose blown in at a high density (3.5lbs per cu. foot). Owen Sound, Ontario, Canada has about 8500 heating degree days (Fahrenheit). It is an interesting design challenge."
I agree with your premise. My (very controversial) opinion is that venting a cathedral ceiling in a cold climate causes more problems than it solves. This is a terrible place for uncontrolled air movement. The big problem with acceptance of unvented ceiling cavities is that they require careful detailing to make sure that they are really really sealed tight. Poking them full of holes for pot lights, skylights, and such can be a recipe for disaster (unless done with great care). Then folks will come along after a failure and say; "see, you should have vented it".
If it were my own structure in a climate that cold, I would add a continuous layer of rigid foam below the densepack (behind the drywall). All seams would be foamed or foil taped. Creates a layer of added insurance, plus a thermal break with the framing. Cavities would be completely air sealed at the eves. I know, controversial...
My current plan involves building down from the 2x6 rafters (after removing 2000 sq. ft of plaster) with plywood gussets and putting 2x3s across up to make my ceiling plane. Then blowing in dense pack cellulose to about R60. The building down already takes care of most of my thermal breaks.I would have picked beneath (or preferably on top) the roof sheathing for the rigid foam. That is closer to what I am reading about unvented roof systems in cold climates.I already know I need to take exceptional care with my air sealing.Since you seem to have thought about this a bit, do you think the rigid foam is necessary with the quantity of cellulose I am planning?
"Since you seem to have thought about this a bit, do you think the rigid foam is necessary with the quantity of cellulose I am planning?"
With a plan like yours, probably not. If you have a kneewall space, the handling of that is a key detail.
I'm going to handle the kneewall space by bringing everthing that is inside the exterior wall into the building envelope. There will be some fairly tricky air sealing (probably polyurethane for a lot of that) to get the top plate and other details in that area.My ceiling insulation will come from the centre peaked are of the roof right down the slope to the bottom. After the insulation and air sealing is redone I will put storage in the kneewall area.For the sloped part I had intended to do one of two things.
1. With the nets they use to blow cellulose just create an area in that middle section which I can blow dense packed.2. Or, just leave it loose in the centre area but put in 25" - 30" of the stuff.I am leaning towards option 1.
Your plan is very well thought out. Bringing the kneewalls 'inside' is inherently better for many reasons.
Nets are good, though another method is to 'blow as you go'. Basically, install a couple of rows of drywall, densepack behind it, repeat. Does not work so well schedule-wise if you sub out the rocking though.
Are you talking about netting cellulose in your rafters? What material were you planning to use? I have experience doing this and with other methods of cathedrlalizing attic insulation. Where is your project?
Edited 5/20/2005 12:41 am ET by RayMoore
There is some kind of netting the manufacturer of the insulation mentions on their website. Have no idea what it looks like or how much it costs.I am going down a lot farther than my rafters. I want R60 which means I need about14" to 15"...I am in Owen Sound, Ontario, Canada. I estimate about 8600 heating degree days (in Fahrenheit).If my house wasn't a 100 years old... there would go a really fine hobby.
I would not use this method in your climate. Cellulose will settle regardless of installation density and at this thickness the settling will be substantial. In a climate as cold as yours, condensation at the roof deck with this method is highly probable. If air is able to reach this surface then the likelyhood increases tenfold. If you do use this method, then venting will be required.
Your better option is a spray in foam, or even better a foam panel above the decking. My disclaimer is that I am responding without having all the information, which is prone to misunderstanding. Can you give us more background? It is hard to give enough information in a forum such as this. Site visits are so much better.
Edited 5/20/2005 11:17 am ET by RayMoore
>> Difficult to get folks to think differently about what they have learned and come to accept as gospel. <<
.. and aren't we glad about that, least we all live in UFI insulated homes ...
My friend has ducts in her attic for both heat and ac and had this done at her house last fall. The main reason for doing it was heat from the ducts causing snow on the roof to melt, thus causing terrible ice dams. The foam made an incredible difference this winter, no ice dams, much warmer upstairs. I expect similar benefits for the cooling season. The only downside was that the installers were a bit careless with the hoses carrying the foam and damaged some of the fiberglass board ductwork.If you have metal ductwork this shouldn't be a concern.
The IRC does allow unvented roofs with foam.
Still, there are specifics, and you need to see if your area has any ammendments that override or have exceptions to that portion of the IRC.
It's an excellent idea to bring the attic space into the envelope of the building. It's the smart thing to do. Bravo for being way ahead of the game.
When I built my house, sealed attics were not allowed, but I did it anyway. Careful detailing and intelligent design are not for the masses, which is why the code often lags building science.
When the code finally does catch up to building science and real-world thermal performance of building envelopes, FG will be banned. Of course it won't be, but it should be.
It seems to make sense that having a flow of air on the underside of the roof would help keep it cool.
it would seem that way. But only if the air actually "carries" any of the heat away from the roof, which requires all sorts of assumptions.
But what about homes in the sun belt. their attics are really, really hot even if they ventilate.
I pull my remote thermometer out of the attic when it gets up to 110º or so. It peaked last week, before our last little cold front, about 98º. My attic is ventilated--call it 75% of maximum possible ventilation--here at 29ºN in central Texas. The roof mass recomes a radiator, so it's heating the air in the attic at a pretty good clip; probably faster than any number of air changes could cool it (even if the relative humidity were not so high as to "use up" most of the cooling potential in the air).
So. it's complicated. I'm trying to talk a buddy into letting me build a hybrid on his carport converted to garage. I'd like to use spray foam throughout, including the ceiling plane, and then ventilate the attic besides.
I'd like to see, empirically, if insulating the attic will help bring it closer to ambient, as opposed to being 20-30º over. I suspect that some ventilation will be required, as thermal lag would eventually heat a sealed insulated attic to above ambient.