InspectAPedia®   -   Search InspectApedia

Severe ice dams, eaves to ground in Poughkeepsie NYUn-vented Roof Cavities & Cathedral Ceilings
Ice Dam & Leak Prevention, Heating / Cooling Effects & Hot Roof Designs

Hot roof designs, aka dense-packed insulated sloped roofs:

What's the best way to design an un-vented or "hot roof" cathedral ceiling or sloped ceiling building? What goes wrong? How can we get the best building life, roof life, and least moisture or leak damage over the life of the structure.

This article series describes various solutions for un-vented cathedral ceilings and similar under-roof spaces, offering advice on how to avoid condensation, leaks, attic mold, & structural damage when roof venting is not possible.

We provide data & design suggestions for hot roof designs or for providing roof and ceiling ventilation, and we describe roof & roof insulation or venting system inspection methods and clues to detect roof venting deficiencies, insulation defects, and attic condensation, mold, or structural damage problems in buildings covered by an un-vented or hot roof or hard-to-vent roof design.

InspectAPedia tolerates no conflicts of interest. We have no relationship with advertisers, products, or services discussed at this website.

Un-Vented Roof Solutions: how to avoid condensation, leaks, damage when roof venting is not possible

Two approaches for insulating cathedral ceilings and flat roofs (C) Carson Dunlop Illustrated Home

[Click to enlarge any image]

Here we argue that while some experts like the "hot roof" design that omits attic or under-roof ventilation entirely, although is good science behind that approach, the science of un-vented roof designs may not adequately tack the challenges of construction mistakes or in-service leaks or damage that occur on building roofs.

There are serious risks of mistakes in construction, variations in building occupancy use and moisture levels, and wear and tear, any of which can lead to very costly surprise rot, mold, or insect damage on buildings where leaks and moisture are trapped in building cavities and remain un-noticed.

Sketch at above left provided courtesy of Carson Dunlop Associates shows the two basic strategies for insulating cathedral ceilings and flat roofs. Venting the roof or building a hot roof with no ventilation.

Worries about the "hot roof" Un-vented Cathedral Ceiling & Hot Roof Designs

As explained in our "hot roof" discussion

at ROOF VENTILATION SPECIFICATIONS

and at HOT ROOF DESIGN PROBLEMS, we don't have confidence in the long term durability of "hot roof designs" because any future roof leak into this cavity produces trapped moisture and rot.

We call this a "hot roof" design because failing to vent the roof from below not only misses a chance to avoid ice dam leaks and condensation damage in cold climates. In hot climates the roof temperature will be much higher on an un-vented roof, resulting in much shorter shingle life. This is less of a concern for slate and similar product roofs.

Also see INSULATION LOCATION for CATHEDRAL CEILINGS for more information.

In buildings where there is no roof venting anyway, an un-vented, well insulated "hot roof" is a second-best alternative to preventing ice dam related leaks in cold climates. Be sure to inspect the roof surface from outside for leaks and damage every year and to fix any damage quickly.

Un-vented cathedral ceiling under-roof spaces

Question: Some older buildings with no roof venting seem to be ok anyway - is there a concern?

Our house was built in 1920 it has never been vented. The second story is finished with the insulation directly on the roof sheeting. This year, after 30 years, we had the house re shingled the roof deck was in perfect condition no problems anywhere.

If we had the the choice I would lean toward venting but because of the way the home was built it's impossible but their has been no negative affect being unvented. - Jim

Reply: Some un-vented homes seem not to suffer, but not all: explanation and some warnings about un-vented cathedral ceilings on older homes

Jim,

I agree that many older homes were often drafty-enough that combined with the good luck of no unusual indoor moisture source (like a recurrent wet basement or crawl space) that they fared pretty well without more aggressive attic venting.

Building ventilation and moisture entry patterns change over the life of a structure

Leaks into cathedral ceiling insulation (C) InspectApedia.com Plowman But one needs to be careful in drawing conclusions from those examples. The way buildings are used, heated, ventilated, insulated, and sealed changes over time. I have inspected homes that were more than 100 years old that had been in good shape as far as moisture problems were concerned, until energy costs led new owners to change the way the house worked.

Water from above or moisture from below was trapped in this vaulted ceiling described by an InspectApedia.com reader - more details are

at HOT ROOF DESIGN vs LEAK RISKS

What was in 1900 a cold, drafty house, was still inexpensive to heat with coal or oil in 1935 (at today's prices), but beginning in the U.S. in the 1970's (the oil embargo and energy crisis) people no longer wanted stunning monthly heating bills.

The addition of layers of siding, storm windows, caulking, and perhaps blown-in insulation in walls and attic floors significantly changed how such houses worked in handling moisture.

Originally water ran into a basement through stone foundation walls, coursed across a dirt or stone floor to a hole where it exited, and moisture that passed into the building easily vented outdoors through many air leak points that created a high air exchange rate (ACH).

But after adding all those energy improvements, moisture that previously leaked right outside found itself trapped in some homes - water still leaked in but moisture had trouble leaving (like checking out of the Hotel California, you could check in but leaving was another matter).

One approach to that problem is to reduce excessive indoor moisture levels.

See MOISTURE CONTROL in BUILDINGS

Even in a 1920 home that has not had a noticeable moisture problem in its attic (sometimes we find a wet, rotted surprise in unvented cathedral ceilings), we might want to improve ventilation in order to keep the roof and attic cooler, extending the shingle life and reducing cooling costs for the home.

Finally, one decides to go ahead and ventilate such a home, as might be done during major renovations that happen to make it easier to provider an air path from soffit to ridge, adequate intake venting is critical lest the outlet venting simply suck heat out of the home in winter.

Reader comment: Replacing a wood shingle roof with asphalt over plywood changes how the building works

Lot of older homes vented naturally thru their wood shingle roof

When you tear off and replace with plywood and an asphalt shingle you are changing the design and could be asking for problems if not properly insulated and ventilated . Ansel 9/4/2011

Reply:

Good point, Ansel, and we agree. Your comment is an example of how changes to a structure over it's life also change how a building works and thus can cause new problems to arise, in this case higher moisture. We might add that a replacement wood shingle roof installed over solid plywood decking won't last as long as wood shingles over ventilated spaced nailers.

Question: We Can't Afford Spray Foam - Can we Just Cut Rigid Foam to Fit?

Spray foam roof © D Friedman at InspectApedia.com We have TJI rafters in a very weird roof area. no venting at all. I need to insulate and cannot afford the $750 minimum in our area for spray in foam. since the TJI has a flange top and bottom, what would be the best combo of rigid foam sizes?

The flanges are 1.25" high and 1" deep and the remaining rafter depth between them is 7". the distance between the flanges within the bay is 21.5". the rafters are 24" OC . Each bay is 5' long for a distance across the roof is 21' please help me calculate. Claire - 9/16/12

in the interim, I figured out how I will do it. I will fill the space between the flanges with 7" strips of 1" foam.

then I will cut and place 2" foam boards and 1" foam boards to fill the bays. tape and vapor barrier to block air

Our photo (above and below left) illustrate before and after spraying foam between the TJI rafters in our own project in 2011.

The "R" value of the roof with the cavity filled will exceed what you can conveniently jam between the TJI's in your own roof and more, it's an air-leak-proof ceiling in the "hot roof" design we discuss below.

To reduce our anxiety about future leaks into this roof structure (the foam and ceiling would hold water from a future roof leak until the roof rotted or the ceiling collapsed) we also opted for a durable standing-seam metal roof above. Photos courtesy Galow Homes.

Reply:

Spray foam roof © D Friedman at InspectApedia.com

Above: spraying open celled foam insulation under the roof of a New York Home.

Also see ROOF NOISE TRANSMISSION

Claire

There is no technical reason why you couldn't cut solid Hi-R foam into slabs and custom fit them between your TJI roof truss bottom chords. And it's true that a spray foam job may cost twice that of installing fiberglass batt insulation in the same space.

But having done some cut-and-fit foam board insulation jobs myself, I have some reservations about the DIY approach using cut sections of foam board insulation in the case you describe.

Not only is the cut-and-paste effort labor intensive, but I worry about a leaky design that makes the insulation job ineffective.

An advantage of sprayed foam under-roof insulation is that it seals perfectly against air leaks. Depending on the foam type used you'll still need a vapor barrier.

Insulating & Ventilating a Roof with Cut Sections of Solid Foam Board

Crawl space insulation with foam board © D Friedman at InspectApedia.com

Our photo (above left) illustrates a project using 2-inch high density styrofoam insulating board under the floor over a (dry) crawl space.

We used cleats to support the foam, not having the convenience of the I-joists that you have in your ceiling.

Foam board insulation job © D Friedman at InspectApedia.com

You can see that even working with some care, it was difficult to keep the foam slabs cut for a perfect fit (green arrow).

A second problem with retrofit foam insulation using cut-foam boards is shown by our photograph of a feeble attempt to insulate around the waste pipe penetrating the floor. (Next photo, below left).

A combination of cutting solid foam board insulating slabs followed by judicious use of a few cans of spray insulating foam can improve the performance of a do-it-yourself cut and paste insulation job.

You may need to use fire-block foam at building floor, wall, or ceiling penetrations, depending on your local building code requirements however.

The R-values of sprayed foam depend on the foam type and weight-rating.

Two common versions are Icynene® half-pound open celled foam and CertainTeed CertaSpray® two pound closed cell foam. Open-celled Icynene LD-C-50 type foam has an R-value of about 3.6/inch.

Question: what's the best way to seal joints of kraft-faced polyiso board insulation in an un-vented roof?

I am planning to insulate an unvented cathedral ceiling in southern Vermont. The ceiling/roof is new construction with 2x10 rafters spaced 12 inches on center. My plan is to cut polyiso foam board to the appropriate width and apply it in the joist bays

I can get kraft paper faced polyiso board 2nds for significantly less cost than retail foil faced sheets.

My first question is what is the difference between the two types of facing, what properties does each have and what is the intended use of each. I have read that kraftpaper faced polyiso board is a class II vapor retarder whereas foil faced polyiso board is a class I vapor barrier.

Because the insulation is going in a roof system in a cold climate where condensation is a leading concern I am wondering if kraft paper faced insulation will provide an adequate vapor barrier? If not, what can I do to create an adequate vapor barrier while using kraft paper faced polyiso board?

I am going to be using 2 layers of 4” thick polyiso board. How would you suggest the layers be bonded together and to the underside of the roof deck?

What is the best way to seal the joints of kraft faced polyiso board? - Anonymous by private email, 206/07/29

Reply:

Cathedral ceiling insulated with solid foam and a poly vapor barrier (C) Daniel FriedmanModerator comments:

IMO I'd be tempted to add a poly vapor barrier on the warm side of the ceiling rafters rather than trying to seal all of the foam edges against the rafters, but I'm a little concerned about accumulated water in areas between multiple vapor barriers in any construction.

For the labor involved, unless it's a low-budget free-labor DIY job I'd be tempted to go to sprayed closed cell foam instead, or to use the furring strip & baffle design we show

at CATHEDRAL CEILING INSULATION to provide an air space above the foam.

In all cases I worry about a hot roof cathedral ceiling under anything but the most bullet-proof roof covering outside.

In the photo I'm standing below a cathedral ceiling that I constructed about 35 years ago, using an approach similar to the one you are planning.

We referred this question to our partner Steve Bliss, over at BuildingAdvisor.com. Steve was traveling so had to confine himself to off-the-cuff reply that follows:

The short answer is that the use of vapor barriers with spray foam in unvented cathedral ceilings is complicated and controversial, so there is no definitive answer.

The general consensus is that open-cell spray foam is not a good choice here in cold climates. Closed-cell foam is preferred, especially in cold climates, as it is considered a Class II vapor retarder an excellent air seal.

The air sealing is probably more important than the permeance, since most moisture transport is by air leaks rather than diffusion.

Using foam board, rather than spray foam, does not provide the air seal of spray foam, so this is a concern. Sealing the board edges with canned foam would probably work but is labor intensive. I’ve done this with a commercial hand-help spray gun and it’s not too bad (if you work cheap).

Foil-faced vs. Kraft paper is a tough one. The foam used in these boards is pretty vapor permeable, so it’s really the facing that controls the permeance.

Kraft paper has a permeance around of about 1.0, just low enough to be considered a Class II vapor retarder. It would probably be fine as long as moisture levels in the cathedral ceiling are not too high.

I agree with you [DJF] that using multiple layers of vapor barriers could be a bit worrisome, but if some condensation occurred within the layers, it probably wouldn’t hurt anything. And, most likely, these surfaces would be warm enough that condensation would not occur.

If you are going to the trouble of cutting and placing foam board between the rafters, why not put in spacers and leave a 1 to 2-inch ventilation channel above. Even without eaves and ridge venting, there is some evidence that an air space can help prevent problems.

Bottom line: Foam board is probably OK. Should be air-sealed in place, especially the bottom layer facing the house. If it’s not air-sealed to the framing, you would need an airtight assembly below, not an easy thing to achieve.

A carefully installed high-tech vapor retarder such as Membrane would be a good choice as it allows for some drying to the interior, but is pricey.

This is really only a best guess as I’m not aware of any good data or case studies of this detail.

Moderator (DF) continues:

Home before second floor cathedral ceililng addition, 3 Willowbrook (C) Daniel Friedman

Above is a cathedral ceiling addition I constructed for a New York home in about 1980.

The first photo (above) shows the low-slope (leaky) shed roof over which we decided to add a cathedral-ceilinged second floor that you can see in the second photo, giving view from the rear of the home.

Cathedral ceiling construction ca 1980 (C) Daniel Friedman

Seeking a very high-R ceiling/roof I combined a baffle design that gave an air space, solid foam insulation, fiberglass insulation to fill the remaining space between rafters, and a vapor barrier on the warm side of the ceiling.

Second floor addition 30 years later 3 Willowbrook Hgts. (C) Daniel Friedman

Above is the same structure seen from the outside about 30 years later.

Below is a view of the interior 25 years later - and with a history of no leaks and no moisture or condensation issues. (Sorry I don't have my photos of the ceiling construction details, but I do have a sketch of how it was done.

Cathedral ceiling insulated with solid foam and a poly vapor barrier (C) Daniel Friedman

The sketch below is discussed in more detail

at CATHEDRAL CEILING INSULATION.

Cathedral ceiling insulation detail (C) S Bliss D FriedmanThe baffle and combined insulation and vapor barrier design shown here works well with soffit and ridge venting but might as Steve Bliss notes, be used even when there is no roof ventilation planned.

I found that the labor of hand-sealing the edges of foam shoved between rafters in a cathedral ceiling was not only too labor intensive but rather imperfect.

More than 45 years ago I constructed a cathedral ceiling insulated with foam cut to fit between rafters but with an air gap between the foam and the underside of the roof deck, as shown in my sketch at the start of this Q&A.

Because I knew the edges of the foam could leak air and moisture, I also included a 6-mil poly vapor barrier on the warm side of the rafters.

You can see this ceiling from the interior side (and me in the corner) a few years ago when the property was finally to be sold.

With more than 35 years of watching and living with this roof and ceiling I have to say it has performed perfectly, with no moisture accumulation, no leaks, and no mold issues.

But had a tree limb or storm damaged the shingles on the roof surface, a leak might have gone undiscovered for long enough to cause some ugly damage. Happily that didn't happen.

An alternative design that I used on other hot roof projects that were built before spraying foam had come to Dutchess County in New York was to add a final layer of 1" foil-faced insulation on the warm side of the rafters.

I taped all of those joints with foil tape, then installed drywall to cover the ceiling. The result was both a higher R-value ceiling and a ceiling with a perm rating of about zero.

Steve Bliss Adds

RE: the air space, with reflective insulation, the direction of heat flow greatly affects the effective R-value of the air space. The greatest gain is with heat flow downward facing an approximately 1” air space; the least is for heat flow upwards – about R-2 vs. R-9.

These numbers are published in REFLECTIVE INSULATION R-VALUES [PDF] from ASHRAE Fundamentals.

This makes it great for floor applications.

The use of foil on either the upper or lower face of the cathedral ceiling insulation should help a lot with summer heat gain (it doesn’t matter which way the foil faces, but it’s more likely to catch dust over the years if facing upward, therefore losing some of its value.

I will probably use foil-faced foam for [a] new bathroom floor ... (over an unheated porch) ...

Properties of Spray Foam Insulation for Roofs

In comparison, any properly-installed spray foam insulating job fits building cavity and mechanical system variations and penetrations perfectly.

Open celled foams are a bit less costly (per inch of finished thickness) than closed cell foams, are sprayed and trimmed, and are lower in R-value.

Closed cell foams (two pound foam) are more dense and heavy, are usually sprayed with little trimming (there may be a concern about the care with which ceiling or wall or floor cavities are filled to an adequate depth) and this foam job may cost 20 to 30% more, gains in both air and moisture barrier properties, adds structural rigidity to the building roof or wall where it is installed, and has an R-value of about R 6.5/inch - almost twice as much.

More details are at INSULATION R-VALUES & PROPERTIES

and at INSULATION CHOICES.

You'll want to compare the R-values per inch of the foam board you planned to use and figure how much thickness you'd need to fit into place to get even close to the R-values of the spray approach. 1-inch Dow Tuff-R board has an R-value of 6.5/inch for one-inch thick boards. That's not making any allowance for heat losses because of looseness of fit or cutting errors.

Once you've installed a makeshift insulation job, you'll be tempted to install the finish ceiling. And once the ceiling is in place you're not going to want to disturb that work for a long time, meaning you may be living with a poor or leaky insulation job and high energy costs.

Take a look at the "hot roof" or "packed" roof insulation discussion above and at the end of this article before making up your mind.

Watch out: when enthusiasts recount differences between the performance of a "packed" or "hot roof" design (presumably very high-R and "tight") and a ventilated roof design (more trouble, more cost, sometimes poorer insulation and higher energy cost) they sometimes may not have realized, or may fail to report that when a leak into the packed roof occurs, water stays there a long time - inviting rot, carpenter ants, mold. The result is a very large repair expense.

Conclusions & Recommendations: Hot vs Vented Roofs

Joe has more graduate-level education and more impressive credentials than I do (Joseph Lstiburek, Ph.D., P.Eng., ASHRAE Fellow at http://www.joelstiburek.com/ ), compared to ABOUT InspectApedia.com and Daniel Friedman's RESUME, and he's an exciting speaker.

And truth be told, despite the fact that we share a middle name (Joe) Dr. Joe is better looking too.

We agree on much of the hot roof theory promoted by Joe, but I maintain some practical objections elaborated below.

First and more significant: TenWolde and Rose have written a seminal work summarizing the issues and their conclusions on the topic of attic and cathedral ceiling ventilation.[7]

We conclude that while attic ventilation can be beneficial under some circumstances and climates, it should not be viewed as the principal strategy to eliminate moisture and other problems in the attic and the roof. Rather, attic ventilation should be part of a broader range of control strategies. Taking all factors into account, we make the following specific recommendations:

In summary, for each of the most commonly cited claims of benefits offered by attic ventilation - reducing moisture problems, minimizing ice dams, ensuring shingle service life, and reducing cooling load - other strategies have been shown to have a stronger and more direct influence.

Consequently, attic ventilation should be shifted away from its position as the centerpiece and focus of regulation. The performance consequences of other design and construction decisions should be given increased consideration.

Consistent with those authors I propose that increased consideration should be given to the effects on the building, roof, heating and cooling costs, and maintenance costs associated with

Watch out for real world snafus, damage, leaks in roofs. In sum, I'm left unsure about the gap between new construction designs and a perfect world where roofs never leak and the roofs I and more importantly, repair and renovation roof contractors have found when we inspected, tore apart, and repaired leaky roofs of both insulated cathedral ceiling homes, and homes with vented roof cavities.

Our expert Steven Bliss commented to offer a final word on Hot Roof Designs:

I agree with you that, in the real world, this is not such a good idea. If there's a flashing leak or other roof leak, you could have a pretty soggy mess that stays wet for a long time and could cause structural decay.

Plenty of people are building hot roofs, but I wouldn't -- except maybe one with spray urethane which won't absorb much water like cellulose would. [37]

Dense packing sloped roofs, residential: Is there Data Supporting Dense-Packed Un-Vented Sloped Roofs?

We moved this roof design topic to INSULATE ROOF with DENSE PACKING

Special Venting (or "packed" hot roof) Problems for Cape Cod Roofs

This discussion is now found at CAPE COD ROOF VENT or DENSE PACK INSULATE

Hot Roofs vs Vented Roofs in Hot Humid Southern Climates - Building Heat Gain

This topic has been moved to HOT ROOFS in HOT HUMID CLIMATES

Hot Roof vs Vented Roofs in Cooler Northern Climates

This article topic has moved to HOT ROOFS vs VENTING in COLD CLIMATES

Roof Leak Risks Need to Be Addressed When Comparing Vented vs Un-Vented Roof Designs

This discussion has moved to HOT ROOF DESIGN vs LEAK RISKS

Problems with Partial Roof Venting

This topic has been moved to PROBLEMS with PARTIAL ROOF VENTILATION

Watch out: If you can't provide enough intake venting to supply adequate air to the roof outlet vent system then it is probably better to not vent at all.

Solutions for un-vented roofs: Avoid cold-climate ice dam leaks & reduce cooling costs in hot climates

Moved to HOT UN-VENTED ROOF DESIGN SOLUTIONS

Design Suggestions for Unvented “Hot” Roof Designs Where Venting is Difficult

Moved to HARD to VENT ROOF SOLUTIONS for HIPS & CATHEDRAL CEILINGS



...

Continue reading at HOT ROOFS in HOT HUMID CLIMATES or select a topic from the closely-related articles below, or see the complete ARTICLE INDEX.

Or see HOT ROOF UN-VENTED ROOF FAQs - questions & replies posted originally at this article

Or see these

Recommended Articles

Suggested citation for this web page

HOT ROOF DESIGN PROBLEMS at InspectApedia.com - online encyclopedia of building & environmental inspection, testing, diagnosis, repair, & problem prevention advice.


Or see this

INDEX to RELATED ARTICLES: ARTICLE INDEX to BUILDING VENTILATION

Or use the SEARCH BOX found below to Ask a Question or Search InspectApedia

Ask a Question or Search InspectApedia

Try the search box just below, or if you prefer, post a question or comment in the Comments box below and we will respond promptly.

Search the InspectApedia website

Note: appearance of your Comment below may be delayed: if your comment contains an image, photograph, web link, or text that looks to the software as if it might be a web link, your posting will appear after it has been approved by a moderator. Apologies for the delay.

Only one image can be added per comment but you can post as many comments, and therefore images, as you like.
You will not receive a notification
when a response to your question has been posted.
Please bookmark this page to make it easy for you to check back for our response.
Our Comment Box is provided by Countable Web Productions countable.ca

Comment Form is loading comments...

Citations & References

In addition to any citations in the article above, a full list is available on request.



ADVERTISEMENT