Asphalt shingles, synthetic underlayment, closed-cell foam and buckled shingles
I am just finishing up a home.The roof was installed about 1 year ago.
I have 6-8 inches of closed cell foam on the attic side of osb roof decking(unvented attic). The roofer used synthetic underlayment(grip rite shinglelayment) on top of the osb and then certaintweed landmark asphalt shingles on top of the underlayment.There are garage sections of my roof that do not have closed cell foam used and a more tradional vented attic.
The problem is my shingles are devolping sections of linear buckling(raised up about1 inch , my best guess) parrallel to the roof rafters and these extend about 10 feet parrell to the roof rafters(about 6 inches wide )as visible from the ground. I dont think the problem is with deformed osb , as the buckles can be easily pushed down.The buckles also seem much more visible in the hotter times of the year and day.This is only occuring in the main roof with closed cell foam on the underside of the osb.The foam from the attic side looks fine with no cracks or separation from rafters.
My questions are:
1) Will the shingles that are buckled up still perform.There must be some movement in the shigles with the heating thru the day and season and I would imagine at the very least I will have some nail attacment issues . I imagine there is some shigle movement as the buckles increse and resolve thru the day/season.Do i need to replace/redo the roof?
2) Why did this occur? Is it shingle installation( installer says its not an install problem).Is it the underlayment , perhaps heated by the closed cell foam application against the osb and then became deformed or malfunctioned( the underlayment company has a tech bullitin that says there product performs to 230F) . Is it a shingle defect ( the shingle company allows the used of unvented attics with there product).
So far I have had the shingle local rep out and he has passed it on to a company engineer and so far I have not heard back
Thanks for any insight or suggestions.
Ted
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Ted,
There's a lot to say on this issue. I'll start with one red flag I noticed: your roofer used the wrong type of underlayment. Most synthetic roofing underlayments can only be used on vented roof assemblies or over vented attics.
Here is the ICC-ES evaluation report for Grip Rite Shinglelayment:
http://www.icc-es.org/reports/pdf_files/ICC-ES/ESR-2945.pdf
The report notes, "Installation is limited to roofs with ventilated attic spaces in accordance with the requirements of the applicable codes."
I doubt that this factor is the main cause of your problem, but this error may prevent you from making a warranty claim to the shingle manufacturer.
It may be unrelated to the buckling, but that underlayment is an ultra-tight 0.03 perms vapor barrier, and with 6+ inches of closed cell foam there's no way for moisture to get out of the OSB. Hopefully the decking (and the trusses or rafters) weren't rain-wetted prior to assembly, and had plenty of drying time between shipment from the mill to the site. It would literally take years to dry through the foam, possibly centuries through the underlayment. Any leak points will only likely be discovered when you re-roof, but would almost certainly require patching some punky sections of OSB. (The good news is that at 6" closed cell foam is pretty structural- you wouldn't be crashing through it walking on the roof even if the OSB eventually became toast in 30-40 years.)
Is the shingle cupping/bucking phenomenon exhibiting at the rafters, or is it something like mid-way between the the rafters?
Ted,
Asphalt shingle buckling is a well-recognized problem. It is usually associated with high humidity in the roof sheathing or underlayment.
Here are some resources to study:
Asphalt Shingle Buckling
How to Minimize Buckling of Asphalt Composition Shingles
Plain Facts About Buckled Shingles
I don't know if you are the homeowner, the roofer, or the GC. If you are the homeowner, you have a valid claim against the contractor, who chose the wrong roofing underlayment. That's a code violation. A reasonable remedy would be to strip the roof and install roofing with the correct underlayment. This work should be performed at the contractor's expense.
If you are the roofer or the GC, I hope you have insurance.
Thanks Martin and Dana. I am the home owner in this scenerio.
Dana I am not sure of exactly where the buckling is occuring either between the rafters or on the rafters . how would one tell. My best recollection is that the buckling seems to be evenly spaced when noted from the ground implying PERHAPS this is occuring at the rafters?
Martin I appreciate the links.Seems to me this is a well known problem in the trade? i was surprised that the shingle arae rep did not know of it? I also found this link ( https://www.greenbuildingadvisor.com/community/forum/energy-efficiency-and-durability/14142/solar-vapor-pressure ) on GBA that talks about underlayment beneath asphalt shingles . Do I understand correctly that solar vapor pressure can cause buckling and one should use a vapor barrier beneath the shingles---
"To prevent solar vapor drive into roof assemblies, Lstiburek and Rudd advocate installing a vapor barrier between the asphalt shingles and the roof sheathing"
Thanks again for the help!
Ted
Ted,
Since I wrote the article on inward solar vapor drive for Energy Design Update -- the article that I quoted in my earlier post -- Joe Lstiburek and Armin Rudd have voiced new doubts on the cause of the phenomenon they observed. The last time I spoke with Armin, he said (to the best of my recollection), "The data are confusing, and it is no longer clear that we were observing inward solar vapor drive through asphalt shingles. I wish I had a million dollars to conduct more research on this issue, but I don't."
One thing is clear: your roof provides evidence that the use of an impermeable roofing underlayment is insufficient to prevent shingle buckling, in spite of Joe Lstiburek's hopeful surmise about a possible solution to the problems in Houston.
Finally, as far as I understand, it's always possible to install vapor-impermeable synthetic roofing underlayment if you have a vented attic. But no manufacturer of such an underlayment allows the use of its product on an unvented cathedral ceiling.
UPDATE-- The shingle rep came out and sent the following opinion:
I have spoken with key staff at CertainTeed in the Insulation, Roofing and Building Science division regarding XXXXX home in Spring City tn
Thoughts are that the issue here is several items combined to make the roof appear the way it does. The issues are deck and framing related along with insulation.
Here is what has collectively been identified. The fact that the roof was dried in with synthetic underlayment with a perm rating of less than one, (most all synthetics are less than one on their perm rating) there was some amount of moisture collected/trapped under that underlayment. Next the closed cell foam was applied as well possible trapping moisture/humidity. Ken told me that he did not use plywood clips are spacers between the sheets of decking, which I have checked with the trade association with OSB and it recommends the use of clips or spacing the sheets 1/8" between the sheets. So from this we may be getting some swelling of the edges with the deck along with movement.
Closed cell foam insulation was applied to the underside of the roof deck but not the actual framing members. The 2x10 framing members are un-insulated except for the depth of the insulation. CertainTeed Insulation group suggests coating the entire rafter with some amount of insulation. The mirroring of the framing member is some Thermal Bridging taking place where the rafter is maintaining a different temperature than the roof deck. It is either hold cold or hot temperature depending on the ambient temperature at the time.
The last item to discuss is the attic itself. This area needs air movement. It should have 3-4 ducts and one or more return air vents in the amount of space in the attic. The HVAC system is incredible in this house with all the humidifier, dehumidifier, and heat recover system. The attic became conditioned space with the addition of the foam insulation to the roof deck.
What could possible improve the appearance? These are speculative ideas at this time but suggestions would include:
1. Addition of the air movement in the attic
2. Applying the foam insulation to the 2x10 rafters, 1-1.5"
3. Hire a company to do moisture checks on the roof assembly. TBM Contractors in Chattanooga has the equipment to do a Thermal Imagery Photography ...it would be done at dusk as the heat source is removed from the building and it starts to cool down.
I would still like to submit roofing shingle samples. I would need 3-4 shingles from an area where the most visible uplifting is taking place. I will send these in and get them evaluated
Everyone here agrees that this is a decking issue.
My comments:
The builder removed a section of shingles about 3x 8 feet .On cutting into the underlayment and peeling it back the roofers noted that the osb was wet (too high up for me to go so I did not see it ) and puckered up at the osb seems from moisture. Unfortunately nobody had a moisture meter to get objective evedence and so will be back up to look further.There was no rot in the osb noted, yet!
Now on further consideration it looks like the roof osb was not separated from each other, and the sythetic low perm underlayment was placed on the roof and lapped ( seems not taped)
Now for the questions;
1) How did the roof decking get wet and swell. Could this be solar drive thru the shingles and up under the underlayment lap joints, thereby getting trapped between the osb and underlayment?The attic side of the foam is tight and about 6 inches of closed cell , so I dont think the water vapor is coming thru the osb from inside the house and condensing on the osb side of the shingle underlayment
2)The shigle rep talks about more foam on the roof rafters to minimize thermal bridging.There are large sections of roof where the undersides of the rafters are covered in foam and there are still buckling pfoblems . I dont recall seeing other foamed roofs with there rafters covered in foam?
3) The shingle rep talks of formal conditioning of my attic space but it is passively conditioned and it stays about 50-60 degrees with a humidity of 40-50%
The builder has been very helpful in all of this but nobody seems to know what to do , and there seem to be no easy(ie inexpensive) ideas. The builder points to another roof with standard felt underlayment and it also shows buckling of the shingles and I guess wonders what role the underlaymet plays? My thought is there are more factors to this buckling than just underlayment , but in my case I do wonder if the underlament placed over an improperly dried osb is the main culprit?
Thanks for any suggestions or guidance
Ted
How sure are you that the foam is closed-cell foam?
As far as possible residual moisture in the OSB, it could easily have been rained on at the yard, in transit, at the jobsite, or on the roof before dried in. The combination of closed-cell foam (if that's in fact what you have) and synthetic underlayment means that any such moisture can hardly get out, as Dana noted, so you could be getting swelling at the panels edges and/or air pockets forming as vapor gets heated and wants to expand. In any case, you have a very unforgiving assembly.
You could easily identify the rafter locations with thermal imaging--if the house is visible from the street, it wouldn't even be necessary to get out of the truck. With a bit of prep work, you could find out if the raised lines in the shingles correspond to framing locations (I would probably do that by putting short pieces of foil tape along a few of the ridges, and then imaging the roof at or after dusk). I'm not sure it's worth the trouble, though--regardless of what you find, your problem will either go away or it won't, and you'll either replace the underlayment and roofing, or you won't.
You could also take moisture meter readings in the sheathing and rafters, by going into attic and using deep wall probes (they look like knitting needles, and attach to the meter with a short wire).
OSB sandwiched between two impermeable layers is never a good idea.
Hi Ted,
I remember your project from July 2011
https://www.greenbuildingadvisor.com/community/forum/energy-efficiency-and-durability/21206/how-insulate-floor-trusses-used-ceiling
It sounds like at least one problem is moisture trapped between two impermeable layers.
I am curious to know if you also insulated in the truss space below the "attic floor" as you were contemplating in the original post?
and how about posting photos showing the buckled shingles
thank you
http://pinterest.com/johnkbrooks/enclosure-failures/
David thaks for the metering hints and yes the roof is closed cell foam to 6+ inches thick.I dont quite understand why it is important to indentify thermal bridging because it is going to be there.Although the buckling is occuring in areas where the rafters were entirely coated as well as not coated by closed cell foam
Martin ,I know NOW the osb is in a bad place but it is what I've got to work with(yikes!). The builder is interested in fixing it but wants to first understand why this is happening and would felt rather than the synthetic underlayment have prevented it .My thinking is that :
1) The osb is getting wet/moist and expanding causing the shingle buckling.The osb is not
spaced and as it expands it compressses some shingles causing the buckling.The
buckling occurs in a direction perpendicular to the long axis of the shingles which is also
the direction of the rafters. I should also point out that one of the reference articles you
mention above suggests buckling occurs even if the osb is properly spaced.
http://www.rci-online.org/interface/2002-07-logan.pdf
2) Now why is the underlayment wet. Is it because of :A) moisture transfer from
inside the attic, B) wet osb as it was not properly dry at the time of underlayment
placement ,or C) solar drive thru the shingles and underneath the lapped underlayment.
How would one figure this out or are there other possibilities?
3) Now the thousands of dollar question is how to fix it?
John , thanks for showing interest. Yes its me again. What we ended up doing was spraying about 3-4 inches of open cell foam between floors including the underside of the attic floor and then also placing about 4 inches of fiberglass batts in the truss space , which is about 16 inches deep. I am attaching some pics taken by the builder on a dormer section of the roof . Just checked out your pinterest enclosure-failures and noticed my roof rafters are there with some pretty ugly pictures. Were you trying to tell me something?
Now a key thing in this is the builder wants to make it right but of course hesitates to trade one problem for another and we are both trying to understand it so the next move fixes it. One thing I ( have not yet discussed with builder) thought of is to try to fix one of the dormer roofs as it could be done in relative isolation from the main roof and used as the "fix experiment" But what to do to fix it or understand it better?
Thanks for any help
Ted
Ted, my reference to identifying the rafter locations was based on something you said, wondering if the bulges are occurring at the rafters or not. If they are, that's a clue.
Your first two pics are confusing to me.... looked for a second and took nothing from them. In the third pic, I can't see where shingles are butted together in those courses. Is it possible the shingles themselves are expanding in the heat and buckling, because they can't move sideways as they heat up? Did the shingle rep actually come out to the house (sounds like he did...)
David, yes the shingle rep was up on the roof about a month ago. The rep reported back as per item 6 above and concluded it was a decking issue. The rep did ask for some shingles to be sent to them and that was done last week. The first 2 pics are of the underlatment and the one with the level was done by the builder over one of the raised areas on the osb where it has puckered up at the seems as it swelled.Can shingles expand?? or is it the decking is moving as it swellls as per this article http://www.rci-online.org/interface/2002-07-logan.pdf
Ted,
There are many factors here, and we can't sort them out over the Internet. A perfect solution would be expensive. This solution would solve the problem, but I'm sure no one wants to pay for it:
1. Strip the shingles and underlayment and throw them away.
2. Install #30 asphalt felt underlayment.
3. Install 2x4s flatways (1.5 inch high) on the roof to create ventilation channels from the soffits to the ridge, with one 2x4 above each rafter.
4. Install new plywood or OSB sheathing on top of the ventilation channels. The sheathing should be installed with H-clips or should be carefully gapped 1/8-inch at the seams.
5. Install #30 asphalt felt on top of the new sheathing, followed by new shingles.
What do we know? The roof has the wrong underlayment. The OSB is now wet.
It's possible that there has been some moisture migration from the interior to the OSB (either via the rafters that protrude through the foam, or through small cracks in the spray foam).
It seems that even with that solution, you might again have buckling problems. It could save your roof deck from rotting, though.
David,
"Even with that solution, you might again have buckling problems."
Wow -- that's an unexpected prediction. Really?
The low-perm underlayment is removed; ventilation channels are installed above the existing sheathing; a new layer of brand-new sheathing is installed (properly gapped) above the ventilation channels -- and you predict that the new asphalt shingles will once again buckle?
What mechanism do you propose to explain the predicted buckling?
Martin, your links above document buckling problems even when the roofing is installed perfectly. You must have posted them for some reason. Do you think his deck is now wet, and that if new felt and shingles are installed, there couldn't possibly be a recurrence? The only possible issue is the synthetic underlayment?
Also, the roofer is only at fault if he knew an unvented assembly was going to be built. I don't know if that was stated by Ted or not.
David,
You wrote, "The only possible issue is the synthetic underlayment?"
No. As I wrote in my last comment, I don't know the list of all the possible issues, because I haven't made a site visit. But I mentioned the possibility that the OSB sheathing may have taken on moisture from the home's interior.
Moreover, I suggested two changes, not one:
- switch the underlayment from a vapor-impermeable underlayment to a vapor-permeable underlayment; and
- install venting above the existing sheathing -- a change which will go a long way toward assuring good moisture management.
Martin, sorry, I read your post too hastily on a tiny screen.... my bad. Framing a new deck over and above the existing does seem likely to work, although it will probably make the builder cry. Question: why two layers of felt?
Hi Ted,
I know your post is almost a decade old, but I am having a near identical issue with a new roof installation of Certainteed shingles. Did you uncover the issue and resolution?
Thanks for the feedback.
Tom
Being that the OP is 10 years old, probably won't get a response.
The roof system described is similar to a closed cell polyurethane SIP roof. OSB - 6" of closed cell spray foam - OSB. One definetly wants to install a BREATHABLE synthetic underlayment (20+ PERMS) and then install an elevated finished roof. You don't want shingles or a metal roof sitting on top of this assembly as the problems mentioned can happen. The OSB needs to have the ability to dry, even if slowly, it cannot dry if a non-breathable membrane is installed and/or the final roof finish is installed directly on top of it.
I would also tape the exterior OSB seams with a breathable tape like SIGA Wig-Luv.
Telegraphing through the shingles means that the OSB decking is not flat or something is happening with the OSB. All materials expand/contract with heat/cold/sunlight.
Here is how I finished off my SIP roof:
I added 2 inches of closed cell foam (and then rockwool batts) to the underside of my roof deck and furred out the rafters with 1.5 inches of polyiso foam.
My roof is due for a re-do, and I share the roof with the other half of the duplex, that I don't own, so I don't think a lifted roof is gonna be feasible.
I believe my roof decking is plywood rather than OSB, from its appearance from the interior before the foam was applied (the house was originally built with skip sheathing, in the early 1900s)
I don't know if the plywood will need replacing anywhere when the shingles come off, but is plywood and felt an adequate assembly for under either replacement asphalt shingles or a standing seam metal roof?
You have Rockwool sandwiched between closed cell foam and polyiso? Is the polyiso foil-faced? That's a risky assembly.
No, the assembly is wood decking, closed cell foam and rockwool between the rafters. The rafters have polyiso foil-faced strips on them to furr out the rafters to make them deep enough to fit the rockwool batts and to not have thermal bridging. (zone 4a)
Ah, got it, thanks.