SIPs on a Timber Frame
lynne_mc
| Posted in General Questions on
Hello,
I know the topic of SIPs has been discussed numerous times over the years, but sometimes understandings change over time (technologies, new studies, etc.), plus every new situation is different. Background: We are building a 1740 sq ft timber frame home at 7500 feet elevation in Colorado. The climate is hot and dry in the summer, and cold and dry in the winter – I believe we are in Climate Zone 5 with semi-arid humidity. Although we get a lot of thunderstorms in the summer, everything dries up quickly soon after. Humidity-related issues are pretty much non-existent here. We can get a lot of snow falls as well, but the snow usually melts and sublimates within a few days due to warming temperatures and intense solar effect. Regarding construction techniques, we are getting different recommendations:
1. Our timber frame company recommends we use a polyurethane-based SIP for the walls and a built-up roof system, which I believe consists of OSB sheathing and rigid foam sheets with staggered seams. This roof would have an inherent vent space, making it a “cold roof.” The reason they recommend SIP walls (vs a framed system) is due to the timber frame. Once in place, we’re told it makes constructing the walls a bit more difficult in that you can’t assemble the walls on the floor and then tilt into place. This added effort will increase the cost of the framing to where you may as well pay a bit more to get the SIP performance. Regarding the built-up roof system:
a. The overhangs will require additional timber structure to support the weight of the overhang. This can be done by extending the timber beams from the interior to exterior of the wall enclosure, thus requiring additional effort to insulate and seal around the timber extrusions.
b. Another option would be to terminate the roof timbers at the SIP wall, and then fasten a timber bracket on the exterior to support the overhang. This is only feasible if there is a timber post on the interior that you can “bolt” the exterior bracket to (assuming a SIP wall). We can probably employ this technique to 12 of 15 timbers. 3 timbers might still have to penetrate the exterior enclosure, although it may be possible to eliminate these as well – not sure yet.
c. If we went with a SIP roof, there would be no need to do any of this, as the SIP panels provide sufficient structural support on their own. However, a SIP roof, especially if we add a cold roof on top of it, is much more expensive. We were hoping to put the savings toward better windows.
2. Our builder recommends the exact opposite. He would prefer to build 2×6 framed walls with blown-in cellulose insulation and use a SIP “hot” roof system. He feels the climate and energy costs don’t justify the expense of SIP walls, nor any other form of advanced energy design nor exterior foam. However, I think he likes the SIP roof due to the ease of construction, and feels making it all weathertight is straightforward with good attention to detail – no cold roof needed, especially given the semi-arid climate in this region.
3. A couple years ago when my husband and I were first researching, we felt an AF wall system with exterior rigid foam was the best way-ahead, but we didn’t realize the added complexities of framing around a timber frame, nor anticipated the pushback (inexperience?) from local builders to using these building methods. We were also leaning towards a SIP roof with a cold roof added. We also considered a metal SIP roof, assuming it negated the need for a cold roof. In the end, we put a pin in it, and decided to press ahead with working the architectural design and figured would get back to the question of walls and roofs later. That time is now here
One more thing to add: labor rates are very high here at this moment. The area is experiencing rapid growth, and lots of new construction is underway.
We are very anxious to hear the recommendations of the GBA community and look forward to your feedback. Which of the options above would you recommend, or is there a better alternative? Although we are committed to being energy-efficient, we are limited by budget, worried about future disasters caused by inexperienced workers, but still want the pursue the best bang-for-the-buck.
Thank you,
Lynne
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Lynne, it's a tough choice, but one that will be decided by your very individual circumstances. I live in a SIP structure that I put up myself, so that's my cred here. I like option 2, what your builder says. A SIP roof exploits the structural quality of SIPs and eliminates a lot of wood and work; a 2x6 wall is not ideal but it's easy and cheap and cellulose is an excellent fill. You're looking for the sweet spot in terms of compromising cost of labor and materials, economy and durability. Plus, if you already have a good working relationship with a builder in a tight economy, this might be the most politic route to take.
About the only reason to use SIP is for quick construction. It doesn't make financial sense in any other application.
As for timber construction, the best way to work with is to build a standard house shell (well insulated, air sealed, frame wall with your choice of double stud or exterior rigid insulation with a vented attic with fluffy insulation) then install the timber beams/trusses inside.
Whether SIP or stud frame, the walls are more than enough to hold up your house, there is not need for extra load bearing timber structure. Trying to integrate the timber beams into the walls just makes your air sealing next to impossible. This also saves you designing load bearing timber trusses, they can be made to look as you want without worrying about holding up the house.
I know it sounds wrong, but a much cheaper, way more efficient and robust construction.
P.S. This is how my cottage up north is built.
Akos, one person's sense is often very different from the next. For example, I would never install fake beams in a house. Some people want a real timberframe, despite the fact that they have to virtually build a second house to enclose it. If conventional financial sense were the only thing at play, we'd all just be setting double-wides on cornfields.
Andy,
Maybe the choice of words was not the best.
They are not fake, for example in my case the timber structure support the interior 2nd story, but this went in after the shell was built.
Building a post frame first and installing walls around it makes it difficult to air seal as the beams are in the way. By reversing the order, you can build a solid shell and still have the post beam look.
Also timber roof trusses usually mean purlins, but these tend to go the wrong way for venting. You end up having to add extra structure above the vent the roof and even then the timber trusses get in the way of a continuous air barrier. Again, by reversing the order and installing the roof first, followed by timbering simplifies both of these details.
I did custom timber framing for fifteen years, and insulated most of the 50 or so homes with polyurethane SIPS, and unvented "hot' roofs. Since then I've learned that unvented roof SIPS can eventually fail, even if the best materials and installers have been used. The joints are typically filled with spray foam, which eventually cracks due to movement in the roof structure (spray foam is not flexible), moisture escapes through the cracks, condenses on the roof shingles or metal and rots the top layer of osb. And in a timber frame, since the joint is usually over a timber, it's impossible to seal it from the interior. So a SIPS roof MUST be vented. Most manufacturers recommend this, but many installers ignore it (It's cheaper - at first). But venting a sips roof means that eventually when the foam cracks, heat still leaks out but it just won't damage anything. A couple of years ago we installed rockwool batts over a board sheathed timber frame roof (with a layer of TJI framing to hold the batts), worked well, but did add to the layers (above the batts was a layer of osb, then 2x4's to form a vent channel, then Advantech roof sheathing. One alternative to doubling the roof framing is multiple layers of recycled foam, attached with long screws over an air barrier layer (Henry or similar). For the walls, you could simply extend your floor deck and build a stud wall on that, or plan for post support inside your floor system, and stud outside the timbers, leaving a 5/8" gap between the two where you can slip in 1/2" drywall. Then add 2" exterior foam for your thermal break. While you are doing all of that, keep in mind that air tightness is the key to a good house.
Bob, the SIP manufacturer I worked with five years ago was incredibly diligent in making sure the pitfalls you describe are NOT inevitable. The interior joint tape that's used in a responsible SIP build is modern high-tech stuff; that monstrous 18" wide strip of butyl that hangs over the ridge beam and is applied to the panels on both sides just seems like a belt and suspenders for a T-Rex.
Bob,
Thank you so much for your response. Your suggestion: "One alternative to doubling the roof framing is multiple layers of recycled foam, attached with long screws over an air barrier layer (Henry or similar)" is (I think) very similar to what I was trying to describe as a "built up" roof system. Also, the framed walls would indeed be installed in the manner you describe (i.e. stud outside the timbers w/5/8" gap for drywall.) However, with the entire timber frame structure hogging up most of the floor space, I would think the walls would need to be constructed in a staging area adjacent to the foundation, and then swung into place somehow. Not sure if that means having a crane onsite for the duration of framing. But however it happens, I'm betting it's going to cost more than framing a house with no timber frame to work around. If the cost is close enough to SIPs, then maybe SIPs make financial sense in this case?
That's great; exactly what is needed. Now all timber framers and all SIPS manufacturers meed to make sure this is done right. Fifteen years ago, after being away from SIPS for a number of years I decided to get more involved. I joined the SIPS association, went to the national convention and talked to installers and manufacturers about the problems, issues and methods to deal with them. No one was interested. They wanted to sell more product and any talk of problems was entirely unwanted. Let's keep sweeping it under the rug. I hope that is changing, but I'm sure it will be slow.
Which OSB polyurethane SIP manufacturer are your using? There is only like 2 or 3 in the entire country.
As mentioned, taping the joints, inside and out, with a proper vapor tape is essential. SIGA is really good tape.
Adding a breathable roof membrane is a good backup and then adding an elevated (1" purlin) metal roof is icing on the cake. If all 3 steps are taken (taping seams, breathable roof membrane, elevated roof), the SIPS can dry out properly if they get wet.
Peter, I just skimmed our folder to see if I could find the name of the SIP manufacturer the TF company recommended. I couldn't find it - it was a year ago when I wrote down that info. I'll keep looking, and post it if we find it.
Just want to make sure I understand your recommendation regarding an elevated (1" purlin) metal roof. I'm not a builder, so please forgive the ignorant question: are you suggesting we could have a standing seam metal roof installed directly over a SIP panel, but by elevating the metal roof via a 1" spacer board, this would allow for a vented "cold roof" system? This would be wonderful, as we always assumed we would have to cover the entire SIP roof with yet another layer of OSB atop a vented channel - and then install the roofing material over that. We are not comfortable going with a hot SIP roof - we'd never get a good night's sleep!
Thx,
Lynne
Lynne,
Yes, you can install a standing seam metal roof directly over a SIP panel by elevating the metal roof via a 1" metal purlin. The metal roof then attaches to these purlins via hidden clip fasteners. The metal purlins are spaced based on loads like wind and snow. Usually 4 feet between purlins is adequate for most roofs.
MBCI makes this roofing system. I used it myself and so far so good. See attached pics...
Thank you! That's very interesting. I previously determined the metal roofing was not cost-effective in our situation, but if it can be installed in a manner that creates a cold-roof for free (in essence), then it narrows the cost delta to where we would get a metal roof upgrade for free...sort of. Assuming we went with the SIP roof, of course.
Thank you everyone for your replies - very much appreciated!
Another question please: I think I understand the failure mechanism for SIPs (moist warm air exfiltrates through the SIP panel, hits a cold surface, condenses, and then begins to decay the OSB). Where we get confused is that OSB, insulation, warm interior air and cold exterior surfaces (the basic ingredients for a failed SIP) are present in most of the other options as well. In fact, since SIPs tend to have LESS air leakage, I don't understand why other framing methods (e.g. tired and true 2x4 framing with fiberglass insulation) which have MORE air leakage, wouldn't cause MORE condensation and MORE OSB rot and decay. If the answer is that these other methods employ air venting of some sort so that the water condensation has a chance to dry, then why not use the same methods for a SIP assembly so that both systems are evaluated equally? If another issue is that SIPs are structural and therefore the OSB failures are more catastrophic, then is this concern negated by a timber frame home, where the timber frame provides the structural support, and rotted OSB can be cut out and repaired more easily (should it happen)? Where I'm trying to go with this is... if we follow the TF company's suggestion (option 1 above), and ensure we have a rain screen in place between the SIP wall and exterior siding, have we really built a home that's any more susceptible to rot than any other wood and OSB-based framing method?
Lynne,
Roofs will get wet over time, that is normal. Metal roofs tend to be more watertight than other roofing systems but water/moisture can and will make it's way to the OSB substrate whether SIPS or conventional truss roof. As long as the OSB can dry, things are usually good. Although excessive wetness will cause OSB to swell and it can still rot, even with drying channels. As long as you keep it as dry as possible, employ a breathable membrane, you should have a lifetime of problem free roofing.
Conventional truss roofs that employ open attics and ones WITHOUT (closed) spray foam underneath them, will dry to the attic side when the OSB gets wet. SIPS can only dry in one direction from the top or bottom. If the top of a SIP gets wet, it can only dry going up. If the bottom of a SIP gets wet, it can only dry going down. So its vital to put a BREATHABLE membrane on top of a SIP Roof and elevate the metal roofing. As metal roofing is not breathable so a channel must be allowed underneath the metal roof to allow for air to flow. Hence the metal elevated purlins.
Lynne,
I think you have a fairly good understanding of the science behind potential SIP failures.
I did want to add a couple of points however as there are some characteristics of SIPS assemblies that make them riskier than some other assemblies.
First, your recent comment seemed to be focused on SIP walls vs stick built walls. I don't think there is too much concern about SIP walls failing. The real danger exists with SIP roofs.
There are two challenges with SIPS- especially SIP roofs that make them riskier than some other assemblies.
1.) They are hard to air seal, especially when attached to a Timber-Frame 'skeleton'. Remember, the SIPs intersect over the structural timbers. It is critical that these seams be sealed on the INSIDE. But how do you tape a SIP joint when there is a rafter/beam in the way? It is possible, but not easy... and I imagine that many builders get this wrong.
2.) Because taping these seams is difficult or impossible, many builders resort to spray foaming the joints (often done anyway). Most SIPs include a channel designed specifically to receive this spray foam. The problem here is that spray foam may not fill the entire channel as you have no way of seeing the foam as you spray it into the channel. Over time, foam can shrink and the SIPS components will move/shrink as well. This opens up air gaps/channels.
*This leads to a very unique problem with SIP Roofs: Three Dimensional Ridge Rot*
Roofs are more susceptible than walls to vapor issues because of the Stack Effect. That is why I wrote earlier that I think few people are too concerned about SIP walls. Remember that cold air is dense and has little space for water vapor. Warm air is far less dense and has plenty of capacity for water droplets. So, the warm air floats above the cool air as the cool air sinks to the floor. This warm, moist air hits the ceiling plane looking for an escape.
Those hard-to-tape seams allow that warm, moist air to enter the edges of each SIP. The channels (which were, in theory, supposed to be perfectly filled with spray foam) create a stack-effect driven network that carries warm, wet air from every SIP seam of the roof and dumps it all on the ridge resulting in rot. Its almost a perfect system for getting wet air to the ridge.
You mentioned that you live in a dry climate a few times. Just remember that the vapor that results in Ridge Rot is not vapor from outside. Rather it is from your home (cooking, showers, breathing, watering plants, gas appliances, etc.) and the ground. That's right- the ground.
Many vapor-related roof issues actually come from the ground. Even in arid climates, the relative humidity of the soil under your home is nearly 100%. Your builder/designer must take special care to keep that ground-source vapor out of the home or else you will really magnify the Ridge Rot issue. This includes adding a Class I vapor retarder under the slab and capillary breaks under footings.
Here is a great article from the Building Science Corp on Ridge Rot:
https://www.buildingscience.com/documents/insights/bsi-036-complex-three-dimensional-air-flow-networks
I had to make the same determination, and this was a long and exhausting process of research and getting quotes directly from SIP manufacturers and framers to see what the comparison *should* be. It's a bit more complicated when you are buying the SIPs from a timber frame company. The cost tradeoff will really depend on your builder and subcontractors. For me, the subcontractors were charging more for working with SIPs because they were unfamiliar with them, the only labor and time savings were to be had on the roof, but no where near the cost to buy and ship the SIPs (which may also require an expensive crane). The walls I was told would have no labor and possibly no time savings, as any site modifications to the SIPs required would significantly delay installation. They may be more common in Colorado, but I'm also really surprised your builder is heavily against exterior rigid foam. I would have thought that would be more common in Colorado. One option to ask your builder is using ZIP-R insulated sheathing, in theory this should simplify a lot of the additional work of adding continuous rigid foam since it's built into the sheathing. It might also be worth asking around to other builders as well if that's an option for you. I'm also a bit surprised to hear about the recommendation of SIP walls because you can't build and tip up site framed walls. SIPs would have the same issue, you need enough room and heavy equipment to be able to lift the SIPs into place for the walls- the same thing you would need if you stick built walls in a staging area and moved them in place.
In the end, I ended up choosing a conventionally framed roof with dense packed fiberglass (less settling than cellulose) with exterior rigid foam, and a venting channel on top of that (mostly for ice dams). The walls are simple 2x4s with mineral wood, with zip sheathing and 2" exterior rigid foam (zipr could be an option here, locally it was very expensive to special order as no one stocks it), and homeslicker venting mesh on the exterior for a venting channel. This hasn't been completed yet, so I guess time will tell if this was a good idea.
Here is my breakdown of the pros and cons:
SIPs Pros:
•Potential labor savings, at least for the roof, roof likely faster to install. I have continually heard that electrical can be quicker. No one I talked to locally here substantiated this claim, however. SIPs are not common in my area.
•Less waste, at least on site.
•Continuous insulation built in.
•May be a more straightforward path to a high *effective* r value wall assembly for builders inexperienced with other wall assemblies.
SIPS Cons:
•Expensive. Not counting shipping, you can expect $10sqft+ for the walls and roofs. Adding deep roofs to get above code minimum R38 to say R49 is going to be even more expensive.
•I have repeatedly heard you need contractors experienced with SIPs, or expect hickups and long delays.
•Labor savings will be dependent on your contractors. Some, like electricians, may charge thousands more just to work with them. It's really something you will just have to ask around and get numbers for to determine.
•May require more expensive lifts to lift panels in place.
•Walls may require site modifications and end up being the same/longer than stick framing. Stick framing exterior walls does not take long.
•People selling the SIPs either claim that moisture failure issues have been solved or claimed to have never heard of them. It's up to you how much you trust this, but my take is that the fixes haven't been around for long enough to really be certain.
•Remodeling- Any changes to electrical, walls extensions, etc. are going to be far more difficult with SIP panels than with traditionally framed walls.
I'm curious if anyone has reasons why Zip-R shouldn't be treated like a SIP ? Specifically, requiring an interior side air barrier in cold climates.
I could see that being a concern, since the interior panel gaps aren’t even sealed with zipr. Perhaps because at least there’s only one sheathing layer and I’m assuming if using zipr a rainscreen ventilation gap would be required to dry to the exterior?
Aunt and uncle have a high end timberframe home, with sip construction, and are battling sip rot right now.
The outer skin of the roof is mush in several places, the only real fix is to stick build a roof over top of the existing panels, or remove and replace all of them.
Sips didn't even outlive the asphalt shingles
How frequent are SIP failures? If the many issues brought up are common place compared to the tiny market share they account for, how could they still be in business? Are the failures all outside of warranties? Blamed on the installers? *If* it truly is really prevalent, why has there not been a class action lawsuit? Or has there?
Carson, wait, what data points to a high proportion of SIP failures? Yes, it was a huge problem in the early days of the industry, but my contacts with three manufacturers in 2013/14 led me to believe that they pulled up their socks and instituted clear installation guidelines. FWIW, my SIP house may be the only one in town, but it's not failing. As a contractor I work every day on conventional houses built since 2000 that are molding, rotting, and falling apart. People just call my work "updating;" they don't condemn their leaky stick homes.
Trevor, please tell us what caused the SIP rot, if you're able. When was this built? How thick were they? Were the interior seams taped? Were the shingles placed directly on the SIPs? These panels don't simply rot on their own--they're subject to some sort of failure mechanism, and it's most helpful to the discussion at hand if you give a little more detail. Are your aunt and uncle in communication with the manufacturer, designer, or builder? Thanks .
Andy, I don’t have real data, that’s why my question was genuine. I have seen a lot of failures posted in various parts of the country and many people on GBA seem to caution against them, but it’s hard to determine the actual failure rate. Hence the question, if they are truly prevalent, how can they be in business? It would stand to reason that the issues aren’t prevalent across all manufacturers, or they must have found a way out of liability. The scary thing for home owners is that the issue may well have been fixed a decade or so ago, but it may take longer for issues to become apparent.
I think the issue was fixed--among professional builders and the manufacturers--but the lag time seems insurmountable, due to the conservative nature of the industry and its consumers. The internet gives anecdotes the upper hand, and SIPA (the industry association) has not overcome it. I'll also acknowledge that while SIPs are a great idea, there are just enough issues that they don't represent THE great idea that will change construction as we know it. I mean, you could probably create a building shell with graham crackers if you get the control layers right. In the end, SIPs are vulnerable to failure, in a way that stick construction is not, and if a system requires such perfect adherence to protocols, it's not going to fly with builders who are not.... perfect.
Andy, do you know why Benson, one of the earliest proponents of SIPs with timberframes, now uses framed panels with dense packed cellulose over foam core sips? Cost? Carbon? Resiliency? Ease of wiring and modifications? All of those?
Additional note on your overhangs- I’ve read that timbers extending out of your air sealing layer is a bad idea. Sealing them well is going to be really difficult since timbers shrink a lot. And then there’s thermal bridging. SIPs would work. I think the solution would depend on the size of the overhang, but for small ones GBA has details of overhangs fastened on after your main roof framing goes up and is sealed. For larger overhangs, your framer should know, but basically they build out a section perpendicular to the main trusses.
It looks like today's article actually addresses this to a degree: https://www.greenbuildingadvisor.com/article/a-sure-path-to-high-performance
All,
First of all, my apologies for not responding earlier, and PROPERLY thanking everyone for taking time and consideration to reply. I'm going to blame my computer! I was refreshing this page for about a week, and no responses ever appeared. After a periodic reboot, and reopening my web-browser - voila! All of a sudden, the newer replies showed up.
It sounds like SIP walls are generally ok, but SIP roofs are iffy. This would seem to bolster the TF architect's suggestion: option 1B above - (SIP walls and built-up cold roof system) would be a generally safe and effective way to get high-performance enclosure. Just need to make every effort to ensure timbers do NOT pass through from interior to exterior. This may pose some challenges, but we'll work through them. Based on everyone's feedback, I'm definitely staying away from a SIP roof. Deep down, I would rather have framed walls with exterior foam - just seems safer in many ways (site mods, last minute electrical changes, etc.) Maybe we can delay this decision until we are closer to a completed design. SIPs seemed to offer so much benefit - such a shame about the inherent risks. However, at least one good thing came of it - if we had never heard about SIPs from the TF community, we would never have thought of building an energy-efficient home at all.
Thank you, everyone!
Just a warning that SIPs may affect some portions of your design. Usually I hear people doing the other way around- SIP roofs and framed walls due to more time savings and less issues with electrical. There are other options as well, like double stud walls.
If you aren't already aware of this and have an interest - you might want to look
at these videos: https://www.youtube.com/channel/UChhBsM9K_Bc9a_YTK7UUlnQ/videos
They show the construction from start to finish of a timber frame home with
Sips - they did it all themselves, didn't hire it done and did a pretty good job.
Have fun !
Without going into too much detail... I would recommend conventional framing. If SIP panels are done wrong it created rot channels as the dew point of small cracks through the plywood starts ruining the plywood. There is not really any fix after that. Plus your heavily reliant of the outer most plywood not getting wet. You can’tuse ice and water on the SIPs either. Conventional always allows for quick and easier repairs for the long term of the house and gives you any option for change down the road. Good luck.
Wade Rikert,
I am curious to know about your observation of rot channels in plywood and also what if any SIP panels use plywood. The bottom side of a SIP panel should not be anywhere near dew point even for very high humidity levels. The upper cold side, under whatever roofing material should not be seeing air exchange sufficient to drive moisture accumulation despite being cold. Also, why can't one use ice and water shield on SIPs, plywood or OSB? Local code for my area demands full ice and water shield under metal roofs.
Peter L's comments (#8) and photo fit more with my understanding of sealing SIPs to beat the seam leak problems. The panels appear quite well sealed. If one drapes 12- 18" sealing tape over the beams where panels meet, wouldn't that effectively seal the one difficult hidden joint.? No transport of interior moisture through the seams should make the panels secure.
I do question the value of 1" standoffs for the metal roofing. A 1" gap will resist air flow pretty well. Mr. Holladay has commented frequently on the magic arrow problem as relates to intended air flow and actual airflow. What one wants to happen is not always achieved.
Such is the case with various siding rain gaps and the perceived risk of fire spread. The conclusion is that gaps 1" and under do not support flame spread despite some predictions of disaster. I suspect that a lack of air flow is the reason. The same lack of air flow should be anticipated for roofing on such low sleepers. Given that rain gaps are vertical not sloped, the pitch of the roof probably is not hugely relevant to increasing air flow.
This lack of air flow might aggravate moisture collection under a metal roof, especially during parts of the year when day/night temperature swings combine with higher relative humidity to create dew conditions under the roofing. I believe I have read about such conditions being a problem in the humid parts of Texas, though I don't think other areas are immune.
A permeable roofing membrane doesn't allow air exchange, just moisture transfer. I do not know if they can act "intelligently" like the interior moisture membranes, nor do I know if bulk water (dew) sitting on a permeable membrane can be guaranteed to not transport moisture back through. Everyone speaks of permeable roofing membrane as being a fail safe application for sheathing, but without sufficient air transport across the membrane to carry away the moisture load I wonder if the sheathing can really be counted on to dry.