Exterior Foam Insulation and Fire Safety
I’d like to see GBA address this issue as well as hear from fellow readers.
With IRC2021 effectively mandating continuous external wall insulation on houses and with the large majority of that being EPS, Polyiso and XPS, I’m wondering about the fire implications. We saw at Grenfell how once an internal fire breaches a window the cladding spreads the fire to all parts of a building. Also, we are experiencing more and more wildfires across the country.
I am concerned.
Yes, I know there are certain things we can do, such as opt for mineral wool insulation (very expensive), or do expensive and complicated detailing restricting ourselves to the use of 3-coat stucco. But this doesn’t address the reality that most new house construction will probably continue to involve flammable foam boards and rainscreen type applications which sound to me like a recipe for fire problems.
What are your thoughts about this? What research has been done on these issues? Is the IRC2021 pushing us down a path that’s going to resort in more, larger and deadlier house fires?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Antonio,
You may find this useful: https://www.buildingscience.com/documents/building-science-insights-newsletters/bsi-098-great-fire-london
Thanks. That was a very good description of the Grenfell fire. I read a lot about it at the time and this described it well for any who have not. I hesitated to mention it in my comment because I'm aware of the important role that building height plays. Nonetheless, as the articles says with regard to the houses it illustrates, "This assembly is not required to pass the NFPA 285 test.".
That's just it. I'd like to find out how these assemblies DO behave under those conditions. I'll look around a little more and see if any testing has been done or if any actual house fires have been studied. I'm definitely concerned about the wildfire issue as well, an external ignition source, since more and more areas are subject to wildfire.
Ideally the price of mineral wool will come down to the point where it's competitive with EPS and Polyiso for exterior insulation. That would help a lot.
Thanks again for the article. I hope others seeing this will read it.
From the Building Science article Malcolm cited, you could keep rain screens to 3/8" depth or less, but that might be a problem for 24"o.c. studs and cladding requiring 16"o.c. support. The article notes that 3/4" rain screens are not as bad as the Greenfell 2" gap, but 3/8" or even less will be sufficient.
One design choice that you have for a ventilated rain screen is whether to provide a vent opening at the top of the rainscreen (for better ventilation, but not required). Obviously the bottom of the rain screen must have an open vent to shed any bulk water that may get beyond the cladding. The bottom vent could have a small mesh 1/8" or less to resist flying embers from a wildfire (as well as other screen for resisting critters). If a fire does occur in the rain screen area, it would get less oxygen with no vent at the top of the rainscreen, by reducing vertical air movement through the rain screen.
One solution to the higher cost of rigid mineral wool for a rainscreen could be to use it as the exterior layer of exterior insulation, and use polyiso for an inside layer of exterior insulation. Nothing preventing you from mixing materials in exterior insulation. Polyiso can burn, but will char in place, and would be protected somewhat by the rigid mineral wool layer. Dupont Thermax polyiso is a glass-fiber-infused polyiso. It has improved fire resistance and can be left exposed (unlike other polyiso that must be covered in habitable locations).
Using fiber cement cladding rather than vinyl siding or wood, can prevent the cladding from burning.
Interior furnishings and the material and landscaping within a couple hundred feet of your building might be more important for preventing and containing fires than the exterior insulation, rain screen and cladding. So there are other ways to protect your flammable exterior insulation, if you must rely on EPS/XPS/GPS for exterior insulation for cost reasons.
Lots of great ideas here. Thanks!
The only issue I have with 3/8" furring over 2" of insulation is that the furring is not thick enough to attach wood siding or stucco lath to. The nails for cedar siding, for example, are supposed to have 1 1/4" penetration into wood. Even 3/4" furring strictly speaking, doesn't do the job. But with 1x4 SPF furring and 2 rink shank nails at each furring strip (24 oc), you get equivalent withdrawal strength. 3/4" furring is also needed for good support for stucco lath.
For this reason I'm using 1x4 SPF furring. I'd be happy to go with thinner furring, but then my siding has to nail all the way through the furring, insulation and sheathing to get to the studs, which is problematic on a number of fronts.
Antonio,
I hadn't thought about the problem of furring thickness limiting the ability to use thinner gaps, and can't think of a solution to that. Practically I find anything under 1/2" can be problematic if there is a WRB over the foam, as it's easy for it to either block the cavity, or bridge it defeating the capillary break.
Typically a vertical chase will need fire stopping at maximum 10' intervals to reduce the chimney effect. For a single storey SFD the chase created by the rainscreen should be less than 10' high, but for multi-storey it is something to look for.
A rain screen gap of 3/8 or less will create enough friction that a fire cannot propagate, and as mentioned above screen can be added that will not let in embers.
I think down stream decisions can reduce the need to worry, such as the type of landscaping and the proximity of flammables. Having rock or concrete or other non flammable surfaces up next to the house will play a major role in fire safety, as will eliminating trees, shrubs (cedars are especially bad) and wood fences close to the house.
If you are looking to fire safe your house also look to the soffits, roofing and gutters as potential ignition spots.
Thanks. You bring up good points. I'll implement them.
plumb_bob,
The BC coastal rain-screen provision has been interpreted as not requiring vertical fire-blocking, but the cavity can not be connected to the roof above.
https://boabc.org/wp-content/uploads/2019/12/18-0007-Fire-blocking-within-rain-screen.pdf
Thanks Malcolm, I have seen that, but it does give a max depth of 1" (I believe this 1" depth may be troublesome in a fire situation) and recommends flashing (fire blocking in practical terms) between floors.
I am curious if California or other jurisdictions have passed code provisions to address these details? These would be a good guideline in designing for fire prone areas.
If your concern is fire spread in vertical cavities like rainscreens, applying intumescent paint to the inside surfaces of the rainscreen would likely help. Intumescent materials expand when they get hot, so they would expand and seal off the rainscreen in the event of a fire, greatly reducing fire spread. That might be something to consider. I'm not aware of any formal testing of such an assembly though, and my concern would be if the intumescent paint would expand enough to really close off the cavity. Any restriction would help, but ideally you'd want the expanded material to seal things off completely to block airflow.
I agree fiber cement siding would help, since it's probably the most fire resistant of the standard cladding materials. I would go with polyiso here too, since it's much less flammable than EPS or XPS. It may help to install rigid mineral wool along the edges to provide a sort of fire barrier to protect the polyiso too, but I'm not sure how beneficial that would be in actual practice, especially with the rainscreen opening right next to it.
If you want to get fancy, you could put some thermal sensors in too for some early warning that there is a serious problem.
It would be good if someone would do some real testing of assemblies for fire zones so that we'd have some good information about better construction techniques for those at risk. I'm not aware of any testing of fancier assemblies though -- most of what I've seen has been in regards to embers getting into attics and things like that, not so much fire resistance of the wall assemblies.
Bill
Some of these links go back over 10 years.
https://www.greenbuildingadvisor.com/question/can-a-rainscreen-become-a-chimney-for-fire
https://www.buildingscience.com/documents/building-science-insights/bsi-129-wildfire
Antonio,
The short answer is design your walls to meet NFPA 285 regulations. That is a fire code required of most commercial bldgs & avoids Grenfell-type fires. NFPA 285 is not generally required for residential construction. It is a pain in the . . . as you must choose the ci Mfr & the insulation boards of interest & then read & follow a [Fire] Technical Evaluation Report (TER) provided by most plastic insulation Mfrs for their products, often different TER’s for their different ci boards. Every NFPA 285 compatible board is covered by a TER. You can’t say design with DuPont’s Thermax & substitute Rmax’s cheaper polyiso. You have to use Rmax’s TER for the equivalent board with different rules for sheathing, cavity insulation & WRBs.
Because many architects are just as concerned as you, they design to NFPA 285 for residential just to be safe.
But the rules are so damn complex that Architects often go to Mineral Wool 1st which results in thicker walls, more stud cavity insulation & moisture problems in certain climates. Because of this complexity, plastic insulation is used more in roofs & foundations where NFPA 285 is not a problem. Plastic insulation creates better, more economic walls, but the complexity limits it.
I’m a mathematician (taught at UT-Austin, Ga Tech, & USF) who has been working on problems in the construction industry for years & have solved the problem for commercial construction with a tool for Architects to use for commercial construction which includes Hotels, Dorms, etc. I’ve received guidance from Sr Partners at Gensler, Perkins&Will (the 2 largest architectural firms) & others. The tool will be released in days. If interested please respond & I’ll provide a link to the preliminary release which is being evaluated by the ci Mfr’s included in the tool.
--tex