Wildland-Urban Interface Concerns Could Torch Some Window Markets
Fire officials and others are increasing efforts to protect homes in “WUI” boundary areas, and the industry needs to be prepared with sound recommendations for fenestration products
An educational presentation from the American Architectural Manufacturers Association. By Grant Muller, Mikron Industries
January 15, 2002
FEATURE ARTICLE | Codes & Standards
January 15, 2002
FEATURE ARTICLE | Codes & Standards
If the danger posed by wildfires to populated areas wasn’t on the minds of fire protection officials before, it certainly was after the summer of 2002. This last “fire season” was one of the worst in the nation’s history, with nearly 6 million acres scorched. Some particularly large fires gained headlines: the Biscuit fire in Oregon that consumed nearly 500,000 acres, the Hayman and Missionary Ridge fires in Colorado which took out a total of some 208,000 acres of forest and 189 residences, and the Rodeo fire in Arizona that claimed 463,000 acres and 423 homes. While none of these approached the damage of the October, 1991 Oakland/Berkeley Hills fire in which 2,449 homes and 437 apartment/condo units were destroyed, the sheer number and size of the 2002 fires took awareness to new levels.
All of this is heightening concern as to what can be done to protect homes located in what has become technically known as “wildland-urban interface” (WUI) boundary areas where suburban residential developments meet fire-prone open grasslands, forests or brushy areas. A web site search engine yields more than 18,000 sites dealing in some way with the WUI issue.
That awareness is spurring a careful examination of building practices, ranging from landscaping guidelines to the performance of building components (including windows) when subjected to impinging exterior fires. Many states, as well as the Federal Emergency Management Administration (FEMA), have released publications addressing what homeowners should consider in protecting their WUI residence from potential fires. The awareness is also prompting action on the regulatory front, mostly (so far) at local levels. Some of this action, as is often the case in such situations, has been precipitous – for example, seeking to ban entire classes of building products or materials without considering comparative performance data
Using procedures similar to those conducted on various window types at the University of California Forest Products Laboratory, Mikron Industries testing showed that it was typically the glass, left, that failed and allowed flame penetration first.
Windows in particular are considered to be one of the most vulnerable portions of a structure when it is exposed to fire. This vulnerability is due to several factors. The thermal shock of direct exposure to flames or the impact of airborne debris could shatter the glass, permitting burning brands or flames to enter the building, virtually assuring its destruction. Conventional wisdom holds that the window frame is also sus ceptible to burn-through under direct flame exposure. Radiant or convective heating, such as from adjacent burning shrubbery, might not break the glass but could ignite or deform the window frame, allowing the glass to fall out and again exposing the building to subsequent entry of flames.
UNDER FIRE IN CALIFORNIA
The first concrete issue involving windows surfaced even before the dreadful 2002 fire season, in spring of 2001, when officials of the Rancho Santa Fe Fire District and San Diego County announced a prohibition on the use of vinyl-framed windows in homes facing a wildland/urban interface, unless a 100-foot buffer zone cleared of combustible vegetation had been established. The prohibition was based on a 1998 NIST study that, on the surface, seemed to indicate that vinyl-framed windows are likely to melt and structurally fail when subjected to the heat of exterior fire, allowing glass to fall out and admitting fire and embers into the home, dooming it to certain destruction. An entire class of product was being excluded from a large market segment--obviously damaging to vinyl window manufacturers but with far-reaching implication to all window producers.
AAMA essentially counseled a time-out and began to work with these fire jurisdictions to forestall unwarranted reaction and to clarify requirements for fire resistant fenestration products for use in WUI areas. The resulting dialogue between the window industry and fire officials produced some constructive results. First, closer examination of the cited NIST study suggests that a wildland fire was seen as generating temperatures as high as 1,100º F. This means that all framing materials would be affected. The study also showed that there was no scientific basis for assuming the heat source used in the testing was representative of the nature of the heat flux actually delivered by wildland fires. In addition, widely accepted flame spread data is applicable to an interior environment only; none exists regarding flame spread rates from the exterior to the interior. Clearly, more research was in order.
To accomplish this, AAMA contacted the University of California Forest Products Laboratory (UCFPL), which was concurrently conducting FEMA-sponsored fire performance tests of exterior building components as a step toward the development of minimum building codes. The UCFPL tests considered the results of direct flame impingement (simulated by exposure to a 300 kW burner) and exposure to the heat likely to be generated by a medium sized ornamental plant burning directly below the window (simulated by exposure to a 150 kW burner for three minutes).
The latter experiment was to test for self-sustained combustion of window frames. The windows were also tested to failure under exposure to the 150 kW burner as a means to simulate the burning of larger plants or the building’s siding just outside the window. A full range of frame materials for single- or double-hung windows with double-pane glass was chosen for the tests. Interestingly, it was the glass that failed first in each and every 300 kW test, with double-pane glass outlasting single-pane significantly. Also, none of the framing materials sustained combustion after the three-minute 150 kW test, although the glass again broke in three cases while the burner was on. In the test-to-failure mode, the glass was once more the weakest link, and even where frames did fail, there was no discernable difference among vinyl or clad wood.
In the test-to-failure mode, the glass was once more the weakest link, and even where frames did fail, there was no discernable difference among vinyl or clad wood.
The conclusion: double-pane glass, especially when made of tempered glass, is the principal line of defense and the clear choice for resistance to flame impingement. Since glass failure, not frame failure, was the primary failure mode, it is not possible to discern performance differences among the frame materials tested. Furthermore, researchers pointed out that, while the possible softening and sagging of vinyl frames under exposure to radiant heating (as suggested by the NIST study) were not within the scope of the UCFPL tests, the sensitive element of a double-hung window is the horizontal interlock where upper and lower sashes meet (i.e., meeting rails) and that aluminum-reinforced frames (especially the meeting rails) “may protect against this kind of failure.”
In other words, all window products failed when subjected to high heat levels, not just vinyl, and the performance was dictated by glass type rather than by window type or the specific framing material. AAMA member company eyewitnesses were present at the UCFPL testing, and the AAMA team was able to rapidly document these new findings and present the data and a videotape of the tests. Faced with this evidence, San Diego County agreed to reconsider the vinyl prohibition within the 100-foot buffer zone if the manufacturer could indeed show that a particular vinyl window product would not distort or sag and develop gaps that could admit fire and embers. It was agreed that this meant the product must meet five specific criteria:
- 1. Frame and sash with welded corners.
- 2. Metal reinforcement in the interlock area.
- 3. Frame and sash profiles certified under the AAMA Vinyl Profile Certification Program.
- 4. Window unit certified and labeled to ANSI/AAMA/NWWDA 101/I.S.2-97 for structural requirements.
- 5. Glazed with annealed or tempered insulating glass.
Documentation was provided to show how AAMA profile and window certification test reports or specifications can be used to identify windows with metal reinforcement. While the hasty prohibition of anything vinyl was forestalled in this one case, the collaboration with the fire officials pointed to the need for a window fire rating specification and an easily identifiable label. AAMA therefore, in fall of 2001, commissioned a Fire Ratings Task Group to work on a possible specification and labeling program to identify windows with those properties and performance acceptable for use in WUI locations. The committee continues to foster dialogue with the fire protection community.
Anyone familiar with a campfire knows that,unless it has been thoroughly extinguished, flames will pop up from the ashes here and there for some time after the main blaze has quieted. It seems to be an apt analogy for this issue, as well. While the performance of building envelope products under exterior fire conditions remains a growing area of interest, there are no concerted national or even statewide efforts to ban certain products, particularly vinyl windows. The issue does seem to have a way of popping up at more or less random venues, however.
This past summer, for example, the city of Prescott, AZ, understandably skittish after dodging the bullet of the massive Rodeo fire, issued an ordinance stating flatly that “Vinyl frame windows in exterior walls are prohibited.” AAMA’s response, similar to that mounted to the San Diego issue, succeeded in having the requirement tabled pending further information. In addition, published information also occasionally surfaces with unnecessary prohibitions against vinyl. For example, in the “Building Materials and Components” chapter of an in-depth booklet published by the State of Colorado, “Firewise Construction,” it says that “vinyl windows seldom ignite [but]…the problem is that vinyl frames melt and structurally fail, allowing the glass to fall away. They are not a good choice…” Again, this has not been proven, especially for vinyl window product designs with metal-reinforced frames. Much of this kind of excessive response is due to anecdotal evidence on the part of fire officials, who ‘know’ that radiant heat from an approaching fire, fire-driven winds and impacts of airborne debris ‘must’ be sufficient to cause the anticipated effects. But this borders on hysterical misapplication of information and is not based on actual research data encompassing all typical design options.
The code development arena is perhaps somewhat less frenetic on the subject, with final language of the 2001 Urban-Wildland Interface Code (one of the so-called “I” codes), due to take effect in 2003, including the following language for Exterior Glazing: “Exterior windows…shall be tempered glass, multilayered glazed panels, glass block, or have a fire-protection rating of not less than 20 minutes. Unapproved plastic or vinyl assemblies shall not be used.” How many states will adopt this particular code remains to be seen, but those already embracing the IBC may be well along that path.
This may seem rather innocuous on the surface but there is concern that the ambiguous language leaves the door open for building officials to exclude products based on framing. Also, it would be difficult to find any 20-minute fire-rated product with standard glass, as most feature wire glass or other costly alternatives. Through AAMA, the industry will be taking a much more alert “heads up” position on future “UWI” code change proposals. Meanwhile, most fire protection organizations are fortunately still focusing on landscaping around the WUI home and creation of a “defensible” buffer zone as of much greater importance than concentrating on building products. Those which do mention window products recommend using double glazed windows, preferable with tempered glass, as well as other techniques such as closable, non-combustible shutters.
PURSUING MORE RESEARCH
More research is needed to properly characterize window behavior when exposed to all aspects of a wildfire and to develop test methods as the basis for any future certification and labeling program. Some issues to resolve are, while window failure seems to be a glass problem more than a framing material problem per se, it would be useful to better quantify the difference between tempered and annealed glass. More accurate simulation of the wildfire environment is also called for. While ground-breaking in itself, the UCFPL tests did not consider issues other than direct fire exposure, such as:
- Simulation of winds preceding a fire to determine if the design pressures are higher than those assumed for climactic
- wind loads.
- Simulation of radiant heat effects due to nearby burning objects.
- Simulation of impact of airborne debris of typical size, type and speed.
Meanwhile, we need to get a handle on the mechanism by which misconceptions about vinyl products are spreading.
Is it through local fire chief meetings? The NFPA? Other associations? The AAMA Fire Ratings Task Group is pursuing a four-step program to develop a performance standard and product certification, but must do more performance research as well:
1. Catalog existing performance research information. So far, little has been found upon which to base the development of a standard. The task group plans to issue (likely at AAMA’s early-February Annual Meeting) a white paper or Technical Information Report (TIR) on the current status of research data. The report would be available publicly by spring of 2003.
2. Become more proactively involved with the WUI fire protection community. We envision a “road show” presentation including video clips of the UCFPL tests for delivery to various interest groups, associations, etc. The idea is to educate and open a dialogue to pre-empt municipalities from acting rashly by prohibiting entire product types.
3. Develop real performance data on design pressure and the effects of radiant heat, direct flame, airborne debris, etc.
4. Develop a test protocol to include all these elements as basis for performance definition, and certification. The task group will seek funding for the research portions of the project, possibly from federal sources.
It is certain to be a long process, with several years likely to be required to develop the sort of data upon which we can base rational decisions. In the meantime, AAMA will watch for situations where we can contribute positively to building codes and WUI fire control recommendations.