By Sean DeCrane
We Must Do Something
What is our work environment? Some may believe that the station is our work environment. in I do not believe that this is accurate. I would refer to the station as our staging area. The station is where we report for work, prepare our equipment, train, eat our meals, and study educational materials–all in preparation for our assignment. Our assignments require us to use the skills we have been taught and hone them so they can be applied when we respond to an incident. The fireground is our work environment, the structures we to which we respond and in which we operate. What dictates how these buildings are designed, constructed, and maintained? The building and fire codes.
When speaking to one of the representatives from Green Bay, who was receiving reports from home about the above mentioned incident, one of the first bits of information we received was about the type of construction. Indications were that the collapse occurred in a newer home with an engineered floor system. The members had done the right thing: as they entered the structure: They tapped the first-floor system, getting feedback that suggested a sound base. It felt solid as they progressed just a short distance; then suddenly it collapsed without warning, sending the two firefighters into the basement, severely injuring Firefighter Brinkley and killing Lieutenant Wolff. There was a feeling at the IAFF convention that we had to address this issue of the engineered products that our firefighters were beginning to encounter. In fact, when the National Institute of Occupational Safety and Health (NIOSH) issued its report (NIOSH 2006-26), it recommended that “building code officials local authorities having jurisdiction should consider modifying the current building codes to require that lightweight trusses be protected with a fire barrier on both top and bottom.”1 Why limit it to local building codes? Why not take it to the national stage? The effort began.
Why Is This Permitted?
R501.2 Requirements. Floor construction shall be capable of accommodating all loads according to Section R301 and of transmitting the resulting loads to the supporting structural elements.2
The issue missing in this language is fire performance. There is no requirement for the floor to perform to a prescribed safety level when exposed to fire. How is this possible in a code that identifies firefighter safety as a concern for code change proposals in its intent?
101.3 Intent. The purpose of this code is to establish the minimum requirements to safeguard the public health, safety and general welfare through structural strength, means of egress facilities, stability, sanitation, adequate light and ventilation, energy conservation, and safety to life and property from fire and other hazards attributed to the built environment and to provide safety to firefighters and emergency responders during emergency operations.3 (underline added)
(2) (Photo courtesy of Georgia Pacific XJ-85.)
The initial effort to meet the NIOSH recommendation of a fire barrier code change was submitted for the 2009 IRC edition requiring a fire-rated barrier on all floor systems. The ICC committee, which is comprised of a number of National Association of Home Builders (NAHB) representatives, was adamantly opposed to a fire barrier system because of difficulties of compliance and additional costs in construction. This proposal would have required a fire barrier, which would require that all through penetrations be protected, including all crawl spaces.
The revised proposal did have some flaws; it was debated at great length in Minneapolis and would also have required the fire protection on all floor systems, including traditional sawn 2 × 10 lumber. Although it was met with resistance from the building industry and some building officials, there was a great amount of support from fire officials and others with concerns about firefighter safety. We believed we had a voting advantage because there was a large representation of the fire service in attendance to vote in support of residential sprinklers. Since the proposal to protect lightweight construction was the first code change proposal after the sprinkler vote, it looked as if it might pass. Unfortunately, many of the fire service representatives in attendance did not realize this safety issue was being heard or did not realize its importance. They left the meeting room after the sprinkler vote. After an extensive debate, the vote was cast, and although more than 700 people voted in support of the code change proposal, it was not enough to overturn the committee’s disapproval. The fire service needed a two-thirds majority. We fell short by approximately 50 votes. Fire service members leaving and a lack of data appeared to limit the cross-over votes needed to override a committee decision.
Between the ICC 2009 and 2012 code cycles, we worked on creating a revised code change proposal. During this time, three reports were released that began to give us the technical data to support the effort to require the protection of lightweight engineered floor systems. Many have asked if this code change is still needed now that the new one- and two-family residential sprinkler requirements have been approved. The problem remains that there are a number of states and local jurisdictions that have refused to accept the residential sprinkler requirement or have outlawed the requirement of residential sprinklers. The fire service knows that the first line of defense is a sprinkler system: If we can limit the fire growth, we cut off its impact on the structural load capabilities. We decided to move forward as we gathered our data.
Tyco Industries commissioned a series of tests (or demonstrations, since a furnace ASTM E119 test format was not used) on engineered I-joist floor assemblies. The demonstration consisted of a typical living room with a couch, chair, table, etc., and a small wastepaper basket next to the couch. The fire was initiated in the waste can and permitted to spread and grow to a flashover level and on to collapse. In a couple of demonstrations, they experienced an approximate nine-minute to 11-minute collapse time from ignition. Time to flashover varied, but it was in the four- to six-minute range.4
To demonstrate that this issue is not just about firefighter safety, we included research by the National Research Center of Canada, which studied the impact of engineered lightweight construction in a basement fire scenario and its impact on occupant egress. Various test scenarios showed situations in which the floor assemblies collapsed while the occupant theoretically had a tenable atmosphere that could have permitted self-rescue.5
The coup de grâce was a series of tests/demonstrations conducted by Underwriters Laboratories in conjunction with the International Association of Fire Chiefs (IAFC), Michigan State University, and the Chicago (IL) Fire Department (CFD) with grant money awarded though the Department of Homeland Security. A number of tests were conducted using a test furnace and temperature parameters of the ASTM E119 test without 100 percent loading but with loading typical of a residential occupancy. The performance of a protected floor assembly was compared with that of an unprotected floor assembly. All the parameters of these tests were detailed in “Structural Collapse: The Hidden Dangers of Residential Fires,” (James M. Dalton of the CFD and Robert G. Backstrom and Steve Kerber of Underwriters Laboratories, Fire Engineering, October 2009)
It is important to note the performance of the unprotected lightweight engineered I-joist assemblies compared with other assemblies. Test results demonstrated that unprotected traditional sawn 2 × 10 lumber construction began to collapse approximately 18 minutes and 30 seconds after ignition time. The unprotected engineered lightweight I-joist assembly showed a wide area failure at the 6-minute mark–a marked difference in performance. Although the results of the Tyco demonstrations and the UL tests seem to be different, if we factor in the time to flashover in the Tyco demonstrations, they are almost identical to the furnace-controlled post-flashover tests at UL.6
(4) (Photo courtesy of Underwriters Laboratories.)
When these floor assemblies were protected with ½-inch gypsum board, the performance increased dramatically, adding an additional eight to 10 minutes of performance. The fire service delegation wanted to return the floor assemblies to the traditional performance levels, so we proposed requiring 5/8-inxg gypsum board while recognizing a number of suggestions from earlier efforts, including a trade-off for residential sprinklers systems.
We took the newly released data and created a revised proposal for the ICC 2012 Code Cycle held in Baltimore, Maryland, this year. During the code action hearings, this fire protection proposal was submitted to the IRC Code Development Committee. Shortly before the hearings began, a collective representation from the fire service and industry consisting of the International Association of Fire Fighters (IAFF), the American Forest & Paper Association, and NAHB met to discuss a possible compromise to the proposal. We agreed on a compromise to submit to the committee chair for approval. This request was denied on technical grounds, and it seemed as if we were back to square one.
R501.3 Fire protection of floors. Floor assemblies, not required elsewhere in this code to be fire resistance rated, shall be provided with a ½ inch gypsum wallboard membrane, 5/8 inch wood structural panel membrane, or equivalent on the underside of the floor framing member.Exceptions:
1. Floor assemblies located directly over a space protected by an automatic sprinkler system in accordance with Section P2904, NFPA13D, or other approved equivalent sprinkler system.
2. Floor assemblies located directly over a crawl space not intended for storage or fuel-fired appliances.
3. Portions of floor assemblies can be unprotected when complying with the following:
3.1 The aggregate area of the unprotected portions shall not exceed 80 square feet per story.
3.2 Fire blocking in accordance with Section R302.11.1 shall be installed along the perimeter of the unprotected portion to separate the unprotected portion from the remainder of the floor assembly.
4. Wood floor assemblies using dimension lumber or structural composite lumber equal to or greater than 2-inch by 10-inch nominal dimension, or other approved floor assemblies demonstrating equivalent fire performance.
Although the language is not perfect, the fire service was successful in incorporating in the residential code language that requires the protection of lightweight construction, whereas before we had no requirements at all. This language is scheduled to go into effect when states and local jurisdictions adopt the 2012 IRC Edition, although a number of states and jurisdictions are looking at adopting it at an earlier time in their cycles to provide greater firefighter safety.
The fire service needs to continue to measure this code change proposal and any local adoptions. There are also groups who will continue to conduct tests in various formats to measure the performance of the code-change language and evaluate the performance. Underwriters Laboratories has established a work group of various stakeholders to continue the discussions evaluating floor systems and proposed protection systems. I am UL will publish a report and a media article when the results are available and the time is right.
I would like to thank the representatives who participated in the discussions that ultimately led to the protection requirement (in random order): Jack Murphy, Ozzie Mirkhah, IAFC Fire & Life Safety Section (IAFC-FLSS), Dennis Pitts, Sam Francis and Kuma Sumathipala, American Forest & Paper Association (AF&PA), Ronny Coleman, Bob Davidson, RJD Code Consultants, Jonathan Humble and Farid Alfawahiri, American Iron and Steel Institute (AISI), Larry Wainwright and Kirk Grundahl, Structural Building Components Association (SBCA), Larry Brown, Ed Sutton and Steve Orlowski, National Association of Home Builders (NAHB), Rich Edgeworth, Pete Van Dorpe and James Dalton, Chicago Fire Department, Robert Backstrom, Robert James and Chris Hasbrook, Underwriters Laboratories, Richard Duffy of the International Association of Fire Fighters (IAFF), and those who provided input and consultation during the process.
1.NIOSH; Fire Fighter Fatality Investigation Report F2006-26, July, 2007.
2 International Code Council, International Residential Code 2009 Edition.
3 International Code Council, International Residential Code 2009 Edition.
4 Tyco Fire Suppression and Building Products; A Technical Analysis: The Performance of Composite Wood Joists Under Realistic Fire Conditions, 2008.
5 National Research Council Canada; Fire Performance of Houses. Phase 1. Study of Unprotected Floor Assemblies in Basement Fire Scenarios, 2008
6 Underwriters Laboratories; Report on Structural Stability of Engineered Lumber in Fire Conditions, 9/30/2008.
Sean DeCrane, a 19-plus year veteran of the fire service, is a battalion chief for the Cleveland Division of Fire, where he is chief of EMS services. He is certified as a State of Ohio fire instructor, fire life safety inspector, and firefighter I and II, a paramedic, and is certified in hazardous materials awareness and operations. He is the International Association of Firefighters (IAFF) representative to the International Code Council (ICC). He has served on the 2009 and 2012 ICC International Fire Code Development Committees and on the ICC’s Fire Council and the Underwriters Laboratories Fire Council and is the lead coordinator for Vision 20/20: National Fire Loss Prevention Agenda’s Strategy 5: Fire Service Coordination in the Building Codes and Standards. He is a co-host of the Fire Engineering podcast “Taming Your Work Environment” and is a presenter at FDIC.