LET’S CALM DOWN THE FIREFIGHTERS

LET’S CALM DOWNTHE FIREFIGHTERS

The following article, “NFBA Tackles Wood Truss Fire Issue,” was published in the April 1992 issue of Rural Builder and is reprinted in its entirety by permission. It was accompanied by an illustration of a charred wood girder with a failed steel beam draped over it.

“The growing clamor to slap a fire hazard label on wood trusses ignores the fact that all structural materials eventually succumb to flames and wood usually isn’t the first.

“After seeing part of a video that links firefighter deaths to collapsed wood trusses, builders attending a seminar at the annual convention of the National Association of Frame Builders also saw a host of statistics which put the hazard potential of wood trusses into perspective.

“ ‘Structural collapse is labeled as the cause of only 5.21 percent of the firefighter fatalities between 1980 and 1989,’ said Kirk Grundahl, a Madison, WI, engineer and technical services director of the Wood Truss Council of America. Grundahl said the statistics were published in Fire Command magazine. Of the 1191 firefighter fatalities during the ’80s, nearly 46 percent were attributed to heart attacks and more than eight percent to falls or being struck by objects.

“ ‘One firefighter death that results from the type of construction is one too many,’ Grundahl said. ‘But we need to see what causes most of the deaths.’ Among firefighters deaths caused solely by collapse or structural failure, 39 might be attributed to all types of wood building materials, compared to 23 resulting from ‘noncombustible’ materials, he noted, citing the same source.

“ All structural materials eventually fail if the fire is hot enough and of sufficient duration,’ Grundahl said, displaying photographs of steel beams hanging limply over charred wood girders still in place. ‘Concrete explodes and steel loses its strength and collapses,’ he said.

“ The fire endurance of wood must be considered along with other elements in the floor or roof system,’ Grundahl said. Tests have shown a fire endurance of 10 minutes for wood joists on 16-in. centers. But two layers of 1/2-in. gypsum wallboard in the ceiling below adds 50 minutes, for a total fire rating of one hour. In fact, multiple layers of anything have greater resistance than the sum of each layer individually,’ he said.

“ Charring,’ he added, ‘actually extends wood’s endurance in fire, because it insulates the interior of the wood member, keeping inner cells cooler than the surface.’ Grundahl urged his audience to become familiar with these and other facts so that they can deal with local challenges to the safety of wood in light construction. ‘Firefighters are powerful people,’ he said. ‘They can influence codes and laws in their jurisdictions.’ And he noted that steel and concrete alternatives contribute to a ‘false sense of security’ because their failures are not always included in the discussion.”

INDUSTRY ATTITUDES

Each building material supplier touts the advantages of its material and the disadvantages of other matcrials. Some people in industry seem to feel that firefighters have no right to information about the hazards of specific materials and types of construction. This sometimes results in personal attacks on those who discuss hazards, calling them “hysterical.” Sometimes the information provided by industry is only half the truth, or it is limited to just one characteristic.

For years the steel industry discussed the effect of heat on steel only up to 1,100°F, at which point the steel is at its normal designed strength. Since the failure range begins at this point, this is rather like telling a person falling from a skyscraper, “You’re doing fine” as he passes the third floor.

In addition, there was no mention of the expansion of steel at this temperature. A normally 100-foot steel member must be 100 feet 9 inches long at 1,000’F. The elongating steel either pushes out the supporting structure, or if the structure resists, the steel buckles.

A letter from a concrete industry representative criticizing an article by Chris Naum on the hazards of concrete under construction totally ignored the catastrophic hazard of mass firefighter deaths in a fire where posttensioned concrete is under construction. In addition, the Prestressed Concrete Institute rebuffed efforts by the National Association of Demolition Contractors to have a permanent designation placed on post-tensioned concrete buildings. Such buildings present different hazards in a fire, and since the industry will not cooperate, fire departments should record and note on preplans the fact that a building is post-tensioned. 1 (See “Cutting Reinforced Concrete,” Fire Engineering, August 1992.)

The industry’ is not prone to pointing out that only a small fraction of concrete is rated fire resistive. Most concrete has no rated fire resistance. This distinction is not made even in some fire service publications.

The Federal Trade Commission penalized the plastics industry for deceptive advertising in that the hazards of plastics were not revealed to the purchasers.

Architectural publications still get ecstatic about the plastic covering of the U.S. Pavilion at Montreal’s Expo ’67 despite the fact that it was totally destroyed by fire in 10 minutes. Fortunately, it was not occupied by hundreds of visitors when the fire occurred.

THE FIRE SERVICE MUST TAKE CARE OF ITSELF

All information from people who have other interests should be taken with a large dose of salt. The fire service must make its own judgments with respect to the hazards of various materials and how they are put together in structures. The way the entire structure is assembled, the connections, may be as important as the material itself.

The picture accompanying the quoted article shows a laminated beam surviving a test while a steel beam is draped over it. Ironically, this photo shows one of the most serious hazards many laminated-wood buildings present for firefighters. True, the laminated wood didn’t burn to destruction, but that is not the only problem. Much laminated timber is supported by unprotected steel columns or is joined with fire-vulnerable metal connectors.

A building with laminated-wood arches was provided with a foamed plastic insulation on the roof. After a total loss, the owners sued the plastic supplier, alleging that the foamed plastic caused the destruction of their otherwise “slow-burning” building. 1 produced pictures showing that the laminated-wood arches fell apart at the metal connections. The suit was dropped.

In addition, the finished appearance of the wood is an important interior decoration feature. Charred wood, however strong, is not acceptable as an interior finish, and restoration is very expensive.

A RECOMMENDED POLICY

From this article and other documents, we can see the arguments the wood industry will put forth to “put firefighters’ problems in perspective,” a public relations term for sympathy soothing syrup.

As a nation we are committed to building most of our structures out of wood. We could say this is a national policy, even though it never was formally expressed. There are many advantages. Wood is a renewable resource. It is easily worked with simple tools by people with little or no training. (Marry Homeowner doesn’t do much welding.) It is a great asset. It has one problem: It is combustible.

The fire service should have a countervailing “policy.” Wooden structures conceal hidden fire and can collapse. It should be the “policy” of the fire service not to be trapped when the wooden structure does its thing. Neither industry nor taxpayers care much about blood and tears. A possibly better justification is that it costs too much to kill and injure firefighters. Every dead firefighter costs the federal taxpayers approximately SI20,000. Dead and injured firefighters cost local government uncounted additional thousands.

It is imperative that every fire department develop operating instructions based on this policy. At least one fire department is being sued for millions for alleged negligence, for going to a defensive mode after several firefighters had been injured by collapsing truss floors. There is always the “expert” who will testify that this is not an “accepted practice.” We don’t have to look far for such dinosaurs. In one statewide officer exam, candidates who “fought” an examination fire problem from the point of view of firefighter safety were denied credit for the question —and the appeal was denied.

FEAR-ALLAYING ARGUMENTS

A number of arguments calculated to “allay firefighters’ fears” are offered in the article quoted above and other material.

Most firefighter deaths are due to heart attacks; only a small percentage of firefighter deaths are due to collapses. So what? What is the connection? How many of the heart attacks occurred while fighting fires in wooden buildings?

Wooden structures can get fire resistance ratings by the application of gypsum board. Comparative fire resistance is measured by time achieved in a limited test. Unfortunately, the use of time ratings almost universally creates the impression in the minds of the uninformed readers that the time is real time in a fire. It appears that persons who definitely know the facts are not averse to propagating this totally false concept. It most certainly does not mean that the structure is safe from collapse for the time mentioned. Nothing is further from the truth.

Note that in the case of combustible test assemblies, the time rating refers to the failure of the structure, usually by collapse. Another much shorter time, “finish rating,” tells when serious temperatures are measured within the test assembly that is between the ceiling and floor. Note further that at one time UL listings called these structures “fire resistive combustible.” Strangely, the word “combustible” has disappeared from most of the listings.

The following lines, included in every edition of Building Construction for the Fire Service, were provided to me by Jack Bono, the recently retired president of Underwriters Laboratories:

“It is erroneous to relate a building’s rated fire resistance to a projected collapse time. The following concept is not correct: ‘The building has a 2-hour rating. If the fire is not controlled in 1 hour and 45 minutes, prepare to withdraw the men and operate from outside.’ Depending upon the fire load, the rate of fire development, and many other factors, such as possible failure of a key structural element or fire barrier, the building may be in distress, or the fire may communicate beyond a fire barrier, in much less than 2 hours. The converse is equally true.

“Since there are so many variables, the limit of confidence one can have is more nearly the concept that a 4hour fire resistive building is more fire resistive than a 2-hour fire resistive building and a 1-hour fire resistive building is less resistive than a 2hour fire resistive building. It is not simple to relate the building ratings to an anticipation of what will occur during a fire, either preplanned or actually in progress.”

Mr. Grundahl apparently thinks it is simple.

The test used, ASTM El 19, may give equal time ratings to combustible structures of wood and gypsum, but the man who developed the concept of applying ASTM El 19 to wooden structures, the famed Dr. S.H. Ingberg of the National Bureau of Standards, now the National Institute of Standards and Technology, warned:

“We must particularly distinguish between the fire resistance of a combustible and a non-combustible structure. While they may be equivalent in time, there is a world of difference in the protection afforded as it concerns the individual building, the hazard to nearby structures and the hazard to life.”’

1 amplify the deficiencies of wood and gypsum structures rated “fire resistive” at length on pages 538-541 of Building Construction for the Fire Service, 3rd ed., Chapter 12. Space here does not allow for full discussion of these deficiencies listed below:

• Today’s fires are much hotter, and high temperatures are reached sooner than in the almost 90-year-old standard fire curve.
• I’he test is conducted on a perfectly built, completely firestopped structure.
• The size of the test structure is
• not realistic.
• The test assumes, most erroneously, that fires burn only upward.
• The test does not simulate fire burning downward into the floor void.
• The test does not simulate fire burning laterally into the void.
• The static load on the test assembly is far lower than the dynamic load of two firefighters making a primary search.
• The test does not provide any penetrations of the gypsum sheath. Structures as built have scores of penetrations.
• The test does not simulate fire starting in the truss void, such as occurs from wiring.
• The test is conducted under negative pressure. This vents explosive and toxic combustion products that accumulate in real-life fires.
• The test does not evaluate the fire-extension potential of piping voids connecting between floors.

One-point failure can’t cause collapse. It is argued that it is not true that a one-point failure can cause the collapse of a truss. Because of “redundant design,” other elements will pick up the load. This might be true if the truss is set up in a test and a single failure is induced. However, in a fire all the redundant elements will be failing at the same time.

The U.S. Forest Products Laboratory (a taxpayer-supported institution for the development of wood) has developed truss frame construction, in which the studs are an integral part of the floor and roof truss: “Truss frame structures are plane structural components. Their design assumes that every member will remain in its assigned position under load. Permanent bracing must provide adequate support to hold every’ truss in its design position and to resist lateral forces due to wind load.”

The bottom chord of a triangular truss providing a long clear span in a commercial building is made up of several pieces of wood joined together with gusset plates. The bottom chord of a truss is under tension. A rope under tension fails if cut in one place. The failure of any panel point (gusset plate joint) is like a cut in a tensioned rope—failure. When one truss fails, the load transferred to other fire-weakened trusses can cause multiple collapses. So much for redundancy.

Other building materials collapse also. Again, so what? The fire service must he concerned about the fire characteristics of all materials. The fact that others fail has no bearing on wood.

Charring protects wood. The charring of wood is said to protect the interior. When a piece of wood is 1 ½inches thick, the loss of substance from charring is far more detrimental than from a heavy timber.

Architects and builders should be educated. This is a diversion, an attempt to put the monkey on somebody else’s back. It might help, but the problem still will be there.

building officials should require better firestopping work. There are no standards for firestopping. Mechanics and inspectors do not understand the necessity of 100 percent perfection in firestopping. If firestopping should, by some miracle, be perfect when installed, there is no protection against future penetrations. If an apartment house is perfectly firestopped along the tenant boundaries, this still leaves the void under and over one apartment on each floor as an undivided explosive gas accumulator.

Sprinklers n ill take care of the problem. The typical 13R sprinkler system installed in multiple dwellings is a partial life-safety system intended to prevent flashover before the occupants can escape. This should work well except for fires that start in the void or extend into the void before the sprinklers can suppress the fire.

A “sprinklered” renovated dormitory at the College of William and Mary was destroyed because of a void fire.

Four Syracuse. New York, firefighters lost their lives in a fire in a fraternity house that had been equipped with a corridor sprinkler system for life safety.

In De Kalb County, Georgia, a fire resulting from a plumber’s propane torch developed in the truss void of a nearly completed three-story apartment house with no contents load. The sprinkler system was operational, but the sprinklers didn’t fuse. A firefighter fell through the second floor into the fire when the truss collapsed. Fortunately, his partner used a hoseline to protect him from the fire until he could be pulled out.

Fire started in a defective ceiling light fixture in an Orlando, Florida, first-floor apartment of an unsprinklered three-story, brick-veneered, plywood-sheathed, truss-floor apartment house. The fire extended to upper floors on the outside of the wood sheathing, concealed by the brick veneer. The truss floor of the third floor sagged more than a foot. If a 13R system had been in place, the result would have been the same.

In a Texas fire I witnessed and photographed, an outside rubbish fire extended through the exterior wall into the truss void of an unsprinklered townhouse.

A 13R sprinkler system is a partial sprinkler system. This is not to denigrate the usefulness of sprinklers in saving the lives of occupants. There have been many successful sprinkler operations. Balancing the potential structural failures with the positive life-safety benefits, I strongly favor the partial system. In fact, I partially sprinklered my house in 1962. However, fire departments should not expect to receive from residential sprinklers the level of structural protection they are accustomed to receiving from standard sprinklers.

No proposals for research discuss the truss void (or trussloft) as a reservoir for toxic explosive carbon monoxide gas. The truss void is an ideal reservoir for carbon monoxide. NIST has told us that CO generation can be as much as 50 times greater in an enclosed void than in the open. Fire departments have not paid as much attention to this problem as they should. The best information I found on this subject was in a “Special Analysis” as part of the NFPA’s 1989 report on firefighter deaths. This lengthy extract is included here because it is important and because the hazard is often unrecognized in the fire service.

“Backdraft is defined as the burning of heated gaseous products of combustion when oxygen is introduced into an environment whose oxygen supply has been depleted due to fire. This burning often occurs with explosive force.

“In general, no obvious signs of backdraft and flashover (e.g., the puffing of smoke out any opening in the building so that the building appears to be breathing; thick smoke and heat with no visible fire; rolling flames across the bottom of a heavy smoke layer) were reported. In several cases, fire fighters specifically reported that smoke was only light as they moved through the building. It is important to be aware, then, that backdraft or flashover can occur with no obvious warning signals. Whenever a fire has been smoldering or burning in an oxygen-starved atmosphere for a long time, there is the potential for large quantities of carbon monoxide and other unburned products of combustion to be present. These may not involve evidence of thick smoke. In larger areas, these dense concentrations of gases and potential fuel could extend a considerable distance from the seat of the fire to positions where they would have been cooled by the atmosphere. Under the right conditions, these concentrations can be ignited and can flash across an area, creating tremendous radiant heat, generally from the ceiling downward. This type of phenomenon has led to a number of fire fighter deaths over the years.

“The second hypothesis —that the suppression approach made rapid fire development more likely—may have been confirmed in a few incidents where there was inadequate ventilation. In one incident, two fire fighters were killed during the overhaul of a clothing store fire that had not been ventilated either horizontally or vertically. The roof had not been opened and the windows were left intact. It was about 45 minutes after the alarm, and the fire was thought to have been knocked down, when the area where the fire fighters were working suddenly became involved in an intense fire. This may have been due to a buildup of heat and unburned products of combustion above a false ceiling or to the sudden introduction of air into an area of the building where there was still some burning and an accumulation of unburned gases. In other cases, applying a hose stream into an unventilated basement or attic space appears to have introduced oxygen and provided the mixing needed to create a backdraft.

“The ventilation of involved structures is essential in limiting risks to fire fighters operating inside, but the ventilation process can expose fire fighters to the very dangers they are trying to reduce. Fire scene managers must recognize conditions that could cause backdraft, flashover, or flameover and could endanger fire fighters operating inside the structure. It is also possible that these fire development phenomena are not factored into local fire scene planning and decision-making as universally as the special hazard would warrant. A careful and thorough investigation of all abnormal situations should be initiated, whether or not there are deaths and injuries. Encountering unusual heat, smoke, or burning at locations remote from the main fire area could indicate problems. Fire fighters at the fire should use care when dealing with these conditions and, following the fire, attempts should be made to understand why the conditions existed.”5

In the case of truss voids, or the voids created by dropped ceilings in renovated old buildings, you might consider using piercing nozzles to penetrate the closed void rather than pulling the ceiling, which admits a large quantity of oxygen.

The U.S. Forest Products Laboratory’ in Madison, Wisconsin, hits received a grant from the U.S. Department of Agriculture to address critical barriers to advances in structural wood engineering associated with concerns for the fire safety of newstructural members such as trusses. I am happy to note that the Project Summary notes that “the concern of fire fighters for the safety of trusses has been most visibly expressed by Francis L. Brannigan in his National Fire Protection Association book Building Construction for the Fire Service. ”

The Reedy Creek Fire District (Disneyworld, Florida) required the use of fire retardant-treated (FRF) trusses for apartments in addition to a 13R sprinkler system. The industry resists this approach, since FRT wood suffers a 25 percent load capacity penalty. 1 have been told that research on the use of surface fire retardant treatments is underway. This method, if applied also to the underside of the plywood floor, might accomplish much. Improving the structural stability of the trusses will not solve all problems, however, as long as the hazard of the accumulation of carbon monoxide in the “trussloft” remains *

Frank Brannigan is accumulating accounts of fires that involve structural elements of all kinds. particularly instances where injuries or near misses or adequate preplanning was involved. Write to him at 204 / Daylily Road, Port Republic, AID 20676.

Endnotes

1. Brannigan, Francis L., Building Construction for the Fire Service, 3rd ed., Chapter 8, National Fire Protection Association.
2. Brannigan, p. 250.
3. Fire Protection Through Modern Building Codes, American Iron and Steel Institute, 1961 ed., p. 123-
4. Brannigan, p. 104, quoted from “Truss Framed Construction,” NAHB Research Foundation Inc.
5. Washburn, A., et al. “1989 Fire Fighter Fatality Report—Special Analysis: Deaths Due to Rapid Fire Progress in Structures.” Fire Command, June 1990, p. 35.