Flame SPREAD–or, more properly, fire growth–is not well appreciated by the fire suppression forces. I doubt that many preplans include “expect rapid fire spread.” Flame spread is a factor in almost all huge multiple-casualty fires. The fuel may be surface finishes, hidden building components, innocent looking decorations, or contents.

Huge losses of life in many major fires were due to flame spread. Among those fires were the following: Chicago`s Iroquois Theater (602 died), Boston`s Cocoanut Grove Nightclub (492), Hartford`s Ringling Circus Fire (168), Nevada`s MGM Grand (88), New York`s Triangle Shirtwaist (145), New York`s Happyland Social Club (87), New Orleans Uptown Lounge (34), and San Juan`s DuPont Plaza Hotel (97).

The largest loss of life in a fire occurred in a church in Santiago, Chile, in which 2,000 died. The fire was in sheeting spread high in the church to simulate clouds.

I have a series of spectacular video clips, including the “unbelievable” large-scale reproduction of the Dublin Disco Fire commissioned by the Irish Government at the British Fire Research station, that make flame spread come alive and will banish the concept “That`s something fire prevention worries about.” Few of you have ever seen a major fire start and grow.


The hazard of old combustible tile concealed above a new code-compliant suspended tile ceiling has cost many lives [Building Construction for the Fire Service, Third Edition, (BCFS3), p. 389]. Two Ohio firefighters died recently when a ceiling tile was pushed up after the evident fire had been extinguished. A burst of fire blew down on the firefighters. The entire basement was filled with fire. This report is based on NIOSH (the National Institute of Occupational Safety and Health) Fire Fighter Fatality Investigative Report 98F-06 [] and my conversation with the investigator.

No code specifically prevents installing the new so-called “fire rated” ceiling below the old ceiling. Mark Chubb, fire code coordinator of the Southern Building Code Congress, informed me that this problem is covered in Section 505.3.2: “Existing interior finish must meet specified flame spread requirements based on ASTM-E84 (interior finish test standard) included exposed finishes in concealed places as defined in Sec 505.1.2.” Many such installations are done without a building permit or inspection.

The burning, hidden, low-density tile generates explosive carbon monoxide at about 50 times the exposed level. Think of it as a bomb.

Some recommendations follow:

If you discover the project when it is started, inform the owner and the building department of the hazard in writing.

If the building department won`t act, tell the owner you will have to use a big line, one ton of water per minute, to remove the new ceiling and thus eliminate the enclosure that makes the bomb. The word bomb is very effective.

When examining a building, pop a ceiling tile and look into the void.

At a fire, equate opening a suspended ceiling with pulling a wire from a bomb. Limit the number of personnel. They should be close to an exit and have a defined path of escape. Beware of “light smoke showing.” Too often, it is just the tip of the fire iceberg.


I received a sales letter from a person attempting to get my assistance in promoting his flame-retardant compound. He wrote that it would “give the product a one-hour flame spread.” His confusion between flame spread and fire resistance is not unique. Are you clear on the difference? (Study Chapter 6, “Principles of Fire Resistance”; Chapter 9, “Fire Growth”; and page 296 of BCFS3.)


Fluorescent lighting has made skylights unnecessary. Often, they are eliminated and lightly patched over. What`s wrong with a rule that says no one goes on the roof of such a death trap before adequate exits from the occupancy have been created? We cannot always expect the good fortune of the Anaheim, California, captain who fell through a parallel chord truss roof of an apartment house into the living room and walked out onto the balcony to report that he was okay.

Bear in mind that skylights were almost universal on roofs in earlier construction. The absence of skylights on an older building should raise the suspicion that weak spots exist in the roof. The concept of opening the roof directly over the fire is sound when only the efficiency of the ventilation is considered and roof beams are solid timbers three inches thick. However, the Gravity Resistance System (GRS) of the roof right over the fire is being attacked by the fire. Opening the roof accelerates the fire and weakens the structure. We were taught to extinguish the fire with the least collateral damage. The worst collateral damage by far is the injury or death of a firefighter. [Add to your file on roof hazards “Beware of Hot Spots” by Captain Jerry Tracy of the Fire Department of New York (Fire Engineering, October 1998, p. 119).]


Speaking of skylights, the U.S. government, seeking to cut lighting bills by in-stalling skylights, suffered two fire losses–one huge and one serious–when contractors cut the roofs with torches. The hazard was not considered when planning for the alteration contract. The light bill savings will never cover the losses.

Fire departments should take a proactive stand when renovations for major buildings are planned, particularly roofs. Roofers use torches and often delay the alarm while they attempt to extinguish the fire, or they believe they have extinguished a fire burrowing into the roof void. The fire`s location causes a long delay in getting water to the scene even if the alarm is prompt. The historic U.S. Treasury Building and an historic New York City synagogue (a photo of the synagogue fire appeared on the November 1988 cover of Fire Engineering) were heavily damaged by fires started by roofers` torches. A serious industrial fire was caused when a roofer`s torch ignited a fiber-reinforced plastic (FRP) skylight.1

Business owners must anticipate all possible hazards and cover them in the precontract discussions when construction work is to be done in an operating facility. The 1953 $45 million General Motors transmission plant fire was caused by a contractor who was hired to cut out some overhead conveyors.


Watch for another hazard on peaked corrugated metal roofs. To provide a cheap skylight, corrugated FRP (e.g., fiberglass) is substituted for a steel panel. There is no curb. Darkness, smoke, or snow on the roof would make the plastic invisible. Take your aerial apparatus out on Sunday and look down on roofs. Shoot pictures for the preplan.

If the plastic of FRP burns out, a mat of glass fibers is left. Do not break up this mat in overhauling. You will release glass fibers that will give you a nasty itch. Bear in mind that many people are misled by the “glass” in fiberglass and consider it noncombustible. For example: A large amount of FRP was used in the construction of a waste treatment plant. Toward the completion of construction, the contractor, thinking the plant was noncombustible, cut the fire insurance so far down that the contractor`s recovery after a fire was only 16 percent.2 The contractor sued the FRP supplier. They had the right of “discovery” to look in the supplier`s files to see if the company knew of the hazard but failed to inform customers–a legal no-no. I was engaged as a consultant. My son Vincent, then a law student and a technician at the agency that is now NIST, told them to look into this aspect. The plaintiff received a substantial settlement. Some plastic panels have better fire resistant properties. (BCFS3, p. 38)


“Five hundred thirty-seven BTU are generated for each cubic foot of oxygen delivered to the fire,” it is stated on page 447 of BCFS3. Mark Emory of the King County (WA) Fire Department asked for a further reference, since he could not find it anywhere including, the NFPA Handbook. Btu and cubic feet are no longer accepted scientific terms because they are not metric. However, on page 1-60 of the 18th edition of the NFPA Handbook, in the paragraph headed “Stoichiometric Oxidant,” we find U “so that organic combustibles all release the same amount of heat per unit mass of consumed oxidant. Thus, it is found that, [here, I found one of the remarkably few typos in 2,500 pages: “13L.J” should be 13 K.J. (kilojoules)] 13 K.J. of energy is released for every gram of oxygen consumed in the combustion of oxygen.”

I checked with Assistant Professor José Torero of the Department of Fire Protection Engineering of the University of Maryland, an expert on the subject. The figure I used is correct; however, it should read “U per cubic foot of oxygen consumed instead of delivered. Most fires will entrain approximately 10 times the oxygen they actually consume.”

Most important: Remember that air provided to fire increases the fire in direct proportion to the quantity supplied. In at least one recent fire, pulling the ceiling below a wide-span wood truss roof probably increased the fire to the extent that the roof collapsed with fatal results. Opening a truss void may likely provide fire beyond the capability of the typical small attack line. Use more thermal imaging devices and piercing nozzles for hidden fire to limit the amount of oxygen added.

One Forest Products Laboratory researcher suggested that some connections be made weaker than others so the truss would sag and give warning. There is no justification for remaining on or under wood trusses on fire. You are playing Russian roulette. There is no “20 minute rule.” There is no warning. Collapse is sudden and catastrophic.

If you must “see what is going on,” opening up the side wall and looking while standing on the aerial platform might be a useful technique. It is safer but, of course, the void might blow when opened up. Truss and wooden I-beam buildings are disposable. Firefighters are not.


I watched a chief officer who had ordered his units out of a trussed floor apartment house struggle with a TV reporter`s questions. She was trying to get him to criticize the building code and the county council. Some street reporters want to get famous like Bob Woodward and Carl Bernstein, the Watergate investigators, by finding deficiencies in governmental services, which they often consider to be run by morons or crooks.

I think the commander should be ready with a statement such as this one: “Only a few dwellings ever burn, so the low-cost building methods like those used here are justified. However, when the lightweight 2 2 4s used for trusses are burning and the fire is spreading through the interconnected voids, we must look to the safety of firefighters. A dead or injured firefighter is a great expense to the taxpayers, and the risk is not justified.” This is much more effective than a statement by the chief a week after the fire.



There is no such thing as too much information in this field. A bit you learned years ago can be the key to a real problem, as this “sea story” demonstrates.

My last fire command was a ship fire in January 1949. The Navy had built a number of wooden interisland vessels to haul cargo and passengers in the far Pacific. Some of them are still in service at the Naval Academy to teach midshipmen “station keeping”–by which ships can operate and maneuver safely in company.

Several of these ships were decommissioned and tied up at the finger piers at The Naval Operating Base in Norfolk, Virginia. A bosun`s mate and several seamen were the ship-keeping detail.

I was called early on a Sunday morning to respond to a fire in one of the ships. (I responded immediately to command all ship fires because I was the only one who knew right-hand side is starboard and left hand is port without looking at a chart in his cap.)

I found very heavy smoke pouring out of a hatch on the forward deck. The acting deputy chief had a charged line stretched, and I am sure he was praying for me to respond and take this problem off his hands.

I stuck my head down the hatch and saw fire burning in wiring on the overhead and heard a motor generator running. I put the bosun`s mate and one of my firefighters into SCBA and sent them to shut down the generator, thinking this was probably the fire. No improvement. Suddenly, I remembered a fire my great mentor Captain Tommy O`Brien of 65 Engine in New York (he retired as assistant chief) had related to me: The other platoon had gotten their brains knocked out at a fire in a large (20 to 30 rooms) apartment on 5th Avenue. O`Brien pointed out that if they had gotten on the floor below, they would have understood the layout of the apartment above.

I told the bosun`s mate to take me aboard a sister ship across the pier. We went down a hatch at the stern starboard side, stepped over the shaft, and moved forward on the port side through a narrow gangway. Casually, he pointed out the oil burner and said, “We lit that off about 20 minutes before the fire.” Big light dawns–there`s the fire.

I got back topside and ordered the first line down the aft hatch and up the engine room gangway to the oil burner. The second line went to the captain`s cabin, directly above, with orders to the truck company to remove the bunk. I told them they would find fire coming up through the hull of the ship. (A double-hulled wooden ship is like a balloon-frame building.)

Later, I was a very unhappy boss when the acting deputy chief told me he had hit fire in that corner on arrival. When you are relieved, be sure to pass along all pertinent information.

Listen to the “war stories.” You never know when some nugget of information or operating hint will save the day.

The Cause

When the ships were brought north from the tropics, a typical household gun-type oil burner was installed to provide heat. When these burners backfired into the cramped engine room working space, a “shipalt” was issued to turn the burners around. They then backfired against the skin of the ship. As indicated above, a wooden ship is double hulled and thus resembles a balloon frame building. The acting deputy chief had hit the visible fire but not the fire in the void.


From time to time, I am asked why I left fire suppression, which I really loved. About six months before the fire discussed above, I was directing operations in the hold of a burning ship when it occurred to me that “if I get killed here, a grateful U.S. government will send my wife and kids $80 a month. I must change jobs and get a job where I could just give good advice.”

My transfer to The Atomic Energy Commission as assistant chief, Safety and Fire Protection Branch, Feed Material Operations, came through shortly after this fire. The Federal Employees Injury Compensation law was changed not long after that–from the worst to one of the best. n

In Memoriam

Chief Dave Gratz died of leukemia in November 1998. I knew him for 40 years as a friend and associate. He served as director of fire services for Montgomery County, Maryland; president of the International Association of Fire Chiefs; and an international representative for the National Fire Protection Association.

He was a dynamic proponent of education for the fire service, as distinguished from training. It was at his urging that I wrote Building Construction for the Fire Service when I was teaching the subject in the fire science program at Montgomery College, Rockville, Maryland. I had protested that I was not an engineer. His argument, which prevailed, was, “If you don`t write it, who will?” He was a giant among his contemporaries.


1. Fire insurance contracts require that the property be insured to at least 80 percent of its value to get full recovery of a loss; otherwise, the payments are reduced proportionally.

2. Record, Factory Mutual, Box 9102, Norwood, MA 02052. I worked on this problem in both the Navy and the Atomic Energy Commission and can recommend the article “Working with Contractors” in the FM Record (third quarter 1998). You must, however, build on this material because the stated policy of the Record is, “As a magazine devoted to property conservation Record does not discuss life safety issues.”

FRANCIS L. (FRANK) BRANNIGAN, SFPE, recipient of Fire Engineering`s first Lifetime Achievement Award, has devoted more than half of his 57-year career to the safety of firefighters in building fires. He is well known for his lectures and videotapes and as the author of Building Construction for the Fire Service, Third Edition, published by the National Fire Protection Association. Brannigan is an editorial advisory board member of Fire Engineering.

No posts to display