By Glenn P. Corbett
For those of us who have been around awhile, we remember the discussion of fire behavior in our basic firefighter training programs of the 1960s and 1970s. We were taught the three sides of the fire triangle (soon to become a tetrahedron and gain a fourth dimension, the mysterious uninhibited chain reaction) and the three methods of heat transfer. We were shocked about the emerging understanding about the dangers of smoke, particularly from plastics. We learned about compartment fires, flashover and backdraft, and the different methods of fire attack. We also estimated fire loads in terms of pounds per square foot of wood or synthetic materials using British thermal units (Btus) per pound.
We conversed in a language that had very few numerical terms: temperature [Fahrenheit (°F)], water volume (gpm), and water pressure (psi). It was that simple.
In recent years, significant advances have included computer modeling of fires, used in building design and fire investigation, and technical firefighting research sponsored by the National Institute of Standards and Technology (NIST) and Underwriters Laboratories. We need to update our basic glossary so the entire fire service can participate in this rapidly changing fire landscape.
Building on the Basics
Although our current foundation of fire behavior knowledge rests on concepts such as the fire tetrahedron and the three (or four, depending on whom you ask) stages of a fire, our fire “descriptors” have tended to be qualitative, or non-numerical-e.g., flashover is described as the simultaneous ignition of all the contents of a room. Today, we should build on this foundation and develop a quantitative or numerical understanding of fire behavior.
Fire behavior is often expressed quantitatively using metric units, not English units; temperatures are measured in Celsius (°C); lengths are measured in meters; and so on. Most technical reports and nearly all computer models use this system. Although it is certainly possible to use English units of measure, it is very uncommon. For example, if we are describing a fire test in a room measuring 12 × 10 × 8 feet that has a ceiling temperature of 1,200°F, a computer fire model would use room dimensions of approximately 3.7 × 3 × 2.5 meters with a ceiling temperature of 650°C.
As we know, the three methods of heat transfer-conduction, convection, and radiation-play a critical role in the analysis of fire spread and growth, including the onset of flashover. We understand how they work: conduction through molecular energy and motion, convection through fluid motion, and radiation through electromagnetic energy. Each method of heat transfer involves the transmission of energy from a high-temperature object to a low-temperature object; energy is typically defined as the ability to do work. Using the language of modern fire protection, we can quantitatively understand how each of these affects fire behavior. Although this article will not use formulas to perform calculations (that is the job of fire behavior textbooks and semester-long courses), we will review the terminology associated with fire dynamics in terms of heat transfer and fire power.
When studying heat transfer, it is important to understand that we are interested in exactly how much energy is actually transferred from one location to another, such as convective heat transfer from a flame (a fluid) to a solid surface such as a wall. We quantitatively express this transfer of heat as a heat flux, essentially the specific amount of heat energy per unit area. The variable q˙” is used to express heat flux for any of the three types of heat transfer.
Now that we know that the variable for heat flux is q˙”, what are the units of measure for the heat energy and area? Typically, heat is expressed in kilowatts (kW), and the area is expressed in square meters (remember, we use the metric system). Therefore, heat transfer is usually presented as kW/m2 (kilowatts per meter squared).
What does this mean, and how does it apply to firefighting? Of course, the effects of heat transfer are all around us. Although we can’t really “see” heat being transferred, we can see the ignition of an exposure from radiant heat, the ignition of cardboard boxes against a hot steel bar joist through wall conduction, and the ignition of an interior finish through convective heat transfer (photo 1).
|(1) The effects of radiant heat are apparent on the top of the roof of an exposed mill during a 1984 conflagration in Passaic, New Jersey. The q˙ ” is most likely at least 15-20 kW/m2; the smoking roof monitor at the top of the building eventually autoignited within minutes under the radiant heat flux. The entire building later became fully involved and was completely destroyed. (Photo by author.)|
Quantitatively, the values of q˙ “can establish whether something will ignite or not ignite and the damage it can do. For example, particleboard will autoignite (ignite without a spark or flame) at approximately 20 kW/m2 after 250 seconds. Exposed skin will burn at 4 kW/m2. Flashover, defined qualitatively in many firefighting texts as a point in fire development in which there is simultaneous ignition of all combustibles in a room, is defined by some fire researchers as 20 kW/m2 heat flux at floor level coming from the hot ceiling layer above.
A review of online fire research resources provides information on a variety of tests exploring heat flux ignition of many materials. In addition, test reports also provide information on flames, plumes, and ceiling layers as to the heat flux they produce at specific temperatures and conditions.
From a practical standpoint, knowing the quantitative basics of heat transfer allows for the solution of real-world problems that occur from time to time. For example, many years ago, I was presented with a political time bomb, a new gasoline tank farm located in a poor neighborhood near a high school. We had to assess the threat to the school in terms of an explosion (an open air vapor cloud deflagration) and a spill fire in the diked area emitting radiant heat. Our quantitative review of the radiant heat exposure on the school revealed that, given the flame height, the distance of the school, and the fire, the radiant heat flux would be insufficient to ignite combustible materials in the school building.
Knowing the language and numbers of heat transfer will allow you to better understand technical testing reports on your personal protective equipment. Recent thermal tests on and the resultant failures of portable radios, for example, use heat flux as one of the testing criteria.
Consider this: Imagine your helmet is equipped with a radiometer (a device to measure radiant heat) on the top of its crown. Knowing that the critical radiant heat flux for flashover is 20 kW/m2, think about how useful it would be if you had real-time information about conditions approaching flashover. We know the importance of air management during an interior attack by monitoring your air supply, so why not know more about the fire conditions around you? In the future, you may be monitoring radiant heat flux while you are conducting interior firefighting operations.
Today, many fire researchers feel that fire power, measured in terms of the rate of heat release, is the single most important factor in fire development. For firefighters, this variable tells how large and significant a fire we will be dealing with.
As highlighted earlier, we have for years preplanned target hazards in terms of fuel load, specifically in terms of Btus per pound of a particular combustible material over the floor area of an anticipated fire. Today, a better assessment would be to use peak heat release rates. For any fire, the rate of heat release typically varies over time, starting small, growing to a crescendo peak, and finally dying down as the fire burns itself out or as ventilation conditions change. Graphically, this “life of fire” appears to be mountain-like, with the slopes of the mountain steep or more gradual, depending on the material burning.
The peak heat release rate-the single highest point of the “mountain”-is typically used when estimating the fire threat since it is usually the worst case impact of the fire. Heat release rate is expressed as the variable Q˙ (sometimes referred to as “big Q˙ “). It is measured in kW, or megawatts (MW) for larger fires.
Examples of peak heat release rates for some common materials include small wastebasket fires ranging from 4 to 50 kW, Christmas trees of between 3 and 5 MW, and polyurethane sofas at approximately 3 MW.
Computer modeling of actual fires has been conducted using NIST’s Fire Dynamics Simulator (FDS) for almost 15 years. One of the first models produced was that of the deadly 1989 Cherry Lane fire in Washington, D.C., which killed two firefighters. The model was used to simulate the fire, including changes in ventilation conditions during the fire. A graph of the varying heat release rates is shown in Figure 1. Note the rapid rise in the heat release rate once the basement sliding door is opened at 140 seconds (a complete technical report of the fire is available online at http://1.usa.gov/149rCcC).
|Figure 1. Heat Release Rate from FDS Simulation|
FDS and its sister software Smokeview (which provides visualization of the computations performed by FDS) are used today not only to study actual fires but also in the design of new “signature” buildings, particularly in the case of “performance-based” designs that do not follow all prescriptive building code requirements but must meet some level of established performance. That level of performance is tied to a specific anticipated/expected fire size (heat release rate in kW or MW) or series of anticipated/expected fires. Essentially, the fire is being selected and modeled in the computer program. Decisions about egress time, presence/absence of fire barriers, fire suppression and detection system designs, and so forth are all in the balance.
Of course, you can see how important it is to select a realistic worst-case scenarios model. It may be up to you to review and accept a building design based on a particular fire size. From experience, I can tell you that you must be very vigilant in reviewing such “design fires.” Sometimes the design fire is grossly underestimated by designers. Call on trusted experts when you are uncertain about making such a decision.
Such decisions are not easy to make. Years ago, I was involved in the code enforcement review of the Alamodome in San Antonio, Texas. Of particular concern was the egress time to empty the seating area during a major fire at field level. Based on significant research and expert opinion, we decided on a 20 MW fire as the worst-case/realistic scenario. This fire size was used to design the smoke management system in the arena, a particularly important system used to aid evacuation by providing a clear path of travel for fleeing occupants. Ultimately, we required an actual 20 MW fire test (with “seed” particulates to create smoke for visualization) in the arena; propane burners were used with the seed to create a buoyant, smoky fire plume. The smoke management system passed the established test criteria.
How might fire size be helpful in firefighting preplans? It is only a matter of time before preplans and fire models will come together to allow firefighters to run fire simulations in target hazards in their communities. Imagine running the models to see how quickly a fire will develop, how it will travel, and the conditions it will create in a building in your response district. Such simulations will help establish building vulnerabilities, fire flow water needs, and possible firefighting strategies.
How about a little Greek language now? Have you ever seen the Greek symbol α? How about the expression ρ? They represent thermal diffusivity and thermal inertia, respectively. They are two important variables used in studying fire behavior, in particular conductive heat transfer. (Look, I promised Greek in the title of this article!)
In the future, firefighters’ technical understanding of fire behavior will become increasingly important in the fire prevention bureau office and on the fireground. Consider taking a course in fire dynamics. I’m sure you’ll find it useful throughout your career.
GLENN P. CORBETT, PE, is the former assistant chief of the Waldwick (NJ) Fire Department, an associate professor of fire science at John Jay College of Criminal Justice in New York City, and a technical editor for Fire Engineering. He served on the Federal Advisory Committee of the National Construction Safety Team and is a member of the Fire Code Advisory Council for New Jersey. He is the coauthor of the late Francis L. Brannigan’s Building Construction for the Fire Service, 5th Edition; editor of Fire Engineering’s Handbook for Firefighter I and II; and an FDIC executive advisory board member. He is the recipient of the 2013 FDIC Tom Brennan Lifetime Achievement Award.
Fire Engineering Archives