Extreme Fire Behavior: Smoke Explosion

April 17, 2006
In the case of both backdraft and smoke explosion, smoke is the fuel. However, the other sides of the fire triangle are quite different.

To this point, we have examined the extreme fire behavior phenomena of flashover and backdraft. Flashover, transition of a compartment fire from growth to fully developed stage, is a heat driven event that depends on the presence of adequate fuel and oxygen. A backdraft is quite different. If ventilation is limited (as it usually is), the available air supply will limit the rate of combustion. As the fire consumes atmospheric oxygen and temperature within the compartment rises, backdraft conditions may develop. Introduction of air into a compartment filled with extremely hot, fuel rich smoke may result in a backdraft.

Many old texts dealing with basic fire behavior or ventilation used the terms smoke explosion and backdraft interchangeably. However, smoke explosion or fire gas explosion and backdraft are quite different phenomena. In the case of both backdraft and smoke explosion, smoke is the fuel. However, the other sides of the fire triangle are quite different. A backdraft requires a high concentration of fuel gas/vapor, low concentration of air, and temperature above the ignition temperature of flammable products of combustion and pyrolysis produces. On the other hand, a smoke explosion requires a mixture of fuel (smoke) and air within the flammable range but will be below the ignition temperature of flammable products of combustion and pyrolysis products (see Figure 1: Explosive/Flammable Range). If the fuel/air mixture had reached its ignition temperature, it would already have ignited.

If a source of ignition is present, the fuel/air mixture will ignite explosively as illustrated in Figure 2. In this incident, a smoke explosion occurred while companies were making an indirect attack on an attic fire in this vacant structure (more on this in a bit). As shown in the "after" photo, the overpressure from the smoke explosion blew out a section of the roof. The firefighters working on the ladder were below and to the side of the opening and were not injured.

Factors that influence the violence of a smoke explosion include the extent to which the structure confines the fuel/air mixture and how close the concentration of fuel and air is to a stoichiometric mixture (ideal for complete combustion). The more confined and closer the concentration is to stoichiometric, the greater the violence of the explosion.

Smoke from an underventilated fire can flow through leakage in a structure to collect in concealed spaces or other compartments within the building. Remember, smoke is fuel! If smoke is present, even if cool and well away from involved compartments there is potential for a smoke explosion. Karlsson and Quintiere (2000) observe that this phenomenon is "seldom observed in enclosure fires". However, while infrequent, the conditions required for a smoke explosion can develop within a structure and present a significant threat to firefighters as illustrated by the following case study.

Case Study Method

In previous articles on extreme fire behavior, the case study method was presented as an excellent approach for developing your knowledge and understanding of fire behavior. Just to review how to approach a case study: Read the questions to be answered first, this provides you with a framework for understanding the information presented. Second, read the case to get an overall understanding of the incident. Last, examine the incident in detail to answer the questions posed at the start of the case. For additional information on the case study method, please refer to Extreme Fire Behavior: Backdraft and Extreme Fire Behavior: Flashover.

Case Study

This incident involved a fire in a tri-level townhouse that resulted in the deaths of a firefighter and company officer. The mid-afternoon fire occurred in the unit located on Side Delta (see Figure 3). This case study will focus on fire behavior related aspects of this incident. However, this case provides an opportunity to learn a number of other important lessons (see the NIOSH and NIST reports for additional information on the incident).

Configuration: This 1600 ft2, tri-level, wood frame apartment building had two levels above ground and a daylight basement (partially above grade on Side C). The first floor was comprised of a single room, divided into living and kitchen areas. A single bedroom and bathroom were located on the second floor.

Due to the roofline of the structure (see Figure 3), there were substantial void spaces behind knee walls on the Alpha and Charlie sides of the second floor (see Figure 4). While NIOSH Report F2005-13 emphasizes the void space hazards of this type of truss, similar voids are likely any time a finished room is located directly beneath a pitched roof.The area of fire origin was behind a knee wall (storage area) on floor two (see Figures 4 and 5). The interconnected void spaces behind the knee walls and above the ceiling permitted the fire and hot smoke to surround the rooms on floor two.

Fuel Profile: Contents were typical of a residential structure. NIOSH report F2005-13 made no specific mention of the contents of the attic storage areas.

Ventilation Profile: While not specified in NIOSH report F2005-13, truss spaces are equipped with roof vents, typically providing 1 ft2 of vent area for each 150 ft2 of attic floor area. These building vents would have potentially provided limited air supply for fire development within the storage area and void spaces.

It is unknown if either exterior door on floor one was open before the arrival of the fire department. However, Engine 1 the first arriving company opened the door and made entry through the doorway on Side C with a hoseline. At approximately the same time, the Incident Commander observed a small amount of flame from the roof above the door on Side C (see Figure 6).

The Captain from Engine 1 tasked a place a positive pressure blower at the door on Side A. Engine 2 stretched a line to this doorway, creating a second opening on floor one. NIOSH Report 2005-3 also discusses the Captain's intention to establish a horizontal exhaust opening on floor two, Side Delta, but found no windows at this location. The report mentions that he observed a skylight on Side Charlie (see Figures 5 and 6). The report did not specify if the Captain vented the skylight.

Fire Development: From its point of origin in the storage area on Side Charlie (Bravo/Charlie corner of the unit), the fire extended throughout the storage area and void spaces on Side C, above the ceiling, and on Side Alpha. Even with the limited ventilation provided by the attic vents, the high wind (30+ mph blowing from Side Charlie) may have accelerated fire development. Fire behavior indicators observed by the first arriving companies included a small amount of flame from the roof peak near the chimney on Side Delta (see Figures 3 and 6) and light colored smoke seeping from the roof shingles and the attic of Exposure Bravo. A short time later flames were observe above the door on Side Charlie (intersection of the involved unit and Exposure Bravo). Engine 2 observed heavy brown smoke filtering down the stairwell as they extended a backup line to the door on Side Alpha. Approximately eight minutes after Engine 1 arrived on scene, an explosion blew the crew from Engine 2 out the door on Side Alpha and caused the window on floor one, Side Delta to fail. Fire and black smoke was blowing out the door on Side Alpha and window on Side Delta. Crews entering the building to attempt a rescue of the crew from Engine 1 encountered fully developed fire conditions on floors 1 and 2.

Initial Tactical Operations: Initial response to this incident was two engines (one with a a Captain and two firefighters and the other a Lieutenant and three firefighters), a truck (Lieutenant and two firefighters), and three additional personnel (Lieutenants) arriving in personally owned vehicles (POV). One of the Lieutenants arrived in his POV and assumed Command just as Engine 1 arrived on scene. Initial reports from bystanders were that children were trapped on floor two. A firefighter (company not specified in NIOSH report F2005-13) placed a PPV fan at the door on Side Alpha. Engine 1 deployed a 1-3/4-inch hoseline through the doorway on Side Charlie to support primary search. Engine 2 stretched a second 1-3/4-inch line through the door on Side Alpha and onto the stairwell to back up the crew from Engine 1 while Truck 1 was setting up (tactical assignment not specified in NIOSH report F2005-13). (Figure 7: 12 Minutes on the Fireground)

Other Incidents Involving Smoke Explosions

Given the relative infrequency of this phenomenon it is useful to examine several other incidents involving smoke explosions. The first occurred in Edinburgh, Scotland, February of 2002 and the second in Gresham, Oregon in February of 2005.

In February 2002, the Lothian & Borders Fire Brigade responded to Sleigh Drive for an apartment fire with persons reported. On arrival, the officer on the first pump observed fire showing from one window on floor one of an ordinary constructed apartment building (brick veneer over concrete block). Crews extended hoselines to attack the fire and support search in the apartment above. The crew working above the fire reported finding a gap between the floor and exterior wall that was permitting smoke to pass from the fire apartment to the upper floors. Shortly after the crews working on the upper floors exited the building to replace air cylinders in their SCBA, an explosion rocked the building. The force of the explosion injured several personnel, falling debris struck several firefighters outside the building, and the crew working in the involved unit received burns due to a dramatic increase in fire intensity and loss of water supply in their hoseline.

In this incident, smoke traveled from the involved unit to the apartment above through a gap between the floor and exterior wall and eventually reached its flammable range. Fire extending to floor two through a void space ignited the flammable fuel/air mixture resulting in a smoke explosion. The violence of the explosion caused failure of a substantial portion of the brick veneer and part of the block wall on floor 2 as well as a partial collapse of floor two.

In the case study as well as the incident on Sleigh Drive, the conditions required for a smoke explosion developed due to normal fire progression and the structural characteristics of the building involved. Conditions in the next incident developed quite differently.

Gresham Fire and Emergency Services responded to a fire in a vacant dwelling in mid-February 2002. The officer of the first arriving company observed dark gray smoke showing from the attic of a two and one-half story, wood frame structure. As additional companies arrived, companies deployed 1-3/4-inch handlines to control the fire on floors 1 and 2 in support of a quick primary search. After the search was completed and all clear, the Incident Commander withdrew companies from the building and ordered an indirect attack to control the increasing volume of fire in the attic. Shortly after initiating the indirect attack from the gable ends, the building was rocked by a violent explosion that blew out a large section of the roof in the center of the structure (see Figure 2).

After its initial effect (large volume of steam production), the indirect attack became less effective. Application of water in the indirect attack cooled the hot smoke in the attic, but also introduced a large volume of air, which mixed with the smoke creating a flammable mixture. Fire burning up from floor two provided the source of ignition, igniting the fuel/air mixture. Companies were withdrawn outside the collapse zone and shifted to a defensive strategy.

Study and Discussion Questions

Use the information presented in the case to answer the following questions.

  1. Was extreme fire behavior involved in this incident? If so, what type of event happened? Use the fire development curve illustrated in figure 9 to work out your answer. Consider the fire burning in the void spaces. Where on the fire development curve do you think it was? What happened after the explosion? (Figure 9: Fire development curve)
How did the fire develop and what factors influenced the occurrence of the extreme fire behavior phenomenon? Give some thought to the potential effect of the 30+ mph wind from Side C. How might this have affected fire development in the void space? What effect do you think applying positive pressure from Side Alpha may have had on fire behavior? What cues were present that may have indicated potential for rapid fire development? Given the limited resources available, what options did the first arriving companies have to address the need for rapid primary search in conjunction with the potential for rapid fire development? How would this differ with the company staffing level and apartment fire assignment used by your department? Compare and contrast this case study with the Keokuk (flashover) and Blaina (backdraft) or events in your own experience. What aspects of these incidents were similar? Which were different?


  • Karlsson, B. & Quintiere, J. (2000). Enclosure fire dynamics. Boca Raton, LA: CRC Press.
  • National Institute for Occupational Safety and Health (NIOSH) (2000) Death in the line of duty, Report F2005-13. Retrieved March 12, 2006 from http://www.cdc.gov/niosh/fire/pdfs/face200513.pdf
  • Gollogly & McGill (n.d.). Case study: Sleigh drive Edinburgh presentation and trainer's notes. Hertfordshire, United Kingdom: Hertfordshire Fire and Rescue Service.

Voice Your Opinion!

To join the conversation, and become an exclusive member of Firehouse, create an account today!