Posts Tagged ‘structural firefighting’

Shielded Fires Part 2

Thursday, February 12th, 2009

The previous post (Shielded Fires) examined US Navy research on the effectiveness of different nozzle techniques when dealing with shielded fires conducted on the ex-USS Shadwell, the US Navy full scale damage control research facility (see Figure 1).

Figure 1. USS Shadwell


The researchers tested two different methods for controlling flaming combustion overhead while moving from the entry point to a location in a compartment where firefighters could make a direct attack on the seat of the fire. The first method involved use of a straight stream or narrow fog pattern and the second involved the use of a medium (60o) fog pattern directed upward at a 45o angle. In both cases, one to three second pulses were used in application of water into the hot gas layer.

US Navy Findings

Analysis of this series of tests resulted in identification of a number of specific findings related to tactics, equipment, and training. Three of these findings were particularly relevant to the differences between straight stream/narrow fog and medium fog applied to control flaming combustion in the upper layer.

  • Pulsed application with a medium fog pattern directed upward at a 45o angle resulted in less disruption of the thermal layer than use of a straight stream/narrow fog pattern.
  • Use of a straight stream/narrow fog resulted in production of a large amount of steam. This was attributed to the fact that the hose streams had to be deflected compartment linings.
  • Water management is important when controlling fire in the upper layer, particularly when using a straight stream/narrow fog. Excess water will only result in excess steam production.

Discussing the findings, the researchers observed that pulsed application of medium fog appeared to be an effective tactic for controlling flaming combustion in the upper layer. This conclusion is supported by consistent reduction of upper layer temperature over the course of the tests involving use of pulsed application of a medium fog pattern.  Previous concerns that this approach would result disruption of thermal layering and excess steam production appeared to be unfounded. This conclusion is supported by the heat flux data at 0.9 M (7′ 10″) and 2.9 M (3′) above the floor. Disruption of thermal layering is indicated by an upward spike in lower level heat flux or equalization of heat flux at the lower and upper levels.


Several questions about the outcome of these tests were posed at the end of the Shielded Fires post.

  • Why did the application of water in a straight stream/narrow fog pattern fail to effectively control flaming combustion in the upper layer?
  • Why did the upper layer temperature fluctuate when a straight stream/narrow fog was used?
  • Why did the upper layer temperature drop consistently when a medium angle fog pattern was used?
  • How did the heat flux measurements correlate with the upper layer temperatures in these two tests?
  • What are the implications of the heat flux data recorded during these tests on tenability within the compartment for both firefighters and unprotected occupants?

Water converted to steam on contact with compartment linings or other hot objects cools the surfaces. This indirectly lowers gas layer temperature as the hot gases will continue to transfer heat to compartment linings and other cooler objects in an attempt to equalize temperature. However, the effect on upper layer temperature is limited, minimizing effectiveness of stream application in controlling flaming combustion in the upper layer. In addition, as gas temperature is not significantly reduced, steam produced on contact with hot surfaces is added to the volume of hot gases, resulting in a less tenable environment.

Ineffectiveness of straight stream/narrow fog attack in controlling flaming combustion in the upper layer and the perception of increased steam production with this type of attack likely have a common cause. Conversion of water to steam requires much more energy than simply heating water from ambient temperature to its boiling point. When water changes phase from liquid to gas (steam) while in the hot gas layer, the temperature of the gases is reduced. This has several consequences. First, sufficient reduction in temperature results in extinguishment of flaming combustion. Second, reduction of gas layer temperature causes a proportional reduction in gas volume. As illustrated in Figure 2, if 35% of the water is truned to steam in the hot gas layer, the total volume of steam and hot fire gases is less than the original volume of hot fire gases alone (Särdqvist, 2002). As this is often difficult to understand, I will provide a more detailed explanation of this in a subsequent post.

Figure 2. Gas Temperature and Relative Volume


Note. Adapted from Water and Other Extinguishing Agents (p. 155) by Stefan Särdqvist, 2002, Karlstad, Sweden: Raddningsverket. Copyright 2002 by Stefan Särdqvist and the Swedish Rescue Services Agency.

Production of the same volume of steam can have far different consequences depending on where it is produced (in the hot gas layer versus on contact with hot surfaces!

Total heat flux includes energy transferred through radiation, convection, and conduction. However, in these full scale fire tests, radiant and convective heat transfer was most significant. Radiant heat transfer is dependent on the temperature of upper layer gases and flaming combustion. Convective heat transfer is dependent on gas temperature, movement of hot gases, and moisture. Reduction in upper layer temperature while maintaining thermal layering minimizes total heat flux at the lower level where firefighters are working.

Other Considerations

These tests were conducted on a ship (see Figure 1) with most of the compartment linings being metal (rather than gypsum board, plaster, or wood as typically encountered in buildings. The fire compartment did not have windows or other ventilation openings that may exist in more typical buildings encountered by structural firefighters. These differences are significant, but do not diminish the importance of the results of these tests and findings by the researchers.

These tests provide substantive evidence in support of the effectiveness of water converted to steam in the hot gas layer (as opposed to on surfaces) in controlling flaming combustion in the hot gas layer. However, this does not diminish the importance of direct application of water onto burning fuel in a direct attack to complete the process of extinguishment.


Särdqvist, S. (2002). Water and other extinguishing agents. Karlstad, Sweden: Swedish Rescue Services Agency.

Scheffey, J., Siegmann, C., Toomey, T., Williams, F., & Farley J. (1997) 1994 Attack Team Workshop: Phase II-Full Scale Offensive Fog Attack Tests, NRL/MR/6180-97-7944. Washington, DC: United States Navy, Naval Sea Systems Command

Remember the Past

In Myth of the Self-Vented Fire I pointed out that every week represents the anniversary of the death of one or more firefighters as a result of extreme fire behavior. Some firefighters have heard about these incidents, but many have not. In an ongoing effort to encourage us to remember the lessons of the past and continue our study of fire behavior, I will occasionally be including brief narratives and links to NIOSH Death in the Line of Duty reports and other documentation in my posts.

February 11, 1998 – 1030
Firefighter Paramedic Patrick Joseph King
Firefighter Anthony E. Lockhart
Chicago Fire Department, Illinois

Firefighter King and Firefighter Lockhart responded on different companies to a report of a structural fire in a tire shop. No visible fire was encountered, there was no excessive heat, and only light smoke was found in most of the building with heavier smoke in the shop area. Ten firefighters were in the interior of the structure when an event that has been described as a flashover or backdraft occurred. The firefighters were disoriented by the effects of the backdraft. Some were able to escape but Firefighter King and Firefighter Lockhart were trapped in the structure. A garage door that self-operated due to fire exposure may have introduced oxygen into the fire area and may have been a factor in the backdraft. The exit efforts of firefighters were complicated by congestion in the building. Within minutes of the backdraft, the building was completely involved in fire and rescue efforts were impossible. Both firefighters died from carbon monoxide poisoning due to inhalation of smoke and soot. Further information related to this incident can be found in NIOSH Fire Fighter Fatality Investigation 98-F-05.

February 9, 2007
Firefighter-Paramedic Apprentice Racheal Michelle Wilson
Baltimore City Fire Department, Maryland

Firefighter Wilson and the members of her fire academy class were attending a live fire training exercise in a vacant rowhouse in Baltimore.

Firefighter Wilson was assigned to a group of apprentices and an instructor designated as Engine 1. Her group advanced a dry attack line into the structure. As they climbed the stairs, the line was charged. Engine 1 encountered and extinguished fire on the second floor but did not check the rest of the second floor for fire prior to proceeding to the third floor. On the third floor, they again encountered and began to extinguish fire.

Fire conditions began to worsen with a marked increase in smoke and heat that appeared to be coming from the second floor. Engine 1 firefighters who were on the stairs began to receive burns from the fire conditions. The instructor for Engine 1 climbed out a window at the top of the stairs and helped one burned firefighter escape to the roof.

Firefighter Wilson appeared at the window in obvious distress and attempted to escape. The windowsill was unusually high (41 inches) and she was unable to escape. Firefighter Wilson momentarily moved away from the window, at which time she advised other firefighters to go down the stairs to escape. When she returned to the window, her SCBA facepiece was off and she was beginning to receive burns. She was able to get her upper body out of the window but she could not make it through. Firefighters on the exterior were unable to pull her through until firefighters were able to gain access on the interior and assist with the effort.

When Firefighter Wilson was pulled to the roof, she was in full cardiac and respiratory arrest. She was immediately removed from the roof and received advanced life support care and transportation to the hospital. She was pronounced dead at 1250 hours. Firefighter Wilson received total body surface burns of 50 percent. The cause of death was listed as thermal burns and asphyxiation.

Further information related to this incident can be found in NIOSH Firefighter Fatality Investigation F2007-09 and the Independent Investigation Report on the Baltimore City Fire Department Live Fire Training Exercise, 145 South Calverton Road, February 9, 2007.

My next post will examine the incident in which Rachael Wilson lost her life in greater detail.

Ed Hartin, MS, EFO, MIFIreE, CFO

Shielded Fires

Monday, February 9th, 2009

Fire control and extinguishment is a fairly straightforward process when water can be applied directly to the burning fuel. In the case of burning ordinary combustibles, the energy required to heat the water to its boiling point and convert it to steam cannot be used to continue the process of pyrolysis and lowers fuel temperature to the point where the fire goes out.

However, this process is complicated when the fire is shielded from direct application of water. Assuming that an offensive strategy is appropriate, there are several options for attacking a shielded fire: 1) Make an indirect attack from the exterior (assuming you can access the involved compartment(s) from an exterior location) or 2) Move inside the building to a point inside the building where a an indirect attack can be made from outside the involved compartment(s), or 3) Move inside to a location where a direct attack can be initiated.

When adequate resources are available offensive attack from the interior addresses both life safety and fire control priorities. Proper hoseline placement and coordination of fire control and ventilation tactics protects civilian occupants and provides a safer work environment for firefighters.

The Ongoing Debate

The topic of fire stream selection (fog versus straight or solid stream) likely generates more energy than a fully developed compartment fire. Firefighters bring a great deal of passion based on experience, knowledge, and ignorance to the discussion.

There is not a single answer to the question of which type of fire steam is best. Fog and straight or solid streams have different performance characteristics and are best suited for different applications. Keep in mind that:

  • An effective fire stream puts the appropriate amount of water in the right form and right location to achieve the desired result.
  • An efficient fire stream accomplishes this with the smallest volume of water and least water damage.

Understanding the effectiveness and efficiency of fire stream application requires both qualitative and quantitative evidence. Firefighters can observe the effects of fire stream application and make a judgment as to effectiveness and efficiency. However, this understanding can be deepened by scientific examination that measures the impact of fire stream application methods.

US Navy Research

In 1994, the United States Navy conducted a series of tests to investigate the aggressive use of water fog for shipboard firefighting (Scheffey, Siegmann, Toomey, Williams, & Farley, 1997). Prior to this time, shipboard firefighting either involved direct attack with a straight stream or narrow fog pattern or indirect attack from outside the involved compartment. This series of tests compared the use of pulsed application of a medium (60o) fog pattern with use of a straight stream in controlling shielded fires and fire conditions involving high temperature and thick smoke conditions that impeded location of the seat of the fire.

The conditions (heat, smoke, and fire gases) associated with these fire scenarios typically does not prevent initial entry into the fire compartment. However, the extra time that it takes to maneuver within a space to locate and attack the seat of the fire does present a significant threat, primarily due to the stage of the fire. Uncontrolled, these fires may continue grow rapidly, potentially resulting in flashover conditions. This is particularly true where the fire is ventilation limited…and entry by the attack team introduces additional air [emphasis added].

While the Navy is concerned with shipboard firefighting, ventilation controlled, shielded fires are commonly encountered by structural firefighters as well.

Test Conditions

The tests were conducted on the ex-USS Shadwell, the Navy’s full-scale damage control research and development platform and involved several different fire scenarios. This post will examine tests involving Fire Threat 1, a growth stage fire involving multiple fuel packages within a compartment to create a well developed growth stage fire approaching flashover (upper layer temperatures in the range of 400o-600o C (752o-1152o F)). In addition, obstructions were placed to preclude the possibility of direct attack from the point of entry. Firefighters were required to control flames in the upper layer in order to penetrate deep enough into the compartment to make a direct attack on the fire.

Compartment Size and Configuration: The compartment used for the test was irregularly shaped (see Figure 2) with approximate dimensions of 8.5 M x 5.4 M (28′ x 17′ 7″) for an approximate floor area of 45.9 M2 (494 ft2).

Figure 1. Compartment Configuration


Fuel Load: Varied fuel types, including wood (red oak) cribs of varied dimensions, 1200 mm x 2400 mm (4′ x 8′) sheets of particle board (two layers of 6.4 mm particle board nailed together to provide a thickness of 13 mm (0.5″), and cardboard boxes of crumpled newspaper. All of the boxes were 457 mm x 381 mm x 305 mm (18″ x 15″ x 12″) and were taped closed after being loosely filled with newspaper. These fuel packages were distributed between three separate fire areas see (Figure 1).

  • Fire Area 1 included a triangular wood crib, three particleboard panels (placed vertically against the compartment walls), and nine cardboard boxes filled with newspaper. Fire in these fuel packages was initiated using 2.8 L (0.5 gallon) of heptanes in a 610 mm (24″) pan.
  • Fire Area 2 included a square wood crib and nine cardboard boxes filled with newspaper. Fire in these fuel packages was initiated using 18.9 L (5 gallons) of heptanes in a 914 mm (36″) square pan.
  • Fire area 3 included a rectangular wood crib and three particleboard panels (placed vertically against the compartment walls). Fire in these fuel packages was initiated using 2.8 L (0.5 gallon) of heptanes in a 610 mm (24″) pan.

Ventilation Profile: Temperature in the upper layer was monitored using thermocouples. Watertight Doors 2-22-2 and 2-21-2 were used to control the air supply to the fire and maintain consistent temperature conditions and flaming combustion in the hot gas layer for each test. When the attack team entered the compartment, air also entered the fire compartment through the entry point at Joining Door 2-16-0 (see Figure 1).

Tactical ventilation was not used in coordination with fire attack during these evolutions. The only ventilation provided while the attack team was engaged in firefighting operations involved the entry point as both exhaust and inlet opening.

Fire Control Procedures: In each of the tests the fire attack team used a 38 mm (1.5″) hoseline with a combination nozzle delivering 360 L/min (95 gpm) at 700 kPa (100 psi). For the pulsed water fog attack, the nozzle team applied oneto three second pulses with a 60o fog pattern directed upward at a 45o angle.  For the straight stream attack, the tactics were the same, but a straight stream or narrow fog pattern was used.

Note: It is important to note that the researchers and Navy firefighters involved in these tests considered a narrow fog pattern and straight stream equivalent within the context of shipboard firefighting. While the characteristics of a narrow fog pattern and straight or solid stream are different, both will penetrate through the hot gas layer and result in conversion of water to steam on contact with compartment linings (rather than within the hot gas layer)

The Tests

Test 14 was performed using traditional straight stream tactics. The nozzle operator applied two pulses with a narrow fog pattern in an attempt to control fire in upper layer. This water application produced a large amount of steam, but failed to control flaming combustion in the hot gases. After application of three more short pulses, the attack team moved to Fire Area 2 (see Figure 1) and commenced a direct attack on the fire in Fire Areas 2 and 3. However, 150 seconds (2 minutes 30 seconds) after commencing fire attack, the fire in Fire Area 1 reignited and flaming combustion in the hot gas layer caused the attack team to withdraw towards the entry point and attempt to regain control of the overhead fire. This was unsuccessful and the attack team withdrew to the entry point. A second attempt was made to enter and control the fire overhead using three long (five second) straight stream application from the doorway. These had minimal effect with continued flaming combustion overhead and involvement of fuel packages in all three fire areas. 420 seconds (seven minutes) after the initial attack, the evolution was terminated.

Test 17 replicated conditions used in Test 14, but pulsed application of a medium (60o) fog pattern was used to control fire in the upper layer, rather than a narrow fog/straight stream. Immediately after making entry, the nozzle operator applied three short pulses directed upward at a 45o Angle in the direction of Fire Area 2. The first pulse appeared to cause the flaming combustion to increase in Fire Area 2, but subsequent pulses controlled flaming combustion overhead. Visibility was reduced slightly, but the attack team was able to advance and make a direct attack on the seat of the fire. The fuel packages in Fire Area 1 reignited, but the fire was quickly controlled.

Influence on the Fire Environment

Quantitative data on factors such as upper level temperature and heat flux (heat transfer per unit area) within the compartment were recorded in addition to qualitative observations by the firefighters and researchers involved in the test. Figure 2 illustrates the temperature changes in the fire compartment during Tests 14 (straight stream/narrow fog) and 17 (medium angle fog) attacks.

Figure 2.  Average Upper Layer Temperature: Tests 14 and 17


Total heat flux (e.g., radiant and convective) was recorded 2.4 M (7′ 10″) and 0.9 M (3′) above the floor. Figures 3 and 4 illustrate conditions recorded during Test 14 (straight stream/narrow fog) and Test 17 (medium angle fog).

The dashed gray lines are provided as a point of reference at 20 kW/m2, 12.5 kW/m2, and 4.5 kW/m2. These correspond to heat flux conditions required for rapid auto ignition of ordinary combustibles, sufficient pyrolysis for piloted ignition of ordinary combustibles, and second degree burns to exposed skin within 30 seconds respectively.

Figure 3. Heat Flux Test 14


Figure 4. Heat Flux Test 17



Based on the information on the US Navy tests presented in this post, consider the following questions:

  1. Why did the application of water in a straight stream/narrow fog pattern fail to effectively control flaming combustion in the upper layer?
  2. Why did the upper layer temperature fluctuate when a straight stream/narrow fog was used?
  3. Why did the upper layer temperature drop consistently when a medium angle fog pattern was used?
  4. How did the heat flux measurements correlate with the upper layer temperatures in these two tests?
  5. What are the implications of the heat flux data recorded during these tests on tenability within the compartment for both firefighters and unprotected occupants?

More to Follow

My next post will examine the answers to these questions and the conclusions reached by the Navy researchers as a result of this series of tests.

Ed Hartin, MS, EFO, MIFireE, CFO


Scheffey, J., Siegmann, C., Toomey, T., Williams, F., & Farley J. (1997) 1994 Attack Team Workshop: Phase II-Full Scale Offensive Fog Attack Tests, NRL/MR/6180-97-7944. Washington, DC: United States Navy, Naval Sea Systems Command

Entrapment Investigation & Lessons Learned

Monday, October 20th, 2008

Structural firefighting agencies can draw some valuable lessons from the wildland firefighting community. Fire behavior training in many structural agencies often begins and ends in recruit academy. For wildland firefighters, fire behavior training involves an extensive, multi-level curriculum (S-190, S290, S-390, S-490 and so on). The wildland community is also more substantively engaged in analysis of fatalities, accidents, and near miss events with the intention of impacting policy, procedure, and performance. This is not to say that they have a perfect safety record, far from it. However, this ongoing effort to identify and implement best practice based on lessons learned is worthy of emulation.

The US Forest Service Technology & Development Program produced a document titled Investigating Wildland Fire Entrapments which outlines the process that should be used and documentation required for entrapment related incidents. Entrapments are:

A situation where personnel are unexpectedly caught in a fire behavior related, life-threatening position where planned escape routes and safety zones are absent, inadequate, or have been compromisedā€¦These situations may or may not result in injury. They include”near misses”¯.

The concept of entrapment applies equally in the structural firefighting environment. I read news accounts of extreme fire behavior related events (e.g., flashover, backdraft) from around the United States on a weekly basis. Flashover, backdraft, or other extreme fire behavior often results in a near miss or minor injury and less frequently in serious injury or fatality. Some (actually very few) of these incidents are documented in the National Firefighter Near Miss Program. As discussed in my last post, the near miss program uses self-reported data. This is extremely useful in determining the individual’s perception of the event and what lessons they took away from the experience. However, the individual reporting the event may or may not have the training or education to recognize what actually happened, determine multiple causal factors, and provide a reasonably objective analysis.

Formal Investigation

If a significant injury occurs, some level of investigation is likely to take place (even if it is limited to a cursory examination of circumstances and conditions by the individual’s supervisor). Traumatic fatalities result in more significant and in many cases multiple investigations by the agency involved, law enforcement agencies, Occupational Safety and Health Administration (state or federal), and potentially the National Institute for Occupational Safety and Health (NIOSH). The purpose of these various investigations is different and not all focus on identifying lessons learned and opportunities for improving organizational performance. However, some reports by the agencies involved, state fire service agencies, and NIOSH take positive steps in this direction. For example:


Near miss events and events involving extreme fire behavior resulting in minor injuries or damage to equipment frequently are not or are inadequately investigated to identify causal factors and lessons learned. Investigation of serious injuries and fatalities in many cases do not adequately address fire behavior and interrelated human factors that may be directly or indirectly related to the cause of the incident. This results in lost opportunities for individual and organizational learning.

Two interrelated challenges make investigating extreme fire behavior events or structural fire entrapments difficult. First is the lack of a formal process or framework for this specific type of investigation and second is potential for investigators lack of specific technical expertise in the area of fire behavior.

A Solution

The US Forest Service uses a team approach to investigating entrapment incidents. The team may include (but is not limited to):

  • Fire Operations Specialist (Operations Section Chief level)
  • Fire Safety Officer
  • Fire Behavior Analyst, with experience in the incident fuel type
  • Fire Weather Meteorologist
  • Fire Equipment Specialists who develop the personal protective equipment (including fire shelters) used on wildland fires
  • Technical Photographer
  • Fire Information Officer

This team is established and begins the investigation as soon as possible after the occurrence of the event to ensure that critical information and evidence is not lost. The investigative process and documentation focuses on accurately describing what happened, when it happened, causal and contributing factors, and recommendations to reduce the risk of future occurrence.

What might this look like in the structural firefighting environment?

Communicating Lessons Learned

Lessons learned must be integrated into appropriate training curriculum to ensure that the lessons are built into organizational culture.

Some agencies have taken steps in this direction. Following the line-of-duty death of Technician Kyle Wilson, Prince William County Department of Fire & Rescue conducted an in-depth investigation which integrated use of computational fluid dynamics (CFD) modeling to describe likely fire conditions and the influence of wind on fire behavior. Following the conclusion of this investigation, the report and related presentations have been distributed widely.

Investigating Wildland Fire Entrapments identifies timeliness as being essential in dissemination of the lessons learned. This presents a significant challenge when faced with a complex event involving a major injury or fatality. However, it is likely that timeliness in communicating lessons learned can be improved without compromising the thoroughness and quality of the investigation.

My next post will examine the US Forest Service’s less formal Peer Review Process which may be used following near miss events or significant events regardless of outcome (possibly concurrently with a formal investigation). Like the entrapment investigation procedure, there are likely some lessons here for the structural firefighting community!

Ed Hartin, MS, EFO, MIFireE, CFO