When a major fire occurs, especially one with associated firefighter injuries or deaths, departments often produce after-action reports, conduct investigations and inspect equipment. All of these are valuable in helping us understand what went right–and wrong–as well as institutionalizing the lessons learned from the specific incident.
But today’s technology allows us to go beyond our more traditional means of learning. Using sophisticated models, we can simulate the movement of fire gases and heat through a building and estimate the response of various fire-protection systems. The National Institute of Standards and Technology (NIST) is the leader in the development of computer-based fire models, which have progressed dramatically as computer technology has improved. Such models allow us to understand from a scientific perspective how fires will play out under various conditions–and use this understanding to enhance firefighter safety.
2 Models
The Fire Research Division at NIST has developed different kinds of fire models for more than 30 years. Currently, it maintains two: Consolidated Fire and Smoke Transport (CFAST) and the Fire Dynamic Simulator (FDS).
CFAST is a “two-zone” model that was developed in the 1980s. It has the advantage of producing a model in just seconds, but the disadvantage of providing only average compartment temperatures. FDS is a computational fluid dynamics (CFD) model developed in the past decade. It provides far more detail of the fire and the gas flow, but calculations can take hours or days to complete.
Both of these models can be used to simulate the movement of smoke and heat through a building. They can also be used to examine the activation time of smoke alarms and sprinkler systems. Both models use the scientific visualization tool Smokeview to visualize model results. Typically, these models are used by engineers to develop and support “performance-based” fire safety designs and to simulate fires as part of the fire investigation process. These fire models are verified and validated against fire test data to ensure that they provide the expected results.
Fire Reconstructions
In conjunction with local fire departments and NIOSH, NIST has developed fire simulations via FDS and Smokeview to assist in the understanding of the fire behavior in several line-of-duty death (LODD) incidents. The fire simulations provide insight into the growth and spread of fire and hot gases through the structures.
FDS, rather than CFAST, is typically used for fire reconstructions. FDS requires the fire building to be modeled in a 3-D volume divided into computation cells; it then numerically computes the density, velocity, temperature, pressure and species concentration of the gas in each cell. The model tracks the generation and movement of fire gases based on the laws of conservation of mass, momentum, species and energy.
Smokeview is a user-friendly post-processing tool that allows FDS’ numerical simulation outputs to be easily displayed with 3-D images. Smokeview can display contours of temperature, velocity and gas concentration in planar slices, plus realistic renderings of the smoke and fire.
Inputs required by FDS include:
- Geometry of the structure
- Computational cell size
- Location of the fire source
- Energy release rate of the fire source
- Mass, geometry and thermo-physical properties of walls, ceilings, floors and furnishings
- Size, location and timing of door and window openings inside and outside of the structure
The selection of input parameters has a significant impact on the outcome of the simulation. In many cases we don’t know all of the materials and fuels that existed at a fire scene. Due to this uncertainty, a range of values is typically used for input into the fire model. As a result, for a given fire, dozens of simulations are run to determine the parameters that generate the simulation that best aligns with the physical damage, photographic and/or video recordings of the fire, fire timeline, fireground recordings and witness statements.
Lessons from Fire Modeling
One of the key features of FDS models is the ability to visualize the fire’s flow path. In many of the incidents NIST and others have analyzed with FDS, the firefighters lost their lives due to a rapid change in the fire environment. In almost all of the LODD cases examined with FDS, a change in ventilation resulted in a major increase in the energy release of the fire that either limited the firefighters’ ability to leave the building or overtook them.
For example, Cherry Road (Washington D.C., 1999), Iowa (1999), Prince William County, Va. (2007) and Texas (2009) all involved situations where the firefighters were positioned in the flow path between the source of the fire, and–due to the change in ventilation–the exit point for the fire gases. In other words, the firefighters were caught between where the fire was and where it wanted to go.
As complicated a technological marvel that FDS and Smokeview are, for the firefighter on the street, it helps to think of fire modeling as an advanced way of visualizing the fire triangle. As the triangle tells you, fuel, heat and oxygen are required to combine via chemical reaction for a fire to exist. Take any one of these away, and the fire cannot exist.
Most structure fires today are ventilation-limited fires. This means that there’s more fuel inside the structure (fuel-rich) than the available ventilation can provide oxygen for. The scenario has been demonstrated on the fireground and with FDS many times: The fire department arrives at the structure fire and only sees smoke. The smoke contains products of combustion that include fuels such as carbon monoxide, carbon particles and unburned hydrocarbons. So the structure, full of smoke, is effectively a large, insulated container full of pre-heated fuel.
The only thing missing from the fire triangle: oxygen. Opening a door or breaking windows will lead to a ventilation-induced flashover. This phenomenon was demonstrated at Cherry Road, and in wind-driven LODD fires, such as those in Prince William County and in Texas. The phenomenon was also demonstrated in the Charleston Sofa Super Store Fire. Significant quantities of hot smoke mixed with fresh oxygen after windows along the front of the store were broken out. As more oxygen was available to be burned, the heat release rate increased and flames extended out of the window openings.
What’s Next
Although the LODD simulations have been effectively used in firefighter training courses, effort is being made at NIST to enable firefighters to use fire models in a more proactive and interactive manner. As the capabilities of the fire models increase and computational speeds improve, the use of models in training must expand in order to aid in the understanding of fire dynamics, particularly the impact of ventilation on a fire and other tactical considerations that will help prevent LODDs in the future.
References
- McGrattan KB, Hostikka S, Floyd JE. Fire Dynamics Simulator (Version 5), User’s Guide. NIST Special Publication 1019-5. National Institute of Standards and Technology: Gaithersburg, Md., October 2007.
- Forney GP. Smokeview (Version 5), A tool for visualizing fire dynamics simulation data, Volume I: User’s Guide. NIST Special Publication 1017-1. National Institute of Standards and Technology: Gaithersburg, Md., August 2007.
- Fire Dynamics Simulator, Technical Reference Guide, 5th edition. NIST Special Publication 1018-5 (four-volume set). National Institute of Standards and Technology: Gaithersburg, Md., and VTT Technical Research Centre of Finland, Espoo, Finland, October 2007.
- Madrzykowski D, Vettori RL. Simulation of the dynamics of the fire at 3146 Cherry Road NE, Washington, D.C., May 30, 1999. NISTIR 6510. National Institute of Standards and Technology, Gaithersburg, Md., April 2000.
- Madrzykowski D, Forney GP, Walton WD. Simulation of the dynamics of a fire in a two-story duplex–Iowa, December 22, 1999. NISTIR 6854. National Institute of Standards and Technology: Gaithersburg, Md., January 2002.
- Technician I Kyle Wilson LODD Report. In Prince William County, Va. Retrieved July 29, 2011, from www.pwcgov.org/default.aspx?topic=040061000230006431.
- Barowy A, Madrzykowski D. Simulation of the dynamics of a wind-driven fire in a single-story ranch-style house–Texas, April 12, 2009.
Bryner NP, Fuss SP, Klein BW et al. Technical study of the Sofa Super Store Fire, South Carolina, June 18, 2007. NIST SP–1118. National Institute of Standards and Technology: Gaithersburg, Md., March 14, 2011.
For More Information
- More details on NIST fire models, reports and test data can be downloaded from www.fire.nist.gov.
- Complete descriptions of the FDS model and Smokeview, as well as the technical references that support the model, are given in references #1—3 in this article and can be downloaded from www.fire.nist.gov/fds/.
