Physical Fire Modeling

1
Physical Fire Modeling
2
Introduction

• Up until now you have been learning the physical
  and chemical underpinnings of fire dynamics.
• These concepts, such as fuel properties, heat
  transfer, ventilation, etc., have been discussed
  separately and then integrated to look at the
  factors that affect fire growth and flame spread.

3
Introduction

• The rest of this course will focus on a variety
  of tools that may be used in the analysis of a
  fire.
• Your understanding of the concepts of fire
  dynamics will help ensure that you can apply
  these tools appropriately.
• In other words, we dont want to be hammering
  nails with the handle of a screwdriver.
• We want to use the right modeling tools for the
  job.

4
Introduction

• Model, simulation, re-creation are words for
  describing a copy of an item or an action.
• In our context, reproducing fire phenomena is
  typically why we would use a model.
• The copy could be in form of a physical model,
  such as the recreation of a full-scale room fire
  or a reduced scale model.
• Other methods of modeling using mathematical
  descriptions of a physical phenomena, such as a
  burning piece of furniture or a room fire, will
  be covered in the following sections.

5
Types of Fire Models Used in Fire Safety Design
6
Types of Fire Models Used in Fire Safety Design
7
Types of Fire Models Used in Fire Safety Design

• Any time that a copy or a model is made, the
  reproduction is never a 100 duplicate.
• NEVER!!!
• Therefore it is important to develop an
  understanding of how close to the original some
  of these simulation tools might bring you, if
  used by a skilled experimentalist, modeler etc.

8
Types of Fire Models Used in Fire Safety Design

• Remember as an investigator, the model can only
  be as good as the data or information from which
  it was developed.
• Your efforts in collecting the data at the scene,
  interviewing witnesses, developing a fire
  timeline are critical to developing a realistic
  model.
• While the results of many of the tools that will
  be presented are quantitative, typically the
  models are a means to test a hypothesis, gain
  insight, and examine the sensitivity of factors
  that may be critical to your investigation.

9
Full - Scale Models

• To introduce full-scale models, lets begin by
  discussing something familiar, a person walking.
• Lets imagine that for purposes of your fire
  investigation timeline, you want to know
  approximately how long it would take a
  middle-aged man with brown hair, approximately 6
  feet tall and weighting approximately 200 lbs, to
  walk 100 yards.
• Your officemate, George, fits that description.
• You have a tape measure, a stopwatch and a big
  parking lot.
• You are ready to re-create the 100 yard walk.
• After Georges walk you have a time measurement
  from your stopwatch.

10
Full - Scale Models

• But, how good are the results of your model?
• How representative is your re-creation of the
  walk?
• Did you have George walk the distance more than
  once?
• Is George consistant?

11
Full - Scale Models

• What other information would you need in order to
  determine how well you simulated the middle-aged
  mans 100 yard walk?

12
Full - Scale Models

• Physical Criteria
• Physical Condition of man
• Healthy athletic or injured or impaired
• Length of legs, typical length of stride
• Was the man carrying, pushing or pulling anything?

13
Full - Scale Models

• Environment
• Flat path vs inclined path of travel
• Straight vs curved
• Paved vs rough terrain
• Day or night
• Weather
• Dry or raining
• Calm or windy

14
Full - Scale Models

• Obstacles
• People
• Traffic
• Physical Barriers
• Distractions

15
Full - Scale Models

• Motivation
• Why was he walking?
• Where was he walking from?
• Where or what was he walking to?

16
Full - Scale Models

• As you can see, there are many parameters that
  need to be considered even in a simple
  simulation or physical model, i.e. its never
  simple.
• In the context of fire behavior, physical
  descriptions and environmental conditions need to
  be accounted for in order to develop a reasonable
  simulation.

17
Full - Scale Models

• The motivational component is not an input or an
  output of fire models.
• Another important means of assessing your answer
  from the 100 yard walk, or from any test, is to
  compare it with data from well documented
  research papers on issue in question.

18
Full Scale Standardized Testing

• Full scale fire testing is conducted on a daily
  basis to support product listings or approvals at
  laboratories such as Underwriters Laboratories
  and Factory Mutual.
• A typical full-scale test may involve the
  ignition of a well-defined commodity, such as
  plastic cups in cardboard boxes on high rack
  storage shelves.
• The ignition source and the heat release rate
  from this fuel package has been well documented
  and has been shown to be repeatable.
• The test scenario is intended to represent a high
  rack warehouse scenario with a high challenge,
  fuel load.

19
Full Scale Standardized Testing

• The types of products tested by these repeatable
  fire conditions are typically sprinklers and
  sprinkler design criteria.
• Tests like this cost tens of thousands of dollars
  each.
• Results from standardized tests can help an
  investigator develop a basis for material and
  fire protection system behavior in a given type
  of full-scale fire situation.
• For example, for the type of tests outlined
  above, the temperature data taken near the
  ceiling of these experiments could be used for
  comparison if you had a fire investigation with a
  similar fuel and arrangement.

20
Room Experiments

• Perhaps a more familiar example of a full-scale
  fire test to fire investigators is the single,
  fully furnished room scenario.
• These may range from a room mock-up burn in a
  trash dumpster laying on its side to a mock-up of
  a living room on a trailer, to a room built
  inside a test facility and finally a room within
  an acquired structure or building of
  opportunity.
• These types of fire tests are conducted around
  the county, typically for training or research
  purposes by fire service or law enforcement
  groups, and researchers.

21
Room Experiments

• The cost of these tests can range, from the cost
  of the furniture and feeding the volunteer staff,
  to tens of thousands of dollars for a fully
  instrumented repeatable experiment.
• Standard Rooms can be used to examine the
  propensity of interior finish materials to
  contribute to room fire growth. NFPA 286 is on
  such test method that utilizes a 2.44 m by 3.66 m
  by 2.44 m tall room with a single doorway in one
  end.
• The open doorway vents into a 2.44 m by 2.44 m
  exhaust hood that is instrumented for oxygen
  consumption calorimetry.

22
Experiments vs Demonstrations

• Many of the above single room tests would be
  considered demonstrations or demos as opposed
  to a scientific experiment.
• For example, a flashover demo would involve
  filling a room or compartment with furniture,
  lighting it on fire, and with video rolling watch
  for flashover to occur.
• Typically an experiment would include
  documentation of the room geometry, fuel types,
  quantities and locations, source of ignition and
  test protocol.
• Usually an experiment would be used to test a
  hypothesis.

23
Experiments vs Demonstrations

• In most cases this means that several experiments
  would need to be conducted for comparison
  purposes and ideally, replicates of each type of
  experiment are conducted to demonstrate the
  behavior is repeatable.
• Experiments take a considerable amount of
  planning and care in order to provide useful
  results.
• Demonstrations can provide an investigator with
  an experience, but experiments will provide an
  investigator with data.
• The data is what provides the support for
  developing technically defensible conclusions on
  origin and cause determination.
• For investigators, there is value to being able
  to contrast and compare two fires, similar in all
  ways but one, in order to see what impact that
  one difference had on the development of the
  fire and the resulting fire damage.

24
Buildings of Opportunity

• Conducting experiments in buildings of
  opportunity enables investigators to examine
  laboratory based theories, to validate computer
  models, or to determine the effectiveness of fire
  protection systems in situ.
• For example, sprinkler effectiveness studies were
  conducted in a dormitory at the University of
  Arkansas in Fayetteville.
• Even though the experiments were conducted with
  the intent of improving the fire safety in
  college dorms, the data can be useful to fire
  investigators as well.

25
Buildings of Opportunity

• Looking at the photos the value of a sprinkler in
  preserving the origin of the fire can be scene.
• The comparison of the before and after photos of
  furnishings can assist the investigator in
  developing a basis for re-construction of similar
  fires.
• In this case, for example, recognizing that the
  milk carton bookcase has been consumed, while
  the remains of the notebooks and papers remain on
  the floor of the room.
• Development of a probabilistic database from
  documented burns in real structures can be used
  to identify trends in fire behavior.

26
Sprinklered room before fire. Sprinklered room
after fire.
27
Non-Sprinklered room before fire. Non-Sprinklered
room after fire.
28
Capabilities and limitations

• Currently, full-scale physical models have the
  greatest potential to reproduce or re-create a
  fire scenario
• provided that the physical model is an accurate
  representation of the actual
• incident fuel package,
• geometry,
• ignition source and
• location and surrounding environmental
  conditions.
• Full-scale experiments are expensive.
• Therefore typically a very limited number, if
  any, replicate tests are conducted.

29
Capabilities and limitations

• Full-scale experiments tend to have less
  experimental control, than laboratory or bench
  scale experiments.
• Conditions which impact the ability to re-create
  a full-scale model and demonstrate repeatability
  include
• limits of the test facility may not represent the
  fire building
• uncertainties due to materials, weather, source
  of ignition
• economics will not allow a full reproduction of
  the area of interest

30
Reduced scale fire experiments

• In most cases, it is just not practical,
  feasible, or necessary to consider a full-scale
  fire experiment.
• For example reconstructing a large warehouse fire
  or a fire involving a shopping mall would not be
  done due to cost, lack of available test space,
  etc.
• However building a reduced scale model or using
  laboratory scale test apparatus may provide
  results useful to the fire analysis.

31
Reduced scale fire experiments

• Reduced scale fire experiment examples
• The simplest version of a reduced scale model can
  be something like a doll house.
• For example an experimental structure can be
  built at reduced physical-scale, such as 1/12th
  where 12 inches real scale equals 1 inch in
  reduced scale.
• Some fire properties, scale directly with the
  physical size but many do not. Quintiere 1995
  provides a good example with a 1/7th scale, 5
  story building atrium model that was constructed
  to study smoke spread through a store after a
  polystyrene and wood Santa Claus display caught
  fire in the atrium.
• Quintiere demonstrates the scaling concepts used
  in the model atrium.

32
Reduced scale fire experiments

• Reduced scale experiments can be a great benefit,
  provided that it is recognized that all fire
  properties to do not scale the same and this
  needs to be accounted for and limits the use of a
  scale model.
• For example convective flows and radiant heat
  transfer do not scale the same, therefore a
  single scale model could only address one or the
  other.

33
Reduced scale fire experiments

• The physics of fire dynamics aside, a simple
  scale model of a structure or neighborhood can be
  of benefit in helping to explain the findings of
  an investigation or for use in demonstrating the
  path of fire flow or people movement.
• Salt water modeling is another means of
  simulating fire and smoke movement in a reduced
  scale environment.
• A reduced scale model is constructed from a clear
  material such as Plexiglas and submerged upside
  down into a tank of fresh water.
• Dyed salt water is allowed to flow into the model
  from the fire source.

34
Reduced scale fire experiments

• As the denser salt water sinks it moves through
  the water until it contacts the ceiling of the
  reduced scale structure and spreads out like a
  ceiling jet.
• The heavy salt water in the upside down room is
  analogous to the less dense smoke that rises in a
  typical room fire scenario.
• No heat transfer effects are considered by salt
  water modeling.
• Many times it is appropriate to model or test a
  small component or a piece of the fire scenario
  room.
• A corner test may be used to examine vertical
  flame spread on wall coverings, laboratory scale
  apparatus such as the cone calorimeter and the
  lateral ignition and flame spread test (LIFT) may
  be used to determine a wide variety of material
  properties such ignition temperature.