NIST Technical Note 1629

 
 
Fire Fighting Tactics Under Wind Driven Fire Conditions:
7-Story Building Experiments

Stephen Kerber
Daniel Madrzykowski

U.S. Department of Commerce

Building and Fire Research Laboratory
National Institute of Standards
and Technology
Gaithersburg, MD 20899
 NIST Technical Note 1629

Fire Fighting Tactics Under Wind Driven Fire Conditions:

7-Story Building Experiments

Stephen Kerber
Daniel Madrzykowski

U.S. Department of Commerce

Building and Fire Research Laboratory
National Institute of Standards
and Technology
Gaithersburg, MD 20899

April 2009

Fire Department of New York City Polytechnic Institute of NYU

Nicholas Scoppetta, Fire Commissioner Jerry MacArthur Hultin, President
Salvatore Cassano, Chief of Department

Department of Homeland Security U.S. Department of Commerce

Janet Napolitano, Secretary Otto J. Wolff, Acting Secretary
Federal Emergency Management Association National Institute of Standards and
Nancy Ward, Acting Administrator Technology
United States Fire Administration Patrick Gallagher, Deputy Director
Glenn A. Gaines, Acting Assistant Administrator
Fire Fighting Tactics Under Wind Driven Fire Conditions:
7-Story Building Experiments

Stephen Kerber
Daniel Madrzykowski
Building and Fire Research Laboratory
National Institute of Standards and Technology

Abstract 

In February 2008, a series of 14 experiments were conducted in a 7-story building to evaluate the
ability of positive pressure ventilation fans, wind control devices and external water application
with floor below nozzles to mitigate the hazards of a wind driven fire in a structure. Each of the
14 experiments started with a fire in a furnished room. The air flow for 12 of the 14 experiments
was intensified by a natural or mechanical wind.. Each of the tactics were evaluated individually
and in conjunction with each other to assess the benefit to fire fighters, as well as occupants in
the structure.

The results of the experiments provide a baseline for the hazards associated with a wind driven
fire and the impact of pressure, ventilation and flow paths within a structure. Wind created
conditions that rapidly caused the environment in the structure to deteriorate by forcing fire gases
through the apartment of origin and into the public corridor and stairwell. These conditions
would be untenable for advancing fire fighters. Each of the tactics were able to reduce the
thermal hazard created by the wind driven fire. Multiple tactics used in conjunction with each
other were very effective at improving conditions for fire fighter operations and occupant egress.

Fire departments that wish to implement the tactics used in this study will need to develop
training and determine appropriate methods for deploying these tactics. Variations in the
methods of deployment may be required due to differences in staffing, equipment, building
stock, typical weather conditions, etc. There is uniformity however, in the physics behind the
wind driven fire condition and the principles of the tactics examined. The data from this
research will help provide the science to identify methods and promulgation of improved
standard operating guidelines (SOG) for the fire service to enhance firefighter safety, fire ground
operations, and use of equipment.

The experiments were conducted by the National Institute of standards and Technology (NIST),
the Fire Department of New York City (FDNY), and the Polytechnic Institute of New York
University with the support of the Department of Homeland security (DHS)/Federal Emergency
Management Agency (FEMA) Assistance to Firefighters Research and Development Grant
Program and the United States Fire Administration.

i
Disclaimer 

Certain trade names or company products are mentioned in the text to specify adequately the
experimental procedure and equipment used. In no case does such identification imply
recommendation or endorsement by the National Institute of Standards and Technology, nor
does it imply that the equipment is the best available for the purpose.

Regarding Non-Metric Units: The policy of the National Institute of Standards and Technology
is to use metric units in all its published materials. To aid in the understanding of this report, in
most cases, measurements are reported in both metric and U.S. customary units.

ii
Acknowledgments 
Experiments of this magnitude require the assistance of many individuals and cooperation of many organizations to
plan and execute. To list everyone that made these experiments possible by name would be impossible. The
following organizations played an important role in bringing this study to fruition and the authors can’t express their
gratitude enough.

The key organization to making this happen was the Fire Department of New York City. The professionalism and
expertise of their staff and fire fighters is first class. We would especially like to thank: Fire Commissioner
Nicholas Scoppetta, COD Salvatore Cassano, Chief of Training Thomas Galvin, Project Coordinator Deputy Chief
Jack Mooney, Incident Management Team; Deputy Chief Joseph Saccente, Deputy Chief Michael McPartland, and
Battalion Chief Joel Gerardi, Project Assistants and subject Matter Experts; Battalion Chief Gerald Tracy, Battalion
Chief George Healy, Battalion Chief Joseph Cunningham, Captain Thomas Yuneman, Captain Mark Foris,
Lieutenant John Ceriello, and FDNY Director of Grant Development Irene Sullivan. We would also like to
acknowledge the FDNY Research and Development group especially Battalion Chief Andy Richter and Captain
Eric Sacknoff, who were instrumental in tool testing and water flow measurement and Firefighter Paul Robinson for
the development of FDNY’s high rise nozzle prototype. Last but not least, we would like to thank all of the FDNY
fire fighters, fire marshals and technical specialists that participated in the experiments and made this report
possible.

The Polytechnic Institute of New York University played a key role in documenting the structure and monitoring the
weather and wind conditions during the experiments. They will continue the research on the impact of wind on fires
in structures through the use of computer modeling. This effort is being lead by Dr. Sunil Kumar. The principal
members of the team involved in the experiments included: Prabodh Panindre, Vishal Prajapati, Susan Mousavi,
Christopher Alvarez, and Ramirez Antonio.

The authors would like to express their appreciation to the many Building and Fire Research (BFRL) staff members,
who provided their technical expertise to make these experiments a success, especially; Roy McLane, Jay McElroy,
Mike Selepak, Anthony Chakalis, Nelson Bryner and Anthony Hamins.

Many other organizations provided assistance that was critical to the performance of this study:
Bureau of Alcohol Tobacco Firearms and Explosives, especially John Pijaca, David Sherman
Chicago Fire Department, especially Battalion Chief Peter Van Dorpe, Battalion Chief Keith Witt and Director
of Training Richard Edgeworth
Delaware County (PA) Emergency Service Training Center, especially Kerby Kerber
DHS/FEMA Assistance to Firefighters Grant Program
Eagle Fire Suppression, especially Kevin Bourke (Capt Eng. Co. 45 FDNY retired)
Fire Protection Research Foundation, especially Casey Grant
Gilbert (AZ) Fire Department, especially Troy Duncan
Governors Island Preservation and Education Corporation, especially Terry Williams
KO Fire Curtain, especially Pat Kilduff and Tom Oswald
Mesa (AZ) Fire Department, especially Josh Friedman
Mink Hollow Systems, especially Todd Wilson
Tempest Technologies, especially Dexter Coffman
U.S. Fire Administration, especially Meredith Lawler

While the preparation and performance of the experiments was challenging and labor intensive, the development of
this report was also challenging and labor intensive. The authors wish to thank Jonathan Kent for the application of
his programming talents to enabling rapid data analysis schemes and the ability to easily meld the timeline
information with the numerical data on each graph.

Last but not least, we would like to thank the reviewers Dr. Andrew Lock, Dr. Glenn Forney, Dr. William
Grosshandler and Mr. Paul Reneke for their timely and thorough reviews of this report.

iii
Nomenclature 

Text:

gpm gallons per minute

FBN 1 Floor below nozzle (high rise nozzle) which is angled up from the floor below
FBN 2 Floor below nozzle (high rise nozzle) with a hook that attaches to the fire floor
IR Infrared
mph miles per hour
MVU Mobile Ventilation Unit
PPE Personal Protective Ensamble
PPV Positive Pressure Ventilation
SCBA Self Contained Breathing Apparatus
WCD Wind Control Device

Figures:

BH Bulkhead
BR Bedroom
double The double window present in the bedrooms of the test apartments; unless
otherwise specified, assume the window of the ignition apartment
LR Living Room
S1 Stairwell doorway on the 1st floor
S3 Stairwell doorway on the 3rd floor
S5 Stairwell doorway on the 5th floor
S7 Stairwell doorway on the 7th floor
single The single window present in the bedroom of the test apartments

iv
Table of Contents 

Abstract.......................................................................................................................................................... i
Disclaimer ....................................................................................................................................................ii
Acknowledgments ................................................................................................................................... iii
Nomenclature.............................................................................................................................................iv
Table of Contents ....................................................................................................................................... v
Executive Summary............................................................................................................................... viii
1 Introduction........................................................................................................................................1
1.1 Background .......................................................................................................................................... 2
1.2 Structure................................................................................................................................................ 6
1.3 Wind ..................................................................................................................................................... 10
1.4 Fuel Load ............................................................................................................................................ 12
1.5 Instrumentation............................................................................................................................... 19
1.6 Estimated Measurement Uncertainty ...................................................................................... 20
1.7 Positive Pressure Ventilation Fans ........................................................................................... 21
1.8 Wind Control Blankets/Curtains ............................................................................................... 22
1.9 Floor Below Water Application Nozzles.................................................................................. 23
2 Experimental Procedure ............................................................................................................. 26
3 Results ............................................................................................................................................... 27
3.1 Experiment 7G .................................................................................................................................. 28
3.2 Experiment 7G2 ............................................................................................................................... 54
3.3 Experiment 7E .................................................................................................................................. 86
3.4 Experiment 7A ................................................................................................................................114
3.5 Experiment 7K................................................................................................................................146
3.6 Experiment 5E ................................................................................................................................173
3.7 Experiment 5E2..............................................................................................................................199
3.8 Experiment 5K................................................................................................................................221
3.9 Experiment 5A ................................................................................................................................247
3.10 Experiment 5G ................................................................................................................................273
3.11 Experiment 5G2 .............................................................................................................................295
3.12 Experiment 3E ................................................................................................................................317
3.13 Experiment 3A ................................................................................................................................347
3.14 Experiment 3G/K...........................................................................................................................372
4 Discussion.......................................................................................................................................411
4.1 Door Control....................................................................................................................................413
4.2 PPV fans to control smoke movement ...................................................................................415
4.3 WCD to control wind driven conditions ................................................................................417
4.4 FBN to control wind driven conditons ...................................................................................418
4.5 Window Failure Time ..................................................................................................................419
5 Tactical Considerations.............................................................................................................421
6 Future Research ...........................................................................................................................423
6.1 Pilot Programs................................................................................................................................423
6.2 Standard Test Methods for Equipment..................................................................................423
7 Summary.........................................................................................................................................423
8 References ......................................................................................................................................426
9 Appendices .....................................................................................................................................429

v
9.1 Appendix A – Detailed Drawings..............................................................................................429
9.2 Appendix B – Fuel Distribution ................................................................................................447
9.3 Appendix C – Wind Measurements and Descriptions.......................................................451
9.4 Appendix D – Detailed Graphs ..................................................................................................497
9.5 Appendix E – Post Fire Images..................................................................................................567

vi
vii
Executive Summary 

The National Institute of Standards and Technology (NIST), the Fire Department of New York
City (FDNY), and the Polytechnic Institute of New York University with the support of the
Department of Homeland Security (DHS)/Federal Emergency Management Agency (FEMA)
Assistance to Firefighters Research and Development Grant Program and the United States Fire
Administration (USFA), have conducted a series of wind driven fire experiments in a seven story
building on Governors Island, New York.

The objective of this study was to improve the safety of fire fighters and building occupants by
developing a better understanding of wind driven fires and wind driven fire fighting tactics,
including structural ventilation and suppression. A series of 14 experiments were conducted to
evaluate the ability of positive pressure ventilation fans (PPV), wind control devices (WCD), and
exterior water application via floor below nozzles (FBN) also known as high rise nozzles to
mitigate the hazards of a wind driven fire in a structure.

Each of the 14 experiments started with a fire in a furnished room. The air flow for 12 of the 14
experiments was intensified by a natural or mechanical wind. Each of the tools was evaluated
individually as well and in conjunction with each other to assess the benefit to fire fighters, as
well as occupants in the structure. The data collected used to examine the impact of the PPV
fans, WCDs, and the exterior water application tactics were temperature, differential pressure,
and gas velocity inside the structure. Each of the experiments was documented with video and
thermal imaging cameras. These experiments also captured video of specific fire phenomena
that are not typically observable on the fire ground.

During these experiments a public corridor and stairwell area was exposed to a wind driven,
post-flashover apartment fire. The door from the apartment to the corridor was open for each of
the experiments. The conditions in the corridor and the stairwell were of critical importance
because that is the portion of the building that firefighters would use to approach the fire
apartment or that occupants from adjoining apartments or adjacent floors would use to exit the
building. Fires in high-rise buildings create unique safety challenges for building occupants and
fire fighters. Smoke and heat spreading through the corridors and the stairs of a building during
a fire can limit building occupants’ ability to escape and can limit fire fighters’ ability to rescue
them. In 2002, there were 7300 reported fires in high rise structures (seven stories or more). The
majority of these high rise fires occurred in residential occupancies, such as apartment buildings.
In fires that originated in apartments, 92 % of the civilian fatalities occurred in incidents where
the fire spreads beyond the room of origin.

All of the fires were ignited in furnished rooms of an apartment. Due to excess fuel
pyrolysis/generation (lack of ventilation) the room of fire origin could not transition to flashover
until windows self-vented and introduced additional fresh air with oxygen to burn. Without a
wind imposed on the vented window, the fire did not spread from the room of origin and never
left the apartment of origin. Even with no externally applied wind, creating a flow path from the
outside, through the fire apartment into the corridor and up the stairs to the open bulkhead on the
roof increased the temperatures and velocities in the corridors and in the stairwell resulting in

viii
hazardous conditions for fire fighters and untenable conditions for occupants on the fire floor and
above in the stairwell.

With an imposed wind of 9 m/s to 11 m/s (20 mph to 25 mph) and a flow path through the fire
floor and exiting out of the bulkhead door on the roof, temperatures in excess of 400 ºC (752 ºF)
and velocities on the order of 10 m/s (22 mph) were measured in the corridor and stairwell above
the fire floor. These extreme thermal conditions are not tenable, even for a firefighter in full
protective gear.

These experiments demonstrated the “extreme” thermal conditions that can be generated by a
“simple room and contents” fire and how these conditions can be extended along a flow path
within a real structure when wind and an open vent are present. Potential tactics which could be
implemented to interrupt and control the flow path are door control from inside the structure and
or a WCD from the floor above the fire. From the floor below the fire, external water application
was demonstrated to be effective in reducing the thermal hazard in the corridor and stairwell.

Identify the Potential for Wind Driven Conditions. Wind conditions should be considered as
part of initial size-up of the incident. Wind conditions can vary widely in an urban environment
due to wind flows around buildings, or shielding by buildings that give the perception on the
ground that no significant wind is present, but another side of the building or a higher elevation
in the building may be exposed to wind conditions. Wind speeds on the order of 10 mph to
20 mph are high enough to create a wind driven fire condition in the structure with an
uncontrolled flow path.

If the fire has vented a window, important information can be gained by observing the behavior
of the flame at the window. If the fire apartment has a high pressure relative to the outside due
to an imposed wind, the flame will “pulse” out of the window to balance the overpressure. If the
flames are being forced out of the window a flow path has been established through the building
and the flow direction maybe favorable to interior fire fighting. If the flames are pulsing or
being forced into the window, condition may not be favorable to interior firefighting and caution
should be used on the approach to the fire floor. Even if flames are being forced out of adjacent
windows in the fire apartment with a high amount of energy, there could still be sufficient energy
flows on the fire floor to create a hazard for firefighters.

Wind driven fire conditions – The wind driven condition can be described as hot gases or
flames flowing horizontally out of the room of fire origin. The wind driven fire condition has
been described as a “blow torch” by firefighters. For our purposes, a wind driven fire condition
exsisted when the fire gases were well mixed and of equally high temperature from the floor to
the ceiling, on the order of at least 400 ºC (752 ºF). For this condition to occur inside a structure,
the fire must be in a flow path. In these experiments the inlet to the flow path was the upwind
window in the room of fire origin. The flow path then went through the aparatment, into the
corridor, and exited out of the bulkhead door on the roof, via the stairwell. Without a flow path,
the wind driven fire condition inside the structure can not occur.

Door Control. Door control is the most basic means to interrupt or control the flow path in the
building. The fire floor stair door should be checked for heat or hot gases flowing around the

ix
edges. The door should only be opened a few inches at first to look for rapid changes in smoke
volume or velocity and/or thermal conditions. If the thermal environment changes quickly, close
the door to interrupt the flowpath. In a smoke filled environment, visual changes to conditions
may not be apparent with out a thermal imager. A similar approach would be used on the door to
the fire apartment.

Impact of PPV. PPV fans alone could not overcome the effects of a wind driven condition.
However when used in conjunction with door control, WCDs, and FBNs, the PPV fans were able
to maintain tenable and clear conditions in the stairwell. The key to successful use of PPV fans
was to mitigate the wind driven fire condition via door control or other tactics. Then the PPVs
can be used to clear the stair and then pressurize the stairwell to provide a safe working
environment. Although the PPV fans, when used alone, could not reverse the flow of a wind
driven fire, PPV fans always improved conditions in the stairwell.

Impact of WCDs. In these experiments, the WCDs reduced the temperatures in the corridor and
the stairwell by more than 50 % within 120 s of deployment. The WCDs also completely
mitigated any velocity due to the external wind. The WCDs were exposed to a variety of
extended thermal conditions without failure.

Impact of externally applied water. In these experiments, the externally applied water streams
were implemented in different ways; a fog stream inserted into the fire room window, a fog
stream flowed from the floor below into the fire room window opening, and a solid water stream
flowed from the floor below into the fire room window opening. In all cases, the water flows
suppressed the fires, thereby causing reductions in temperature in the corridor and the stairwell
of at least 50 %. The water flow rates used in these experiments were between 125 gpm and
200 gpm, demonstrating that a relatively small amount of water applied directly to the burning
fuels can have a significant impact.

Stored Energy. Wind driven fire conditions can generate and transfer energy throughout the
flow path. When doors or WCDs are used to stop the wind driven fire conditions, energy and
fuel may be trapped on the fire floor. These experimental results indicate that the thermal
conditions due to the residual heat on the fire floor were still of a level which could pose a
hazard to firefighters in full PPE. However, when used in combination with PPV fans to force
cool air into the stairwell and out through the fire floor, and or with the cooling effect from water
streams, the fire floor temperatures were reduced to tenable conditions for fire fighters in full
PPE in minutes.

The data from this research will help provide the science to identify methods and promulgation
of improved standard operating guidelines (SOG) for the fire service to enhance firefighter
safety, fire ground operations, and use of equipment. If the demonstrated technologies continue
to prove effective in the field trials and pilot programs, the next step may be to examine the need
for standards and standardized test methods to define a minimum level of acceptable
performance of these devices.

x
1 Introduction 

The National Institute of Standards and Technology (NIST), the Fire Department of New York City
(FDNY), and the Polytechnic Institute of New York University with the support of the Department of
Homeland Security (DHS)/Federal Emergency Management Agency (FEMA) Assistance to Firefighters
Research and Development Grant Program and the United States Fire Administration (USFA), have
conducted a series of wind driven fire experiments in a seven story building on Governors Island, New
York. The experiments were conducted in February 2008.

The objective of this study was to improve the safety of firefighters and building occupants by enabling
a better understanding of wind driven firefighting tactics, including structural ventilation and
suppression. This was achieved by investigating technical issues that address the teaching of the
dynamics of fire phenomena and prediction of fire intensity and growth under wind driven conditions.
The data from this research will also help to identify methods and promulgation of improved standard
operating guidelines (SOG) for the fire service to enhance firefighter safety, fire ground operations, and
use of equipment.

A series of 14 experiments were conducted in a seven story building to evaluate the ability of positive
pressure ventilation fans (PPV), wind control devices (WCD) and floor below nozzles (FBN) to mitigate
the hazards of a wind driven fire in a structure. Each of the 14 experiments started with a fire in a
furnished room. The air flow for 12 of the 14 experiments was intensified by a natural or mechanical
wind.. Each of the tools was evaluated individually as well and in conjunction with each other to assess
the benefit to fire fighters, as well as potential occupants in the structure.

These field experiments followed a series of experiments that were conducted in NIST’s Large Fire
Facility from November 2007 through January 2008. The NIST laboratory experiments demonstrated
and quantified the impact of wind on a fire in a structure and the impact of WCDs and externally applied
water streams on mitigating the hazard from a wind driven fire. The experiments were conducted in a
laboratory in order to provide the best levels of control on the experiments and have the capability of
making heat release rate and high quality gas concentration measurements which would be difficult and
cost prohibitive to make in an acquired structure in the field. The laboratory experiments are fully
documented in the report, “Fire Fighting Tactics Under Wind Driven Conditions: Laboratory
Experiments”[1].

As implied by the title, the laboratory experiments were only one portion of the research needed to fully
analyze the impact of wind on a fire resistive structure fire and demonstrate potential methods (tactics)
for improving firefighter safety and effectiveness. The experiments on Governors Island were required
to examine the effects of natural ventilation (door control) and PPV on the fire and the means of egress
in the building alone and in conjunction with the use of WCDs and FBNs. Tactics incorporating
devices, such as WCDs, to control the ventilation conditions or the use of a special fire nozzle from the
floor below the fire floor have been tried by the fire service under “real fire” conditions with varying
levels of success. Unfortunately, there is no data to understand the capabilities and limitations of these
fire fighting approaches. This study provides real scale data, to guide the development of appropriate
tactical operations for use under wind driven conditions.

1
1.1 Background 

Fires in high-rise buildings create unique safety challenges for building occupants and fire fighters.
Smoke and heat spreading through the corridors and the stairs of a building during a fire can limit
building occupants’ ability to escape and can limit fire fighters’ ability to rescue them. In 2002, there
were 7300 reported fires in high rise structures (structures 7 stories or more). The majority of these high
rise fires occurred in residential occupancies, such as apartment buildings. In fires that originated in
apartments, 92 % of the civilian fatalities occurred in incidents where the fire spreads beyond the room
of origin [2].

Changes in the building’s ventilation, such as the opening of doors or windows can increase the growth
of the fire and allow it to spread beyond the room of fire origin. This can also increase the spread of fire
gases through the building. In some cases, such as the Cook County Administration Building fire in
October 2003, the fire flow into the corridors and the stairway prevented fire fighters from suppressing
the fire from inside the structure. This fire resulted in 6 building occupant fatalities and several fire
fighter injuries in the stairway [3].

The failure of a window in the fire apartment in the presence of an external wind can create significant
and rapid increases in the heat production of a fire. Combined with open doors to corridors, stairs, or
downwind apartments, many wind driven fire incidents have resulted in fire fighter fatalities and injuries
[4, 5].

1.1.1 Historical Wind Driven Fires 

Recognition of wind driven fire conditions has been taken into account in forest fires and large area
conflagrations for more than 100 years. This is due in part to the fact that some of the most destructive
and deadly conflagrations in the United States such as the Great Pestigo, WI fire and the Great Chicago
fire were wind driven events. Both of these fires started on the same day, October 8, 1871. The Pestigo
fire resulted in 1152 fatalities and more than 1.2 million acres burned. The Chicago fire resulted in more
than 250 fatalities, and 17400 structures destroyed over a 2000 acre area [6].The magnitude of these
fires was, in part, the result of strong south winds combined with “tinder dry” conditions [7].

While wildland fire managers and officers training includes weather conditions in their evaluation of
incident conditions (size-up), typically structural firefighters and fire officers do not receive this type of
training [8, 9, 10, 11]. Wildland fire fighter training manuals dedicate almost half of their fire behavior
chapter to weather with significant sections on wind [12]. Structural fire fighter training manuals, which
are approximately 1000 pages in length, dedicate a page or less to the interaction of wind and structural
fire behavior [13, 14, 15]. As a result, structure fires that may have been affected by wind conditions
have typically not been recognized as such or well documented, with some notable exceptions. A few
such exceptions are presented in the following sections.

1.1.2 Experience of the Fire Department of New York City 

The Fire Department of New York City (FDNY) began to recognize that wind driven fires, particularly
those in multiple story, residential occupancies of fire resistive construction (Class I) were challenging

2
their resources, their tactics, and their safety. Norman and Tracy and others in the department began to
look at the challenges and results of wind driven fires, with the goal of changing the tactics in order to
improve the safety and effectiveness of their members [16, 17, 18]. A listing of notable FDNY wind
driven fire incidents is given in Table 1.1-1 [16, 19]. While it might appear that the frequency of
occurrence has increased, the reality may be that the recognition of wind driven fires has increased in the
department. In fact, the FDNY has developed a training DVD, Fighting Wind Driven Fires in High Rise
Multiple Dwellings, which was created in November of 2007 with the objective of developing an
awareness for wind effects in a structure, and identifying how to control the hazard or find shelter from
the hazard by controlling doors and preparing areas of refuge [20].

Another factor Norman [16] identifies is that the fire does not have to be 20 stories or more above
ground for wind to be a factor. Table 1.1-1 demonstrates that these FDNY wind driven fire incidents
have occurred as low as the 3rd story above ground. NFPA data shows that the majority of fires in high
rise buildings occur below the 7th floor [2].

Table 1.1-1. FDNY Wind driven fire incidents.

Date Location Victims Stories Fire Floor
1/23/80 30 Montrose Avenue, Brooklyn 1 civilian fatality 16 11th
23 Horace Harding
2/11/89 3 civilian fatalities 16 14th
Expressway, Queens
11/2/94 Park Ave, Bronx 2 civilian fatalities 20 18th
40-20 Beach Channel Drive,
1/5/96 1 firefighter fatality 13 3rd
Queens
1/7/97 1 Lincoln Place, Manhattan 18 firefighters injured 42 28th

12/18/98 77 Vandalia Avenue, Brooklyn 3 firefighter fatalities 10 10th

4 civilian fatalities, 9 firefighters
12/23/98 124 West 60th st, Manhattan 51 19th
injured
30 firefighters injured, 4 civilians
4/23/01 Waterside Plaza, Manhattan 37 24th
injured
20 Confucious Place,
9/9/04 12 firefighters burned 44 37th
Manhattan
40-20 Beach Channel Drive,
1/26/06 3 firefighters burned 13 6th
Queens
2/26/06 20 Moshulu Parkway, Bronx 3 firefighters burned 41 24th & 25th
1700 Bedford Avenue, 1 firefighter fatality 4 firefighters
1/03/08 25 14th
Brooklyn burned, 4 civilians injured
3/28/08 Grand Avenue, Manhattan 1 civilian fatality, 45 injured 26 4th

4/2/08 Sutter Ave, Brooklyn 3 firefighters injured 22 5th

Other wind related fire fighter line of duty deaths have occurred in New York City in smaller buildings
of ordinary construction (Type III) such as the “Black Sunday Fire.” This fire started on the third story
of a four story apartment building. The average wind speed was 12 mph with gusts up to 45 mph. Fire

3
fighters searching for victims on the floor above the fire reported that, “fire was blowing into the
hallway.” The rapid spread of fire to the 4th floor left 6 firefighters trapped. Their only option was to
deploy out of the windows to ground, resulting in the death of two of the firefighters and serious injuries
to the other four [21].

Buildings and topographical features alone or in combination deflect wind and as a result cause changes
in wind speed and direction or localized wind effects around a building. In cities, this may be referred to
as “building-spawned” wind. All buildings, regardless of size, can block wind, which may cause “local
areas of amplified winds around corners and enhanced turbulence in building wakes” [22]. When wind
hits the face of a structure it will seek the path of least resistance to move around it. For a multi-story
building with a flat face on the upwind side, it has been demonstrated that some of the wind will go over
the building, a portion of the wind will go around the building and a portion of the wind will be
deflected downward, and develop a vortex near the ground. [23].

1.1.3 U.S. Wind Driven Fire Experience 

These wind driven fire incidents are not limited to New York City. Houston, TX., st. Louis, MO., and
Prince William County, VA. are just a few of the other localities in the United States that have
experienced losses to wind driven fires [5, 24, 25, 26]. These incidents ranged from a fire that started on
the 5th floor of a 41 story, fire resistive building to a fire that started on the wood deck outside a two
story, wood frame, single family home.

Recently a near miss was documented in a wood frame Cape Cod-type house in Long Island, NY.
Firefighters working to extinguish the fire, making entry from the front of the house, had flames pushed
over them by the wind entering the structure from the rear [27].

A search of the National Fire Fighter Near-Miss Reporting system database also shows a variety of fire
incidents and structures where wind caused a significant change in fire conditions resulting in rapid
increases in thermal hazard to the fire fighters [28, 29, 30, 31, 32, 33, 34].

1.1.4 Wind Driven Tactics Research 

What tactics or tools are appropriate for use with a wind driven fire and how should the tactics or tools
be implemented? In order to answer this question, the problem has to be fully defined. The wind driven
fire hazard that has been examined occurs in a high rise building of fire resistive construction with
internal corridors and interior stairs. The Vandalia fire incident in which three FDNY fire fighters died
exemplifies scenarios which result in untenable conditions in a public corridor. The door to the fire
apartment was left open. As a result, there was nothing to keep the fire or the smoke contained in the
apartment of fire origin. A door to a stair was opened and the stair was vented to the outside or an
apartment on the downwind side of the building is opened. If the fire apartment was on the upwind side
of the building and the window failed a ventilation path would be in place for flames to sweep through
the apartment of origin and out into the corridor, making it impossible and untenable for firefighters to
approach the fire apartment.

4
Norman summarizes tactics that FDNY has researched to address this condition, 1) breaching, 2)
suppressing the fire with an exterior water stream, and 3) controlling the flow of wind into the fire
apartment with a window fire blanket or curtain [16]. Breaching involves making a hole from a
protected stair and continuing to breach walls until a hole for a hoseline can be made in the wall of the
fire apartment or the fire could be attacked from an adjoining balcony. Exterior hose streams have been
used when the fire apartment is in reach of an aerial apparatus stream. For apartments on higher floors,
an applicator pipe or Navy fog applicator may be used to apply water into the window of a fire
apartment on the floor above. The use of a wind control device deployed from the floor above the fire
floor to block an open window to a fire apartment on the upwind side of the building has been
researched by FDNY. In fact, the department issued wind control devices to special Operations Units
[16].

Positive Pressure Ventilation (PPV) is being used by fire departments on smaller structures, such as
single family homes, to control the fire flow by introducing pressure from the front door and venting the
house through a strategic exit opening. If done correctly, this tactic can remove significant amounts of
heat and smoke from the structure, thus improving the fire fighters’ working environment and improving
the chances of survival for the building occupants. NIST has completed several studies which have a
two fold impact: 1) providing guidance on the safe use of PPV and 2) characterizing and validating the
modeling of PPV with a computational fluid dynamics (CFD) model, so that the model can be used as a
training tool for the fire service [35, 36, 37, 38].

In 2006, NIST research began to examine the use of PPV in high-rise firefighting. To accomplish this
task, NIST partnered with the Chicago Fire Department (CFD), FDNY and the Toledo Fire and Rescue
Department. In a vacant 30-story high-rise in Toledo, the capability of PPV to pressurize the stair was
demonstrated in an extensive series of pressure experiments [39]. This study was followed with a series
of fire experiments conducted in a 16-story high-rise in Chicago [40]. The results of the fire
experiments demonstrated the ability of properly sized and placed PPV fans to pressurize stairways in a
high rise building and clear them of heat and smoke even with post-flashover fires open to the corridor
on the fire floor. Near the end of the test series in Chicago, experiments were conducted to examine the
impact of wind on an apartment fire and the potential for a wind control device and/or a large PPV fan to
control the hazard and protect the corridor. The experiments conducted on the 3rd floor demonstrated
that introducing a wind to a post flashover room fire can result in “blow torch” flames through the
apartment and into the corridor in less than 30 s. The experiments also showed that a wind control
device could in fact negate the impact of the wind and that PPV fans may have a role in mitigating the
hazard from wind driven fires.

In November 2007, the National Institute of Standards and Technology, with the support of the Fire
Protection Research Foundation and the U.S. Fire Administration, began conducting eight fire
experiments to examine the impact of wind on fire spread through a multi-room structure and examine
the capabilities of wind-control devices (WCD) and externally applied water to mitigate the hazard. The
principle measurements used to examine the impact of the WCDs and the external water application
tactics were heat release rate, temperature, heat flux, and velocity inside the structure. Measurements of
oxygen, carbon dioxide, carbon monoxide, total hydrocarbons and differential pressures were also
measured in the bedroom (room of fire origin) and in a furnished living room (room in the flow path).
Each of the experiments was recorded with video and thermal imaging cameras.

5
The experiments were designed to expose a public corridor area to a wind driven post-flashover
apartment fire. The door from the apartment to the corridor was open for each of the experiments. The
conditions in the corridor were of critical importance because that is the portion of the building that
firefighters would use to approach the fire apartment or that occupants from an adjoining apartment
would use to exit the building.

The fires were started in the bedroom of the apartment. Prior to the failure or venting of the bedroom
window, which was on the upwind side of the experimental apartment, the heat release rate from the fire
was on the order of 1 MW. After window failure, but prior to implementing either of the mitigating
tactics, the heat release rates from the post-flashover structure fire were typically between 15 MW and
20 MW. When the door from the apartment to the corridor was open, temperatures in the corridor area
near the open doorway, 1.52 m (5.00 ft) below the ceiling, were in excess of 600 °C (1112 °F) for each
of the experiments. The heat fluxes measured in the same location, during the same experiments, were
in excess of 70 kW/m². These extreme thermal conditions are not tenable, even for a firefighter in fully
protective gear. These conditions were attained within 30 s of the window failure.

In these experiments, the WCDs reduced the temperatures in corridor outside the doorway by more than
50 % within 60 s of deployment. The heat fluxes were reduced by at least 70 % during this same time
period. The WCDs also mitigated completely any gas velocity due to the external wind.

The externally applied water streams were implemented in three different ways; a fog stream across the
face of the window opening, a fog stream into the window opening, and a solid water stream into the
window opening. The fog stream across the window was not effective at reducing the thermal
conditions in the corridor. The fog stream in the window decreased the corridor temperature by at least
20 % and the corresponding heat flux measures by at least 30 %. The solid streams experiments resulted
in corridor temperature and heat flux reductions of at least 40 % within 60 s of application. None of the
water applications had a significant impact on reducing the gas velocities in the structure. In some cases
the gas velocity increased during water application.

These experiments demonstrated that the two potential tactics which could be implemented from either
the floor above the fire, in the case of a WCD, or from the floor below the fire, in the case of external
water application, were demonstrated to be effective in reducing the thermal hazard in the corridor.

The experiments documented below, conducted in a seven story building, analyzed the impact of wind
on actual structure fire and demonstrate potential tactics for improving firefighter safety and
effectiveness.

1.2 Structure 

The experiments were conducted in a 7-story fire resistant residential apartment building located on
Governors Island in New York Harbor, south of Manhattan and west of Brooklyn. This location was
important due to the prevailing winds during the winter months. The structure was located on the
northwestern side of the island as shown in Figure 1.2-1. The structure was “dog bone” shaped and was
constructed of concrete block with concrete slab floors. Each elevation of the structure can be seen in
Figure 1.2-2 through Figure 1.2-5. More detailed drawings with dimensions are located in Appendices
Appendix A – Detailed Drawings.

6
The structure was symmetric about the fire wall in center of the building. Each side had 6 apartments (A
through F and G through L), a scissor stairwell and an elevator bank. There were no openings between
the two sides of the structure; therefore they were treated as two separate experimental volumes for the
analysis. The floor plan was the same for floors 2 through 7. For the experiments, the first floor had all
of the apartments and storage rooms sealed off and only utilized the lobby and the stairwells. All four
stairways extended up above the seventh floor and had a bulkhead doorway that opened up to the roof.
A detail view of the scissor stairwell can be seen in Figure 1.2-7. Opening the bulkhead door was
important for flow path control of smoke and hot gases.

Apartments A, E, G and K were furnished and utilized for the experiments. Apartments A and G were
two bedroom apartments with a living room, kitchen and bathroom. Apartments E and K were three
bedroom apartments with a living room, kitchen and bathroom. In each of the experiments the kitchen
and bathroom were walled off with gypsum board to minimize the amount of overhaul that was required
between experiments. Figure 1.2-8 shows the experimental floor plan with the areas of the building that
were not utilized removed for clarity. The doors to the apartments not utilized for the experiments
always remained closed.

All of the walls in the apartments were made of painted gypsum board and the walls in the public
corridors and stairwells were painted concrete block. All of the doors to the apartments were steel and
all of the doors to the bedrooms were hollow core wood doors. The doors to the bedrooms being used
for the experiments were left open as well as the fire apartment entrance door. The opening and closing
of other doors was noted in the timelines.

All of the windows were double pane in aluminum frames. In the bedrooms used for ignition, there
were two windows, referred to as a “double” window with 4 pane sections and a “single” window with
two pane sections. The double window was closest to the point of ignition and in the wind driven
experiments was on the upwind side. Each of the pane sections in the double window measured 0.61 m
(2.00 ft) wide and 0.53 m (1.75 ft) high. The single window was located on the adjacent exterior wall.
Each of the two pane sections, an upper and lower, measured 0.71 m (2.33 ft) wide and 0.53 m (1.75 ft)
high.

The living room had a larger window which was divided in to 5 glass sections, a single pane center
section with a windows on each side that had an upper and lower pane section The glass pane area of
the center section measured 0.83 m (2.71 ft) wide and 1.22 m (4.00 ft) high. Each of the side panes
measured 0.56 m (1.83 ft) wide and 0.53 m (1.75 ft) high.

The distance between the floor and the lower edge of the glass portion of all of the windows in the
bedrooms and the living rooms was 0.91 m (3.00 ft). The dimensions of overall window frames and
window openings in the brick façade are provided in the detailed drawings in Appendix A.

7
Figure 1.2-1. Governors Island Map (Experimental Structure Circled)

Figure 1.2-2. Front of structure (Side A) Figure 1.2-3. Left side of structure (Side B)

Figure 1.2-4. Rear of structure (Side C) Figure 1.2-5. Right side of structure (Side D)

8
Figure 1.2-6. Floor Plan for Floors 2-7 (Fire Apartments Circled)

Figure 1.2-7. Detail of scissor stairs

9
 Figure 1.2-8. Cut Away View of Experimental Area

1.3 Wind 

The wind for these experiments was essential. Ideally the natural wind would be sufficient to create the
wind driven fire conditions for the experiments. Historically and during previous site visits the natural
wind was from the desired direction at the desired magnitudes. Unfortunately the natural wind was not
present during the week in which these experiments were scheduled and the option of waiting for the
right wind was not an acceptable. There was one experiment in which the natural wind was sufficient
and that was the experiment in Apartment 3G/K.

The direction and magnitude of the wind was measured in multiple locations by representatives of
Polytechnic University. They had anemometers on and around the structure to measure the prevailing

10
winds prior to the experiments for planning purposes and during the experiments. Their data can be
found in Appendix C – Wind Measurements and Descriptions.

Because natural wind was not available, wind was created mechanically with the use of a Mobile
Ventilation Unit (MVU). The MVU was a 1.2 m (48 in) gasoline powered fan that was able to provide a
simulated 9 m/s to 11 m/s (20 mph to 25 mph) wind that was repeatable for all of the experiments
(Figure 1.3-1 through Figure 1.3-3).

The MVU was secured to a lift which was able to position the fan as high as the 7th floor to allow the
simulated wind to be directed straight into the desired opening. The MVU was positioned 4 m to 8 m
from the opening and the fan speed was adjusted to achieve the 9 m/s to 11 m/s (20 mph to 25 mph)
wind in the opening. The MVU was positioned to be centered on the opening and varied slightly based
on the obstructions around the building and the terrain.

Figure 1.3-1. MVU on Lift Figure 1.3-2. MVU Positioned at Window

Figure 1.3-3. Close-up View of MVU

11
1.4 Fuel Load 

The fuel load was comprised of common items that may be found in an apartment building. The corner
bedroom and the living room of each of the apartments utilized (A, E, G and K) were furnished.
Apartments G and K had a second bedroom furnished (Figure 1.2-8). The bedrooms had an average fuel
load of 21 kg/m2 (4.3 lb/ft2) and the living rooms had an average fuel load of 15 kg/m2 (3.1 lb/ft2). This
accounted for the combustible furnishings and not the metal objects such as the bed frames or lamps.
The combustible wood floor was also not included in these averages. Table 1.4-1 shows all of the items
used and their respective labels. Figure 1.4-1 through Figure 1.4-4 show the floor plan and layout of the
fuel load items. In these figures the fuel load items are labelled letters that correspond to Table 1.4-1.
For example, the bed in each bed room is labeled “a-e”. This means that items a through e from Table
1.4-1 are co-located in that position. In this example the mattress, box spring, head board, bedding and
pillows are used for the bed fuel package. Figure 1.4-5 through Figure 1.4-32 show photographs of each
apartment and the fuel load arrangement. These figures are representative of the similar layouts for each
floor. For example the layout for Apartment 7A is similar to 5A and 3A.

12
Table 1.4-1. Fuel Load Labels and Description
H W D
Label Item (m) (m) (m) Materials Mass (kg)
A Mattress 0.2 1.5 2.0 fiber pad/polyurethane foam 30.4
B Box spring 0.2 1.5 2.0 wood / felt 31.7
C Head Board 1.1 1.4 0.0 wood 17.3
D Bedding * 2.6 2.0 - cotton/polyester 5.0
E Pillows -2 0.3 0.7 0.1 cotton/polyester 1.0
Upholstered Chair
F (Brown) 0.7 0.7 0.7 polyurethane foam/wood 24.1
G Upholstered Chair (Red) 0.8 0.9 0.9 polyurethane foam/wood 28.9
H Chair (Red w/ wood arms) 0.8 0.6 0.8 polyurethane foam/wood 14.5
I Desk Chair (Green) 1.1 0.6 0.6 polyurethane foam/wood 16.7
J Desk Chair (Brown) 0.7 0.4 0.4 wood 7.0
K Coffee Table 0.6 0.6 0.5 particle board/wood 9.7
Desk (Dark Brown w/
L blue) 0.8 1.3 0.8 particle board/wood 29.9
M Desk (Light Brown) 0.8 1.1 0.6 particle board/wood 26.4
N Dresser (Dark Brown) 0.8 1.5 0.5 particle board/wood 80.9
O Dresser (Light Brown) 0.6 1.8 0.5 particle board/wood 73.3
P Nightstand (Light Brown) 0.6 0.5 0.4 particle board/wood 20.9
Q Nightstand (Dark Brown) 0.6 0.7 0.5 particle board/wood 32.6
R Round Table 0.6 0.7 0.7 particle board/wood 12.1
S Sofa 0.9 1.8 1.0 polyurethane foam/wood 53.0
T TV (Small) 0.4 0.7 0.4 Plastic 23.3
U TV (Large) 0.6 0.7 0.5 Plastic 36.9
V Trash Can - - - Plastic 0.3
* nominal H and W area of sheets and comforter

13
Figure 1.4-1. Apartment A Fuel Layout Figure 1.4-2. Apartment K Fuel Layout

Figure 1.4-3. Apartment G Fuel Layout Figure 1.4-4. Apartment K Fuel Layout

14
1.4.1 Apartment A Photographs 

Figure 1.4-5. View From Doorway Figure 1.4-6. Living Room (View 1)

Figure 1.4-7. Living room (View 2) Figure 1.4-8. View Down Hallway

Figure 1.4-9. Ignition Bedroom (View 1) Figure 1.4-10. Ignition Bedroom (View 2)

15
1.4.2 Apartment E Photographs 

Figure 1.4-11. View From Doorway Figure 1.4-12. Living Room (View 1)

Figure 1.4-13. Living Room (View 2) Figure 1.4-14. Ignition Bedroom (View 1)

Figure 1.4-15. Ignition Bedroom (View 2) Figure 1.4-16. Ignition Bedroom (View 3)

16
1.4.3 Apartment G Photographs 

Figure 1.4-17. View From Doorway Figure 1.4-18. Living Room (View 1)

Figure 1.4-19. Living Room (View 2) Figure 1.4-20. Target Bedroom (View 1)

Figure 1.4-21. Target Bedroom (View 2) Figure 1.4-22. Ignition Bedroom (View 1)

Figure 1.4-23. Ignition Bedroom (View 1) Figure 1.4-24. Ignition Bedroom (View 3)

17
1.4.4 Apartment K Photographs 

Figure 1.4-25. View From Doorway Figure 1.4-26. Living Room (View 1)

Figure 1.4-27. Living Room (View 2) Figure 1.4-28. Target Bedroom (View 1)

Figure 1.4-29. Target Bedroom (View 2) Figure 1.4-30. Ignition Bedroom (View 1)

Figure 1.4-31. Ignition Bedroom (View 2) Figure 1.4-32. Ignition Bedroom (View 3)

18
1.5 Instrumentation 

The measurements taken during the experiments included differential pressure, gas temperature, gas
velocity, video recording and thermal imaging. Differential pressure transducers were located
throughout the stairwells, 1.2 m (4 ft) above the floor on every landing (Figure 1.5-1). A copper tube
was run from the transducers through the wall from the D or J Apartment into the stairwell to measure
the pressure increase.

Gas temperature was measured with bare-bead, Chromel-Alumel (type K) thermocouples, with a
0.5 mm (0.02 in) nominal diameter. Thermocouple arrays were located in the ignition bedrooms, public
hallways and stairwells (Figure 1.5-4). Each thermocouple location had an array of thermocouples with
measurement locations of 0.03 m, 0.3 m, 0.6 m, 0.9 m, 1.2 m,1.5 m, 1.8 m and 2.1 m (1 in, 1 ft, 2 ft, 3 ft,
4 ft, 5 ft, 6 ft and 7 ft) below the ceiling (Figure 1.5-2).

Gas velocity was measured utilizing differential pressure transducers connected to bidirectional velocity
probes. These probes were located in the fire apartment doorways, doorways into the stairwell as well
as the stairwell doorways of the fire floors. There were three probes on the vertical centerline of each
doorway located at 0.3 m (1 ft) from the top of the doorway, the center of the doorway and 0.3 m (1 ft)
from the bottom of the doorway. Thermocouples were co-located with the bidirectional probes to
complete the gas velocity measurement.

Video cameras and thermal imaging cameras were placed inside and outside the building to monitor
both smoke and heat conditions throughout each experiment. As many as eight video camera views and
two thermal imaging views were recorded during each test (Figure 1.5-3).

Figure 1.5-1. Pressure Transducer

Figure 1.5-3. Video and Thermal Imaging Cameras Figure 1.5-2. Bidirectional Probes and Thermocouples

19
 Figure 1.5-4. Instrumentation Locations on Floors 3, 5 and 7

1.6 Estimated Measurement Uncertainty 

There are different components of uncertainty in the length, differential pressure and gas temperature
data reported here. Uncertainties are grouped into two categories according to the method used to
estimate them. Type A uncertainties are those which are evaluated by statistical methods, and Type B
are those which are evaluated by other means [41]. Type B analysis of systematic uncertainties involves
estimating the upper (+ a) and lower (- a) limits for the quantity in question such that the probability that
the value would be in the interval (± a) is essentially 100 %. After estimating uncertainties by either
Type A or B analysis, the uncertainties are combined in quadrature to yield the combined standard
uncertainty. Multiplying the combined standard uncertainty by a coverage factor of two results in the
expanded uncertainty which corresponds to a 95 % confidence interval (2σ).

Each length measurement was taken carefully. However due to some issues, such as obstructions and
unleveled terrain there was a total expanded uncertainty of ± 6 % associated with the measurements.

20
 Differential pressure reading uncertainty components were derived from pressure transducer instrument
specifications and previous experience with pressure transducers. The transducers were factory
calibrated and the zero and span of each was checked in the laboratory prior to the experiments yielding
an accuracy of ±1% [42]. The total expanded uncertainty was estimated at 10 %.

The standard uncertainty in temperature of the thermocouple wire itself is ± 2.2 °C at 277 °C and
increases to ± 9.5 °C at 871 °C as determined by the wire manufacturer [43]. The variation of the
temperature in the environment surrounding the thermocouple is known to be much greater than that of
the wire uncertainty [44, 45]. Small diameter thermocouples were used to limit the impact of radiative
heating and cooling. The estimated total expanded uncertainty for temperature in these experiments is
±15 %.

Bi-directional probes and single thermocouples were used to measure the velocity. The bi-directional
probes used similar pressure transducers as those used for the differential pressure measurements
discussed above. Bare-bead Type K thermocouple are co-located with the probe. The estimated total
expanded uncertainty for temperature in these experiments is ±15 %.

An error bar representative of the estimated uncertainty for each measurement is given on every data
graph. The bar is applied near the peak measurement on the graph.

1.7 Positive Pressure Ventilation Fans 

Several of the same model 0.7 m (27 in) positive pressure ventilation (PPV) fans were utilized for these
experiments (Figure 1.7-1 and Figure 1.7-2). They were powered by a 9.0 hp gasoline engine and were
used to ventilate and pressurize the stairwells of the structure during the experiments. Fans were
positioned 1.8 m (6 ft) from the doorways that they were flowing into and tilted back 10 degrees from
vertical. Fans were located at the front lobby doors, base of the stairwells on the first floor and two
floors below the fire floor depending upon the experimental configuration.

Previous research has shown that PPV fans positioned correctly are able to create a pressure sufficient
enough to keep products of combustion out of the stairwell by creating a higher pressure than the fire
creates [39, 40]. These experiments follow the guidelines produced by the previous research to
determine if the fans can be effective at keeping the combustion products out of the stairwell under wind
driven conditions.

21
 Figure 1.7-1. Front View of PPV Fans Figure 1.7-2. Rear View of PPV Fans

1.8 Wind Control Blankets/Curtains 

Two different Wind Control Devices (WCD) were used during these experiments. The first was a Wind
Control Blanket (WCD). The WCD measured 3.0 m (10 ft) by 3.7 m (12 ft), was made of a silica fabric
aluminized with a foil material and weighed approximately 14.0 kg (30.9 lb). It had a strap on each
corner and one in the top center to secure it in the desired location. The WCD required a team of at least
two fire fighters positioned above and a team of at least two fire fighters below the fire apartment to
deploy and secure the blanket. One of the 3.7 m (12 ft) sides of the WCD was weighted with a chain to
assist with deployment.

The second device was the Wind Control Curtain (WCD). The WCD measured 1.8 m (6 ft) by 2.4 m
(8 ft), was made of a treated fiberglass material and weighed approximately 11.8 kg (26 lb). It was
reinforced with metal rods and had a rope on each corner to secure it. The WCD could be deployed by a
single fire fighter above the fire apartment and required an additional fire fighter below to secure the
bottom of the curtain.

2.4 m (8 ft)

1.8 m (6 ft)

Figure 1.8-1. Wind Control Device Stowed Figure 1.8-2. Wind Control Devi ce Open

22
 3.7 m (12 ft)

3.0 m (10 ft)

Figure 1.8-3. Wind Control Device Stowed Figure 1.8-4. Wind Control Device Open

Figure 1.8-5. Wind Control Device Deployed Figure 1.8-6. Wind Control Device Deployed

1.9 Floor Below Water Application Nozzles 

Two different types of Floor Below Nozzles (FBN) were utilized. The first, referred to as FBN 1, was a
bent pipe with the ability to attach a nozzle of choice. This FBN was designed to be positioned straight
out of the floor below the fire and by adjusting the distance out of the window the crew can direct the
stream into the floor above (Figure 1.9-1).

The second type of FBN, referred to as FBN 2, was also of a bent pipe design. The difference was that
this nozzle was designed to hook onto the window sill of the floor above. There were several prototypes
of the FBN 2, varying the configuration to get the tip of the nozzle into the above apartment, but for the
purposes of this analysis, they both will be grouped together (Figure 1.9-2).

23
Both types of FBN were able to accommodate any tip or water distribution device. Both a smooth bore
and a fog tip were used for these experiments. The smooth bore was a 1 1/8 inch tip (Figure 1.9-1). The
fog tip was considered any tip that created a broken water stream. Three versions of the fog tip that
were used included a navy nozzle (Figure 1.9-3), sprinkler head type tip (Figure 1.9-4) and a common
fog tip (Figure 1.9-7). Figure 1.9-5and Figure 1.9-6 show each of the FBN operating into the floor
above.

Figure 1.9-1. FBN 1 w/ a Smooth Bore Nozzle Figure 1.9-2. One version of FBN 2

Figure 1.9-3. FBN 2 with a Fog Nozzle (Navy Nozzle) Figure 1.9-4. FBN 2 with a Fog Nozzle (Sprinkler)

24
Figure 1.9-5. Floor Below Nozzle 1 Flowing Figure 1.9-6. Floor Below Nozzle 2 Flowing

Figure 1.9-7. FBN 1 w/ Fog Tip Figure 1.9-8. Another Version of FBN 2

25
2 Experimental Procedure 

Prior to the ignition of each experiment the MVU was placed into position and set to the appropriate
speed. Three experiments 7G, 7G2 and 3G/K were an exception to this and utilized the natural wind.
The computer acquisition system was started and data were collected from each instrument every
second. Video cameras recording the experiment were also started at this time and monitored
throughout the experiment. After the recording of background data, ignition was initiated remotely.
Most of the experiments had the fire ignited in the corner bedroom of the apartment next to the bed.
When the fire was initiated in the living room, the point of ignition was on the seat of one of the sofas.

The ignition source consisted of a cardboard book of 20 matches that was ignited by an electrically
heated wire. The matchbook was placed near the bottom of an 8.5 L (9 qt) polypropylene waste
container with a mass of 0.315 kg. The height of the waste container was 0.3 m (10.5 in) with interior
dimensions at the top opening of 0.23 m (9.0 in) by 0.20 m (8.0 in). Approximately 0.3 kg of dry
newspaper was added to the waste container. The majority of the newspaper was folded flat, and placed
on edge along the sides of the waste container. Four sheets of newspaper were crumpled into “balls”
approximately 100 mm in diameter and placed on top of the electric match in the center of the waste
container. Heat release rate experiments were conducted for this fuel package under an oxygen
consumption calorimeter at NIST. The peak heat release of the waste container and the newspaper was
on the order of 30 to 35 kW for the two heat release rate experiments conducted. [1]

Every experiment began with all of the doors and windows closed with the exception of the door to the
furnished bedrooms and the main entrance door from the public corridor to the fire apartment. After
ignition, the fire was allowed to grow until it failed the windows in the room or became ventilation
limited. The purpose of the experiments was to get a wind driven fire condition. Therefore when an
apartment became ventilation limited because of lack of window failure additional doors were opened
remotely such as the stairwell door to provide additional air for the fire to grow. In some cases,
windows were manually vented to provide additional air flow.

Once wind driven conditions were achieved many different openings were made to simulate the
operations of a fire department. These operations included opening the front lobby doors for access,
entering the stairwell, opening the door from the stairwell to the fire floor, and opening the bulkhead
door at the top of the stairwell. In addition to these operations each of the tools to be tested, PPV fans,
Wind Control Devices and Floor Below Nozzles were deployed and evaluated.

The experiments were run until the fire burned down and smoke production was minimal. The
experimental duration varied between 23 min and 53 min depending on the growth of the fire and the
impact of the tactics. Detailed tables of the events are documented chronologically in the experimental
timelines.

26
3 Results 

During the series of wind driven fire experiments on Governors Island a large number of parameters
which affect the flow of combustion products through a building and the ability to control the heat and
velocity of those products of combustion were examined. In the data, the impacts of controlling doors,
Positive Pressure Ventilation fans (PPV), Wind Control Devices (WCD), and Floor Below Nozzles
(FBN) can be seen.

An overview of each experiment is provided with a list of experimental objectives. A figure in each
experiment’s results shows the pertinent locations such as doors and windows that change during the
events of the experiment. A timeline shows all of the events that take place to accomplish the
experimental objectives and the time at which they occur. The figures after the timelines show the
locations for the cameras and instruments that are presented in the observation, temperature, pressure,
velocity and discussion sections.

The temperature section has three graphs for each experiment. Each graph has lines and labels that
indicate the changing events during the experiment. The first presents temperatures versus time
throughout the fire apartment, public corridor and fire floor landing inside the stairwell at 2.1 m (7 ft)
above the floor. This was the temperature approximately 0.3 m (1 ft) below the ceiling. The second
graph presents the temperature at the same locations as the first graph but at the 1.2 m (4 ft) elevation
above the floor. The third graph shows the temperatures at 1.2 m (4 ft) above the floor on the stairwell
landings of floors 1, 2, 4, 6 and the top of the stairwell or at the bulkhead door.

The pressure section has a graph that shows the pressure versus time at 1.2 m (4 ft) above the landing on
each floor. The graph has lines and labels that indicate the changing events during the experiment. The
pressures in the stairwell are of the most interest when the PPV fans are being utilized to pressurize the
stairwell to slow or stop the flow of fire gases into the stairwell.

The velocity section has a graph that shows the average doorway velocity versus time for the doors at
the base of the stair, top of the stair, entrance to the stair from the fire floor and from the apartment to
the public corridor. Velocity measurements were taken 0.3 m (1 ft) from the top of the door, 0.3 m (1 ft)
from the bottom of the door and in the middle of the door on the centerline of the doorway. These
measurements were averaged to provide the bulk flow through the doorway.

Detailed graphs are provided for each measurement point for each experiment in Appendix D – Detailed
Graphs. The graphs include time histories of temperatures in multiple locations on the fire floor,
temperatures throughout the stairwell, velocities of the fire floor, velocities in the stairwell and pressures
in the stairwells. The legends indicate each instruments height above the floor in meters. The vertical
axis of the temperature graphs provides the temperature in Centigrade on the left and Fahrenheit on the
right. The vertical axis of the velocity graphs provides the velocity in meter per second of the left and
miles per hour on the right. The horizontal axis provides the time in seconds. The experiment event
timeline is given across the top of each graph. Time “zero” was when ignition took place.

27
3.1 Experiment 7G 

The first experiment in the structure was located on the 7th floor in apartment 7G. Experiment 7G was
ignited in the bedroom furthest from the open apartment door (Figure 3.1-1). This experiment was
different from the other experiments, in that no external wind was being imposed to the structure and
there was little to no natural wind. The events that took place during the experiment to accomplish the
experimental objectives bulleted below are shown in Table 3.1-1.

Experimental Objectives
• Examine fire development with no mitigation and little/no wind
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Utilize portable PPV fans to control smoke movement
• Utilize a MVU to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.1-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

28
Figure 3.1-1. 7th floor apartment G floor plan with locations pertinent to Experiment 7G

29
Table 3.1-1: Experiment 7G Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
344 345 Bedroom double Bottom left pane failed
365 372 Front lobby Doors opened
402 405 1st floor stair A Door opened
438 439 Bedroom double Top left pane failed
439 437 Bedroom double Top right pane failed
847 849 Bedroom double Bottom right pane failed
970 973 Bedroom single Top pane failed
1195 1196 Bedroom single Bottom pane failed
1224 1225 7th floor stair A Door opened 0.08 m (3 in)
1302 1307 Bulkhead door Opened
1346 1352 7th floor stair A Door opened
1533 1542 1st floor stair A 27 inch fan activated
1569 1573 Bulkhead door Closed
1715 1719 7th floor stair A Door closed
1839 1845 7th floor stair A Door opened
1847 1854 Bulkhead Door Opened & closed
2060 2069 5th floor stair A Door opened
2060 2069 5th floor stair A 27 inch fan activated
2306 2313 Front lobby MVU started
2401 2405 5th floor stair A 27 inch fan deactivated
2401 2405 5th floor stair A Door closed
2475 2487 Front lobby MVU stopped
2600 2600 Experiment completed

30
Figure 3.1-2. Instrumentation and camera locations with labels.

31
3.1.1 Observations 

The observations are presented as a series of images captured from eight camera locations, six were
video cameras and two were thermal imaging cameras. The camera positions inside the building are
shown in Figure 3.1-2.

Figure 3.1-3 through Figure 3.1-16 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 44 min (2640 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom
with a view of the single window straight across, double window to the right and ignition trashcan at the
foot of the bed. The top right view (LIVINGRM) was of the living room looking across to the windows.
The corridor and corridor thermal imaging views (CORRIDOR and CORR IR) show the corridor
leading up to the open apartment doorway. The next two views (STAIR and STAIR IR) show the
stairwell with a visual and a thermal imaging camera from the floor below the fire looking up to the fire
floor. The bottom left view (ROOF) shows the bulkhead door at the top of the stairwell and the bottom
right view (OUTSIDE) shows an external view of the fire apartment.

Figure 3.1-3 shows the conditions at the time of ignition. At this point, the six video views are clear and
unobstructed. However, the thermal images provided limited thermal contrast because the surfaces in
the view are at nearly equal temperature.

The images in Figure 3.1-4 were captured 240 s (4 min) after ignition. The fire from the trash container
spread to the bed. A smoke layer formed in the bedroom and descended to the floor level. Smoke
spread through the living room and out into the corridor with a layer height of approximately 0.9 m (3 ft)
above the floor. The corridor IR view showed heat leaving the top of the apartment doorway and
entering the corridor. The stair views show little smoke in the stair and no heat entering the stair.

Figure 3.1-5 shows the images at 480 s (8 min) after ignition. The fire in the bedroom was ventilation
limited and visibility was greatly reduced on the fire floor. Three of the four panes of glass in the
bedroom double window had failed and smoke and heat was seeping around the gaps in the stairwell
door. Smoke could also be seen leaking around the door at the roof level.

The images in Figure 3.1-6 were captured 720 s (12 min) after ignition. The fire in the bedroom was
receiving oxygen from the failed double window and was beginning to increase in size. The smoke
layer was still to the floor in the living room and the corridor. The light in the living room was
reflecting off of the smoke but there was no fire in the living room. The corridor IR camera was
saturated with heat but the door to the apartment was still visible. The smoke in the stairwell was
moving up the stairwell toward the roof and the door to the stairwell was heating up.

Figure 3.1-7 shows the images at 960 s (16 min) after ignition. The single window in the bedroom had
failed and the fire in the bedroom was continuing to grow on the bed. There was still zero visibility in
the living room or corridor. The corridor IR view shows heat continuing to flow into the corridor. The
conditions in the stairwell remained the same from the previous figure and no flames are visible coming
from the bedroom windows.

32
Figure 3.1-8 shows the images at 1200 s (20 min) after ignition. The bed was fully involved in flames
and the room was transitioning to flashover. Heat was continuing to flow into the corridor and visibility
was still poor. There was increased smoke and heat leaking into the stairwell. Flames were burning out
of the single bedroom window but only light smoke was visible from the double bedroom window.

The images in Figure 3.1-9 were captured 1440 s (24 min) after ignition. The door to the stairwell had
been partially open for 120 s (2 min) and fully open for 90 s (1.5 min) and the bulkhead door had been
open for 150 s (2.5 min). The bedroom had transitioned to flashover and was free burning. The amount
of smoke in the living room and corridor was greatly reduced and visibility was improved as the fire was
receiving plenty of air from the open windows and the stairwell. Increased heat was flowing into the
corridor and cool air was entering the fire apartment through the bottom of the apartment doorway. The
smoke and heat entering the stairwell was increased as the door was opened. The smoke was moving up
through the stairwell and out of the open bulkhead doorway. The camera on the roof was completely
obscured by smoke coming out of the stairwell. The flames had pulled into the bedroom and increased
smoke was coming out of the double bedroom window.

The images in Figure 3.1-10 were captured 1530 s (25.5 min) after ignition. This time was just prior to
PPV fan activation. The bedroom was still well involved in flames but the flames pulled back from
burning the gases to the surfaces of the items in the room. Visibility in the living room and corridor
remained the same as the previous figure but the same amount of heat and smoke was flowing into the
corridor and stairwell. Visibility was returning to the roof area and the smoke flowing from the
bedroom windows was decreased.

Figure 3.1-11 shows the images at 1650 s (27.5 min) after ignition. At this time the PPV fan at the base
of the stair has been operating for 120 s (2 min) and the bulkhead door was closed for 90 s (1.5 min).
The fan has increased the pressure in the stairwell stopping the flow of smoke and heat into the stairwell.
Visibility was improved in the corridor but the fire is still forcing heat out of the fire apartment and into
the corridor. Residual smoke in the stairwell was being forced out of the gaps around the roof door.

Figure 3.1-12 shows the images at 1680 s (28 min), 30 s after the previous figure. Conditions continued
to improve in the stairwell as the smoke and heat flow we stopped with the use of the PPV fan.

Figure 3.1-13 shows the images at 1920 s (32 min) after ignition. At this time the PPV fan at the base of
the stair has been operating for 380 s (6.3 min). The conditions continue to improve as the fire burns
down. The heat from the fire is still able to exit the fire apartment but is not able to enter the stairwell.
Figure 3.1-14 shows the images at 2160 s (36 min), after a second portable PPV fan was added at the 5th
floor flowing into the stairwell. This added flow stops the flow of heat out of the fire apartment and
reverses all flow back into the apartment. The stairwell and corridor now had complete visibility.

The final 2 figures, Figure 3.1-15 and Figure 3.1-16 show the conditions after the MVU was used
outside of the structure blowing into the front doors. All of the smoke was cleared up to the bedroom
and conditions were returned to near ambient. The fire had reduced to a pile of burning embers. The
stair IR camera was knocked over by the flow from the MVU so that image was lost. There was no fire
extension beyond the room of origin.

33
Figure 3.1-3. Experiment 7G, ignition.

34
Figure 3.1-4. Experiment 7G, 240s after ignition.

35
Figure 3.1-5. Experiment 7G, 480s after ignition.

36
Figure 3.1-6. Experiment 7G, 720s after ignition.

37
Figure 3.1-7. Experiment 7G, 960s after ignition.

38
Figure 3.1-8. Experiment 7G, 1200s after ignition.

39
Figure 3.1-9 Experiment 7G, 1440s after ignition.

40
Figure 3.1-10 Experiment 7G, 1530s after ignition.

41
Figure 3.1-11 Experiment 7G, 1650s after ignition.

42
Figure 3.1-12 Experiment 7G, 1680s after ignition.

43
Figure 3.1-13 Experiment 7G, 1920s after ignition.

44
Figure 3.1-14 Experiment 7G, 2160s after ignition.

45
Figure 3.1-15 Experiment 7G, 2400s after ignition.

46
Figure 3.1-16 Experiment 7G, 2640s after ignition.

47
3.1.2 Temperatures 

Figure 3.1-17 through Figure 3.1-19 provide temperature data for five measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.1-2. The events that occurred during the experiment, as presented in
Table 3.1-1, are listed on the top of the figures with lines at the time of the event. A time history for
each thermocouple at each measurement location is located in Appendix D – Detailed Graphs.

Figure 3.1-17 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 7K in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
temperature in the bedroom rose to above 650 ºC (1202 ºF) before becoming ventilation limited and
decreasing. At 345 s, the bottom left pane in the bedroom failed causing the temperature to decrease to
approximately 400 ºC (752 ºF) as the fire responded to the availability of the cool outside air.
Additional window panes failed and the fire grew and increased the temperature to above 800 ºC
(1472 ºF). The fire became ventilation limited again and decreased back down to 650 ºC (1202 ºF) until
the single window failed providing more oxygen. The temperature remained above 800 ºC (1472 ºF) for
approximately 500 s. Once the fan was activated at the base of the stairwell with the 7th floor door open,
the temperatures in the bedroom cooled to 600 ºC (1112 ºF). This decrease in temperature cannot be
entirely attributed to the fan as the fire was beginning to burn down at the time of fan activation.

The three temperature measurement locations in the corridor behaved similarly which is consistent with
the steady two layer environment that existed remote from the fire. Temperatures steadily increased as
the fire in the bedroom increased. Temperatures increased to approximately 200 ºC (392 ºF) up until the
fan was activated. The fan flow decreased the corridor temperatures to around 100 ºC (212 ºF) before
the stairwell door was closed and the temperatures recovered. When the stairwell door was opened
again and the fan flow could cool the temperatures decreased up until the end of the experiment.

The temperature in the stairwell remained near ambient until the stairwell door was opened. Once open
the stairwell temperature at 2.1 m (7 ft) above the landing increased to 150 ºC (302 ºF). The fan
activation decreased this temperature back down to ambient even with the door completely open on the
fire floor. Closing and opening the door on the fire floor caused short duration temperature spikes as the
pressure barrier created by the fan equalized the flow.

Figure 3.1-18 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. The bedroom temperature increased
to 400 ºC (752 ºF) before the fire became ventilation limited. The temperatures reacted to the failing
windows and then the 1.2 m (4 ft) temperature began to track with the 2.1 m (7 ft) temperature after
1000 s which indicates flashover conditions in the bedroom. Beyond 1000 s the temperatures reacted
the same as the previous figure.

The temperatures in the corridor steadily increased to approximately 150 ºC (302 ºF), until
approximately 1350 s when the 7th floor stair door (bulkhead) was opened. At this time the temperature
near the stairwell slightly decreased and the temperature near the fire apartment slightly increased. The
activation of the fan decreased all of the corridor temperatures down to approximately 50 ºC (122 ºF).

48
Removing the fan flow by closing the door allowed temperatures to increase to around 100 ºC (212 ºF)
before returning to ambient by opening the stair door.

The stairwell temperatures at 1.2 m on the landing of the fire floor remained below 100 ºC (212 ºF) with
the door open and no mechanical ventilation but decreased to ambient with the fan operating. The
addition of another fan or the MVU had no impact because the fire had burned down by the time of their
implementation.

Figure 3.1-19 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, 6 and at the
bulkhead. As the fire was initially growing some hot fire gases leaked into the stairwell and increased
the temperature at the top of the stairwell by 20 ºC (36 ºF). The cool air being pulled into the stairwell
through stack effect was cooling the lower floors of the structure. When the 7th floor door was partially
opened at 1225 s the temperature at the top of the stair increased again to 50 ºC (122 ºF) and cooled
briefly as the bulkhead door was opened and then increased to a peak of 90 ºC (194 ºF). The fan
activation decreased this temperature back down to 20 ºC (68 ºF) and kept it low for the remainder of the
experiment.
Ex
S7 BH pe
BR S1 BR B R B :o :o S S5 r im
:d : R p 5 : MV ent
BR oub LBY st air : dou : sing : sin peni BH S7 S B S S 7
e n i
: o p MV c l o s c
: I g le : o o b l g ng : o : o 1 H 7 : o g n e n U e d U s om
nit fail pen pen le cle e fail le cle 0.0 pen pen : f an : clos : clos pen /clos /fan star t /f an topp ple
ion ing ing ing ar ing ar 8m ing ing on ed ed ing ing on ing off ed ted

1200 2192
TC BR7GW 2.1m
TC Corr7GW 2.1m
TC Corr7KW 2.1m
1000 TC CorrStair7W 2.1m
1832
TC Stair7W 2.1m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.1-17. 7G Temperature vs. time at 2.1 m above the floor

49
 Ex
S7 BH pe
BR S1 B BR B : :o S S5 r im
:d : R R o p 5 :
L s : : s
BR oub BY t air dou ing sin ni : p e B H S7 S B S
e
S7 nin op : MV c los MVU ent c
: I g le : op op ble le gle ng 0 : op : op 1: f H: c 7: c : op g/c en/f U st ed/f sto omp
nit fail en en f e e a l e l a a a
ion ing ing ing clea ailing clea .08m ning ning n on losed osed ning osing n on r ting n of pped leted
r r f
1200 2192
TC BR7GW 1.2m
TC Corr7GW 1.2m
TC Corr7KW 1.2m
1000 TC CorrStair7W 1.2m
1832
TC Stair7W 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.1-18. 7G Temperature vs. time at 1.2 m above the floor

Ex
S BH pe
BR S1 BR B R B R 7: op :o S5 S5 r im
:d L : : : p e : : c MV ent
BR oub BY st air dou sing : sin eni BH S7 S S
BH S7 7: ing pe n o MV U
l o s U c
: I g le : op op ble le n :
gle g 0 op op : 1 e
: cl ope /cl n/fa sta d/f a sto omp
nit fail en en f e e : fa : c
ion ing ing ing clea ailing clea .08m ning ning n on losed osed ning osing n on r ting n of pped leted
r r f
120 248
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
100 TC Stair 2W
212
TC Stair1 MidW

Temperature (F)
Temperature (C)

80 176

60 140

40 104

20 68

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.1-19. 7G stairwell temperature vs. time

3.1.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.1-20. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an

50
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
5 Pa and only fluctuated slightly. Once the doors to the base of the stair were opened the pressures
began to fluctuate but remained below 15 Pa. Opening the door at the top of the stairwell decreased
pressures back to ambient conditions around 5 Pa. With the portable PPV fan operating at the base of
the stair and the door to the 7th floor open the pressures averaged 15 Pa to 22 Pa. When the door to the
7th floor was closed the pressures increased to an average of 23 Pa to 30 Pa. After a second portable fan
was added on the 5th floor and the 7th floor door was open the pressures averaged 20 Pa to 38 Pa. After
the MVU was operating with the 7th floor door open the average pressures ranged from 55 Pa to 130 Pa.
Ex
BH pe
BR S1 B R B R B S 7: o :o S S5 r im
:d L : s : : R : p
d o si n si e n B H S 7 p e 5 : M : c los VU ent c
M
BR oub BY t air ub gle ng ing : o : o S B S S 7 n in o p V U s ed
: I g le : op op 1: H: 7: : o g/ en o
nit fail en en le l
cle f aili e cle 0.08 peni peni f an clos clos peni closi /f an ta /fa stop mpl
ion ing ing ing ar n g ar m ng ng on ed ed ng ng on r ting n off ped eted

200
DP Stair 7W
DP Stair 6W
DP Stair 5W
150 DP Stair 4W
DP Stair 3W
DP Stair 2W
DP Stair 1W
Pressure (Pa)

100

50

-50
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.1-20. 7G stairwell pressures vs. time

3.1.4 Velocities 

Average doorway velocities from 4 doors; the fire apartment door, the door from the fire floor to the
stairwell, the door at the base of the stair and the door at the bulkhead, were recorded versus time and
shown in Figure 3.1-21. The three bi-directional probes in each door were averaged to provide the bulk
flow through the doors. A time history for each probe at each measurement location is located in
Appendix D – Detailed Graphs.

The bulk flow out of the fire apartment steadily increased with the size of the fire. Prior to providing a
vent, by opening the stair door, the bulk velocity was out of the apartment. Once the stair was opened
and the bulkhead door was opened the bulk flow increased to a range of 1 m/s to 3 m/s (2 mph to 7 mph)
out of the fire floor and a range of 2 m/s to 3 m/s (4 mph to 7 mph) out of the top of the stairwell. Once
the PPV fan was activated these velocities converged at 0 m/s as the flow from the fan was equalizing
the bulk flow from the fire. Once the MVU was activated there were large velocities through the
structure and out through the fire apartment open windows. Flow in through the fire apartment door
peaked at approximately 8 m/s (18 mph)

51
 Ex
S7 BH pe
BR S1 B BR B : :o S S5 r im
:d : R R o p 5 :
L s : : s
BR oub BY t air dou ing sin ni : p e B H S7 S B S
e
S7 nin op : MV c los MVU ent c
: I g le : op op ble le gle ng 0 : op : op 1: f H: c 7: c : op g/c en/f U st ed/f sto omp
nit fail en en f e e a l e l a a a
ion ing ing ing clea ailing clea .08m ning ning n on losed osed ning osing n on r ting n of pped leted
r r f
9 20

6 13

3 7

Velocity (mph)
Velocity (m/s)

0 0

-3 -7

-6 -13
Exp 7G Bulkhead
-9 Exp 7G Stair1 -20
Exp 7G Stair7
Exp 7G Apt7G
-12 -27
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.1-21. 7G Velocities vs. time

3.1.5 Discussion 

This experiment served as a good baseline as to how the bedroom fires will behave without the influence
of wind. Smoke quickly filled the fire apartment and corridor before becoming ventilation limited.
Temperatures reached a threshold when the windows failed providing conditions for the fire to grow.
Although high temperatures reached the corridor, the second bedroom and living room did not ignite and
burn. The fire was contained to the room of origin.

The fire was deliberately ignited near the window to best allow for window failure. It took 345 s
(5.75 min) to fail the double pane in the bottom left of the window closest to the flames. The internal
pane of the double pane window failed much earlier but acted as an insulator protecting the outer pane
from failure. The flame actually had to come into contact with the pane to fail it. It took 848 s
(14.1 min) until all four panes of window failed. The air that was introduced through the first failed
pane cooled the room causing the window to take longer to fail. The single window remote from the
area of ignition took 972 s (16.2 min) to fail the top panes and 1196 s (19.9 min) to fail the bottom
panes. There were flames coming out of the top portion of the window for more than 180 s (3 min)
before the bottom panes fell out of the frame.

The opening of the stairwell door and the bulkhead door, although remote from the bedroom fire, caused
the fire in the bedroom to intensify when open. As these vents were open the temperatures above 0.9 m
(3 ft) rose by as much as 100 ºC (180 ºF) while the temperatures below 0.9 m (3 ft) decreased by as
much as 75 ºC (135 ºF). Even with 3 open windows in the bedroom hot gases flowing into the stairwell
intensified as the stairwell acted like a chimney.

The use of a single 0.7 m (27 in) portable fan was able to raise the pressures in the stairwell sufficiently
to prevent smoke flow into the stairwell. Stairwell pressures achieved ranged from 15 Pa to 30 Pa with

52
no doors open and with the door on the 7th floor open. The fan was implemented shortly after the fire
reached its peak and maintained a smoke free stair for the duration of the experiment. Temperatures
were cooled up to the fire apartment, and into the fire apartment once the fire burned down to a pile of
rubble. The use of the MVU to ventilate was extremely effective at removing smoke and hot gases but
was only used for overhaul type conditions.

53
3.2 Experiment 7G2 

The second experiment in the structure was located in the same apartment as the first experiment, 7G.
Experiment 7G2 was ignited in the middle bedroom in a trashcan at the base of the bed, and in the living
room on the top of sofa that backed to the kitchen (Figure 3.2-1). The fire in the bedroom from the first
experiment, Experiment 7G, did not extend beyond the room of origin so the apartment was cleared of
smoke and experiment 7G2 was ignited. Both windows in the back bedroom failed in the previous
experiment so they remained open during the duration of this experiment. This experiment also utilized
natural ventilation conditions with no simulated wind and very little natural wind. The events that took
place during the experiment to accomplish the experimental objectives below are shown in Table 3.2-1.

Experimental Objectives
• Create baseline data for fire conditions with no/little wind
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.2-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

54
Figure 3.2-1. 7th floor apartment G floor plan with locations pertinent to Experiment 7G2

55
Table 3.2-1: Experiment 7G2 Timeline
Start (s) End (s) Event Location Event Description
0 0 Middle bedroom and living room Ignition
351 356 Front lobby Doors opened
365 367 1st floor stair A Door opened
642 648 7th floor stair A Door opened
702 703 Bedroom double Top left pane failed
755 762 1st floor stair A Door closed
766 767 Bedroom double Top right pane failed
828 830 1st floor stair A Door opened
887 891 Bulkhead door Opened
959 960 Bedroom double Bottom left pane failed
982 999 1st floor stair A 27 inch fan activated
1075 1076 Bulkhead door Closed
1353 1358 1st floor stair A 27 inch fan deactivated
1500 1502 Bulkhead door Opened
1635 1643 1st floor stair A 27 inch fan activated
1766 1767 Bulkhead door Closed
1779 Living room window Bottom left pane failed
1794 Living room window Center pane failed
1805 Living room window Top left pane failed
1843 Living room window Top right pane failed
1870 1886 5th floor stair A Door opened
1870 1886 5th floor stair A 27 inch fan activated
1874 1875 Living room window Completely failed
1947 1966 7th floor stair A Door closed
2050 2054 7th floor stair A Door opened
2133 2136 7th floor stair A Door opened 0.08 m (3 in)
2229 2232 7th floor stair A Door opened
2351 2355 5th floor stair A 27 inch fan deactivated
2351 2355 5th floor stair A Door closed
2376 2376 1st floor stair A 27 inch fan deactivated
2554 2563 1st floor stair A 27 inch fan activated
2696 2699 Bulkhead door Opened
2771 2774 Bulkhead door Closed
3159 3159 Experiment completed

56
Figure 3.2-2. Instrumentation and camera locations with labels.

57
3.2.1 Observations 

The observations are presented as a series of images captured from eight camera locations, six were
video cameras and two were thermal imaging cameras. The camera positions inside the building are
shown in Figure 3.2-2.

Figure 3.2-3 through Figure 3.2-21 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 48 min (2880 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom
which was burned in the 7G experiment, the bedroom furthest from the apartment entrance; the failed
single window is straight across the view, and the failed double window is to the right of the view. The
top right view (LIVINGRM) was of the living room looking across to the windows. The corridor and
corridor thermal imaging views (CORRIDOR and CORR IR) show the corridor leading up to the open
apartment doorway. The next two views (STAIR and STAIR IR) show the stairwell with a visual and a
thermal imaging camera from the floor below the fire looking up to the fire floor. The bottom left view
(ROOF) shows the bulkhead door at the top of the stairwell and the bottom right view (OUTSIDE)
shows an external view of the fire apartment.

Figure 3.2-3 shows the conditions at the time of ignition. At this point, the six video views are clear and
unobstructed. The thermal images show that there is some residual heat in the corridor and the stairway
at the start of the experiment.

The images in Figure 3.2-4 were captured 240 s (4 min) after ignition. Fire involved half of the living
room sofa where one of the two ignitions had occurred. A smoke layer had formed in the living room
and was approximately 1.8m (6 ft) above the floor. The fire in the living room was directly visible from
the corridor and corridor IR views.

Figure 3.2-5 shows the images at 480 s (8 min) after ignition. Visibility was low in the corridor and
living room as the smoke layer had dropped to the floor in those areas. Smoke was visibly escaping the
structure around the closed bulkhead door and through the failed windows in the bedroom furthest from
the apartment entrance. Despite the smoke obscuration in the living room and corridor, visibility is still
high in this bedroom due to the vented windows.

The images in Figure 3.2-6 were captured 720 s (12 min) after ignition. The 7th floor stairwell door had
been opened and smoke began to descend the stairwell. A significant volume of smoke was leaking past
the bulkhead door. The double window of the bedroom closest to the apartment entrance partially
failed; flames and smoke were venting from the opening. Smoke density in the bedroom furthest from
the apartment entrance was also significantly higher than four minutes prior.

Figure 3.2-7 shows the images at 960 s (16 min) after ignition. At this point the bulkhead door had been
opened and visibility was reduced to zero in the outdoor roof view. Two additional panes in the window
of the bedroom closest to the apartment entrance had failed and smoke as well as flames continued to
vent from it. Limited visibility had returned to the floor level of the corridor. A clear heat layer is
visible in the corridor IR and stair IR views as hot gasses pass through the ceiling of those areas.

58
Opening the bulkhead door lifted the smoke level in the stairwell but also increased the flow of hot gases
through the corridor and into the stairwell.

Figure 3.2-8 shows the images at 1200 s (20 min) after ignition. At this point the window in the
bedroom closest to the apartment entrance had completely failed and flames continued to vent from it.
A portable PPV fan was activated and was ventilating the stairwell and then the bulkhead door was
closed to pressurize the stairwell. The fan was operating for 200 s (3.4 min) and had improved
conditions in the stairwell and corridor. The bulkhead door had been closed and little or no smoke is
visibly leaking around it. Hot gas flow is not visible in either the corridor IR or the stair IR view.
Visibility was high in all views except the living room, where visibility was limited.

The images in Figure 3.2-9 were captured 1440 s (24 min) after ignition. The condition of the window
in the bedroom closest to the apartment entrance remained the same. The PPV fan was deactivated and
the conditions in the stairwell worsened. Smoke began to fill the corridor again and hot gas flow was
visible in the corridor IR at the corridor ceiling.

Figure 3.2-10 shows the images at 1680 s (28 min) after ignition. The bulkhead door had been opened
and smoke flow through it was visible. The PPV fan was reactivated and conditions were improving in
the stairwell and corridor. At this point the ignition sofa was fully involved and there was sufficient
visibility in the living room to see it from the living room floor. Limited hot gas flow was visible in the
corridor IR view because the camera switched to the extended mode (low sensitivity), and visibility was
good in the corridor view.

Figure 3.2-11 shows the images at 1760 s (29 min and 20 s) after ignition. The fire in the living room
was developing and increased heat was flowing into the corridor. Visibility in the lower layer of the
corridor was still good as the fire was pulling fresh air from the stairwell.

The images in Figure 3.2-12 show the conditions at 1800 s (30 min) after ignition. The fire in the living
room was becoming ventilation limited and the smoke layer had descended to the floor. One of the
living room windows had failed and smoke and flames were flowing from the window. The bulkhead
door had been shut for 36 s and the fan was establishing a pressure equilibrium to limit smoke and heat
flow into the stairwell

Figure 3.2-13 shows the images at 1920 s (32 min) after ignition. The living room window had
completely failed and flames were pushing out of it. The flames pulled to the air provided by the
windows but the living room did not transition to flashover yet. The corridor IR view was saturated,
which limits the contrast in the view, but heat was clearly visible in the stair IR view. Smoke was
leaking around the closed bulkhead door.

Figure 3.2-14 shows the images at 1980 s (33 min) after ignition. The stair door to the 7th floor was
closed to examine the impact of the PPV fan on smoke leakage around the door. The living room had
transitioned to flashover and the fire was at its peak. The stairwell door was easily visible and there was
no obvious heat flow around the door. The fan was forcing the residual smoke in the stairwell out
around the cracks of the bulkhead door at the top of the stairwell.

59
Figure 3.2-15 shows the images at 2070 s (34 min and 30 s) after ignition. This was captured 15 s after
the 7th floor stair door was reopened. The living room was still well involved in flames and the heat has
stopped at the stairwell door by the fans pressure barrier. Smoke flow into the stairwell was greatly
reduced and the visibility in the corridor was improved at the lower level.

Figure 3.2-16 shows the images at 2160 s (36 min) after ignition. Flames and smoke were continuing to
vent from both the living room window and the window in the bedroom closest to the apartment
entrance. A significant volume of smoke continued to leak around the closed bulkhead door. Visibility
at the floor level of the corridor was significantly lowered as a result of smoke flow being stopped at the
7th floor stair A door, which was opened .08m (3 in).

Figure 3.2-17 shows the images at 2400 s (40 min) after ignition. At this point the 27 inch fans on the
5th and 1st floor stairwells were turned off, 43 s and 22 s prior, respectively. Significant hot gas flow is
visible in the corridor IR and stair IR; smoke flow is visible in the cameras of those views. The
increased heat flow into the corridor and stairwell caused both thermal imaging cameras to transition to
extended mode. Smoke was descending down the stairwell making the stairwell door no longer visible.

Figure 3.2-18 shows the images at 2550 s (42 min and 30 s) after ignition. These were the conditions
just prior to the fan at the first floor being started. The fire was burning down but the conditions in the
stairwell were still poor. Figure 3.2-19 shows the images at 2580 s (43 min), 20 s after the fan at the
first floor was reactivated. Smoke and heat flow into the stairwell were reduced.

Figure 3.2-20 shows the images at 2640 s (44 min) after ignition. Conditions were continuing to
improve in the stairwell with the fan operating. The fire was burning down so the fan flow was clearing
the corridor in addition to the stairwell. The smoke that was in the stair was being forced out around the
closed bulkhead door.

Figure 3.2-21 shows the images at 2880 s (48 min) after ignition. At this point the fire had largely
burned down and smoke had been cleared.

60
Figure 3.2-3. Experiment 7G2, ignition.

61
Figure 3.2-4. Experiment 7G2, 240s after ignition.

62
Figure 3.2-5. Experiment 7G2, 480s after ignition.

63
Figure 3.2-6. Experiment 7G2, 720s after ignition.

64
Figure 3.2-7. Experiment 7G2, 960s after ignition.

65
Figure 3.2-8. Experiment 7G2, 1200s after ignition.

66
Figure 3.2-9. Experiment 7G2, 1440s after ignition.

67
Figure 3.2-10. Experiment 7G2, 1680s after ignition.

68
Figure 3.2-11. Experiment 7G2, 1760s after ignition.

69
Figure 3.2-12. Experiment 7G2, 1800s after ignition.

70
Figure 3.2-13. Experiment 7G2, 1920s after ignition.

71
Figure 3.2-14. Experiment 7G2, 1980s after ignition.

72
Figure 3.2-15. Experiment 7G2, 2070s after ignition.

73
Figure 3.2-16. Experiment 7G2, 2160s after ignition.

74
Figure 3.2-17. Experiment 7G2, 2400s after ignition.

75
Figure 3.2-18. Experiment 7G2, 2550s after ignition.

76
Figure 3.2-19. Experiment 7G2, 2580s after ignition.

77
Figure 3.2-20. Experiment 7G2, 2640s after ignition.

78
Figure 3.2-21. Experiment 7G2, 2880s after ignition.

79
3.2.2 Temperatures 

Figure 3.2-22 through Figure 3.2-26 provide temperature data for four measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.2-2. There was no temperature measurement location in the fire
apartment for this experiment as it was damaged in the previous experiment in the same apartment. The
events that occurred during the experiment are listed on the top of the figures with lines at the time of
the event. A time history for each thermocouple at each measurement location is located in Appendix D
– Detailed Graphs.

Figure 3.2-22 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 4 measurement
locations, just outside the fire apartment in the corridor, just outside Apartment 7K in the corridor, just
outside the A stairwell doorway, and just inside the A stairwell on the landing. The three temperature
measurement locations in the corridor behaved similarly which is consistent with the steady two layer
environment that existed remote from the fire. Temperatures steadily increased as the fire in the
bedroom and living room increased. Temperatures increased to approximately 250 ºC (482 ºF) up until
the fan was activated. The fan flow decreased the corridor temperatures to below 100 ºC (212 ºF). After
the fan was tuned off, and the fire in the living room began to escalate, the corridor temperatures
increased to 200 ºC (392 ºF). When the bulkhead door was opened the corridor temperature peaked at
300 ºC (572 ºF). As the fire in the living room transitioned to flashover the temperature just outside the
fire apartment reached 500 ºC (932 ºF). Closer to the stairwell when the fan was operating the corridor
temperature did not exceed 300 ºC (572 ºF)

The temperature in the stairwell remained near ambient until the stairwell door was opened. Once open
the stairwell temperature at 2.1 m (7 ft) above the landing increased to 150 ºC (302 ºF). The fan
activation decreased this temperature back down to ambient even with the door completely open on the
fire floor. The fan was turned off at 1359 s and the temperature quickly increased back to 150 ºC
(302 ºF). There were a number of door opening and closings during the experiment and the stairwell
temperature fluctuated between ambient and 150 ºC (302 ºF). Once the fire floor door was opened it
took approximately 60 s to 90 s to establish the pressure barrier and return the temperature back to
ambient. When the living room fire was near its peak the fans were turned off and the stairwell
temperatures increased above 250 ºC (482 ºF). Activating the fan again at approximately 2560 s reduces
the stairwell temperature from 250 ºC (482 ºF) to 50 ºC (122 ºF) in approximately 90 s.

Figure 3.2-23 shows the temperatures at 1.2 m (4 ft) which would be more representative of where
firefighters would be operating or occupants would be evacuating. The temperatures in the corridor
increased to approximately 120 ºC (248 ºF), and leveled off until the bulkhead door was opened. This
event caused the temperature just outside the fire apartment to increase to 250 ºC (482 ºF). Turning the
PPV fan on reduced this temperature quickly to below 100 ºC (212 ºF). Similar increases occurred again
as the fan was tuned off and the bulkhead was opened. At approximately 1965 s the 7th floor stair door
was closed and the temperatures rapidly increased as the hot gas descended. Temperatures increased
from 100 ºC (212 ºF) to above 400 ºC (752 ºF) in seconds. Reopening the door reduced temperatures
just as quickly without raising the temperatures that much in the stairwell because of the flow of the fan.
When the fans were turned off at 2376 s the temperatures throughout the fire floor increased from
200 ºC (360 ºF) to 400 ºC (720 ºF) in less than 10 s. Opening the bulkhead door while the fan was

80
flowing, reducing pressurization and increasing ventilation, also caused temperatures to escalate quickly
throughout the fire floor.

The stairwell temperatures at 1.2 m on the landing of the fire floor reached as high as 200 ºC (392 ºF)
with the door open and no mechanical ventilation but decreased to near ambient with the fan operating.
The addition of another fan at the 5th floor while the fire in the living room was at its peak decreased
stairwell temperatures from 60 ºC (140 ºF) to 20 ºC (68 ºF).

Figure 3.2-24 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, 6 and at the
bulkhead door. Below the fire floor there was no significant change in temperature. Above the fire
floor, at the bulkhead door, temperatures reached as high as 160 ºC (320 ºF) under normal ventilation
conditions. Temperatures remained below 90 ºC (194 ºF) while fan were operating.

S
M BR LR L 5: s S5
R S7
ul S
tip LB S S : do 1: s S B
: wi : w t air : o S : clo
B n in o S
le Y: 1: 7: ub ta 1: H S1 BH S1 H S1 BH do d pe S7 7: pe 7: se S1 S1 BH BH
ign op op op le ir c op op : f c : f op : f w ow n : o n o d : : o
it io en en en fail los en en an los an en an clos f ail cl /fan clos pen 0. 0 pen /f an fan fan pen clos
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing off off on ing ed

600 1112
TC Corr7GW 2.1m
TC Corr7KW 2.1m
TC CorrStair7W 2.1m
500 TC Stair7W 2.1m 932

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.2-22. 7G2 Temperature vs. time at 2.1 m above the floor

81
 S
M BR LR L 5: s S5
ul : S : R: t a S7 :c
tip LB S S do 1: s S B wi w ir :
le Y: 1: 7: ub ta 1: H S1 BH S1 BH S1 BH ndo ind ope S7 S7: ope S7: lose S1 S1 BH BH
i
ign op op op le r c op op : f c : f op : f w o w n : o n o d : : o
it io en en en fail los en en an los an en an clos f ail cl /fan clos pen 0. 0 pen /f an fan fan pen clos
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing off off on ing ed

600 1112
TC Corr7GW 1.2m
TC Corr7KW 1.2m
TC CorrStair7W 1.2m
500 TC Stair7W 1.2m 932

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.2-23. 7G2 Temperature vs. time at 1.2 m above the floor

S
M BR LR L 5: s S5
ul S1 : w R: t ai S7
L
tip B S S do : s S B:
in wi r o S : o S : clo
le Y: 1: 7: ub ta 1: H S1 BH S1 H S1 BH do nd pe S7 7: pe 7: se S1 S1 BH BH
B
ign op op op le ir c op op : f c : f op : f w ow n : o n o d : : o
it io en en en fail los en en an los an en an clos f ail cl /fan clos pen 0. 0 pen /f an fan fan pen clos
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing off off on ing ed

200 392
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
150 302

Temperature (F)
Temperature (C)

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.2-24. 7G2 stairwell temperature vs. time

3.2.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.2-25. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
adjacent apartment on the same floor. With all of the doors to the stairwell closed the pressures
remained below 5 Pa and only fluctuated slightly. Once the doors to the base of the stair were opened

82
the pressures began to fluctuate but remained below 9 Pa. Opening the door at the top of the stairwell
decreased pressures back to ambient conditions around 5 Pa. With the portable PPV fan operating at the
base of the stair and the door to the 7th floor open the pressures averaged 13 Pa to 26 Pa. After a second
portable fan was added on the 5th floor and the 7th floor door was open the pressures ranged from
approximately 15 Pa to 30 Pa.
S
M BR LR L 5: s S5
ul : S : R: ta S7 :c
tip LB S d 1 : w i w ir :
le Y: 1: S7: oub st a S1: BH S1 BH S1 BH S1 BH ndo ind ope S7 S7: ope S7: lose S1 S1 BH BH
ign op op op le ir c op op : f : o : w o w n : o n o d : : o
it io en en en fail los en en an clos f an pen f an clos f ail cl /fan clos pen 0.0 pen /f an f an fan pen clos
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
40
DP Stair 7W
DP Stair 6W
DP Stair 5W
30 DP Stair 4W
DP Stair 3W
DP Stair 2W
DP Stair 1W
Pressure (Pa)

20

10

-10
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.2-25. 7G2 stairwell pressures vs. time

3.2.4 Velocities 

Average doorway velocities from 4 doors; the fire apartment door, the door from the fire floor to the
stairwell, the door at the base of the stair and the door at the bulkhead, were recorded versus time and
shown in Figure 3.2-26. The three bi-direction probes in each door were averaged to provide the bulk
flow through the doors. A time history for each probe at each measurement location is located in
Appendix D – Detailed Graphs.

The bulk flow out of the fire apartment steadily increased with the size of the fire. Prior to providing a
vent, by opening the stair door, the bulk velocity was out of the apartment. Once the stair was opened
and the bulkhead door was opened the bulk flow increased to approximately 2 m/s (4 mph) out of the
fire floor and 4 m/s (9 mph) out of the top of the stairwell. Once the PPV fan was activated these
velocities converged at 0 m/s as the flow from the fan was equalizing the bulk flow from the fire.

83
 S
M BR LR L 5: s S5
ul : S : R: t a S7 :c
tip LB S S do 1: s S B wi w ir :
le Y: 1: 7: ub ta 1: H S1 BH S1 BH S1 BH ndo ind ope S7 S7: ope S7: lose S1 S1 BH BH
i
ign op op op le r c op op : f c : f op : f w o w n : o n o d : : o
it io en en en fail los en en an los an en an clos f ail cl /fan clos pen 0. 0 pen /f an fan fan pen clos
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing off off on ing ed

4 9

2 4

0 0

Velocity (mph)
Velocity (m/s)

-2 -4

-4 -9

Exp 7G2 Bulkhead

-6 Exp 7G2 Stair1
-13
Exp 7G2 Stair7
Exp 7G2 Apt7G
-8 -18
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.2-26. 7G2 Velocities vs. time

3.2.5 Discussion 

This experiment served as a good baseline as to how a fire involving two rooms, a bedroom and living
room, behaved without the influence of wind. The fire in the bedroom developed much faster than the
living room, most likely because of the available oxygen from the open window in the bedroom from the
previous experiment and the relatively small size. Smoke quickly filled the fire apartment and corridor
before becoming ventilation limited. The fire in the living room reacted quicker to the changes in
ventilation by opening the stairwell and bulkhead doors. The oxygen from the stairwell allowed the fire
to grow and fail the windows, which drove the room to flashover.

It took 703 s (11.8 min) for the first dual pane window in the bedroom to fail, and 1053 s (17.6 min) for
the entire window to fail. This time would have been longer or the window may not have failed if it was
not for the open windows in the adjacent bedroom because the fire in the living room would have used
up all of the oxygen coming from the stairwell. The living room window began to fail 1779 s (29.7 min)
and completely failed at 1875 s (31.3 min). It appeared from the video that the flames had to touch the
windows in order to fail them.

The opening of the stairwell door and the bulkhead door caused the fire in the bedroom and living room
to intensify. The fire in this experiment was closer to the corridor and therefore reacted faster to the
additional oxygen from the open door. As these vents were open the temperatures above 1.2 m (4 ft)
rose by as much as 200 ºC (360 ºF). Even with 3 open windows in the rear bedroom, 2 open bedrooms
in the other furnished bedroom and all of the living room windows open, hot gases flowing into the
stairwell intensified as the stairwell acted like a chimney.

The single 0.7 m (27 in) portable fan was able to raise the pressures in the stairwell sufficiently to
prevent smoke flow into the stairwell. When the stairwell door was opened there was typically a 60 s to
90 s period of turbulence until the pressures and flows reached steady state to stabilize the temperatures.

84
Stairwell pressures achieved ranged from 15 Pa to 30 Pa with no doors open and with the door on the
7th floor open. This was consistent with the first experiment. A fan was implemented multiple times
and was able to create a smoke free stair or greatly reduce the flow of hot gases. The temperatures were
greatly cooled up to the fire apartment on multiple occasions, and into the fire apartment once the fire
burned down to a pile of rubble.

85
3.3 Experiment 7E 

The third experiment in the structure was located on the 7th floor in apartment 7E, on the other half of
the building from the first two experiments. Experiment 7E was ignited in the bedroom furthest from
the open apartment door (Figure 3.3-1). This experiment was the first experiment with a simulated wind
provided by the MVU. The simulated wind was imposed on the bedroom double window at a
magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). Prior to the experiment the inner pane
of the bedroom window was removed, leaving a single pane window. The events that took place during
the experiment to accomplish the experimental objectives below are shown in Table 3.3-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.3-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

86
Figure 3.3-1. 7th floor apartment E floor plan with locations pertinent to Experiment 7E

87
Table 3.3-1: Experiment 7E Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
163 163 Bedroom double Bottom right pane failed
163 165 7th floor stair A Door opened 0.08 m (3 in)
470 471 Bedroom double Top right pane failed
475 484 Front lobby Doors opened
492 497 1st floor stair A Door opened
534 535 Bedroom double Bottom left pane failed
640 645 Front lobby 27 inch fan activated
663 667 7th floor stair A Door closed
683 727 Bedroom double Top left pane failed
705 710 Bulkhead door Opened
782 807 1st floor stair A 27 inch fan moved
855 858 5th floor stair A Door opened
855 858 5th floor stair A 27 inch fan activated
1021 1033 Bedroom double Air flow adjusted
1029 1032 Bulkhead door Closed
1094 1107 7th floor stair A Door opened 0.08 m (3 in)
1222 1232 5th floor stair A 27 inch fan deactivated
1222 1232 5th floor stair A Door closed
1262 1266 7th floor stair A Door opened 1/2
1310 1314 Bulkhead door Opened
1379 1450 Living room window Failure
1469 1471 Bulkhead door Closed
1800 Experiment completed

88
Figure 3.3-2. Instrumentation and camera locations with labels.

89
3.3.1 Observations 

The observations are presented as a series of images captured from eight camera locations, six were
video cameras and two were thermal imaging cameras. The camera positions inside the building are
shown in Figure 3.3-2.

Figure 3.3-3 through Figure 3.3-17 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 28 min (1680 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 7E, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor and corridor
thermal imaging views (CORRIDOR and CORR IR) show the corridor leading up to the open apartment
doorway. The next two views (STAIR and STAIR IR) show the stairwell with a visual and a thermal
imaging camera from the floor below the fire looking up to the fire floor. The bottom left view (ROOF)
shows the bulkhead door at the top of the stairwell and the bottom right view (OUTSIDE) shows an
external view of the fire apartment.

Figure 3.3-3 shows the images at the time of ignition. At this point, the six video views were clear and
unobstructed. However, the thermal images provided limited thermal contrast because the surfaces in
the view are at nearly equal temperature.

Figure 3.3-4 shows the images at 240 s (4 min) after ignition. The fire from the trash container had
spread onto the bed, and the smoke layer had descended to the floor of the ignition bedroom. The
bedroom window had begun to fail approximately 75 s prior and the 7th floor stairwell A door was
opened 0.08 m (3 in). The simulated wind forced the smoke throughout the public corridor and the
stairwell, reducing visibility. Despite low visibility in the camera views for the corridor and stair, the
temperature of the gas flow does not exceed 200ºC (392ºF) at the ceiling level. At this point, a small
amount of smoke is visibly leaking around the bulkhead door.

Figure 3.3-5 shows the images at 480 s (8 min) after ignition. Visibility had not improved significantly
from 240 s in the bedroom, living room, and corridor views. A second pane of the bedroom double
failed 9 s prior and visibility in the stairwell view was slightly improved. Temperatures in the corridor
visibly increased in the corridor IR.

Figure 3.3-6 shows the images at 720 s (12 min) after ignition. Flames involved the bedroom, reached
the camera in the bedroom corner, and were visibly passing through the living room. The bulkhead door
had been opened and the 7th floor stair A door had been closed, but visibility had not yet returned to the
stairwell view. A portable PPV fan was activated but the impact on stairwell visibility can’t be
determined because the smoke that reached the camera deposited on the lens of the camera. Smoke was
however exiting the open bulkhead door.

90
Figure 3.3-7 shows the images at 960 s (16 min) after ignition. At this point the bedroom double
window had completely failed. Some visibility returned to the stairwell, and the stair IR view showed
extreme contrast in temperature of the closed 7th floor stair A door and the rest of the stairwell.

Figure 3.3-8 shows the images at 1090 s (18 min and 10 s) after ignition. These were the conditions
prior to opening the door from the 7th floor to the stairwell. The wind driven conditions are being held
up at the door with no vent, hence no flow path. The door has heated up substantially as shown by the
contrast in the thermal imaging view.

Figure 3.3-9 shows the images at 1170 s (19 min and 30 s) after ignition. Flames still involved most of
the bedroom but none were visible elsewhere in the apartment. The door to the stairwell was opened
0.08 m (3 in) 70 s prior, and the two portable PPV fans were trying to keep the hot gases out of the
stairwell and were not completely successful against the wind driven conditions.

Figure 3.3-10 shows the images at 1200 s (20 min) after ignition. The bulkhead door had been closed
170 s prior and the 7th floor stair A door had been reopened 0.08 m (3 in) 100 s prior. Smoke had not
yet cleared the roof view.

Figure 3.3-11 shows the images at 1320 s (22 min) after ignition. The bulkhead door was opened
approximately 10 s prior to this figure being captured. The wind forced the hot gases from the bedroom
past the living room and ignited the sofa in the right side of the image. The wind driven fire gases were
moving up the stairwell and through the bulkhead doorway.

Figure 3.3-12 shows the images at 1350 s (22 min and 30 s) after ignition, and 30 s after the sofa caught
fire. The fire on the sofa had intensified and flames were beginning to be forced into the corridor. The
fan was still operating on the first floor and the hot gases were traveling up the stairwell and not down to
the floor below.

Figure 3.3-13 shows the images at 1380 s (23 min) after ignition. The fire in the bedroom was burning
down but the fire in the living room was intensifying. The flames in the corridor reached the stairwell
doorway and the view in the stairwell was beginning to clear up. The stair IR view shows high levels of
heat coming out of the fire floor. The corridor IR camera view was saturated with heat and was
unusable.

Figure 3.3-14 shows the images at 1440 s (24 min) after ignition. Flames were pushing out of the living
room window which had failed. The bulkhead and 7th floor stair A doors had been opened prior to the
window failing. The flames now involved all of the living room and corridor. Even though the living
room window vented, flames were still traveling to the stairwell. Visibility was high in the stairwell
view, as most of the heat and smoke traveled up the stairs.

Figure 3.3-15 shows the images at 1480 s (24 min and 40 s) after ignition. The living room was still
fully involved, with flames being forced out of the window and extending above the roof line. The
corridor camera still showed a view although it was on the brink of failure. Flames were no longer
visible in the stairwell view but the thermal imaging view showed heat continuing to enter the stairwell.

91
Figure 3.3-16 shows the images at 1560 s (26 min) after ignition. The fire in the living room was still
venting out of the windows from top to bottom and the internal camera became covered with soot,
eliminating that view. The camera in the corridor finally failed and that thermal view was still not
usable. The bulkhead door had been closed for 90 s and the pressure from the fan was maintaining
visibility in the stair and limiting the heat flow.

Figure 3.3-17 shows the images at 1680 s (28 min) after ignition. Limited combustion was occurring in
the bedroom where ignition occurred and there was zero visibility in the living room view, however
flames were still pushing from the living room window. The fan was continuing to improve conditions
in the stairwell.

92
Figure 3.3-3. Experiment 7E, ignition.

93
Figure 3.3-4. Experiment 7E, 240s after ignition.

94
Figure 3.3-5. Experiment 7E, 480s after ignition.

95
Figure 3.3-6. Experiment 7E, 720s after ignition.

96
Figure 3.3-7. Experiment 7E, 960s after ignition.

97
Figure 3.3-8. Experiment 7E, 1090s after ignition.

98
Figure 3.3-9. Experiment 7E, 1170s after ignition.

99
Figure 3.3-10. Experiment 7E, 1200s after ignition.

100
Figure 3.3-11. Experiment 7E, 1320s after ignition.

101
Figure 3.3-12. Experiment 7E, 1350s after ignition.

102
Figure 3.3-13. Experiment 7E, 1380s after ignition.

103
Figure 3.3-14. Experiment 7E, 1440s after ignition.

104
Figure 3.3-15. Experiment 7E, 1480s after ignition.

105
Figure 3.3-16. Experiment 7E, 1560s after ignition.

106
Figure 3.3-17. Experiment 7E, 1680s after ignition.

107
3.3.2 Temperatures 

Figure 3.3-18 through Figure 3.3-20 provide temperature data for five measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.3-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.3-18 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 7A in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
temperature in the bedroom rose to approximately 500 ºC (932 ºF) before the bottom right window pane
failed. This influx of cool air lowered the bedroom temperature to below 200 ºC (392 ºF) and slowed
fire growth. It took until about 550 s until the fire in the bedroom really got developed and approached
flashover. At 600 s the bedroom 2.1 m (7 ft) temperature exceeded 600 ºC (1112 ºF) and it remained
above that temperature until approximately 1400 s when the fuel load in the bedroom was burning
down.

The three temperature measurement locations in the corridor behaved similarly in reaction to the events
with the higher temperatures at positions closer to the fire apartment. Corridor temperatures reached
400 ºC (752 ºF) before they became ventilation limited between 700s and 1100 s. Temperatures in the
corridor remained very steady until a ventilation path was created. The 7th floor stair door was opened
0.08 m (3 in) at 1100 s which caused the corridor temperature outside the fire apartment to increase from
approximately 400 ºC (752 ºF) to 550 ºC (1022 ºF). The temperatures declined again due to the lack of
oxygen until the bulkhead door was opened. Once opened a wind driven condition developed increasing
the corridor temperatures to above 600 ºC (1112 ºF). The temperatures decreased when the bulkhead
door was closed and the PPV fan was flowing.

The temperature in the stairwell remained near ambient until the window in the bedroom failed. The
stairwell temperature at 2.1 m (7 ft) above the landing increased to above 250 ºC (482 ºF) even with the
door closed. The fan activation decreased this temperature back down to ambient. When the stair door
was opened 0.08 m (3 in) at 1100 s the stairwell temperature increased to 300 ºC (572 ºF). After the
bulkhead was opened and the wind driven conditioned developed the stair temperature exceeded 500 ºC
(932 ºF). Closing the bulkhead and allowing the fan to pressurize caused the stair temperature to
decrease.

Figure 3.3-19 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. The bedroom temperature increased
to 500 ºC (932 ºF) before the bedroom window partially failed, cooling the room. The temperatures
cooled due to the failing windows and then the 1.2 m (4 ft) temperature began to track with the 2.1 m
(7 ft) temperature after 600 s which indicated flashover conditions in the bedroom. The 1.2 m bedroom
temperature remained above that temperature until approximately 1400 s when the fire in the bedroom
began to decay.

Corridor temperatures reached approximately 300 ºC (572 ºF) to 600 ºC (1112 ºF) before they became
ventilation limited between 700s and 1100 s. Temperatures in the corridor remained very steady until a

108
ventilation path was created. The 7th floor stair door was opened 0.08 m (3 in) at 1100 s which caused
the corridor temperature outside the fire apartment to increase from 350 ºC (662 ºF) to more than 600 ºC
(1202 ºF). The temperatures decreased again with the lack of oxygen until the bulkhead door was
opened. Once opened a wind driven condition developed increasing the corridor temperatures to above
800 ºC (1472 ºF). The temperatures decreased when the bulkhead door was closed and the PPV fan was
flowing.

The temperature in the stairwell remained near ambient until the window in the bedroom failed. The
stairwell temperature at 2.1 m (7 ft) above the landing increased to above 250 ºC (482 ºF) even with the
door closed. The fan activation decreased this temperature back down to ambient. When the stair door
was opened 0.08 m (3 in) at 1100 s the stairwell temperature increased to 250 ºC (482 ºF). After the
bulkhead was opened and the wind driven conditioned developed the stair temperature exceeded 500 ºC
(932 ºF). Closing the bulkhead and allowing the fan to pressurize caused the stair temperature to
decrease.

Figure 3.3-20 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, and at bulkhead.
The thermocouple on the sixth floor failed to work. After the window failed hot gases were forced
around the cracks in the stairwell door, increasing the temperature at the bulkhead door to 180 ºC
(356 ºF). The fan was activated and the bulkhead was opened lowering the temperature to ambient.
Prior to fan activation temperatures were increasing all the way down to the first floor. After the fan
was activated the temperatures below the 4th floor never elevated above ambient.
ted
l i ng .08m a r g n ed .08m off alf i l ing mpl e
i e
cl ovin f an o t o
fa 0 ing ing on d g us 0 fan h g fa tc
ion le ing en nin ble / ed adj ing ed/ ed nin ow ed en
I gni t d o ub op e n :
e
op open : fan clos ope dou f an mopen clos flow open clos open ope wind clos er im
: : : Y : Y : : : : : : : : : : : : p
BR BR S7 LB S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S 5 S 7 B H LR B H Ex
2000 3632
TC BR7EE 2.1m
TC Corr7EE 2.1m
TC Corr7AE 2.1m
TC CorrStair7E 2.1m
TC Stair7E 2.1m
1500 2732

Temperature (F)
Temperature (C)

1000 1832

500 932

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.3-18. 7E Temperature vs. time at 2.1 m above the floor

109
 ed
m d .08m off alf l et
a il ing 0.08 le ar ng o n t e 0 n ai ling o mp
g i s h f c
n ef g c a
nin ning n on ed ning ble mov n/f an sed adju ning ed/f ned ning dow sed nt
itio ubl nin pe s e u s e me
: I gn : do : ope Y: o : ope Y: fa : clo : op : do : f an : ope : clo r flow : ope : clo : ope : op : win : clo p er i
BR BR S7 LB S 1 LB S 7 B H B R S 1 S 5 BH Ai S7 S 5 S 7 B H LR B H Ex
1200 2192
TC BR7EE 1.2m
TC Corr7EE 1.2m
TC Corr7AE 1.2m
1000 TC CorrStair7E 1.2m
1832
TC Stair7E 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.3-19. 7E Temperature vs. time at 1.2 m above the floor

ted
l i ng .08m a r g n ed .08m off alf i l ing mpl e
i e
cl ovin f an o t o
fa 0 ing ing on d g us 0 fan h g fa tc
ion le ing en nin ble / ed adj ing ed/ ed nin ow ed en
I gni t d o ub op e n :
e
op open : fan clos ope dou f an mopen clos flow open clos open ope wind clos er im
: : : Y : Y : : : : : : : : : : : : p
BR BR S7 LB S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S 5 S 7 B H LR B H Ex
600 1112
TC Bulkhead MidE
TC Stair 6E
TC Stair 4E
500 TC Stair 2E
932
TC Stair1 MidE

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.3-20. 7E stairwell temperature vs. time

3.3.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.3-21. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
5 Pa and only fluctuated slightly. Once the bottom right window pane failed in the bedroom the
pressures in the stairwell increased to approximately 12 Pa and steadily increased to 35 Pa to 45 Pa.
Next, the front lobby doors and first floor stair door were opened. This allowed the pressure to vent and
created a flow out of the first floor lobby. At 600 s the pressures from top to bottom on each floor were

110
40 Pa, 30 Pa, 24 Pa, 16 Pa, 11 Pa, 8 Pa and 5 Pa respectively. A fan was added at the front lobby due to
the smoke coming out of the front doors, which increased the pressure in the stair further. The bulkhead
door was opened and the pressure in the entire stair dropped below 10 Pa as the gases moved to the vent.

After the smoke moved up the stairwell, the fan was moved from the lobby to the base of the stairwell.
This increased the stairwell pressures to a range of 5 Pa to 18 Pa. Adding a second PPV fan at floor
5 increased the stairwell pressure range to 5 Pa to 22 Pa. The bulkhead door was closed and the
stairwell pressure range increased to from 5 Pa to 30 Pa. Opening the fire floor door 0.08m (3 in) and
allowing more wind driven flow into the stair increased the pressure range further to 20 Pa to 44 Pa.
With the fire floor door completely opened the pressure range was 20 Pa to 58 Pa. After the living room
window failed and the flow was split the pressures became steadier at 16 Pa to 24 Pa.

ted
l i ng .08m a r g n ed .08m off alf i l ing mpl e
i 0 e o
c l o v i n f an d t n a o
ion le ing
fa ing ing on d g jus ing 0 d/f a ed h ing ow f ed tc
ni t d o ub op e n en e nin ble / e ad e n en
I g : op open : fan clos ope dou f an m open clos flow open clos open ope wind clos er im
: : : Y : Y : : : : : : : : : : : : p
BR BR S7 LB S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S5 S7 BH LR BH Ex
80
DP Stair 7E
DP Stair 6E
DP Stair 5E
60 DP Stair 4E
DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.3-21. 7E stairwell pressures vs. time

3.3.4 Velocities 

Average doorway velocities from 4 doors; the fire apartment door, the door from the fire floor to the
stairwell, the door at the base of the stair and the door at the bulkhead, were recorded versus time and
shown in Figure 3.3-22. The three bi-direction probes in each door were averaged to provide the bulk
flow through the doors. A time history for each probe at each measurement location is located in
Appendix D – Detailed Graphs.

The flows out of the fire apartment were as high as 5 m/s with the stairwell door closed. When the
bulkhead door was opened with the fan flow the bulk flow velocity out of the top of the stairwell peaked
at 4 m/s. The flow out of the fire apartment with the stairwell opened was approximately 5 m/s and
increased to 20 m/s with the bulkhead door opened. The flow into the stairwell was approximately
9 m/s and the flow out of the bulkhead door was 13 m/s with the wind driven flow combined with the
PPV fan flow.

111
 ed
m d .08m off alf l et
a il ing 0.08 le ar ng o n t e 0 n ai ling o mp
g i s h f c
n ef g c a
nin ning n on ed ning ble mov n/f an sed adju ning ed/f ned ning dow sed nt
itio ubl nin pe s e u s e me
: I gn : do : ope Y: o : ope Y: fa : clo : op : do : f an : ope : clo r flow : ope : clo : ope : op : win : clo p er i
BR BR S7 LB S 1 LB S 7 B H B R S 1 S 5 BH Ai S7 S 5 S 7 B H LR B H Ex
30 67
Exp 7E Bulkhead
Exp 7E Stair1
Exp 7E Stair7
Exp 7E Apt7E
20 45

Velocity (mph)
Velocity (m/s)

10 22

0 0

-10 -22

-20 -45
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.3-22. 7E Velocities vs. time

3.3.5 Discussion 

Experiment 7E was the first experiment to introduce a simulated wind. A wind driven fire was created
but it took some coordinated opening of doors to get the conditions right, but when the wind driven
condition was created it forced flames all the way from the ignition bedroom to the stairwell door in the
corridor. These conditions would not be conducive to an interior direct frontal fire fighter attack. The
missing inner pane of the bedroom window caused much earlier failure as compared to experiment 7G.
One of the bedroom windows failed at 165 s causing cooling which delayed the wind driven condition
and other window failure, but when the conditions developed the area in the vent path became untenable
very quickly. The entire bedroom double window was failed at 729 s.

At approximately 700 s, the bedroom is fully involved in fire (post-flashover). Even though
temperatures in the bedroom were in excess of 600 ºC (1112 ºF), the temperatures in the public hall and
the stair were well below that but increasing. This was mainly due to the fact that there was no flow
path for the fire gases to follow because the door to the stair was closed. Even with the door closed,
there was enough leakage to increase the temperatures significantly in the stairwell and at the top of the
stairs. Opening the bulkhead door, with the fire floor closed improved conditions in the stairwell.
Opening the bulkhead door, with the fire floor door open deteriorated conditions in the entire flow path
from the bedroom to the top of the stairs. When the first floor was open and the fire floor door and
bulkhead door were closed the gases forced around the fire floor door traveled down the stairs and
exited the front of the building 7 floors below. This flow path greatly decreases safety of occupants
evacuating and requires fire fighters to utilize their SCBA (self contained breathing apparatus) upon
entering the structure. In addition, visibility was limited on all 7 floors of the stairwell which would
slow fire department operations.

The use of the PPV fans greatly improved conditions in the stairwell. They were not able to keep them
completely free of smoke under wind driven conditions but they were very effective and moving the

112
flow up the stairwell and holding the smoke and hot gases when the fire floor door was closed. With the
door open the flow was slowed but not stopped. After the bulkhead door was opened the temperatures
in the public hall and the stair continued to increase until the fan was moved into place at the base of the
stair on the first floor. Then the 5th floor door was opened and another fan was added at that location. It
is important to note the temperatures in the stair were reduced and kept at ambient conditions with two
fans. A small increase in temperature occurred when the stair door on the 7th floor was opened 3 inches.
Turning off the fan on 5 and opening the stair door on the 7th floor results in a significant temperature
increase in the stair. Opening the bulkhead door at approximately 1370 s, was followed by ignition of
the sofa in the living room. This increased the temperature in the public hallway until the living room
window self vents and the bulkhead was closed. At this point the single 27 inch fan on the first floor
was working to cool the stair and the public hall by pushing air back to Apartment 7E.

113
3.4 Experiment 7A 

The fourth experiment in the structure was located on the 7th floor in apartment 7A. Experiment 7A was
ignited in the bedroom furthest from the open apartment door (Figure 3.4-1). This experiment also
utilized a simulated wind provided by the MVU. The simulated wind was subjected to the bedroom
double window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The middle
bedroom was also furnished and its door remained open during the experiment. The events that took
place during the experiment to accomplish the experimental objectives below are shown in Table 3.4-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Examine impact of the deployment of a WCD on fire room window
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.4-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

114
Figure 3.4-1. 7th floor apartment A floor plan with locations pertinent to Experiment 7A

115
Table 3.4-1: Experiment 7A Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
282 286 7th floor stair A Door opened and closed
316 316 Bedroom double Bottom right pane failed
432 433 Bedroom double Top right pane failed
453 454 Bedroom double Bottom left pane failed
587 596 Front lobby Doors opened
608 609 Bedroom double Top left pane failed
615 620 1st floor stair A Door opened
684 686 Bulkhead door Opened
746 749 7th floor stair A Door opened (1/2)
813 825 7th floor stair A Door closed
840 848 Bedroom double WCD partially deployed
896 904 Bedroom double WCD fully deployed
984 986 7th floor stair A Door opened 0.08 m (3 in)
1056 1060 7th floor stair A Door opened
1089 1091 7th floor stair A Door closed
1128 1135 1st floor stair A 27 inch fan activated
1168 1170 Bulkhead door Closed
1194 1199 7th floor stair A Door opened
1311 1314 Living room window Pane failed
1324 1325 Living room window Pane failed
1327 1328 Living room window Pane failed
1344 1344 Living room window Pane failed
1356 1356 Living room window Pane failed
1647 1647 Bedroom double FBN 2 started flowing (Fog)
1772 1793 Bedroom double WCD removed
1932 1932 Bedroom double FBN 2 stopped
2011 2016 1st floor stair A 27 inch fan deactivated
2092 2096 1st floor stair A 27 inch fan activated
2534 2534 Experiment completed

116
Figure 3.4-2. Instrumentation and camera locations with labels.

117
3.4.1 Observations 

The observations are presented as a series of images captured from eight camera locations, six were
video cameras and two were thermal imaging cameras. The camera positions inside the building are
shown in Figure 3.4-2.

Figure 3.4-3 through Figure 3.4-21 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 40 min (2400 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 7A, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor and corridor
thermal imaging views (CORRIDOR and CORR IR) show the corridor leading up to the open apartment
doorway; the corridor IR had been damaged in the previous experiment and was not functioning
properly, however it does provide some information. The next two views (STAIR and STAIR IR) show
the stairwell with a visual and a thermal imaging camera from the floor below the fire looking up to the
fire floor. The bottom left view (ROOF) shows the bulkhead door at the top of the stairwell and the
bottom right view (OUTSIDE) shows an external view of the fire apartment.

Figure 3.4-3 shows the images at the time of ignition. At this point, the six video views were clear and
unobstructed. The corridor thermal imaging camera was also damaged from the previous experiment
but still had a view so it was included in the recording. This experiment shares the corridor from
Experiment 7E so the corridor is thermally damaged and there is a bit of a lingering haze of smoke in the
living room.

Figure 3.4-4 shows the images at 240 s (4 min) after ignition. The smoke layer descended to the floor
throughout the 7A apartment and the public corridor. The stairwell, which was closed at the 7th floor,
remained clear of smoke. The outside view shows all of the windows intact and the fire fighting crew in
place to deploy a WCD.

Figure 3.4-5 shows the images at 480 s (8 min) after ignition. The 7th floor stair A door had been
opened and closed 194 s prior, this had allowed a significant volume of smoke into the stairwell;
visibility in the stairwell view had reduced to zero. The right side of the bedroom double window had
failed in three sections 164 s, 47 s, and 26 s prior and the fire had progressed significantly in the
bedroom. The increasing temperature of the 7th floor stair A door was visible in the stair IR view.

Figure 3.4-6 shows the images at 630 s (10 min and 30 s) after ignition. The bedroom had become fully
involved in flame and the view in the living room was obscured by soot. Flames were visible leaving
the fire apartment and entering the corridor. Visibility was lost in the stairwell and heat was being
forced around the gaps in the doorway. Flames are visible at the bedroom window but all smoke and
flames are being forced into the window by the simulated wind.

118
Figure 3.4-7 shows the images at 720 s (12 min) after ignition. The bulkhead door had been opened 35 s
prior. Flames were pushing into the public corridor through the apartment entrance. The stairwell,
which was closed at the 7th floor, was clearing of smoke below the fire floor as the smoke moved up
through the open bulkhead doorway. More flames are visible at the bedroom window but all smoke and
flames are being forced into the window by the simulated wind.

Figure 3.4-8 shows the images at 780 s (13 min) after ignition. Flames now involved the entire
apartment from the bedroom, through the living room and extended into the corridor. Hot gases were
flowing into the stairwell as shown in the stairwell thermal imaging camera. The corridor thermal
imaging camera shows high heat levels but there is no usable image.

Figure 3.4-9 shows the images at 910 s (15 min and 10 s) after ignition. The WCD was deployed and
positioned over the bedroom double window blocking the wind exposure. The bedroom was fully
involved in flames and the living room view was still obscured by smoke. The corridor and corridor IR
camera failed due to thermal exposure from the wind driven fire. The conditions in the stairwell
improved substantially with visibility returning and all of the combustion products venting vertically
through the open bulkhead doorway. The stairwell door had been closed slowing the flow into the
stairwell.

Figure 3.4-10 shows the images at 960 s (16 min) after ignition. None of the fire floor was visible from
the video views. The closed stairwell remained clear below the fire floor but smoke was leaking around
the stair door and flowing through the open bulkhead doorway.

Figure 3.4-11 shows the images at 1080 s (18 min) after ignition. The 7th floor stair A door had been
fully opened 20 s prior and the smoke is flowing into the stairwell similar to Experiment 7G, when there
was no wind effects. The fire in the bedroom was visible again and it was burning toward the window
where some air was leaking in through the bottom of the blanket where it was not completely sealed
over the window.

Figure 3.4-12 shows the images at 1200 s (20 min) after ignition. The bulkhead door had been closed
30 s prior and stairwell A was being pressurized by a 27 inch fan placed at the 1st floor entrance to that
stairwell. No smoke was visible in the stair view and no smoke was leaking from the bulkhead door.

Figure 3.4-13 shows the images at 1440 s (24 min) after ignition. The living room window had begun to
fail 129 s prior. There was still no smoke or heat flow in the stairwell. The WCD was continuing to
hold up to the extreme temperatures it was being subjected to. The silver coating on the WCD was
missing in spots but the WCD itself was still completely intact to stop the wind.

Figure 3.4-14 shows the images at 1680 s (28 min) after ignition. FBN 2 was inserted into the bottom of
the window and started flowing water 32 s prior, but no visible change is apparent in the bedroom or
living room views because they are obscured by soot deposition.

Figure 3.4-15 shows the images at 1830 s (30 min and 30 s) after ignition. The WCD had been removed
40 s prior and the FBN was continuing to flow. The bulk of the fire was extinguished in the bedroom
but flames are still visible in the living room. Conditions remained the same in the stairwell with the
portable PPV fan continuing to keep it free of smoke and heat.

119
Figure 3.4-16 shows the images at 1920 s (32 min) after ignition. The FBN 2 was continuing to flow
into the ignition bedroom, and without the WCD, visibility was improved in that view. Flames only
continued to burn on a shielded item in the corner of the bedroom. A red glow was still visible in the
living room view but no heat was being forced into the stairwell.

Figure 3.4-17 shows the images at 2010 s (33 min and 30 s) after ignition. Suppression had stopped.
The portable fan was still operating at the base of the stairwell. Figure 3.4-18 shows the images at
2090 s (34 min and 50 s) after ignition. This was approximately 75 s after the fan was turned off and the
smoke was being forced into the stairwell, reducing visibility. Figure 3.4-19 shows the images at 2120 s
(35 min and 20 s), 25 s after the fan was turned back on clearing the stair of smoke.

Figure 3.4-20 shows the images at 2160 s (36 min) after ignition and Figure 3.4-21 shows the images at
2400 s (40 min) after ignition. No significant change in conditions was visible; the fan continued to
keep the stairwell free of smoke while the fire was allowed to burn down.

120
Figure 3.4-3. Experiment 7A, ignition.

121
Figure 3.4-4. Experiment 7A, 240s after ignition.

122
Figure 3.4-5. Experiment 7A, 480s after ignition.

123
Figure 3.4-6. Experiment 7A, 630s after ignition.

124
Figure 3.4-7. Experiment 7A, 720s after ignition.

125
Figure 3.4-8. Experiment 7A, 780s after ignition.

126
Figure 3.4-9. Experiment 7A, 910s after ignition.

127
Figure 3.4-10. Experiment 7A, 960s after ignition.

128
Figure 3.4-11. Experiment 7A, 1080s after ignition.

129
Figure 3.4-12. Experiment 7A, 1200s after ignition.

130
Figure 3.4-13. Experiment 7A, 1440s after ignition.

131
Figure 3.4-14. Experiment 7A, 1680s after ignition.

132
Figure 3.4-15. Experiment 7A, 1830s after ignition.

133
Figure 3.4-16. Experiment 7A, 1920s after ignition.

134
Figure 3.4-17. Experiment 7A, 2010s after ignition.

135
Figure 3.4-18. Experiment 7A, 2090s after ignition.

136
Figure 3.4-19. Experiment 7A, 2120s after ignition.

137
Figure 3.4-20. Experiment 7A, 2160s after ignition.

138
Figure 3.4-21. Experiment 7A, 2400s after ignition.

139
3.4.2 Temperatures 

Figure 3.4-22 through Figure 3.4-24 provide temperature data for five measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.4-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.4-22 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 7E in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
temperature in the bedroom increased to more than 450 ºC (842 ºF) until a window pane failed lowering
the temperature. The temperature recovered and then two more panes failed decreasing the bedroom
temperature again. At approximately 540 s the bedroom transitioned to flashover, quickly driving the
bedroom temperature to above 800 ºC (1472 ºF). The WCD was deployed and the bedroom temperature
increased briefly until it was completely in place, blocking the entire window, and then decreased from
1000 ºC (1832 ºF) to below 500 ºC (932 ºF) in 180 s.

The 2.1 m (7 ft) temperatures in the corridor measurement locations slowly increased as the fire in the
bedroom grew. As the bedroom flashed over and a wind driven condition was developed the corridor
temperatures increased to above 600 ºC (1112 ºF). After the WCD was in place the corridor
temperatures decreased from over 600 ºC (1112 ºF) to below 300 ºC (572 ºF) in 100 s. These
temperatures recovered up to almost 400 ºC (932 ºF) before the FBN was activated and the WCD was
removed lowering the temperatures.

The 2.1 m (7 ft) temperature in the stairwell increased as hot gases were forced into the stairwell from
gaps around the door. The 7th floor door was opened at 750 s which caused the stairwell temperature to
increase from 150 ºC (302 ºF) to 700 ºC (1292 ºF) in seconds. Closing the door and the deployment of
the WCD drove the temperature back down to below 200 ºC (392 ºF). A portable PPV fan was activated
and that kept the stairwell below 150 ºC (302 ºF) for the remainder of the experiment.

Figure 3.4-23 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. These temperatures behaved very
similarly to the temperatures at 2.1 m (7 ft) after the wind driven condition was developed. This
indicates that after that time there was a uniform well mixed environment. After deployment of the
WCD the 1.2 m (4 ft) temperatures in the corridor dropped to 100 ºC (212 ºF) to 200 ºC (392 ºF) below
the 2.1 m (7 ft) temperatures re-establishing a two layer environment by removing the impact of the
wind.

Figure 3.4-24 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, 6 and at the
bulkhead. The temperatures below the fire floor remained at ambient during the entire experiment.
While the wind driven condition was flowing into the stairwell the bulkhead door was opened allowing
the hot gases to flow up the stairwell. The temperature at the bulkhead door increased to 100 ºC
(212 ºF) until the bulkhead door was opened. The temperature dropped briefly and then increased to
above 600 ºC (1112 ºF) quickly as the flow path was established through the open 7th floor door. As he
fire floor door was closed the temperature at the top of the stair decreased to below 100 ºC (212 ºF).

140
After the PPV fan was activated the temperature remained below 100 ºC (212 ºF) for the remainder of
the experiment.
)
) l e) ble )
u ble ub b le) ou uble
o d d
g o
(d e ( d m ou ( do te
sin ng r f D c 8 in g ar g (d CD d ( p le
i l e
clo il lea ha C la .0
W p 0 ail le in W pp om
n g/ e fa ing e c g ng g g in g g w f w c low ing to tc
i tio enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn op do : o do op op op clo pla W op op clo fan clo op win win FB re FB fan fan r im
:I : : Y : : : : : : : : : : : : : : : : : : : : pe
BR S7 BR LB BR S1 BH S7 S7 BR BR S7 S7 S7 S1 BH S7 LR LR BR BR BR S1 S1 Ex
1200 2192
TC BR7AE 2.1m
TC Corr7AE 2.1m
TC Corr7EE 2.1m
1000 TC CorrStair7E 2.1m
1832
TC Stair7E 2.1m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.4-22. 7A Temperature vs. time at 2.1 m above the floor

)
) l e) ble )
u ble ub b le) ou uble
o o d d
(d e ( d m ou ( o te
s
g
in ng r f D c 8 n g r (d CD d (d p le
i l i a g e
clo il lea ha C la .0
W p 0 ail le in W pp om
n g/ e fa ing e c g ng g g in g g w f w c low ing to tc
ti io enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn p o o o p p p lo la p p lo n lo p in in B e B n n r im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: fa 1: fa x pe
B B
S B L B S B S S B B S S S S B S L L B B B S S R R E
1200 2192
TC BR7AE 1.2m
TC Corr7AE 1.2m
TC Corr7EE 1.2m
1000 TC CorrStair7E 1.2m
1832
TC Stair7E 1.2m
Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.4-23. 7A Temperature vs. time at 1.2 m above the floor

141
 ) e)
le) ble le ) ubl le)
b u o b
ou o ub (d ou ed
g (d e ( d m do CD d (d et
si n
i ng a r l f D a c 0 8 l in g ar g
(
e pl
clo il le ha C l . ai le in W pp o m
g/ e fa ing e c g ng g W p 0 g w f w c low ing to tc
n g in g
i tio enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn op do : o do op op op clo pla W op op clo fan clo op win win FB re FB fan fan r im
:I : : Y : : : : : : : : : : : : : : : : : : : : pe
BR S7 BR LB BR S1 BH S7 S7 BR BR S7 S7 S7 S1 BH S7 LR LR BR BR BR S1 S1 Ex

800 1472
TC Bulkhead MidE
TC Stair 6E
TC Stair 4E
TC Stair 2E
TC Stair1 MidE
600 1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.4-24. 7A stairwell temperature vs. time

3.4.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.4-25. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
2 Pa and only fluctuated slightly. After the bedroom window began to fail the wind that leaked around
the gaps in the stairwell door increased the pressure in the closed stairwell to closer to 5 Pa. The base of
the stair was opened which dropped the pressure and the bulkhead was opened which dropped the
pressures to close to zero. There was a spike to 12 Pa in the 7th floor stairwell pressure when the door
was opened at 750 s. The stairwell was opened again after the WCD was deployed and there was no
significant increase in stairwell pressure. At 1135 s a portable PPV fan was activated increasing the
stairwell pressures from 5 Pa to 25 Pa depending on the distance from the fan. These pressures were
with the 7th floor door to the stair open. At 1790 s the WCD was removed increasing the pressures in
the stair to 15 Pa to 22 Pa. There was no pressure increase at floors 1 and 2. Turning the fan off
decreased the pressures to below 8 Pa throughout the stairwell and turning it back on returned the
pressures back to previous levels.

142
 ) e)
le) ble le ) ubl le)
b u o b
ou o ub (d ou ed
g (d e ( d m do CD d (d et
s i n
i ng ar l f D a c 0 8 l in g ar g
(
e pl
clo il le ha C l . ai le in W pp o m
g/ e fa ing e c g ng g W p 0 g w f w c low ing to tc
n g in g
i tio enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn op do : o do op op op clo pla W op op clo fan clo op win win FB re FB fan fan r im
:I : : Y : : : : : : : : : : : : : : : : : : : : pe
BR S 7 B R LB B R S 1 B H S 7 S 7 B R B R S 7 S 7 S 7 S 1 B H S 7 LR LR B R B R B R S 1 S 1 Ex

30
DP Stair 7E
DP Stair 6E
DP Stair 5E
DP Stair 4E
DP Stair 3E
20 DP Stair 2E
DP Stair 1E
Pressure (Pa)

10

-10
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.4-25. 7A stairwell pressures vs. time

3.4.4 Velocities 

Average doorway velocities from 4 doors; the fire apartment door, the door from the fire floor to the
stairwell, the door at the base of the stair and the door at the bulkhead, were recorded versus time and
shown in Figure 3.4-26. The three bi-direction probes in each door were averaged to provide the bulk
flow through the doors. A time history for each probe at each measurement location is located in
Appendix D – Detailed Graphs. The velocity out of the open apartment door steadily increased as the
fire grew. Opening the 7th floor stair door initiated the wind driven condition which had a peak velocity
of 21 m/s (45 mph). The velocity into the stairwell and out of the top of the stairwell peaked at
approximately 12 m/s (27 mph). After the WCD was deployed the velocities decreased even with the
stairwell doors open which would be expected. With the use of the fan, the velocities were less than
2 m/s (4 mph) toward the fire apartment.

143
 ) e)
le) ble le ) ubl le)
b u o b
ou o ub (d ou ed
g (d e ( d m do CD d (d et
si n
i ng a r l f D a c 0 8 l in g ar g
(
e pl
clo il le ha C l . ai le in W pp o m
g/ e fa ing e c g ng g W p 0 g w f w c low ing to tc
n g in g
i tio enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn op do : o do op op op clo pla W op op clo fan clo op win win FB re FB fan fan r im
:I : : Y : : : : : : : : : : : : : : : : : : : : pe
BR S7 BR LB BR S1 BH S7 S7 BR BR S7 S7 S7 S1 BH S7 LR LR BR BR BR S1 S1 Ex
30 67
Exp 7A Bulkhead
Exp 7A Stair1
Exp 7A Stair7
Exp 7A Apt7A
20 45

Velocity (mph)
Velocity (m/s)

10 22

0 0

-10 -22

-20 -45
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 3.4-26. 7A Velocities vs. time

3.4.5 Discussion 

Experiment 7A utilized a simulated wind created by the MVU. A wind driven fire was created from the
fire ignited in the bedroom. Flames were forced out of the apartment and into the corridor with the
stair door closed and conditions worsened further when the stairwell door was opened allowing the hot
gases to flow throughout the stairwell. These conditions would not be conducive to an interior direct
frontal fire fighter attack. Stairwell temperatures were in excess of 600ºC (1112 ºF) with the door open.

The first pane of the bedroom window failed at 316 s and the window completely failed at 609 s. The
temperature in the center of the bedroom at 1.2 m (4 ft) at the time of window failure was approximately
350 ºC (662 ºF). The temperature in the center of the room at the time of complete window failure was
in excess of 800 ºC (1472 ºF). This demonstrated that when a portion of the window failed the wind
being introduced through the opening can cool and slow the failure of the rest of this type of window.

This experiment highlighted the importance of door control. A wind driven condition was created on
the fire floor and then the stairwell door was opened. With the bulkhead door also open the flow path
between the fire floor and the top of the stairwell increased from less than 50 ºC (122 ºF) to over 400 ºC
(752 ºF) in less than 15 s. The stairwell temperature peaked at 700 ºC (1292 ºF) in 35 s at which time
the fire floor stairwell door was closed. Closing the door lowered temperatures to less than 200 ºC
(392 ºF) quickly.

There was a small delay in getting the WCD in place, but once it was fully deployed over the window
the impact was beneficial to conditions throughout the structure. Fire apartment and corridor
temperatures reduced by more than half in less than 30 s. More importantly the stairwell temperatures at
the fire floor were reduced to ambient from 400 ºC (752 ºF) and the stairwell temperatures above the fire

144
floor were reduced from 250 ºC (482 ºF) to less than 100 ºC (212 ºF). Without the WCD, even with the
door to the fire floor shut the stairwell at the fire floor and above became untenable for occupants as
well as fire fighters as hot gases continued to be forced around the gaps around the fire floor door.
These temperatures would have forced the fire fighters to stay below the fire floor with no ability to
advance up the stairwell until the WCD was deployed.

The impact of the WCD was observed and then a portable PPV fan was used to control smoke and heat
movement. The fan was activated at the base of the stair and the stairwell temperatures never increased
above 100 ºC (212 ºF) although the corridor temperature right inside the stairwell door was in excess of
300 ºC (572 ºF). Even with the living room well involved in fire the stairwell remained free of smoke
and heat with the fire floor stairwell door open. When it was briefly turned off for 78 s during the end of
the experiment, the smoke moved into and down the stairwell reducing visibility. Within 30 s of
restarting the fan, the stairwell was free of smoke.

145
3.5 Experiment 7K 

The fifth experiment in the structure was located on the 7th floor in apartment 7K. Experiment 7K was
also ignited in the bedroom furthest from the open apartment door (Figure 3.5-1). This experiment also
utilized a simulated wind provided by the MVU. The simulated wind was subjected to the bedroom
double window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The events that
took place during the experiment to accomplish the experimental objectives below are shown in Table
3.5-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Examine impact of the deployment of a WCD on fire room window
• Utilize portable PPV fans to control smoke movement
• Control wind driven conditions with FBN

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.5-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

146
Figure 3.5-1. 7th floor apartment K floor plan with locations pertinent to Experiment 7K

147
Table 3.5-1: Experiment 7K Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
164 167 7th floor stair A Door closed
240 241 Bedroom double Bottom left pane failed
240 241 7th floor stair A Door opened 0.08 m (3 in)
248 249 Bedroom double Bottom right pane failed
251 252 7th floor stair A Door closed
263 263 Bedroom double Top left pane failed
286 287 Bedroom double Top right pane failed
327 Bulkhead door Opened
326 332 7th floor stair A Door opened
361 363 Bulkhead door Closed
368 385 Front lobby Doors opened
391 395 7th floor stair A Door closed
406 410 1st floor stair A Door opened
461 467 Bulkhead door Opened
486 491 7th floor stair A Door opened
547 552 7th floor stair A Door closed
554 577 Living room window Failure, starting along top
587 588 Bedroom double WCD deployed
641 642 7th floor stair A Door opened 0.08 m (3 in)
669 672 7th floor stair A Door opened
733 738 7th floor stair A Door closed
766 777 1st floor stair A 27 inch fan activated
849 851 Bulkhead door Closed
879 881 7th floor stair A Door opened
1505 1512 Bedroom double WCD removed
1589 1589 Living room window FBN 2 started flowing (Fog)
1665 1665 Living room window FBN 2 stopped
1683 Bedroom double Air flow removed
1745 1745 Experiment completed

148
Figure 3.5-2. Instrumentation and camera locations with labels.

149
3.5.1 Observations 

The observations are presented as a series of images captured from eight camera locations, seven of
which were video cameras and one of which was an IR camera. The camera positions inside the
building are shown in Figure 3.5-2.

Figure 3.5-3 through Figure 3.5-17 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 28 min (1680 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 7K, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor view
(CORRIDOR) shows the corridor leading up to the open apartment doorway. Next to that, the outside
lobby view (LOBBY) shows the area just outside the lobby doors. The next two views (STAIR and
STAIR IR) show the stairwell with a visual and a thermal imaging camera from the floor below the fire
looking up to the fire floor. The bottom left view (ROOF) shows the bulkhead door at the top of the
stairwell and the bottom right view (OUTSIDE) shows an external view of the fire apartment.

Figure 3.5-3 shows the images at the time of ignition. At this point, the seven video views were clear
and unobstructed. However, the thermal image from the stairwell provided limited thermal contrast
because the surfaces in the view are at nearly equal temperature. This was the second experiment in this
half of the building so there is preexisting smoke and heat damage to the corridor and stairwell. The 7th
floor door to the stair was open at the start of the experiment.

Figure 3.5-4 shows the images at 240 s (4 min) after ignition. The bottom left window pane had just
begun to fail and the wind was moving smoke and heat throughout the floor. The stair IR camera began
to fail prior to this view and subsequently would provide limited information. The smoke layer
throughout the fire floor had descended to the floor and was beginning to leak into the stairwell, which
was closed on the 7th floor.

Figure 3.5-5 shows the images at 340 s (5 min and 40 s) after ignition. The bedroom window had
completely failed and the 7th floor stair door had been opened again for 10 s. Flames were visible
throughout the bedroom and moving through the living room. Visibility was lost in the corridor and
stairwell. There was enough smoke being forced around the closed bulkhead door to obscure the roof
camera view. Notice how flames were pulsing back out of the bedroom window against the air flow.

Figure 3.5-6 shows the images at 480 s (8 min) after ignition. The stairwell had been closed on the fire
floor more than 80 s prior and the bulkhead and 1st floor stair A doors had been opened approximately
20 s and 70 s prior, respectively. Smoke had not cleared the stairwell; visibility in this view was still
obscured. However, a significant volume of smoke was flowing from the bulkhead door. The bedroom
was still full of flames and the flames continued to pulse out of the window toward the simulated wind.

Figure 3.5-7 shows the images at 550 s (9 min and 10 s) after ignition. At this time the stair door was
being closed and bulkhead door was opened, allowing a flow path from the bedroom to the top of the

150
stairwell. Visibility returned to the living room showing flames involving the sofas and floor area.
Flames were flowing out of the fire apartment and out into the corridor. Smoke was thinning in the
stairwell as flaming combustion was increased due to the available oxygen from the complete flow path.

Figure 3.5-8 shows the images at 585 s (9 min and 45 s) after ignition. This was captured just before
deployment of the WCD over the bedroom window. Flames were still pulsing out of the bedroom
window and the living room window was beginning to fail. Visibility was decreased in the living room,
corridor and stairwell.

Figure 3.5-9 shows the images at 620 s (10 min and 20 s) after ignition. The WCD was deployed over
the bedroom window cutting off the wind from the fire. The smoke out of the living room window had
decreased and turned white.

Figure 3.5-10 shows the images at 720 s (12 min) after ignition. The burning had decreased in the
bedroom and visibility was still poor in the living room and corridor. The visibility in the stairwell was
improved slightly as the smoke was moving up the stairwell through the open bulkhead door.

Figure 3.5-11 shows the images at 960 s (16 min) after ignition. The bulkhead door had been closed
110 s prior and the 7th floor stair A door had been closed and opened approximately 220 s and 75 s prior,
respectively. A PPV fan was activated at the base of the stairwell and conditions in the stairwell were
improved and the cloudy view was due to soot deposited on the camera lens.

Figure 3.5-12 shows the images at 1200 s (20 min) after ignition and Figure 3.5-13 shows the images at
1440 s (24 min) after ignition. No appreciable change in conditions was visible from the video views.
The stairwell remained free of smoke and there was no visible burning in the living room.

Figure 3.5-14 shows the images at 1500 s (25 min) after ignition. The living room was rekindling and
flames became visible in the video view. Smoke production was increasing out of the living room
window. The smoke condition in the stairwell was unchanged. This was also the time 5 s before the
WCD was removed allowing the reintroduction of the simulated wind.

Figure 3.5-15 shows the images at 1560 s (26 min) after ignition. The WCD was removed 55 s prior and
the wind was introduced through the window. Flames in the living room increased and were forced out
of the living room windows. Smoke and heat were flowing back into the stairwell against the pressure
from the fan.

Figure 3.5-16 shows the images at 1600 s (26 min and 40 s) after ignition. FBN 2 was inserted into the
living room window and was flowing with a fog nozzle for 20 s at the time of this image. Large
amounts of steam were seen coming from the living room. Visibility was still minimal in the living
room, corridor and stairwell.

Figure 3.5-17 shows the images at 1680 s (28 min) after ignition. The view in the living room had
partially returned and there were no visible flames. The conditions in the stairwell were improved and
smoke flow out of the apartment visible from the outside was minimal.

151
Figure 3.5-3. Experiment 7K, ignition.

152
Figure 3.5-4. Experiment 7K, 240s after ignition.

153
Figure 3.5-5. Experiment 7K, 340s after ignition.

154
Figure 3.5-6. Experiment 7K, 480s after ignition.

155
Figure 3.5-7. Experiment 7K, 550s after ignition.

156
Figure 3.5-8. Experiment 7K, 585s after ignition.

157
Figure 3.5-9. Experiment 7K, 620s after ignition.

158
Figure 3.5-10. Experiment 7K, 720s after ignition.

159
Figure 3.5-11. Experiment 7K, 960s after ignition.

160
Figure 3.5-12. Experiment 7K, 1200s after ignition.

161
Figure 3.5-13. Experiment 7K, 1440s after ignition.

162
Figure 3.5-14. Experiment 7K, 1500s after ignition.

163
Figure 3.5-15. Experiment 7K, 1560s after ignition.

164
Figure 3.5-16. Experiment 7K, 1600s after ignition.

165
Figure 3.5-17. Experiment 7K, 1680s after ignition.

166
3.5.2 Temperatures 

Figure 3.5-18 through Figure 3.5-20 provide temperature data for five measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.5-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.5-18 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 4 measurement
locations, just outside the fire apartment in the corridor, just outside Apartment 7G in the corridor, just
outside the A stairwell doorway, and just inside the A stairwell on the landing. The bedroom 2.1 m
(7 ft) thermocouple did not function correctly so the 1.8 m (6 ft) thermocouple is presented in its place.
The bedroom temperature in this experiment increased to approximately 700 ºC (1292 ºF) prior to
window failure and continued to increase to above 900 ºC (1652 ºF) prior to WCD deployment. After
WCD deployment the temperature decreased by half to approximately 450 ºC (842 ºF) in approximately
90 s. This temperature increased again to above 600 ºC (1112 ºF) as the fire in the living room grew and
increased again after the WCD was removed at 1505 s.

The temperatures just outside the fire apartment in the corridor increased quickly after the door to the
stairwell on the fire floor was opened. The 2.1 m (7 ft) temperature at this location increased from
approximately 300 ºC (572 ºF) to 700 ºC (1292 ºF) in less than 10 s. The wind driven condition
appeared to stop and the temperatures decreased as the fire downstream became ventilation limited and
the door to the 7th floor stairwell was closed. The bulkhead door was opened and then the 7th floor stair
door was reopened and the temperature increased quickly again from 400 ºC (752 ºF) to 800 ºC
(1472 ºF) in 10 s. The wind driven condition was reestablished and then the WCD was deployed over
the bedroom window reducing the temperature in the corridor just outside the apartment from 700 ºC
(1292 ºF) to less than 300 ºC (572 ºF) in 60 s. Towards the end of the experiment the WCD was
removed and the fire in the living room increased to flashover. This drove the temperature up to 700 ºC
(1292 ºF) and that was quickly reduced to below 200 ºC (392 ºF) within 10 s of the FBN activation.

The temperatures at the other end of the corridor near the entrance to the stairwell and Apartment 7G
responded to the events similarly. This location increased to approximately 400 ºC (752 ºF) during the
wind driven conditions and decreased when the door to the stairwell was closed or the WCD was in
place. When the WCD was in place, stair door was opened and the PPV fan was activated the corridor
temperatures reduced to ambient levels. These locations increased in temperature when the WCD was
removed and the living room flashed over but quickly decreased after the FBN was activated.

The temperature at 2.1 m (7 ft) above the landing in the stairwell increased substantially when the door
was opened at both 330 s and 495 s. This location’s temperature increased to as high as 400 ºC (752 ºF)
but decreased to less than 100 ºC (212 ºF) after the door was closed. With the WCD in place and a PPV
fan operating at the base of the stairwell the temperatures in the stairwell remained ambient. When the
WCD was removed and the stair was open the temperature increased to 200 ºC (392 ºF) but decreased
quickly after the activation of the FBN.

Figure 3.5-19 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. The temperature trends at this

167
elevation are very similar to those at 2.1 m (7 ft). The temperatures responded quickly to the presence
of a flow path into the stairwell. When the 7th floor stair door was opened the temperatures in the
corridor rose quickly and when the stair door was closed the corridor temperatures decreased quickly.
When the stair door was opened at 495 s the temperature right outside of the fire apartment increased
from 300 ºC (572 ºF) to as high as 800 ºC (1472 ºF) before the door was closed. After the WCD was
deployed and the door was reopened the corridor temperature did not increase at all.

The temperature on the stairwell landing at 1.2 m (4 ft) would be important because that is where the fire
fighting crew would be advancing from after opening the door to the fire floor. When the door was
opened this temperature increased from approximately 50 ºC (122 ºF) to over 300 ºC (572 ºF) in a matter
of seconds twice in this experiment. Both times the door closing returned temperatures to less than
100 ºC (212 ºF). After the WCD was applied the stairwell door was opened and the same temperature
measurement location increased from 50 ºC (122 ºF) to 100 ºC (212 ºF) which would be conducive to
fire fighter advancement toward the fire apartment.

Figure 3.5-20 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, 6 and at the
landing above the 7th floor with access to the roof. The door to the fire floor was opened twice during
wind driven fire conditions. Both times the temperature at the top of the stairwell increased from less
than 50 ºC (122 ºF) to above 250 ºC (482 ºF) in less than 10 s. Temperatures on the sixth floor, one
floor below the fire floor, saw temperature increases to above 200 ºC (392 ºF). Closing the fire floor
decreased these temperatures to below 50 ºC (122 ºF). Slight temperature increases were recorded at the
fourth floor which indicates that smoke made it to at least that point in the stairwell. With the WCD in
place and the wind effects blocked the temperature at the top of the stairwell increased from 40 ºC
(104 ºF) to 80 ºC (176 ºF) and there was no temperature increase below the fire floor.
BR: WCD removed (double)
BR: placing WCD (double)

Experiment completed
S7: opening 0.08m

S7: opening 0.08m

LR: window failing

Removing air flow

BR: double failing

LR: FBN stopped

LR: FBN flowing
LBY: opening
BH: opening

BH: opening
S7: opening

S1: opening

S7: opening

S7: opening
BR: Ignition

BH: closed

BH: closed
S7: closed

S7: closed

S7: closed

S7: closed

S7: closed
S1: fan on
S7: open

1200 2192
TC BR7KW 1.8m
TC Corr7KW 2.1m
TC Corr7GW 2.1m
1000 TC CorrStair7W 2.1m
1832
TC Stair7W 2.1m
Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.5-18. 7K Temperature vs. time at 2.1 m above the floor

168
 Temperature (C) Temperature (C)

0
100
200
300
400
0
200
400
600
800
1000
1200

0
0
BR: Ignition BR: Ignition

S7: closed S7: closed

BR: double failing BR: double failing

200
200
S7: opening 0.08m S7: opening 0.08m
S7: closed S7: closed
S7: opening S7: opening
BH: opening BH: opening
BH: closed BH: closed

400
400
LBY: opening LBY: opening
S7: closed S7: closed
S1: opening S1: opening
BH: opening BH: opening
S7: opening S7: opening

600
600
S7: closed S7: closed
LR: window failing LR: window failing

Figure 3.5-20. 7K stairwell temperature vs. time

BR: placing WCD (double) BR: placing WCD (double)
S7: opening 0.08m S7: opening 0.08m
S7: open S7: open

800
800
S7: closed S7: closed
S1: fan on S1: fan on
BH: closed BH: closed
S7: opening S7: opening

Time (s)
Time (s)
Figure 3.5-19. 7K Temperature vs. time at 1.2 m above the floor
TC Stair7W 1.2m
TC BR7KW 1.2m
TC Corr7KW 1.2m
TC Corr7GW 1.2m

1000
1000
TC CorrStair7W 1.2m

169
1200
1200

TC Stair 2W
TC Stair 4W
TC Stair 6W

TC Stair1 MidW
TC Bulkhead MidW

1400
1400

BR: WCD removed (double) BR: WCD removed (double)

LR: FBN flowing LR: FBN flowing
LR: FBN stopped LR: FBN stopped

1600
1600

Removing air flow Removing air flow

Experiment completed Experiment completed

1800
1800

32
32

212
392
572
752
392
752
1112
1472
1832
2192

Temperature (F) Temperature (F)

3.5.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.5-21. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
6 Pa and only fluctuated slightly. At 240 s, shortly after the bedroom window completely failed, the
stair door opened slightly then closed causing a spike in the stairwell pressure. After the stairwell door
was completely opened at 330 s, and the other doors in the stairwell were closed the pressures increased
45 Pa to 75 Pa with the higher pressures at the top of the stairwell. When the fire floor stairwell door
was reopened at 410 s, with the 1st floor door to the stairwell open and the bulkhead door open, the
pressure in the stairwell increased 15 Pa to 25 Pa at the top of the stairwell and remained ambient below
the fire floor.

Later in the experiment the stairwell was closed and a PPV fan was activated at the base of the stairwell.
With the bulkhead open to ventilate the stairwell the pressures ranged from 25 Pa at the bottom to 6 Pa
at the bulkhead door. The bulkhead was then closed to pressurize the stairwell and the stairwell pressure
increased to approximately 25 Pa throughout. The fire floor door to the stairwell was then opened with
the WCD in place and no wind impact and the pressures ranged from 25 Pa at the bottom of the stairwell
to 10 Pa at the top in the area of the fire floor. Removing the WCD at 1505 s and allowing the wind to
combine with the fan pressures increased the stairwell pressures to above 25 Pa throughout the stairwell.

BR: WCD removed (double)

BR: placing WCD (double)

Experiment completed
S7: opening 0.08m

S7: opening 0.08m

LR: window failing

Removing air flow

BR: double failing

LR: FBN stopped

LR: FBN flowing
LBY: opening
BH: opening

BH: opening
S7: opening

S1: opening

S7: opening

S7: opening
BR: Ignition

BH: closed

BH: closed
S7: closed

S7: closed

S7: closed

S7: closed

S7: closed
S1: fan on
S7: open

100
DP Stair 7W
DP Stair 6W
DP Stair 5W
80 DP Stair 4W
DP Stair 3W
DP Stair 2W
60 DP Stair 1W
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.5-21. 7K stairwell pressures vs. time

3.5.4 Velocities 

Average doorway velocities from 3 doors; the fire apartment door, the door from the fire floor to the

170
stairwell and the door at the base of the stair, were recorded versus time and shown in Figure 3.5-22.
The three bi-direction probes in each door were averaged to provide the bulk flow through the doors. A
time history for each probe at each measurement location is located in Appendix D – Detailed Graphs.
The velocities at the apartment doorway and stairwell doorway increase in response to the stairwell door
opening and the creation of a flow path. The velocity at the apartment doorway increased from
approximately 5 m/s to 17 m/s twice when the door was opened, and the stairwell door velocity
increased from 2 m/s to 7 m/s twice. The flow from the fire apartment to the stairwell had to turn 180
degrees to enter the stairwell which most likely accounted for the difference in velocity between the fire
apartment and the stairwell.

With the WCD in place the flow into the stairwell was decreased to approximately 1 m/s and when the
PPV fan was activated and the bulkhead was closed the flow at the fire floor door was 0 m/s to 1 m/s
toward the fire apartment. The bulk flow out of the fire apartment with the fan operating slowed as the
fire burned down from 4 m/s down to 0 m/s before the WCD was removed. This indicates that the flow
from the fan increases the pressure to stop or slow the flow into the stairwell but does not over power
and drive the flow back through the fire apartment, feeding large amounts of oxygen to the fire. Fan
flow does make it to the fire at the lower elevations because of lower temperatures and lower pressure
differences but the impact on fire growth from this flow was not obvious in this experiment.

BR: WCD removed (double)

BR: placing WCD (double)

Experiment completed
S7: opening 0.08m

S7: opening 0.08m

LR: window failing

Removing air flow

BR: double failing

LR: FBN stopped

LR: FBN flowing
LBY: opening
BH: opening

BH: opening
S7: opening

S1: opening

S7: opening

S7: opening
BR: Ignition

BH: closed

BH: closed
S7: closed

S7: closed

S7: closed

S7: closed

S7: closed
S1: fan on
S7: open

20 45
Exp 7K Stair1
Exp 7K Stair7
Exp 7K Apt7K
15 34

Velocity (mph)
Velocity (m/s)

10 22

5 11

0 0

-5 -11
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.5-22. 7K Velocities vs. time

3.5.5 Discussion 

Experiment 7K utilized the simulated wind from the MVU and created wind driven conditions on the 7th
floor. Fire gases were forced from the bedroom into the living room, igniting the living room and into

171
the corridor and stairwell. These conditions would not be conducive to an interior direct frontal fire
fighter attack. stairwell temperatures were in excess of 400 ºC (752 ºF) with the door open.

The first pane of the bedroom window failed at 180 s and the window completely failed at 224 s. The
temperature in the center of the bedroom at 1.2 m (4 ft) at the time of window failure was approximately
500 ºC (932 ºF). During this experiment the fire grew rapidly prior to window failure and was not
cooled by the window failure. The window failure allowed the fire to intensify immediately.

Just like the previous experiment, this experiment highlighted the importance of door control. A wind
driven condition was created on the fire floor and then the stairwell door was opened. Simply opening
the stairwell door caused a 300 ºC (572 ºF) to 400 ºC (752 ºF) increase in temperature throughout the
fire floor and stairwell in a matter of seconds. Closing the 7th floor stair door approximately 1 minute
after opening it resulted in a significant temperature decrease in both the corridor and the stairwell.
Beginning with the opening of the first floor stair door at 470 s, a flow path was opened up from the
bedroom through the corridor and up the stair to the bulkhead. After the 7th floor stair door and the
bulkhead door open the temperatures in the corridor increased to over 700 ºC (1292 ºF), floor to ceiling.

The 7th floor stair door was closed and then 30 s later the WCD was deployed over the vented bedroom
window. This resulted in a decrease in temperature in the bedroom of approximately 500 ºC (900 ºF).
The temperatures downstream also decreased from approximately 700 ºC (1300 ºF) to 180 ºC (350 ºF) at
1.2 m (4 ft) above the floor in the public corridor and 7th floor stair temperatures went from in excess of
300 ºC (570 ºF) down to 50 ºC (122 ºF) at 1.2 m (4 ft) above the floor in the stair. The 7th floor stair
door was opened and closed with the WCD still deployed causing an increase in temperature in the
stairwell from 40 ºC (104 ºF) to 80 ºC (176 ºF). This minor increase in temperature would be conducive
to a direct frontal fire attack.

The PPV fan on the 1st floor was activated at 840 s. This dropped the temperatures in the stairwell to
less than 30 ºC (85 ºF). Opening the 7th floor stair door at 945 s, with the fan operating had the effect of
reducing the temperatures in the corridor to ambient. While using the fan and the WCD, conditions were
close to ambient for firefighters in both the corridor and the stair. The WCD was removed at 1505 s and
the fire in the living room transitioned to flashover at 1515 s and the temperatures in the corridor
reached over 700 ºC (1292 ºF). With the fan operating the temperatures in the corridor remained below
200 ºC (392 ºF). The fan could not stop the flow, but it limited its impact with the fire floor door open.

The removal of the WCD at 1505 seconds resulted in post-flashover conditions in the living room and
an temperatures in the public hallway in excess of 700 ºC (1292 ºF). An FBN 2 with a fog nozzle,
narrow pattern, flowing approximately 225 gpm of water into the living room window was put into
operation at 1575 s and suppressed the fire in less than 60 s. Temperatures in the corridor and stairwell
decreased to below 200 ºC (392 ºF) in seconds and continued to decrease until the experiment was
ended, even with the simulated wind still flowing through the apartment.

172
3.6 Experiment 5E 

The sixth experiment in the structure was located on the 5th floor in apartment 5E. Experiment 5E was
ignited in the bedroom furthest from the open apartment door (Figure 3.6-1). This experiment also
utilized a simulated wind provided by the MVU. The simulated wind was subjected to the bedroom
double window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The events that
took place during the experiment to accomplish the experimental objectives below are shown in Table
3.6-1.

Experimental Objectives
• Deploy WCD prior to window failure
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Allow WCD to be subjected to fire conditions for a long period of time
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.6-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

173
Figure 3.6-1. 5th floor apartment E floor plan with locations pertinent to Experiment 5E

174
Table 3.6-1: Experiment 5E Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
88 98 Bedroom double WCD deployed
233 233 Bedroom double Protected window failing
297 303 5th floor stair A Door opened
373 376 Bulkhead door Opened
744 752 Front lobby Door opened
752 758 1st floor stair A Door opened
822 Bedroom single Cracked at top
876 892 5th floor stair A Door opened & closed 3x
899 902 5th floor stair A Door opened
979 981 Bedroom single Manually vented (top pane only)
1060 1064 Bulkhead door Closed
1112 1115 1st floor stair A 27 inch fan activated
1308 1315 Bedroom single WCD deployed
1363 1394 Bedroom double Air flow removed
1380 1383 5th floor stair A Door closed
1401 1406 Bulkhead door Opened
1455 1456 Bulkhead door Closed
1515 1518 5th floor stair A Door opened
1619 1622 1st floor stair A 27 inch fan deactivated
1630 1630 Experiment completed

175
Figure 3.6-2. Instrumentation and camera locations with labels.

176
3.6.1 Observations 

The observations are presented as a series of images captured from eight camera locations, seven were
video cameras and one was a thermal imaging cameras. The camera positions inside the building are
shown in Figure 3.6-2.

Figure 3.6-3 through Figure 3.6-16 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 44 min (2640 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 5E, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor view
(CORRIDOR) shows the corridor leading up to the open apartment doorway. Next to that, the outside
lobby view (LOBBY) shows the area just outside the lobby doors. The next two views (STAIR and
STAIR IR) show the stairwell with a visual and a thermal imaging camera from the floor below the fire
looking up to the fire floor. The bottom left view (ROOF) shows the bulkhead door at the top of the
stairwell and the bottom right view (OUTSIDE) shows an external view of the fire apartment.

Figure 3.6-3 shows the images at the time of ignition. At this point, the seven video views were clear
and unobstructed. The thermal image in the stairwell provided limited thermal contrast because the
surfaces in the view are at nearly equal temperature.

Figure 3.6-4 shows the images at 120 s (2 min) after ignition. The WCD was deployed over the double
bedroom window prior to failure from fire exposure. The fire involved the area at the end of the bed
where the trashcan was positioned. A smoke layer was visible throughout the fire floor at approximately
1.5 m (5 ft) above the floor.

Figure 3.6-5 shows the images at 240 s (4 min) after ignition. The window behind the WCD had just
failed. At this point the smoke layer had descended to the floor in the bedroom, corridor, and living
room views. Smoke had also begun to leak around the closed 5th floor stair A door into the stairwell.

Figure 3.6-6 shows the images at 310 s (5 min and 10 s) after ignition. The door to the stairwell was
opened 7 s prior and smoke was flowing into the stairwell but was unaffected by the wind subjected to
the bedroom window. Visibility was still very limited in the bedroom, living room and corridor.

Figure 3.6-7 shows the images at 395 s (6 min and 35 s) after ignition. This image shows the conditions
20 s after the bulkhead door was opened. Large amounts of smoke are visible entering the fire floor and
exiting the bulkhead doorway. The smoke in the stairwell lifted from the landing of the fire floor but
visibility was still minimal on the fire floor.

Figure 3.6-8 shows the images at 480 s (8 min) after ignition. Smoke was flowing up the stairwell
through the open bulkhead door. Conditions had not changed significantly since the last images were
recorded. Figure 3.6-9 shows the images at 720 s (12 min) after ignition. The smoke flow continued
through this point with a visible decrease in smoke density in the public corridor. The thermal imaging

177
view showed that heat flow into the stairwell was increasing. An area of discoloration formed on the
WCD.

Figure 3.6-10 shows the images at 960 s (16 min) after ignition. The 5th floor stair A door had been
open and closed several times in an attempt to fail the unprotected window in the bedroom. At this
point, the smoke density on the fire floor had decreased and the view in the bedroom indicated that the
fire had smothered and was not producing much smoke. The thin smoke out of the bulkhead also
indicated that burning had decreased. The lobby and stair door on the first floor had also been opened
approximately 200 s prior.

Figure 3.6-11 shows the images at 980 s (16 min and 20 s) after ignition. The outside view shows the
unprotected window in the bedroom being manually ventilated with a pike pole from the floor above to
try to give the fire more oxygen.

Figure 3.6-12 shows the images at 1200 s (20 min) after ignition. The bedroom single window had been
vented manually 220 s prior and smoke was flowing out of it. Smoke density in the public corridor had
further decreased as the fire apartment was drawing air from the stairwell. The flames in the bedroom
had increased with the added oxygen from the ventilated window pane and open stairwell. A PPV fan
was activated 85 s prior with the bulkhead door closed to pressurize the stairwell. There was still
burning in the bedroom and smoke flowing from the open window pane. Smoke and heat flow into the
stairwell was reduced.

Figure 3.6-13 shows the images at 1305 s (21 min and 45 s) after ignition and Figure 3.6-14 shows the
images at 1320 s (22 min) after ignition. These images were captured just before and just after the
second WCD was deployed over the single bedroom window that was manually ventilated.

Figure 3.6-15 shows the images at 1440 s (24 min) after ignition. A WCD had been deployed over the
bedroom single window approximately 120 s prior, the 5th floor stair A door had been closed 57 s prior,
and the bulkhead door had been open approximately 35 s prior. Smoke density in corridor, bedroom,
and living room views was such that visibility was zero. The stairwell was clear of smoke, and very
little smoke was venting from the open bulkhead door with the PPV fan operating.

Figure 3.6-16 shows the images at 1560 s (26 min) after ignition. At this time visibility was obscured in
the living room and corridor. Ignition of all fuel surfaces in the bedroom is visible in the bedroom view.
Despite the 5th floor stair A door being opened at 1518 s after ignition, smoke did not fill or descend the
stairwell with the PPV fan operating.

178
Figure 3.6-3. Experiment 5E, ignition.

179
Figure 3.6-4. Experiment 5E, 120s after ignition.

180
Figure 3.6-5. Experiment 5E, 240s after ignition.

181
Figure 3.6-6. Experiment 5E, 310s after ignition.

182
Figure 3.6-7. Experiment 5E, 395s after ignition.

183
Figure 3.6-8. Experiment 5E, 480s after ignition.

184
Figure 3.6-9. Experiment 5E, 720s after ignition.

185
Figure 3.6-10. Experiment 5E, 960s after ignition.

186
Figure 3.6-11. Experiment 5E, 980s after ignition.

187
Figure 3.6-12. Experiment 5E, 1200s after ignition.

188
Figure 3.6-13. Experiment 5E, 1305s after ignition.

189
Figure 3.6-14. Experiment 5E, 1320s after ignition.

190
Figure 3.6-15. Experiment 5E, 1440s after ignition.

191
Figure 3.6-16. Experiment 5E, 1560s after ignition.

192
3.6.2 Temperatures 

Figure 3.6-17 through Figure 3.6-19 provide temperature data for five measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.6-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.6-17 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 5A in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
temperature in the bedroom increased to a maximum of approximately 600 ºC (1112 ºF) before the fire
became ventilation limited and the temperature declined. The temperature declined to 200 ºC (392 ºF)
before responding to the added oxygen from the open stairwell and bulkhead doors. With the added
oxygen and burning the temperature increased back above 600 ºC (1112 ºF). Bedroom temperatures
remained around 500 ºC (1292 ºF) and increased to a peak of 800 ºC (1472 ºF) after the single bedroom
window was manually ventilated and the PPV fan was activated. The temperature changes were all
gradual and conditions remained ventilation limited for most of the experiment.

During the entire experiment the temperatures in the corridor, right outside the fire apartment, at 2.1 m
(7 ft) above the floor do not exceed 200 ºC (392 ºF). The temperature at this position did not change
significantly in response to the changes in ventilation or bedroom fire involvement and remained steady
for the duration of the experiment. The two other corridor temperature measurement locations at the
opposite end of the corridor remained near 100 ºC (212 ºF) for the duration of the experiment. They also
remained very steady and did not change significantly in response to the changing events. The 2.1 m
(7 ft) temperature in the stairwell remained ambient when the door to the fire floor was closed and
increased as high as 80 ºC (176 ºF) when the door was open.

Figure 3.6-18 shows the temperatures at 1.2 m (4 ft) which would be more representative of where
firefighters would be operating or occupants would be evacuating. The temperature at this elevation in
the bedroom behaved very similarly to the 2.1 m (7 ft) temperature. This was consistent with a well
mixed ventilation limited fire environment. These temperatures were within 100 ºC (212 ºF) for most of
the experiment until about 1200 s when there was enough oxygen from the broken window to create
flashover conditions in the bedroom. From that time on the bedroom temperatures were uniform
throughout. All of the other 1.2 m (4 ft) temperatures on the fire floor remained at or below 100 ºC
(212 ºF) for the duration of the experiment.

Figure 3.6-19 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4 and at the landing
above the 7th floor near the bulkhead door with access to the roof. The temperature above the fire floor
at the top of the stairwell was the only measurement above the fire floor and the temperature increase
peaked at approximately 80 ºC (176 ºF) when the fire was at its peak. The PPV fan did not completely
keep all of the smoke and heat out of the stairwell when the fire was at its peak.

193
 BR BR
:p S5 :p Ex
lac :o lac pe
i ng B R pe B ing Rem r im
WC : do ni n R:
BR S B H
L B S g S s i B
WC ov B H B S ent
: Ig D (d uble 5: op : op Y: op 1: op /clos 5: op ngle H: c S1: f D (s ing a S5: c : op H: c 5: op com
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le ing ing ing ing ing 3x ing ing ed on le t
) ) ow ed ing ed ing ed
1000 1832

800 1472

Temperature (F)
Temperature (C)

600 1112

TC BR5EE 2.1m
TC Corr5EE 2.1m
400 TC Corr5AE 2.1m 752
TC CorrStair5E 2.1m
TC Stair5E 2.1m

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.6-17. 5E Temperature vs. time at 2.1 m above the floor

BR BR
:p S5 :p Ex
lac :o lac pe
i ng B R pe BR ing Rem r im
WC : do ni n
BR u S5 B H
L : s
B Y S 1 g/ c S 5 i n B
WC o v i S B H B S ent
D
: Ig (d ble : op : op : op : op los : op gle H: c 1: f D (s g a 5: c : op H: c 5: op com
S n
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le ing ing ing ing ing 3x ing ing ed on le t
) ) ow ed ing ed ing ed
1200 2192
TC BR5EE 1.2m
TC Corr5EE 1.2m
TC Corr5AE 1.2m
1000 TC CorrStair5E 1.2m
1832
TC Stair5E 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.6-18. 5E Temperature vs. time at 1.2 m above the floor

194
 BR BR
:p S5 :p Ex
lac :o lac pe
i ng B R pe B ing Rem r im
WC : do ni n R:
BR S B H
L B S g S s i B
WC ov B H B S ent
: Ig D (d uble 5: op : op Y: op 1: op /clos 5: op ngle H: c S1: f D (s ing a S5: c : op H: c 5: op com
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le ing ing ing ing ing 3x ing ing ed on le t
) ) ow ed ing ed ing ed
100 212
TC Bulkhead MidE
TC Stair 4E
TC Stair 2E
TC Stair1 MidE
80 176

Temperature (F)
Temperature (C)

60 140

40 104

20 68

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.6-19. 5E stairwell temperature vs. time

3.6.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.6-20. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
3 Pa and only fluctuated slightly, except for the 7th floor pressure, which fluctuated around 6 Pa. This
was probably due to the heat damage to the floor and the door from the experiments on the 7th floor.

Opening the bulkhead door while the first floor stairwell door remained closed lowered the pressures in
the stairwell below ambient conditions. Opening the 1st floor stair door returned pressures to near
ambient. The WCD was in place prior to window failure so no impact of the simulated wind was seen
in the stairwell pressures. The only significant pressure increase came from the PPV fan at the base of
the stairwell being activated at 1115 s. Pressures in the area of the fire floor only increased to an
average of approximately 6 Pa with the bulkhead closed but decreased to 0 Pa with the bulkhead door
open.

195
 BR BR
:p S5 :p Ex
lac :o lac pe
i ng B R pe B ing Rem r im
WC : do ni n R:
BR S B H
L B S g S s i B
WC ov B H B S ent
: Ig D (d uble 5: op : op Y: op 1: op /clos 5: op ngle H: c S1: f D (s ing a S5: c : op H: c 5: op com
nit oub fail en en en e n i ng e fa lo a in i r lo e lo e pl
ion le ing ing ing
) ing ing 3x ning iling sed n on gle) flow sed ning sed ning eted

25
DP Stair 7E
DP Stair 6E
20 DP Stair 5E
DP Stair 3E
DP Stair 2E
15 DP Stair 1E
Pressure (Pa)

10

-5

-10
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.6-20. 5E stairwell pressures vs. time

3.6.4 Velocities 

Average doorway velocities from 4 doors; the fire apartment door, the door from the fire floor to the
stairwell, the door at the base of the stair and the door at the bulkhead, were recorded versus time and
shown in Figure 3.6-21. The three bi-direction probes in each door were averaged to provide the bulk
flow through the doors. A time history for each probe at each measurement location is located in
Appendix D – Detailed Graphs.

Due to the WCD being deployed prior to window failure all measured velocities on the fire floor were
less than 1 m/s (2 mph). Flow velocity out of the bulkhead door was also less than 1 m/s (2 mph). The
bulk flow into the 1st floor stair door was approximately 2 m/s (4 mph) with the PPV fan operating.

196
 BR BR
:p S5 :p Ex
lac :o lac pe
i ng B R pe B ing Rem r im
WC : do ni n R:
BR S B H
L B S g S s i B
WC ov B H B S ent
: Ig D (d uble 5: op : op Y: op 1: op /clos 5: op ngle H: c S1: f D (s ing a S5: c : op H: c 5: op com
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le ing ing ing ing ing 3x ing ing ed on le t
) ) ow ed ing ed ing ed
4 9
Exp 5E Bulkhead
Exp 5E Stair1
Exp 5E Stair5
3 Exp 5E Apt5E 7

2 4

Velocity (mph)
Velocity (m/s)

1 2

0 0

-1 -2

-2 -4
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 3.6-21. 5E Velocities vs. time

3.6.5 Discussion 

Experiment 5E was the first experiment on the 5th floor. A WCD was deployed from the 6th floor over
the bedroom double window where the simulated wind was positioned. This simulates a firefighting
crew that goes to the floor above the fire and finds a wind condition subjected to that area/side of the
building and the fire apartment below. As a preventative measure the crew deploys the WCD over the
fire apartment window.

The window under the WCD failed at 233 s after ignition. Due to the WCD blocking the wind effects a
wind driven condition was not created. The single window on the other side of the bedroom from where
ignition occurred did not self-vent even after 990 s into the experiment and the fire had reached a
ventilation limited, steady state condition. Temperatures at the 1.2 m (4 ft) from the floor elevation were
steady around 450 ºC (842 ºF) for over 200 s. In order to increase the fire size and expose the WCD to
higher temperatures the top pane of the single window in the bedroom was manually vented. This event,
along with providing air from the stairwell, allowed the room to flashover and continue to burn at high
temperatures for the remainder of the experiment.

Until the window in the bedroom was ventilated the only source of oxygen to the fire was from the
stairwell which was remote from the bedroom. Opening the fire floor stairwell door did not provide
much oxygen to the fire because the stairwell filled up with smoke quickly and mixed with any air being
pulled up the stairwell. When the bulkhead door was opened this changed and the fire was able to pull
fresh air as the smoke moved up and out of the stairwell. This allowed the fire in the bedroom to
intensify, but because of how remote the bedroom was from the source of oxygen, the concrete walls
and ceiling, and the distance to fuel through the hallway to the living room the fire did not spread.
When more oxygen was provided by ventilating the top pane of the single window the fire intensified
further but there was still not enough oxygen, and by that time the fuel had burned down quite a bit.

197
Temperatures in the corridor were not greatly impacted by ventilation and conditions were conducive to
a direct frontal attack for the duration of the experiment.

Another objective of this experiment was to subject the WCD to high temperature conditions for a long
period of time to see how well they would hold up. If the WCD fails during fire attack the crews being
shielded from the wind driven conditions could be caught in the flow path which is not desired so the
WCD must be able to withstand these conditions. While the WCD changed color from red to grey but
did not fail during this experiment. Using the 1.2 m (4 ft) temperature in the center of the room to
estimate the conditions the blanket was exposed to indicate that the WCD was subjected to temperatures
in excess of 400 ºC (752 ºF) for more than 1100 s, including temperatures in excess of 700 ºC (1292 ºF)
for at least 200 s. Flames were in contact with the WCD and there was no failure and the possible
effects of a wind driven fire were never present during the experiment.

The PPV fan on the 1st floor was activated at 1115 s. This increased the pressures on the 1st and 2nd
floors above 15 Pa as would be expected as compared to previous experiments. The pressures above
this height in the stairwell were much lower than expected which suggests that stairwell doors were left
open or were blown open during the experiment. This reduced the effectiveness of the fan but it still
improved conditions in the area of the fire floor as seen when the fan was turned off near the end of the
experiment and smoke flow into the stairwell increased. This highlights the importance of door control
when utilizing PPV.

198
3.7 Experiment 5E2 

The seventh experiment in the structure was located in the same apartment as the sixth experiment, 5E.
Experiment 5E2 was ignited in the living room on the top of sofa that backed to the kitchen (Figure
3.7-1). The fire in the bedroom from the sixth experiment did not extend beyond the room of origin so
the apartment was cleared of smoke and the experiment 5E2 was ignited. Both windows in the back
bedroom failed in the previous experiment so they remained open during this experiment. This
experiment also utilized a simulated wind provided by the MVU. The simulated wind was subjected to
the living room window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The
events that took place during the experiment to accomplish the experimental objectives below are shown
in Table 3.7-1.

Experimental Objectives
• Create wind driven fire conditions from a living room fire
• Examine building flows with different door opening configurations
• Utilize portable PPV fans to control smoke movement
• Control wind driven conditions with FBN

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.7-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

199
Figure 3.7-1. 5th floor apartment E floor plan with locations pertinent to Experiment 5E2

200
Table 3.7-1: Experiment 5E2 Timeline
Start (s) End (s) Event Location Event Description
0 0 Living room Ignition (front doors + stair doors 1 and 5 opened)
271 273 Bulkhead door Opened
594 596 Living room window Center pane failed
628 628 Living room window Top right pane failed
681 681 Living room window Top left pane failed
698 698 Living room window Bottom left pane failed
716 719 Bulkhead door Closed
725 725 Living room Bottom right pane failed
809 810 Front lobby 27 inch fan activated
837 841 Front lobby Left door closed
851 860 Front lobby 27 inch fan moved, toward door
880 883 Front lobby 27 inch fan moved, into door
903 909 Living room window FBN 1 started flowing (Fog)
915 920 Front lobby Fan moved closer to door
973 976 Living room window FBN 1 stopped
979 981 Front lobby Left door opened
1020 1020 Experiment completed

201
Figure 3.7-2. Instrumentation and camera locations with labels.

202
3.7.1 Observations 

The observations are presented as a series of images captured from eight camera locations, seven were
video cameras and one was a thermal imaging camera. The camera positions inside the building are
shown in Figure 3.7-2.

Figure 3.7-3 through Figure 3.7-12 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 16 min (960 s) after ignition. Images were captured every 4 min
and at times when events had taken place that demonstrate the experimental objectives. Each image
view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom which
was burned in the 5E experiment, the bedroom furthest from the apartment entrance; the failed single
window is straight across the view, and the failed double window is to the right of the view. The top
right view (LIVINGRM) was of the living room looking across to the windows. The corridor view
(CORRIDOR) shows the corridor leading up to the open apartment doorway. Next to that, the outside
lobby view (LOBBY) shows the area just outside the lobby doors. The next two views (STAIR and
STAIR IR) show the stairwell with a visual and a thermal imaging camera from the floor below the fire
looking up to the fire floor. The bottom left view (ROOF) shows the bulkhead door at the top of the
stairwell and the bottom right view (OUTSIDE) shows an external view of the fire apartment with a
focus on the living room window.

Figure 3.7-3 shows the images at the time of ignition. At this point, the seven video views were clear
and unobstructed. The effects of the previous experiment (experiment 5E) are visible in the corridor and
bedroom views and some smoke damage from that experiment is still present in the living room. The
door to the fire floor is open for the start of the experiment as seen in the stairwell views and the doors to
the lobby and to the base of the stairs are open as shown in the lobby view.

Figure 3.7-4 shows the images at 240 s (4 min) after ignition. At this point, the fire that was ignited on
the left sofa was growing on the seat of the sofa. Smoke density has increased in the living room,
bedroom, and corridor views. Smoke is visibly flowing up the stairwell.

Figure 3.7-5 shows the images at 480 s (8 min) after ignition. The bulkhead door had been opened 208 s
prior. The sofa is still not fully involved, but the smoke density throughout the apartment increased,
limiting visibility in the corridor, bedroom, and living room views. The heat and volume of the smoke
flow up the stairwell has increased. The fire has become ventilation limited.

Figure 3.7-6 shows the images at 660 s (11 min) after ignition. The living room window began to fail
65 s prior to these images. The added oxygen and impact of the wind has increased the fire growth and
spread significantly. The living room was becoming fully involved in flames and flames were being
forced through the apartment hallway and into the rear bedroom from the previous experiment.
Visibility in the stairwell has been reduced and the amount of heat being forced into the stairwell has
greatly increased in the thermal imaging view. The amount of smoke flowing out of the open bulkhead
door has increased.

203
Figure 3.7-7 shows the images at 690 s (11 min and 30 s) after ignition. The living room and bedroom
have become completely involved in flame and the wind driven condition has developed. Visibility has
decreased to zero in the stairwell and the heat being forced up and down the stairwell has increased as
indicated by the thermal imaging camera switching to its high heat mode.

Figure 3.7-8 shows the images at 720 s (12 min) after ignition. The living room window was failing
incrementally; for the past 2 minutes the majority of the window opening was clear. At this point,
flames were visible in the stair camera view; the stair IR view was saturated, but reflected a surface
temperature in the stairwell of 260 ºC (500 ºF).

Figure 3.7-9 shows the images at 750 s (12 min and 30 s) after ignition. The bulkhead door had been
closed for 30 s and the lobby image shows smoke being forced out of the front lobby doors. This shows
the change in flow path and the ability to force smoke and heat down 5 stories in as little as 30 s. Both
the stairwell video camera and thermal imaging camera failed due to heat exposure. Flames are no
longer visible in the bedroom but are visible in the corridor. The flames are also pulsing out of the
living room against the wind.

Figure 3.7-10 shows the images at 890 s (14 min and 50 s) after ignition. This was 13 s before the FBN
was flowing. The wind driven condition still existed but few camera views were still usable. The living
room camera still showed full room involvement.

Figure 3.7-11 shows the images at 915 s (15 min and 15 s) after ignition. This was approximately 10 s
after the FBN 1 with a fog pattern was flowing into the living room. The fire was beginning to be
suppressed in the living room and the flow from the nozzle can be seen in the outside view.

Figure 3.7-12 shows the images at 960 s (16 min) after ignition. At this point FBN 1 had been flowing
into the living room window for almost 60 s. The fire appears to be limited to some of the furnishings.
The burning, hot gas layer had been extinguished. Most of the other cameras were either non-
operational or obstructed by soot deposition.

204
Figure 3.7-3. Experiment 5E2, ignition.

205
Figure 3.7-4. Experiment 5E2, 240s after ignition.

206
Figure 3.7-5. Experiment 5E2, 480s after ignition.

207
Figure 3.7-6. Experiment 5E2, 660s after ignition.

208
Figure 3.7-7. Experiment 5E2, 690s after ignition.

209
Figure 3.7-8. Experiment 5E2, 720s after ignition.

210
Figure 3.7-9. Experiment 5E2, 750s after ignition.

211
Figure 3.7-10. Experiment 5E2, 890s after ignition.

212
Figure 3.7-11. Experiment 5E2, 915s after ignition.

213
Figure 3.7-12. Experiment 5E2, 960s after ignition.

214
3.7.2 Temperatures 

Figure 3.7-13 through Figure 3.7-15 provide temperature data for four measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.7-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.7-13 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 4 measurement
locations, just outside the fire apartment in the corridor, just outside Apartment 5A in the corridor, just
outside the A stairwell doorway, and just inside the A stairwell on the landing. The bedroom
measurement location was compromised during the previous experiment and is not included in this
analysis. There were no temperature measurements in the living room, so the closest measurement to
the fire was the corridor just outside the fire apartment. The fire reached a steady state with the corridor
temperature at approximately 150 ºC (302 ºF) as the smoke layer descended to the height of the seat of
the sofa. The bulkhead door was opened to provide oxygen and the temperature steadily increased to
approximately 300 ºC (572 ºF), while the fire on the sofa increased. The increased fire size caused the
window to begin to fail and the conditions changed rapidly. The temperature at 2.1 m (7 ft) cooled
briefly as the cold air was forced into the living room but once conditions changed the temperature
increased from less than 300 ºC (572 ºF) to more than 800 ºC (1472 ºF) in seconds. The bulkhead door
was closed to change the flow path and the temperature steadily decreased from 900 ºC (1652 ºF) to
450 ºC (842 ºF) over the next 200 s. At 903 s the FBN was activated and the temperature dropped to
approximately 150 ºC (302 ºF) in the 70 s the nozzle was flowing.

The two other corridor measurement locations responded similarly to the events as the one closest to the
fire apartment, except their peak temperatures were lower. The measurement location in the corridor
next to the open stairwell door reached 800 ºC (1472 ºF) and the measurement location next to
Apartment A reached 600 ºC (1112 ºF). This temperature difference also shows the difference in
temperature between being in the flow path and being just outside the flow path. The temperature of the
corridor after the nozzle had been flowed for 70 s was conducive to a direct frontal fire attack.

The temperature in the stairwell also reached 600 ºC (1112 ºF) at the peak of the wind driven condition.
Once the bulkhead door was closed the temperature steadily decreased to approximately 300 ºC (572 ºF)
before the FBN was flowing. After the FBN was flowing the temperature decreased to 150 ºC (302 ºF).

Figure 3.7-14 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. The temperatures at all of the
measurement locations at the 1.2 m (4 ft) elevation remained below 50 ºC (122 ºF) as the fire was
developing in the living room. At approximately 550 s the living room was transitioning to flashover
and the temperatures increased to as high as 250 ºC (482 ºF) in the corridor as the window failed. The
temperatures increased as the fresh air was mixing in the living room. At approximately 675 s the flow
path from the living room to the stairwell was well mixed and the temperatures were the same as the
2.1 m (7 ft) elevation showing that there was no longer a cool layer and the temperatures were
increasing from floor to ceiling. After the FBN was flowed the temperatures decreased to approximately
100 ºC (212 ºF) which would allow fire fighters to advance to the fire apartment.

215
Figure 3.7-15 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4 and at the
bulkhead. The temperature at the top of the stairwell started the experiment elevated from the hot gases
trapped in the top of the stairwell from the previous experiment. During the wind driven condition with
the bulkhead door open temperatures above the fire floor reached as high as 1200 ºC (2192 ºF) and
decreased quickly when the bulkhead door was closed because this area was no longer part of the flow
path. During the wind driven condition temperatures below the fire floor on the 4th floor reached 400 ºC
(752 ºF) and the 2nd floor temperature was above 150 ºC (302 ºF). After suppression was started with
the FBN the temperature below the fire floor decreased to below 100 ºC (212 ºF).

Ex
p
LR
:w LR L L L L L LR LB er im
:w B
BY Y Y : B R B Y : F Y e
LR BH ind L : : F : B : l nt
: Ig :o ow BH: indo BY: : left fan f an BN fan N s eft o com
nit pe
nin fai clo w c fan clo mo mo flow mo top pen ple
io lin se lea o se ve ve in ve pe in te
n g g d r n d d d g d d g d

1200 2192
TC Corr5EE 2.1m
TC Corr5AE 2.1m
TC CorrStair5E 2.1m
1000 TC Stair5E 2.1m 1832

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 3.7-13. 5E2 Temperature vs. time at 2.1 m above the floor

Ex
LR pe
:w LR L
LB B B R L L L L R L B r im
BH ind :w
L Y Y: Y: : F BY: : FB Y: l ent
LR ow BH: indo BY: : left fan f an BN fan N s eft o com
: Ig :o
nit pe
nin fai clo w c fan clo mo mo flow mo top pen ple
io lin se lea o se ve ve in ve pe in te
n g g d r n d d d g d d g d

1200 2192
TC Corr5EE 1.2m
TC Corr5AE 1.2m
TC CorrStair5E 1.2m
1000 TC Stair5E 1.2m 1832
Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)

216
Figure 3.7-14. 5E2 Temperature vs. time at 1.2 m above the floor

Ex
p
LR
:w LR L L L L L LR LB er im
BH ind :w BY BY BY R: BY : F Y: en
LR i L : : : F : B l t
: Ig :o ow BH: ndo BY: left fan f an BN fan N s eft o com
nit pe
nin fai clo w c fan clo mo mo flow mo top pen ple
io lin se lea o se ve ve in ve pe in te
n g g d r n d d d g d d g d

1500 2732
TC Bulkhead MidE
TC Stair 4E
TC Stair 2E
TC Stair1 MidE
1200 2192

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 3.7-15. 5E2 stairwell temperature vs. time

3.7.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.7-16. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
5 Pa and only fluctuated slightly, except for the 7th floor pressure which fluctuated around 8 Pa. This
was most likely due to the heat accumulation in the top of the stairwell in the area of the 7th floor.

The pressures all remained constant until the windows in the living room began to fail. After window
failure the pressures in the stairwell in the area of the fire floor increased to approximately 5 Pa. This
minimal impact was due to the top and bottom of the stairwell being open. When the bulkhead door was
closed the pressure increased quickly. The 7th floor increased from less than 10 Pa to more than 80 Pa in
seconds. The other floors from the 6th to the 2nd increased to approximately 70 Pa, 65 Pa, 50 Pa, 35 Pa
and 20 Pa respectively.

Adding the PPV fan at the front lobby door at 810 s increased the lower floor pressures. As it was
moved closer to the stairwell when the smoke flow out of the front of the building decreased the
pressures increased. Flowing the FBN and stopping the wind driven fire condition caused the pressures
in the stairwell to decrease with the decreased energy production from the fire.

217
 Ex
p
LR
:w LR L L L L L LR LB er im
BH ind :w BY BY BY R: BY : F Y: en
LR i L : : : F : B l t
: Ig :o ow BH: ndo BY: left f an fan BN fan N s eft o com
nit pe fai clo w c fan clo mo mo flow mo top pe pl
io nin lin se lea o se ve ve in ve pe nin ete
n g g d r n d d d g d d g d

100
DP Stair 7E
DP Stair 6E
DP Stair 5E
80 DP Stair 4E
DP Stair 3E
DP Stair 2E
60 DP Stair 1E
Pressure (Pa)

40

20

-20
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 3.7-16. 5E2 stairwell pressures vs. time

3.7.4 Velocities 

Average doorway velocities from 4 doors; the fire apartment door, the door from the fire floor to the
stairwell, the door at the base of the stair and the door at the bulkhead, were recorded versus time and
shown in Figure 3.7-17. The three bi-direction probes in each door were averaged to provide the bulk
flow through the doors. A time history for each probe at each measurement location is located in
Appendix D – Detailed Graphs.

The fire floor stairwell door remained open for the duration of the experiment enabling a flow path to
the stairwell. The velocity at the apartment doorway increased in response to the growth of the living
room fire and increased further once the window failed. The velocity increased to a peak of
approximately 9 m/s (20 mph) before the bulkhead door was closed. The flow from the fire apartment
had to turn 180 degrees to enter the stairwell which most likely accounted for the difference in velocity
between the fire apartment and the stairwell. The flow through the bulkhead door was very similar to
the flow leaving the fire apartment as would be expected but the velocity measurement beyond 650 s
was not reliable because of the extreme heat the pressure transducers were subjected too.

218
 Ex
p
LR
:w LR L L L L L LR LB er im
BH ind :w BY BY BY R: BY : F Y: en
LR i L : : : F : B l t
: Ig :o ow BH: ndo BY: left fan f an BN fan N s eft o com
nit pe
nin fai clo w c fan clo mo mo flow mo top pen ple
io lin se lea o se ve ve in ve pe in te
n g g d r n d d d g d d g d

15 34
Exp 5E2 Bulkhead
Exp 5E2 Stair1
Exp 5E2 Stair5
10 Exp 5E2 Apt5E 22

5 11

Velocity (mph)
Velocity (m/s)

0 0

-5 -11

-10 -22

-15 -34
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 3.7-17. 5E2 Velocities vs. time

3.7.5 Discussion 

This experiment utilized the simulated wind directed into the living room and a wind driven condition
was created which caused extremely hazardous conditions inside the building. Temperatures exceeded
600 ºC (1112 ºF) throughout the fire floor at any elevation above the floor. These same temperatures
entered the stairwell extended to the top of the stairwell and created poor conditions down the stairwell
and into the lobby. Large amounts of smoke were forced out of the front lobby door displaying
untenable conditions throughout the entire stairwell. The condition created during this experiment were
not conducive to a direct frontal fire attack and firefighters remote from the fire in the stairwell would
have been threatened if they were in the flow path due to extreme temperatures.

The fire became ventilation limited and was beginning to burn itself out, even with the windows open in
the bedroom from the previous experiment. The bulkhead door was opened and that allowed the smoke
layer to move up the stairwell and create a condition for the fire to grow. The fire transitioned to
flashover and failed the window which triggered the wind driven condition. The hot gases were
primarily being forced to the stairwell and up the stairs to the open bulkhead door. Once the bulkhead
door was closed the conditions changed. The flow was forced to travel down the stairs to the front door
but did so slowly. Without the smoke moving upward the wind driven fire in the corridor became
ventilation limited and the temperatures in the corridor and stairwell decreased by nearly half. This
scenario also had the fire floor stairwell door open the entire experiment. This door is a key door to
limit the impact to the rest of the structure. With this door open other floors were exposed to untenable
conditions for firefighters and occupants.

A FBN 1 with a fog nozzle, narrow pattern, flowing approximately 10.1 L/s (160 gpm) of water into the
living room window was put into operation at 903 s and was allowed to flow for about 70s.
Temperatures in the corridor and stairwell decreased to below 200 ºC (392 ºF) very quickly and

219
continued to decrease until the experiment was ended, even with the simulated wind still flowing
through the apartment. The water introduced to the seat of the fire changed an environment with
burning gases to just burning objects. The fire was not able to be extinguished completely with the
FBN, but it reduced its energy output significantly.

220
3.8 Experiment 5K 

The eighth experiment was located on the 5th floor in apartment 5K. Experiment 5K was ignited in the
bedroom furthest from the open apartment door (Figure 3.8-1). This experiment also utilized a
simulated wind provided by the MVU. The simulated wind was subjected to the bedroom double
window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The events that took
place during the experiment to accomplish the experimental objectives bulleted below are shown in
Table 3.8-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Examine impact of the partial deployment of a WCD on fire room window
• Control wind driven conditions with FBN in conjunction with WCD
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.8-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

221
Figure 3.8-1. 5th floor apartment K floor plan with locations pertinent to Experiment 5K

222
Table 3.8-1: Experiment 5K Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
303 304 Bedroom double Bottom left pane failed
452 468 5th floor stair A Door opened
487 494 Front lobby Doors opened
513 518 1st floor stair A Door opened
693 695 Bedroom double Bottom right pane failed
696 700 Bulkhead door Opened
798 798 Bedroom double Top left pane failed
830 831 Bedroom double Top right pane failed
953 975 Bedroom double WCD deploying (1/2 of window)
1012 1012 Living room window Starting to fail
1043 1045 Bedroom double FBN 1 started flowing (Fog)
1124 1124 Bedroom double FBN 1 stopped
1180 1184 Bulkhead door Closed
1219 1219 Living room window FBN 1 started flowing (Fog)
1295 1295 Living room window FBN 1 stopped
1456 1464 1st floor stair A 27 inch fan activated
1501 1503 Bulkhead door Opened
1590 1592 Bulkhead door Closed
1873 1885 Bedroom double Air flow removed
1920 1920 Experiment completed

223
Figure 3.8-2. Instrumentation and camera locations with labels.

224
3.8.1 Observations 

The observations are presented as a series of images captured from eight camera locations, all eight were
video cameras. The camera positions inside the building are shown in Figure 3.8-2. Another outside
view was added for this experiment in place of the thermal imaging view.

Figure 3.8-3 through Figure 3.8-16 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 32 min (1920 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 5K, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor view
(CORRIDOR) shows the corridor leading up to the open apartment doorway. Next to that, the outside
lobby view (LOBBY) shows the area just outside the lobby doors. The view (STAIR) shows the
stairwell with a camera from the floor below the fire looking up to the fire floor. Next to that is the
tower view (TOWER) which shows an external view of the fire apartment. The bottom left view
(ROOF) shows the bulkhead door at the top of the stairwell and the bottom right view (OUTSIDE)
shows an alternate view of the outside of the apartment from a window in another wing of the building.

Figure 3.8-3 shows the images at the time of ignition. At this point, the eight video views were clear
and unobstructed. The bedroom view shows the ignition trashcan near the double window and a sheet of
gypsum board covering the single window. The single window was covered to prevent any loss of flow
from the simulated wind.

Figure 3.8-4 shows the images at 240 s (4 min) after ignition. The fire had extended to the bed from the
trash container where ignition took place. The smoke layer had descended almost to the floor in the
ignition bedroom, but was approximately 1.5m (5 ft) from the floor in the living room, and was just
visible at the ceiling level in the public corridor. Flames are visible from the exterior tower view but the
window was still intact.

Figure 3.8-5 shows the images at 480 s (8 min) after ignition. One pane from the bedroom double
window failed 176 s prior and the 5th floor stair A door was opened 12 s prior. The smoke layer had
descended to the floor in the corridor, living room, and bedroom views; visibility in these views is near
zero beyond the flames in the bedroom view. Smoke had begun to push down the stairwell. There was
no visible smoke leakage from the bulkhead door at this time.

Figure 3.8-6 shows the images at 720 s (12 min) after ignition. The bulkhead door had been opened 20 s
prior and smoke was venting from it. There was very little visibility in the stairwell view. The living
room view shows flames beginning to enter the living room.

Figure 3.8-7 shows the images at 830 s (13 min and 50 s) after ignition. The bedroom was completely
full of flames, there were flames involving the living room and moving toward the corridor. There was
minimal visibility in the corridor and no visibility in the stairwell. Smoke was flowing out of the
bulkhead doorway indicating that the stairwell was full of smoke to the roof.

225
Figure 3.8-8 shows the images at 960 s (16 min) after ignition. The bedroom double window had
completely failed by this point. Flames could be seen in the stairwell view coming from the public
corridor. The volume of smoke venting from the bulkhead had reduced the visibility in the roof view to
zero. The WCD was being moved into place over the left half of the bedroom double window. White
smoke was pushing out of the fire apartment in the area of the living room window.

Figure 3.8-9 shows the images at 975 s (16 min and 15 s) after ignition. The WCD device remained in
place, blocking one half of the window. This was done to gain insight into the question, if the WCD
was not big enough to block an entire window is there a benefit to blocking as much of it as possible?
All other conditions in the images are similar to those from 15 s prior.

Figure 3.8-10 shows the images at 1040 s (17 min and 20 s) after ignition. The living room window has
failed approximately 20 s prior and there were flames throughout the bedroom, living room, and
corridor, reaching the stairwell. These were the conditions 5 s before the activation of the FBN.

Figure 3.8-11 shows the images at 1080 s (18 min) after ignition. The FBN 1 with a fog nozzle has been
flowing for 35 s into the bedroom window. There was still some burning in the bedroom and there does
not appear to be much impact on the amount of flames in the living room. The visibility in the stairwell
and room views was still zero.

Figure 3.8-12 shows the images at 1200 s (20 min) after ignition. At this point the corridor view had
been damaged and no longer provided video. The bedroom view was similarly lost by this time. FBN 1
began flowing into the bedroom double window 155 s prior and ceased 76 s prior. The bulkhead had
been closed 16 s prior. Combustion in the living room was visible from that view. Flames were still
extending from the failed living room window. This view was captured 20 s prior to FBN 1 flow into
the living room.

Figure 3.8-13 shows the images at 1260 s (21 min) after ignition. The FBN was relocated to the living
room window and flowed into the living room 41 s prior to this set of images. The water was still
flowing and the visible flame in the living room was suppressed.

Figure 3.8-14, Figure 3.8-15, and Figure 3.8-16 show the images at 1440 s (24 min), 1680 s (28 min),
and 1920 s (32 min) after ignition, respectively. By this time the internal views on the fire floor were
limited, and the external views indicate that a majority of the fire had been suppressed. The FBN was
stopped at 1295 s (21.6 min). The use of the PPV fan rekindled a fire in part of the living room and
cleared the stairwell of smoke.

226
Figure 3.8-3. Experiment 5K, ignition.

227
Figure 3.8-4. Experiment 5K, 240s after ignition.

228
Figure 3.8-5. Experiment 5K, 480s after ignition.

229
Figure 3.8-6. Experiment 5K, 720s after ignition.

230
Figure 3.8-7. Experiment 5K, 830s after ignition.

231
Figure 3.8-8. Experiment 5K, 960s after ignition.

232
Figure 3.8-9. Experiment 5K, 975s after ignition.

233
Figure 3.8-10. Experiment 5K, 1040s after ignition.

234
Figure 3.8-11. Experiment 5K, 1080s after ignition.

235
Figure 3.8-12. Experiment 5K, 1200s after ignition.

236
Figure 3.8-13. Experiment 5K, 1260s after ignition.

237
Figure 3.8-14. Experiment 5K, 1440s after ignition.

238
Figure 3.8-15. Experiment 5K, 1680s after ignition.

239
Figure 3.8-16. Experiment 5K, 1920s after ignition.

240
3.8.2 Temperatures 

Figure 3.8-17 through Figure 3.8-19 provide temperature data for five measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.8-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.8-17 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 5G in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
bedroom temperature increased to approximately 500 ºC (932 ºF) by 315 s prior to the bottom left
window pane failing. The wind introduced into the bedroom cooled the room. Doors were opened to
provide more ventilation to the fire. Ultimately the bulkhead door needed to be opened to allow the fire
to grow, fail more window panes, and develop a wind driven fire. At approximately 700 s the
temperature increased from 350 ºC (662 ºF) to 700 ºC (1292 ºF) and ultimately to 900 ºC (1652 ºF) after
the bedroom double window completely failed. The WCD was deployed at 975 s over half of the failed
window and the bedroom temperature slightly increased. At 1048 s the FBN was directed into the
bedroom and the bedroom temperature decreased from approximately 900 ºC (1652 ºF) to 100 ºC
(212 ºF) very quickly and remained there for the remainder of the experiment.

The corridor temperatures remained below 250 ºC (482 ºF) until 825 s when the wind driven condition
was developed triggered by full bedroom involvement in flames extending into the living room. At this
time the temperature in the corridor right outside the fire apartment increased from 250 ºC (482 ºF) to
650 ºC (1202 ºF) in seconds. The two other corridor temperatures increased from approximately 200 ºC
(392 ºF) to above 400 ºC (752 ºF) in seconds and climbed to a steady state of approximately 600 ºC
(1112 ºF) in less than 1 min. The deployment of the WCD over half of the window had no significant
impact on the corridor temperatures. During the flowing of the FBN into the bedroom the corridor
temperatures decreased from above 600 ºC (1112 ºF) to approximately 300 ºC (572 ºF). The
temperature just outside the fire apartment recovered to above 400 ºC (752 ºF) because of the continued
burning in the living room but decreased once the FBN was directed into the living room suppressing
the remaining fire.

The stairwell 2.1 m (7 ft) temperature remained below 200 ºC (392 ºF) until the wind driven condition
was developed. The temperature spiked quickly from 200 ºC (392 ºF) to 350 ºC (662 ºF) and steadily
increased to over 600 ºC (1112 ºF). The WCD deployment had little effect on the temperature and the
FBN activation decreased the stairwell temperature from 550 ºC (1022 ºF) to 250 ºC (482 ºF) while the
water was flowing. The wind driven condition was eliminated once the bedroom was suppressed and
the temperature decreased further once the living room was extinguished.

Figure 3.8-18 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. The temperature at this elevation in
the bedroom behaved very similarly to the 2.1 m (7 ft) temperature. This was consistent with a well
mixed ventilation limited fire. This temperature was within 100 ºC (212 ºF) for most of the experiment
until about 825 s when there was enough oxygen from the broken window to create flashover conditions
in the bedroom. From that time on the bedroom temperatures were uniform throughout. At

241
approximately 825 s the flow path from the bedroom to the stairwell was well mixed and the
temperatures were the same as the 2.1 m (7 ft) elevation showing that there was no longer a cool layer
and the temperatures were increased from floor to ceiling. After the FBN was flowed into the bedroom
and living room the temperatures decreased to approximately 200 ºC (212 ºF) which would allow fire
fighters to advance to the fire apartment.

Figure 3.8-19 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, 6 and at the
bulkhead. Temperatures in the stairwell above the fire floor began to escalate from 50 ºC (122 ºF), once
the bulkhead door was opened and created a flow path through the bedroom and out of the top of the
stairwell, to over 550 ºC (1022 ºF). The temperature on the 4th floor reached 300 ºC (572 ºF) before the
deployment of the WCD. Blocking half of the flow through the bedroom window slowed the flow
moving down the stairwell and decreased the temperature back down to 50 ºC (122 ºF). When the FBN
was flowed into the living room and the bulkhead was closed the fire gases were pushed down the
stairwell creating a temperature spike of 150 ºC (302 ºF) for a short period of time.
BR BR BR:
:p : F FB Ex
lac
BR B N
BR ng LR: N flo sto
i LR LR: R e per im
:d L : W w w p : F m e
BR oub S5 BY S1 BH dou CD indo ing ped B FB B BH B ovi nt
: Ig le : op : op : op : op ble (d
o w f (do (do H: c N flo N sto S1: f : op H: c ng ai comp
nit fail en en en en a e
a l
ion ing ing ing ing ing clea uble iling uble uble osed wing ped n on ning osed r flow leted
p l
r ) ) )
1800 3272
TC BR5KW 2.1m
TC Corr5KW 2.1m
TC Corr5GW 2.1m
1500 TC CorrStair5W 2.1m
2732
TC Stair5W 2.1m

Temperature (F)
Temperature (C)

1200 2192

900 1652

600 1112

300 572

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 3.8-17. 5K Temperature vs. time at 2.1 m above the floor

242
 BR BR BR:
:p : F FB Ex
lac
BR B N
BR ing LR: N flo sto LR LR: R e per im
:d L : W w w p : F m e
BR oub S5 BY S1 BH dou CD indo ing ped B FB B BH B ovi nt
: Ig le : op : op : op : op ble (d
o w f (do (do H: c N flo N sto S1: f : op H: c ng ai comp
ni t f e e e e a e
ion ailing ning ning ning ning clea uble ailing uble uble losed wing pped n on ning losed r flow leted
r ) ) )
1500 2732
TC BR5KW 1.2m
TC Corr5KW 1.2m
TC Corr5GW 1.2m
1200 TC CorrStair5W 1.2m 2192
TC Stair5W 1.2m

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 3.8-18. 5K Temperature vs. time at 1.2 m above the floor
BR BR BR:
:p : F FB Ex
lac
BR B i ng L R B N f N s t LR LR: R e per im
:d R W : w l o o p m e
BR oub S5 LBY S1 BH : dou CD indo wing ped :
B FB B
F BH B ovi nt
: Ig le : op : op : op : op ble (d
o w f (do (do H: c N flo N sto S1: f : op H: c ng ai comp
nit fail en en en en a e
ion ing ing ing ing ing clea uble ailing uble uble losed wing pped n on ning losed r flow leted
r ) ) )
800 1472
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
600 1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 3.8-19. 5K stairwell temperature vs. time

3.8.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.8-20. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures fluctuated with
an average pressure of approximately 10 Pa. The natural wind, number of damaged apartments from
previous experiments and leaky stairwell doors caused the higher ambient pressures. After a window
pane failed in the bedroom and the door to the stairwell was opened the pressure in the stair increased to
40 Pa at the top of the stairwell and approximately 25 Pa in the area of the fire floor. Opening the

243
bulkhead decreased the pressure in the stairwell to below 15 Pa. The fire floor pressure remained above
25 Pa with the elevated temperatures from the wind driven fire and decreased once the WCD was
deployed. The bulkhead door was closed at 1185 s and the pressure in the top three floors of the
stairwell increased to above 30 Pa. Suppressing the living room fire decreased the pressures and adding
the PPV fan increased the pressures in the entire stairwell to between 15 Pa and 25 Pa.

BR BR BR:
:p : F FB Ex
lac
BR B
BR ng LR: N flo sto
i N
LR LR: R e per im
:d W w mo
L
B R ou b S 5 B Y S 1 B H : d C in w i p p : F F B B vin ent c
: Ig le : o : o : o : o oub D ( dow ng ( ed ( BH BN N s S1 H: o BH g a om
nit fail pen pen pen pen le cle doub fail doub doub : clos flow topp : fan pen : clos i r f pl e
low ted
ion ing ing ing ing ing ar le) ing le) le) ed ing ed on ing ed

60
DP Stair 7W
DP Stair 6W
50 DP Stair 5W
DP Stair 4W
DP Stair 3W
40 DP Stair 2W
DP Stair 1W
Pressure (Pa)

30

20

10

-10
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 3.8-20. 5K stairwell pressures vs. time

3.8.4 Velocities 

Average doorway velocities from 3 doors; the fire apartment door, the door from the fire floor to the
stairwell, and the door at the base of the stairwell, were recorded versus time and shown in Figure
3.8-21. The three bi-direction probes in each door were averaged to provide the bulk flow through the
doors. A time history for each probe at each measurement location is located in Appendix D – Detailed
Graphs. During the wind driven condition the velocity out of the fire apartment was as high as 9 m/s
(20 mph). The velocity in through the 1st floor stair door was less than 1 m/s (2 mph) for most of the
experiment and increased as the PPV fan was added later in the experiment.

244
 BR BR BR:
:p : F FB Ex
lac
BR B N
BR ing LR: N flo sto LR LR: R e per im
:d L : W w w p : F m e
BR oub S5 BY S1 BH dou CD indo ing ped B FB B BH B ovi nt
: Ig le : op : op : op : op ble (d
o w f (do (do H: c N flo N sto S1: f : op H: c ng ai comp
ni t f e e e e a e
ion ailing ning ning ning ning clea uble ailing uble uble losed wing pped n on ning losed r flow leted
r ) ) )
15 34
Exp 5K Stair5
Exp 5K Stair1
Exp 5K Apt5K
12 27

9 20

Velocity (mph)
Velocity (m/s)

6 13

3 7

0 0

-3 -7
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 3.8-21. 5K Velocities vs. time

3.8.5 Discussion 

This experiment utilized the simulated wind directed into the bedroom and a wind driven condition was
created which caused extremely hazardous conditions inside the building. Temperatures exceeded
600 ºC (1112 ºF) throughout the fire floor at any elevation above the floor. These same temperatures
entered the stairwell, extended to the top of the stairwell and created poor conditions from the top of the
stairwell and down a floor to the 4th floor. The conditions created during this experiment were not
conducive to a direct frontal fire attack and firefighters remote from the fire in the stairwell would have
been threatened if they were in the flow path due to extreme temperatures.

The first pane of the bedroom window failed at 304 s and the window completely failed at 831 s. The
temperature in the center of the bedroom at 1.2 m (4 ft) at the time of window failure was approximately
500 ºC (932 ºF). The temperature in the center of the room at the time of complete window failure was
in excess of 900 ºC (1652 ºF). This demonstrates that when a portion of the window fails, the wind
being introduced through the opening cools and slows the failure of the rest of the window. The
majority of the window did not fail until the room transitioned to flashover.

Similar to previous experiments the fire took a significant amount of time to develop even with a failed
window pane in the fire room. The air was forced into the bedroom but had nowhere to travel to with
the exception of leaks around doors. Once the stair door was opened the fire did not grow significantly.
The first floor stair door was opened but even with a downward flow the smoke layer remains down to
the floor level. Once the bulkhead door was opened the flow changes to up the stairwell and the fire was
able to grow quickly. The open bulkhead door contributed to the development of the wind driven
condition. A flow path is required for a wind driven fire condition to occur.

The WCD device was deployed to block one half of the bedroom window. This was to gain insight into
the question, if the WCD was not big enough to block an entire window, or if it can not be positioned

245
into place, is there a benefit to blocking as much of it as possible and if you block half of the flow do
you create better conditions in the structure? With the WCD in place, the conditions on the fire floor
did not improve much. The greatest impact was in the stairwell. With half of the window blocked the
temperatures below the fire floor returned to near ambient. This indicates that with the entire double
window unblocked there was enough flow to split and move both up and down the stairwell. However
with half the inlet area the flow did not have enough energy to split and the flow path up the stairwell
could support the total flow.

The 10.1 L/s (160 gpm) flow from the FBN 1 with a fog nozzle, directed into the bedroom, suppressed
the bedroom fire. When the nozzle was flowing, the temperatures throughout the fire floor decreased.
During the flowing of the FBN the corridor temperatures decreased from above 600 ºC (1112 ºF) to
approximately 300 ºC (572 ºF). These temperatures were still above those that support fire fighters
advancing down the corridor but the burning gases were stopped. Stairwell temperatures dropped to
below 200 ºC (392 ºF) which would allow control to be gained of that door if it was lost previously. The
flow into the bedroom did not suppress the living room fire which had vented itself out of the living
room windows. The nozzle was redeployed to the living room window on the floor below and the fire
was suppressed. This created conditions that would allow firefighters to advance and complete
extinguishment and other operations on the fire floor.

The use of the PPV fan after fire suppression reduced temperatures in the stairwell and returned
visibility faster than if natural ventilation was used. Even with the fire suppressed, there was still smoke
and hot gases being produced and the fan allowed these combustion products to be limited to the floor of
origin and not spread throughout the stairwell.

246
3.9 Experiment 5A 
The ninth experiment was located on the 5th floor in apartment 5A. Experiment 5A was ignited in the
bedroom furthest from the open apartment door (Figure 3.8-1). This experiment also utilized a
simulated wind provided by the MVU. The simulated wind was subjected to the bedroom double
window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The events that took
place during the experiment to accomplish the experimental objectives bulleted below are shown in
Table 3.9-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Examine impact of the deployment of a WCD on fire room window
• Control wind driven conditions with FBN in conjunction with WCD
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.8-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

247
Figure 3.9-1. 5th floor apartment A floor plan with locations pertinent to Experiment 5A

248
Table 3.9-1: Experiment 5A Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
234 234 Bedroom double Bottom right pane failed
322 325 Front lobby Right door opened
359 361 1st floor stair A Door opened
430 436 5th floor stair A Door opened
528 532 Bulkhead door Opened
588 588 Bedroom double Top right pane failed
605 606 Bedroom single Top pane failing
635 635 Bedroom double Bottom left pane failed
660 661 Bedroom double Top left pane failed
820 825 Bulkhead door Closed
831 832 Bedroom single Bottom pane failed
847 847 Bedroom single Clear
895 902 Bedroom double WCD deployed
986 996 Front lobby 2 - 27 inch fans activated
1304 1336 Bedroom double WCD bottom corner peeled back
1373 1373 Bedroom double FBN 2 started flowing (fog), blocked by WCD
1378 1382 Bedroom double WCD bottom corner peeled away from nozzle
1404 1404 Bedroom double FBN 2 stopped
1573 1573 Experiment completed

249
Figure 3.9-2. Instrumentation and camera locations with labels.

250
3.9.1 Observations 

The observations are presented as a series of images captured from eight camera locations, seven were
video cameras and one was a thermal imaging cameras. The camera positions inside the building are
shown in Figure 3.9-2.

Figure 3.9-3 through Figure 3.9-16 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 24 min (1440 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 5A, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor view
(CORRIDOR) shows the corridor leading up to the open apartment doorway. Next to that, the outside
lobby view (LOBBY) shows the area just outside the lobby doors. The next two views (STAIR and
STAIR IR) show the stairwell with a visual and a thermal imaging camera from the floor below the fire
looking up to the fire floor. The bottom left view (ROOF) shows the bulkhead door at the top of the
stairwell and the bottom right view (OUTSIDE) shows an external view of the fire apartment.

Figure 3.9-3 shows the images at the time of ignition. At this point, the seven video views were clear
and unobstructed. However, the thermal image provided limited thermal contrast because the surfaces
in the view are at nearly equal temperature. Soot and thermal damage visible in the corridor and stair
views are from previous experiments

Figure 3.9-4 shows the images at 240 s (4 min) after ignition. The bedroom double window began to
fail 6 seconds prior. Smoke had reached the floor in the living room, corridor, and bedroom views; a
significant volume of smoke was leaking around the closed 5th floor stair A door as well.

Figure 3.9-5 shows the images at 420 s (7 min) after ignition. At this point the front lobby door is open
and the 1st Floor door to stair A has been open for a minute. The smoke level in stair A is moving below
the 5th floor landing and the thermal imager view shows that the area around the stair door has become
hotter.

Figure 3.9-6 shows the images at 440 s (7 min and 20 s) after ignition. These images were recorded 4 s
after the 5th stair door was opened. The fire continued to burn in the bedroom. Smoke obscuration in
the stair increased after the door to the fire floor was opened.

Figure 3.9-7 shows the images at 480 s (8 min) after ignition. No changes had been made since the 5th
stair door was opened. Conditions in the stair continued to decay. Visibility was nearly zero in the
stairwell view, and the stair IR view was saturated by hot gases.

Figure 3.9-8 shows the images at 600 s (10 min) after ignition. These images were recorded
approximately 1 min after the bulkhead door was opened and 12 seconds after the top right pane of the
bedroom double window failed. The increased ventilation to the fire apartment enabled the flames to
spread to the living room. The visibility in the corridor and the stair had improved by a small amount.

251
Figure 3.9-9 shows the images at 720 s (12 min) after ignition. At this point the bedroom double
window had completely failed and the top portion of the bedroom single window had cracked and was
leaking smoke. The fire had intensified in the bedroom. Flames were no longer visible in the living
room. Smoke was venting from the open bulkhead door.

Figure 3.9-10 shows the images at 890 s (14 min and 50 s) after ignition. At 847 s the bedroom single
window failed (opened) completely. As a result of both the window and door openings to the bedroom
the excess fuel in the hot gas layer burned away and the fire was reduced to flames attached to the
burning objects in the room. The bulkhead door was closed approximately 1 min before the images
were recorded. The conditions in the stair had not changed significantly

Figure 3.9-11 shows the images at 910 s (15 min and 10 s) after ignition. The WCD deployment, over
the bedroom double window opening, was completed 8 s before the images were recorded. Flames were
extending out of the bedroom single window opening.

Figure 3.9-12 shows the images at 960 s (16 min) after ignition. The flames which had been pushing out
of the bedroom single window had withdrawn by this point. The hot gas layer in the living room had
increased in thickness.

Figure 3.9-13 shows the images at 1200 s (20 min) after ignition. The two 27 inch fans positioned in
front of the lobby doors were activated 204 s prior. The fuel throughout the bedroom had largely burned
out. The fans had improved visibility from the stair through the corridor and into the fire apartment, by
venting the smoke through the bedroom single window opening.

Figure 3.9-14 shows the images at 1365 s (22 min and 45 s) after ignition; approximately 30 s after the
WCD was partially removed. The reintroduction of wind into the bedroom increased the fire size and
flames were observed exiting the bedroom single window opening.

Figure 3.9-15 shows the images at 1385 s (23 min and 5 s) after ignition. FBN 2 had been deployed and
was flowing 160 gpm. The fire in the bedroom was suppressed and steam was being forced out of the
bedroom window openings. Visibility in the stair improved.

Figure 3.9-16 shows the images at 1440 s (24 min) after ignition. The water had flowed into the
bedroom unobstructed for approximately 20 s. Some debris in the bedroom was still burning. Smoke
had not yet cleared.

252
Figure 3.9-3. Experiment 5A, ignition.

253
Figure 3.9-4. Experiment 5A, 240s after ignition.

254
Figure 3.9-5. Experiment 5A, 420s after ignition.

255
Figure 3.9-6. Experiment 5A, 440s after ignition.

256
Figure 3.9-7. Experiment 5A, 480s after ignition.

257
Figure 3.9-8. Experiment 5A, 600s after ignition.

258
Figure 3.9-9. Experiment 5A, 720s after ignition.

259
Figure 3.9-10. Experiment 5A, 890s after ignition.

260
Figure 3.9-11. Experiment 5A, 910s after ignition.

261
Figure 3.9-12. Experiment 5A, 960s after ignition.

262
Figure 3.9-13. Experiment 5A, 1200s after ignition.

263
Figure 3.9-14. Experiment 5A, 1365s after ignition.

264
Figure 3.9-15. Experiment 5A, 1385s after ignition.

265
Figure 3.9-16. Experiment 5A, 1440s after ignition.

266
3.9.2 Temperatures 

Figure 3.9-17 through Figure 3.9-19 provide temperature data for five measurement locations on the fire
floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.9-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.9-17 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 5E in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. Opening
the bulkhead door at 532 s, coupled with the partially opened bedroom double window resulted in the
bedroom transitioning through flashover. As the bedroom double window vented completely,
temperatures throughout the 5th floor increased. Note that the temperatures in the flow path were higher,
while the temperature measurement position in front of Apartment 5E, out of the flow path, provided the
lowest temperature. The deployment of the WCD at approximately 900 s reduced the temperature in the
bedroom by almost 200 ºC (392 ºF). The addition of the two PPV fans in front of the lobby door
provided additional temperature reductions in the stair and throughout the 5th floor. Partial removal of
the WCD at approximately 1330 s resulted in increased temperatures in the bedroom and downstream
locations, until the fire was suppressed with the FBN.

Figure 3.9-18 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. The temperature time histories at
1.2 m (4 ft) above the floor in this experiment are very similar to the temperature time histories at 2.1 m
(7 ft) above the floor.

Figure 3.9-19 shows the temperatures 1.2 m (4 ft) above the landings on floors 1, 2, 4 and at the landing
above the 7th floor with access to the roof. The temperatures at the bulkhead level began to increase first
as the hot gases that leaked past the 5th floor stair door rose to the top of the stairwell. After the 5th floor
stair door was opened, the temperatures at the bulkhead decreased, while the temperatures at the
remaining measurement positions in the stairwell increased. Opening the bulkhead door reduced the
temperatures throughout the stairwell, until the bedroom transitioned to flashover. Then the
temperatures at the 4th floor landing and the bulkhead landing increased. After the bedroom single
window failed, the temperature at the bulkhead landing decreased, but the temperatures increased at the
4th and 2nd floor landings. Once the WCD was deployed, the temperatures throughout the stairwell were
reduced. The addition of the PPV fans brought all the temperatures to near ambient conditions, with the
exception of the bulkhead landing. The temperatures at the bulkhead landing were down to
approximately 100 ºC (212 ºF), until the WCD was partially removed. Then the temperatures at the 4th
floor landing and the bulkhead increased until the end of the experiment. The temperature at the 4th
floor decreased to ambient temperatures during the period of water application.

267
 le) le) e) le)
a r (d o ub (d oub doubl (doub eted
g le ( pl
i ng in l f c ing ar r
lea g WC on
D D g ed
WC flowin stopp nt com
n le fail t open ng ng in g le ha le fail le cle ed c g
i o i i e n s i n
ni t ub i gh en en en ub ng ub os ngl aci an ov BN FBN er im e
: Ig R: do BY: r 1: op 5: op H: op R: do R: si R: do H: cl R: si R: pl BY: f :m :F : p
BR B L S S B B B B B B B L BR BR BR Ex
1500 2732
TC BR5AE 2.1m
TC Corr5AE 2.1m
TC CorrStair5E 2.1m
1200 TC Stair5E 2.1m 2192
TC Corr5EE 2.1m

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 3.9-17. 5A Temperature vs. time at 2.1 m above the floor
le) le) e) le)
a r (d o ub (d oub doubl (doub eted
g le ( pl
i ng in l f c ing ar r
lea g WC on
D D g ed
WC flowin stopp nt com
n le fail t open ng ng in g le ha le fail le cle ed c g
i o i i l e n s i n
ni t ub i gh en en en ub ng ub os ng aci an ov BN FBN er im e
: Ig R: do BY: r 1: op 5: op H: op R: do R: si R: do H: cl R: si R: pl BY: f :m :F : p
BR B L S S B B B B B B B L BR BR BR Ex
1200 2192
TC BR5AE 1.2m
TC Corr5AE 1.2m
TC Corr5EE 1.2m
1000 TC CorrStair5E 1.2m
1832
TC Stair5E 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 3.9-18. 5A Temperature vs. time at 1.2 m above the floor

268
 le) le) e) le)
a r (d o ub (d oub doubl (doub eted
g le ( pl
i ng in l f c ing ar r
lea g WC on
D D g ed
WC flowin stopp nt com
n le fail t open ng ng in g le ha le fail le cle ed c g
i o i i e n s i n
ni t ub i gh en en en ub ng ub os ngl aci an ov BN FBN er im e
: Ig R: do BY: r 1: op 5: op H: op R: do R: si R: do H: cl R: si R: pl BY: f :m :F : p
BR B L S S B B B B B B B L BR BR BR Ex
500 932
TC Bulkhead MidE
TC Stair 4E
TC Stair 2E
TC Stair1 MidE
400 752

Temperature (F)
Temperature (C)

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 3.9-19. 5A stairwell temperature vs. time

3.9.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.9-20. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
5 Pa and only fluctuated slightly. After the 5th floor stair door was opened the pressure in the stair
increased with the pressures above the fire floor above 30 Pa.

The impact of opening the bulkhead door was a drop in pressure throughout the height of the stairwell.
The pressures increased again for the period between 588 s (venting of the top right portion of the
bedroom double window and 606 s (venting of the top portion of the bedroom single window).
Pressures in the stair remained low until the bulkhead door was closed at 825 s.

Pressures in the stairwell ranged from near 0 Pa at the 1st floor up to approximately 60 Pa at the floors
above the fire floor. Deploying the WCD reduced all of the pressures in the stairwell to less than 10 Pa.
The PPV fans increased the pressures in the stair to a range of 10 to 20 Pa. Once the WCD was partially
removed, the pressure from the wind drove the pressures in the stairwell up to approximately 40 Pa. The
period of suppression, led to another decrease in pressure in the stairwell.

269
 le) le) ) le)
r o ub o ub uble oub ed
i ng le a (d d
( (d o ( d l et
ing lf c iling lear ar WCDn CD ing pped omp
fail open g ha c l e W w c
itio
n le ht nin ning ning ble gle fa ble d c g
se gle cin ns
o g lo to t
vi n N f N s e n
ub g e u u
: Ign : do Y: ri : ope : ope : op : do : sin : do : clo : sin : pla Y: fa : mo : FB : FB per im
BR B R LB S 1 S 5 B H B R B R BR BH B R B R LB B R B R B R Ex
80
DP Stair 7E
DP Stair 6E
DP Stair 5E
60 DP Stair 4E
DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 3.9-20. 5A stairwell pressures vs. time

3.9.4 Velocities 

Average doorway velocities from 3 doors; the fire apartment door, the door from the fire floor to the
stairwell, and the door at the base of the stair, were recorded versus time and shown in Figure 3.9-21.
The three bi-directional probes in each door were averaged to provide the bulk flow through the doors.
The peak velocities on the fire floor occurred with the bulkhead door open and half of the bedroom
double window open. With the WCD in place, the PPV fans were able to reverse the flow through the
fire floor. A time history for each probe at each measurement location is located in Appendix D –
Detailed Graphs.

The velocity reinforces the concept of the flow path. Prior to the failure of the upwind window in the
bedroom, (room of fire origin), there is no significant flow out of the apartment 5A. After panes of the
upwind window began to self vent, the flow out of the apartment began to increase. Providing a
potential flow path between the fire apartment and the 1st floor, by opening the lobby doors, the 1st floor
stair door and the 5th floor stair door, increased the flow from the fire apartment to approximately
2 m/s (4 mph). Given that the combustion products entering the stairwell are buoyant, they have
tendency to flow up instead of down the stairs. Hence when the bulkhead door was opened to provide
an flow path to the roof, the velocity into the stair increased to approximately 4 m/s (9 mph). With this
“preferred” flow path from the upwind window to the bulkhead open the fire continued to grow,
additional portions of the window failed and the velocity out of the apartment and into the stair peaked
at approximately 6 m/s (13 mph). The impact of closing the bulkhead door at 825 s after ignition
reduced the flow out of the fire apartment to less than 1 m/s (2 mph). With the WCD was in place and
the bulkhead door closed, the wind driven conditions were stopped. By using PPV fans on the 1st floor,
the flow into the stairwell was reversed as fresh air was pushed from the higher pressure in the stairwell
onto the fire floor and back through the fire apartment. This flow continued until the WCD was
removed and suppression began.

270
 le) le) e) le)
a r (d o ub (d oub doubl (doub eted
g le ( pl
i ng in l f c ing ar r
lea g WC on
D D g ed
WC flowin stopp nt com
n le fail t open ng ng in g le ha le fail le cle ed c g
i o i i e n s i n
ni t ub i gh en en en ub ng ub os ngl aci an ov BN FBN er im e
: Ig R: do BY: r 1: op 5: op H: op R: do R: si R: do H: cl R: si R: pl BY: f :m :F : p
BR B L S S B B B B B B B L BR BR BR Ex
10 22
Exp 5A Stair1
Exp 5A Stair5
8 Exp 5A Apt5A 18

6 13

Velocity (mph)
Velocity (m/s)

4 9

2 4

0 0

-2 -4

-4 -9
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 3.9-21. 5A Velocities vs. time

3.9.5 Discussion 

In this experiment, a simulated wind driven fire condition was created which caused extremely
hazardous conditions inside the building. The peak temperatures were generated by the creation of a
flow path between the bedroom and the vented stairwell. The failure (self venting) of half of the
bedroom double window combined with the open door between the fire floor and the stairwell and the
open bulkhead door at the top of the stair led to the creation of a flow path. Temperatures in the flow
path between the open bedroom window and the 5th floor stair exceeded 400 ºC (752 ºF) throughout the
fire floor at both the 2.1 m (7 ft) and 1.2 m (4 ft) elevations above floor level. Heat and smoke entered
the stair, resulting in poor visibility and increased temperatures both above and below the fire floor. In
addition, the ventilation condition caused by the venting and doors resulted in a peak gas velocity of
approximately 8 m/s (18 mph) on the fire floor. The thermal conditions created during this experiment
were not conducive to a direct frontal fire attack and firefighters remote from the fire in the stairwell
would have been threatened if they were in the flow path due to extreme temperatures.

The first pane of the bedroom window failed at 234 s and the window completely failed at 661 s. The
temperature in the center of the bedroom at 1.2 m (4 ft) at the time of window failure was approximately
400 ºC (752 ºF). The temperature in the center of the room at the time of complete window failure was
in excess of 900 ºC (1652 ºF). This demonstrates that when a portion of the window fails the wind being
introduced through the opening cools and slows the development of the fire and the failure of the
remaining sections of this type of window. However, given the fuel load, the partial failure of the
window and the ventilation from the open stair door on the fire floor provided the oxygen required to
achieve flashover. The majority of the window did not fail until after the room transitioned to flashover.

Similar to previous experiments, the fire took a significant amount of time to develop even with a failed
window pane in the fire room. The air was forced into the bedroom but had nowhere to travel to with
the exception of leaks around doors. Once the stair door was opened the fire began to grow slowly.

271
With the first floor stair door opened heat and smoke were pushed down the stairs. This flow was
reversed at approximately 530 s when the bulkhead door was opened. Once the bulkhead door was
opened, the pressure in the stairwell dropped and the hot gases flowed up the stairwell. This increase in
flow through the fire apartment enabled the fire grow rapidly. The open bulkhead door provided the
flow path needed to draw fire out of the room of origin and enable the thermal hazards of a wind driven
condition on the fire floor. In this experiment the most notable changes in temperature, pressure and
velocity were caused by opening or closing the bulkhead door. This shows that interrupting the flow
path, anywhere on the flow path, can have a significant effect on the thermal conditions in the structure.

The WCD device was only deployed over the upwind bedroom double window only. This was done to
examine the effect that another open vent to the outside (bedroom single window) would have on the
conditions in the building and what impact, if any, it would have on the WCD. The deployment was
completed at approximately 900 s; more than a minute after the bulkhead door was closed. With the
WCD in place the temperatures in the bedroom decreased from approximately 800 ºC (1472 ºF) to
600 ºC (1112 ºF). Temperatures at other locations on the fire floor and the stairwell also decreased.

The introduction of PPV, approximately 90 s after the deployment of the WCD, pressurized the stairwell
and as a result reversed the flow of combustion products back through the fire apartment to be vented
out of the bedroom single window opening. The PPV fans improved the visibility and reduced the
temperatures in the stairwell.

The 10.1 L/s (160 gpm) flow from the FBN 2 with a fog nozzle, directed into the bedroom, rapidly
suppressed the decaying bedroom fire. The FBN only flowed water for approximately 30 s. While the
water was flowing, the temperature in the bedroom decreased from approximately 600 ºC (1112 ºF) to
approximately 100 ºC (212 ºF). Since the WCD was partially removed, after the water flow was
stopped, the temperatures increased due to the wind pushing heat from the bedroom through the fire
floor and into the stairwell.

272
3.10 Experiment 5G 

The tenth experiment was located on the 5th floor in apartment 5G. Experiment 5G was ignited in the
bedroom furthest from the open apartment door (Figure 3.10-1). This experiment also utilized a
simulated wind provided by the MVU. The simulated wind was subjected to the bedroom double
window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The events that took
place during the experiment to accomplish the experimental objectives bulleted below are shown in
Table 3.10-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Examine impact of the deployment of a WCD on fire room window
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.10-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

273
Figure 3.10-1. 5th floor apartment G floor plan with locations pertinent to Experiment 5G

274
Table 3.10-1: Experiment 5G Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom Ignition
425 425 Bedroom double Bottom left pane failed
491 493 Front lobby Left door opened
503 506 1st floor stair A Door opened
541 549 5th floor stair A Door opened
569 576 Bulkhead door Opened 0.02 m (0.75 in)
602 602 Bedroom double Top right pane failed
628 628 Bedroom double Bottom right pane failed
632 638 Bulkhead door Opened (all the way)
648 648 Bedroom double Top left pane failed
671 Bedroom single Failed
877 879 Bedroom single WCD deployed
959 1008 Living room window Manually vented
1326 1326 Experiment completed

275
Figure 3.10-2. Instrumentation and camera locations with labels.

276
3.10.1 Observations 

The observations are presented as a series of images captured from eight camera locations, seven were
video cameras and one was a thermal imaging camera. The camera positions inside the building are
shown in Figure 3.10-2.

Figure 3.10-3 through Figure 3.10-12 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 28 min (1680 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 5G, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor view
(CORRIDOR) shows the corridor leading up to the open apartment doorway. Next to that, the outside
lobby view (LOBBY) shows the area just outside the lobby doors. The next two views (STAIR and
STAIR IR) show the stairwell with a visual and a thermal imaging camera from the floor below the fire
looking up to the fire floor. The bottom left view (ROOF) shows the bulkhead door at the top of the
stairwell, this view is disconnected shortly after ignition. The bottom, right view (OUTSIDE) shows an
external view of the fire apartment.

Figure 3.10-3 shows the images at the time of ignition. At this point, the seven video views were clear
and unobstructed. However, the thermal image provided limited thermal contrast because the surfaces
in the view are at nearly equal temperature. The damage to the corridor and stairwell from Experiment
5K can be seen as the existing condition in these areas.

Figure 3.10-4 shows the images at 240 s (4 min) after ignition. The fire ignited in the trash container at
the base of the bed has just spread to the comforter and sheets on the bed. The smoke layer in the
ignition bedroom has descended to about 0.9 m (3 ft) above the floor. The smoke in the living room or
corridor was light in color and not very dense. At this point the roof view had malfunctioned and was
lost for the remainder of this experiment.

Figure 3.10-5 shows the images at 480 s (8 min) after ignition. The lower left pane of the bedroom
double window had failed 55 s prior. Smoke had filled the bedroom, living room, and corridor reducing
the visibility in those views to nearly zero. Smoke which was leaking past the 5th floor stair A door was
descending the stairwell.

Figure 3.10-6 shows the images at 550 s (9 min and 10 s) after ignition. The stairwell door on the 5th
floor was opened and visibility in the stairwell on the floor below the fire was reduced further. Heat was
also flowing down the stairwell as seen in the thermal imaging view. The lobby doors were opened 44 s
prior creating a flow path down the stairwell. No flames are visible from the Tower View and the
bedroom camera has been obscured by soot.

Figure 3.10-7 shows the images at 575 s (9 min and 35 s) after ignition. Flames were now visible in the
living room and smoke was flowing out of the lobby doors. Flames were visible from the corridor and
the density of the smoke in the stairwell was decreasing. There was still only one failed window pane in

277
the bedroom so the fire was still growing after being cooled initially by the introduction of cold air from
the outside.

Figure 3.10-8 shows the images at 630 s (10 min and 30 s) after ignition. The remainder of the bedroom
double window’s panes had failed and flames were pulsing back out toward the simulated wind.
Visibility worsened in the corridor and stairwell and more heat was being forced down the stairwell.
The amount of smoke coming out of the lobby from the stairwell was increasing.

Figure 3.10-9 shows the images at 720 s (12 min) after ignition. The bulkhead door had been opened
completely 82 s prior. Smoke density in the stairwell was noticeably reduced. Flames were no longer
pushing from the bedroom double window. Smoke was still coming out of the lobby and the thermal
imaging camera had malfunctioned due to extreme heat exposure.

Figure 3.10-10 shows the images at 885 s (14 min and 45 s) after ignition. A WCD was deployed over
the bedroom single window to prevent the loss of wind flow through that window and create worse
conditions downstream in the structure. Visibility was beginning to return to the living room and no
flames were visible. Flames were pulsing out of the bedroom double window again.

Figure 3.10-11 shows the images at 960 s (16 min) after ignition. Smoke density in the stairwell had
decreased noticeably from 80 s prior when the WCD was deployed over the bedroom single window
indicating less burning and more flow moving up the stairwell. There was no longer smoke coming
from the lobby doors.

Figure 3.10-12 shows the images at 1200 s (20 min) after ignition. The living room windows had been
manually ventilated at 1008 s to see if there was any fire in the living room and to see if it would
intensify. At this time there did not appear to be any fire in the living room. Smoke had largely cleared
the stairwell at this point. Visibility in the corridor had also improved significantly. The tower view
was temporarily disabled during this period.

278
Figure 3.10-3. Experiment 5G, ignition.

279
Figure 3.10-4. Experiment 5G, 240s after ignition.

280
Figure 3.10-5. Experiment 5G, 480s after ignition.

281
Figure 3.10-6. Experiment 5G, 550s after ignition.

282
Figure 3.10-7. Experiment 5G, 575s after ignition.

283
Figure 3.10-8. Experiment 5G, 630s after ignition.

284
Figure 3.10-9. Experiment 5G, 720s after ignition.

285
Figure 3.10-10. Experiment 5G, 885s after ignition.

286
Figure 3.10-11. Experiment 5G, 960s after ignition.

287
Figure 3.10-12. Experiment 5G, 1200s after ignition.

288
3.10.2 Temperatures 

Figure 3.10-13 through Figure 3.10-15 provide temperature data for five measurement locations on the
fire floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.10-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location can be found in Appendix D – Detailed Graphs.

Figure 3.10-13 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 5K in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
temperature in the bedroom increased to above 300 ºC (572 ºF) and became ventilation limited. The
temperature remained constant until the window began to fail. After the first pane failed the temperature
increased to over 600 ºC (1112 ºF). The temperature decreased again and then climbed above 800 ºC
(1472 ºF) with the fire floor stairwell door opened. The temperature in the bedroom remained above
700 ºC (1292 ºF) for the remainder of the experiment.

Corridor and stairwell temperatures at 2.1 m (7 ft) remained below 200 ºC (392 ºF) until the 5th floor
stairwell door was opened and the temperatures increased to above 300 ºC (572 ºF). The corridor
temperature peaked at approximately 450 ºC (842 ºF). A wind driven fire never fully developed and the
temperatures decreased after 800 s because the fuel was burning down in the bedroom and the living
room never ignited.

Figure 3.10-14 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. The temperature in the bedroom at
1.2 m (4 ft) increased to approximately 250 ºC (482 ºF) and became ventilation limited. The
temperature remained constant until the window began to fail. After the first pane failed the temperature
increased to over 500 ºC (932 ºF). The temperature decreased again and then climbed above 800 ºC
(1472 ºF) with the fire floor stairwell door opened. The bedroom transitioned to flashover and the
temperatures at every elevation remained above 700 ºC (1292 ºF) for the remainder of the experiment.

Corridor and stairwell temperatures at 1.2 m (4 ft) remained below 200 ºC (392 ºF) until the 5th floor
stairwell door was opened and the temperatures increased to above 300 ºC (572 ºF). The corridor
temperature tracked with the 2.1 m (7 ft) elevation for the remainder of the experiment, while the wind
was flowing into the bedroom creating a well mixed flow and no lower layer.

Figure 3.10-15 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, 6 and at the
landing above the 7th floor with access to the roof. After the fire floor stairwell door was opened the
temperature on the landing above and below the fire floor increased to over 200 ºC (392 ºF). Once the
bulkhead door was opened the temperature below the fire floor decreased and the temperature above the
fire floor increased to 340 ºC (644 ºF) and the temperature at the top of the stairwell increased to 250 ºC
(482 ºF).

289
 le)
i ng ted
i ng ning 02
m
a r D (s enting pl e
a il
e f ft ope ning ning ning n
0 .
ec
le C
g W dow v com
ni tion ubl e e e e ub l acin ent
: Ig : do BY: le 1: op 5: op H: op H: op R: do : pl R: wi
n er im
BR BR L S S B B B BR L E xp
1200 2192
TC BR5GW 2.1m
TC Corr5GW 2.1m
TC Corr5KW 2.1m
1000 TC CorrStair5W 2.1m
1832
TC Stair5W 2.1m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.10-13. 5G Temperature vs. time at 2.1 m above the floor
)
gle ted
g 2 m (sin ting le
ili n nin
g
0. 0 lea
r D
WC w ven co mp
itio
n le fa ope ing ing ing le c ng o ent
: Ig
n
: d oub Y: left : open : open : open : open : doub : p laci : wind er im
BR BR LB S1 S5 BH BH BR BR LR E xp
1200 2192
TC BR5GW 1.2m
TC Corr5GW 1.2m
TC Corr5KW 1.2m
1000 TC CorrStair5W 1.2m
1832
TC Stair5W 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.10-14. 5G Temperature vs. time at 1.2 m above the floor

290
 le)
i ng ted
i ng ning 02
m
a r D (s enting pl e
a il
e f ft ope ning ning ning n
0 .
ec
le C
g W dow v com
ni tion ubl e e e e ub l acin ent
: Ig : do BY: le 1: op 5: op H: op H: op R: do : pl R: wi
n er im
BR BR L S S B B B BR L E xp
400 752
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
300 572

Temperature (F)
Temperature (C)

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.10-15. 5G stairwell temperature vs. time

3.10.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.10-16. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
8 Pa and only fluctuated slightly. When the window in the fire room failed the pressures in the stairwell
increased slightly due to the leaks around the fire floor door. After the fire floor door was opened the
pressure in the top of the stairwell increased to as high as 68 Pa at the top of the stairwell and
approximately 50 Pa in the area of the fire floor. Pressures below the 4th floor remained below 20 Pa
due to the 1st floor door being opened. The bulkhead door was opened at 635 s and the pressures
decreased to below 10 Pa with the exception of the fire floor.

291
 le)
i ng ted
i ng ning 02
m
a r D (s enting pl e
a il 0
e f ft ope ning ning ning n
.
ec
le C
gW wv com
ni tion ub l e e e e ubl acin ndo ent
: Ig : do BY: le 1: op 5: op H: op H: op R: do R: pl R: w i er im
BR BR L S S B B B B L E xp
80
DP Stair 7W
DP Stair 6W
DP Stair 5W
60 DP Stair 4W
DP Stair 3W
DP Stair 2W
DP Stair 1W
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.10-16. 5G stairwell pressures vs. time

3.10.4 Velocities 

Average doorway velocities from 4 doors; the fire apartment door, the door from the fire floor to the
stairwell, the door at the base of the stair and the door at the bulkhead, were recorded versus time and
shown in Figure 3.10-17. The three bi-direction probes in each door were averaged to provide the bulk
flow through the doors. A time history for each probe at each measurement location is located in
Appendix D – Detailed Graphs.

Velocities out of the fire apartment and into the corridor were approximately 1 m/s (2 mph) after the
windows failed and increased to 3 m/s (7 mph) with the fire floor door open and increased further to
approximately 7 m/s (16 mph) after the bulkhead door was opened. Flows out of the bulkhead door
reached 10 m/s (22 mph) at the peak and bulk flows into the 1st floor stairwell door were approximately
1 m/s (2.2 mph).

292
 le)
i ng ted
i ng ning 02
m
a r D (s enting pl e
a il
e f ft ope ning ning ning n
0 .
ec
le C
g W dow v com
ni tion ubl e e e e ub l acin ent
: Ig : do BY: le 1: op 5: op H: op H: op R: do : pl R: wi
n er im
BR BR L S S B B B BR L E xp
10 22

5 11

Velocity (mph)
Velocity (m/s)

0 0

-5 -11

-10 Exp 5G Bulkhead -22

Exp 5G Stair1
Exp 5G Stair5
Exp 5G Apt5G
-15 -34
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.10-17. 5G Velocities vs. time

3.10.5 Discussion 

This experiment began in a similar way to many of the others, with a growing fire in the bedroom that
became ventilation limited until the window failed. The difference in this experiment was that a fully
developed wind driven condition was never created and the fire ran out of fuel before flames could be
forced into the corridor. A possible reason for this could be that the single window in the bedroom
failed allowing a loss of flow and more burning locally to consume the fuel faster. The source of wind
was directed at the bedroom double window and when the single window failed that became the path of
least resistance. There was more flow entering the room than could leave out the single window so flow
did force hot gases into the corridor and hallway but flames being forced into the living room from floor
to ceiling did not occur. Flames temporarily were forced into the living room, but nothing was ignited.
A WCD was deployed to block the loss of flow out of the single window, but by that time in the
experiment, there was not enough fuel remaining in the bedroom to generate the energy and conditions
for flashover.

While a fully developed wind driven condition was not achieved a possible tactic was highlighted. If it
can be determined from the team above the fire apartment that the wind is only subjected to one side of
the fire room and there is a second window on the downwind side of the fire room that has not self
ventilated then manually ventilating that window could reduce the impact of the wind down stream by
providing a low pressure relief locally to the fire. Another vent should not be made on the upwind side
or if it is suspected that the wind could shift, because this could lead to additional wind pushing into the
fire room, which could add to the wind driven fire flows in the structure.

The first pane of the bedroom window failed at 425 s and the window completely failed at 648 s. The
temperature in the center of the bedroom at 1.2 m (4 ft) at the time of window failure was approximately
240 ºC (464 ºF). The temperature in the center of the room at the time of complete window failure was
in excess of 800 ºC (1472 ºF). This demonstrates that when a portion of the window fails the wind being

293
introduced through the opening cools and slows the failure of the rest of the window. The majority of
the window did not fail until the room transitioned to flashover. The single window in the room failed at
671 s when the room temperature was approximately 850 ºC (1562 ºF).

Once again during this experiment the fire could not grow until a ventilation path was created. When
the flow path was made to go down the stairs and out the 1st floor the fire grew but did so slowly. When
the path was transitioned to travel up the stairs and out of the bulkhead the fire peaked but had run low
on fuel. The hot gases traveled past the living room but did not ignite it because of lack of oxygen in the
living room and lack of a flow path toward the living room windows. When the living room window
was manually ventilated to try to flow past the sofas the fire in the bedroom had already burned down
and would not spread.

294
3.11 Experiment 5G2 

The eleventh experiment in the structure was located in the same apartment as the tenth experiment, 5G.
Experiment 5G2 was ignited in the living room on the top of sofa that backed to the bedroom (Figure
3.11-1). The fire in the bedroom did not extend beyond the room of origin so the apartment was cleared
of smoke and the experiment was ignited. Both windows in the back bedroom had failed in the previous
experiment so they remained open during this experiment. The living room window was also manually
ventilated during the previous experiment so this experiment began with the window open. This
experiment also utilized a simulated wind provided by the MVU. The simulated wind was subjected
into the living room window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The
events that took place during the experiment to accomplish the experimental objectives below are shown
in Table 3.11-1.

Experimental Objectives
• Create wind driven fire conditions from a living room fire
• Examine building flows with different door opening configurations
• Examine impact of the deployment of a WCD on fire room window
• Utilize portable PPV fans to control smoke movement
• Control wind driven conditions with FBN in conjunction with WCD

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.11-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

295
Figure 3.11-1. 5th floor apartment G floor plan with locations pertinent to Experiment 5G2

296
Table 3.11-1: Experiment 5G2 Timeline
Start (s) End (s) Event Location Event Description
0 0 Living room Ignition (window vented)
39 44 1st floor stair A 27 inch fan deactivated
177 187 5th floor stair A Door closed
518 521 5th floor stair A Door opened
543 547 Bulkhead door Opened
687 706 Living room window WCD deployed
746 748 Front lobby 2 - 27 inch fans activated
801 803 Living room window WCD pulled back at bottom right
819 825 Bulkhead door Closed
823 834 Living room window FBN 2 inserted, bottom right replaced
879 879 Living room window FBN 2 started flowing water (fog)
932 932 Living room window FBN 2 stopped
933 944 Living room window WCD pulled back at bottom right
953 953 Living room window FBN 2 started flowing water (fog)
978 978 Living room window FBN 2 stopped
1084 1085 Living room window WCD replaced
1125 1157 Front lobby One 27 inch fan moved inside
1138 1141 Front lobby Right door opened
1172 1175 Front lobby Right door closed
1350 1350 Experiment completed

297
Figure 3.11-2. Instrumentation and camera locations with labels.

298
3.11.1 Observations 

The observations are presented as a series of images captured from eight camera locations, seven were
video cameras and one was a thermal imaging camera. The camera positions inside the building are
shown in Figure 3.11-2.

Figure 3.11-3 through Figure 3.11-15 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 28 min (1680 s) after ignition. Images were captured every
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (BEDROOM) at the top left of each figure shows the bedroom of
apartment 5G, the bedroom furthest from the apartment entrance with a view of the single window
straight across, double window to the right and ignition trashcan at the foot of the bed. The top right
view (LIVINGRM) was of the living room looking across to the windows. The corridor view
(CORRIDOR) shows the corridor leading up to the open apartment doorway. Next to that, the outside
lobby view (LOBBY) shows the area just outside the lobby doors. The next two views (STAIR and
STAIR IR) show the stairwell with a visual and a thermal imaging camera from the floor below the fire
looking up to the fire floor. The bottom left view (ROOF) shows the bulkhead door at the top of the
stairwell, this view is disconnected shortly after ignition. The bottom, right view (OUTSIDE) shows an
external view of the fire apartment.

Figure 3.11-3 shows the images at the time of ignition. The bedroom and roof views were still missing
from the previous experiment (experiment 5G). The living room window was vented in the previous
experiment. The firefighters in the corridor view were performing a remote ignition of the living room
and retreated from that area shortly after.

Figure 3.11-4 shows the images at 240 s (4 min) after ignition. At this point the sofa where ignition
occurred was mostly involved in flames, and those flames were visible from the corridor view, which
looks directly into the living room area through the apartment entrance. While there was no visible
smoke layer, the amount of smoke in the living room, stairwell, and corridor had increased noticeably.

Figure 3.11-5 shows the images at 480 s (8 min) after ignition. At this point the sofa which provided the
ignition location in the living room was fully involved. Visibility in the corridor has reduced to zero.
Conditions in the stairwell had not changed noticeably from 240 s prior.

Figure 3.11-6 shows the images at 565 s (9 min and 25 s) after ignition. The living room was
transitioning to flashover and flames were beginning to spread into the corridor. High heat conditions
have made it to the open stairwell and smoke was exiting the front lobby doors. The bulkhead door was
opened 18 s earlier.

Figure 3.11-7 shows the images at 695 s (11 min and 35 s) after ignition. The WCD deployment began
over the living room window 8 s prior to this image. The view in the living room and corridor was
darkened down due to under-ventilated conditions. Visibility in the stairwell was zero and heat was still
visible coming down the stairs.

299
Figure 3.11-8 shows the images at 720 s (12 min) after ignition. At this point the corridor camera view
was lost for the remainder of the test. Significant volumes of smoke were venting down the stairwell
and out the open lobby doors. The conditions in the stairwell remained the same from the previous set
of images.

Figure 3.11-9 shows the images at 820 s (13 min and 40 s) after ignition. This image showed the WCD
being pulled back so that the FBN 2 could be inserted from below. Flames were visible in the living
room and conditions in the stairwell were still poor. Two PPV fans were operating outside the lobby
doors. Shortly after this image the WCD was placed back into place to cover the FBN 2.

Figure 3.11-10 shows the images at 945 s (15 min and 45 s) after ignition. The FBN 2 with a fog nozzle
was flowed into the living room for 53 s and then the WCD was pulled back to observe its effectiveness.
There were no longer flames visible from the outside or inside views. There was no longer smoke
coming out of the lobby doors and the stairwell view was not clear but could be obscured by soot
deposited on the lens of the camera.

Figure 3.11-11 shows the images at 960 s (16 min) after ignition. At this point conditions had not
changed significantly from 15 s prior. The FBN 2 was flowed again with the WCD pulled back to
further extinguish the fire. Some steam was coming out of the living room and conditions in the
stairwell were improved.

Figure 3.11-12 shows the images at 1200 s (20 min) after ignition. The fire in the living room was
largely extinguished, and visibility had returned to all of the remaining views. One PPV fan was moved
inside to the 1st floor doorway. It was activated 44 s prior and the visibility in the stairwell improved.

300
Figure 3.11-3. Experiment 5G2, ignition.

301
Figure 3.11-4. Experiment 5G2, 240s after ignition.

302
Figure 3.11-5. Experiment 5G2, 480s after ignition.

303
Figure 3.11-6. Experiment 5G2, 565s after ignition.

304
Figure 3.11-7. Experiment 5G2, 695s after ignition.

305
Figure 3.11-8. Experiment 5G2, 720s after ignition.

306
Figure 3.11-9. Experiment 5G2, 820s after ignition.

307
Figure 3.11-10. Experiment 5G2, 945s after ignition.

308
Figure 3.11-11. Experiment 5G2, 960s after ignition.

309
Figure 3.11-12. Experiment 5G2, 1200s after ignition.

310
3.11.2 Temperatures 

Figure 3.11-13 through Figure 3.11-15 provide temperature data for four measurement locations on the
fire floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.11-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.11-13 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 4 measurement
locations; just outside the fire apartment in the corridor, just outside Apartment 7K in the corridor, just
outside the A stairwell doorway, and just inside the A stairwell on the landing. The temperatures in the
hallway increased as the fire in the living room increased. The fire transitioned to flashover at
approximately 565 s, 30 s after the bulkhead door was opened. Shortly after this time the temperatures
in the flow path increased from 250 ºC (482 ºF) to over 600 ºC (1112 ºF) in seconds. The temperature in
the corridor peaked at 800 ºC (1472 ºF) and decreased to just over 600 ºC (1112 ºF) before the WCD
was deployed. The WCD maintained temperatures below 600 ºC (1112 ºF) by eliminating the wind but
air was still available to the fire from the open bedroom windows. Temperatures dropped further when
the FBN was directed into the living room. The water decreased the temperature to below 300 ºC
(572 ºF) in less than 1 min of flowing.

Figure 3.11-14 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. These temperatures were almost
identical to the 2.1 m (7 ft) measurements because the wind was flowing into the apartment for the
duration of the experiment. The wind was mixing all the gases and there was no layering in the corridor
or stairwell.

Figure 3.11-15 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 4, 6 and at the
bulkhead. After the 5th floor stairwell door was opened there was a rush of heat into the stairwell,
including increases in temperature all the way down to the lobby of 100 ºC (212 ºF). The bulkhead door
was opened 16 s later and the flow transitioned to up the stairwell strictly. Temperatures as high as
600 ºC (1112 ºF) were measured at the floor above the fire and the bulkhead temperature peaked at
approximately 400 ºC (752 ºF). The WCD maintained stairwell temperatures below 500 ºC (932 ºF) and
the FBN decreased the temperatures to less than 200 ºC (392 ºF) quickly. The combination of these
tactics would allow a fire fighting crew to advance to the fire apartment.

311
 Ex
LR p
LR
:
LR
: : W LR LR: LR: LR LR: LR: W LB er im e
LR S B p l a L B m o C :F F B m o :F F B C Y: nt
: I g S1: f S5: c 5: op H: op cing Y: f a ving BH: c D rep BN fl N sto ving BN fl N sto D rep f an m comp
nit i an l os eni eni W C ns WC los lac owi p WC owi p l ac ov
on off ed ng ng D on D ed ed ng p ed D ng p ed ed ed leted

1000 1832
TC Corr5GW 2.1m
TC Corr5KW 2.1m
TC CorrStair5W 2.1m
TC Stair5W 2.1m
800 1472

Temperature (F)
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.11-13. 5G2 Temperature vs. time at 2.1 m above the floor
Ex
LR p
LR
:
LR
: : W LR LR: LR: LR LR: LR: W LB er im e
LR S B pl a LB m o C : F FB m o : F FB C Y: nt
H B D B B D
: I g S1: f S5: c 5: op : op cing Y: f a ving H: c rep N fl N sto ving N fl N sto rep f an m comp
nit i an l e e W n W l l o W o l
on off osed ni ng ning CD s on CD osed aced wing pped CD wing pped aced oved leted

1000 1832
TC Corr5GW 1.2m
TC Corr5KW 1.2m
TC CorrStair5W 1.2m
TC Stair5W 1.2m
800 1472

Temperature (F)
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.11-14. 5G2 Temperature vs. time at 1.2 m above the floor

312
 Ex
LR p
LR
:
LR
: : W LR LR: LR: LR LR: LR: W LB er im e
LR S B p l a L B m o C :F F B m o :F F B C Y: nt
: I g S1: f S5: c 5: op H: op cing Y: f a ving BH: c D rep BN fl N sto ving BN fl N sto D rep f an m comp
nit i an l os eni eni W C ns WC los lac owi p WC owi p l ac ov
on off ed ng ng D on D ed ed ng p ed D ng p ed ed ed leted

800 1472
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
600 1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.11-15. 5G2 stairwell temperature vs. time

3.11.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.11-16. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. The wind was subjected to the fire apartment from ignition time on so
there was no ambient pressure. The initial pressures were 20 Pa to 40 Pa at the top of the stairwell and 0
Pa to 20 Pa in the lower portion of the stairwell with the fire floor door open and the 1st floor stair door
open. When the 5th floor door was closed the pressures all dropped below 20 Pa. Opening the fire floor
door again when the fire was growing increased the pressures to as high as 60 Pa. The pressures
dropped again when the bulkhead door was opened 16 s later. Pressures remained low until the WCD
was pulled back for the nozzle and the PPV fans were turned on. Pressures reached as high as 80 Pa
with the effects of the fire, wind and fans.

313
 Ex
LR p
LR
:
LR
: : W LR LR: LR: LR LR: LR: W LB er im e
LR S B p l a L B m o C :F F B m o :F F B C Y: nt c
: I g S 1: f S 5: c 5: o H: o cin Y: v B D B N v B N D fa
nit i an l os pen pen g W f ans ing W H: clo repla N flow stop ing W N flow stop repla n mo omple
on off ed i ng ing CD on C D sed ced i ng p ed C D i ng p ed ced ved ted

80
DP Stair 7W
DP Stair 6W
DP Stair 5W
60 DP Stair 4W
DP Stair 3W
DP Stair 2W
40 DP Stair 1W
Pressure (Pa)

20

-20

-40
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.11-16. 5G2 stairwell pressures vs. time

3.11.4 Velocities 

Average doorway velocities from 3 doors; the fire apartment door, the door from the fire floor to the
stairwell and the door at the base of the stair, were recorded versus time and shown in Figure 3.11-17.
The three bi-direction probes in each door were averaged to provide the bulk flow through the doors. A
time history for each probe at each measurement location is located in Appendix D – Detailed Graphs.

The initial velocities while the fire was growing were approximately 3 m/s (7 mph) into the fire
apartment and down the stairwell to the first floor. This bulk velocity decreased to less than 1 m/s
(2 mph) when the stairwell was closed. The peak velocity out of the fire apartment and into the stairwell
was approximately 7 m/s (16 mph) when the fire was at its peak and the stair and bulkhead doors were
open. Deploying the WCD slowed the velocity and suppressing the fire decreased the velocity to less
than 2 m/s (4 mph).

314
 Ex
LR p
LR
:
LR
: : W LR LR: LR: LR LR: LR: W LB er im e
LR S B p l a L B m o C :F F B m o :F F B C Y: nt
: I g S1: f S5: c 5: op H: op cing Y: f a ving BH: c D rep BN fl N sto ving BN fl N sto D rep f an m comp
nit i an l os eni eni W C ns WC los lac owi p WC owi p l ac ov
on off ed ng ng D on D ed ed ng p ed D ng p ed ed ed leted

8 18
Exp 5G2 Stair1
Exp 5G2 Stair5
Exp 5G2 Apt5G
6 13

4 9

Velocity (mph)
Velocity (m/s)

2 4

0 0

-2 -4

-4 -9
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 3.11-17. 5G2 Velocities vs. time

3.11.5 Discussion 

Wind driven fire conditions were created utilizing the simulated wind from the MVU. Flames entered
the corridor and temperatures in the stairwell exceeded 600 ºC (1112 ºF). Large amounts of smoke were
forced out of the front lobby door displaying untenable conditions throughout the entire stairwell. The
condition created during this experiment were not conducive to a direct frontal fire attack and
firefighters remote from the fire in the stairwell would have been threatened if they were in the flow path
due to extreme temperatures.

The wind was able to flow into the living room through the open door so the fire had to grow with cool
air moving past it, which it was able to do. As in many of the experiments, the fire did not grow quickly
when the flow path was down the stairs but transitioned quickly when the bulkhead door was opened.
The fire transitioned to flashover and failed to result in a wind driven condition. The hot gases were
primarily being forced to the stairwell and up the stairs to the open bulkhead door. Once the WCD was
deployed the conditions changed. The flow was stopped but the fire still was able to burn because of the
air available from the open windows in the bedrooms and the open stairwell. The fire floor door is a
key door to control to limit the impact to the rest of the structure. With this door open all floors were
threatened with untenable conditions for firefighters and occupants.

The WCD was effective at minimizing the impact of the wind, but in this scenario there was still plenty
of oxygen available to the fire from the opening made from the previous experiment. The WCD relies
on limiting the oxygen to slow the burning and lower the temperatures. The WCD held up well to
extreme temperature above 500 ºC (932 ºF) for more than 150 s. The WCD was also controlled by the
crew to be able to pull a corner of the WCD away to insert the FBN. At one point, the WCD was
wrapped around the FBN, but that was fixed by the deployment crews.

315
The PPV fans were activated to push the smoke up the stairwell once the WCD was deployed. The fans
were effective when the wind was blocked but were less effective when the WCD was pulled back for
the nozzle to be inserted. The fans just don’t increase the pressure enough to stop the higher flow from
the wind. Stairwell temperatures in the area of the fire floor decreased by over 200 ºC (392 ºF) when the
fans were activated and the WCD was in place. After the fire was suppressed the fans did a good job of
keeping the smoke out of the stair, even with a little wind effect.

The FBN 2 with a fog nozzle was flowed at approximately 160 gpm for 60 s. The fire in the living room
was suppressed but not extinguished. The temperature in the corridor decreased significantly from
above 600 ºC (1112 ºF) to under 300 ºC (572 ºF) in seconds. The FBN performed well in conjunction
with the WCD but this was very labor intensive and required practice by the crews involved to execute
properly.

316
3.12 Experiment 3E 

The twelfth experiment in the structure was located on the 3rd floor in apartment 3E. Experiment 3E
was ignited in both the bedroom and the living room in an effort have both rooms involved in fire
simultaneously to examine the impact of a FBN deployed in the bedroom (Figure 3.11-1). This
experiment also utilized a simulated wind provided by the MVU. The simulated wind was subjected to
the bedroom window at a magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The events
that took place during the experiment to accomplish the experimental objectives below are shown in
Table 3.12-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom and living room fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Control wind driven conditions with FBN
• Examine use of an aerosol generation suppression device
• Utilize portable PPV fans to control smoke movement

The results for the experiment are presented in the following sections: observations, temperature,
pressure, velocity, and discussion. Figure 3.11-2 shows the camera names and locations used in the
observation section and the instrumentation names and locations used in the temperature, pressure and
velocity sections.

317
Figure 3.12-1. 3rd floor apartment E floor plan with locations pertinent to Experiment 3E

318
Table 3.12-1: Experiment 3E Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom and living room Ignition
272 274 Front lobby Doors opened
300 303 1st floor stair A Door opened
318 324 3rd floor stair A Door opened
379 383 Bulkhead door Opened
493 494 Bedroom double Bottom right pane failed
500 501 Bedroom double Top right pane failed
525 526 Bedroom double Top left pane failed
632 632 Bedroom double Bottom left pane failed
943 943 Bedroom double FBN 2 started flowing (Smooth Bore)
1017 1017 Bedroom double FBN 2 stopped
1094 1097 Living room window Center pane vented manually
1264 1264 Living room window Top right pane failed
1301 1301 Bedroom double FBN 2 started flowing (Smooth Bore)
1347 1348 Bedroom double FBN 2 stopped
1362 1362 Living room window Aearosol generating device activated
1368 1403 Living room window WCD deploying
1513 1517 Living room window WCD removed
1551 1554 Bulkhead door Closed
1648 1680 1st floor stair A 27 inch fan on
1653 1678 Front lobby 2 - 27 inch fans activated
2155 2156 Living room window FBN 1 started flowing (Smooth Bore)
2175 2175 Living room window FBN 1 stopped
2298 2298 Experiment completed

319
Figure 3.12-2. Instrumentation and camera locations with labels.

3.12.1 Observations 

The observations are presented as a series of images captured from eight video camera locations. The
camera positions inside the building are shown in Figure 3.12-2.

Figure 3.12-3 through Figure 3.12-19 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 36 min and 30 s (2190 s) after ignition. Images were captured

320
every 4 min and at times when events had taken place that demonstrate the experimental objectives.
Each image view is labeled. The first view (LIV RM A) at the top left of each figure shows the living
room of apartment 3A; across the view are the windows. The top right view (LIV RM E) was of the
living room in apartment 3E looking across to the windows. The corridor view (CORRIDOR) shows
the corridor leading up to the open apartment doorway. Next to that, the outside lobby view (LOBBY)
shows the area just outside the lobby doors. The view (STAIR) shows the stairwell with a camera from
the floor below the fire looking up to the fire floor. Next to that is the outside view (OUTSIDE) which
shows an alternate view of the outside of the apartment from a window in another wing of the building.
The bottom left view (ROOF) shows the bulkhead door at the top of the stairwell and the bottom right
view (TOWER) shows an external view of the fire apartment.

Figure 3.12-3 shows the images at the time of ignition. At this point, the eight video views were clear
and unobstructed. As noted earlier, no thermal imagers were used for this experiment.

Figure 3.12-4 shows the images at 240 s (4 min) after ignition. At this point smoke from the fire in the
bedroom had reduced the visibility in the living room of apartment 3E and the public corridor to nearly
zero. Some smoke had leaked around the closed 3rd floor stair A door, and a small amount of smoke
was visible at the ceiling level of the living room in apartment 3A.

Figure 3.12-5 shows the images at 480 s (8 min) after ignition. By this time the lobby doors, the 1st floor
stair A door, the 3rd floor stair A door, and the bulkhead doors had been opened 206, 177, 156, and 97 s
prior, respectively. A significant volume of smoke can be seen venting from the bulkhead door. The
smoke layer had moved into the stairwell. The views of the living room in apartment 3E and the public
corridor are still obscured. The smoke in living room 3A had increased, but a clear smoke layer was not
yet visible in the living room 3A view.

Figure 3.12-6 shows the images at 600 s (10 min) after ignition. At this point three of the four panes of
the bedroom double window were broken by the fire. Flames can be seen pushing out of the window,
against the wind. The amount of smoke in the stairwell had increased and had descended below the fire
floor level. Significant amounts of smoke were flowing out of the bulkhead doorway.

Figure 3.12-7 shows the images at 720 s (12 min) after ignition. The bedroom double window in the
ignition bedroom had completely failed at 632 s. Smoke from the bulkhead door had flooded the roof
area outside the door completely obscuring the view. The smoke layer had descended about halfway
down stairway A between floor 2 and 3. The smoke in the living room of apartment 3A continued to
reduce visibility in that view. Smoke can be seen leaking to the outside from the living room windows in
apartment 3E.

Figure 3.12-8 shows the images at 930 s (15 min and 30 s) after ignition. These images show the
conditions prior to the start of suppression. FBN 2 is being put into position. The window frame has
been partially detached from the building.

Figure 3.12-9 shows the images at 960 s (16 min) after ignition. FBN 2 began flowing approximately
200 gpm in a solid stream into the ignition bedroom 17 s prior. The effects of FBN 2 are not yet visible
in any of the 8 views however.

321
Figure 3.12-10 shows the images at 1200 s (20 min) after ignition. The water flow was stopped at
1017 s. The fire in the bedroom was reduced. The center pane of apartment 3E’s living room window
was vented manually at 1097 s. The smoke layer had descended the stairwell from floor 3 to floor 2.
Flames were visible in the living room E view.

Figure 3.12-11 shows the images at 1360 s (22 min and 40 s) after ignition. The fire in the living room
of apartment 3E continued to burn. The flames from the living room can be seen exiting the window.
During the time period from 1301 s to 1348 s, the FBN 2 nozzle was flowing into the bedroom double
window opening. The stream did not have a “straight shot” down the hallway. There was no visible
suppression effect on the fire in the living room from the water that was discharged into the bedroom.

At 1362 s after ignition, an aearosol generating device was activated and lowered into the living room
window opening from above. Within seconds, the deployment of a WCD was started in an attempt to
keep the fire extinguishing agent in the living room. The agent suppressed the flames in the living room
for approximately 50 seconds. At 1420 s, the furnishings in the living room appeared to be fully
involved again. Figure 3.12-12 shows the images at 1440 s (24 min) after ignition. The WCD, which
did not fully cover the window, can be seen in the outside view. Smoke flow out of the bulkhead door
increased significantly.

Figure 3.12-13 shows the images at 1560 s (26 min) after ignition. The bulkhead door had just been
closed. The WCD was removed approximately 45 s prior. Flames from the apartment 3E doorway can
be seen in the center of the corridor view. Smokey conditions remained in apartment 3A.

Figure 3.12-14 shows the images at 1680 s (28 min) after ignition. Smoke had descended down the
stairwell view, reducing visibility in that view. The fire in the living room continued to burn.

Figure 3.12-15 shows the images at 1710 s (28 min and 30 s) after ignition. All three of the 27 inch PPV
fans were operating by this time. Two PPV fans were located outside in front of the lobby doors, and
one PPV fan was located in front of the 1st floor doorway to stair A. As a result, the stair was cleared of
smoke and additional smoke can be seen pushing around the edges of the closed bulkhead door.

Figure 3.12-16 shows the images at 1920 s (32 min) after ignition. Flames from the living room of
apartment 3E were visible in the corridor view. The stair remained clear of smoke.

Figure 3.12-17 shows the images at 2155 s (35 min and 55 s) after ignition. These images were
recorded at the time that FBN 2, began to flow water into the living room. The flow rate was
approximately 210 gpm through a smooth bore tip.

Figure 3.12-18 shows the images at 2160 s (36 min) after ignition. FBN 2 continued to flow water into
the window of living room E. Flames were no longer visible in the living room E view.

Figure 3.12-19 shows the images at 2190 s (36 min and 30 s) after ignition. These conditions are
representative of the conditions at the conclusion of the experiment. The fires in the bed room and
living room were extinguished with the FBNs. The stair was kept clear from smoke with the fans on the
1st floor. The smoke that had leaked into apartment A appeared to be at its worst at 16 minutes after

322
ignition. The smoke began to slowly dissipate after that however visible smoke remained in the living
room of apartment A at the conclusion of the experiment.

323
Figure 3.12-3. Experiment 3E, ignition.

324
Figure 3.12-4. Experiment 3E, 240s after ignition.

325
Figure 3.12-5. Experiment 3E, 480s after ignition.

326
Figure 3.12-6. Experiment 3E, 600s after ignition.

327
Figure 3.12-7. Experiment 3E, 720s after ignition.

328
Figure 3.12-8. Experiment 3E, 930s after ignition.

329
Figure 3.12-9. Experiment 3E, 960s after ignition.

330
Figure 3.12-10. Experiment 3E, 1200s after ignition.

331
Figure 3.12-11. Experiment 3E, 1360s after ignition.

332
Figure 3.12-12. Experiment 3E, 1440s after ignition.

333
Figure 3.12-13. Experiment 3E, 1560s after ignition.

334
Figure 3.12-14. Experiment 3E, 1680s after ignition.

335
Figure 3.12-15. Experiment 3E, 1710s after ignition.

336
Figure 3.12-16. Experiment 3E, 1920s after ignition.

337
Figure 3.12-17. Experiment 3E, 2155s after ignition.

338
Figure 3.12-18. Experiment 3E, 2160s after ignition.

339
Figure 3.12-19. Experiment 3E, 2190s after ignition.

340
3.12.2 Temperatures 

Figure 3.12-20 through Figure 3.12-22 provide temperature data for five measurement locations on the
fire floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.4-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.12-20 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment, 3E, in the corridor, just outside Apartment 3A in
the corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
temperature in the bedroom peaked 500 ºC (932 ºF), approximately 200 s after ignition. The
temperature in the bedroom began to decrease due to ventilation limitations. After opening the 3rd floor
door to the stair, at 320 s, the temperature in the bedroom began to increase again. After opening the
bulkhead door, the temperatures in the bedroom went up to approximately 600 ºC (1112 ºF) and stayed
at that level until the bedroom double window on the upwind side began to fail. Initially the wind driven
flow cooled the bedroom. Then as the one side of the bedroom double window self vented the fire in the
bedroom, quickly transitioned to flashover. At 600 s, temperatures throughout the fire floor are in
excess of 400 ºC (752 ºF). The use of the FBN dramatically decreased the temperature in the bedroom.
For the remainder of the experiment, the temperatures in the bedroom remained at less than 100 ºC
(212 ºF) until the end of the experiment.

Figure 3.12-21 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. These temperatures at 1.2 m (4 ft)
followed the same trends as described in Figure 3.12-20. Until the living room window was manually
vented at approximately 1100 s. The temperatures at 1.2 m (4 ft above the floor) exceed the
temperatures at 2.1 m (7 ft) above the floor in the areas outside of the apartment due to the horizontal
wind driven hot gas flows. The temperatures on the fire floor remained elevated until the PPV fans were
activated at which point the temperatures cooled to less than 100 ºC (212 ºF), with the exception of a
few brief spikes.

Figure 3.12-22 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, and at the landing
above the 7th floor with access to the roof. The sensors located at floors 4 and 6 were damaged in
previous experiments. The temperature at the bulkhead level tracked with the fire development in the
bedroom from the time that the 3rd floor stair door was opened through suppression. At approximately
1100 s through the end of the experiments, the stair temperatures were driven by the living room fire.
After the initial suppression of the fire, temperatures in the bedroom were within the tenable range for
fully protected firefighters.

341
 ) )
le) ble le) ble
o ub do u o u b d ou d ted
s ng ar g (d ed ( ing g (d ed ( ate D ved g d mple
n l i n i l n i v C o e
itio ng g fai cle wi pp f a w i p p a ct W e m i n pp o
g g
Ign peni enin enin enin uble uble flo sto w flo sto g r d on low to nt c
le o p p p o o B N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
ltip Y: : o : o : o : d : d : F : F : w : F : F : d : p : W : c Y: : fa R: F R: F xper
Mu LB S1 S3 BH BR BR B R B R LR B R B R L R L R LR B H LB S 1 L L E
1200 2192
TC BR3EE 2.1m
TC Corr3EE 2.1m
TC Corr3AE 2.1m
1000 TC CorrStair3E 2.1m
1832
TC Stair3E 2.1m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 3.12-20. 3E Temperature vs. time at 2.1 m above the floor
) le) ) le)
ble ub ble ub ed
(d ou (do (d ou (do ted d l et
o ns g il ing lear i n g ped iling ing ped tiva CD ove i n g ped omp
it i fa c w p f a w p c W em p c
Ign nin ing ing ing le le flo sto ow flo sto ce a ing r ed s on n w
flo sto ent
i p le : ope open open open doub doub F BN FBN wind FBN FBN devi plac WCD clos : fan fan o FBN FBN er im
l t Y : : : : : : : : : : : : : : Y : : : p
Mu LB S1 S3 BH BR BR B R B R LR B R B R L R L R LR B H LB S 1 L R LR E x
1000 1832
TC BR3EE 1.2m
TC Corr3EE 1.2m
TC Corr3AE 1.2m
800 TC CorrStair3E 1.2m 1472
TC Stair3E 1.2m

Temperature (F)
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 3.12-21. 3E Temperature vs. time at 1.2 m above the floor

342
 ) )
le) ble le) ble
o ub do u o u b d ou d ted
s ng ar g (d ed ( ing g (d ed ( ate D ved g d mple
n l i n i l n i v C o e
itio ng g fai cle wi pp f a w i p p a ct W e m i n pp o
g g
Ign peni enin enin enin uble uble flo sto w flo sto g r d on low to nt c
le o p p p o o B N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
ltip Y: : o : o : o : d : d : F : F : w : F : F : d : p : W : c Y: : fa R: F R: F xper
Mu LB S1 S3 BH BR BR B R B R LR B R B R L R L R LR B H LB S 1 L L E
600 1112
TC Bulkhead MidE
TC Stair 2E
TC Stair1 MidE
500 932

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 3.12-22. 3E stairwell temperature vs. time

3.12.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.12-23. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
5 Pa and only fluctuated slightly. During the period that the bulkhead door was open (380 s to 1550 s),
the pressures decreased as the stairwell became part of the flow path exhausting combustion products
out of the building. Once the bulkhead door was closed, the pressure increased in the stair. After the
PPV fans were activated, pressures in the lower portion of the stairwell increased, while the pressures in
the upper portion decreased due to cooling.

343
 ) )
le) ble le) ble
o ub do u o u b d ou d ted
s ng r g (d ed ( ing g (d ed ( ate D ved g d mple
n l i a n i l n i v C o e
itio ng g g g fai cle lowi topp w fa lowi topp act g W rem on
i n pp o
low to nt c
e Ign peni enin enin enin uble uble f s f s d
N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
l o p p p o o B
ltip Y: : o : o : o : d : d R: F R: F R: w R: F R: F R: d R: p R: W H: c BY: 1: fa R: F R: F xper
Mu LB S1 S3 BH BR BR B B L B B L L L B L S L L E
80
DP Stair 7E
DP Stair 6E
DP Stair 4E
60 DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 3.12-23. 3E stairwell pressures vs. time

3.12.4 Velocities 

Average doorway velocities from 2 doorways; the doorway from the first floor to the stairwell, and the
doorway at the bulkhead, were recorded versus time and shown in Figure 3.12-24. The three bi-
direction probes in each door were averaged to provide the bulk flow through the doors. A time history
for each probe at each measurement location is located in Appendix D – Detailed Graphs. Once the
bulkhead door was opened the velocity (negative direction equates to flow out of the doorway) increased
from zero to approximately 10 m/s (22 mph) as the bedroom flashed over and the bedroom double
window opening was cleared. While the fire in the bedroom was being actively suppressed, the velocity
of the flow out of the bulkhead went to zero. After suppression stopped and the living room window
was manually vented the flow had an average velocity of approximately 8 m/s (18 mph) until the
bulkhead door was closed. Hence the second application of water from the bedroom, the suppression
device nor the partial WCD had an impact on the flow through the stairwell and the bulkhead doorway.

344
 ) )
le) ble le) ble
o ub do u o u b d ou d ted
s ng ar g (d ed ( ing g (d ed ( ate D ved g d mple
n l i n i l n i v C o e
itio ng g fai cle wi pp f a w i p p a ct W e m i n pp o
g g
Ign peni enin enin enin uble uble flo sto w flo sto g r d on low to nt c
le o p p p o o B N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
ltip Y: : o : o : o : d : d : F : F : w : F : F : d : p : W : c Y: : fa R: F R: F xper
Mu LB S1 S3 BH BR BR B R B R LR B R B R L R L R LR B H LB S 1 L L E
3 7

0 0

-3 -7

Velocity (mph)
Velocity (m/s)

-6 -13

-9 -20

-12 -27
Exp 3E Bulkhead
Exp 3E Stair1
-15 -34
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 3.12-24. 3E Velocities vs. time

3.12.5 Discussion 

The development of this fire was similar to previous fires. The fire grew until it became ventilation
limited, the energy release and hence the temperature decreased until additional ventilation was provided
via a window opening and/or ventilation from the stair. Hazardous thermal conditions for firefighters in
full structural firefighting PPE are limited to the room of origin until a flow path through the structure is
created. In this experiment, that flow path went from the open bedroom double window through
apartment 3E, through the corridor and into the stairwell to the bulkhead doorway on the roof. As long
as that flow path existed and there was a post flashover fire in the bedroom, hazardous thermal
conditions existed on the fire floor for firefighters.

Suppressing the bedroom fire with the FBN 2 mitigated the thermal hazard within seconds. The second
portion of the fire experiment began with the manual ventilation of the living room. While the fire in the
living developed to the point where all of the fuels in the room were burning, it never created the level
of thermal hazard that the bedroom fire did. The living room was rapidly extinguished with an
externally applied water stream.

Positive pressure ventilation was used to clear and cool the stairwell in the latter stages of the fire.
Again nothing that had not been demonstrated under more severe thermal conditions in previous
experiments.

Based on the manufacturer’s guide, the aerosol generator has been found to be effective in contained
spaces [46]. In this experiment, the aerosol generator suppression device had a short term,
approximately 50 s, impact on the fire in the living room. The WCD was deployed after the suppression
device was placed via rope into the living room window in an effort to contain the extinguishing agent in

345
the fire room. However, given the “wind driven” ventilation flowing through the living room and the
limited coverage of the WCD, this was not an ideal application for this type of device.

346
3.13 Experiment 3A 

The thirteenth experiment in the structure was located on the 3rd floor in apartment 3A. Experiment 3A
was ignited in both the bedroom and the living room in an effort have both rooms involved in fire to
examine the impact of a FBN deployed in the bedroom (Figure 3.11-1). This experiment also utilized a
simulated wind provided by the MVU. The simulated wind was aimed into the bedroom window with a
magnitude of approximately 9 m/s to 11 m/s (20 mph to 25 mph). The events that took place during the
experiment to accomplish the experimental objectives below are shown in Table 3.13-1.

Experimental Objectives
• Create wind driven fire conditions from a bedroom fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Control wind driven conditions with FBN
• Utilize portable PPV fans to control smoke movement

347
Figure 3.13-1. 3rd floor apartment A floor plan with locations pertinent to Experiment 3A

348
Table 3.13-1: Experiment 3A Timeline
Start (s) End (s) Event Location Event Description
0 0 Bedroom and living room Ignition
279 283 Front lobby Right door opened
286 290 1st floor stair A Door opened
288 295 3rd floor stair A Opened and closed
329 331 3rd floor stair A Opened
381 383 Bulkhead door Opened
496 497 Bedroom double Top left pane failed
504 504 Bedroom double Top right pane failed
559 561 Bedroom double Bottom right pane failed
610 611 Bedroom double Bottom left pane failed
695 695 Bedroom double FBN 1 started flowing (Smooth Bore)
732 732 Bedroom double FBN 1 stopped
772 773 Living room window Center pane failed
838 838 Living room window Top left pane failed
860 860 Living room window Top right pane failed
877 877 Living room window Bottom left pane failed
902 902 Living room window Bottom right pane failed
922 922 Living room window FBN 1 started flowing (Smooth Bore)
932 932 Living room window FBN 1 stopped
954 954 Living room window FBN 1 started flowing (Smooth Bore)
976 976 Living room window FBN 1 stopped
1030 1044 1st floor stair A 27 inch fan added
1077 1079 Bulkhead door Closed
1200 1200 Experiment completed

349
Figure 3.13-2. Instrumentation and camera locations with labels.

3.13.1 Observations 

The observations are presented as a series of images captured from eight video camera locations. The
camera positions inside the building are shown in Figure 3.13-2.

Figure 3.13-3 through Figure 3.13-15 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 20 min (1200 s) after ignition. Images were captured every

350
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (LIV RM A) at the top left of each figure shows the living room
of apartment 3A; across the view are the windows. The top right view (LIV RM E) was of the living
room in apartment 3E looking across to the windows. The corridor view (CORRIDOR) shows the
corridor leading up to the open apartment doorway. Next to that, the outside lobby view (LOBBY)
shows the area just outside the lobby doors. The view (STAIR) shows the stairwell with a camera from
the floor below the fire looking up to the fire floor. Next to that is the outside view (OUTSIDE) which
shows an alternate view of the outside of the apartment from a window in another wing of the building.
The bottom left view (ROOF) shows the bulkhead door at the top of the stairwell and the bottom right
view (TOWER) shows an external view of the fire apartment.

Figure 3.13-3 shows the images at the time of ignition. At this point, the eight video views were clear
and unobstructed. However, the view of the living room from the previous experiment was dark due to
the fire damage it sustained from the previous experiment. The damage to the corridor from the
previous experiment was also visible.

Figure 3.13-4 shows the images at 240 s (4 min) after ignition. Visibility in the corridor and the 3A
living room was nearly zero at this point. No smoke was visible leaking past the closed 3rd floor stair A
door. No smoke was visible from the outside of the structure at this time.

Figure 3.13-5 shows the images at 480 s (8 min) after ignition. By this time the lobby doors, the 1st floor
stair A door, the 3rd floor stair A door, and the bulkhead doors had been opened 200s, 190s, 153s, and
98 s prior, respectively. Smoke was flowing up the stairwell through the open 3rd floor stair A door and
out of the open bulkhead door. Visibility was returned slightly to the corridor view.

Figure 3.13-6 shows the images at 565 s (9 min and 25 s) after ignition. Three panes of the bedroom
double window had failed and flames were being blown through the living room. Visibility in the
stairwell and in the corridor was greatly reduced. Thick black smoke was still flowing out of the open
bulkhead door.

Figure 3.13-7 shows the images at 625 s (10 min and 25 s) after ignition. The smoke and hot gases
continued to flow through the apartment and into the corridor reducing visibility but there were no
flames visible from the interior views. Flames were visible through the smoke in the stairwell view.
There was an increase in the amount of smoke flowing out of the bulkhead doorway.

Figure 3.13-8 shows the images at 690 s (11 min and 30 s) after ignition. These were the conditions
just prior to flowing the floor below nozzle. Flames were still visible at the bedroom window, visibility
in the living room was zero and hot gases were flowing through the corridor into the stairwell. The
stairwell view was lost due to increased heat on the landing below the fire floor. Smoke was leaking
heavily from the living room windows and visibility on the roof was reduced from the smoke coming
from the stairwell. Light smoke was visible coming from the lobby of the structure.

Figure 3.13-9 shows the images at 705 s (11 min and 45 s) after ignition. FBN 1 with a 1 ⅛ inch smooth
bore nozzle had been flowing approximately 200 gpm for 10 s. The flames in the bedroom were
reduced and the remaining views were very similar to the previous set.

351
Figure 3.13-10 shows the images at 720 s (12 min) after ignition. By this time the ignition bedroom
double window had completely failed and FBN 1 had been started flowing water into it for 25 s. Smoke
was leaking around the living room window. Visibility in all of the internal views is zero. Increased
smoke and steam were coming from the bulkhead doorway.

Figure 3.13-11 shows the images at 840 s (14 min) after ignition. The FBN was stopped approximately
90 s prior and the living room window began to fail 67 s prior. Flames were visible coming from the
failed living room window. Increased smoke was coming out of the lobby doorway and decreased
smoke was visible coming from the bulkhead doorway.

Figure 3.13-12 shows the images at 910 s (15 min and 10 s) after ignition. Wind driven conditions were
still present and the FBN was repositioned to the living room window. Flames were still visible in and
from the living room window. There was no longer smoke coming from the lobby doorway.

Figure 3.13-13 shows the images at 935 s (15 min and 35 s) after ignition. The FBN was flowed into the
living room for 10 s and the views show a large amount of steam coming from the structure and
decreased visibility at the roof level. There were still no interior views at this time.

Figure 3.13-14 shows the images at 960 s (16 min) after ignition. FBN 1 was flowing water into the
living room for the second time. A large volume of smoke and steam was pushing out of that window.

Figure 3.13-15 shows the images at 1200 s (20 min) after ignition. The experiment was completed by
this time. The fire had largely burned down, but visibility had not returned to the internal views. Steam
was visible flowing from the living room windows and there was no longer smoke coming from the
bulkhead door. A PPV fan was added at the 1st floor stair door to pressurize and remove the smoke from
the stairwell.

352
Figure 3.13-3. Experiment 3A, ignition.

353
Figure 3.13-4. Experiment 3A, 240s after ignition.

354
Figure 3.13-5. Experiment 3A, 480s after ignition.

355
Figure 3.13-6. Experiment 3A, 565s after ignition.

356
Figure 3.13-7. Experiment 3A, 625s after ignition.

357
Figure 3.13-8. Experiment 3A, 690s after ignition.

358
Figure 3.13-9. Experiment 3A, 705s after ignition.

359
Figure 3.13-10. Experiment 3A, 720s after ignition.

360
Figure 3.13-11. Experiment 3A, 840s after ignition.

361
Figure 3.13-12. Experiment 3A, 910s after ignition.

362
Figure 3.13-13. Experiment 3A, 935s after ignition.

363
Figure 3.13-14. Experiment 3A, 960s after ignition.

364
Figure 3.13-15. Experiment 3A, 1200s after ignition.

365
3.13.2 Temperatures 

Figure 3.13-16 through Figure 3.13-18 provide temperature data for five measurement locations on the
fire floor at two elevations and throughout the height of the stairwell. The locations of the measurement
points can be seen in Figure 3.13-2. The events that occurred during the experiment are listed on the top
of the figures with lines at the time of the event. A time history for each thermocouple at each
measurement location is located in Appendix D – Detailed Graphs.

Figure 3.13-16 shows the temperatures versus time at 2.1 m (7 ft) above the floor at 5 measurement
locations, the bedroom, just outside the fire apartment in the corridor, just outside Apartment 3E in the
corridor, just outside the A stairwell doorway, and just inside the A stairwell on the landing. The
temperature in the bedroom increased to above 300 ºC (572 ºF) before it became ventilation limited and
the temperature decreased. Opening stairwell doors maintained the temperature but opening the
bulkhead door allowed the bedroom to transition to flashover and fail the window subjected to the wind.
Once the wind was able to enter the apartment the temperature decreased briefly as the cold air flowed
in and then the temperature increased quickly to 800 ºC (1112 ºF) in the bedroom, over 500 ºC (932 ºF)
in the corridor and over 300 ºC (572 ºF) in the stairwell. The temperatures all decreased quickly to
approximately 200 ºC (392 ºF) or below after the FBN flowed for 37 s.

Figure 3.13-17 shows the temperatures at 1.2 m (4 ft) which would be more representative to where
firefighters would be operating or occupants would be evacuating. These temperatures were almost
identical to the 2.1 m (7 ft) measurements because the wind was flowing into the apartment for the
duration of the experiment. The wind was mixing all the gases and there was no layering in the corridor
or stairwell.

Figure 3.13-18 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, and at the landing
above the 7th floor with access to the roof. The temperatures in the stairwell remained near ambient until
the bedroom window failed. Prior to suppression the temperature above the fire floor, near the bulkhead
door was above 400 ºC (752 ºF). The temperature on the 2nd floor landing, below the fire floor had
reached 350 ºC (662 ºF) and the 1st floor increased to above 100 ºC (212 ºF). The temperatures
recovered slightly as the fire in the living room increased but the temperatures remained below 200 ºC
(392 ºF). The stairwell temperatures returned to ambient once the FBN flowed into the living room and
the PPV fan was activated.

366
 BR BR:
: F FB Ex
L S3 B N pe
Mu BY :o
pe
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
ltip : rig do : w p w :
win FB FB : FB S e
le h S 1 n ing 3:S B H: ub d o u i ng
p e d
in do d N N F B N N 1 : f B nt c
t
ign op op : / c l op op le f b le (do (do w f o w fl o sto fl o
a
sto n a H: c omp
itio eni eni osi eni eni aili cle ub ub aili cle wi pp wi pp dd los let
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

1200 2192
TC BR3AE 2.1m
TC Corr3AE 2.1m
TC Corr3EE 2.1m
1000 TC CorrStair3E 2.1m
1832
TC Stair3E 2.1m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 3.13-16. 3A Temperature vs. time at 2.1 m above the floor
BR BR:
: F FB Ex
S3 B N pe
Mu LBY :o
p
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
:
ltip rig en p w
do : do win pe in win : F FB : F FB S e
le h t
S 1 :
S
ing 3: H: B ub u g d d o d BN N BN N : fa BH nt co
1
ign op op / c o p o p le ble (do (do w ow f st f s t n : m
l o l
itio eni eni losi eni eni faili cle ub ub f aili cle owi pp owi pp dd clos plet o a
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

1500 2732
TC BR3AE 1.2m
TC Corr3AE 1.2m
TC Corr3EE 1.2m
1200 TC CorrStair3E 1.2m 2192
TC Stair3E 1.2m

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 3.13-17. 3A Temperature vs. time at 1.2 m above the floor

367
 BR BR:
: F FB Ex
L S3 B N pe
Mu BY :o
pe
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
ltip : rig do : w p w :
win FB FB : FB S e
le h S 1 n ing 3:S B H: ub d o u i ng
p e d
in do d N N F B N N 1 : f B nt c
t
ign op op : / c l op op le f b le (do (do w f o w fl o sto fl o
a
sto n a H: c omp
itio eni eni osi eni eni aili cle ub ub aili cle wi pp wi pp dd los let
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

500 932
TC Bulkhead MidE
TC Stair 2E
TC Stair1 MidE
400 752

Temperature (F)
Temperature (C)

300 572

200 392

100 212

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 3.13-18. 3A stairwell temperature vs. time

3.13.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.13-19. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained below
5 Pa and only fluctuated slightly. The pressure slightly increased up to 10 Pa above the fire floor after
the fire floor door was opened. Once the bulkhead door was opened this pressure dropped below zero as
the stairwell became the flow path. When the wind driven fire was at its peak the pressures in the area
of the fire floor reached as high as 20 Pa. As the fire was suppressed by the FBN the pressure also
dropped to below 10 Pa. Adding the fan at the end of the experiment increased the stairwell pressures to
an average of 20 Pa with the bulkhead door closed.

368
 BR BR:
: F FB Ex
L S3 B N pe
Mu BY :o
pe
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
ltip : rig do : w p w :
win FB FB : FB S e
le h t
S 1 :
n ing 3:S B H: ub d o u i ng
p e d
in do d N N F B N N 1 : f a BH nt co
ign op op / c op op le b le (do (do w f o w fl sto fl sto n a : c m
itio eni eni osi eni eni l faili cle ub ub ailin cle win ppe win ppe dd los plete
o o
ns ng ng ng ng ng ng ar le) le) g ar g d g d ed ed d

40
DP Stair 7E
DP Stair 6E
DP Stair 4E
30 DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

20

10

-10
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 3.13-19. 3A stairwell pressures vs. time

3.13.4 Velocities 

Average doorway velocities from 2 doors; the door at the base of the stair and the door at the bulkhead,
were recorded versus time and shown in Figure 3.13-20. The three bi-direction probes in each door
were averaged to provide the bulk flow through the doors. A time history for each probe at each
measurement location is located in Appendix D – Detailed Graphs.

The average flow through the 1st floor stair door was less than 2 m/s (4 mph). The bulk flow out of the
bulkhead door peaked around 14 m/s (31 mph) and decreased after suppression to around 5 m/s
(11 mph) and ultimately decreased to 0 m/s after the activation of the fan and the closing of the bulkhead
door.

369
 BR BR:
: F FB Ex
L S3 B N pe
Mu BY :o
pe
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
ltip : rig do : w p w :
win FB FB : FB S e
le h S 1 n ing 3:S B H: ub d o u i ng
p e d
in do d N N F B N N 1 : f B nt c
t
ign op op : / c l op op le f b le (do (do w f o w fl o sto fl o
a
sto n a H: c omp
itio eni eni osi eni eni aili cle ub ub aili cle wi pp wi pp dd los let
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

3 7

0 0

-3 -7

Velocity (mph)
Velocity (m/s)

-6 -13

-9 -20

-12 -27

-15 -34
Exp 3A Bulkhead
Exp 3A Stair1
-18 -40
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 3.13-20. 3A Velocities vs. time

3.13.5 Discussion 

Experiment 3A examined the impact of a smooth bore FBN 1 on a wind driven, multi-room apartment
fire. The fire in the bedroom achieved post-flashover (temperatures floor to ceiling in excess of 600 ºC
(1100 ºF), shortly after the windows began to self vent at 497 s. The peak temperatures were generated
by the creation of a flow path between the bedroom and the vented stairwell. The failure (self venting)
of half of the bedroom double window combined with the open door between the fire floor and the
stairwell and the open bulkhead door at the top of the stair led to the creation of a flow path.
Temperatures in the flow path between the open bedroom window and the 5th floor stair exceeded
400 ºC (752 ºF) throughout the fire floor at both the 2.1 m (7 ft) and 1.2 m (4 ft) elevations above floor
level. Heat and smoke entered the stair, resulting in poor visibility and increased temperatures both
above and below the fire floor. In addition, the ventilation condition caused by the venting and doors
resulted in a peak gas velocity of approximately 14 m/s (31 mph) out of the bulkhead doorway. The
thermal conditions created during this experiment were not conducive to a direct frontal fire attack and
firefighters remote from the fire in the stairwell would have been threatened if they were in the flow path
due to extreme temperatures.

The first pane of the bedroom window failed at 497 s and the window completely failed at 611 s. The
temperature in the center of the bedroom at 1.2 m (4 ft) at the time of window failure was approximately
400 ºC (752 ºF). The temperature in the center of the room at the time of complete window failure was
in excess of 900 ºC (1652 ºF). The partial failure of the window and the ventilation from the open stair
door on the fire floor provided the oxygen required to achieve flashover. The majority of the window
did not fail until after the room transitioned to flashover.

The FBN caused an immediate reduction of temperature on the fire floor and in the stairwell. The FBN
extinguished the fire in the bedroom, but not the fire in the living room. However, the wind driven
conditions were minimized by the application of water into the bedroom reducing temperatures to the

370
point where firefighters could enter the corridor and advance on the fire. The fire in the living room was
allowed to re-develop. The FBN was repositioned to make an attack on the living room and extinguished
the fire.

The addition of the PPV fan at the base of the stair in conjunction with closing the bulkhead door to
increase the stairwell pressure kept the smoke out of the stairwell from the knocked down fire. This
would allow crews to overhaul in a more tenable atmosphere.

371
3.14 Experiment 3G/K 

The fourteenth and last experiment in the structure involved two apartments on the 3rd floor, apartments
3G and 3K. Apartment 3G was on the up wind side of the building and was used as the ignition
apartment. The fire was ignited in three locations simultaneously; both of the bedrooms and the living
room as shown in Figure 3.14-1. In this experiment, the natural wind played a significant role. The
wind was coming from the WNW and hitting the side of the building at an average wind speed in the
range of 7 m/s to 9 m/s (15 mph to 20 mph), with gusts of up to 13 m/s (29 mph), as shown in Appendix
C – Wind Measurements and Descriptions. Later in this experiment a simulated wind provided by the
MVU was also utilized. The simulated wind was aimed into the bedroom window with a magnitude of
approximately 9 m/s to 11 m/s (20 mph to 25 mph). Apartment K was located on the down wind side of
the apartment and was loaded with fuel. The flow path on the 3rd floor was arranged to go from the most
remote bedroom in apartment G through the open apartment door into the public corridor into the open
door of apartment K to the open bedroom windows on the downwind side of the building. The events
that took place during the experiment to accomplish the experimental objectives below are shown in
Table 3.14-1.

Experimental Objectives
• Create wind driven fire conditions from an apartment fire
• Examine window failure times and conditions
• Examine building flows with different door opening configurations
• Allow wind driven fire to vent through another apartment down wind (horizontal flow path)
• Utilize portable PPV fans to control smoke movement
• Allow WCD to be subjected to fire conditions for a long period of time
• Control wind driven conditions with FBN

372
Figure 3.14-1. 3rd floor apartment G and K floor plans with locations pertinent to Experiment 3G/K

373
Table 3.14-1: Experiment 3G/K Timeline
Start (s) End (s) Event Location Event Description
0 0 All 3 rooms in G Ignition
247 250 Front lobby Left door opened
262 267 1st floor stair A Door opened
303 303 Bedroom single Bottom pane failed
352 413 Bedroom single Top pane failed
423 427 3rd floor stair A Door opened
452 455 Bulkhead door Opened
516 521 Bulkhead door Closed
713 716 Bulkhead door Opened
743 772 Front lobby 2 - 27 inch fans activated
770 785 1st floor stair A 27 inch fan activated
831 834 Bulkhead door Closed
1035 1045 Living room window Vented manually
1106 Bedroom double Top right pane failed
1141 1142 Bedroom single WCD deployed
1167 1167 Bedroom double Pane failed
1168 1168 Bedroom double Pane failed
1193 1194 Bedroom double Bottom right pane failed
1257 1263 Living room window WCD deployed
1467 1495 Bedroom double Air flow introduced
1660 1661 Middle bedroom double WCD deployed
1820 1831 Bedroom double Air flow removed
2078 2082 Bedroom double FBN 1 started flowing (Smooth Bore)
2220 2220 Bedroom double FBN 1 stopped
2644 2657 Living room window WCD peeled away for stream
2677 2683 Living room window FBN 1 started flowing (Smooth Bore)
2765 2765 Living room window FBN 1 stopped
3006 3006 Experiment completed

374
Figure 3.14-2. Instrumentation and camera locations with labels.

3.14.1 Observations 

The observations are presented as a series of images captured from eight video camera locations. The
camera positions inside the building are shown in Figure 3.14-2.

Figure 3.14-3 through Figure 3.14-28 present sets of eight images one from each camera position, at a
given time, from the time of ignition to 48 min (2880 s) after ignition. Images were captured every

375
4 min and at times when events had taken place that demonstrate the experimental objectives. Each
image view is labeled. The first view (LIV RM G) at the top left of each figure shows the living room
of apartment 3G; across the view are the windows. The top right view (CORRIDOR) shows the corridor
leading up to the open apartment doorway. The view (STAIR) shows the stairwell with a camera from
the floor below the fire looking up to the fire floor. Next to that is the view (TOWER) which shows an
exterior view of apartment 3G. The view (LIV RM K) shows the living room of apartment 3K with the
windows strait across the view. Next to that, the view (ROOF) shows the bulkhead door at the top of the
stairwell. The bottom, left view (LOBBY) shows the area just outside the lobby doors. The bottom,
right view (not labeled) shows the 3K apartment (downwind side) from ground level outside the
building.

Figure 3.14-3 shows the images at the time of ignition. At this point, the eight video views were clear
and unobstructed. The corridor view appears dark due to the limited available light.

Figure 3.14-4 shows the images at 240 s (4 min) after ignition. At this point the smoke layer in
apartment 3G had descended to the floor throughout. The smoke layer in the 3K apartment had
descended to about 1.2 m (4 ft) above the floor in the living room. No smoke is leaking into the
stairwell.

Figure 3.14-5 shows the images at 420 s (7 min) after ignition. Visibility was reduced to nearly zero in
both living room views. A few seconds earlier, the bedroom single window (located on the upwind side
of the building) finished self venting as shown in the tower view. Smoke is venting out of the open
windows in the apartment 3K bedroom windows.

Figure 3.14-6 shows the images at 480 s (8 min) after ignition. Smoke continued to vent from the open
3K bedroom windows. The door from the fire floor to the stair was opened at 427 s (7.1 min). This
allowed smoke to be pushed down the 3rd floor A stairwell and vent out of the open lobby doors.
The bulkhead door was opened at 455 s (7.6 min) and closed again at 521 s (8.7 min). Very little smoke
flowed up and out of the open bulkhead doorway.

Figure 3.14-7 shows the images at 580 s (9 min and 40 s) after ignition. Flames can be seen moving
through the living room of apartment 3G and the thick looking smoke is rolling through the corridor.
Smoke continued to vent through both the open bedroom windows in apartment 3 K and from the open
front lobby doors. Just a few seconds before these images were recorded, the fire in the most remote
bedroom in apartment 3G became underventilated to the point where flames were no longer visible.
Flames were not readily seen from the tower video. This changed was followed by large puffs of black
smoke being pushed out of the open bedroom single window against the prevailing wind.

Figure 3.14-8 shows the images at 600 s (10 min) after ignition. The puffs of smoke that had pushed out
of the upwind bedroom window of apartment 3G, quickly transformed into “balls of fire” being pushed
out of the upwind window. This phenomenon continued at a steady rate until the bulkhead door was
opened at 716 s (11.9 min) after ignition.

Figure 3.14-9 shows the images at 720 s (12 min) after ignition. The bulkhead door had been opened 4 s
prior, but no smoke was pushing from it. Visibility was still obscured in all the internal views. Large

376
volumes of smoke were flowing out of the open bedroom windows in apartment 3K and out of the open
lobby doors. A small amount of smoke was still being pushed out of the open window in apartment 3G.

Figure 3.14-10 shows the images at 780 s (13 min) after ignition. Approximately 10 s before these
images were recorded, the two 27 inch PPV in front of the lobby doors were started. This reduced the
amount of smoke coming out of the lobby doorway. Efforts were being made to start the PPV fan that
was located in front of the doorway to stair A, at this time.

Figure 3.14-11 shows the images at 960 s (16 min) after ignition. The bulkhead door had been closed
126 s prior and the 27 inch fan at the 1st floor stair A door had been operating for approximately
3 minutes. Visibility had improved slightly in the stairwell view as the smoke density decreased. No
smoke was venting through the lobby doors at this point. The smoke venting out of the open downwind
bedroom windows in apartment 3K appeared to have gotten lighter in color and had increased in
volume.

Figure 3.14-12 shows the images at 1120 s (18 min and 40 s) after ignition. A few seconds earlier the
bedroom double window in apartment 3G failed, resulting in significant flame extension out of the
double window as shown in the tower view. This reduced the amount of smoke that was being vented
on the downwind side through apartment 3K.

Figure 3.14-13 shows the images at 1200 s (20 min) after ignition. In an effort to maintain flow into the
public corridor and apartment 3K, a WCD had been placed over the bedroom single window in the 3G
apartment bedroom furthest from the apartment exit 57 s prior. The flames pushing from that bedroom’s
double window intensified significantly in part because of the smoke being driven from the living room
into the bedroom.

Figure 3.14-14 shows the images at 1440 s (24 min) after ignition. A WCD had been placed over the
living room window 164 s prior. This closed off both of the upwind openings in apartment 3G. The
amount of flame extension from the bedroom double window was reduced and fire was visible in the
apartment 3G living room. However no smoke was flowing out of the down wind apartment.

Figure 3.14-15 shows the images at 1520 s (25 min and 20 s) after ignition. At this time the MVU had
been moved into position and was being adjusted to blow the heat back through apartment 3G. This
would challenge the WCDs that had been deployed and potentially generate heat behind them.

Figure 3.14-16 shows the images at 1680 s (28 min) after ignition. A WCD had been placed over the
second bedroom double window 19 s prior in an effort to maintain a closed vent in that room. The stair
was kept clear. A small amount of smoke was venting from the open bedroom windows in apartment
3K.

Figure 3.14-17 shows the images at 1775 s (29 min and 35 s) after ignition. A significant change in the
conditions in the living room in apartment 3G began 5 s prior. Large volume of billowing black smoke
pushed around the WCD. Flames from the living room window followed. The smoke vented on the
downwind side of the building through apartment 3K became darker in color and increased in volume.

377
Figure 3.14-18 shows the images at 1820 s (30 min and 20 s) after ignition. Flames coming up from
behind the WCD over the living room window were being driven by the wind into the 4th floor, where
safety crews were positioned. The amount of smoke venting from apartment 3K had increased
significantly and had gotten very dark in color. Given the concern that the conditions on 4th floor were
decaying, the simulated wind was removed from the bedroom double window in apartment 3G.

Figure 3.14-19 shows the images at 1920 s (32 min) after ignition. The MVU was removed
approximately 90 s prior. The flow quickly reversed, with flames extending out of the bedroom double
window for a brief period. Flame extension from behind the living room WCD stopped and the smoke
that was venting from the 3K bedroom windows stopped as well.

Figure 3.14-20 shows the images at 2070 s (34 min and 30 s) after ignition. These images were captured
to show the conditions just before suppression was started with a FBN.

Figure 3.14-21 shows the images at 2100 s (35 min) after ignition. Suppression with FBN 1 began at
2082 s after ignition. The 1-1/8th inch, smooth bore nozzle tip was used, flowing approximately 200
gpm. Water flow continued until 2200 s. Copious amounts of white vapor flowed out of the bedroom
double window opening. A very small amount of smoke was venting from the open bedroom windows
in apartment 3K. The stair was kept clear by the PPV fans.

Figure 3.14-22 shows the images at 2160 s (36 min) after ignition. Suppression continued with no
visible changes from the previous images.

Figure 3.14-23 shows the images at 2260 s (37 min and 40 s) after ignition. Suppression stopped at
2220 s. Flames were present in the living room in apartment 3G. The glow could be seen in both the
living room view and through the WCD in the tower view. Smoke began to vent from apartment 3K
again.

Figure 3.14-24 shows the images at 2400 s (40 min) after ignition. No ventilation or suppression
conditions were changed between 2220 s and 2400 s. The flames in the living room were no longer
visible in the living room view. Smoke continued to vent from behind the WCD over the living room
window opening in apartment 3G and the from the open bedroom windows in apartment 3K.

Figure 3.14-25 shows the images at 2640 s (44 min) after ignition. A portion of the WCD over the
living room opening was being moved to in order to use a FBN to flow water directly into the living
room. As air was allowed into the living room, the flames could be seen again.

Figure 3.14-26 shows the images at 2685 s (44 min and 45 s) after ignition. A few seconds before these
images were recorded, FBN 1, with the same smooth bore nozzle used earlier in the experiment began to
flow approximately 200 gpm of water into the living room. The flames in the living room were quickly
extinguished.

Figure 3.14-27 shows the images at 2715 s (45 min and 15 s) after ignition. FBN 1 was still applying
water into the living room. Significant volumes of white vapor poured out of the bedroom double
window. White vapor could also be seen venting out of apartment 3K.

378
Figure 3.14-28 shows the images at 2880 s (48 min) after ignition. FBN 1 flowed in to the living room
for approximately 90 s. These images were recorded after the suppression effort was completed. Vapor
was still coming out of the upwind bedroom. A very modest amount of smoke was still being vented
from the open windows in the downwind apartment. The PPV fans continued to keep the stairwell free
from smoke.

379
Figure 3.14-3. Experiment 3G/K, ignition.

380
Figure 3.14-4. Experiment 3G/K, 240s after ignition.

381
Figure 3.14-5. Experiment 3G/K, 420s after ignition.

382
Figure 3.14-6. Experiment 3G/K, 480s after ignition.

383
Figure 3.14-7. Experiment 3G/K, 580s after ignition.

384
Figure 3.14-8. Experiment 3G/K, 600s after ignition.

385
Figure 3.14-9. Experiment 3G/K, 720s after ignition.

386
Figure 3.14-10. Experiment 3G/K, 780s after ignition.

387
Figure 3.14-11. Experiment 3G/K, 960s after ignition.

388
Figure 3.14-12. Experiment 3G/K, 1120s after ignition.

389
Figure 3.14-13. Experiment 3G/K, 1200s after ignition.

390
Figure 3.14-14. Experiment 3G/K, 1440s after ignition.

391
Figure 3.14-15. Experiment 3G/K, 1520s after ignition.

392
Figure 3.14-16. Experiment 3G/K, 1680s after ignition.

393
Figure 3.14-17. Experiment 3G/K, 1775s after ignition.

394
Figure 3.14-18. Experiment 3G/K, 1820s after ignition.

395
Figure 3.14-19. Experiment 3G/K, 1920s after ignition.

396
Figure 3.14-20. Experiment 3G/K, 2070s after ignition.

397
Figure 3.14-21. Experiment 3G/K, 2100s after ignition.

398
Figure 3.14-22. Experiment 3G/K, 2160s after ignition.

399
Figure 3.14-23. Experiment 3G/K, 2260s after ignition.

400
Figure 3.14-24. Experiment 3G/K, 2400s after ignition.

401
Figure 3.14-25. Experiment 3G/K, 2640s after ignition.

402
Figure 3.14-26. Experiment 3G/K, 2685s after ignition.

403
Figure 3.14-27. Experiment 3G/K, 2715s after ignition.

404
Figure 3.14-28. Experiment 3G/K, 2880s after ignition.

405
3.14.2 Temperatures 

Figure 3.14-29 through Figure 3.14-31 provide temperature data for several measurement locations on
the fire floor at two elevations and throughout the height of the stairwell. The specific locations are
identified with the description of each graph. The locations of the measurement positions can be seen in
Figure 3.14-2. The events that occurred during the experiment are listed on the top of the figures with
lines at the time of the event. A time history for each thermocouple at each measurement location is
located in Appendix D – Detailed Graphs.

Figure 3.14-29 shows the temperatures versus time at 3 measurement locations; the bedroom, just
outside the fire apartment in the corridor, just outside the A stairwell doorway. In this graph the
bedroom data was measured at 1.8 m (6 ft) above the floor and the corridor positions were measured at
2.1 m (7 ft) above the floor. The bedroom temperature time history was very similar to the previous
experiments in that the temperature increased to a peak range of 400 ºC (752 ºF) to 500 ºC (932 ºF) and
then became ventilation limited. As portions of the upwind bedroom single window failed the
temperature decreased, again similar to previous experiments. A difference here was that once the
bedroom single window failed, a horizontal flow path existed from the bedroom in apartment 3G to
apartment 3K; however this did not provide enough ventilation to enable the bedroom to transition to
flashover. In previous cases, opening the bulkhead door led to flashover. In this case the natural wind
impacted the entire building. The predominant flow was down and out the front lobby. Closing the
bulkhead coincided with the transition to flashover. Post flashover, the temperatures in the bedroom
began to follow previous experiment trends again, with the temperatures remaining high until
suppression.

The temperatures in the corridor reached peak temperatures in the range of 300 ºC (572 ºF) to 400 ºC
(752 ºF) following flashover in the bedroom. These temperatures remained fairly stable until the living
room window was vented. This event was followed by the failure of the bedroom double window. The
resulting wind flow reduced the temperatures in the corridor to less than 100 ºC (212 ºF). The
temperatures in the corridor increased to approximately 200 ºC (392 ºF) once WCDs were deployed to
block the natural wind and the simulated wind from the MVU turned the flow around and pushed the hot
gases back through the apartment. Once the living room transitioned to flashover, the temperatures in
the corridor exceeded 600 ºC (1112 ºF). After removing the simulated wind, the corridor temperatures
decreased to approximately 200 ºC (392 ºF). Use of the FBN in the bedroom resulted in a temperature
increase in the corridor. suppression in the living room with the FBN results in all temperatures on the
fire floor decreasing to approximately 100 ºC (212 ºF).

Figure 3.14-30 shows the temperatures versus time at 3 measurement locations; the bedroom, just
outside the fire apartment in the corridor, and just outside the A stairwell doorway at 1.2 m (4 ft) which
would be more representative of where firefighters would be operating or occupants would be
evacuating. The bedroom and corridor positions track well with the temperatures closer to the ceiling.

Figure 3.14-31 shows the temperatures 1.2 m (4 ft) above the landing on floors 1, 2, 6 and at the
bulkhead landing above the 7th floor with access to the roof. The highest temperatures occurred when
the bulkhead door was closed and they occurred at the two floor levels below the fire floor since prior to
the failure of multiple windows in the fire apartment, the predominant flow path was down the stair way
and out the front doors. After the bulkhead door was opened, and the PPV fans were turned on, all of

406
the temperatures decreased to less than 25 ºC (77 ºF) for the remainder of the experiment. This
experiment was different from the other experiments with regard to the bulkhead and ventilation in that
the wind was pushing fresh air into the bulkhead door. As a result it did not function as a hot gas
exhaust vent but as a cool air intake port.

) )
le e) le
ng u bl ou b
s i o d
g g D( r ow D (d d ( d
ns ning g
in ear ilin ilin C lea CD ir fl WC ved ing ppe
D
C ng e
io il
it pe g fa cl g g g on
a
f fa W c W a g o ow to W owi topp
n o n n in d in d w le g le g g i n m fl s g fl s
Ig t ni le le ni n e n s on e o b in b in in c re in
le lef e ing ing e pe los pe fan n los ind ou lac ou lac uc pla w BN BN ov N N
l tip Y: : op : s : s : op : o : c : o Y: : fa : c : w : d : p : d : p rod 2: flo : F : F : m : FB : FB
u t r
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R
1200 2192
TC BR3GW 1.8m
TC Corr3GW 2.1m
TC CorrStair3W 2.1m
1000 TC BR3KW 1.8m 1832

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.14-29. 3G/K Temperature vs. time at 2.1 m above the floor
) )
le e) le
ng u bl o ub
s i o d
g g D( r ow D (d d ( d
o ns ning il
g
in ear ailin ilin C lea CD ir fl WC ved ing ppe
D
C i n g pe
i i
t p e g f a cl g g g n f f a W c W a g o ow to W w p
n o g n l g o st o
Ig t o nin le le nin nin ed nin s on ed ow ble in ble ing ing ci rem f s in fl
p le lef pe ing ing pe pe los pe fan an los ind ou lac ou lac duc pla ow BN BN ov B N B N
ti : o s s o o c o : f c w d p d p
: : : : tro 2 r f : : : l F F m F :F
ul Y : : : : : : : Y : : : : :
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R
1200 2192
TC BR3GW 1.2m
TC Corr3GW 1.2m
TC CorrStair3W 1.2m
1000 TC BR3KW 1.2m 1832
Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.14-30. 3G/K Temperature vs. time at 1.2 m above the floor

407
 )
e) e) le
n gl u bl o ub
i o d
g (s w D (d (
n s i ng
in g r
a i l in iling CD ar CD flo C ed ng ped D
C ng e
d
l e r W i W owi topp
n itio pen g fai cle g g g o n
w
fa fa W cl W ai g ov ow top
g g n m l g
n n e
Ig t o nin le le nin ni ed ni s on ed o bl in bl in in c re f s e g i in f l s
le lef e ing ing e pe los pe fan n los ind ou lac ou lac uc pla w BN BN ov N N
l tip Y: : op : s : s : op : o : c : o Y: : fa : c : w : d : p : d : p rod 2: flo : F : F : m : FB : FB
u t r
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R
250 482
TC Bulkhead MidW
TC Stair 6W
TC Stair 2W
TC Stair1 MidW
200 392

Temperature (F)
Temperature (C)

150 302

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.14-31. 3G/K stairwell temperature vs. time

3.14.3 Pressures 

Average pressure measurements in the stairwell are shown in Figure 3.14-32. The differential pressure
measurements were made at 1.2 m (4 ft) on the stairwell landing of every floor and referenced to an
apartment on the same floor. With all of the doors to the stairwell closed the pressures remained within
± 10 Pa. Opening the lobby and stair door on the 1st floor (on the downwind side of the building)
reduced the pressures in the stairway. Opening the stair door to the fire floor resulted in increased
pressures above the fire floor, while the pressures stayed low below the fire floor. Once the PPV fans
were used, the pressures increased in the stairwell and for each of the positions remained relatively
steady through the remainder of the experiment. The exception was the pressure measurement on the
fire floor which had swings between -30 Pa and 25 Pa, due in part to the multiple openings on the fire
floor, and the methods attempted to control the wind flows.

408
 )
e) e) le
n gl u bl o ub
i o d
g (s w D (d (
n s i ng
in g r
a i l in iling CD ar CD flo C ed ng ped D
C ng e
d
l e r W i W owi topp
n itio pen g fai cle g g g o n
w
fa fa W cl W ai g ov ow top
g g n m l g
n n e
Ig t o nin le le nin ni ed ni s on ed o bl in bl in in c re f s e g i in f sl
le lef e ing ing e pe los pe fan n los ind ou lac ou lac uc pla w BN BN ov N N
l tip Y: : op : s : s : op : o : c : o Y: : fa : c : w : d : p : d : p rod 2: flo : F : F : m : FB : FB
u t r
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR LR LR L R
40
DP Stair 7W
DP Stair 6W
DP Stair 5W
DP Stair 4W
DP Stair 3W
20 DP Stair 2W
DP Stair 1W
Pressure (Pa)

-20

-40
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 3.14-32. 3G/K stairwell pressures vs. time

3.14.4 Discussion 

This experiment was unique in this series of experiments, in that it had a natural wind condition that had
an effect on the entire structure. Yet in many respects the results of this experiment were similar to the
results of the experiments with a simulated wind that was only imposed on one window.
The fire developed and became fuel rich or ventilation limited. After the window self vented and the
bedroom transition to post flashover conditions, the resulting thermal conditions were not conducive to a
direct frontal fire attack and firefighters remote from the fire in the corridor would have been threatened
if they were in the flow path due to extreme temperatures.

The wind completely changed the flow path in the stairwell for this experiment. No hot gas or smoke
was vented out of the bulkhead doorway on the roof in this experiment. This was counter to the
experience in the other experiments, where opening the bulkhead door resulted in flashover. In this
experiment, the wind pushed the flow down the stairs and out the front lobby doors, which were on the
downwind side of the building.

The self venting of the bedroom double window demonstrated how building arrangement and wind
conditions may exist such that, by venting other windows in the fire aparment, the wind driven fire can
be vented out of the building, reducing the hazardous conditions in the public corridor and the stair. The
concern here would be the lack of control if the wind were to change abruptly.

PPV fans were used successfully to stop the downward flow and clear the stairwell. In this experiment,
as is generally true, door control and co-ordination were the keys to the successful use of PPV.

409
In this experiment, the WCDs were used more as tools to “replace the vented windows” for the
experiment rather than for a fire fighting tactic. This experiment did create a significant thermal
exposure for the WCD over the bedroom and living room windows, including exposure to direct flame
impingement. In both cases, the WCDs continued to maintain their functionality throughout the
experiment and blocked the wind.

The FBN 1 was discharged into both the bedroom and the living room and suppressed the fires in a short
period of time, resulting in conditions in the corridor which would have allowed firefighters in full PPE
to have safe access to the fire floor.

The fire did not spread to Apartment 3K. The peak temperature measurements in the most remote
bedroom, furthest distance from ignition on the flow path at 1.2 m (4 ft) above the floor, were
approximately 150 ºC (302 ºF) during the natural wind portion of the experiment and were
approximately 200 ºC (392 ºF) during the simulated wind portion of the experiment. Clearly untenable
conditions for unprotected building occupants and potentially hazardous conditions for protected fire
fighters. Removing the air reduced the temperatures in this remote room to less than 50 ºC (122 ºF)
within seconds. During the suppression of the living room fire, the temperatures in the bedroom
increased to nearly 100 ºC (212 ºF), indicative of steam flowing toward the open windows in the
bedroom.

410
4 Discussion 

The overall objective of this study was to improve the safety of firefighters and building occupants. If
fire conditions in the building are such that fire fighters can not safely do their job, clearly the options
available to the occupants will be limited and they will be at great risk. A means to evaluate the
effectiveness of PPV, WCD and/or externally applied water is to examine the tactic’s capability to
mitigrate the hazard and provide survivable conditions in the stairwell and corridor for firefighters in full
PPE. To assess these conditions and the effectiveness of the tactics, a brief discussion of firefighter
tenability is required.

The fire environment provides many challenges; reduced visibility, toxic combustion products, thermal
energy and potential for structural collapse. Since a fire resistive building was used in this study, the
challenges are limited to the first three items listed above. Firefighters may be equipped to deal with
these challenges with thermal imagers to improve visibility, self contained breathing apparatus (SCBA)
to protect against the combustion products for a limited time, and PPE to absorb thermal energy for a
limited time. How long the PPE can protect the firefighter from a thermal injury is based on many
factors; thermal storage capacity of the gear, condition of the PPE, moisture content of the PPE, fit of
the PPE, insulation under the PPE (station uniform), and the rate of energy (heat) transfer to the PPE.

The rate of heat transfer is of predominate interest in examining the results of these experiments.
Unfortunately there is no single measure, as the heat is transferred in different ways. The two principle
means of heat transfer of concern here are convection which is a function of the gas temperature and gas
velocity and radiation which is a function of the gas temperature and the composition of the fire gases.
In the wind driven tests post-window failure, the majority of the heat transfer, even in positions near the
floor is a combination of convection and radiation. In other words, hot fire gases flowing over a
firefighter and hot gases and/or hot surfaces in the compartment radiating energy to the firefighter. One
of the more extreme examples of this combination of convective and radiative heat transfer is direct
flame impingement.

In the ideal situation PPE is designed to protect a firefighter from temperatures up to 260 ºC (500 ºF) for
5 minutes [47]. However, that does not account for the heat flux that the PPE is exposed to along with
the elevated temperature, nor the increased rate of heat transfer due to the relatively high velocity
convective flows. Just prior to flashover, the heat flux from the upper hot gas layer to the floor
approaches 20 kW/m2. Post-flashover heat flux conditions range from 60 kW/m2 to more than
160 kW/m2. Based on previous research at NIST, a firefighter in full PPE, exposed to temperatures in
excess of 260 ºC (500 ºF) combined with heat fluxes in excess of 20 kW/m2 suggest that survival time
would be limited to less than 30 s [48, 49, 50]. In these experiments, during a wind driven condition,
temperatures on the order of at least 400 ºC (752 ºF) occurred in the corridor and portions, prior to using
one of the mitigating tactics, indicating that conditions in the corridor and portions of the stair may not
be not survivable for a firefighter in full PPE.

The first two experiments (7G and 7G2) were conducted without an imposed wind condition to examine
the development of the fire and the impact of PPV. The next eleven experiments were conducted with
simulated wind of 9 m/s to 11 m/s (20 mph to 25 mph), typically imposed on a window in the room of
fire origin. These experiments were used to develop wind driven fire conditions in the structure and
then examine different means of controlling the flow path through the structure, i.e. doors, fans and

411
WCDs, to mitigate the hazard. Suppression tactics using appliances from the floor below were also
examined in these experiments.

The last experiment (3GK) was focused mainly on horizontal flow path from the upwind side of the
building to the downwind side. For this experiment, natural wind was available and was used. The
MVU driven flow was also used in the latter part of in this experiment.

Each of the experiments had a list of objectives that were to be examined. In this section, those
objectives will be examined across the experimental series and examples will be given focusing on the
specific tactic or tactics used. The effectiveness will be examined mainly in terms of temperature, but
velocity and pressure will also be considered as the conditions are related and references will be made
back to the specific figures in the experiment.

Wind driven fire conditions – The wind driven condition can be described as hot gases or flames
flowing horizontally out of the room of fire origin. The wind driven fire condition has been described as
a “blow torch”. For our purposes, a wind driven fire condition was when the fire gases were well mixed
and of equally high temperature from the floor to the ceiling, on the order of at least 400 ºC (752 ºF).
This condition was generated in eleven of the fourteen experiments. The three exceptions were
experiments 7G, 7G2 and 5E. Experiments 7G and 7G2 were conducted with no imposed external
wind, hence no wind driven fire condition. Experiment 5E had an imposed wind on the upwind window
of the fire room, however a WCD was deployed prior to window failure, as a result there was no wind
driven fire condition in Apartment 5E.

An example of the uniform floor to ceiling temperatures is given below in Figure 4-1. These
temperature measurements show the first 400 s after ignition at the corridor location outside of
apartment 7K. The fire was started in the bedroom, approximately 15 m (50 ft) away from the
measurement location. The temperatures increased gradually and they were stratified with the hotter
gases near the ceiling. Once the flow path was completed by opening of the bulkhead door on the roof,
temperatures from floor to ceiling increased to approximately 600 ºC (1112 ºF) within seconds. These
hot gases which were typically flowing with velocities of 5 m/s (11 mph) or more are lethal to
firefighters in full PPE. There is no safe position to “get low and let it blow” since the hot gases are well
mixed and nearly uniform in temperature from floor to ceiling. Hence, it is critical that firefighters do
not get caught in the flow path, downwind of a wind driven fire.

412
 B R S 7: o
: do
BR
: Ig S 7: uble peni ng S7: S7: BH:
ope
LB
ope BH: cl Y: ope S7: cl
ni ti o cl o s fail i 0.08 cl os ning ning osed ning o se
n ed ng m ed d
1000 1832
TC Corr7KW C
TC Corr7KW 2.1m
TC Corr7KW 1.8m
800 TC Corr7KW 1.5m 1472
TC Corr7KW 1.2m
TC Corr7KW 0.9m
TC Corr7KW 0.6m

Temperature (F)
Temperature (C)

600 1112

400 752

200 392

0 32
0 50 100 150 200 250 300 350 400
Time (s)
Figure 4-1. Temperature vs time from the vertical thermocouple array at the corridor position outside of apartment 7K.

If the flow path is favorable to an interior attack, the heat will be forced out of the structure by the wind
and the cool air will be at the firefighters back as they approach the fire. Conditions adverse to an
interior attack will be developed when the flow of air is into the structure through the fire room, such
that heat is forced into the structure by the wind or the building ventilation path. In these cases, the
flames or hot gases will meet the firefighters head on. In order for this condition to exist a number of
conditions have to be met; a ventilation limited fire, which generated significant quantities of excess fuel
in the form of combustion products, additional fuel down wind of the room of origin and most
importantly a flow path. Without a flow path, the wind driven fire condition can not occur.

4.1 Door Control 

Door control is the most basic means to control the flow path in the building. Figure 4.1-1 and Figure
4.1-2 show the impact of the opening and closing doors in the flow path. Figure 4.1-1 shows a portion of
the temperature time history from Experiment 5E2. Time “zero” equals the beginning of window
failure. At this point in time a flow path exists through the apartment into the stairwell and up to the
open bulkhead door. At approximately the 70 s mark, the majority of the living room window had self
vented and the living room has transitioned to a post-flashover condition with flames floor to ceiling.
Post-flashover, all of the temperature measurements at 1.2 m (4 ft) above the floor along the flow path
on the fire floor were in excess of 400 ºC (752 ºF). Once the bulkhead door was closed, the wind driven
condition stopped. This experiment was challenging because the living room was still able to get air
from other open windows in the apartment, therefore the living room continued to burn but the flow of
hot gases into the corridor and stairwell slowed down. Just prior to the bulkhead door being closed, the
velocity of the hot gas flowing out of the bulkhead was approximately 10 m/s (22 mph) and the
temperature at the bulkhead was over 1000 ºC (1832 ºF), Figure 3.7-15 and Figure 3.7-17. Within
seconds of closing the bulkhead door, the velocity at the bulkhead was reduced to 0 m/s and within 90 s
the temperature at the bulkhead had decreased to approximately 200 ºC (392 ºF).

413
However, the thermal conditions on the fire floor were decreased slowly due to the continued burning in
the living room. PPV fans were turned on at the lobby doors, however, with the bulkhead door closed
and the imposed wind still blowing into the apartment, the pressures in the stairwell at the fire floor were
in the 50 Pa to 60 Pa range, more than the fan at the position could overcome, Figure 3.7-16.

A FBN 1, flowing approximately 160 gpm with a narrow fog pattern, was discharged into the living
room window opening. After a few seconds of adjustment, water flowed into the living room and
temperatures on the fire floor were reduced to approximately 100 ºC (212 ºF) within 70 s.

These results demonstrated that closing a door to interrupt the flow path, even a door remote from the
fire floor, in this case the bulkhead door, can stop the wind driven flow condition. However, the further
the door is away from the fire apartment the larger the area that may still be filled with hot gases (stored
fuel and energy). In this experiment, applying water from the floor below was needed to control the fire
in the living room and dissipate the heat on the fire floor, thus rendering the area safe for further
firefighter operations.
LR: r
win l ea ed ved oved lowing oved ped ening
do w ed d ow c on clos mo m f m top
N s : left o
p
f ail c l o s
w in : f an : left : f an : f an FBN : f an F B
ing H: R : B Y B Y B Y B Y R: B Y R: B Y
B L L L L L L L L L
1200 2192
TC Corr5EE 1.2m
TC Corr5AE 1.2m
TC CorrStair5E 1.2m
1000 TC Stair5E 1.2m 1832

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
-50 0 50 100 150 200 250 300 350 400
Time (s)
Figure 4.1-1. Temperature vs time, several locations on the fire floor at 1.2 m (4 ft) above the floor, experiment 5E2. T=0 is the
time of the beginning of window failure.

Experiment 5G2 demonstrated the impact that opening doors to create a flow path can have on the
development of wind driven conditions. Figure 4.1-2 shows a portion of the temperature time history
from experiment 5G2, where time zero is defined at the time that the door to the stairwell on the fire
floor was opened. By opening that door, a flow path was developed that went from the open living room
window through the stairwell and down through the open 1st floor stair door to the outside. As a result
this action increased the temperature in the stair on the fire floor and at adjacent floors above and below
the fire floor by approximately 100 ºC (212 ºF), Figure 3.11-15. The hot gas velocity flowing out of the
apartment and into the stair were approximately 2 m/s (4 mph) as a result of the stair door opening on
the fire floor and flow out of the stairwell at the 1st floor was about 1 m/s (2 mph).

414
Approximately 20 seconds later the bulkhead door was opened. This gave the buoyant hot gases a flow
path that was up and out. Within 30 s of the opening of the bulkhead door, temperatures in the flow
path on the fire floor had exceeded 600 ºC (1112 ºF). Temperatures in the stairwell above the fire floor
increased to almost 600 ºC (1112 ºF) at the 6th floor landing and 400 ºC (752 ºF), Figure 3.11-15.
The velocity of the hot gases flowing out of the apartment and into the stairwell increased to
approximately 6 m/s (13 mph), Figure 3.11-17.

S5: openi BH: openi

ng ng
1000 1832
TC Corr5GW 1.2m
TC Corr5KW 1.2m
TC CorrStair5W 1.2m
TC Stair5W 1.2m
800 1472

Temperature (F)
Temperature (C)

600 1112

400 752

200 392

0 32
-20 0 20 40 60 80 100 120 140 160
Time (s)
Figure 4.1-2. 5G2 Temperature vs time, several locations on the fire floor at 1.2 m (4 ft) above the floor. T=0 is the time of opening
the 5th floor stairwell door.

4.2 PPV fans to control smoke movement 

As documented in several previous studies [39,40], PPV fans can be used effectively in high rise
buildings to pressurize stairwells to protect them from the inflow of combustion products based on the
natural pressures that are generated from fire. Figure 4.2-1 and Figure 4.2-2 provide examples of the
impact that a single 27 inch PPV fan can have on the temperatures in the stairwell from a non-wind
driven fire on the 7th floor, during experiments 7G and 7G2 respectively. In each case, time zero equals
the time that the fan team had the fan in place, started and operating at maximum rpm. With cool air
from outside the structure being forced into the stairwell, the temperatures decreased to near ambient
temperatures on the fire floor level within 60 s of activation.

It has also been noted in many of the wind driven experiments that one or two 27 inch PPV fans could
not exceed the pressures generated by a wind driven fire condition. In experiment G2, with one fan
operating on the 1st floor and one fan operating on the 5th floor with the bulkhead door closed and the
fire floor door open stairwell pressures were in the range of 15 Pa to 30 Pa.

In comparison, experiments 5A and 5G had imposed wind conditions. With the 1st floor and 5th floor
(fire floor) doors open to the stairwell and the bulkhead door closed, the differential pressures at the fire
floor and up exceeded 40 Pa.

415
The key to successful use of PPV fans is to mitigate the wind driven fire condition via door control or
other tactics. Then the PPVs can be used to clear the stair and then pressurize the stairwell to provide a
safe working environment. Even if the PPV fans, when used alone, could not reverse the flow of a wind
driven fire, they always improved conditions in the stairwell.

S1: fan o BH: clos d

n ed S7: close
200 392
TC Bulkhead MidW
TC Stair7 MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
150 302
TC Stair1 MidW

Temperature (F)
Temperature (C)

100 212

50 122

0 32
-20 0 20 40 60 80 100 120 140 160 180 200
Time (s)
Figure 4.2-1. 7G Stairwell temperature vs time. T=0 is the time that the fan was operating at the 1st floor doorway to the stairwell.

S1: fan o g
n BH openin
250 482
TC Bulkhead MidW
TC Stair7 MidW
TC Stair 6W
200 TC Stair 4W 392
TC Stair 2W
TC Stair1 MidW

Temperature (F)
Temperature (C)

150 302

100 212

50 122

0 32
-20 0 20 40 60 80 100 120 140
Time (s)
Figure 4.2-2. 7G2 Stairwell temperature vs time. T=0 is the time that the fan was operating at the 1st floor doorway to the
stairwell.

416
4.3 WCD to control wind driven conditions 

As discussed earlier, if the flow path can be interrupted, wind driven fire conditions cannot exist. All of
the WCDs used in this study and in previous studies [1, 40] demonstrated that they were effective
interrupting the flow path and were able to withstand the fire conditions to which they were exposed
without failing. Once the WCD was deployed and covered the window completely, the flow path was
interrupted, the velocities in the flow path decreased, and the wind driven condition stopped. As shown
below in Figure 4.3-1, the firefighters would then be faced with the residual fire and stored energy (hot
gases and hot walls, ceiling and floor) and fuels (smoke and remaining furnishings or interior finish) in
the structure.
Figure 4.3-1 and Figure 4.3-2 demonstrate the impact that WCDs can have on a post flashover wind
driven fire environment. Time zero is the point in time that the WCD was in place for each experiment.
In both figures is can be seen that the temperatures began to decrease at locations remote from the room
of origin almost immediately. Within 120 s of deployment, the temperatures had decreased by at least
50% from the temperatures at the time of WCD deployment.

BR: WC
D in pla S 7: ope
ce (d ning 0.0 n ed
ouble) 8m S 7: o pe S7: clos
1200 2192
TC BR7AE 1.2m
TC Corr7AE 1.2m
TC Corr7EE 1.2m
1000 TC CorrStair7E 1.2m 1832
TC Stair7E 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
-20 0 20 40 60 80 100 120 140 160 180 200
Time (s)
Figure 4.3-1. 7A Temperature vs time at 1.2 m above the floor. T=0, is the time that the WCD was deployed.

417
BR: pla
cing WC S
D (doub 7: opening 0.0 S 7: ope S7: clos S1: fan
le) 8m n ed on
1200 2192
TC BR7KW 1.2m
TC Corr7KW 1.2m
TC Corr7GW 1.2m
1000 TC CorrStair7W 1.2m 1832
TC Stair7W 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
-20 0 20 40 60 80 100 120 140 160 180 200
Time (s)
Figure 4.3-2. 7K Temperature vs time at 1.2 m (4 ft) above the floor. T=0, is the time that the WCD was deployed.

4.4 FBN to control wind driven conditons 

The results of the experiments with the FBN demonstrated that a relatively small amount of water flow,
125 gpm to 200 gpm had a significant effect on suppressing the fire in the room of origin and reducing
the temperatures downwind of the fire apartment. In the case of the FBNs or externally applied water,
the temperature reductions were the most significant in the room of fire origin, reducing the
temperatures to approximately 100 ºC (212 ºF) within 60 s of the start of suppression. Figure 4.4-1 and
Figure 4.4-2, from experiments 3A and 5K respectively, show the temperatures at 1.2 m (4 ft) above the
floor in the bedroom (room of fire origin) and outside of the fire apartment at locations in the corridor
and in the stairwell on the fire floor. In Experiment 3A, Figure 4.4-1, the post flashover wind driven fire
was suppressed with FBN1, which was fitted with a 1-1/8th inch smooth bore tip, and flowing
approximately 160 gpm of water. Notice that after the initial bedroom fire was suppressed, the water
flow was stopped and the FBN was moved to suppress the fire that the wind had spread to the living
room. This provided additional reductions in temperatures on the fire floor.

Figure 4.4-2 shows a similar scenario from Experiment 5K. In this experiment FBN1 was fitted with an
adjustable fog tip. The nozzle was adjusted to a narrow setting of approximately 30º and had a flow rate
of 160 gpm. Again, the nozzle provided rapid suppression of the bedroom. The fog stream did not
have as much impact downstream as did the solid stream when bounced off the ceiling. Just as in 3A,
the FBN was moved to suppress the fire that the wind had spread to the living room. This further
reduced the temperatures on the fire floor. In both cases, the FBN was able to reduce the hazards and
provide a tenable environment for firefighters in full PPE.

418
 BR: BR
FBN : FBN
fl ow stopp LR: win l ear i ng pe d i ng pe d
i ng
(dou ed (dou dow f a in d ow c BN flow BN stop BN fl ow BN stop
ble) ble) il ing w F F F F
LR: LR: LR: LR: LR:
1200 2192
TC BR3AE 1.2m
TC Corr3AE 1.2m
TC Corr3EE 1.2m
1000 TC CorrStair3E 1.2m 1832
TC Stair3E 1.2m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
-30 0 30 60 90 120 150 180 210 240 270 300
Time (s)
Figure 4.4-1. 3A Temperature vs time at 1.2 m (4 ft) above the floor. T=0 is the time that suppression was started.

le) ble)
g (doub ed (dou flowing stopped
N flowin N stopp sed
BR: FB BR: FB BH: clo LR: FBN LR: FBN
1500 2732
TC BR5KW 1.2m
TC Corr5KW 1.2m
TC Corr5GW 1.2m
TC CorrStair5W 1.2m
1200 2192
TC Stair5W 1.2m

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
-30 0 30 60 90 120 150 180 210 240 270 300
Time (s)
Figure 4.4-2. 5K Temperature vs time at 1.2 m (4 ft) above the floor. T=0 is the time that suppression was started.

4.5 Window Failure Time 

The windows in the apartments were double pane, aluminum framed windows. Whereas the focus of the
study was not on window construction or window failure modes, the data collected on window failure
provides some insight. Only the failure times from the bedroom double windows is examined because
those were the experiments in which the point of ignition was in the same location relative to the
window and where the windows had not been previous exposed to fire conditions. These results are
summarized in Table 4.5-1.

419
Based on what appeared to be a long time for window failure in Experiment 7G and since flashover and
the wind driven conditions could not occur until the window failed, the inner pane of glass was removed
for Experiment 7E. This resulted in a failure time of 163 s and appeared to delay the onset of flashover
because of the early failure time. In both of these cases the gas temperature in the bedroom at 2.1 m
above the floor, was approximately 500 ºC (932 ºF).

In the remaining experiments, the inner window pane was left in place, but the glass was scored with a
glass cutter to increase points of stress to incur earlier failure of the inner pane. In these cases the failure
times ranged from a low of 233 s to a high of 496 s and averaged 342 s. Gas temperatures in the room,
near the ceiling, at the time of failure, ranged from 300 º C (572 ºF) to 700 º C (1292).

In each of the experiments, it appeared that flames made contact with the windows prior to failure of the
glass. The failure of the glass was one of the enabling events for the wind driven fire condition inside
the structure, as it served as the fresh air inlet to any potential flow path that might evolve based on
opening of doors or other windows.

420
Table 4.5-1. Failure Time of Bedroom Double Window and Approximate Upper Layer Gas Temperature

Experiment Time to Initial Approximate Gas Comments

Failure (s) Temperature (º C)
7G 344 550 no wind
7E 163 550 inner pane of glass removed
7A 316 400 Inner pane scored
7K 240 700 1.8m in BR, Inner pane scored
5E 233 550 WCD pre-deployed, Inner pane scored
5K 303 550 Inner pane scored
5A 234 500 Inner pane scored
5G 425 300 Inner pane scored
3E 493 550 Inner pane scored
3A 496 550 Inner pane scored

5 Tactical Considerations 

Fire departments that wish to implement the tactics used in this study will need to develop training and
determine appropriate methods for deploying these tactics. Variations in the methods of deployment
may be required due to differences in staffing, equipment, building stock, typical weather conditions,
etc. There is uniformity however, in the physics behind the wind driven fire condition and the principles
of the tactics examined. These are discussed below.

Wind is a factor. As shown in these experiments, wind can significantly increase the thermal
hazards of a fire in a structure. Wind conditions will vary at different elevations above the ground floor,
on different sides of a building, due to the effects of surrounding structures or topography, or just
changes in the wind itself. Therefore wind needs to be considered as part of the intial “size-up” of the
fire conditions and continue to be monitored and reported on throughout the fire incident.

Smoke is Fuel. A ventilation limited (fuel rich) condition developed prior to the failure of the
windows. Oxygen depleted combustion products, containing carbon dioxide, carbon monoxide and
unburned hydrocarbons, filled the rooms of the structure. Once the window failed, the fresh air
provided the oxygen needed to sustain the transition through flashover, which caused a significant
increase in temperature.

Venting does not always equal cooling. In these experiments, the cool air forced into the
broken upwind window resulted in an initial period of cooling in the room of origin, typically followed
by a transition to flashover, if a flow path was available.

Fire induced flows. Velocities within the structure exceeded 5 m/s (11 mph), just due to the fire
growth and the flow path that was set-up between the window opening and the open bulkhead door on
the roof.

421
 Avoid the flow path. The directional nature of the fire gas flow was demonstrated with thermal
conditions which were significantly higher in the “flow” portion of the corridor as opposed to the
position remote from the flowpath. Thermal conditions in the flow path were often higher than
400 ºC (752 ºF) and were not consistent with firefighter survival.

Control the flow path. Wind driven conditions with untenable thermal conditions can not occur
if there is no flow path through the structure. If the door to the stair on the fire floor is closed, the spread
of fire and heat will be limited and wind driven “blow torch” conditions will not occur. In these
experiments, after the fire room window had failed, opening the fire floor stair door and the bulkhead
door, typically provided the flow path that led to wind driven “blow torch” conditions on the fire floor.
Therefore it is importment to control and coordinate the opening and closing of doors in a structure.
This study has also shown that WCDs, when deployed in a way to completely cover the window
opening, were effective in mitigating the effects of the wind and controlling inlet of the flow path. The
experiments demonstrated the importance of coordinating and controlling the opening and closing of
doors, which could provide a flow path, to avoid placing fire fighters in the flow path. Fire fighter
training would be beneficial to identify and prepare areas which could be used as safe areas of refuge,
which are out of the flow path.

Use of PPV. Two 27 in PPV fans could not overcome the effects of a wind driven condition.
However when used in conjunction with door control, WCDs, and FBNs (tactics that stop the wind
driven condition) the PPV fans were able to maintain tenable and clear conditions in the stairwell. In
these experiments, PPV use always improved conditions in the stairwell.

Impact of WCDs. In these experiments, the WCDs reduced the temperatures in the corridor and
the stairwell by more than 50 % within 120 s of deployment. The WCDs also completely mitigated any
velocity due to the external wind. The WCDs were exposed to a variety of extended thermal conditions
without failure. The WCD must cover the window opening completely to be effective. Pre-deployment
of a WCD over the fire room upwind window, prevented the development of wind driven fire
conditions. The benefit of using a WCD, compared to using the apartment door or the stairwell door, is
that the flow path is interrupted at the entry point. This improves all of the conditions along the entire
potential flow path.

Impact of externally applied water. In these experiments, the externally applied water streams
were implemented in different ways; a fog stream inserted into the fire room window, a fog stream
flowed from the floor below into the fire room window opening, and a solid water stream flowed from
the floor below into the fire room window opening. In all cases, the water flows suppressed the fires,
thereby causing reductions of at least 50 % in temperature in the corridor and the stairwell. The water
flow rates used in these experiments were between 125 gpm and 200 gpm, demonstrating that a
relatively small amount of water directly applied to burning fuels can have a significant impact.

Wind driven fire conditions can generate and transfer energy throughout the flow path. When doors or
WCDs are used to stop the wind driven fire conditions, energy and fuel may be trapped on the fire floor.
These experimental results indicate that the thermal conditions due to the residual heat on the fire floor,
were still of a level which could pose a hazard to firefighters in full PPE. However when used in
combination with PPV fans to force cool air into the stairwell and out through the fire floor, and or with

422
the cooling effect from water streams, the fire floor temperatures were reduced to tenable conditions for
fire fighters in full PPE in minutes.

6 Future Research 
The results from this series of full scale experiments in a 7-story, fire resistive building demonstrated
that WCDs, and externally applied water streams, have the potential to reduce the hazard from a wind
driven apartment fire. The resulting conditions in the corridor offered a fire environment with an
improved level of safety for firefighters, although not an environment free from hazard. When PPV was
used, the stair was cleared of smoke and hot gases.

The constraints of the fire resistive construction, used in these experiments, preclude insight on the
response of ordinary or wood frame construction buildings to wind. For example, in the case of a wood
framed structures, this study did not fully address all of the hazards that could be generated by a wind
driven structure fire such as a shorter time until structural collapse. Therefore it is important to take the
lessons from these experiments and the laboratory based experiments [1] to further advance the
understanding of these tactics in the field. Experiments in real wood framed buildings with realistic fuel
loads are required to further the understanding of the capabilities and limitations of implementing fire
fighting tactics with PPV, WCDs and external hose streams in non fire resistive buildings.

6.1 Pilot Programs 

FDNY has developed a training program on wind driven fires to educate their members about the
importance of considering wind conditions when sizing up a fire, and to develop an understanding of
flows within the building and how to control those flows with doors and PPV fans. Based on the
outcome of the Governors Island experiments, FDNY has implemented a pilot program that includes
training on tactics to mitigate wind driven fire hazards and deployment of PPV fans, WCDs and
external hose streams which could be used in high rises. The Chicago Fire Department is planning to
implement a pilot study on the use of PPV fans, WCDs, and external hose streams in the near future
[51]

6.2 Standard Test Methods for Equipment 

As the research and field trials continue, there are many commercially available products that are being
examined and there are many prototype firefighting tools that are being offered for use in the
experiments. If the technologies demonstrated continue to prove effective in the field trials and pilot
programs, the next step may be to examine the need for standards and standardized test methods to
define a minimum level of acceptable performance of these devices.

7 Summary 

The National Institute of Standards and Technology (NIST), the Fire Department of New York City

423
(FDNY), and the Polytechnic Institute of New York University with the support of the Department of
Homeland Security (DHS)/Federal Emergency Management Agency (FEMA) Assistance to Firefighters
Research and Development Grant Program and the United States Fire Administration (USFA), have
conducted a series of wind driven fire experiments in a 7-story, building on Governors Island, New
York.

The objective of this study was to improve the safety of firefighters and building occupants by enabling
a better understanding of wind driven fires and wind driven firefighting tactics, including structural
ventilation and suppression. A series of 14 experiments were conducted to evaluate the ability of
positive pressure ventilation fans (PPV), wind control devices (WCD) and floor below nozzles (FBN) to
mitigate the hazards of a wind driven fire in a structure.

Each of the 14 experiments started with a fire in a furnished room. The air flow for 12 of the 14
experiments was intensified by a natural or mechanical wind. Each of the tools was evaluated
individually as well and in conjunction with each other to assess the benefit to fire fighters, as well as
potential occupants in the structure. The measurements used to examine the impact of the WCDs and
the external water application tactics were temperature, differential pressure, and gas velocity inside the
structure. Each of the experiments was recorded with video and thermal imaging cameras. These
experiments also provided visual documentation of fire phenomena that are not typically observable on
the fire ground.

The experiments were designed to expose a public corridor and stairwell area to a wind driven, post-
flashover apartment fire. The door from the apartment to the corridor was open for each of the
experiments. The conditions in the corridor and the stairwell were of critical importance because that is
the portion of the building that firefighters would use to approach the fire apartment or that occupants
from an adjoining apartments or adjacent floors would use to exit the building.

All of the fires were ignited in furnished rooms of an apartment. Due to excess fuel generation (or lack
of ventilation) the room of fire origin could not transition to flashover until windows self-vented and
introduced additional fresh air with oxygen to burn. Without a wind imposed on the vented window, the
fire did not spread from the room of origin and never left the apartment of origin. Even with no
externally applied wind, creating a flow path from the outside, through the fire apartment into the
corridor and up the stairs to the bulkhead on the roof increased the temperatures and velocities in the
corridors and in the stairwell to create hazardous conditions for fire fighters and untenable conditions for
occupants on the fire floor and above in the stairwell.

With an imposed wind of 9 m/s to 11 m/s (20 mph to 25 mph) and flow path through the fire floor and
exiting out of the bulkhead, temperatures in excess of 400 ºC (752 ºF) and velocities on the order of
10 m/s (22 mph) were measured in the corridor and stairwell above the fire floor. These extreme
thermal conditions are not teneable, even for a firefighter in full protective gear.

These experiments demonstrated the “extreme” thermal conditions that can be generated by a “simple
room and contents” fire and how these conditions can be extended along a flow path within a real
structure when wind and an open vent are present. Potential tactics which could be implemented to
interrupt and control the flow path are door control from inside the structure and or a WCD from the

424
floor above the fire. From the floor below the fire, external water application was demonstrated to be
effective in reducing the thermal hazard in the corridor and stairwell.

Impact of PPV. PPV fans alone could not overcome the effects of a wind driven condition. However
when used in conjunction with door control, WCDs, and FBNs the PPV fans were able to maintain
tenable and clear conditions in the stairwell.

Impact of WCDs. In these experiments, the WCDs reduced the temperatures in the corridor and the
stairwell by more than 50 % within 120 s of deployment. The WCDs also completely mitigated any
velocity due to the external wind. The WCDs were exposed to a variety of extended thermal conditions
without failure.

Impact of externally applied water. In these experiments, the externally applied water streams were
implemented in different ways; a fog stream inserted into the fire room window, a fog stream flowed
from the floor below into the fire room window opening, and a solid water stream flowed from the floor
below into the fire room window opening. In all cases, the water flows suppressed the fires, thereby
causing reductions in temperature in the corridor and the stairwell of at least 50 %. The water flow rates
used in these experiments were between 160 gpm and 200 gpm, demonstrating that a relatively small
amount of water applied directly to the burn fuels can have a significant impact.

These experimental results also indicate that the post deployment thermal conditions for flow path
control using a WCD, after the development of wind driven conditions, were still of a level which could
pose a hazard to firefighters in full PPE. However, when used in combination with PPV fans to force
cool air into the stairwell and out through the fire floor, and or with the cooling effect from an externally
applied water stream, the fire floor temperatures can be reduced to near ambient conditions in minutes.

The experiments also provided potential guidance for firefighters as a part of a fire size up and approach
to the room of fire origin: note wind conditions in the area of the fire, look for “pulsing flames”, or
flames not exiting a window opening, examine smoke conditions around closed doors in the potential
flow path, and maintain control of doors in the flow path. Even if flames are being forced out of
adjacent windows with a high amount of energy, there could still be sufficient energy flows on the fire
floor to create a hazard for firefighters.

The data from this research will also help to identify methods and promulgation of improved standard
operating guidelines (SOG) for the fire service to enhance firefighter safety, fire ground operations, and
use of equipment. If the demonstrated technologies continue to prove effective in the field trials and
pilot programs, the next step may be to examine the need for standards and standardized test methods to
define a minimum level of acceptable performance of these devices.

425
8 References 

1. Madrzykowski, D., and Kerber, S., Fire Fighting Tactics Under Wind Driven Conditions:
Laboratory Experiments. National Institute of Standards and Technology, Gaithersburg, MD,
NIST TN 1618, January 2009.
2. Hall, John, R., High-rise Building Fires. National Fire Protection Association, Fire Analysis and
Research Division, 1 Batterymarch Park, Quincy, MA., August 2005.
3. Madrzykowski, D., and Walton, W.D., Cook County Administration Building Fire, 69 West
Washington, Chicago, Illinois, October 17, 2003: Heat Release Rate Experiments and FDS
simulations. National Institute of Standards and Technology, Gaithersburg, MD, NIST SP 1021,
July 2004.
4. NIOSH F99-01, Three firefighters die in a 10-Story High-Rise Apartment Building – New York,
NIOSH Fire Fighter Fatality Investigation and Prevention Program, Morgantown, WV., August
1999.
5. NIOSH F2001-33, High-Rise Apartment Fire Claims the Life of One Career Fire Fighter (Captain)
and Injures Another Career Fire Fighter (Captain) – Texas, NIOSH Fire Fighter Fatality
Investigation and Prevention Program, Morgantown, WV., October 2002.
6. http://www.nfpa.org/itemDetail.asp?categoryID=14398&itemID=34426&URL=Learning/Public%2
0education/Fire%20Prevention%20Week/About%20Fire%20Prevention%20Week. National
Fire Protection Association, About Fire Prevention Week. Downloaded 9/11/08.
7. http://www.crh.noaa.gov/grr/history/?m=10&d=8 NOAA National Weather Service Weather
Forecast Office, Weather History of October 8th. Downloaded 9/11/08.
8. NFPA 1051, Standard for Wildland Fire Fighter Professional Qualifications, National Fire
Protection Association, 1 Batterymarch Park, Quincy, MA. 2007 ed.
9. NFPA 1001, Standard for Fire Fighter Professional Qualifications, National Fire Protection
Association, 1 Batterymarch Park, Quincy, MA. 2008 ed.
10. NFPA 1021, Standard for Fire Officer Professional Qualifications, National Fire Protection
Association, 1 Batterymarch Park, Quincy, MA. 2003 ed.
11. Andrews, P.L.; Bevins, C.D., Seli, R.C., BehavePlus fire modeling system, version 4.0: User’s
Guide. USDA, Forest Service, Rocky Mountain Research Station, General Technical Report
RMRS-GTR-106WWW, July 2008.
12. Fundamentals of Wildland Fire Fighting, International Fire Service Training Association,
Oklahoma State University, Stillwater, OK 3rd ed. 1998.
13. Essentials of Fire Fighting, International Fire Service Training Association, Oklahoma State
University, Stillwater, OK 5th ed., 2008.
14. Firefighter’s Handbook, Essentials of Firefighting and Emergency Response, 2nd ed. Thomson
Delmar Learning, Clifton Park, NY, 2004.
15. Fundamentals of Fire Fighter Skills, Jones and Bartlett Publishers, Sudbury, MA., 2004.
16. Norman, John, “Extreme Wind Driven Fireproof Multiple Dwelling Fires”, With New York
Firefighters (WNYF), New York, NY, 1st/2007.
17. Tracy, Gerald, “1 Lincoln Plaza, Operations of the First-Arriving Units at a High-Rise Multiple
Dwelling Fire”, With New York Firefighters (WNYF), New York, NY., 2nd/1997.
18. Daly, James, D., and Healy, George, “Wind-Driven Queens Fire Provokes Several Maydays”, With
New York Firefighters (WNYF), New York, NY., 3rd/2006.
19. Tracy, Gerald, Personal Communication.

426
20. Fighting Wind Driven Fires in High Rise Multiple Dwellings, FDNY, 9 Metrotech Center,
Brooklyn, NY., 2007.
21. NIOSH F2005-03, Career lieutenant and career fire fighter die and four career fire fighters are
seriously injured during a three alarm apartment fire - New York. NIOSH Fire Fighter Fatality
Investigation and Prevention Program, Morgantown, WV. Issued January 2006, Revised January
2007.
22. Liu, Henry, “Wind Engineering: A Handbook for Structural Engineers”. Prentice Hall, Englewood
Cliffs, NJ 1991. pp62-64.
23. Simiu, Emil and Scanlan, Robert H., “Wind Effects on Structures, Fundamentals and Applications
to Design”, 3rd ed.. John Wiley & Sons, New York, NY, 1996, p 520.
24. Investigation Number 02-50-10, Captain Jay Jahnke, Houston Fire Department, October 13, 2001 ,
State Fire Marshal’s Office Line of Duty Death Investigation, Texas Department of Insurance,
Austin, Texas.
25. NIOSH F98-26, Eight-Alarm Fire in a 27-story High-Rise Apartment Building for the Elderly
Nearly Claims the Life of One Fire Fighter – Missouri. NIOSH Fire Fighter Fatality
Investigation and Prevention Program, Morgantown, WV. February 23, 1999.
26. NIOSH F2007-12, Career Fire Fighter Dies in Wind Driven Residential Structure Fire – Virginia.
NIOSH Fire Fighter Fatality Investigation and Prevention Program, Morgantown, WV. Issued
May 2008, Revised June 2008.
27. Goldfeder, W., Front Door Left Open by Occupants Advances Fire Onto Firefighters, Firehouse,
Vol. 33, No. 3, Melville, NY., March 2008.
28. “Report # 05-0000531.” National Firefighter Near-Miss Reporting System. 18 September 2005.
11 September 2008 <http://www.firefighternearmiss.com/gogglemini/h05-0000531.html>.
29. “Report # 06-0000164.” “National Firefighter Near-Miss Reporting System. 15 March 2006.
11 September 2008 <http://www.firefighternearmiss.com/googlemini/h06-0000164.html>.
30. “Report # 06-0000186.” National Firefighter Near-Miss Reporting System. 24 March 2006.
11 September 2008 <http://www.firefighternearmiss.com/googlemini/h06-0000186.html>.
31. “Report # 06-0000501.” National Firefighter Near-Miss Reporting System. 4 October 2006.
11 September 2008 < http://www.firefighternearmiss.com/googlemini/h06-0000501.html>.
32. “Report # 07-0000805.” National Firefighter Near-Miss Reporting System. 19 March 2007.
11 September 2008 <http://www.firefighternearmiss.com/googlemini/h07-0000805.html>.
33. “Report # 07-0000960.” National Firefighter Near-Miss Reporting System. 14 June 2007.
11 September 2008 <http://www.firefighternearmiss.com/googlemini/h07-00960.html>.
34. “Report # 08-0000154.” National Firefighter Near-Miss Reporting System. 25 March 2008.
11 September 2008 <http://www.firefighternearmiss.com/googlemini/h08-0000154.html>.
35. Kerber, S. and Walton, W. D. “Characterizing Positive Pressure Ventilation Using Computational
Fluid Dynamics”. National Institute of Standards and Technology, Gaithersburg, MD., NISTIR
7065, 2003.
36. Kerber, S. and Walton, W. D. “Effect of Positive Pressure Ventilation on a Room Fire.” National
Institute of Standards and Technology, Gaithersburg, MD., NISTIR 7213, 2005.
37. Kerber, S. and Walton, W. D. “Full Scale Evaluation of Positive Pressure Ventilation in a Fire
Fighter Training Building.” National Institute of Standards and Technology, Gaithersburg, MD.,
NISTIR 7342, 2006.
38. Kerber, S. “Evaluation of the Ability of Fire Dynamic Simulator to Simulate Positive Pressure
Ventilation in the Laboratory and Practical Scenarios.” National Institute of Standards and
Technology, Gaithersburg, MD., NISTIR 7315, 2006.

427
39. Kerber, S., Madrzykowski, D., Stroup, D. “Evaluating Positive Pressure Ventilation In Large
Structures: High-Rise Pressure Experiments.” National Institute of Standards and Technology,
Gaithersburg, MD., NISTIR 7412, 2007.
40. Kerber, S., Madrzykowski, D., “Evaluating Positive Pressure Ventilation In Large Structures: High-
Rise Fire Experiments.” National Institute of Standards and Technology, Gaithersburg, MD.,
NISTIR 7468, 2007.
41. Taylor, B.N., and Kuyatt, C.E., “Guidelines for Evaluating and Expressing the Uncertainty of NIST
Measurement Results”, National Institute of Standards and Technology, Gaithersburg. MD.,
NIST TN 1297, January 1993.
42. Setra Model 264, Very Low Pressure Transducer Data Sheet Rev E., Setra Systems, Boxborough,
MA., December 2002.
43. Omega Engineering Inc., The Temperature Handbook, Vol. MM, pages Z-39-40, Stamford, CT.,
2004.
44. Blevins, L.G., "Behavior of Bare and Aspirated Thermocouples in Compartment Fires", National
Heat Transfer Conference, 33rd Proceedings. HTD99-280. August 15-17, 1999, Albuquerque,
NM, 1999.
45. Pitts, W.M., E. Braun, R.D. Peacock, H.E. Mitler, E. L. Johnsson, P.A. Reneke, and L.G.Blevins,
"Temperature Uncertainties for Bare-Bead and Aspirated Thermocouple Measurements in Fire
Environments," Thermal Measurements: The Foundation of Fire Standards. American Society
for Testing and Materials (ASTM). Proceedings. ASTM STP 1427. December 3, 2001, Dallas,
TX.
46. Fit 5 Product Information Sheet, ARA Safety Inc., Vancouver BC 2008
47. Abeles, F.J.; Del Vecchio, R.J.; and Himel, V.H., A Firefighter’s Integrated Life Protection System:
Phase I, Design and Performance Requirements. Grumman Aerospace Corporation, Bethpage,
New York, September 1974.
48. Peacock, Richard, D., Krasny, John, F., Rockett, John, A., and Huang, Dingyi, Protecting
Firefighters Exposed in Room Fires, Part 2: Performance of Turnout Coat Materials Under
Actual Fire Conditions. Fire Technology, Vol. 26, No. 3, August 1990, pp 202-222.
49. Donnelly, M.K., Davis, W.D., Lawson, J.R., amd Selepak, M.J., Thermal Environment for
Electronic Equipment Used by First Responders. National Institute of Standards and
Technology, Gaithersburg, MD, NIST TN 1474, January 2006.
50. Madrzykowski, D., Fatal Training Fires: Fire Analysis for the Fire Service. Interflam 2007.
International Interflam Conference, 11th Proceedings London, England, September 3-5, 2007.
Interscience Communications, London, England. pp 1169-1180.
51. Personal communication with Richard Edgeworth, Director of Training, Chicago Fire Department,
March 18, 2009.

428
9 Appendices 
9.1 Appendix A – Detailed Drawings  
Polytechnic University and the Bureau of Alcohol Tobacco Firearms and Explosives (ATF)
measured the structure and their detailed drawings are included in this appendix.

429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
9.2 Appendix B – Fuel Distribution 
Table 9.2-1. Fuel Distribution for the A Apartments
Apartment 3A Apartment 5A Apartment 7A
Living Living Living
Item Bedroom Room Bedroom Room Bedroom Room
Mattress 1 1 1
Box Spring 1 1 1
Head Board 1 1 1
Bedding/Pillows 1 1 1
Upholstered Chair (Brown) 1 1 1
Upholstered Chair (Red) 1 1 1
Chair (Red w/ wood arms) 1 1 1
Desk Chair (Green) 2 1 1
Desk Chair (Brown) 1 1 1
Coffee Table 1 1 1
Desk (Dark Brown w/ blue top) 1 1 1
Desk (Light Brown) 1 1 1 1 1 1
Dresser (Dark Brown) 1 1 1
Dresser (Light Brown) 1 1 1
Table Lamp 2 2 2 2 2 2
Nightstand (Light Brown) 1 1 2 2
Nightstand (Dark Brown) 1 1 1
Round Table 1 1 1
Sofa 1 1 1
Floor Lamp 1 1 1
TV (Small) 1
TV (Large) 1 1 1 1

447
Table 9.2-2. Fuel Distribution for the E Apartments
Apartment 3E Apartment 5E Apartment 7E
Living Living Living
Item Bedroom Room Bedroom Room Bedroom Room
Mattress 1 1 1
Box Spring 1 1 1
Head Board 1 1 1
Bedding/Pillows 1 1 1
Upholstered Chair (Brown) 1
Upholstered Chair (Red) 1 1 1
Chair (Red w/ wood arms) 1 1 1
Desk Chair (Green) 2 1 1
Desk Chair (Brown) 1 1 1
Coffee Table 1 1 1
Desk (Dark Brown w/ blue top) 1 1 1
Desk (Light Brown) 1 1 1 1 1 1
Dresser (Dark Brown) 1 1 1
Dresser (Light Brown) 1 1 1
Table Lamp 2 2 2 2 2 2
Nightstand (Light Brown) 2 2 2
Nightstand (Dark Brown) 1 1 1
Round Table 1 1 1
Sofa 2 2 2
Floor Lamp 1 1 1
TV (Small) 1
TV (Large) 1 1 1 1 1

448
Table 9.2-3. Fuel distribution for the G Apartments
Apartment 3G Apartment 5G Apartment 7G
Living Living Living
Item Bedrooms Room Bedrooms Room Bedrooms Room
Mattress 2 2 2
Box Spring 2 2 2
Head Board 2 2 2
Bedding/Pillows 2 2 2
Upholstered Chair (Brown) 1 1
Upholstered Chair (Red) 2 2 2
Chair (Red w/ wood arms) 1 1 1
Desk Chair (Green) 2 2 2
Desk Chair (Brown) 1 1 2
Coffee Table 1 1 1
Desk (Dark Brown w/ blue top) 2 2 2
Desk (Light Brown) 2 2 2
Dresser (Dark Brown) 2 2 2
Dresser (Light Brown) 1 1 1
Table Lamp 2 2 3 2 3 2
Nightstand (Light Brown) 2 2 2
Nightstand (Dark Brown) 2 2 2
Round Table 2 2 2
Sofa 1 1 1
Floor Lamp 1 1 1
TV (Small) 1 1
TV (Large) 1 1 1 1

449
Table 9.2-4. Fuel distribution for the K Apartments
Apartment 3K Apartment 5K Apartment 7K
Living Living Living
Item Bedrooms Room Bedrooms Room Bedrooms Room
Mattress 2 2 2
Box Spring 2 2 2
Head Board 2 2 2
Bedding/Pillows 2 2 2
Upholstered Chair (Brown) 1 1
Upholstered Chair (Red) 2 2 2
Chair (Red w/ wood arms) 1 1 1
Desk Chair (Green) 1 2 2
Desk Chair (Brown) 1 1
Coffee Table 1 1 1
Desk (Dark Brown w/ blue top) 2 2 2
Desk (Light Brown) 2 2 2
Dresser (Dark Brown) 2 2 2
Dresser (Light Brown) 1 1 1
Table Lamp 3 2 1 2 3 3
Nightstand (Light Brown) 2 2 2
Nightstand (Dark Brown) 2 2 2
Round Table 2 2 2
Sofa 2 2 2
Floor Lamp 1 1 1
TV (Small) 1 1
TV (Large) 1 1 2 1

450
9.3 Appendix C – Wind Measurements and Descriptions 

Polytechnic University positioned weather stations around the structure to measure wind speed
and direction. This appendix includes their data and supporting documentation.

451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
9.4 Appendix D – Detailed Graphs 
9.4.1 Experiment 7G 
9.4.1.1 Temperatures
Ex
S7 BH pe
BR S1 B B B : : o S S 5 r im
: : R R R o pe 5: :c MV ent
BR dou LBY stair : dou : sing : sin peni BH S7 S S
BH S7 7: ing pe n o M V U
l o s U s com
b
: Ig le : o o b le g ng : o : o 1 o n e d
nit fail pen pen le cle fail le cl 0.0 pen pen : fan : clos : clos pen /clos /f an star t /fan topp ple
ion ing ing ing ar ing ear 8m ing ing on ed ed ing ing on ing off ed ted

1200 2192
TC BR7GW 2.1m
TC BR7GW 1.8m
TC BR7GW 1.5m
1000 TC BR7GW 1.2m 1832
TC BR7GW 0.6m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-1. Temperature versus time for the 7G bedroom thermocouple tree, Experiment 7G
Ex
S7 BH pe
BR S1 B R BR B :o :o S5 S5 r im
:d L : s : : s R : p e B p e : M : c MV ent
BR u o B Y t a d o i n s n H S B S S nin p o V l o
i
: Ig ble : op r op ble gle u i n gle ng 0 : op : op 1: f H: c 7: c : op g/c en/f U st ed/f U sto comp
i 7 S 7 s
nit fail en en f e e a l e l a a a
ion ing ing ing clea ailing clea .08m ning ning n on losed osed ning osing n on r ting n of pped leted
r r f
400 752
TC Corr7GW C
TC Corr7GW 2.1m
TC Corr7GW 1.8m
TC Corr7GW 1.5m
TC Corr7GW 1.2m
300 TC Corr7GW 0.9m
572

Temperature (F)
TC Corr7GW 0.6m
Temperature (C)

200 392

100 212

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-2. Temperature versus time for the 7G hall thermocouple tree, Experiment 7G

497
 Ex
S7 BH pe
BR S1 B B B : : o S S 5 r im
: : R R R o pe 5: :c MV ent
BR dou LBY stair : dou : sing : sin peni BH S7 S B S S 7 n in o p
MV U
l o s U s com
b
: Ig le : o o b le g ng : o : o 1 H 7 : o g e n e d
nit fail pen pen le cle fail le cl 0.0 pen pen : fan : clos : clos pen /clos /f an star t /fan topp ple
ion ing ing ing ar in g e ar 8 m in g i ng o n e d e d i ng in g o n i n g off ed ted

250 482
TC Corr7KW C
TC Corr7KW 2.1m
TC Corr7KW 1.8m
200 TC Corr7KW 1.5m 392
TC Corr7KW 1.2m
TC Corr7KW 0.9m

Temperature (F)
TC Corr7KW 0.6m
Temperature (C)

150 302

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-3. Temperature versus time for the 7K hall thermocouple tree, Experiment 7G
Ex
S7 BH pe
BR S1 B R BR B : :o S S5 r im
:d L : st : : R : o p p e 5 : M : c MV ent
BR u o B Y a d o s i n s e n B H S B S S nin p o V l o
i
: Ig ble : op r op ble gle u i n gle ng 0 : op : op 1: f H: c 7: c : op g/c en/f U st ed/f U sto comp
i 7 S 7 s
nit fail en en f e e a l e l a a a
ion ing ing ing clea ailing clea .08m ning ning n on losed osed ning osing n on r ting n of pped leted
r r f
400 752
TC CorrStair7W C
TC CorrStair7W 2.1m
TC CorrStair7W 1.8m
TC CorrStair7W 1.5m
TC CorrStair7W 1.2m
300 TC CorrStair7W 0.9m
572

Temperature (F)
TC CorrStair7W 0.6m
Temperature (C)

200 392

100 212

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-4. Temperature versus time for the 7th floor stair hall thermocouple tree, Experiment 7G

498
 Ex
S7 BH pe
BR S1 B B B : : o S S 5 r im
: : R R R o pe 5: :c MV ent
BR dou LBY stair : dou : sing : sin peni BH S7 S B S S 7 n in o p
MV U
l o s U s com
b
: Ig le : o o b le g ng : o : o 1 H 7 : o g e n e d
nit fail pen pen le cle fail le cl 0.0 pen pen : fan : clos : clos pen /clos /f an star t /fan topp ple
ion ing ing ing ar in g e ar 8 m in g i ng o n e d e d i ng in g o n i n g off ed ted

250 482
TC Stair7W 2.1m
TC Stair7W 1.8m
TC Stair7W 1.5m
200 TC Stair7W 1.2m 392
TC Stair7W 0.9m
TC Stair7W 0.6m

Temperature (F)
Temperature (C)

150 302

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-5. Temperature versus time for the 7th floor stairwell thermocouple tree, Experiment 7G
Ex
S7 BH pe
BR S1 B R BR B : :o S S5 r im
:d L : st : : R : o p p e 5 : M : c MV ent
BR u o B Y a d o s i n s e n B H S B S S nin p o V l o
i
: Ig ble : op r op ble gle u i n gle ng 0 : op : op 1: f H: c 7: c : op g/c en/f U st ed/f U sto comp
i 7 S 7 s
nit fail en en f e e a l e l a a a
ion ing ing ing clea ailing clea .08m ning ning n on losed osed ning osing n on r ting n of pped leted
r r f
120 248
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
100 TC Stair 2W 212
TC Stair1 MidW

Temperature (F)
Temperature (C)

80 176

60 140

40 104

20 68

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-6. Temperature versus time for the stair thermocouples, Experiment 7G

499
9.4.1.2 Pressures
Ex
S BH pe
BR S1 B R B R B 7: o :o S5 S5 r im
:d : : R p p : MV ent
BR ou LBY stair :d s : s
ou ing in nin H: e B S 7 S B S S 7
e n in
: o p
MV U
cl o s U c
b
: Ig le : o o ble le gle g o :o 1: H: 7: : o g/c n e sta d/fa sto omp
e
nit fail pen pen cle faili cle 0.08 peni peni fan clos clos peni losin /fan r pp l
ion ing ing ing ar n g ar m n g ng o n ed e d ng g o n
ti ng off ed eted
n

200
DP Stair 7W
DP Stair 6W
DP Stair 5W
150 DP Stair 4W
DP Stair 3W
DP Stair 2W
DP Stair 1W
Pressure (Pa)

100

50

-50
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-7. Differential pressure versus time for the stairwell bi-directional probes, Experiment 7G

9.4.1.3 Velocities
Ex
S7 BH pe
BR S1 B R BR B :o :o S5 S5 r im
:d : R p : MV ent
BR ou LBY stair : dou : sing : sin peni BH S7 S B S S 7
e n in
: o p
MV U
cl o s U s com
b
: Ig le : o o b le g ng : o :o 1 H 7 :o g e n e d
nit fail pen pen le cle fail le cl 0.0 pen pen : fan : clos : clos pen /clos /f an star t /fan topp ple
ion ing ing ing ar in g e ar 8 m in i
g ng o n ed e d ng i in g o n i ng off ed ted

6 13

3 7

0 0

Velocity (mph)
Velocity (m/s)

-3 -7

-6 -13

-9 VEL Stair7 TopW

-20
VEL Stair7 MidW
VEL Stair7 BotW
-12 -27
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-8. Velocity versus time for the 7th floor stairwell bi-directional probes, Experiment 7G

500
 Ex
S7 BH pe
BR S1 B B B : : o S S 5 r im
: : R R R o pe 5: :c MV ent
BR dou LBY stair : dou : sing : sin peni BH S7 S B S S 7 n in o p
MV U
l o s U s com
b
: Ig le : o o b le g ng : o : o 1 H 7 : o g e n e d
nit fail pen pen le cle fail le cl 0.0 pen pen : fan : clos : clos pen /clos /f an star t /fan topp ple
ion ing ing ing ar in g e ar 8 m in g i ng o n e d e d i ng in g o n i n g off ed ted

10 22
VEL Stair1 TopW
VEL Stair1 MidW
8 VEL Stair1 BotW 18

6 13

Velocity (mph)
Velocity (m/s)

4 9

2 4

0 0

-2 -4

-4 -9
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-9. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 7G
Ex
S7 BH pe
BR S1 B R BR B : :o S S5 r im
:d L : st : : R : o p p e 5 : M : c MV ent
BR u o B Y a d o s i n s e n B H S B S S nin p o V l o
i
: Ig ble : op r op ble gle u i n gle ng 0 : op : op 1: f H: c 7: c : op g/c en/f U st ed/f U sto comp
i 7 S 7 s
nit fail en en f e e a l e l a a a
ion ing ing ing clea ailing clea .08m ning ning n on losed osed ning osing n on r ting n of pped leted
r r f
1 2

0 0

-1 -2

Velocity (mph)
Velocity (m/s)

-2 -4

-3 -7

-4 VEL Bulkhead TopW

-9
VEL Bulkhead MidW
VEL Bulkhead BotW
-5 -11
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-10. Velocity versus time for the bulkhead bi-directional probes, Experiment 7G

501
 Ex
S7 BH pe
BR S1 B B B : : o S S 5 r im
: : R R R o pe 5: :c MV ent
BR dou LBY stair : dou : sing : sin peni BH S7 S B S S 7 n in o p
MV U
l o s U s com
b
: Ig le : o o b le g ng : o : o 1 H 7 : o g e n e d
nit fail pen pen le cle fail le cl 0.0 pen pen : fan : clos : clos pen /clos /f an star t /fan topp ple
ion ing ing ing ar in g e ar 8 m in g i ng o n e d e d i ng in g o n i n g off ed ted

6 13

3 7

0 0

Velocity (mph)
Velocity (m/s)

-3 -7

-6 -13

-9 VEL Apt7G TopW

-20
VEL Apt7G MidW
VEL Apt7G BotW
-12 -27
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-11. Velocity versus time for the apartment 7G doorway bi-directional probes, Experiment 7G

9.4.2 Experiment 7G2 
9.4.2.1 Temperatures
S
Mu BR LR L 5: s S5
R S7
l
S
tip LB S S : do 1: s S B
: wi : w tair : o S : clo
B n
le Y: 1: 7: ub ta 1: H S B S H S B do nd pe S7 7: pe 7: se S S BH B
i o S
ign op op op le ir c op op 1: H c 1: op 1: H c w ow n/ : c op n op d/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
200 392
TC BR7GW 2.1m
TC BR7GW 1.8m
TC BR7GW 1.5m
TC BR7GW 1.2m
TC BR7GW 0.6m
150 302

Temperature (F)
Temperature (C)

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-12. Temperature versus time for the 7G2 bedroom thermocouple tree, Experiment 7G2

502
 S
Mu BR LR L 5: s S5
R S7
l
S
tip LB S S : do 1: s S B
: wi : w tair : o S : clo
B n
le Y: 1: 7: ub ta 1: H S B S H S B do nd pe S7 7: pe 7: se S S BH B
i o S
ign op op op le ir c op op 1: H c 1: op 1: H c w ow n/ : c op n op d/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
800 1472
TC Corr7GW C
TC Corr7GW 2.1m
TC Corr7GW 1.8m
TC Corr7GW 1.5m
TC Corr7GW 1.2m
600 TC Corr7GW 0.9m
1112

Temperature (F)
TC Corr7GW 0.6m
Temperature (C)

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-13. Temperature versus time for the 7G2 hall thermocouple tree, Experiment 7G2
S
Mu BR LR L 5: s S5
: S1 : w R: tai S7
l tip LB S S do : s S B B ind win r o S : o S : clo
le Y 1 7 u t 1 H S B S H S B p S 7 p 7 s S S BH B
ign : op : op : op ble air c : op op 1: H c 1: op 1: H c ow dow en/ 7: c : op en : op ed/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
400 752
TC Corr7KW C
TC Corr7KW 2.1m
TC Corr7KW 1.8m
TC Corr7KW 1.5m
TC Corr7KW 1.2m
300 TC Corr7KW 0.9m
572

Temperature (F)
TC Corr7KW 0.6m
Temperature (C)

200 392

100 212

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-14. Temperature versus time for the 7K hall thermocouple tree, Experiment 7G2

503
 S
Mu BR LR L 5: s S5
R S7
l
S
tip LB S S : do 1: s S B
: wi : w tair : o S : clo
B n
le Y: 1: 7: ub ta 1: H S B S H S B do nd pe S7 7: pe 7: se S S BH B
i o S
ign op op op le ir c op op 1: H c 1: op 1: H c w ow n/ : c op n op d/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
500 932
TC CorrStair7W C
TC CorrStair7W 2.1m
TC CorrStair7W 1.8m
400 TC CorrStair7W 1.5m 752
TC CorrStair7W 1.2m
TC CorrStair7W 0.9m

Temperature (F)
TC CorrStair7W 0.6m
Temperature (C)

300 572

200 392

100 212

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-15. Temperature versus time for the 7th floor stair hall thermocouple tree, Experiment 7G2
S
Mu BR LR L 5: s S5
: S1 : w R: tai S7
l tip LB S S do : s S B B ind win r o S : o S : clo
le Y 1 7 u t 1 H S B S H S B p S 7 p 7 s S S BH B
ign : op : op : op ble air c : op op 1: H c 1: op 1: H c ow dow en/ 7: c : op en : op ed/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
400 752
TC Stair7W 2.1m
TC Stair7W 1.8m
TC Stair7W 1.5m
TC Stair7W 1.2m
TC Stair7W 0.9m
300 TC Stair7W 0.6m
572

Temperature (F)
Temperature (C)

200 392

100 212

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-16. Temperature versus time for the 7th floor stairwell thermocouple tree, Experiment 7G2

504
 S
Mu BR LR L 5: s S5
R S7
l
S
tip LB S S : do 1: s S B
: wi : w tair : o S : clo
B n
le Y: 1: 7: ub ta 1: H S B S H S B do nd pe S7 7: pe 7: se S S BH B
i o S
ign op op op le ir c op op 1: H c 1: op 1: H c w ow n/ : c op n op d/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
200 392
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
150 302

Temperature (F)
Temperature (C)

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-17. Temperature versus time for the stair thermocouples, Experiment 7G2

9.4.2.2 Pressures
S
Mu BR LR L 5: s S5
: S1 : w R: tai S7 :c
l tip LB S S do : S i w ro : l
le Y: 1: 7 u st B
1 H S B S B H S1 BH do ind pe S7 7: pe S7: ose S1 S1 BH BH
n S o
ign op op : op ble air c : op op 1: H c 1: op : f c w o w n/ : c o p n op d/ : f : f op c
f f 0
itio en en en fail los en en an los an en an los fail cl fan los en .0 en fan an an en los
ns ing ing ing ing ed ing ing on ed of ing on ed ing ear on ed ing 8m ing of of on ing ed
f f f
40
DP Stair 7W
DP Stair 6W
DP Stair 5W
30 DP Stair 4W
DP Stair 3W
DP Stair 2W
DP Stair 1W
Pressure (Pa)

20

10

-10
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-18. Differential pressure versus time for the stairwell bi-directional probes, Experiment 7G2

505
9.4.2.3 Velocities
S
Mu BR LR L 5: s S5
R S7
l
S
tip LB S S : do 1: s S B
: wi : w tair : :c
le Y: 1: 7: ub ta 1: H S B S H S B do ind pe S7 7: pe S7: lose S S BH B
B n o S o
ign op op op le ir c op op 1: H c 1: op 1: H c w ow n/ : c op n op d/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
4 9

2 4

Velocity (mph)
Velocity (m/s)

0 0

-2 -4

VEL Stair7 TopW

VEL Stair7 MidW
VEL Stair7 BotW
-4 -9
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-19. Velocity versus time for the 7th floor stairwell bi-directional probes, Experiment 7G2
S
Mu BR LR L 5: s S5
R S7
l
S
tip LB S S : do 1: s S B
: wi : w tair : o S : clo
B n
le Y: 1: 7: ub ta 1: H S B S H S B do d pe S7 7: pe 7: se S S BH B
in o S
ign op op op le ir c op op 1: H c 1: op 1: H c w ow n/ : c op n op d/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
4 9

2 4

Velocity (mph)
Velocity (m/s)

0 0

-2 -4

VEL Stair1 TopW

VEL Stair1 MidW
VEL Stair1 BotW
-4 -9
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-20. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 7G2

506
 S
Mu BR LR L 5: s S5
R S7
l
S
tip LB S S : do 1: s S B
: wi : w tair : o S : clo
B n
le Y: 1: 7: ub ta 1: H S B S H S B do nd pe S7 7: pe 7: se S S BH B
i o S
ign op op op le ir c op op 1: H c 1: op 1: H c w ow n/ : c op n op d/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
2 4

0 0

-2 -4

Velocity (mph)
Velocity (m/s)

-4 -9

-6 -13

-8 VEL Bulkhead TopW

-18
VEL Bulkhead MidW
VEL Bulkhead BotW
-10 -22
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-21. Velocity versus time for the bulkhead bi-directional probes, Experiment 7G2
S
Mu BR LR L 5: s S5
: S1 : w R: tai S7
l tip LB S S do : s S B B ind win r o S : o S : clo
le Y 1 7 u t 1 H S B S H S B p S 7 p 7 s S S BH B
ign : op : op : op ble air c : op op 1: H c 1: op 1: H c ow dow en/ 7: c : op en : op ed/ 1: 1: op H c
itio en en en fail los en en fan los fan en fan los fail cl fan los en 0.0 en fan fan fan en los
ns ing ing ing ing ed ing ing on ed off ing on ed ing ear on ed ing 8m ing of off on ing ed
f
9 20

6 13

Velocity (mph)
Velocity (m/s)

3 7

0 0

-3 -7
VEL Apt7G TopW
VEL Apt7G MidW
VEL Apt7G BotW
-6 -13
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-22. Velocity versus time for the apartment 7G doorway bi-directional probes, Experiment 7G2

507
9.4.3 Experiment 7E 
9.4.3.1 Temperatures
ted
l ing 0.08
m
ar n g n ed .08m off alf i lin
g
mpl e
a i le o t 0 n a o
f ing g on d g c vi n us fa h g f tc
ion le ing en n i n nin ble mo /fa ed adj ning ed/ ned nin ow ed en
I gni t do ub open :
e
op ope : fan clos ope dou fan open clos flow ope clos ope ope wind clos er im
: : : Y : Y : : : : : : : : : : : : p
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S 5 S 7 B H LR B H Ex
2500 4532
TC BR7EE 2.1m
TC BR7EE 1.8m
TC BR7EE 1.5m
2000 TC BR7EE 1.2m 3632
TC BR7EE 0.9m
TC BR7EE 0.6m

Temperature (F)
Temperature (C)

1500 2732

1000 1832

500 932

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-23. Temperature versus time for the 7E bedroom thermocouple tree, Experiment 7E
ted
l ing 0.08
m
a r g n ed .08m off alf i l ing mpl e
fa i g e
cl vin an o s t 0 /fan h g a o
ion le ing nin ing on d ing le o f d dju ing d ed in wf d en
tc
gni t do ub open o pe pen fan lose pen oub an m pen/ lose ow a pen lose pen pen indo lose r im
: I : : Y: : o Y: : c : o : d :f c o
: o H: ir fl 7: o 5: c 7: o H: R: w H: c pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 B A S S S B L B Ex
1200 2192
TC Corr7EE C
TC Corr7EE 2.1m
TC Corr7EE 1.8m
1000 TC Corr7EE 1.5m 1832
TC Corr7EE 1.2m
TC Corr7EE 0.9m

Temperature (F)
TC Corr7EE 0.6m
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-24. Temperature versus time for the 7E hall thermocouple tree, Experiment 7E

508
 ted
l ing 0.08
m
a r g n ed .08m off alf i l ing mpl e
i e
cl vin n o t fa o
fa ing g on d g us 0 fan h g c
ion le ing en n i n nin ble mo /fa ed adj ning ed/ ned nin ow ed me
nt
I gni t do ub open :
e
op ope : fan clos ope dou fan open clos flow ope clos ope ope wind clos r i
: : : Y : Y : : : : : : : : : : : : pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S 5 S 7 B H LR B H Ex
1000 1832
TC Corr7AE C
TC Corr7AE 2.1m
TC Corr7AE 1.8m
800 TC Corr7AE 1.5m 1472
TC Corr7AE 1.2m
TC Corr7AE 0.9m

Temperature (F)
TC Corr7AE 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-25. Temperature versus time for the 7A hall thermocouple tree, Experiment 7E
ted
l i ng 0 8m
. a r g n ed .0 8 m off f
l i l in g
mpl e
ai g 0 g cle vin an o st 0 /fan ha g fa tc
o
n ef nin ing on d ing le dju ing ed in w d
n itio oubl penin o
o f d d
pe pen fan lose pen oub an m pen/ lose ow a pen lose pen pen indo lose im en
:I g :d :o Y: : o Y: : c : o : d : f : o H: c ir fl 7: o 5: c 7: o H: o R: w H: c r
pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 B A S S S B L B Ex
1000 1832
TC CorrStair7E C
TC CorrStair7E 2.1m
TC CorrStair7E 1.8m
800 TC CorrStair7E 1.5m 1472
TC CorrStair7E 1.2m
TC CorrStair7E 0.9m

Temperature (F)
TC CorrStair7E 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-26. Temperature versus time for the 7th floor stair hall thermocouple tree, Experiment 7E

509
 ted
l ing 0.08
m
a r g n ed .08m off alf i l ing mpl e
i e
cl vin n o t fa o
fa ing g on d g us 0 fan h g c
ion le ing en n i n nin ble mo /fa ed adj ning ed/ ned nin ow ed me
nt
I gni t do ub open :
e
op ope : fan clos ope dou fan open clos flow ope clos ope ope wind clos r i
: : : Y : Y : : : : : : : : : : : : pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S 5 S 7 B H LR B H Ex
800 1472
TC Stair7E C
TC Stair7E 2.1m
TC Stair7E 1.8m
TC Stair7E 1.5m
TC Stair7E 1.2m
600 TC Stair7E 0.9m
1112

Temperature (F)
TC Stair7E 0.6m
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-27. Temperature versus time for the 7th floor stairwell thermocouple tree, Experiment 7E
ted
l i ng 0 8m
. a r g n ed .0 8 m off f
l i l in g
mpl e
ai g 0 g cle vin an o st 0 /fan ha g a o
n ef nin ing on d ing le dju ing ed in wf d tc
n itio oubl penin o
o f d d
pe pen fan lose pen oub an m pen/ lose ow a pen lose pen pen indo lose im en
:I g :d :o Y: : o Y: : c : o : d : f : o H: c ir fl 7: o 5: c 7: o H: o R: w H: c r
pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 B A S S S B L B Ex
600 1112
TC Bulkhead MidE
TC Stair 6E
TC Stair 4E
500 TC Stair 2E 932
TC Stair1 MidE

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-28. Temperature versus time for the stair thermocouples, Experiment 7E

510
9.4.3.2 Pressures
ted
l ing 0.08
m
a r g n ed .0 8m off lf i l ing mpl e
i
a g g e
cl vin an o s t 0 an h a fa co
n ef nin ing on d ing le dju ing d/f ed ing w d nt
n itio oubl penin o
o /f d
pe pen fan lose pen oub an m pen lose ow a pen lose pen pen indo lose r im e
:I g :d :o Y: : o Y: : c : o : d : f : o H: c ir fl 7: o 5: c 7: o H: o R: w H: c pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 B A S S S B L B Ex
80
DP Stair 7E
DP Stair 6E
DP Stair 5E
60 DP Stair 4E
DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-29. Differential pressure versus time for the stairwell bi-directional probes, Experiment 7E

9.4.3.3 Velocities
ted
l ing 0.08
m
a r g n ed .0 8m off lf i l in g
mpl e
i
a g g cle vin an o st 0 /fan ha g fa o
n ef nin ing on d ing le dju ing ed in w d tc
n itio oubl penin o
o f d d
pe pen fan lose pen oub an m pen/ lose ow a pen lose pen pen indo lose im en
:I g :d :o Y: : o Y: : c : o : d : f : o H: c ir fl 7: o 5: c 7: o H: o R: w H: c r
pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 B A S S S B L B Ex
12 27
VEL Stair7 TopE
VEL Stair7 MidE
10 VEL Stair7 BotE 22

8 18

Velocity (mph)
Velocity (m/s)

6 13

4 9

2 4

0 0

-2 -4
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-30. Velocity versus time for the 7th floor stairwell bi-directional probes, Experiment 7E

511
 ted
l ing 0.08
m
a r g n ed .08m off alf i l ing mpl e
i e
cl vin n o t fa o
fa ing g on d g us 0 fan h g c
ion le ing en n i n nin ble mo /fa ed adj ning ed/ ned nin ow ed me
nt
I gni t do ub open :
e
op ope : fan clos ope dou fan open clos flow ope clos ope ope wind clos r i
: : : Y : Y : : : : : : : : : : : : pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S 5 S 7 B H LR B H Ex
6 13
VEL Stair1 TopE
VEL Stair1 MidE
VEL Stair1 BotE
4 9

2 4

Velocity (mph)
Velocity (m/s)

0 0

-2 -4

-4 -9

-6 -13
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-31. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 7E
ted
l i ng 0 8m
. a r g n ed .0 8 m off f
l i l in g
mpl e
ai g 0 g cle vin an o st 0 /fan ha g a o
n ef nin ing on d ing le dju ing ed in wf d tc
n itio oubl penin o
o f d d
pe pen fan lose pen oub an m pen/ lose ow a pen lose pen pen indo lose im en
:I g :d :o Y: : o Y: : c : o : d : f : o H: c ir fl 7: o 5: c 7: o H: o R: w H: c r
pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 B A S S S B L B Ex
5 11

0 0

Velocity (mph)
Velocity (m/s)

-5 -11

-10 -22

-15 -34
VEL Bulkhead TopE
VEL Bulkhead MidE
VEL Bulkhead BotE
-20 -45
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-32. Velocity versus time for the bulkhead bi-directional probes, Experiment 7E

512
 ted
l ing 0.08
m
a r g n ed .08m off alf i l ing mpl e
i e
cl vin n o t fa o
fa ing g on d g us 0 fan h g c
ion le ing en n i n nin ble mo /fa ed adj ning ed/ ned nin ow ed me
nt
I gni t do ub open :
e
op ope : fan clos ope dou fan open clos flow ope clos ope ope wind clos r i
: : : Y : Y : : : : : : : : : : : : pe
BR BR S7 L B S 1 LB S 7 B H B R S 1 S 5 BH Air S7 S 5 S 7 B H LR B H Ex
30 67
VEL Apt7E TopE
VEL Apt7E MidE
25 VEL Apt7E BotE 56

20 45

Velocity (mph)
Velocity (m/s)

15 34

10 22

5 11

0 0

-5 -11
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-33. Velocity versus time for the apartment 7E doorway bi-directional probes, Experiment 7E

9.4.4 Experiment 7A 
9.4.4.1 Temperatures
) le) )
b le) ble le) oub ble
u b
ou o ou ( do
d u te
d
g
in ng (d e (d m g ar g (d CD d ( le
s r l f D c 8 in e p
clo il
i lea ha C la .0
W p 0 ail le in W pp om
n g/ e fa ing e c g ng g g in g g w f w c low ing to tc
ti io enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn p o o o p p p lo la p p lo n l o p in in B e B n n r im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xpe
B B
S B L B S B S S B B S S S S B S L L B B B S S R R E

1200 2192
TC BR7AE 2.1m
TC BR7AE 1.8m
TC BR7AE 1.5m
1000 TC BR7AE 1.2m 1832
TC BR7AE 0.9m
TC BR7AE 0.6m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-34. Temperature versus time for the 7A bedroom thermocouple tree, Experiment 7A

513
 ) e)
le) ble le) ubl le)
b
u ou b do ub d
g (do ( d d ou D ( (do le
te
s n
i ng r f D c e 8m n g r ( C d p
i l i a g e
clo il lea ha C pla 0.0 ail le in W pp om
on i ng/ le fa ning le c ing ing ing d ng W in ing d n d i ng w f w c flow ving sto ff n ntc
iti n b e b n n n e c i D n n e o s n d d N e o o o N o o e
gn pe ou op ou pe pe pe los la C pe pe los n lo pe in in B em B n n er
im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xp
B S B LB B S B S S B B S S S S B S LR LR B B B S S E

1200 2192
TC Corr7AE C
TC Corr7AE 2.1m
TC Corr7AE 1.8m
1000 TC Corr7AE 1.5m
1832
TC Corr7AE 1.2m
TC Corr7AE 0.9m

Temperature (F)
TC Corr7AE 0.6m
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-35. Temperature versus time for the 7A hall thermocouple tree, Experiment 7A
) e)
le) ble le) ubl le)
b u b d ub o
ou o ou D ( (do te
d
i n g (d e (d m g (d le
s i ng r l f D c 8 in ar g C e d p
clo il lea ha C la .0
W p 0 ail le in W pp om
n g/ e fa ing e c g ng g g in g g w f w c low ing to tc
ti io enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn p o o o p p p lo la p p lo n l o p in in B e B n n r im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xpe
B B
S B L B S B S S B B S S S S B S L L B B B S S R R E

800 1472
TC Corr7EE C
TC Corr7EE 2.1m
TC Corr7EE 1.8m
TC Corr7EE 1.5m
TC Corr7EE 1.2m
600 TC Corr7EE 0.9m
1112

Temperature (F)
TC Corr7EE 0.6m
Temperature (C)

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-36. Temperature versus time for the 7E hall thermocouple tree, Experiment 7A

514
 ) e)
le) ble le) ubl le)
b
u ou b do ub d
g (do ( d d ou D ( (do le
te
s n
i ng r f D c e 8m n g r ( C d p
l i a g e
clo il
i lea ha C pla 0.0 ail le in W pp om
on i ng/ le fa ning le c ing ing ing d ng W in ing d n d i ng w f w c flow ving sto ff n ntc
iti n b e b n n n e c i D n n e o s n d d N e o o o N o o e
gn pe ou op ou pe pe pe los la C pe pe los n lo pe in in B em B n n er
im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xp
B S B LB B S B S S B B S S S S B S LR LR B B B S S E

1200 2192
TC CorrStair7E C
TC CorrStair7E 2.1m
TC CorrStair7E 1.8m
1000 TC CorrStair7E 1.5m
1832
TC CorrStair7E 1.2m
TC CorrStair7E 0.9m

Temperature (F)
TC CorrStair7E 0.6m
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-37. Temperature versus time for the 7th floor stair hall thermocouple tree, Experiment 7A
) e)
le) ble le) ubl le)
b u b d ub o
ou o ou D ( (do te
d
i n g (d e (d m g (d le
s i ng r l f D c 8 in ar g C e d p
clo il lea ha C la .0
W p 0 ail le in W pp om
n g/ e fa ing e c g ng g g in g g w f w c low ing to tc
ti io enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn p o o o p p p lo la p p lo n l o p in in B e B n n r im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xpe
B B
S B L B S B S S B B S S S S B S L L B B B S S R R E
1000 1832
TC Stair7E C
TC Stair7E 2.1m
TC Stair7E 1.8m
800 TC Stair7E 1.5m 1472
TC Stair7E 1.2m
TC Stair7E 0.9m

Temperature (F)
TC Stair7E 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-38. Temperature versus time for the 7th floor stairwell thermocouple tree, Experiment 7A

515
 ) e)
le) ble le) ubl le)
b
u ou b do ub d
g (do ( d d ou D ( (do le
te
s n
i ng r f D c e 8m n g r ( C d p
i l i a g e
clo il lea ha C pla 0.0 ail le in W pp om
on i ng/ le fa ning le c ing ing ing d ng W in ing d n d i ng w f w c flow ving sto ff n ntc
iti n b e b n n n e c i D n n e o s n d d N e o o o N o o e
gn pe ou op ou pe pe pe los la C pe pe los n lo pe in in B em B n n er
im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xp
B S B LB B S B S S B B S S S S B S LR LR B B B S S E

800 1472
TC Bulkhead MidE
TC Stair 6E
TC Stair 4E
TC Stair 2E
TC Stair1 MidE
600 1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-39. Temperature versus time for the stair thermocouples, Experiment 7A

9.4.4.2 Pressures
)
) le) ble )
uble ub b le) ou uble
o o
(d (d ou ( do
d ed
g g ar g (d CD d ( et
s in ng
i a r l f D ace 08m l in e pl
clo il ha C l . ai le in W m
le W p 0 pp o
n g/ e fa ing e c g ng g g in g g w f w c low ing to tc
i tio enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn op do : o do op op op clo pla W op op clo fan clo op win win FB re FB f an fan r im
R:I 7 : R: BY R: 1: H: 7: 7: R: R: 7: 7: 7: 1: H: 7: R: R: R: R: R: 1: 1: xpe
B S B L B S B S S B B S S S S B S L L B B B S S E

30
DP Stair 7E
DP Stair 6E
DP Stair 5E
DP Stair 4E
DP Stair 3E
20 DP Stair 2E
DP Stair 1E
Pressure (Pa)

10

-10
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-40. Differential pressure versus time for the stairwell bi-directional probes, Experiment 7A

516
9.4.4.3 Velocities
)
) le) ble )
uble ub b le) ou uble
o o u ( o d d
g (d e (d m do CD d (d te
i n g r D c 8 g r ( p le
s in lf in a g e
clo il lea ha C pla 0.0 ail le in W pp om
on i ng/ le fa ning le c ing ing ing d ng W in ing d n d i ng w f w c flow ving sto ff n ntc
ti i e o o o e
ni en ub pe ub en en en se ac CD en en se o s en nd nd N m N o o im
: I g : op : do Y: o : do : op : op : op : clo : pl : W : op : op : clo : fan : clo : op : wi : wi : FB : re : FB : f an : fan p er
B R S 7 B R LB B R S 1 B H S 7 S 7 B R B R S 7 S 7 S 7 S 1 B H S 7 LR LR B R B R B R S 1 S 1 Ex
20 45
VEL Stair7 TopE
VEL Stair7 MidE
VEL Stair7 BotE
15 34

Velocity (mph)
Velocity (m/s)

10 22

5 11

0 0

-5 -11
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-41. Velocity versus time for the 7th floor stairwell bi-directional probes, Experiment 7A
) e)
le) ble le) ubl le)
b
u ou b d ub o d
g (do ( d dou D ( (do le
te
s n
i ng r f D c e 8m n g r ( C d p
i l i a g e
clo il lea ha C pla 0.0 ail le in W pp om
on i ng/ le fa ning le c ing ing ing d ng W in ing d n d i ng w f w c flow ving sto ff n ntc
iti n b e b n n n e c i D n n e o s n d d N e o o o N o o e
gn pe ou op ou pe pe pe los la C pe pe los n lo pe in in B em B n n r im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa pe
B B
S B L B S B S S B B S S S S B S L L B B B S S R R Ex
6 13
VEL Stair1 TopE
VEL Stair1 MidE
VEL Stair1 BotE
4 9

Velocity (mph)
Velocity (m/s)

2 4

0 0

-2 -4

-4 -9
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-42. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 7A

517
 ) e)
le) ble le) ubl le)
b
u ou b do ub d
g (do ( d d ou D ( (do le
te
s n
i ng r f D c e 8m n g r ( C d p
l i a g e
clo il
i lea ha C pla 0.0 ail le in W pp om
on i ng/ le fa ning le c ing ing ing d ng W in ing d n d i ng w f w c flow ving sto ff n ntc
iti n b e b n n n e c i D n n e o s n d d N e o o o N o o e
gn pe ou op ou pe pe pe los la C pe pe los n lo pe in in B em B n n er
im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xp
B S B LB B S B S S B B S S S S B S LR LR B B B S S E
5 11

0 0

Velocity (mph)
Velocity (m/s)

-5 -11

-10 -22

-15 -34
VEL Bulkhead TopE
VEL Bulkhead MidE
VEL Bulkhead BotE
-20 -45
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-43. Velocity versus time for the bulkhead bi-directional probes, Experiment 7A
) e)
le) ble le) ubl le)
b u b d ub o
ou o ou D ( (do te
d
i n g (d e (d m g (d le
s i ng r l f D c 8 in ar g C e d p
clo il lea ha C la .0
W p 0 ail le in W pp om
n g/ e fa ing e c g ng g g in g g w f w c low ing to tc
ti io enin ubl pen ubl enin eni enin sed cin CD enin en sed on sed enin do do N f m ov N s off on en
gn p o o o p p p lo la p p lo n l o p in in B e B n n r im
R: I 7: o R: d Y: R: d 1: o H: o 7: o 7: c R: p R: W 7: o 7: o 7: c 1: fa H: c 7: o : w : w R: F R: r R: F 1: f a 1: fa xpe
B B
S B L B S B S S B B S S S S B S L L B B B S S R R E
40 89
VEL Apt7A TopE
VEL Apt7A MidE
VEL Apt7A BotE
30 67

Velocity (mph)
Velocity (m/s)

20 45

10 22

0 0

-10 -22
0 300 600 900 1200 1500 1800 2100 2400 2700
Time (s)
Figure 9.4-44. Velocity versus time for the apartment 7A doorway bi-directional probes, Experiment 7A

518
 Temperature (C) Temperature (C)

0
200
400
600
800
1000
0
200
400
600
800
1000
1200

0
0
BR: Ignition BR: Ignition

S7: closed S7: closed

BR: double failing BR: double failing

200
200
S7: opening 0.08m S7: opening 0.08m
S7: closed S7: closed
S7: opening S7: opening
9.4.5.1 Temperatures

BH: opening BH: opening

9.4.5 Experiment 7K 

BH: closed BH: closed

400
400
LBY: opening LBY: opening
S7: closed S7: closed
S1: opening S1: opening
BH: opening BH: opening
S7: opening S7: opening

600
600
S7: closed S7: closed
LR: window failing LR: window failing
BR: placing WCD (double) BR: placing WCD (double)
S7: opening 0.08m S7: opening 0.08m
S7: open S7: open

800
800
S7: closed S7: closed
S1: fan on S1: fan on
BH: closed BH: closed
S7: opening S7: opening

Time (s)
Time (s)

519
1000
1000

TC Corr7KW C

TC Corr7KW 0.6m
TC Corr7KW 0.9m
TC Corr7KW 1.2m
TC Corr7KW 1.5m
TC Corr7KW 1.8m
TC Corr7KW 2.1m

1200
1200

Figure 9.4-46. Temperature versus time for the 7K hall thermocouple tree, Experiment 7K
TC BR7KW 0.6m
TC BR7KW 0.9m
TC BR7KW 1.2m
TC BR7KW 1.5m
TC BR7KW 1.8m

1400
1400

Figure 9.4-45. Temperature versus time for the 7K bedroom thermocouple tree, Experiment 7K

BR: WCD removed (double) BR: WCD removed (double)

LR: FBN flowing LR: FBN flowing
LR: FBN stopped LR: FBN stopped

1600
1600

Removing air flow Removing air flow

Experiment completed Experiment completed

1800
1800

32
32

392
752
392
752

1112
1472
1832
1112
1472
1832
2192

Temperature (F) Temperature (F)

 Temperature (C) Temperature (C)

0
100
200
300
400
500
600
0
100
200
300
400
500
600

0
0
BR: Ignition BR: Ignition

S7: closed S7: closed

BR: double failing BR: double failing

200
200
S7: opening 0.08m S7: opening 0.08m
S7: closed S7: closed
S7: opening S7: opening
BH: opening BH: opening
BH: closed BH: closed

400
400
LBY: opening LBY: opening
S7: closed S7: closed
S1: opening S1: opening
BH: opening BH: opening
S7: opening S7: opening

600
600
S7: closed S7: closed
LR: window failing LR: window failing
BR: placing WCD (double) BR: placing WCD (double)
S7: opening 0.08m S7: opening 0.08m
S7: open S7: open

800
800
S7: closed S7: closed
S1: fan on S1: fan on
BH: closed BH: closed
S7: opening S7: opening

Time (s)
Time (s)

520
1000
1000
TC Corr7GW C

1200
1200
TC Corr7GW 0.6m
TC Corr7GW 0.9m
TC Corr7GW 1.2m
TC Corr7GW 1.5m
TC Corr7GW 1.8m
TC Corr7GW 2.1m

TC CorrStair7W 0.6m
TC CorrStair7W 0.9m
TC CorrStair7W 1.2m
TC CorrStair7W 1.5m
TC CorrStair7W 1.8m
TC CorrStair7W 2.1m
Figure 9.4-47. Temperature versus time for the 7G hall thermocouple tree, Experiment 7K

1400
1400

BR: WCD removed (double) BR: WCD removed (double)

Figure 9.4-48. Temperature versus time for the 7th floor stair hall thermocouple tree, Experiment 7K
LR: FBN flowing LR: FBN flowing
LR: FBN stopped LR: FBN stopped

1600
1600

Removing air flow Removing air flow

Experiment completed Experiment completed

1800
1800

32
32

212
392
572
752
932
212
392
572
752
932

1112
1112

Temperature (F) Temperature (F)

 Temperature (C) Temperature (C)

0
100
200
300
400
0
100
200
300
400
500

0
0
BR: Ignition BR: Ignition

S7: closed S7: closed

BR: double failing BR: double failing

200
200
S7: opening 0.08m S7: opening 0.08m
S7: closed S7: closed
S7: opening S7: opening
BH: opening BH: opening
BH: closed BH: closed

400
400
LBY: opening LBY: opening
S7: closed S7: closed
S1: opening S1: opening
BH: opening BH: opening
S7: opening S7: opening

600
600
S7: closed S7: closed
LR: window failing LR: window failing
BR: placing WCD (double) BR: placing WCD (double)
S7: opening 0.08m S7: opening 0.08m
S7: open S7: open

800
800
S7: closed S7: closed
S1: fan on S1: fan on
BH: closed BH: closed
S7: opening S7: opening

Time (s)
Time (s)

521
1000
1000

1200
1200
TC Stair7W 0.6m
TC Stair7W 0.9m
TC Stair7W 1.2m
TC Stair7W 1.5m
TC Stair7W 1.8m
TC Stair7W 2.1m

Figure 9.4-50. Temperature versus time for the stair thermocouples, Experiment 7K
TC Stair 2W
TC Stair 4W
TC Stair 6W

TC Stair1 MidW
TC Bulkhead MidW

1400
1400

BR: WCD removed (double) BR: WCD removed (double)

Figure 9.4-49. Temperature versus time for the 7th floor stairwell thermocouple tree, Experiment 7K

LR: FBN flowing LR: FBN flowing

LR: FBN stopped LR: FBN stopped

1600
1600

Removing air flow Removing air flow

Experiment completed Experiment completed

1800
1800

32
32

212
392
572
752
212
392
572
752
932

Temperature (F) Temperature (F)

 Velocity (m/s) Pressure (Pa)

10
-20
20
40
60
80

-2
0
2
4
6
8
0
100

0
0
BR: Ignition BR: Ignition

S7: closed S7: closed

BR: double failing BR: double failing

200
9.4.5.3 Velocities
9.4.5.2 Pressures

200
S7: opening 0.08m S7: opening 0.08m
S7: closed S7: closed
S7: opening S7: opening
BH: opening BH: opening
BH: closed BH: closed

400
LBY: opening

400
LBY: opening
S7: closed S7: closed
S1: opening S1: opening
BH: opening BH: opening
S7: opening S7: opening

600
S7: closed S7: closed

600
LR: window failing LR: window failing
BR: placing WCD (double) BR: placing WCD (double)
S7: opening 0.08m S7: opening 0.08m
S7: open S7: open

800
S7: closed S7: closed

800
S1: fan on S1: fan on
BH: closed BH: closed
S7: opening

Time (s)
S7: opening

522
Time (s)

1000
1000

1200
1200
DP Stair 1W
DP Stair 2W
DP Stair 3W
DP Stair 4W
DP Stair 5W
DP Stair 6W
DP Stair 7W

VEL Stair7 BotW

VEL Stair7 MidW
VEL Stair7 TopW

1400
BR: WCD removed (double)

Figure 9.4-52. Velocity versus time for the 7th floor stairwell bi-directional probes, Experiment 7K
1400

LR: FBN flowing

Figure 9.4-51. Differential pressure versus time for the stairwell bi-directional probes, Experiment 7K

BR: WCD removed (double)

LR: FBN stopped

1600
LR: FBN flowing
Removing air flow
LR: FBN stopped
1600

Experiment completed
Removing air flow
Experiment completed

1800
0
4
9

-4
13
18
22
1800

Velocity (mph)
 Velocity (m/s) Velocity (m/s)

-20
-10
10
20
30

0
-6
-4
-2
0
2
4

0
0
BR: Ignition BR: Ignition

S7: closed S7: closed

BR: double failing BR: double failing

200
200
S7: opening 0.08m S7: opening 0.08m
S7: closed S7: closed

VEL Stair1 BotW

VEL Stair1 MidW
VEL Stair1 TopW
S7: opening S7: opening
BH: opening BH: opening
BH: closed BH: closed

400
400
LBY: opening LBY: opening
S7: closed S7: closed
S1: opening S1: opening
BH: opening BH: opening
S7: opening S7: opening

600
600
S7: closed S7: closed
LR: window failing LR: window failing
BR: placing WCD (double) BR: placing WCD (double)
S7: opening 0.08m S7: opening 0.08m
S7: open S7: open

800
800
S7: closed S7: closed
S1: fan on S1: fan on
BH: closed BH: closed
S7: opening S7: opening

Time (s)
Time (s)

523
1000
1000

1200
1200

VEL Apt7K MidW

VEL Apt7K TopW

1400
1400

Figure 9.4-53. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 7K

BR: WCD removed (double) BR: WCD removed (double)

LR: FBN flowing LR: FBN flowing

Figure 9.4-54. Velocity versus time for the apartment 7K doorway bi-directional probes, Experiment 7K
LR: FBN stopped LR: FBN stopped

1600
1600

Removing air flow Removing air flow

Experiment completed Experiment completed

1800
1800

0
0
4
9

-9
-4

22
45
67

-45
-22
-13

Velocity (mph) Velocity (mph)

9.4.6 Experiment 5E  
9.4.6.1 Temperatures
BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S 5 B H
LB
Y S 1
ing
/ c S 5
:s W
ing BH S CD ing S5 H:o v B B S5 ent c
: I g (d b l : o : o : o : o lo : o l 1 o H : o om
nit oub e fail pen pen pen pen sing pen e f ail : clos : f an (sing air fl : clos pen : clos pen ple
ion le) ing ing ing ing ing 3x ing ing ed on le ow e d in g ed ing ted
)
1200 2192
TC BR5EE 2.1m
TC BR5EE 1.8m
TC BR5EE 1.5m
1000 TC BR5EE 1.2m 1832
TC BR5EE 0.9m
TC BR5EE 0.6m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-55. Temperature versus time for the 5E bedroom thermocouple tree, Experiment 5E
BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R en BR ing Rem r im
WC : do S B L B S i ng S : s W o B S ent
BR u H Y / i n B C v i H B
: I g D (d ble 5: op : op : op 1: op clos 5: op gle H: c S1: f D (s ng a S5: c : op H: c 5: op com
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le ing ing ing ing ing 3x ing ing ed on le t
) ) ow ed ing ed ing ed
250 482
TC Corr5EE C
TC Corr5EE 2.1m
TC Corr5EE 1.8m
200 TC Corr5EE 1.5m 392
TC Corr5EE 1.2m
TC Corr5EE 0.9m

Temperature (F)
TC Corr5EE 0.6m
Temperature (C)

150 302

100 212

50 122

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-56. Temperature versus time for the 5E hall thermocouple tree, Experiment 5E

524
 BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S5 B H
L B Y
i n : s
S 1 g/ c S 5 i n B S
WC o v in S B H B S5 ent c
: I g (d ble : op : op : op : op los : op gle H: c 1: f D (s g a 5: c : op H: c : op om
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le) ing ing ing ing ing 3x ing ing ed on le t
) ow ed ing ed ing ed
180 356
TC Corr5AE C
TC Corr5AE 2.1m
TC Corr5AE 1.8m
150 TC Corr5AE 1.5m 302
TC Corr5AE 1.2m
TC Corr5AE 0.9m

Temperature (F)
TC Corr5AE 0.6m
Temperature (C)

120 248

90 194

60 140

30 86

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-57. Temperature versus time for the 5A hall thermocouple tree, Experiment 5E
BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S 5 B H
LB
Y S 1
ing
/ c S 5
:s W
ing BH S CD ing S5 H:o v B B S5 ent c
: I g (d b l : o : o : o : o lo : o l 1 o H : o om
nit oub e fail pen pen pen pen sing pen e f ail : clos : f an (sing air fl : clos pen : clos pen ple
ion le ing ing ing ing ing 3x ing ing ed on le ow e d in g ed ing ted
) )
150 302
TC CorrStair5E 2.1m
TC CorrStair5E 1.8m
TC CorrStair5E 1.5m
120 TC CorrStair5E 1.2m 248
TC CorrStair5E 0.9m
TC CorrStair5E 0.6m

Temperature (F)
Temperature (C)

90 194

60 140

30 86

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-58. Temperature versus time for the 5th floor stair hall thermocouple tree, Experiment 5E

525
 BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S5 B H
L B Y
i n : s
S 1 g/ c S 5 i n B S
WC o v in S B H B S5 ent c
: I g (d ble : op : op : op : op los : op gle H: c 1: f D (s g a 5: c : op H: c : op om
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le) ing ing ing ing ing 3x ing ing ed on le t
) ow ed ing ed ing ed
120 248
TC Stair5E C
TC Stair5E 2.1m
TC Stair5E 1.8m
100 TC Stair5E 1.5m 212
TC Stair5E 1.2m
TC Stair5E 0.9m

Temperature (F)
TC Stair5E 0.6m
Temperature (C)

80 176

60 140

40 104

20 68

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-59. Temperature versus time for the 5th floor stairwell thermocouple tree, Experiment 5E
BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S 5 B H
LB
Y S 1
ing
/ c S 5
:s
i n B S
WC o v in S B H B S5 ent c
: I g (d b l : o : o : o : o lo : o gl H 1 D g 5 : o H : o om
nit oub e fail pen pen pen pen sing pen e f ail : clos : f an (sing air fl : clos pen : clos pen ple
ion le ing ing ing ing ing 3x ing ing ed on le o w e d in g ed ing ted
) )
100 212
TC Bulkhead MidE
TC Stair 4E
TC Stair 2E
80 TC Stair1 MidE 176

Temperature (F)
Temperature (C)

60 140

40 104

20 68

0 32
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-60. Temperature versus time for the stair thermocouples, Experiment 5E

526
9.4.6.2 Pressures
BR BR
:p S5 :p Ex
lac :o lac pe
i ng B R pe B i n Re r im
W : n R g m
B R C D d oub S 5 B H LB
Y S1 ing/ c S5 : sin B S
WC ov
D in S BH B S5 ent c
: I g (d l : o : o : o : o lo : o gl H 1 g 5 : o H : o om
nit oub e fail pen pen p e p e si n p e : : (s a :c p :
nin nin g 3 enin f ailin close f an o ingle ir flo lose enin close enin plete
p
ion le ing ing ing g g x g g d n w d g d g d
) )
25
DP Stair 7E
DP Stair 6E
20 DP Stair 5E
DP Stair 3E
DP Stair 2E
15 DP Stair 1E
Pressure (Pa)

10

-5

-10
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-61. Differential pressure versus time for the stairwell bi-directional probes, Experiment 5E

9.4.6.3 Velocities
BR BR
:p S5 :p Ex
lac : o lac pe
i ng B R pe BR ing Rem r im
WC : do ni n
BR S B L B S g S : si W C o v B S ent
: I g D (d uble 5: op H: op Y: op 1: op / clos 5: op ngle BH: c S1: f D (s ing a S5: c H: op BH: c 5: op com
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le) ing ing ing ing ing 3x ing ing ed on le t
) ow ed ing ed ing ed
2.5 6
VEL Stair5 TopE
VEL Stair5 MidE
2.0 VEL Stair5 BotE 4

1.5 3

Velocity (mph)
Velocity (m/s)

1.0 2

0.5 1

0.0 0

-0.5 -1

-1.0 -2
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-62. Velocity versus time for the 5th floor stairwell bi-directional probes, Experiment 5E

527
 BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S5 B H
L B Y
i n : s
S 1 g/ c S 5 i n B S
WC o v in S B H B S5 ent c
: I g (d ble : op : op : op : op los : op gle H: c 1: f D (s g a 5: c : op H: c : op om
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le) ing ing ing ing ing 3x ing ing ed on le t
) ow ed ing ed ing ed
5 11
VEL Stair1 TopE
VEL Stair1 BotE
4 9

3 7

Velocity (mph)
Velocity (m/s)

2 4

1 2

0 0

-1 -2
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-63. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 5E
BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S 5 B H
LB
Y S 1
ing
/ c S 5
:s
i n B S
WC o v in S B H B S5 ent c
: I g (d b l : o : o : o : o lo : o gl H 1 D g 5 : o H : o om
nit oub e fail pen pen pen pen sing pen e f ail : clos : f an (sing air fl : clos pen : clos pen ple
ion le ing ing ing ing ing 3x ing ing ed on le o w e d in g ed ing ted
) )
2 4

1 2

0 0

Velocity (mph)
Velocity (m/s)

-1 -2

-2 -4

-3 VEL Bulkhead TopE

-7
VEL Bulkhead MidE
VEL Bulkhead BotE
-4 -9
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-64. Velocity versus time for the bulkhead bi-directional probes, Experiment 5E

528
 BR BR
:p S5 :p Ex
lac : op lac pe
i ng B R e B ing Rem r im
WC : do n R
BR D u S5 B H
L B Y
i n : s
S 1 g/ c S 5 i n B S
WC o v in S B H B S5 ent c
: I g (d ble : op : op : op : op los : op gle H: c 1: f D (s g a 5: c : op H: c : op om
nit oub fail en en en en ing en f ail los an ing ir fl los en los en ple
ion le) ing ing ing ing ing 3x ing ing ed on le t
) ow ed ing ed ing ed
1.0 2

0.5 1

Velocity (mph)
Velocity (m/s)

0.0 0

-0.5 -1

-1.0 -2
VEL Apt5E TopE
VEL Apt5E MidE
VEL Apt5E BotE
-1.5 -3
0 200 400 600 800 1000 1200 1400 1600 1800
Time (s)
Figure 9.4-65. Velocity versus time for the apartment 5E doorway bi-directional probes, Experiment 5E

9.4.7 Experiment 5E2 
9.4.7.1 Temperatures
Ex
LR pe
LR L L r
:w :w L
LB B B R
Y
L
Y
L
:
L BY : F BY: im en
R
LR BH ind B in Y F
LB : l : f a : f a B : f a B N l e f t c
: Ig :o ow H: do Y: eft n n N n s t o om
nit pe
nin fai clo w c f an clo mo mo flow mo top pen ple
ion lin se lea o se ve ve in ve pe in te
g g d r n d d d g d d g d

1200 2192
TC BR5EE 2.1m
TC BR5EE 1.8m
TC BR5EE 1.5m
1000 TC BR5EE 1.2m 1832
TC BR5EE 0.9m
TC BR5EE 0.6m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-66. Temperature versus time for the 5E2 bedroom thermocouple tree, Experiment 5E2

529
 Ex
LR pe
:w LR
:w LB LB LB R LB R: BY r im e
L L L
BH ind L Y Y: Y: : F Y : F B : l nt
LR ow BH: indo BY: : left fan f an BN f an N s eft o com
: Ig :o
pe fai c w f a c m m fl o m to p p
nit
ion nin lin lose clea n o lose ove ove win ove ppe enin lete
g g d r n d d d g d d g d

1200 2192
TC Corr5EE C
TC Corr5EE 2.1m
TC Corr5EE 1.8m
1000 TC Corr5EE 1.5m 1832
TC Corr5EE 1.2m
TC Corr5EE 0.9m

Temperature (F)
TC Corr5EE 0.6m
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-67. Temperature versus time for the 5E2 hall thermocouple tree, Experiment 5E2
Ex
p
LR
:w LR L L L L L LR LB er im
: B B B R B : Y e
LR BH ind w L Y Y: Y: : F Y: FB : l nt
: Ig :o ow BH: indo BY: : left fan f an BN f an N s eft o com
nit pe
nin fai clo w c f an clo mo mo flow mo top pen ple
ion lin se lea o se ve ve in ve pe in te
g g d r n d d d g d d g d

800 1472
TC Corr5AE C
TC Corr5AE 2.1m
TC Corr5AE 1.8m
TC Corr5AE 1.5m
TC Corr5AE 1.2m
600 TC Corr5AE 0.9m
1112

Temperature (F)
TC Corr5AE 0.6m
Temperature (C)

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-68. Temperature versus time for the 5A hall thermocouple tree, Experiment 5E2

530
 Ex
LR pe
:w LR
:w LB LB LB R LB R: BY r im e
L L L
BH ind L Y Y: Y: : F Y : F B : l nt
LR ow BH: indo BY: : left fan f an BN f an N s eft o com
: Ig :o
pe fai c w f a c m m fl o m to p p
nit
ion nin lin lose clea n o lose ove ove win ove ppe enin lete
g g d r n d d d g d d g d

1000 1832
TC CorrStair5E C
TC CorrStair5E 2.1m
TC CorrStair5E 1.8m
800 TC CorrStair5E 1.5m 1472
TC CorrStair5E 1.2m
TC CorrStair5E 0.9m

Temperature (F)
TC CorrStair5E 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-69. Temperature versus time for the 5th floor stair hall thermocouple tree, Experiment 5E2
Ex
p
LR
:w LR L L L L L LR LB er im
: B B B R B : Y e
LR BH ind w L Y Y: Y: : F Y: FB : l nt
: Ig :o ow BH: indo BY: : left fan f an BN f an N s eft o com
nit pe
nin fai clo w c f an clo mo mo flow mo top pen ple
ion lin se lea o se ve ve in ve pe in te
g g d r n d d d g d d g d

800 1472
TC Stair5E C
TC Stair5E 2.1m
TC Stair5E 1.8m
TC Stair5E 1.5m
TC Stair5E 1.2m
600 TC Stair5E 0.9m
1112

Temperature (F)
TC Stair5E 0.6m
Temperature (C)

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-70. Temperature versus time for the 5th floor stairwell thermocouple tree, Experiment 5E2

531
 Ex
LR pe
:w LR
:w LB LB LB R LB R: BY r im e
L L L
BH ind L Y Y: Y: : F Y : F B : l nt
LR ow BH: indo BY: : left fan f an BN f an N s eft o com
: Ig :o
pe fai c w f a c m m fl o m to p p
nit
ion nin lin lose clea n o lose ove ove win ove ppe enin lete
g g d r n d d d g d d g d

1500 2732
TC Bulkhead MidE
TC Stair 4E
TC Stair 2E
1200 TC Stair1 MidE 2192

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-71. Temperature versus time for the stair thermocouples, Experiment 5E2

9.4.7.2 Pressures
Ex
p
LR
:w LR L L L L L LR LB er im
: w B R
BY BY : BY : F Y: en
LR BH ind L Y : : F : B l t
: Ig :o ow BH: indo BY: : left f an f an BN fan N s eft o com
nit pe f ai clo w c f an clo mo mo flow mo top pe pl
ion nin lin se lea o se ve ve in ve pe nin ete
g g d r n d d d g d d g d

100
DP Stair 7E
DP Stair 6E
DP Stair 5E
80 DP Stair 4E
DP Stair 3E
DP Stair 2E
60 DP Stair 1E
Pressure (Pa)

40

20

-20
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-72. Differential pressure versus time for the stairwell bi-directional probes, Experiment 5E2

532
9.4.7.3 Velocities
Ex
p
LR
:w LR L L L L L LR LB er im
: w B B Y B Y R: B Y : F Y: en
LR BH ind L Y : : F : B l t
: Ig :o ow BH: indo BY: : left fan f an BN f an N s eft o com
nit pe
nin fai clo w c f an clo mo mo flow mo top pen ple
ion lin se lea o se ve ve in ve pe in te
g g d r n d d d g d d g d

12 27
VEL Stair5 TopE
VEL Stair5 MidE
VEL Stair5 BotE
9 20

Velocity (mph)
Velocity (m/s)

6 13

3 7

0 0

-3 -7
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-73. Velocity versus time for the 5th floor stairwell bi-directional probes, Experiment 5E2
Ex
LR pe
LR L L r
:w :w L
LB B B R
Y
L
Y
L
:
L BY : F BY: im en
R
LR BH ind B in Y F
LB : l : f a : f a B : f a B N l e f t c
: Ig :o ow H: do Y: eft n n N n s t o om
nit pe
nin fai clo w c f an clo mo mo flow mo top pen ple
ion lin se lea o se ve ve in ve pe in te
g g d r n d d d g d d g d

2.0 4
VEL Stair1 TopE
VEL Stair1 MidE
VEL Stair1 BotE
1.5 3

1.0 2

Velocity (mph)
Velocity (m/s)

0.5 1

0.0 0

-0.5 -1

-1.0 -2
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-74. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 5E2

533
 Ex
LR pe
:w LR
:w LB LB LB R LB R: BY r im e
L L L
BH ind L Y Y: Y: : F Y : F B : l nt
LR ow BH: indo BY: : left fan f an BN f an N s eft o com
: Ig :o
pe fai c w f a c m m fl o m to p p
nit
ion nin lin lose clea n o lose ove ove win ove ppe enin lete
g g d r n d d d g d d g d

10 22

5 11

0 0

Velocity (mph)
Velocity (m/s)

-5 -11

-10 -22

-15 VEL Bulkhead TopE

-34
VEL Bulkhead MidE
VEL Bulkhead BotE
-20 -45
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-75. Velocity versus time for the bulkhead bi-directional probes, Experiment 5E2
Ex
p
LR
:w LR L L L L L LR LB er im
: B B B R B : Y e
LR BH ind w L Y Y: Y: : F Y: FB : l nt
: Ig :o ow BH: indo BY: : left fan f an BN f an N s eft o com
nit pe
nin fai clo w c f an clo mo mo flow mo top pen ple
ion lin se lea o se ve ve in ve pe in te
g g d r n d d d g d d g d

12 27
VEL Apt5E TopE
VEL Apt5E MidE
10 VEL Apt5E BotE 22

8 18

Velocity (mph)
Velocity (m/s)

6 13

4 9

2 4

0 0

-2 -4
0 100 200 300 400 500 600 700 800 900 1000 1100
Time (s)
Figure 9.4-76. Velocity versus time for the apartment 5E doorway bi-directional probes, Experiment 5E2

534
9.4.8 Experiment 5K 
9.4.8.1 Temperatures
BR BR BR:
:p : F FB Ex
lac
BR B N
BR ing LR: N flo sto LR LR: Re per im
:d W w m o en
BR oub S5 LBY: S1 BH: : dou CD indo wing pped : F
B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w (d (d : c flo s to : f a pe : c ai
nit fail en en en en o f o o p
ion ing ing ing ing ing clear uble) ailing uble) uble) losed wing pped n on ning losed r flow leted

1200 2192
TC BR5KW 1.8m
TC BR5KW 1.5m
TC BR5KW 1.2m
1000 TC BR5KW 0.9m 1832
TC BR5KW 0.6m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-77. Temperature versus time for the 5K bedroom thermocouple tree, Experiment 5K
BR BR BR:
:p : F FB Ex
lac
BR B N
BR ng LR: N flo sto
i LR LR: Re per im
:d L : W w w p : F m e
BR oub S5 BY: S1 BH: dou CD indo ing ped B FB B BH B ovi nt
: I g le : op op : op op ble (d
o w f (do (do H: c N flo N sto S1: f : op H: c ng ai comp
nit fail en en en en a e
ion ing ing ing ing ing clea uble) ailing uble) uble) losed wing pped n on ning losed r flow leted
r
2500 4532
TC Corr5KW 2.1m
TC Corr5KW 1.8m
TC Corr5KW 1.5m
2000 TC Corr5KW 1.2m 3632
TC Corr5KW 0.9m
TC Corr5KW 0.6m

Temperature (F)
Temperature (C)

1500 2732

1000 1832

500 932

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-78. Temperature versus time for the 5K hall thermocouple tree, Experiment 5K

535
 BR BR BR:
:p : F FB Ex
lac
BR B i n g LR B N N s
f t L LR Re per im
: R W : w l o o R :F m o en
BR doub S5 LBY: S1 BH: : dou CD indo wing pped :
B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w ( ( : s
flo to : f a pe : c
nit fail en en en en o f d o do c ai p
ion ing ing ing ing ing clear uble) ailing uble) uble) losed wing pped n on ning losed r flow leted

1000 1832
TC Corr5GW C
TC Corr5GW 2.1m
TC Corr5GW 1.8m
800 TC Corr5GW 1.5m 1472
TC Corr5GW 1.2m
TC Corr5GW 0.9m

Temperature (F)
TC Corr5GW 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-79. Temperature versus time for the 5G hall thermocouple tree, Experiment 5K
BR BR BR:
:p : F FB Ex
lac
BR B N
BR ing LR: N flo sto LR LR: Re per im
:d W w p m o en
BR oub S5 LBY: S1 BH: : dou CD indo wing ped : F
B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w ( ( : s
nit fail en en en en o f d o d o c flo to : fa p e :c ai p
ion ing ing ing ing ing clear uble) ailing uble) uble) losed wing pped n on ning losed r flow leted

1000 1832
TC CorrStair5W C
TC CorrStair5W 2.1m
TC CorrStair5W 1.8m
800 TC CorrStair5W 1.5m 1472
TC CorrStair5W 1.2m
TC CorrStair5W 0.9m

Temperature (F)
TC CorrStair5W 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-80. Temperature versus time for the 5th floor stair hall thermocouple tree, Experiment 5K

536
 BR BR BR:
:p : F FB Ex
lac
BR B i n g LR B N N s
f t L LR Re per im
: R W : w l o o R :F m o en
BR doub S5 LBY: S1 BH: : dou CD indo wing pped :
B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w ( ( : s
flo to : f a pe : c
nit fail en en en en o f d o do c ai p
ion ing ing ing ing ing clear uble) ailing uble) uble) losed wing pped n on ning losed r flow leted

800 1472
TC Stair5W 2.1m
TC Stair5W 1.8m
TC Stair5W 1.5m
TC Stair5W 1.2m
TC Stair5W 0.9m
600 TC Stair5W 0.6m
1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-81. Temperature versus time for the 5th floor stairwell thermocouple tree, Experiment 5K
BR BR BR:
:p : F FB Ex
lac
BR B N
BR ing LR: N flo sto LR LR: Re per im
:d W w p m o en
BR oub S5 LBY: S1 BH: : dou CD indo wing ped : F
B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w ( ( : s
nit fail en en en en o f d o d o c flo to : fa p e :c ai p
ion ing ing ing ing ing clea uble ailing uble uble) losed wing pped n on ning losed r flow leted
r ) )
800 1472
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
600 1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-82. Temperature versus time for the stair thermocouples, Experiment 5K

537
9.4.8.2 Pressures
BR BR BR:
:p : F FB Ex
lac
BR B i ng LR: BN fl N sto L LR Re per im
: R W w o R : mo
BR doub S5 LBY: S1 BH : dou CD indo wing pped BH: FBN FBN S1 BH: o BH vin ent c
: I g le :o o :o :o b l (d w (d (d : f s t : : g a om
ni t f p e p e p e p e e o f o o c l o o f a p e c i r f pl e
ion ailing ning ning ning ning clear uble) ailing uble) uble losed wing pped n on ning losed low ted
)
60
DP Stair 7W
DP Stair 6W
50 DP Stair 5W
DP Stair 4W
DP Stair 3W
40 DP Stair 2W
DP Stair 1W
Pressure (Pa)

30

20

10

-10
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-83. Differential pressure versus time for the stairwell bi-directional probes, Experiment 5K

9.4.8.3 Velocities
BR BR BR:
:p : F FB Ex
lac
BR B N
BR ing LR: N flo sto LR LR: Re per im
:d L : W w w p : F m o en
BR oub S5 BY: S1 BH: dou CD indo ing ped B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w ( ( : s
nit fail en en en en o f d o d o c fl o to : f a p e :c ai p
ion ing ing ing ing ing clear uble) ailing uble) uble) losed wing pped n on ning losed r flow leted

20 45
VEL Stair5 MidW
VEL Stair5 BotW

15 34

Velocity (mph)
Velocity (m/s)

10 22

5 11

0 0

-5 -11
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-84. Velocity versus time for the 5th floor stairwell bi-directional probes, Experiment 5K

538
 BR BR BR:
:p : F FB Ex
lac
BR B i n g LR B N N s
f t L LR Re per im
: R W : w l o o R :F m o en
BR doub S5 LBY: S1 BH: : dou CD indo wing pped :
B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w ( ( : s
flo to : f a pe : c
nit fail en en en en o f d o do c ai p
ion ing ing ing ing ing clear uble) ailing uble) uble) losed wing pped n on ning losed r flow leted

4 9
VEL Stair1 TopW
VEL Stair1 MidW
VEL Stair1 BotW
3 7

Velocity (mph)
Velocity (m/s)

2 4

1 2

0 0

-1 -2
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-85. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 5K
BR BR BR:
:p : F FB Ex
lac
BR B N
BR ing LR: N flo sto LR LR: Re per im
:d W w p m o en
BR oub S5 LBY: S1 BH: : dou CD indo wing ped : F
B H FB N B N S1 BH: o BH ving t com
: I g le : op op : op op ble (d w ( ( : s
nit fail en en en en o f d o d o c flo to : fa p e :c ai p
ion ing ing ing ing ing clea uble ailing uble uble) losed wing pped n on ning losed r flow leted
r ) )
12 27
VEL Apt5K TopW
VEL Apt5K MidW
9 VEL Apt5K BotW 20

6 13

Velocity (mph)
Velocity (m/s)

3 7

0 0

-3 -7

-6 -13

-9 -20
0 200 400 600 800 1000 1200 1400 1600 1800 2000
Time (s)
Figure 9.4-86. Velocity versus time for the apartment 5K doorway bi-directional probes, Experiment 5K

539
9.4.9 Experiment 5A 
9.4.9.1 Temperatures
le) le) le) le)
a r (d oub (d oub doub (doub eted
g e g ( ed pl
ing in f cl g ar r D D
WC lowin topp t com
n le fail t open ng ng in g le hal f ailin le cle d c lea g WC on g f s n
o ub i g h eni eni ub ngle ub e ci n n s n
niti en ose ngl a a ovi BN FBN er im e
: I g R: do BY: r 1: op 5: op H: op R: do R: si R: do H: cl R: si R: pl BY: f :m :F : p
BR B L S S B B B B B B B L BR B R B R Ex
1500 2732
TC BR5AE 2.1m
TC BR5AE 1.8m
TC BR5AE 1.5m
1200 TC BR5AE 1.2m 2192
TC BR5AE 0.9m
TC BR5AE 0.6m

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-87. Temperature versus time for the 5A bedroom thermocouple tree, Experiment 5A
e) e) ) e)
ub l ubl ble ubl d
g n g c le ar
r D (do D (do (dou d (do plete
in i f g a ar C ing oppe com
n fail open g g g hal f ailin cle C
cl e g W o n g W low t t
gn itio ouble right peni n penin penin ouble i ngle ouble losed ingle lacin f ans o vin BN f BN s r im en
: I : d Y: : o : o : o : d : s : d : c : s : p Y: : m :F : F pe
BR B R LB S1 S 5 B H BR BR B R B H B R B R LB BR B R B R Ex
1000 1832
TC Corr5AE C
TC Corr5AE 2.1m
TC Corr5AE 1.8m
800 TC Corr5AE 1.5m 1472
TC Corr5AE 1.2m
TC Corr5AE 0.9m

Temperature (F)
TC Corr5AE 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-88. Temperature versus time for the 5A hall thermocouple tree, Experiment 5A

540
 e) e) ) e)
ub l ubl ble ubl d
g n g c le ar
r D (do D (do (dou d (do plete
in i f g a ar C ing oppe com
n fail open g g g hal f ailin cle C
cl e g W o n g W low t t
gn itio ouble right peni n penin penin ouble i ngle ouble losed ingle lacin f ans o vin BN f BN s r im en
: I : d Y: : o : o : o : d : s : d : c : s : p Y: : m :F : F pe
BR B R LB S1 S 5 B H BR BR B R B H B R B R LB BR B R B R Ex
250 482
TC Corr5EE C
TC Corr5EE 1.8m
TC Corr5EE 1.5m
200 TC Corr5EE 1.2m 392
TC Corr5EE 0.9m
TC Corr5EE 2.1m

Temperature (F)
Temperature (C)

150 302

100 212

50 122

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-89. Temperature versus time for the 5E hall thermocouple tree, Experiment 5A
le) ) )
oub ble ble) ble
g le a r (d (dou (dou (dou leted
ing enin fc g ar ar WCD n D g ed
WC owin topp t com
p
fail p g g g hal f ailin cle cl e
n i tion ouble right o eni n enin penin ouble i ngle ouble losed ingle lacing f ans o oving BN fl BN s im en
: I g R: d BY: 1: op 5: op H: o R: d R: s R: d H: c R: s R: p BY: :m :F : F per
BR B L S S B B B B B B B L BR B R B R Ex
1000 1832
TC CorrStair5E C
TC CorrStair5E 2.1m
TC CorrStair5E 1.8m
800 TC CorrStair5E 1.5m 1472
TC CorrStair5E 1.2m
TC CorrStair5E 0.9m

Temperature (F)
TC CorrStair5E 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-90. Temperature versus time for the 5th floor stair hall thermocouple tree, Experiment 5A

541
 e) e) ) e)
ub l ubl ble ubl d
g n g c le ar
r D (do D (do (dou d (do plete
in i f g a ar C ing oppe com
n fail open g g g hal f ailin cle C
cl e g W o n g W low t t
gn itio ouble right peni n penin penin ouble i ngle ouble losed ingle lacin f ans o vin BN f BN s r im en
: I : d Y: : o : o : o : d : s : d : c : s : p Y: : m :F : F pe
BR B R LB S1 S 5 B H BR BR B R B H B R B R LB BR B R B R Ex
800 1472
TC Stair5E C
TC Stair5E 2.1m
TC Stair5E 1.8m
TC Stair5E 1.5m
TC Stair5E 1.2m
600 TC Stair5E 0.9m
1112

Temperature (F)
TC Stair5E 0.6m
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-91. Temperature versus time for the 5th floor stairwell thermocouple tree, Experiment 5A
le) ) )
oub ble ble) ble
g le a r (d (dou (dou (dou leted
ing enin fc g ar ar WCD n D g ed
WC owin topp t com
p
fail p g g g hal f ailin cle cl e
n i tion ouble right o eni n enin penin ouble i ngle ouble losed ingle lacing f ans o oving BN fl BN s im en
: I g R: d BY: 1: op 5: op H: o R: d R: s R: d H: c R: s R: p BY: :m :F : F per
BR B L S S B B B B B B B L BR B R B R Ex
500 932
TC Bulkhead MidE
TC Stair 4E
TC Stair 2E
400 TC Stair1 MidE 752

Temperature (F)
Temperature (C)

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-92. Temperature versus time for the stair thermocouples, Experiment 5A

542
9.4.9.2 Pressures
) ) )
ble ble le) ble
a r (d ou (d ou oub (dou ted
g ni ng le d
D g ( ed pl e
li n l f c ng ar r CD
WClowin topp t com
n e fai ope g g n g e ha f aili e cle c l ea g W on g
l t n n i l l d f s n
itio ou
b igh eni eni en ub ngle ub e le i n s
los ing lac fan
in
ov BN BN im e
: I gn r p p p o
: d BY: 1: o 5: o H: o R: d R: s R: d
i o
: c R: s R: p BY: : m R: F R: F xper
BR B R L S S B B B B B H B B L R
B B B E
80
DP Stair 7E
DP Stair 6E
DP Stair 5E
60 DP Stair 4E
DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-93. Differential pressure versus time for the stairwell bi-directional probes, Experiment 5A

9.4.9.3 Velocities
le) le) le) le)
a r (d oub (d oub doub (doub eted
g e g ( ed pl
ing in f cl g ar r D D
WC lowin topp t com
n le fail t open ng ng in g le hal f ailin le cle d c lea g WC on g f s n
o ub i g h eni eni ub ngle ub e ci n n s n
niti en ose ngl a a ovi BN FBN er im e
: I g R: do BY: r 1: op 5: op H: op R: do R: si R: do H: cl R: si R: pl BY: f :m :F : p
BR B L S S B B B B B B B L BR B R B R Ex
12 27
VEL Stair5 TopE
VEL Stair5 MidE
VEL Stair5 BotE
9 20

6 13

Velocity (mph)
Velocity (m/s)

3 7

0 0

-3 -7

-6 -13
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-94. Velocity versus time for the 5th floor stairwell bi-directional probes, Experiment 5A

543
 e) e) ) e)
ub l ubl ble ubl d
g n g c le ar
r D (do D (do (dou d (do plete
in i f g a ar C ing oppe com
n fail open g g g hal f ailin cle C
cl e g W o n g W low t t
gn itio ouble right peni n penin penin ouble i ngle ouble losed ingle lacin f ans o vin BN f BN s r im en
: I : d Y: : o : o : o : d : s : d : c : s : p Y: : m :F : F pe
BR B R LB S1 S 5 B H BR BR B R B H B R B R LB BR B R B R Ex
2.5 6
VEL Stair1 TopE
VEL Stair1 BotE
2.0 4

1.5 3

Velocity (mph)
Velocity (m/s)

1.0 2

0.5 1

0.0 0

-0.5 -1

-1.0 -2
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-95. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 5A
le) ) )
oub ble ble) ble
g le a r (d (dou (dou (dou leted
ing enin fc g ar ar WCD n D g ed
WC owin topp t com
p
fail p g g g hal f ailin cle cl e
n i tion ouble right o eni n enin penin ouble i ngle ouble losed ingle lacing f ans o oving BN fl BN s im en
: I g R: d BY: 1: op 5: op H: o R: d R: s R: d H: c R: s R: p BY: :m :F : F per
BR B L S S B B B B B B B L BR B R B R Ex
12 27
VEL Apt5A TopE
VEL Apt5A MidE
VEL Apt5A BotE
9 20

6 13

Velocity (mph)
Velocity (m/s)

3 7

0 0

-3 -7

-6 -13
0 200 400 600 800 1000 1200 1400 1600
Time (s)
Figure 9.4-96. Velocity versus time for the apartment 5A doorway bi-directional probes, Experiment 5A

544
9.4.10 Experiment 5G 
9.4.10.1 Temperatures
gle) ted
g g 2 m r D (sin ting pl e
ilin enin 0. 0 lea WC ve n com
ition e fa p g g ng ec cing indow ent
ubl : left o penin penin openi open doubl
: Ig
n
: d o o o H: : : : p l a w er im
BR BR L B Y S 1: S 5 : B BH BR BR LR: E xp
1500 2732
TC BR5GW 2.1m
TC BR5GW 1.8m
TC BR5GW 1.5m
1200 TC BR5GW 1.2m 2192
TC BR5GW 0.9m
TC BR5GW 0.6m

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-97. Temperature versus time for the 5G bedroom thermocouple tree, Experiment 5G
gle) ted
g 2m (sin ting le
fa il i n
en i n g
0 . 0
cle a r
W C D
v n
e co mp
ition u ble left op ening ening pening pen ouble c ing indow ent
: Ig
n o
: d BY: o p p
o H: o H: o R: d l a
: p R: w er im
BR BR L S 1: S 5 : B B B BR L E xp
600 1112
TC Corr5GW 2.1m
TC Corr5GW 1.8m
TC Corr5GW 1.5m
500 TC Corr5GW 1.2m 932
TC Corr5GW 0.9m
TC Corr5GW 0.6m

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-98. Temperature versus time for the 5G hall thermocouple tree, Experiment 5G

545
 gle) ted
g g 2 m r D (sin ting pl e
i n n 0 a
fail en i 0 . cle W C ve n com
niti
on u ble left op ening ening pening pen ouble c ing indow ent
: Ig o
: d BY: o p p
o H: o H: o R: d l a
: p R: w er im
BR BR L S 1: S 5 : B B B BR L E xp
400 752
TC Corr5KW 2.1m
TC Corr5KW 1.8m
TC Corr5KW 1.5m
TC Corr5KW 1.2m
TC Corr5KW 0.9m
300 TC Corr5KW 0.6m
572

Temperature (F)
Temperature (C)

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-99. Temperature versus time for the 5K hall thermocouple tree, Experiment 5G
gle) ted
g g 2 m r D (sin ting pl e
i n n 0 a
fail eni 0 . cle W C ve n com
niti
on u ble left op ening ening pening pen ouble c ing indow ent
: Ig o
: d BY: op op H: o H: o R: d l a
: p R: w er im
BR BR L S 1: S 5 : B B B BR L E xp
600 1112
TC CorrStair5W 2.1m
TC CorrStair5W 1.8m
TC CorrStair5W 1.5m
500 TC CorrStair5W 1.2m 932
TC CorrStair5W 0.9m
TC CorrStair5W 0.6m

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-100. Temperature versus time for the 5th floor stair hall thermocouple tree, Experiment 5G

546
 gle) ted
g g 2 m r D (sin ting pl e
i n n 0 a
fail en i 0 . cle W C ve n com
niti
on u ble left op ening ening pening pen ouble c ing indow ent
: Ig o
: d BY: o p p
o H: o H: o R: d l a
: p R: w er im
BR BR L S 1: S 5 : B B B BR L E xp
500 932
TC Stair5W 2.1m
TC Stair5W 1.8m
TC Stair5W 1.5m
400 TC Stair5W 1.2m 752
TC Stair5W 0.9m
TC Stair5W 0.6m

Temperature (F)
Temperature (C)

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-101. Temperature versus time for the 5th floor stairwell thermocouple tree, Experiment 5G
gle) ted
g g 2 m r D (sin ting pl e
i n n 0 a
fail eni 0 . cle W C ve n com
niti
on u ble left op ening ening pening pen ouble c ing indow ent
: Ig o
: d BY: op op H: o H: o R: d l a
: p R: w er im
BR BR L S 1: S 5 : B B B BR L E xp
400 752
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
300 572

Temperature (F)
Temperature (C)

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-102. Temperature versus time for the stair thermocouples, Experiment 5G

547
9.4.10.2 Pressures
gle) ted
g g 2 m r D (sin ting pl e
i n n 0 a n
il
le fa ope ing
ni
ing
0 . le C
le c ng W ow v
e com
niti
on ing nt
: Ig : d oub Y: left : open : open : open : open : doub : placi : wind er im
e
BR BR LB S1 S5 BH BH BR BR LR E xp
80
DP Stair 7W
DP Stair 6W
DP Stair 5W
60 DP Stair 4W
DP Stair 3W
DP Stair 2W
DP Stair 1W
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-103. Differential pressure versus time for the stairwell bi-directional probes, Experiment 5G

9.4.10.3 Velocities
gle) ted
g 2m (sin ting le
fa il i n
en i n g
0 . 0
cle a r
W C D
ve n co mp
niti
on u ble left op ening ening pening pen ouble c ing indow ent
: Ig o
: d BY: o p p
o H: o H: o R: d l a
: p R: w er im
BR BR L S 1: S 5 : B B B BR L E xp
8 18
VEL Stair5 TopW
VEL Stair5 MidW
VEL Stair5 BotW
6 13

Velocity (mph)
Velocity (m/s)

4 9

2 4

0 0

-2 -4
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-104. Velocity versus time for the 5th floor stairwell bi-directional probes, Experiment 5G

548
 gle) ted
g g 2 m r D (sin ting pl e
i n n 0 a
fail en i 0 . cle W C ve n com
niti
on u ble left op ening ening pening pen ouble c ing indow ent
: Ig o
: d BY: o p p
o H: o H: o R: d l a
: p R: w er im
BR BR L S 1: S 5 : B B B BR L E xp
10 22
VEL Apt5G TopW
VEL Apt5G MidW
VEL Apt5G BotW
8 18

6 13

Velocity (mph)
Velocity (m/s)

4 9

2 4

0 0

-2 -4
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-105. Velocity versus time for the apartment 5G doorway bi-directional probes, Experiment 5G

9.4.11 Experiment 5G2 
9.4.11.1 Temperatures
Ex
LR L p
LR
:
LR
: m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR S B H p l a L B o B C D F B F B o F B F B C D Y: nt
: Ig S1: f S5: c 5: op : op cing Y: f a ving H: c rep N fl N sto ving N fl N sto rep f an m comp
niti a l o e e W n W l o s l ac o p W o p l ac ov
on n off s ed n i ng ng n i CD
s on C D ed ed ngwi p ed C D
wi ng p ed ed ed leted

400 752
TC BR5GW 2.1m
TC BR5GW 1.8m
TC BR5GW 1.5m
TC BR5GW 1.2m
TC BR5GW 0.9m
300 TC BR5GW 0.6m
572

Temperature (F)
Temperature (C)

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-106. Temperature versus time for the 5G2 bedroom thermocouple tree, Experiment 5G2

549
 Ex
LR L p
LR
:
LR
:m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR S5 B H p L
l ac B Y o v B C D FB F B o FB F B C D Y: nt
S S
: Ig 1: f 5: c : op : op ing : in H r N N vin N N r f an co
niti an los eni eni W C f ans g WC : clos epl ac flowi stopp g WC flowi stopp eplac mov mplet
on off ed ng ng D on D ed ed ng ed D ng ed ed ed ed

1000 1832
TC Corr5GW 2.1m
TC Corr5GW 1.8m
TC Corr5GW 1.5m
800 TC Corr5GW 1.2m 1472
TC Corr5GW 0.9m
TC Corr5GW 0.6m

Temperature (F)
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-107. Temperature versus time for the 5G2 hall thermocouple tree, Experiment 5G2
Ex
LR L p
LR
:
LR
: m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR p L
S 5 B H l ac B Y o v B C D FB F B o FB F B CD Y: f nt c
S S
: Ig 1: f 5: c : op : op ing in H Nf N v N N a o
niti an l e e
:
W f an g W : cl epl
r lo sto ing W flo sto repl n m mp
on off osed ning ning CD s on C D osed aced wing pped CD wing pped aced oved leted

600 1112
TC Corr5KW 2.1m
TC Corr5KW 1.8m
TC Corr5KW 1.5m
500 TC Corr5KW 1.2m 932
TC Corr5KW 0.9m
TC Corr5KW 0.6m

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-108. Temperature versus time for the 5K hall thermocouple tree, Experiment 5G2

550
 Ex
LR L p
LR
:
LR
:m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR S5 B H p L
l ac B Y o v B C D FB F B o FB F B C D Y: nt
S S
: Ig 1: f 5: c : op : op ing : in H r N N vin N N r f an co
niti an los eni eni W C f ans g WC : clos epl ac flowi stopp g WC flowi stopp eplac mov mplet
on off ed ng ng D on D ed ed ng ed D ng ed ed ed ed

1000 1832
TC CorrStair5W 2.1m
TC CorrStair5W 1.8m
TC CorrStair5W 1.5m
800 TC CorrStair5W 1.2m 1472
TC CorrStair5W 0.9m
TC CorrStair5W 0.6m

Temperature (F)
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-109. Temperature versus time for the 5th floor stair hall thermocouple tree, Experiment 5G2
Ex
LR L p
LR
:
LR
: m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR p L
S 5 B H l ac B Y o v B C D FB F B o FB F B CD Y: f nt c
S S
: Ig 1: f 5: c : op : op ing in H Nf N v N N a o
niti an l e e
:
W f an g W : cl epl
r lo sto ing W flo sto repl n m mp
on off osed ning ning CD s on C D osed aced wing pped CD wing pped aced oved leted

800 1472
TC Stair5W 2.1m
TC Stair5W 1.8m
TC Stair5W 1.5m
TC Stair5W 1.2m
TC Stair5W 0.9m
600 TC Stair5W 0.6m
1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-110. Temperature versus time for the 5th floor stairwell thermocouple tree, Experiment 5G2

551
 Ex
LR L p
LR
:
LR
:m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR S5 B H p L
l ac B Y o v B C D FB F B o FB F B C D Y: nt
S S
: Ig 1: f 5: c : op : op ing : in H r N N vin N N r f an co
niti an los eni eni W C f ans g WC : clos epl ac flowi stopp g WC flowi stopp eplac mov mplet
on off ed ng ng D on D ed ed ng ed D ng ed ed ed ed

800 1472
TC Bulkhead MidW
TC Stair 6W
TC Stair 4W
TC Stair 2W
TC Stair1 MidW
600 1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-111. Temperature versus time for the stair thermocouples, Experiment 5G2

9.4.11.2 Pressures
Ex
LR p
LR
:
LR
: : W LR LR: LR: LR LR: LR: W LB er im e
S B p l L m C : F F B m : F F B C Y: nt c
LR
: I g S 1: f S 5: c 5: o H: o acin BY: ovin BH D re BN N ov B N D fa
niti an l os pen pen g W f ans g W : clo pla flow stop ing W N flow stop repla n mo omple
on off ed ing ing CD on C D sed ced ing ped C D ing ped ced ved ted

80
DP Stair 7W
DP Stair 6W
DP Stair 5W
60 DP Stair 4W
DP Stair 3W
DP Stair 2W
40 DP Stair 1W
Pressure (Pa)

20

-20

-40
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-112. Differential pressure versus time for the stairwell bi-directional probes, Experiment 5G2

552
9.4.11.3 Velocities
Ex
LR L p
LR
:
LR
: m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR p L
S 5 B H l ac B Y o v B C D FB F B o F B F B C D Y: nt
S S
: Ig 1: f 5: c : op : op ing : in H r N N vin N N r f an co
niti an los eni eni W C f ans g WC : clos epl ac flowi stopp g WC flowi stopp eplac mov mplet
on off ed ng ng D on D ed ed ng ed D ng ed ed ed ed

10 22
VEL Stair5 TopW
VEL Stair5 MidW
VEL Stair5 BotW
8 18

6 13

Velocity (mph)
Velocity (m/s)

4 9

2 4

0 0

-2 -4
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-113. Velocity versus time for the 5th floor stairwell bi-directional probes, Experiment 5G2
Ex
LR L p
LR
:
LR
: m : W LR: LR: LR: m LR: LR: R: W LB er im e
LR S B H p l a L B o B C D F B F B o F B F B C D Y: nt
: Ig S1: f S5: c 5: op : op cing Y: f a ving H: c rep N fl N sto ving N fl N sto rep f an m comp
niti a l o e e W n W l o s l ac o p W o p l ac ov
on n off s ed n i ng ng n i CD
s on C D ed ed ngwi p ed C D
wi ng p ed ed ed leted

10 22
VEL Apt5G TopW
VEL Apt5G MidW
VEL Apt5G BotW
8 18

6 13

Velocity (mph)
Velocity (m/s)

4 9

2 4

0 0

-2 -4
0 200 400 600 800 1000 1200 1400
Time (s)
Figure 9.4-114. Velocity versus time for the apartment 5G doorway bi-directional probes, Experiment 5G2

553
9.4.12 Experiment 3E 
9.4.12.1 Temperatures
) )
le) ble le) ble
ub dou ub dou d ed
s g r
o o
(d ed ( ing g (d ed ( ate D ved d p let
n l i n a g
n pp fai l n pp cti v C o g e
n p p co m
it io ing g fai cle wi wi W m wi
g g
Ign pen enin enin enin uble uble flo sto dow flo sto ice a ing D re ed s on on flo sto ent
le o p p o o B N B N in B N N
B e v la c C s
lo f an n B N B N im
p
ltip Y: : o : o : o : d : d : F : F : w : F : F : d : p : W : c Y: : f a R: F R: F xper
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 L L E
1200 2192
TC BR3EE 2.1m
TC BR3EE 1.8m
TC BR3EE 1.5m
1000 TC BR3EE 1.2m 1832
TC BR3EE 0.9m
TC BR3EE 0.6m

Temperature (F)
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-115. Temperature versus time for the 3E bedroom thermocouple tree, Experiment 3E
) )
le) ble le) ble
ub dou ub dou d ed
s g r
o o
(d ed ( ing g (d ed ( ate D ved d p let
n l i n a g
n pp fai l n pp cti v C o g e
n p p co m
it io ing g fai cle wi wi W m wi
g g
Ign pen enin enin enin uble uble flo sto dow flo sto ice a ing D re ed s on on flo sto ent
le o p p o o B N B N in B N N
B e v la c C l os f an n B N B N im
p
ltip Y: : o : o : o : d : d : F : F : w : F : F : d : p : W : c Y: : f a R: F R: F xper
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 L L E
1000 1832
TC Corr3EE C
TC Corr3EE 2.1m
TC Corr3EE 1.8m
800 TC Corr3EE 1.5m 1472
TC Corr3EE 1.2m
TC Corr3EE 0.9m

Temperature (F)
TC Corr3EE 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-116. Temperature versus time for the 3E hall thermocouple tree, Experiment 3E

554
 ) )
le) ble le) ble
o ub dou o ub dou d ted
d ( d ( e d d mple
n s il i ng ear i ng ( ped iling ing ( ped tivat CD ove i n g pe
it io ing ng ng ing le fa le cl w p
flo sto w flo sto
fa w p ac W em on low to nt c
p o
gn n i i g r d
N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
p le I ope pen pen open doub doub B a r
lti Y: : o : o : : : : F : F : w : F : F : d : p : W : c Y: : f : F : F pe
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 LR LR E x
800 1472
TC Corr3AE C
TC Corr3AE 2.1m
TC Corr3AE 1.8m
TC Corr3AE 1.5m
TC Corr3AE 1.2m
600 TC Corr3AE 0.9m
1112

Temperature (F)
TC Corr3AE 0.6m
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-117. Temperature versus time for the 3A hall thermocouple tree, Experiment 3E
) le) ) le)
u ble oub u ble oub ed
do (d do (d ed d let
n s il i ng ear i ng ( ped iling ing ( ped tivat CD ove i n g ped omp
o
it i ing ng ng ing le fa le c l p f a p ac W em p
gn
w
flo sto w flo sto
w g r d on low to nt c
l
n i i
e I ope pen pen open doub doub B N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
p : F : F : w : F : F : d : p : W : c Y: : f a r
lti Y: : o : o : : : : F : F pe
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 LR LR E x
1000 1832
TC CorrStair3E C
TC CorrStair3E 2.1m
TC CorrStair3E 1.8m
800 TC CorrStair3E 1.5m 1472
TC CorrStair3E 1.2m
TC CorrStair3E 0.9m

Temperature (F)
TC CorrStair3E 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-118. Temperature versus time for the 3rd floor stair hall thermocouple tree, Experiment 3E

555
 ) )
le) ble le) ble
o ub dou o ub dou d ted
d ( d ( e d d mple
n s il i ng ear i ng ( ped iling ing ( ped tivat CD ove i n g pe
it io ing ng ng ing le fa le cl w p
flo sto w flo sto
fa w p ac W em on low to nt c
p o
gn n i i g r d
N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
p le I ope pen pen open doub doub B a r
lti Y: : o : o : : : : F : F : w : F : F : d : p : W : c Y: : f : F : F pe
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 LR LR E x
800 1472
TC Stair3E C
TC Stair3E 2.1m
TC Stair3E 1.8m
TC Stair3E 1.5m
TC Stair3E 1.2m
600 TC Stair3E 0.9m
1112

Temperature (F)
TC Stair3E 0.6m
Temperature (C)

400 752

200 392

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-119. Temperature versus time for the 3rd floor stairwell thermocouple tree, Experiment 3E
) le) ) le)
u ble oub u ble oub ed
do (d do (d ed d let
n s il i ng ear i ng ( ped iling ing ( ped tivat CD ove i n g ped omp
o
it i ing ng ng ing le fa le c l p f a p ac W em p
gn
w
flo sto w flo sto
w g r d on low to nt c
l
n i i
e I ope pen pen open doub doub B N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
p : F : F : w : F : F : d : p : W : c Y: : f a r
lti Y: : o : o : : : : F : F pe
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 LR LR E x
600 1112
TC Bulkhead MidE
TC Stair 2E
TC Stair1 MidE
500 932

Temperature (F)
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-120. Temperature versus time for the stair thermocouples, Experiment 3E

556
9.4.12.2 Pressures
) )
le) ble le) ble
o ub dou o ub dou d ted
d ( d ( e d d mple
n s il i ng ear i ng ( ped iling ing ( ped tivat CD ove i n g pe
it io ing g g g fa cl w p
flo sto w flo sto
fa w p ac W em
on low to nt c
p o
e Ign pen enin enin enin uble uble g r d
N BN indo BN BN evice lacin CD lose f ans n on BN f BN s im e
l o p p p o o B r
ltip Y: : o : o : o : d : d R: F R: F R: w R: F R: F R: d R: p R: W H: c BY: 1: fa R: F R: F xpe
Mu LB S1 S3 BH BR BR B B L B B L L L B L S L L E
80
DP Stair 7E
DP Stair 6E
DP Stair 4E
60 DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

40

20

-20
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-121. Differential pressure versus time for the stairwell bi-directional probes, Experiment 3E

9.4.12.3 Velocities
) )
le) ble le) ble
ub dou ub dou d ed
s g r
o o
(d ed ( ing g (d ed ( ate D ved d p let
n l i n a g
n pp fai l n pp cti v C o g e
n p p co m
it io ing g fai cle wi wi W m wi
g g
Ign pen enin enin enin uble uble flo sto dow flo sto ice a ing D re ed s on on flo sto ent
le o p p o o B N B N in B N N
B e v la c C s
lo f an n B N B N im
p
ltip Y: : o : o : o : d : d : F : F : w : F : F : d : p : W : c Y: : f a R: F R: F xper
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 L L E
4 9

3 7

2 4

Velocity (mph)
Velocity (m/s)

1 2

0 0

-1 -2

-2 -4
VEL Stair1 TopE
VEL Stair1 BotE
-3 -7
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-122. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 3E

557
 ) )
le) ble le) ble
o ub dou o ub dou d ted
d ( d ( e d d mple
n s il i ng ear i ng ( ped iling ing ( ped tivat CD ove i n g pe
it io ing ng ng ing le fa le cl w p
flo sto w flo sto
fa w p ac W em on low to nt c
p o
gn n i i g r d
N BN indo BN BN evice lacin CD lose fans n on BN f BN s im e
p le I ope pen pen open doub doub B a r
lti Y: : o : o : : : : F : F : w : F : F : d : p : W : c Y: : f : F : F pe
Mu LB S1 S3 BH BR BR B R B R LR B R B R LR LR LR B H LB S 1 LR LR E x
3 7

0 0

-3 -7

Velocity (mph)
Velocity (m/s)

VEL Bulkhead TopE

-6 VEL Bulkhead MidE -13
VEL Bulkhead BotE

-9 -20

-12 -27

-15 -34
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Time (s)
Figure 9.4-123. Velocity versus time for the bulkhead bi-directional probes, Experiment 3E

9.4.13 Experiment 3A 
9.4.13.1 Temperatures
BR BR:
: F FB Ex
L S 3 B BN N LR pe
Mu BY :o R B R fl s to : LR L R LR L R LR r im
ltip : rig pe : d : ow p w : w : : F : : F S e
le h S1 nin S3 BH ou dou ing ed ind p in F B B N F B B N 1: f B nt c
ign t op : op g/ c : op : op ble ble (do (do ow dow N fl sto N flo sto an a H: c omp
it io e e lo e e fa c u u f a c o p p d l l
ns ning ning sing ning ning iling lear ble) ble) iling lear wing ped wing ped ded osed eted

1500 2732
TC BR3AE 2.1m
TC BR3AE 1.8m
TC BR3AE 1.5m
1200 TC BR3AE 1.2m 2192
TC BR3AE 0.9m
TC BR3AE 0.6m

Temperature (F)
Temperature (C)

900 1652

600 1112

300 572

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-124. Temperature versus time for the 3A bedroom thermocouple tree, Experiment 3A

558
 BR BR:
: F FB Ex
L S3 B BN N LR pe
Mu B Y :o R B R fl s to : LR L R LR L R LR r im
ltip : rig pe :d : ow p w :w : :F : :F S e
le S 1 ni n S 3 B H ou d o i n p e in d in F B B F B B 1 : B nt
h
ign t op : op g/ c : op : op ble uble (do (do ow f dow N flo sto N flo sto an a H: c comp
g d N N f
it io eni eni losi eni eni faili cle ub ub aili cle wi pp wi pp dd los let
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

1000 1832
TC Corr3AE C
TC Corr3AE 2.1m
TC Corr3AE 1.8m
800 TC Corr3AE 1.5m 1472
TC Corr3AE 1.2m
TC Corr3AE 0.9m

Temperature (F)
TC Corr3AE 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-125. Temperature versus time for the 3A hall thermocouple tree, Experiment 3A
BR BR:
: F FB Ex
S3 B N pe
Mu LBY :o
p
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
:
ltip rig en p w
do : do win pe in win : F FB : F FB S e
le h t
S 1 :
S
ing 3: H: B ub u g d d o d BN N BN N : f a BH nt co
1
ign op op / c o p o p le ble (do (do w ow f s t f s t n : m
it io eni eni losi eni eni faili cle ub ub faili cle lowi opp lowi opp add clos plet
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

600 1112
TC Corr3EE C
TC Corr3EE 2.1m
TC Corr3EE 1.8m
500 TC Corr3EE 1.5m 932
TC Corr3EE 1.2m
TC Corr3EE 0.9m

Temperature (F)
TC Corr3EE 0.6m
Temperature (C)

400 752

300 572

200 392

100 212

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-126. Temperature versus time for the 3E hall thermocouple tree, Experiment 3A

559
 BR BR:
: F FB Ex
L S3 B BN N LR pe
Mu B Y :o R B R fl s to : LR L R LR L R LR r im
ltip : rig pe :d : ow p w :w : :F : :F S e
le S 1 ni n S 3 B H ou d o i n p e in d in F B B F B B 1 : B nt
h
ign t op : op g/ c : op : op ble uble (do (do ow f dow N flo sto N flo sto an a H: c comp
g d N N f
it io eni eni losi eni eni faili cle ub ub aili cle wi pp wi pp dd los let
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

1200 2192
TC CorrStair3E C
TC CorrStair3E 2.1m
TC CorrStair3E 1.8m
1000 TC CorrStair3E 1.5m 1832
TC CorrStair3E 1.2m
TC CorrStair3E 0.9m

Temperature (F)
TC CorrStair3E 0.6m
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-127. Temperature versus time for the 3rd floor stair hall thermocouple tree, Experiment 3A
BR BR:
: F FB Ex
S3 B N pe
Mu LBY :o
p
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
:
ltip rig en p w
do : do win pe in win : F FB : F FB S e
le h t
S 1 :
S
ing 3: H: B ub u g d d o d BN N BN N : f a BH nt co
1
ign op op / c o p o p le ble (do (do w ow f s t f s t n : m
it io eni eni losi eni eni faili cle ub ub faili cle lowi opp lowi opp add clos plet
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

1000 1832
TC Stair3E C
TC Stair3E 2.1m
TC Stair3E 1.8m
800 TC Stair3E 1.5m 1472
TC Stair3E 1.2m
TC Stair3E 0.9m

Temperature (F)
TC Stair3E 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-128. Temperature versus time for the 3rd floor stairwell thermocouple tree, Experiment 3A

560
 BR BR:
: F FB Ex
L S3 B BN N LR pe
Mu B Y :o R B R fl s to : LR L R LR L R LR r im
ltip : rig pe :d : ow p w :w : :F : :F S e
le S 1 ni n S 3 B H ou d o i n p e in d in F B B F B B 1 : B nt
h
ign t op : op g/ c : op : op ble uble (do (do ow f dow N flo sto N flo sto an a H: c comp
g d N N f
it io eni eni losi eni eni faili cle ub ub aili cle wi pp wi pp dd los let
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

500 932
TC Bulkhead MidE
TC Stair 2E
TC Stair1 MidE
400 752

Temperature (F)
Temperature (C)

300 572

200 392

100 212

0 32
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-129. Temperature versus time for the stair thermocouples, Experiment 3A

9.4.13.2 Pressures
BR BR:
: F FB Ex
L S 3 BR BN N LR pe
Mu BY :o B f
R: low op s t : L R L L R L L R r im
ltip : rig pe :d win : w R: F : FB R: F : FB S1 e
le h t
S 1 :
n i ng
S 3 :
B H:
o u b
d o u i ng
p e d d o in d B N N B N N : f a BH nt co
ign op op / c o p o p le b le (do (do w f o w fl s t fl s t n : m
it io eni eni losi eni eni fail
ing cle ub ub ailin cle win ppe win ppe dd clos plete
o o o o a
ns ng ng ng ng ng ar le) le) g ar g d g d ed ed d

40
DP Stair 7E
DP Stair 6E
DP Stair 4E
30 DP Stair 3E
DP Stair 2E
DP Stair 1E
Pressure (Pa)

20

10

-10
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-130. Differential pressure versus time for the stairwell bi-directional probes, Experiment 3A

561
9.4.13.3 Velocities
BR BR:
: F FB Ex
L S 3 B BN N LR pe
Mu BY :o R B R fl s to : LR L R LR L R LR r im
ltip : rig p : o : : :
S1 enin S3 BH dou : dou wing pped wind wind : FB FBN : FB FBN S1: f en
t
le h
ign t op : op g/ c : op : op ble ble (do (do ow ow N fl s Nf s an BH: com
l f f o t
it io eni eni osi eni eni aili cle ub ub aili cle wi pp wi pp dd clos plet
o l o to a
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

5 11

0 0

Velocity (mph)
Velocity (m/s)

-5 -11

-10 -22

-15 -34
VEL Bulkhead TopE
VEL Bulkhead MidE
VEL Bulkhead BotE
-20 -45
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-131. Velocity versus time for the bulkhead bi-directional probes, Experiment 3A
BR BR:
: F FB Ex
S3 B N pe
Mu LBY :o
p
BR
: BR N flo sto LR: LR: LR LR: LR LR: r im
:
ltip rig en do : do win ppe win win : F FB : F FB S1: e
le h S 1 i n S 3 B H u u g d d d B N B N f B nt co
ign t op : op g/ c : op : op ble b ( ( ow o N s N s an H: m
it io eni eni losi eni eni faili le cle doub doub faili w cle flowi topp flowi topp add clos plet
ns ng ng ng ng ng ng ar le) le) ng ar ng ed ng ed ed ed ed

6 13

4 9

2 4

Velocity (mph)
Velocity (m/s)

0 0

-2 -4

-4 -9
VEL Stair1 TopE
VEL Stair1 BotE
-6 -13
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (s)
Figure 9.4-132. Velocity versus time for the 1st floor stairwell bi-directional probes, Experiment 3A

562
9.4.14 Experiment 3G/K 
9.4.14.1 Temperatures
)
e) e) le
n gl u bl ou b
i o d
g (s w D (d (
n s i ng
in g r
a i l in iling CD ar CD flo C ed ng ped D
C i ng pe
d
itio pen g fail cle g g g n fa fa W cle W air g W ov owi top W w o p
n o g n l g o st
Ig t o nin le le nin nin ed nin s on ed ow ble in ble ing ing ci rem f s in fl
p le lef pe ing ing pe pe los pe fan an los ind ou lac ou lac duc pla ow BN BN ov BN B N
ti : s s o c o :
o : : :o : : : Y :f : : c w d p d p
: : : : tro 2 r fl : : : F F m F :F
ul Y : : :
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R

300 572
TC BR3KW 1.8m
TC BR3KW 1.5m
TC BR3KW 1.2m
250 TC BR3KW 0.9m 482
TC BR3KW 0.6m

Temperature (F)
Temperature (C)

200 392

150 302

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-133. Temperature versus time for the 3G_K bedroom thermocouple tree, Experiment 3G_K
) )
le e) le
ng u bl o ub
i o d
g (s w D (d (
n s i ng
in g r
a i l in iling CD ar CD flo C ed ng ped D
C ng e
d
l e r W i W owi topp
n itio pen g fai cle g g g o n
w
fa fa W cl W ai g ov ow top
g g n m l g
Ig t o nin le le nin nin ed nin s on ed o ble in ble in ing ci re f s in fl s
le lef e ing ing e pe los pe fan n los ind ou lac ou lac uc pla w BN BN ov B N B N
tl ip Y: : op : s : s : op : o : c : o Y: : fa : c : w : d : p : d : p rod 2: flo : F : F m
: : F : F
u t r
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R
800 1472
TC Corr3KW C
TC Corr3KW 2.1m
TC Corr3KW 1.8m
TC Corr3KW 1.5m
TC Corr3KW 0.9m
600 TC Corr3KW 0.6m
1112

Temperature (F)
Temperature (C)

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-134. Temperature versus time for the 3G_K hall thermocouple tree, Experiment 3G_K

563
 ) ) )
le le uble
ng b
o u do
si
s i ng g n g g D ( r D l ow C D d (d d ( D d
n in a r l
i li i n C a C f e ng e C i ng pe
itio pen g fail cle g g g n fa fai W cle W air g W ov owi topp W w p
n o g n l g o sto
Ig t o nin le le nin nin ed nin s on ed ow ble in ble ing ing ci rem f s in fl
p le lef pe ing ing pe pe los pe fan an los ind ou lac ou lac duc pla ow BN BN ov BN B N
ti : o : s : s : o : o : c : o Y: : f : c : w : d : p : d : p ro 2: fl : F : F :m :F :F
ul Y : t r
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R

1200 2192
TC Corr3GW C
TC Corr3GW 2.1m
TC Corr3GW 1.8m
1000 TC Corr3GW 1.5m
1832
TC Corr3GW 1.2m
TC Corr3GW 0.9m

Temperature (F)
TC Corr3GW 0.6m
Temperature (C)

800 1472

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-135. Temperature versus time for the 7K hall thermocouple tree, Experiment 3G_K
)
e) e) le
n gl u bl ou b
i o d
g (s w D (d (
n s i ng
in g r
a i l in iling CD ar CD flo C ed ng ped D
C i ng pe
d
itio pen g fail cle g g g n fa fa W cle W air g W ov owi top W w o p
n o g n l g o st
Ig t o nin le le nin nin ed nin s on ed ow ble in ble ing ing ci rem f s in fl
p le lef pe ing ing pe pe los pe fan an los ind ou lac ou lac duc pla ow BN BN ov BN B N
ti : s s o c o :
o : : :o : : : Y :f : : c w d p d p
: : : : tro 2 r fl : : : F F m F :F
ul Y : : :
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R
1000 1832
TC CorrStair3W C
TC CorrStair3W 2.1m
TC CorrStair3W 1.8m
800 TC CorrStair3W 1.5m 1472
TC CorrStair3W 1.2m
TC CorrStair3W 0.9m

Temperature (F)
TC CorrStair3W 0.6m
Temperature (C)

600 1112

400 752

200 392

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-136. Temperature versus time for the 7th floor stair hall thermocouple tree, Experiment 3G_K

564
 ) ) )
le le uble
ng b
o u do
si
s i ng g n g g D ( r D l ow C D d (d d ( D d
n in a r l
i li i n C a C f e ng e C i ng pe
itio pen g fail cle g g g n fa fai W cle W air g W ov owi topp W w p
n o g n l g o sto
Ig t o nin le le nin nin ed nin s on ed ow ble in ble ing ing ci rem f s in fl
p le lef pe ing ing pe pe los pe fan an los ind ou lac ou lac duc pla ow BN BN ov BN B N
ti : o : s : s : o : o : c : o Y: : f : c : w : d : p : d : p ro 2: fl : F : F :m :F :F
ul Y : t r
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R

300 572
TC Stair3W 1.5m
TC Stair3W 1.2m
TC Stair3W 0.9m
250 TC Stair3W 0.6m 482

Temperature (F)
Temperature (C)

200 392

150 302

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-137. Temperature versus time for the 7th floor stairwell thermocouple tree, Experiment 3G_K
)
e) e) le
n gl u bl ou b
i o d
g (s w D (d (
n s i ng
in g r
a i l in iling CD ar CD flo C ed ng ped D
C i ng pe
d
itio pen g fail cle g g g n fa fa W cle W air g W ov owi top W w o p
n o g n l g o st
Ig t o nin le le nin nin ed nin s on ed ow ble in ble ing ing ci rem f s in fl
p le lef pe ing ing pe pe los pe fan an los ind ou lac ou lac duc pla ow BN BN ov BN B N
ti : s s o c o :
o : : :o : : : Y :f : : c w d p d p
: : : : tro 2 r fl : : : F F m F :F
ul Y : : :
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R
250 482
TC Bulkhead MidW
TC Stair 6W
TC Stair 2W
200 TC Stair1 MidW 392

Temperature (F)
Temperature (C)

150 302

100 212

50 122

0 32
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-138. Temperature versus time for the stair thermocouples, Experiment 3G_K

565
9.4.14.2 Pressures
) )
le e) le
ng u bl ou b
i o d
g g (s w D (d d (
ns ning in g r
a i lin ilin CD ear CD r flo WC ved ing pe D
C ng e
d
o l
iti pe g fai cle g g g o n a l
f fa W c W a g o ow top i W owi topp
n n n w e g e g g i n m l g l
Ig t o nin le le nin ni ed ni s on ed o bl in bl in in c re f s in f s
le lef e ing ing e pe los pe fan n los ind ou lac ou lac uc pla w BN BN ov BN B N
tl ip Y: : op : s : s : op : o : c : o Y: : fa : c : w : d : p : d : p rod 2: flo : F : F : :F :F
m
u t r
M LB S1 BR BR S3 BH BH BH LB S1 BH LR BR BR BR LR In BR Ai BR BR L R LR L R

40
DP Stair 7W
DP Stair 6W
DP Stair 5W
DP Stair 4W
DP Stair 3W
20 DP Stair 2W
DP Stair 1W
Pressure (Pa)

-20

-40
0 300 600 900 1200 1500 1800 2100 2400 2700 3000
Time (s)
Figure 9.4-139. Differential pressure versus time for the stairwell bi-directional probes, Experiment 3G_K

566
9.5 Appendix E – Post Fire Images 

9.5.1 Apartment 7G 

Figure 9.5-1. 7G apartment entrance Figure 9.5-2. To the corridor from the living room

Figure 9.5-3. Down the corridor toward apartment 7K Figure 9.5-4. Through the bedrooms from the living room

Figure 9.5-5. 7G apartment entrance Figure 9.5-6. Down the stairwell from the fire floor

567
9.5.2 Apartment 7E 

Figure 9.5-7. 7E apartment entrance Figure 9.5-8. Living room from entrance

Figure 9.5-9. Living room window Figure 9.5-10. Bedroom double window

Figure 9.5-11. Bedroom single window Figure 9.5-12. Bedroom single window

568
9.5.3 Apartment 7A 

Figure 9.5-13. Living room window Figure 9.5-14. Bedroom double window

Figure 9.5-15. Bedroom single window Figure 9.5-16. Bedroom door from ignition bedroom

Figure 9.5-17. Living room from bedroom-side corner Figure 9.5-18. 7A apartment entrance from living room

569
9.5.4 Apartment 7K 

Figure 9.5-19. Apartment 7K entrance Figure 9.5-20. Living room from entrance

Figure 9.5-21. Living room to bedroom hallway Figure 9.5-22. Bedroom 2 window

Figure 9.5-23. Bedroom double window Figure 9.5-24. Bedroom single window

570
9.5.5 Apartment 5E 

Figure 9.5-25. Living room from entrance Figure 9.5-26. Living room window

Figure 9.5-27. Bedroom windows from adjacent bedroom Figure 9.5-28. Bedroom double from adjacent bedroom

Figure 9.5-29. 5E apartment entrance from the inside Figure 9.5-30. 5E apartment entrance from the corridor

571
9.5.6 Apartment 5K 

Figure 9.5-31. 5K apartment entrance Figure 9.5-32. Living room from entrance

Figure 9.5-33. Living room window and kitchen Figure 9.5-34. Bedroom 2 from hallway

Figure 9.5-35. Bedroom double window Figure 9.5-36. Bedroom single window

572
9.5.7 Apartment 5A 

Figure 9.5-37. 5A apartment entrance Figure 9.5-38. Living room window

Figure 9.5-39. Living room toward bedroom-side wall Figure 9.5-40. Living room toward wall opposite window

Figure 9.5-41. Bedroom single window Figure 9.5-42. Bedroom double window

573
9.5.8 Apartment 5G 

Figure 9.5-43. 5G apartment entrance Figure 9.5-44. Toward entrance from living room

Figure 9.5-45. Bedroom 2 from hallway Figure 9.5-46. bedroom 2 toward hallway

Figure 9.5-47. Bedroom double window Figure 9.5-48. Bedroom single window

574
9.5.9 Apartment 3A 

Figure 9.5-49. 3A apartment entrance Figure 9.5-50. Living room from entrance

Figure 9.5-51. From apartment door into corridor Figure 9.5-52. Bedroom double window

Figure 9.5-53. Bedroom single window Figure 9.5-54. Toward apartment entrance from living room

575
9.5.10 Apartment 3E 

Figure 9.5-55. Living room from entrance Figure 9.5-56. Living room window

Figure 9.5-57. 3E apartment entrance from living room Figure 9.5-58. Living room to bedroom hallway

Figure 9.5-59. Bedroom double window Figure 9.5-60. Bedroom single window

576
9.5.11 Apartments 3G and 3K 
9.5.11.1 3G

Figure 9.5-61. Living room from entrance Figure 9.5-62. Bedroom double window

Figure 9.5-63. Bedroom single window Figure 9.5-64. Bedroom doorway from far corner

Figure 9.5-65. Living room through bedroom 2 Figure 9.5-66. Living room window

577
9.5.11.2 Public Corridor

Figure 9.5-67. Corridor target

578
9.5.11.3 3K

Figure 9.5-68. Into corridor from 3K apartment entrance Figure 9.5-69. Living room window

Figure 9.5-70. Bedroom 2 window Figure 9.5-71. Bedroom double window

Figure 9.5-72. Bedroom single window Figure 9.5-73. Down the corridor from 3K apartment entrance

579