The smoke clears
It is conceived as a world-renowned cultural landmark: Covering two city blocks and six stories high, the Regional Performing Arts Center in Philadelphia is designed with a distinctive architectural profile, the centerpiece of which is a towering, 150-ft.-high, barrel-shaped glass atrium.
Designed by Rafael Viñoly Architects P.C., New York City, the facility's ancillary uses-in addition to its role as a world-class performing arts center-include meeting rooms, office space and dining areas. The atrium itself encloses a 2,500-seat concert hall and a 750-seat theater. Considered a mixed-use facility and open to the public, the structure is classified as a high-rise occupancy based on the height of its highest occupied level, a roof garden and terrace.
Given its intended status as a performing arts center of the first rank, taking its place architecturally and culturally alongside the finest and most prestigious in the world, aesthetic concerns were paramount. Yet the project's stakeholders were confronted with thorny and complex issues surrounding fire safety generally, and smoke control specifically.
As defined in a fire-protection engineering analysis published by consulting engineer Hughes Associates Inc., Baltimore, "stakeholders" are those constituencies with vested interests in fire-safety issues. For the Regional Performing Arts Center, as with most projects of this scope, they included the owner, the architect, the authority having jurisdiction (AHJ), the construction manager, the insurance company, the fire-protection engineers as well as other engineers responsible for the design of systems. The key, according to the analysis, is first to identify relevant stakeholders and involve them closely throughout each stage of a project.
A factor of equal importance is to leverage the capabilities of highly sophisticated and enhanced computer-modeling techniques to analyze and test smoke movement and control. Smoke-movement modeling can result in major dollar savings in the design of active and passive fire-protection systems, and can help protect the integrity of architectural designs.
While computerized smoke-control modeling techniques vary by building design, local conditions and requirements, the practice is becoming more prevalent and extremely reliable. And with more structures supporting mixed-use spaces that incorporate hotel and convention facilities, theaters, offices, restaurants and retail space, the need to design effective smoke-control systems has become more critical. Novel computer tools can be instrumental in achieving those goals.
Predicting smoke movement
In this case, Hughes Associates employed a program that has been used to design smoke-control systems for structures as large and complex as the World Trade Center in New York City. Originally developed by the Gaithersburg, Md.-based National Institute of Standards and Technology (NIST) to study "dirty environments" within buildings, CONTAM is a modeling method now used to examine a wide range of possible variables and their effects on particulate movement. These variables include:
Outside air temperature.
Movements induced by heating, ventilation and air-conditioning (HVAC) systems.
In those structures where-for a variety of reasons-CONTAM may not be appropriate, a complex modeling approach developed in the aerospace industry known as large-eddy-simulation (LES) computational fluid dynamics (CFD) may be employed. By dividing a space into thousands or even millions of small cells and calculating conditions within each cell and across their boundaries, LES/CFD can work well for the analysis of uniquely shaped spaces, atria or similar large-volume interior spaces.
Other, less recent developments in smoke-movement modeling techniques are provided by computer models such as NIST's CFAST-which stands for fire and smoke-transport modeling in the C programming language-and analytical techniques such as those presented in NFPA 92B, a standard published by the National Fire Protection Association, Quincy, Mass. Last, the widely used text published by the American Society of Heating, Refrigeration and Air-conditioning Engineers (ASHRAE), Atlanta, "Design of Smoke Management Systems," is an important resource.
Each, however, has its limitations. CFAST, for example, measures smoke movement from room to room, modeling the resulting smoke layer in each room; the model presents difficulties in taking into account other factors such as stack or wind effects. On the other hand, NFPA 92B and the ASHRAE text specify methods for calculating individual pieces of the puzzle. The methods can look at smoke development or other individual variables, but building teams attempting to tie them all together to evaluate wind and stack effects simultaneously may find the methods laborious and time-consuming.
For the process of estimating smoke layer location and composition within the atrium of the Regional Performing Arts Center, the team took an effective if challenging approach. Engineers used modeling by hand calculation in combination with computer models such as CFAST and CFD.
Practical and aesthetic concerns
For the design and construction of this major project, the applicable code is the 1997 edition of the City of Philadelphia Building Code, which is based on the Building Officials and Code Administrators' National Building Code (BOCA), 1996 Edition, with local amendments.
Among the provisions of PHIL is a requirement for smoke control for all atria that connect more than two levels, which clearly means that some form of smoke control-active, passive or alternative-is required for the Regional Performing Arts Center. The requirement mandates that smoke must be maintained to a minimum of six feet above the highest level of exit access for a minimum of 20 minutes. While the code allows passive (nonmechanical) smoke control, the design analysis must demonstrate that the intent of the requirement is met.
Distilled to its most basic elements, the challenge lay in how to meet the intent of the time and smoke-height requirement for the atrium without including exhaust fans or ductwork in the glass. Providing fans and ductwork in the glass dome was neither practical nor in keeping with the aesthetic goals of the design.
Using the aforementioned modeling techniques, however, an analysis of smoke movement and composition within the atrium led to a demonstrable conclusion. While the volume of smoke produced from a given fire within a tall, open space is larger than that for a smaller, enclosed space, the relative amount of particulate and energy produced, overall, is the same. A larger, taller, open space has the effect of reducing the concentration of particulate within the air space. In the atrium of the Regional Performing Arts Center, modeling showed that the hazard from exposure to smoke by the building's occupants, at the highest level of exit access, was limited.
There were myriad other factors and complexities to take into account and analyze, of course, but the central point is that each of the stakeholder constituencies came to understand and readily accept both the analysis by computer modeling and how it led to a different approach to smoke management. Were it not for the capabilities of the modeling, the literal meaning of the code requirement-designed to protect building occupants from smoke hazard while exiting the building in an emergency-could have forced expensive changes in the design or significantly altered the aesthetic character of its design.
In the past, predicting how effectively a smoke-control system would work-while incorporating several variables, such as wind and stack effect-was, in practice, limited to testing systems after the building was constructed. This approach made it costly to make changes after the fact. Now, through early and ongoing involvement by all the stakeholders throughout the design process, and with the sophistication of computerized smoke-control modeling techniques, at least two distinct benefits emerge:
Design changes, if needed, are far less costly and intrusive than construction changes.
Where design changes are less feasible for aesthetic reasons, solutions present themselves that either meet the intent of code requirements or provide documentation and rationales for using the alternative designs.
The latter was certainly the case for the building team developing the Regional Performing Arts Center. The resulting decisions and design development provide testimony to the effectiveness of the use of computer smoke and fire modeling. When it is in use, the fire-safety performance of the structure-if ever tested-will further confirm the approach.
Eric Rosenbaum is director of A & E Services for Hughes Associates, a Baltimore-based fire science and engineering firm. He was chairman of the Society of Fire Protection Engineerings' committee that drafted the "Guide to Performance-Based Fire-Protection Analysis and Design of Buildings," published this year.