A rational approach for fire-resistance evaluation of double-tee, prestressed concrete slabs in parking structures Puneet Kumar and Venkatesh K. R. Kodur ouble-tee, prestressed concrete slabs offer numerous advantages over traditional slab systems in terms of Dhigher load-carrying capacity, better use of space, cost-effectiveness, and optimized production. Consequently, the use of these slabs as a structural solution has gained pop- ularity in recent decades. Double-tee slabs are often used in parking structures, where their longer spans reduce the total number of required columns, allowing better use of space. The two main functions of double-tee slabs in parking struc- ■ This paper proposes a rational approach to evaluate tures include transferring applied surface loads to framing the fire resistance of double-tee slabs in parking members and providing fire compartmentation. In the event structures using a finite element analysis (FEA) of a fire, double-tee slabs are required to contain fire spread model to analyze realistic vehicle fire scenarios and from one floor to another while sustaining applied loading, loading conditions. without collapse, for a certain fire-exposure duration; this is defined as the fire resistance of a double-tee slab. Cur- ■ Appropriate vehicle fire scenarios were developed rently, fire-resistance requirements for a double-tee slab are based on a thorough literature review, and the FEA assessed using a prescriptive approach based only on stan- model was validated by comparing results from the dard fire exposure conditions, as defined by ASTM E119.1 model to a fire test on a full-scale double-tee slab. However, fire exposure in a parking structure often results from burning vehicles and can be significantly different from ■ Case studies analyzed with the proposed rational standard fire exposure. Vehicle fires are typically charac- approach indicate that fire-resistance predictions terized by a rapid temperature rise and quick decay over a for double-tee slabs under current prescriptive short duration, whereas standard fire exposure encompasses approaches for evaluating fire resistance are overly a longer burning duration and no decay phase. Figure 1 conservative. compares the evolution of fire temperatures for two typical vehicle fires in parking structures (parking fire 1 and 2) with PCI Journal (ISSN 0887-9672) V. 65, No. 2, March–April 2020. 1 PCI Journal is published bimonthly by the Precast/Prestressed Concrete Institute, 200 W. Adams St., Suite 2100, Chicago, IL 60606. standard fire exposure (ASTM E119 fire). Figure 1 shows Copyright © 2020, Precast/Prestressed Concrete Institute. The Precast/Prestressed Concrete Institute is not responsible for statements made that temperatures in parking structure fires quickly attain by authors of papers in PCI Journal. Original manuscripts and discussion on published papers are accepted on review in accordance with the higher peak temperatures and subside in a shorter amount Precast/Prestressed Concrete Institute’s peer-review process. No payment is offered. 20 PCI Journal | March–April 2020 1200 1000 800 ASTME 119 Fire Parking Fire 1 600 Parking Fire 2 400 Temperature, ˚C 200 0 0 50 100 150 200 Time, min Figure 1. Evolution of fire temperatures in vehicle and standard fire exposures. Note: °C = (°F – 32)/1.8. of time than the standard fire, thus producing less thermal Currently, the fire resistance of double-tee slabs is evaluat- impact on structural members. Therefore, the fire resistance of ed mainly through tabulated fire ratings, which are derived double-tee slabs, based on standard fire exposure, may not be from standard fire tests and are a function of equivalent slab a realistic indication of fire performance. thickness and clear cover thickness to prestressing strands.9,10 The minimum concrete cover keeps the temperature in the Most of the previous fire-resistance studies on double-tee prestressing strands below a critical temperature to satisfy slabs are based on standard fire exposure, and there is limited the load-carrying functionality of the slab. The minimum slab guidance for evaluating their fire resistance under vehicle fire thickness limits the temperature rise on the unexposed slab exposure. The earlier studies on finite-element-based models surface to satisfy compartment functionality. However, these for prestressed concrete double-tee slabs were by Franssen two provisions for evaluating fire resistance are overly con- and Bruls.2 However, this model did not account for shear servative because they do not account for all critical factors stresses, and therefore, shear capacity of the slab was deter- governing the fire response of double-tee slabs. As an alterna- mined using simplified equations from prescriptive codes. tive to tabulated fire ratings, design specifications also provide Kodur and Hatinger3 developed a two-dimensional numerical simplified fire design equations to evaluate the fire resistance model that could simulate the fire response of double-tee of slabs.9,10 This procedure evaluates moment capacity deg- slabs under standard and realistic fire exposure. A series of radation based on sectional temperatures under standard fire parametric studies was carried out to characterize parame- exposure to check for failure of the member at any given time. ters governing the fire resistance of double-tee slabs, and it However, no guidance is provided on evaluating sectional was concluded that fire scenario and failure criteria have a temperatures under realistic vehicle fire scenarios (such as significant influence on the fire resistance of double-tee slabs. those in parking structures), which limits the applicability of However, there is still a lack of validated advanced three-di- the design equations to standard fire exposure. mensional numerical models for tracing the thermo-mechani- cal behavior of double tees under realistic fire exposure from To overcome these limitations, a rational approach was de- the start of the fire to burnout conditions. On the other hand, veloped for evaluating the fire resistance of double-tee slabs several advanced numerical models do exist in literature for under realistic fire and loading conditions that can occur in a other structural members, such as beams, columns, slabs, and parking structure. This approach considers critical factors for walls.4–8 This lack of advanced numerical models is hindering evaluating the fire resistance of double-tee slabs, including the development of design guidelines for the fire-resistance varying fire characteristics, member geometry, loading and evaluation of double-tee slabs under vehicle fire exposure. support conditions, temperature-dependent material properties, PCI Journal | March–April 2020 21 geometric and material nonlinearity, and realistic failure limit December 31, 2017, and there have been a few other fire inci- states. This rational approach was developed using a three-di- dents involving the burning of three to seven cars.12,15 Howev- mensional finite element analysis (FEA) model in structural er, the available statistics indicate that such incidents are rare analysis software, where the response of the slab is traced from in parking structures and correspond to a very small probabil- the initial burning stage through the decay stage of the fire ity of occurrence. Therefore, vehicle fire scenarios involving (burnout conditions). The applicability of this approach to dou- one to three vehicles make up the majority of possible vehicle ble-tee slabs in parking structures is illustrated through case fire scenarios in parking structures. studies that incorporate slab dimensions, loading conditions, and fire scenarios that are typical in parking structures. Evolution of fire temperatures Characteristics of vehicle fires Temperature evolution during a vehicle fire can vary signifi- cantly and has been studied both experimentally and numeri- Vehicles are made of highly combustible materials and cally. One such experimental study was conducted by Mangs contain significant amounts of flammable fuel at any point in and Keski-Rahkonen.16 Three full-scale experiments were con- time and, therefore, are prone to fire incidents due to acciden- ducted on individual passenger cars equipped with standard tal (such as electrical and mechanical malfunctions, colli- service accessories and 0.03 m3 (30 L [7.9 gal.]) of gasoline in sions, and the like) or intentional (vandalism) causes. From each fuel tank. Fire was ignited under the engine compartment 2006 to 2010, fire departments in the United States alone or in the passenger cabin using a heptane tray. In the first test, responded to an average of 152,300 vehicle fires per year, the ventilation conditions of the car were varied by keeping which caused an average of 209 civilian casualties and direct the left door 100 mm (3.9 in.) ajar with the window complete- property loss of $536 million per year.11 About 69% of these ly open and the right door window 50 mm (2.0 in.) open. In fires were caused by electrical and mechanical malfunctions the second and third tests, all doors were kept closed along in vehicles, which indicates a high probability of accidental with one completely open window and three windows open vehicle fires in parking structures. Hence, it is imperative 50 mm (2.0 in.). The vehicle fire behavior was defined by to design parking structures to withstand the effects of such measuring the rate of heat release, mass loss, smoke produc- probable vehicle fires. One of the key input parameters tion rate, heat flux, and temperatures above and inside the car. affecting fire design is determining an appropriate fire sce- In all three tests, a peak rate of heat release of 2 MW (6.8 × nario that can occur in a parking structure. Identifying a fire 106 BTU/hr) and a peak temperature of 1000°C (1832°F) was scenario includes defining the number of vehicles involved measured, and a rapid decrease in temperature and rate of heat in the fire, the resulting fire temperatures, and the location of release was observed after 40 minutes of burning. the fire relative to structural members.