Analysis, Design & Performance
Full Session with Abstracts
339072-1 - Experimental study on long-span composite floor beams subject to fire
Friday, April 20
9:30 AM - 10:30 AM
The National Fire Research Laboratory (NFRL) at the National Institute of Standards and Technology (NIST) has the unique capability of testing large-scale structures under realistic fire conditions. A series of long-span composite floor beams subjected to combined gravity and fire loads is currently being tested. Composite floor system was selected because of its widespread use in construction and design of steel-framed office buildings. Although there have been some experimental studies on composite beams, most of them were conducted using a standard furnace fire, and test beam spans were limited to 10 m or shorter due to the size limitations of the furnaces. Analyzing the fire behavior of long-span composite beams is challenging since there are many influencing factors, including gravity connections, concrete slabs, and shear connectors, that often leads to complex fire-structure interaction modeling of the overall system. In standard fire tests, however, measuring the connection behavior was often ignored or very limited to evaluate its effect on the overall system response to fire. The focus of this experimental study is to measure the realistic fire performance of composite floor beams and systems, including connections, under realistic fire loads.
A total of five 12.8-m span composite beams were designed and constructed in accordance with U.S codes and standards. Each specimen was constructed with a 1.83-m wide lightweight concrete slab casted over a 7.6-cm deep composite, all fluted, galvanized steel decking, and supported on a W18x35 floor beam using 19 mm shear studs at 300 mm. Test variables include: (i) types of shear connections (welded/bolted double-angles versus single-plate) and (ii) slab end condition at the centerlines of support columns (simple versus continuous slabs). A uniform floor load is applied using six equally-spaced point loads along the centerline of the floor beam. The first specimen is tested at ambient temperature to measure its flexural moment capacity. The other four specimens are loaded to 50 % of their bending moment capacity measured at ambient temperature, followed by a compartment fire of 4 MW using natural gas. For compartmentation, gypsum panels with a total open area of 3 m2 are constructed to confine the fire below the composite beam specimen. The thermal load is controlled using three purpose-built, 1 m x 1.5 m natural gas burners distributed on the floor along the centerline of the beam. The total heat release rate of 4 MW is held constant for 3.5 hours, followed by a 30-minute cool-down period. Both qualitative and quantitative measurement techniques are utilized to measure the characteristics of a test fire, thermal and structural responses to fire. The test results will be used for validation of prediction models and for technical information necessary to advance performance-based engineering of composite structures under realistic fire.