A series of compartment fire experiments was conducted on long-span steel-concrete composite floor beams designed and constructed following U.S. building codes and standards. The test program consisted of five 12.8 m long composite floor beam specimens with various end support conditions. Each specimen was constructed as a partially-composite beam consisting of a W18×35 steel beam and an 83 mm thick lightweight concrete slab cast on top of 76 mm deep ribbed steel decking units. Test variables included two types of simple shear connections (shear-tab and welded/bolted double-angle connections) and the slab continuity over girders. One specimen with the double-angle connections at the ends was tested at ambient temperature and the remaining four specimens were tested under simultaneous mechanical and fire loading. This report, Part 1, presents details of the test setup, specimens, design basis of fire loading, instrumentation, and the behavior of the composite beam with double-angle connections at ambient temperature. The ambient temperature test indicated that the composite beam specimen failed by a shear stud near the west end, followed by concrete breakout failure and yielding of the steel beam. The measured moment capacity was approximately 80% of the calculated flexural strength. The double-angle connection at the west end failed by weld fracture, which caused collapse of the composite beam. The ambient behavior of the composite beam specimen presented herein will serve as a baseline to compare with the composite beam assemblies tested under combined mechanical loads and fire exposure, which are presented in a subsequent report; Part 2 (Choe et al. 2019). The datasets obtained from these tests provide technical information to advance performance-based design of composite floor assemblies in steel-framed buildings subject to fire.

This report presents the results of compartment fire experiments on four 12.8 m long composite floor beams with various end support conditions. Each specimen was constructed as a partiallycomposite beam, consisting of a lightweight concrete slab cast on 76 mm (3 in.) deep ribbed steel decking and a W18×35 steel beam acting compositely via 19 mm (3/4 in.) diameter steel headed stud anchors welded on the top flange. Test variables included two types of simple shear connections (shear-tab and welded-bolted double-angle connections) and the presence or absence of slab continuity over girders. Hydraulic actuators were used to apply gravity loads to the specimen under fire conditions. The natural gas-fueled burners were used to create a compartment fire with the maximum heat release rate of 4000 kW. This study evaluated the characteristics of the fire loading and thermal and structural responses of the specimens. The test results indicated that the thermal restraint conditions significantly affected the behavior and failure modes of the specimens with simple shear connections. The specimens resisted gravity loads at large vertical displacements near midspan (approximately a ratio of the span length over 20) without collapse under fire loading. Various limit states and vulnerabilities to fires were observed, including local buckling of the steel beams near supports, flexural failure (yielding of steel beams and concrete fracture near restrained end supports), and connection failure (weld shear or bolt shear) during heating and cooling which could lead to partial or total collapse of the floor system.

No abstract

The policy of the National Institute of Standards and Technology is to use metric units in all its published materials. Because this report is intended for the U.S. building construction industry, which uses inch-pound units, it is more practical and less confusing to use inch-pound units, in some cases, rather than metric units. However, in most cases, units are presented in both metric and the inch-pound system. Certain commercial entities, equipment, products, or materials are identified in this document in order to describe a procedure or concept adequately. Such identification is not intended to imply recommendation or endorsement by the National Institute of Standards and Technology, nor is it intended to imply that the entities, products, materials, or equipment are necessarily the best available for the purpose.Another policy of the National Institute of Standards and Technology is to include statements of uncertainty with all NIST measurements. In this document, however, some measurements of authors outside of NIST are presented, for which uncertainties were not reported and are unknown. iii National Institute of Standards and Technology ABSTRACTThis report presents the experimental design and results of a series of localized fire tests on structural steel I-shaped beams. A total of nine tests were conducted in the National Fire Research Laboratory, including thermal tests (Tests 1 through 5) and four-point bending tests at ambient (Test 6) and elevated temperatures (Tests 7 through 9). The specimens were nominally 6.2-m long W16×26 beams made of ASTM A992 steel. Each specimen was vertically supported with either a (i) simple support condition, or (ii) double-angles bolted to laterally braced support columns. The thermally-loaded specimens were exposed to fire generated using a natural gas burner. The burner had an area of 1 m 2 and was located 1.1 m below the bottom flange of the beam at midspan. A four-point flexural loading scheme was used to apply concentrated forces at two locations 2.44 m apart around midspan. The recorded data included temperatures, heat release rates from the burner, and structural measurements including forces, displacements and strains. The repeatability of several measurements was evaluated. The test results indicated that the heating rate of the specimen was sensitive to the prescribed heat release rate versus time relationship used in each test. However, the thermal gradient developed in the fire-exposed cross sections of the beam never achieved linearity under the localized fire exposure. When the exposed (bottom) flange temperature was maintained to 616 °C, the load-bearing capacity of the simply-supported beam was reduced to 67 % of its room-temperature capacity. When the same simply-supported beam was initially loaded to 67 % of its room-temperature capacity and then exposed to a growing (t-squared) fire, the exposed flange temperature of the beam at failure was 663 °C. When the beam specimen was supported by double-angle connections and subjected to the same bending mo...

No abstract