Shallow-depth composite floor systems reduce floor depth by placing deep decking (or precast concrete panels) on the bottom flange of a steel beam that is partially encased in concrete. For ease of construction, an asymmetric I-beam is needed to allow the decking to be dropped vertically onto the bottom flange. Hot-rolled asymmetric I-beams (termed “A-shapes”) are being researched for future large-scale production in the United States. The purpose of the research presented herein was to evaluate the behavior of A-shapes through a full-scale test of a 6.1×12.2 m (20×40 ft) shallow-depth composite floor system. The floor system consisted of A-shapes that spanned between stub columns and deep decking spanned between the beams. The composite floor system had a total depth of 257 mm (10.1 in.). Testing of the floor system included analyzing the torsional response of the A-shape during construction. Vibration testing was performed on the constructed floor system by conducting heel drop and walking tests. The serviceability of these floor systems was further evaluated by conducting service live load testing in locations around the slab. The floor system performed well, with a deflection corresponding to L/690 for a 3.8 kPa (80 psf) equivalent live load. Finally, the ultimate capacity of the floor system was quantified by loading the center beam to failure. Significant composite strength was gained and exceeded 125% of the full composite flexural capacity quantified with a standard effective concrete width. The flexural rigidity of the floor system had a stiffness greater than the gross moment of inertia using twice the standard effective width. The overstrength and increased stiffness are a result of more concrete being engaged in the composite section, which was verified by measuring concrete strain on the slab surface. The full-scale testing conducted on shallow-depth floor systems with A-shapes indicated that it is a viable option for building floor systems.
No takes yet. Share an insight, caveat, or question.
Ottmers et al. (2025) studied this question.