Parsley, Yura, and Jirsa 2000
Push-out tests were performed on eight rectangular CFTs. Three different mechanisms were identified as responsible for the shear transfer between the steel tube and concrete core. Design guidelines were presented along with a proposed bond strength equation and strength reduction factor.
Experimental Study, Results, and Discussion
Eight push-out tests were performed on rectangular CFTs. Two different D/t ratios and two different methods of load application to the steel tube were tested, thus there were four sets of duplicate tests. The steel tubes had a thickness of 0.25 in and a side length of either 8 in or 10 in. The yield strength of the steel was 48 ksi and the compressive strength of the concrete was either 5.8 ksi or 6.5 ksi. Load was applied to the top of the specimen on a portion of the concrete core which extended past the steel tube. At the bottom of the specimen, 1 in. of the steel tube was left unfilled. Half of the specimens were placed with the entire steel cross-section in contact with the reaction surface, thus loading the entire cross-section. The other half of the specimens were supported by shear tabs which were welded to the outer surface of the steel tube. The inside surfaces of the steel tube were not treated and no mechanical shear connectors were used.
Instrumentation was provided to measure the slip between the steel tube and concrete core, as well as, provide information on the distribution of axial load at various cross-sections. With this data, the load-slip response and the distribution of axial load along the length of the member were presented. From these graphs, three shear mechanisms were identified. The first mechanism, adhesion, refers to the chemical bond between the steel and concrete. This mechanism was only active in the early stages of loading and was identified by a very stiff response. The second mechanism, friction, is the stress that develops as a result of the roughness of the interface. This mechanism was active for the entire experiment and was identified by a constant rate of force transfer along the length of the member. The third mechanism, wedging, occurs when the slip is resisted by indentations or geometric irregularities in the steel tube. It was active at different times during the experiments, particularly under two circumstances. First, wedging was active immediately after the adhesion was overcome for the smaller specimens, before the initial indentations in the steel tubes were pushed outward by the concrete core. Second, wedging was active for high values of slip for the specimens with shear tabs, as the rotation of the shear tabs caused distortion of the steel tube. Wedging was identified by non-zero stiffness in the load-slip response.
To formulate an equation for the bond strength, CFTs were related to thin-walled pressure vessels. By assuming that the concrete behaves like a fluid, applying a uniform pressure to the steel tube, the parameter t/D^2 was identified as significant. The experimental results, augmented by two additional results from another study, were fit to a linear equation of t/D^2. A strength reduction factor of 0.9 was selected as appropriate for design of CFT columns.
Reference
Parsley, M. A., Yura, J. A., and Jirsa, J. O. (2000). “Push-Out Behavior of Rectangular Concrete-Filled Steel Tubes,” Composite and Hybrid Systems, ACI SP-196, Aboutaha, R. S. and Bracci, J . M. (eds.), American Concrete Institute, Farmington Hills, Michigan.