Fujimoto et al. 2004
The behavior of short circular and rectangular CFT beam-columns was investigated through a series of experiments and the development of an analytical model. The experimental program included a wide range of test parameters including high strength materials. Using a fiber model, moment-thrust curvature relationships were developed and compared to the experimental results.
Experimental Study, Results and Discussions
Sixty-five CFT specimens were tested (33 circular and 32 rectangular). The test parameters under investigation were: nominal steel tensile strength (with values of 58, 86, and 113 ksi), nominal concrete strength (with values of 2.9, 5.8, and 11.6 ksi), D/t ratio of the steel tubes (obtained by the changing the diameter or width of the specimen), and axial load. All specimens had an L/D ratio of 3.0. The circular tubes were manufactured by bending and seam welding a flat plate. The square tubes were manufactured by welding together two cold formed channel sections. End plates were welded to both ends of the tubes.
The tests were conducted at three different locations; consequently, three different loading schemes were used. The circular specimens were in a pinned-pinned condition and subjected to constant axial load and monotonically increasing end moments. Some of the rectangular specimens were in a pinned-pinned condition and subjected to monotonically increasing axial load with constant eccentricity. The remaining rectangular specimens were in a fixed-free configuration and subjected to constant axial load and monotonically increasing end moment. Tables of the test results as well as figures of moment curvature relationships were provided.
Many of the specimens (nine circular and eight rectangular) failed by cracking in the weld connecting the end plate. These failures were attributed to poor workmanship and the results of those tests were omitted from the discussion. Most of the specimens performed well and exhibited high ductility. However, some of the specimens manufactured with 58 ksi steel exhibited a significant drop in strength after the maximum value was reached. This was attributed to either the inability of normal strength steel to provide sufficient confinement to high strength concrete or local buckling of the steel tube before the concrete could be confined. The specimens manufactured with 113 ksi steel experienced significant strength deterioration after local buckling of the steel tube. The authors noted that since the steel was high strength, it was carrying a larger portion of the load which led to a larger decrease upon local buckling.
The fiber method was used to derive moment-thrust curvature relationships for the columns. The concrete material model for the circular sections accounted for confinement with the increased strength and improved behavior after attaining maximum strength. The concrete material model for the rectangular sections only included the improved behavior after attaining maximum strength. Size effects in the concrete were also considered. The steel material model was derived assuming a value for hoop stress and assuming the von Mises yield criterion. Local buckling was taken into account within the steel material model for the rectangular steel tubes. The values for ultimate moment from the experiments were compared to those obtained from the fiber model as well as two other simpler methods. The fiber model was shown to provide the best estimate.
Fujimoto, T., Mukai, A., Nishiyama, I., and Sakino, K. (2004). “Behavior of Eccentrically Loaded Concrete-Filled Steel Tubular Columns,” Journal of Structural Engineering, ASCE, Vol. 130, No. 2, February, pp. 203-212.