Sakino 1995
Circular CFTs were tested to investigate their flexural behavior. The moment capacity of the specimens was examined and the test results were compared with the AIJ (1987) design code provisions.
Experimental Study Results and Discussions
Twenty-eight circular CFT specimens were tested under constant axial load and monotically increasing uniform bending moment. Low, medium, and high strength materials were used to manufacture the specimens. The steel yield strength and concrete compressive strength ranged from 41.2 to 121.1 ksi and from 3.55 to 11.26 ksi, respectively. Different D/t ratios were selected for each steel grade, varying from 26.9 to 152. The axial load ratios (P/Po) were chosen between 0.15 and 0.8. The L/D ratio was constant and equal to 3. The ends of the specimens were welded to 1.57 in. thick end plates. The end plates were attached to cylindrical bearings to simulate pinned end conditions.
It was observed that the D/t ratio and axial load ratio were the dominant factors affecting the flexural behavior. The specimens having D/t ratios larger than 100 showed a rapid decrease in strength. However, the response was ductile when the D/t ratio was lower than 50. The specimens having high strength steel, high strength concrete, and a low D/t ratio also achieved ductile response, even under high axial loads. Excluding the specimens with small D/t values, local buckling was observed for most of the specimens, and it generally took place either at the peak point or in the post-peak region of the load-deformation responses.
Analytical Study
Interaction curves for the specimens were obtained analytically using the stress block approach presented in the AIJ (1987) design code provisions. Two moment-thrust analyses were performed for the specimens. In the first one, no reduction factor for concrete strength was introduced. However, in the second one, a 0.85 reduction factor was applied to account for the concrete stress block. It was found that the analysis with full concrete strength gave conservative strength estimates for low D/t ratios but unconservative strength estimates for high D/t ratios. The enhancement in capacity observed in the case of low D/t ratios was attributed to the confinement of the concrete and strain hardening of the steel. On the other hand, the analysis that included a reduction in the concrete strength predicted the capacity of the specimens conservatively. However, the capacity predictions for the low D/t specimens were found to be too conservative.
Reference
Sakino, K. (1995). “Elastoplastic Behavior of Concrete Filled Circular Steel Tubular Beam-Columns,” Proceedings of the Third Joint Technical Coordinating Committee Meeting, U.S.-Japan Cooperative Research Program, Phase 5: Composite and Hybrid Structures, Honolulu, Hawaii, June 26-28, 1995, National Science Foundation, Arlington, Virginia.