Seng, Sophea. Investigation on cold-formed steel built-up box beams with large cross-section. Master's Degree(Engineering and Technology). Thammasat University. Thammasat University Library. : Thammasat University, 2017.
Investigation on cold-formed steel built-up box beams with large cross-section
Abstract:
This research performed experimental studies and finite element analyses of cold-formed steel built-up box beams with large cross-section. Torsional rigidity of open single C-section is weak since cold-formed steel cross-section comprises of thin elements, and the single C-section is unsymmetrical. Modes of buckling of cold-formed steel member consist of local, distortional, and lateral-torsional buckling. In this study, flexural tests on built-up box beams assembled by two C-sections were conducted. Twelve built-up specimens with large cross-sections of C20019 and C30024 were selected. Material properties of cold-formed steel are: tensile yield strength for 1.9 mm thickness and for 2.4 mm thickness. Stiffening plates and self-drilling screws were used as connection system on both top and bottom flanges of built-up box beams. Four-point bending test was employed to achieve the condition of pure moment. Influences of span length and connection spacing on the behavior of built-up box beams have been discussed, in which span lengths are 2,800 mm and 3,800 mm, and connection spacings are L/2, L/3, L/4, L/6 (L is the clear span length). It was observed that there was no failure of stiffening plate and self-drilling screw. From the results of maximum loads, it was observed that the largest different percentages of maximum load compared with the specimen in case of connection spacing of L/2 in each series test including C20019-3M, C20019-4M, and C30024-4M were 6.77%, 5.29%, and -1.99% respectively. Hence, there was slight difference of maximum load when the connection spacing changed. Two failure modes were found including local buckling and the combination of local buckling and distortional buckling. In additions, when the span length decreased from 3,800 mm to 2,800 mm, the maximum load increased in the range of 32% to 42% with the average increase of 36%. Nonlinear finite element analyses have been conducted by using ABAQUS to replicate the results with the experimental studies in term of maximum loads, failure modes, vertical deflections, and strains. The comparison results between the experimental studies and finite element analyses were found to be in a good agreement. The difference percentage of maximum load between experimental and numerical results were in the range of 0.08% to 12.04%. The buckling modes and strain values obtained from numerical results appeared to be the same with the experimental results. Thus, the present of finite element analyses are reliable and can be used to predict the maximum load and failure mode with an acceptable error.
Thammasat University. Thammasat University Library