ABSTRACT

In orthotropic steel decks, the rib-to-crossbeam joint is one of the fatigue-prone details. However, relevant detail category and the assessment method are not explicitly covered in current major codes. This paper focuses at the finite element modeling of the rib-to-crossbeam joint. Calculation results of two finite element models adopting solid elements and shell elements are compared with experimental results to investigate the finite element modeling method for the rib-to-crossbeam joint. The experimental results show that the distortional stress of rib-to-crossbeam joint is positively correlated with the weld length between rib and crossbeam. The calculation results reveal that for the rib-to-crossbeam joint, the results of shell model and solid model are closer when choosing the intersection point between the crossbeam and the rib as the origin to calculate the frontier distance which is the distance between the weld toe and the strain gauge. Under this scheme, when the frontier distance is shorter than 3.5 mm, the calculation results of shell model are higher than solid model. When the frontier distance is between 3.5 mm and 6 mm, the calculation results of shell model are smaller than solid model. When the frontier distance is larger than 6 mm, the results of shell model and solid model converged. Compared with the experimental results, either the solid model or the shell model gives higher estimations.