Numerical models verification of cracked tubular T, Y and K-joints under combined loads

Abstract

This paper summarizes the key steps involved in the construction of an accurate and consistent finite element model for general cracked tubular T, Y and K-joints. The joint under consideration contains a surface crack which can be of any length and located at any position along the brace–chord intersection. Welding details along the brace–chord intersection, compatible with the American Welding Society (AWS) specifications [American Welding Society. Structural welding code-steel, ANSI/AWS, 15th ed., Miami, 2000], are included in the geometrical model. In order to develop a systematic and consistent modelling procedure, the whole process is divided into four key steps. They are, namely, (1) construction of a consistent geometrical model of the joint with welding details, (2) determination of cracked surface to define the semi-elliptical surface crack profile, (3) generation of well-graded finite element meshes, and (4) stress intensity factor studies around the crack front. To produce a well-graded finite element mesh, a sub-zone technique is used in the mesh generation whereby the entire structure is divided into several sub-zones with each zone consisting of different types of elements and mesh densities. The stress intensity factors (SIFs) are evaluated using the standard J-integral method. Two full-scale T and K-joint specimens were tested to failure under axial load (AX), in-plane bending (IPB), and out-of-plane bending (OPB). In the tests, the rate of crack propagation was monitored carefully using the alternating current potential drop (ACPD) technique. Using the known material parameters C and m, the experimental SIFs were obtained, and they are found to be in complete agreement with the computed SIFs obtained from the generated models. Hence, the proposed finite element models are both efficient and reliable.