Applying Gurson Type of Damage Models to Low Constraint Tests of High Strength Steels and Welds

Abstract

Pipelines experiencing displacement-controlled loading need to have adequate strain capacity. Large tensile strain capacity can only be achieved when the failure processes are ductile. In ductile failure analyses, the strain capacity may be determined by two approaches. The first approach uses the conventional fracture mechanics criteria, such as the attainment of the critical crack tip opening displacement, to assess the onset of the crack propagation. The other approach uses damage mechanics models in which the onset and propagation of cracks are controlled by the nucleation, growth, and coalescence of voids in the material. The damage mechanics models can provide some insights of the ductile failure processes as they have more physical mechanisms built in the constitutive model. In this paper, the Gurson-Tvergaard-Needleman (GTN) model is applied to two types of low-constraint tests: curved wide plates and back-bend specimens. The wide plate test is considered more representatives of full-scale pipes than the conventional laboratory-sized specimens, but requires large-capacity machines. The back-bend test is a newly developed low-constraint laboratory-sized test specimen. A relatively simple approach to determine the damage parameters of the GTN model is discussed and the transferability of damage parameters between those two test types is also analyzed.