The relation of tensile specimen size and geometry effects to unique constitutive parameters for ductile materials

Abstract
Direct observation of solid deformation at all points in a material test specimen is generally not possible. This becomes important when an internal point experiences the maximum effective deformation of the entire specimen. The simple reduction of load-displacement data to material constitutive parameters, as for inelastic tensile specimen necking, is precluded. Although the mechanics of necking has received considerable attention, the complementary issue of constitutive parameter determination from inelastic specimen experimental data has not. This investigation addresses both aspects of the problem for HY-100 steel. An experimental-computational procedure is demonstrated, which generates material constitutive parameters valid at strains beyond those at the onset of necking. Neck deformation, specimen size and specimen geometry effects are used to ensure the uniqueness of the parameters in the sense that a single uniaxial continuum true-stress–true-strain curve predicts the behaviour of laboratory specimens with different geometries. The solution parameters for HY-100 steel are developed, in the context of an incremental elastic–plastic constitutive formulation, by this procedure. These results support the view that tensile specimen necking is the result of the interaction between imperfection-free specimen geometry, material behaviour and applied load for relatively unsophisticated constitutive formulations if the constitutive parameters are accurately determined. In addition, broken symmetry of specimen deformation is computationally predicted and experimentally confirmed for long, slender specimen geometries.