Influence of experimental boundary conditions on the calibration of a ductile fracture criterion

Abstract Hybrid experimental-numerical approach is the commonly used technique to identify the material parameters of ductile fracture criteria. In this work, attention is paid to the precise definition of the Finite Element (FE) model of fracture tests by applying real boundary conditions coming from the local displacement field. The aim is to investigate the effect of the way to reproduce experimental boundary conditions on the accuracy of the hybrid experimental-numerical approach. A stress triaxiality and Lode angle based ductile fracture criterion is used to predict the onset of ductile fracture for AA6061-T6 aluminum alloy sheets. To calibrate this criterion, a series of fracture tests is carried out up to fracture. Notched tensile specimens with different radius, tensile specimens with a central hole and shear specimens are used to cover a wide range of stress states. Strain distribution over the surface is measured with digital image correlation technique. Two numerical models are defined for each geometry type, a first one constrained with real experimental boundary conditions and a second one with ideal boundary conditions, to predict stress and strain distributions. By comparing results of both approaches with experiments, it is shown that the model constrained by local experimental boundary conditions provides a better agreement with the experiments than the simplified model does, which underlines the sensitivity of the hybrid method to the adopted boundary conditions. Furthermore, several fracture test combinations are used to calibrate the ductile criterion and the effect of the calibration tests is discussed.