Calibration and evaluation of seven fracture models

Over the past 5 years, there has been increasing interest of the automotive, aerospace, aluminum, and steel industries in numerical simulation of the fracture process of typical structural materials. Accordingly, there is a pressure on the developers of leading commercial codes, such as ABAQUS, LS-DYNA, and PAM-CRASH to implement reliable fracture criteria into those codes. Even though there are several options to address fracture in these and other commercial codes, no guidelines are given for the users as to which fracture criterion is suitable for a particular application and how to calibrate a given material for fracture. The objective of the present paper is to address the above issues and present a thorough comparative study of seven fracture criteria that are included in libraries of material models of non-linear finite element codes. A set of 15 tests recently conducted by the authors on 2024-T351 aluminum alloy is taken as a reference for the present study. The plane stress prevails in all these tests. These experiments are compared with the constant equivalent strain criterion, the Xue-Wierzbicki (X-W) fracture criterion, the Wilkins (W), the Johnson-Cook (J-C) and the CrachFEM fracture models. Additionally, the maximum shear (MS) stress model, and the fracture forming limit diagram (FFLD) are included in the present evaluation. All criteria are formulated in the general 3-D case for the power law hardening materials and then are specified for the plane stress condition. The advantage of working with plane stress is that there is one-to-one mapping from the stress to the strain space. Therefore, the fracture criteria formulated in the stress space can be compared with those expressed in the strain space and vice versa. Fracture loci for all seven cases were constructed in the space of the equivalent fracture strain and the stress triaxiality. Interesting observations were made regarding the range of applicability and expected errors of some of the most common fracture criteria. Besides evaluating the applicability of several fracture criteria, a detailed calibration procedure for each criterion is presented in the present paper. It was found rather unexpectedly that the MS stress fracture model closely follows the trend of all tests except the round bar tensile tests. The X-W criterion and the CrachFEM models predict correctly fracture in all types of experiments. The W criterion is working well in certain ranges of the stress triaxiality. (c) 2005 Elsevier Ltd. All rights reserved.