Comparison of the CEL and ALE approaches for the simulation of orthogonal cutting of 15-5PH and 42CrMo4 materials

Abstract

The modeling of cutting processes by the finite element method represents a great challenge for scientific researchers. Indeed, the plastic strain and strain rate in cutting induce a high risk of mesh distortion, especially for models based on a Lagrangian formulation which were the first to be developed. To overcome this problem other formulations, such as the Arbitrary Eulerian-Lagrangian (ALE), which is a combination of the classical formulations (Eulerian and Lagrangian), have been developed. Despite very interesting results in the simulation of orthogonal cutting, the implementation of this type of model remains delicate, and in particular the choice of initial geometrical conditions which can lead to a sudden stop of the calculation induced by a mesh distortion. To overcome these limitations, the Coupled Eulerian-Lagrangian (CEL) approach has been developed to simulate the orthogonal cutting process. In this paper, the CEL and ALE approaches were compared to simulate the orthogonal cutting process. The first objective was to compare the quality of the predicted physical quantities (forces, chips) over a wide range of cutting speeds and cutting thicknesses for two different materials. The second objective was to evaluate the time required for a complete simulation, integrating the time to set up the initial conditions (chip geometry) and finally the computation time of the machining operation. This work has shown that ALE and CEL formulations lead to very similar quantitative results that are pretty close to the experimental results. The ALE formulation has the advantage of having a reduced computation time compared to the CEL formulation. However, the preparation of the simulation is easy and robust for the CEL, whereas the selection of initial geometrical conditions for the ALE calculation requires several trials and errors before finding stable conditions. The CEL formulation is therefore a solution to be favored when launching several simulations in an automated way thanks to its reliability.