AN ASSESSMENT OF NUMERICAL PARAMETERS INFLUENCING SPRINGBACK IN EXPLICIT FINITE-ELEMENT ANALYSIS OF SHEET-METAL FORMING PROCESS

Springback is one of the key factors influencing the quality of stampe d sheet metal parts in sheet metal manufacture. There have been divers e efforts to evaluate and/or decrease springback for a long time. Anal ytical methods have been mainly developed, at first, for analyzing par ts of two-dimensional or simple three-dimensional geometry. Since it i s almost impossible to apply analytical methods to fully three-dimensi onal parts in general, numerical methods including the finite element method (FEM) have recently been developed and applied to sheet metal p arts of general shape as a result of rapid progress of computing envir onments. However, the results of springback simulated by FEM may vary greatly according to how numerical factors are set, for example elemen t size, punch velocity, contact penalty constant, damping ratio, etc. Therefore, a comprehensive assessment of numerical factors influencing springback is required. In this work, numerical factors influencing s pringback have been evaluated quantitatively using the Taguchi method. To clarify the effect of each factor, the U-draw bending process is c hosen as an evaluation problem because of its large springback. In the analysis, the explicit time integration method is used in the simulat ion of forming stage, whereas the implicit time integration method is applied to the springback stage in order to get the static solution re adily. Numerical experimentation has been carried out under various co mbinations of numerical factors. From comparison of the numerical resu lts with the experimental results, factors important for springback ha ve been assessed. (C) 1998 Elsevier Science S.A. All rights reserved.