Some phenomenological and computational aspects of sheet metal blanking simulation M. Rachik a,* , J.M. Roelandt a , A. Maillard b a Universite  de Technologie de Compie Ágne, GSM, Lab. Roberval, BP 20529, 60205 Compie Ágne, France b CETIM, Service Me Âtaux en Feuilles, 52 avenue Fe Âlix Louat, BP. 80067, 60304 Senlis Cedex, France Received 17 April 2001; received in revised form 27 February 2002; accepted 28 June 2002 Abstract In this paper, we present a comprehensive experimental and numerical study of the sheet metal blanking process. Various blanking tests involving different materials and geometry are investigated and the numerical results are compared with experimental data. Forthe numerical aspects of this study, the main topics discussed are sheet metal constitutivemodel, the numerical integration algorithm and mesh adaptivity. Taking into account the complexity of the blanking process, we chose an explicit ®nite element code to overcome the convergence problems. OurnumericalmodelisthusbasedonadynamicexplicitschemeassociatedwiththeALEformulationformeshadaptivity.Sincethechoiceof the sheet metal constitutive model is important to achieve the product shape prediction and the burr height estimation, we compare the Parndtl±Reuss plasticity model with Gurson±Tvergaard±Needleman coupled plasticity-damage model. The comparisons between numerical resultsandmeasurementsshowthatthemaximumpunchforceisstronglyrelatedtotheplastic¯owbutthattheburrheightestimationrequires damage modelling. In addition, Gurson's modi®ed model greatly improves the punch force prediction. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Blanking; Forming process simulation; Ductile fracture; Explicit dynamic; ALE 1. Introduction Sheet metal blanking is an industrial process widely used in automotive, electronic and several other industrial appli- cations. It consists in separating a blank from a sheet by means of a high-localised shear deformation due to the action of a punch. The pioneering work of Johnson and Slater [1] shows that the phenomena involved in the blank- ing operation have been well known for a long time. The authors gave an interesting schematic representation of punch force versus punch penetration diagram where the phases associated with the different phenomena are well identi®ed. They clearly show that the maximum punch force is strongly related to the plastic ¯ow and that it does not depend on the initiation and the propagation of a fracture. It should be pointed out that amongst several existing sheet metal forming processes, the blanking process stands apart since it leads to plastic shearing followed by the creation and the propagation of cracks (Fig. 1). Despite its widespread use, the design of the metal blanking is often based on trial and error tests, that is a time consuming procedure. Over the last few decades, several researches have been devoted to the modelling of the blanking process. The earlier work of Atkins [2] and Zhou and Wierzbicki [3] concerned the development of some analytical models. These simple models can be used to estimate the punch force but they are not able to inves- tigate all the phenomena involved. Moreover, they are limited to plane strain problems. In this sense, the ®nite element method seems to be a powerful tool that can improve the knowledge of sheet metal blanking since it is more adapted to the complex constitutive models and the general boundary conditions. For these reasons, the numer- ical simulation of sheet metal blanking has gained a wide- spread interest within the computational mechanics community. In this way, several ®nite element approaches have been put forward to model the blanking process. These models deal with different aspects of the shearing process and are more or less able to describe all the phenomena involved. They are mainly characterised by the way that the material behaviour is modelled (with or without ductile fracture assessment) and the mesh adaptivity is handled. Maiti et al. [4] performed the assessment of the in¯uence of some process parameters like clearance and friction on the punch force. They used an elastoplastic model with the Journal of Materials Processing Technology 128 (2002) 256±265 * Corresponding author. E-mail address: mohamed.rachik@utc.fr (M. Rachik). 0924-0136/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII:S0924-0136(02)00460-0