Material-data-driven prediction of sheared surface tears of fine blanked parts

Abstract

Abstract. Fine blanking is a cost-effective manufacturing technology for mass-producing sheet-metal components with high sheared surface quality. For steels with higher carbon content, the quality of fine blanking products is significantly influenced by microstructural characteristics, such as the morphology and distribution of carbides, which can be controlled through heat treatment [1]. Precisely, there is a relationship between spheroidization of carbides and the occurrence of tears at the sheared surface of fine blanked parts especially at the tip of a gear [2]. Furthermore, material characteristics can vary both along a sheet-metal coil and from coil to coil despite tight tolerances [3] leading to unpredictable tearing. To monitor the fluctuation of material characteristics at regular intervals along sheet-metal, non-destructive testing (NDT) is used before the process providing information about the microstructure. While in the state of the art, the data from NDT was used to calculate the mechanical properties and to optimize the process based on these properties [4], in this paper, NDT data is used to predict the sheared surface tears of fine blanked parts, without the reduction of the content-rich data to mechanical properties. For this purpose, an experiment was conducted on the fine blanking of the steel 42CrMo4+A to produce components resembling a gear shape. Prior to the manufacturing process, the material was measured using an eddy current sensor, and subsequently the tearing of the fine blanked parts was evaluated. For the prediction of sheared surface tears, linear regression methods were used, and a feature selection was done to find the excitation frequencies of the eddy current sensor with the highest impact on the tearing. It was shown that the eddy current measurements along the coil contain valuable information about the tearing of the fine blanked part.