Experimental and numerical study of pinching phenomena in sheet metal rolling processes

Abstract

Steel sheets are an essential raw material for a wide range of applications, such as household appliances, packaging, construction, shipbuilding, industrial machinery, automotive, and energy industries. The worldwide market for steel sheets faces intense competition and increasing demand for light-weight metal components to reduce CO2 emissions. As a result, newly developed grades of advanced high-strength steel (AHSS) have gained attention, especially from the automotive industry. AHSS allows for down-gauging due to its higher strength compared to other conventional steel grades. However, due to the low thickness of the sheets, rolling of AHSS is a critical process that may suffer from instabilities, such as pinching. Pinching represents a complex type of phenomena related to inhomogeneous stress distributions in the strip, which may arise from disruptions during the rolling process. Similarly to other shape defects, pinches can be related to uneven strip deformations in the roll bite, which result in inhomogeneous stress distributions across the strip’s width. Pinching defects in steel sheets appear as surface marks, wrinkling, repetitive rippled areas, and local ruptures. In the most severe cases, the strip breaks completely, causing damage to the rolls and considerable manufacturing downtime.Controlling the stability and enhancing the performance of the rolling process are top priorities for steel manufacturers. These tasks aim to minimize the occurrence of defects, ensure consistent product quality, and enhance the efficiency of the manufacturing process. Therefore, better understanding of instability phenomena like pinching is required for determining suitable solutions to prevent them and to obtain a stable rolling process. However, despite being a commonly reported issue among steel manufacturers, pinching has been poorly understood in terms of its underlying mechanism. Currently, there is a lack of research examining the mechanisms behind pinches, both in terms of experimental and numerical investigations. Without a comprehensive understanding of these phenomena, it is unfeasible to develop effective measures to prevent pinches and ensure stable operations of rolling mills. Therefore, the aims of this study are: firstly, to identify the mechanism and possible causes of pinching, and secondly, to develop a simulation tool that can be used to analyze pinching phenomena and design guidelines for the selection of robust production settings in cold rolling mills. To this end, both experimental study and numerical modelling are performed, as presented in this work.The experimental investigation of pinching phenomena presented in this work provide an in-depth understanding of the circumstances that lead to pinching through a series of cold rolling tests and the analysis and characterization of pinching defects.To study pinching phenomena, an appropriate tool is needed to replicate and investigate actual pinching events. Simulation models are essential for predicting the occurrence of pinching during the rolling process. However, existing numerical models of rolling do not succeed to reproduce the occurrence of pinching. This is because pinching is a complex phenomenon that depends on the strong interplay between local deformations within the roll bite and the stress state outside the roll bite. To capture this complexity, a numerical tool must be capable of modeling the process both at a millimeter (or sub-millimeter) scale within the roll bite and at a meter scale outside the roll bite. Moreover, to effectively study pinching events, a three-dimensional rolling model is necessary, as the distribution of stresses and strains across the strip's width is a crucial factor. The finite element method (FEM) is a well-established numerical tool for simulating metal forming processes, and is therefore a suitable technique for analyzing and predicting defects during rolling. However, accounting for all the relevant physics of the rolling process in a conventional 3D FEM model would result in an unfeasible computational time. This work proposes a numerical strategy to decrease the computational expense of 3D sheet rolling FEM simulations. The method involves coupling a global model, which represents the behavior and stress state of the strip outside the roll bite, with a local model that reproduces the deformation mechanics inside the roll bite. The global model is a shell finite element model of the sheet, while the local model is a high resolution 2D plane strain model of the roll bite. The developed approach has been validated by comparing its results to those of a conventional full 3D rolling model under stable rolling conditions. Additionally, this model has been employed to carry out a qualitative analysis of instability phenomena that arise during thin strip rolling. Such phenomena include flatness defects that result from disruptions in the frictional conditions. The simulation results demonstrate that locally varying friction induce local variations in the thickness strain, which cause stress re-distributions in the rolled sheet, resulting in flatness defects. Therefore, the proposed model offers a cost-effective alternative to more expensive 3D FEM models in the analysis of complex instability phenomena that can lead to defects during sheet metal rolling processes.