To reduce the cost of Lithium-Ion battery production, waste detection is essential. The reliable detection of defects in electrodes enables quality control and a well-founded decision about whether a piece of electrode is scrap or not, which leads to an increase in cell quality and cell safety down the production line.
A widely used inline system for defect detection is an optical detection system based on line cameras and flash lights. The electrode is scanned by the cameras and subsequently, brightness differences on the surface are detected and processed inline. The characteristics of the defect image allow for an automated classification of the defects based on brightness, form and compactness. Furthermore, the detailed imaging of defects sets the basis for the identification of causes and estimation of the defect severity.
The poster provides an insight into the eight defined defect classes and the methodology of an inline optical defect inspection system. These include pin holes, stripes, cracks, contaminations and others. By using the defined defect classes a large proportion of the defects arising during electrode manufacturing is described and their criticality can be estimated with help of current literature. Some of the mentioned defects are negated during the calendaring step, but others show no major change to their features. Therefore, the defect detection after the electrode calendaring is of high interest as well but not as easy because of the increasing electrode gloss. Due to this change of defect features and the subsequent cutting and stacking, the previously known position of the defect on the electrode coil is lost. To be able to still track the defects it is necessary to link them with a data-matrix-code (DMC) in the sense of “tracking and tracing” in order to be able to automatically dispose scrap or track the influence of defects on the cell performance.