Condition-based maintenance of machine tool spindle units through vibration analysis

Abstract

Manufacturing enterprises need a robust condition-based maintenance (CBM) program that can provide accurate health information about equipment conditions, and reasonable predictions about how that condition is likely to change in the future. This information enables the enterprise to avoid unexpected downtime, keep production running, and lower the direct and indirect costs due to maintenance activities. The spindle unit of a machine tool is a critical subsystem that is responsible for the metal removal process, and due to this criticality, spindles are often the subject of vibration monitoring CBM programs. However, state-of-the-art spindle monitoring guidelines are insufficient to guarantee functional performance of the spindle. Standard guidelines evaluate spindle health based on vibration root-mean-square levels, or other simple statistics, which are not motivated by functional performance characteristics of the spindle, namely the material removal rate, and the final geometric accuracy and surface finish of the workpiece. In this work, a new framework for modal analysis is proposed for CBM of machine tool spindle units, which is based on fundamentals of machining dynamics and stability lobe theory. The framework is shown to have superior capability for tracking spindle health as it relates the vibration signatures to the limiting depth of cut of the manufacturing process. Additionally, two different methods for conducting the modal analysis testing are examined, which allows for improved data collection and addresses the practical challenges of modal testing in the production environment. The first method is based on artificial excitation by a dedicated mechanical impulse generator. The second method is based on operational modal analysis techniques that are applied to the spindle during cutting.