Accurate identification of spindle modal parameters is critical to realizing a new generation of “smart” machine tools with built-in self-diagnosis capability. This paper describes a new approach to extracting spindle modal parameters from the output measured during operation, based upon stochastic subspace identification. The technique accounts for structural dynamic behavior, associated with the spindle rotation, that is not present when the spindle remains stationary. Experimental results conducted on a customized spindle test bed under different speed-load combinations confirm the effectiveness of the new technique for on-line spindle condition monitoring.

This content is only available via PDF.
You do not currently have access to this content.