Diagnosing Spindle Defects Using 4-D Holospectrum

A spindle is one of the most commonly used mechanical devices in many industrial sectors, such as the large rotating machinery industry and the machine tool industry. Diagnosing spindle defects, such as unbalance, misalignment, and crack, plays a very important role in reducing operation and maintenance costs. This article presents a new method of diagnosing spindle defects using a newly developed sensor signal processing technique called four-dimensional (4-D) holospectrum. The experimental results show that the spindle defects, including unbalance, misalignment, and crack, have distinct features in the 4-D holospectrum. Compared to the conventional spectral analysis and/or modal analysis methods, the 4-D holospectrum method is more suitable for computer-automated diagnosis.