摘要
This paper examines the relationship between the ball screw preload variation and detected vibration signals. A preload-adjustable ball screw feed drive system based on a modified double nut structure is constructed. Meanwhile, a lumped dynamic model to study the preload variation of the system is proposed. A MEMS accelerometer based sensing system is implemented for obtaining experimental data from the constructed feed drive system. Signals at the ball nuts are acquired and analyzed. Through spectrum analysis of the processed signals and the results of mathematical dynamic modeling, the preload variation can be diagnosed in two indexes: peak frequency shift and the magnitude variation of its peak frequency. The experimental results, agreeing with the trend of mathematical modeling, show peak frequencies display increasing values as the preload steps up and the peak power levels exhibit rising trends as the adjusted preload level of the ball screw increases. This study provides an approach to monitor the health status of the used ball screw, which can be significant for better performance of the feed drive system. © 2011 Elsevier Ltd. Allrightsreserved.