Serious wear of the cutter will result in poor machining accuracy and machining efficiency. It is very important to promptly know the cutter has been in a critical wear status and replace the cutter before the damage occurs on the workpiece. By using empirical method, an on-line diagnosis method for detecting the occurrence of critical wear of a cutter was proposed in this study. The variation trend of the cutting vibration caused by the cutter wear was analyzed and the threshold of detection was determined based on a lot of experiment data. In addition to check the variation of vibration, the Fast Fourier Transform method was also applied to check the frequency characteristics of the cutting vibration. Based on the developed detection algorithm, a computer assisted on-line critical-cutter-wear monitoring and control system was developed in LabView language. The verification experiment result has shown that the system can detect the occurrence of critical wear of cutter and instantly stop the machine for cutter replacement.