This article presents a FFT method in statistic quality control. Most previous statistic quality control tools can not analysis the quality data in both time and frequency. In this paper, the FFT method is used to transforming our view of the quality signal from time-based to frequency-based, to provide a basis for current decision during production as to when to hunt for cause of variation and take action intend to correct them, and when to leave a process alone. In spite of the apparent simplicity of the using FFT, the enterprise involved in trying to manage and improve quality find that it use calls for an entirely new point of view. The article conclusion proved by a practical example.