The sound features of the weld pool status in plasma arc welding were systematically investigated after the sound signal was collected with a microphone. The results show that it is difficult to extract information about the weld pool status directly in time domain although the sound signal varies with the weld pool behaviors to some extent. The frequency spectra of the sound signal contain plenty of information about the weld pool behaviors. It is shown from the analysis of the sound generation mechanism that the sound signal of plasma arc welding is mainly caused by the weld pool oscillation, the power source fluctuation and so on. R S algorithm is designed to determine the weld pool status, and it is able to offer the feedback information for the closed loop control of the penetration quality of plasma arc welding. [ The results features of the weld pool status in plasma arc welding were systematically investigated after the sound signal was collected with a microphone. The results show that it is difficult to extract information about the weld pool status directly in time domain although the sound signal varies with the weld of the weld pool behaviors to some extent. The frequency spectra of the sound signal contain plenty of information about the weld pool behaviors. It is shown from the analysis of the sound generation mechanism that the sound signal of plasma arc welding is mainly caused by the weld. pool oscillation, the power source fluctuation and so on. RS algorithm is designed to determine the weld pool status, and it is able to offer the feedback information for closed loop control of the penetration quality of plasma arc welding. [