Development of a Novel Commutation Method which Drastically Suppresses Commutation Failure of a Matrix Converter

Abstract

This paper proposes the safest novel commutation method for a matrix converter. There are two conventional commutation methods which depend on the polarity of the input line voltage (it is called “voltage commutation”) and depend on the polarity of the output current (it is called “current commutation”). However, problem of the voltage commutation is that commutation failure occurs around zero of the input line voltage. It is difficult to detect its polarity due to depending on offset and delay of the sensor. Similarly, the current commutation failure occurs around zero of the load current. A cause of these detection errors are a detection delay and an offset of a sensor. The proposed commutation method combines the input voltage commutation and the load current commutation. Therefore, the proposed commutation method can decrease the commutation failure without high accuracy sensor. In addition, a voltage error compensation based on the proposed commutation method is proposed in this paper. The proposed method can simply compensate for the commutation error of the output voltage and the input current at the same time.
The effects of the proposed method are confirmed by experimental results with a 750W induction motor and a R-L load. Those results confirm that the proposed commutation can decrease commutation failure. Moreover, the total harmonics distortion of the input current and the output current are 2.6 point and 0.9 point lower than that of the condition without the compensation at 100% output power.