ABSTRACT
The pulse-width modulation (PWM) is a technique that provides an arbitrary shape of voltage by adjusting the ratio of on and off duties. The sinusoidal PWM can be realized by analog circuits, whereas the space vector PWM consumes a DSP computing power since PWM duties are calculated by a partitioning method in each PWM period. With the PWM techniques, the semiconductor switches are not allowed to operate in the active region. As physical switching devices, IGBTs are most widely used. But, power MOS-FETs are suitable for low voltage, low power applications where fast switching is required. Besides, physical switches cause conduction losses due to the switch on-drop voltage during the conduction stage. As the voltage blocking capability increases, the on-drop voltage increases. However, practical semiconductor switches show non-ideal behavior: During the turn on transient, the (collector-emitter) voltage that a switch bolsters drops only after the switch (collector) current builds up.
