The 3-phase BLDC motor(Brushless DC motor) is widely used for its simple structure and high efficiency. The torque of BLDC motor is calculated multiplying its trapezoidal back EMF by phase currents. Therefore, the torque is ideally uniform if each pha...
The 3-phase BLDC motor(Brushless DC motor) is widely used for its simple structure and high efficiency. The torque of BLDC motor is calculated multiplying its trapezoidal back EMF by phase currents. Therefore, the torque is ideally uniform if each phase currents are controlled having 120° phase difference and the shape of the back EMF is trapezoidal.
It is required to measure the exact position to drive the BLDC motor, because the maximum torque is generated by synchronizing the phase current with the flux linkage which is represented by the permanent magnet rotor. Generally, hall effect sensors, encoders and resolvers are used for the position sensor. However, there are some significant disadvantage using these sensors in the aspects of the drive cost, the volume needs, the noise immunity and the performance reliability. Therefore, It has been investigated in several studies to driving BLDC motors without the position sensors. The conventional sensorless methods are following.
It is the most widely used sensorless method detecting the ZCP(Zero Crossing Point) of the back EMF. However, estimating the exact CP(Commutation Point) with this algorithm requires phase delay filters and additional circuits. Moreover, the estimation performance decreases in the low and high speed region.
Another position estimation method detecting 3rd harmonics of the back EMF requires the neutral point of the BLDC motor, which is not available in common, and additional circuits are also needed.
The ZCP of the back EMF is also detected by measuring the freewheeling current of the open phase, but it is very hard to measure the freewheeling current in the low speed region and additional circuits are also needed.
In this paper, a new sensorless method for BLDC motors is proposed with estimating back EMF using Idc and Idc_Model error. And then estimating angle and speed using back EMF. there is no additional circuits and it is robust to parameter errors. The simulations and experimental results are shown to verify the proposed method.