Controlling BLDC Motors in Trapezoidal Form

One of the easiest motor control methods for brushless DC motors is the trapezoidal, six-step, or 120° block commutation control. Optimum torque generation requires applying square-wave currents to motor phases in alignment with the trapezoidal back-EMF profile of BLDC motor. MOSFETs of the inverter drive can exhibit only six combinations of on/off states. Therefore, this method has another name—the six-step—resulting in six possible orientations of the stator field within the plane of rotation of the magnetic field of the rotor.

Depending on the desired direction of rotation of the motor, the six possible inverter states must follow a specific sequence. This is necessary so that the orientation of the stator and rotor magnetic fields produces the maximum torque. There are two ways of sensing the rotor position for determining proper commutation timing—sensing through Hall-effect sensors on the motor, or a sensorless way of back-EMF sensing of the rotating motor phases.

Of the two, using sensors requires no voltage or current feedback signals for proper operation. Rather, the position feedback from the Hall sensors is adequate to determine the proper sequence for energizing the motor phases. Hall sensors in strategic positions in the motor can sense rotor position as a result of the rotating magnetic field of the permanent magnets in the rotor. Trapezoidal control using sensors is easier to implement, as it allows for proper commutation even during startups—the information about the rotor position is available even at zero speed.

For trapezoidal control without sensors, the proper motor commutation sequence depends on the back-EMF that the motor’s rotation generates. Such trapezoidal control requires energizing only two motor phases at a time. As the non-energized phases have no current flowing through them, it is possible to sense the back-EMF they are producing during the non-energized times. Typically, such back-EMF positional feedback in BLDC motors is trapezoidal and is either linearly increasing or decreasing. Therefore, most positional feedback techniques using back-EMF use a zero-crossing detection for determining the moment when it crosses a reference point. This can be either half the DC bus voltage or the neutral motor voltage.

Sensorless control has a major drawback. As the magnitude of the back-EMF is proportional to the rotational speed, the rotor must be rotating at a minimum speed to generate a back-EMF of adequate magnitude for sensing the rotor position properly. Therefore, it is necessary to use a startup mechanism for kick-starting the motor until it reaches an adequate rotational speed.

Although it is easier to implement a trapezoidal control with sensors, the Hall sensors add an increased cost. Additionally, signals from Hall sensors may be noisy and may require hardware or software filtering. The motor also requires more wiring, which in some environments, may be a challenge. On the other hand, sensorless control is more complex. It is necessary to tune it to meet specific loads or operating conditions and may face difficulties in starting up under heavy loads. That makes sensorless control well-suited for applications with a well-known load profile that increases with speed, such as for a fan.