Brief Description of the PWM Drive Control Principle for Brushless DC Motors

14 min readFeb 7, 2025

Brushless DC Motor (BLDCM) does not have brushes or commutators and requires PWM pulse wave control. Compared to traditional brushed DC motors, it has swapped the positions of stator and rotor (stator winding as rotor, neodymium permanent magnet as rotor, Hall sensor replaces carbon brush for commutation). Compared to traditional brushed DC motors, brushless motors require a dedicated drive control circuit, but they have higher efficiency, lower power consumption, and noise. They can be controlled by servo and perform variable frequency speed regulation.

In traditional brushed DC motors, the rotating part is the winding, while in brushless motors, regardless of whether it’s an “inner rotor” or “outer rotor” structure, the rotating part (rotor) is always a permanent magnet (neodymium iron boron magnet). The stator is the part that produces the rotating magnetic field, mainly composed of silicon steel sheets and windings. This article aims to simply and clearly introduce the working principle of brushless motors as a prelude to writing FOC vector control algorithm-related content.

Left-Hand Rule

The Left-Hand Rule (also known as Fleming’s right-hand rule) was first proposed by British electrical engineer John Ambrose Fleming in 1885. The specific content is that the three fingers of the left hand are mutually perpendicular, and when the thumb points to the direction of the current (amperage), the index finger points to the direction of the magnetic field, and the middle finger points to the direction of the electric current:

The force F (measured in Newtons) on a conductor is equal to the product of the magnetic flux density B (measured in Webers per square meter), electric current I (measured in amperes), and the length of the conductor L (measured in meters):

The left-hand rule states that: a conductor in a magnetic field will experience a force. When the magnetic field is constant, the larger the electric current, the longer the conductor, the greater the force it experiences.

Right-Hand Screw Rule

Also known as Ampere’s Rule, the right-hand screw rule was first proposed by French physicist André-Marie Ampère in 1820. It is used to determine the direction of the magnetic field produced by an electric current. Hold a spiral-shaped conductor with your right hand, and if your four fingers point towards the direction of the electric current flowing through the coil, then:

Note: In electromagnetism, N represents the north pole (North), and S represents the south pole (South).

Basic Principles of Brushed DC Motors

If two permanent magnets are placed in a hollow cylindrical shape with their north and south poles facing each other, and an electric current is passed through a coil placed between them, according to the right-hand screw rule, the coil will produce N and S magnetic fields at its ends. If one end of the coil is attracted to the permanent magnet, the other end will be repelled, causing the coil to rotate in one direction until the magnetic field produced by the coil is opposite to that of the permanent magnets (i.e., the N pole of the coil is attracted to the S pole of the permanent magnet, and vice versa).

If the electric current flowing through the coil is reversed, the magnetic fields produced by the coil will also change polarity. The previously attracting poles will now repel each other, causing the coil to rotate in a new direction. This is the basic working principle of traditional brushed DC motors.

Brushed DC Motors are mainly composed of stator, rotor, commutator, and brushes. The stator is fixed and usually made of permanent magnets. The commutator changes the direction of the electric current at specific positions to ensure that the magnetic field produced by the rotor remains at a certain angle relative to the stator’s magnetic field, allowing continuous rotation under the action of magnetic forces.

I hope this helps! Let me know if you have any further questions or need any clarification.

The N and S poles in the diagram above are made of permanent magnets, while the rotor is composed of three sets of windings. The two ends of the brushes are connected to the power source and the commutator, respectively. Each set of windings has an insulating material between them.

A brushed DC motor can be easily speed-controlled by adjusting the voltage. However, it has some drawbacks: the brushes tend to wear out quickly, produce a lot of noise, and require regular replacement; in addition, there will be friction dust on the contact surface between the brushes and the commutator.

Brushless Motor Working Principle

A brushless DC motor uses a microcontroller to replace the brushes and uses Hall sensors or zero-crossing detection of induced electromotive force to detect the rotor position in real-time. The basic process is as follows: the microcontroller reads the state of the Hall sensor or induced electromotive force, and then controls the three-phase windings through a driver circuit to complete the commutation.

The following diagram shows the internal structure of a three-phase brushless DC motor:

Note that the windings are typically wound in multiple groups with symmetrical layout.

The following diagram shows a typical brushless DC motor driver circuit:

Since the brushless DC motor requires additional microcontrollers, Hall sensors, and zero-crossing detection circuits, its production cost is higher. However, it also offers more flexible programming control.

Difference in Structure between Brushless and Brushed Motors

In a brushed motor, the stator (rotor) is located at the center of the motor, while the rotor (stator) with windings is located on the outside:

In contrast, a brushless motor has the stator at the center and the rotor on the outside:

Note: The windings in both diagrams are wound in multiple groups with symmetrical layout.

Brushless Motor Commutation Principle

The commutation of a brushless motor relies on Hall sensors or zero-crossing detection of induced electromotive force. Hall sensors are magnetic field-sensitive components that change their output signal to high and low levels when passing through the north and south poles, respectively. Typically, three Hall sensors are placed at the commutator edges and arranged at 120° electrical angles:

When the rotor commutes, the Hall sensors output high and low level changes, allowing the detection of rotor commutation. Each Hall sensor changes its state once per magnetic field rotation period, resulting in 6 state changes for each Hall sensor.

In addition to Hall sensor-based commutation schemes, zero-crossing detection-based commutation control will be discussed in more detail in subsequent content.

Main Performance Indicators of Brushless Motors

The following table summarizes the main performance indicators of brushless DC motors:

| Performance Indicator | Parameter Explanation | | — — | — — |

Note: The table is not complete as it was truncated. If you would like me to continue translating, please let me know!

PWM Control Principle

When using PWM pulse wave drive to control a three-phase brushless motor, only two phases are energized at a time (one phase connected to the positive terminal of the power source and one phase connected to the negative terminal of the power source, while the other phase is floating). At this time, current flows from the positive terminal of the power source into one phase, then back out through the other phase to the negative terminal of the power source.

Disconnect the power supply from A phase, allowing it to float.
Enable the upper bridge arm power transistor for B phase, connecting it to the positive terminal of the power source (since simultaneous conduction of both transistors would cause a short circuit, one must be disconnected before enabling).
Disconnect the upper bridge arm power transistor for C phase and enable the lower bridge arm power transistor, allowing C phase to connect to the negative terminal of the power source.

In this state, current flows from the positive terminal of the power source through B phase into the motor, then out through C phase back to the negative terminal of the power source via the lower bridge arm power transistor.

The energized coils produce a magnetic field that interacts with the stator’s magnetic field to cause rotation. If the state of the Hall sensors is used to change the conduction sequence of the power transistors, the rotor will rotate to the next position. Each change in conduction sequence constitutes one commutation, and six commutations are required for each rotation period. If the Hall sensor output states are used according to the following table:

The table illustrates the state of rotation for a counterclockwise rotation. If the rotation is clockwise, the conduction sequence will be reversed.

Note: During rotation, the stator’s magnetic field always leads the rotor’s magnetic field by an appropriate angle. By adjusting this angle difference, the motor can produce higher torque.

Since each phase of the motor is spaced 120° apart (electrical angle), the output waveforms of the three Hall sensors will also be spaced 120° apart. Additionally, each Hall sensor’s output state changes every 180°, and since there are six sectors in total, the combination of output states from all three Hall sensors will change every 60° electrical angle. The following diagram illustrates this relationship between Hall sensor output states and electrical angle:

DC motors’ speed and voltage are directly proportional, meaning that by changing the PWM duty cycle of the power MOSFETs, you can control the speed (equivalent to changing the effective voltage on the motor’s phase lines). The PWM frequency is usually around 20KHz. There are several common ways to regulate the speed:

Half-Bridge: One phase’s lower bridge arm is always enabled, and only the upper bridge arm of another phase is controlled. For example, in the first state of the table above, set B phase positive and C phase negative, turn on power MOSFETs 1 and 2, and current flows from the positive terminal of the power source through FET 1 to motor phase B, then through phase C to FET 2, and finally back to the negative terminal of the power source. At this time, you can only control the upper bridge arm’s duty cycle while keeping the lower bridge arm always enabled.
Full-Bridge: One phase’s upper bridge arm is always enabled, and only the lower bridge arm of another phase is controlled. For example, keep MOSFETs 1 and 2 open, but only control MOSFET 3.
Complementary Regulation: Complementally regulate two-phase bridge arms’ power MOSFETs. For example, MOSFETs 1 and 4 are a complementary output pair, while MOSFETs 2 and 3 are another complementary output pair. The advantages of this method are that when the upper FET is turned off, current can immediately be taken over by the lower FET without going through the internal freewheeling diode of the power MOSFET, thus avoiding damage to the power MOSFET due to large currents. This method is relatively common in actual development work.

A Simple and Intuitive Summary

Brushless DC Motor (BLDC) belongs to synchronous motors. Below is a physical picture of a 1400KV three-phase brushless motor for RC models:

The stator of a brushless motor is usually composed of three star-connected coil windings, while the rotor uses a permanent magnet. By properly energizing the stator, a rotating magnetic field can be generated on the stator, making the rotor’s fixed magnetic poles follow this rotating magnetic field in an orderly manner:

Brushless motors are generally driven by a three-phase inverter circuit composed of six MOSFETs. This inverter circuit is usually made up of three half-bridge MOSFET circuits (each consisting of two MOSFETs and an output line taken out between them). By controlling the switching sequence of the upper and lower bridge arm MOSFETs, a rotating magnetic field can be generated on the stator:

Note: The upper and lower bridge arms cannot be turned on at the same time, otherwise a short circuit will occur. Therefore, dead-time control must be introduced to avoid simultaneous conduction of the same phase’s upper and lower bridge arms.

Further Discussion on Phase-Shifting Principles

As mentioned earlier, in actual motor control scenarios, it is necessary to obtain the current position of the rotor and calculate the next step for which bridge arm to turn on, so that the motor can start rotating. Obtaining the rotor position can be achieved using sensing and non-sensing methods.

Sensing Methods

Hall Sensor-Based Sensing Method

Brushless motors usually use three switch-type Hall sensors to detect the rotor’s position, each installed 120° apart as shown in the diagram: Here is the translated Markdown content in English:

When the N-pole of the rotor is close to the Hall sensor, a high level output is produced. When the N-pole is far from the Hall sensor, a low level output is produced. When the rotor rotates one circle, the waveform will be as follows:

Using sensitive Hall sensors increases the production cost of brushless motors and installation costs. If the sensor fails, it may cause the motor to malfunction.

Zero-Crossing Detection

The zero-crossing control strategy mainly includes back-EMF method, inductance method, and current-sensing diode method. The back-EMF method is the most widely used and mature. According to Lenz’s law (the direction of induced current always opposes the change of magnetic flux), the polarity of back-EMF is opposite to the main voltage on the winding. The formula for calculating back-EMF is as follows:

After the brushless motor production process, the rotor magnetic field and winding turns are fixed parameters. Only the angle speed (rotor speed) can determine the back-EMF value. When a phase changes direction, one winding is positive, another is negative, and the third is open-circuit. By detecting the zero-crossing points of each phase’s back-EMF, we can obtain the 6 positions of the rotor within an electrical period.

The following diagram shows the waveform of current and back-EMF for each phase during one rotation cycle:

Note: Each phase’s back-EMF exists in both positive-to-negative and negative-to-positive states. Therefore, a three-phase brushless motor has 6 zero-crossing states.

In practical development work, the zero-crossing detection based on back-EMF mainly includes ADC sampling, comparator detection, and current-sensing methods. The last one is a zero-Crossing FOC control scheme, while the first two are zero-Crossing PWM control schemes. The following sections will introduce these two schemes.

Note: When the rotor speed is very slow, the amplitude of back-EMF is very low, making it difficult to detect the zero-crossing point.

Zero-Crossing Detection using ADC

When the brushless motor rotates, the back-EMF at the zero-crossing point will be floating. By detecting each phase’s voltage relative to ground and comparing it with the DC bus voltage, we can determine whether a zero-crossing event has occurred. In other words, the zero-Crossing detection using ADC is achieved by simultaneously measuring the end voltage and DC bus voltage and then comparing them.

The following diagram shows an example of an ADC zero-Crossing detection circuit:

To simplify the calculation process, the end voltage and DC bus voltage often use the same division ratio. For example, in a 12V brushless motor control scheme, we can use a division ratio of to ensure that the DC bus voltage and end voltage are within the range that can be measured by the ADC on the motor control chip:

Zero-Crossing Detection using Comparator

When the brushless motor rotates, the back-EMF at the zero-crossing point will be floating. By detecting each phase’s voltage relative to ground and comparing it with the neutral point voltage, we can determine whether a zero-crossing event has occurred. Typically, the brushless motor does not have a neutral point, making it impossible to directly measure the neutral point voltage.

The following diagram shows an example of a comparator-based zero-Crossing detection circuit:

In a zero-crossing detection scheme based on comparators, three-phase windings can be connected to a common point using resistors with the same resistance value, thus building a neutral point. The neutral point voltage and terminal voltage are then compared by a comparator to obtain a zero-crossing signal.

The above is the principle diagram of a zero-crossing detection circuit based on comparators. By using resistors with the same resistance value R46, R47, and R48 to connect each phase, a virtual neutral point is constructed.

Defects of PWM control method

Although the PWM motor control algorithm is relatively simple and has a low hardware BOM cost, it also has several significant disadvantages:

Using PWM pulse width modulation to drive brushless motors, since the phase current only has on and off states, even when the duty cycle and average current are small, the peak current in the coil will still be relatively large. Since heat loss is proportional to the square of the current, the power consumption and heat dissipation of PWM-controlled motors will also be relatively high.
Due to the discontinuous electromagnetic field caused by PWM control, the motor’s torque will exhibit oscillations. Therefore, PWM is not suitable for applications where high accuracy is required, as it will lead to a decrease in control precision. In addition, the oscillation of torque will cause system vibration and noise, which will interfere with the operation of other chips and sensors on the PCB.

To solve these problems, German engineer F.Blaschke proposed the vector control method (Vector Control) in the 1970s. This method involves representing three-phase output current and voltage as vectors during processing. Its essence is to transform complex three-phase AC signals into direct-current controllable two-phase orthogonal currents through a series of coordinate transformations, thereby decoupling complex current relationships and making AC motors simple and controllable. This control method is also known as field-oriented control (FOC, Field Oriented Control). I will write a separate article on this topic later.

Note: During the writing process of this article, the author also found an English text with illustrations 《How Brushless DC Motor Works?》, which can be read together with this article to facilitate a quick understanding and mastery of the PWM control principle of three-phase brushless motors.