Brushless-DC motor systems for the uninitiated

After only a few years of being completely immersed in the culture of motor drivers, it is very easy for me to take for granted the fact that most people are not familiar with electromechanical spinning. This has gotten me into multiple conversations in which I should have proverbially thrown out the shovel rather than digging myself into a deeper hole.

But I feel that I have an obligation to try to educate others on this topic, because I rarely see motors covered in college courses outside of abstract control theory. My goal here is to attempt to explain brushless-DC motors simply.

In all motors, there is a concept called “commutation” that describes the process of switching current flow (in some way) in order to move the physical rotor shaft. Movement occurs when current flowing through a coil generates a magnetic field that opposes or attracts some existing field, typically generated by a permanent magnet. This force causes movement in the rotor (the rotating part of the motor) with respect to the stator (the stationary part of the motor).

Magnets are a good analogy for commutation. When you place two magnets on a table with the same poles facing, they will push each other away. They will stop moving once they get far enough away from each other. If you then take one magnet and move it toward the second magnet, the second magnet will be pushed away again. If you continue doing this, the magnet will continue to move; this is a linear example of commutation.

Brushed-DC motors implement mechanical commutation, meaning that the physical structure of the motor actually causes the motor to commutate. Brushes make contact with a commutator (aptly named) and as the motor spins the current through, the motor coil will alternate polarity. This allows the field generated by the permanent magnet on the stator to always be in opposition to the magnetic field on the rotor, and therefore to always generate force. Mechanical commutation means that a brushed-DC motor will spin simply by applying some voltage across the motor windings, as shown in Figure 1.

Figure 1

Brushed-DC motor construction

Brushed-DC motor construction

At this point many readers are probably looking at the title of this article in confusion, since I am not yet talking about brushless-DC motors. However in order to explain “brushless,” I needed to first explain what the brush is being used for in the first place.

The genesis of a brushless motor is relatively simple from the outset: most of the issues with brushed-DC motors arise from the brushes. The brushes may spark, wear out, cause significant noise, contribute a significant amount of power loss, result in a significant speed limitation and can’t be easily cooled. This means that you shouldn’t use brushed-DC motors around anything flammable, on anything that needs a long lifetime, on any application that needs to be quiet or requires high efficiency, on any high-speed system, or on any high-power system. Those are significant limitations! You can solve these problems by doing away with the brushes, but that unfortunately eliminates the mechanical commutation.

A lack of mechanical commutation causes additional problems. The motor will still need to commutate; however, you (the designer) are now responsible for it. A brushless motor uses electrical commutation, which is a fancy way of saying “you need to ensure that the current in the motor is always generating a magnetic field that will move the rotor.” But looking at Figure 2, how do you know where the rotor is so that you can know how to apply current to move it?

Figure 2

Brushless-DC motor construction

Brushless-DC motor construction

The first major architecture decision for brushless-DC motor systems is the distinction between “sensored” and “sensorless.” (As I write this article, I am constantly reminded about the great travesty that both of those words are constantly flagged for spellchecking as “censored” and “senseless.”) You need to know where the rotor is, and you have two primary methods to figure it out:

  • Sensored approaches typically use Hall-effect sensors (or Hall elements) or encoders to detect the rotor position. While encoders can give very accurate angle feedback, they can be expensive. Hall-effect sensors are a popular type of magnetic sensor. In three-phase brushless-DC motors, implementing three Hall-effect sensors provides a simple six steps for commutation.
  • Sensorless methods of position estimation typically involve measuring or estimating the back-electromotive force (EMF) generated by the motor when it is spinning. Back-EMF is a complicated topic best suited for a different time, but simply put, it is a voltage generated on the motor coil that is a function of the motor speed as well as the motor load. Sensorless methods by nature are estimation, and often require complicated calculations. Sensorless approaches become exponentially more difficult as the motor speed decreases (such as for a position-controlled servo), since back-EMF decreases with speed.

The second major architecture decision for brushless-DC motor systems is the control method. If you know – or think that you know – where the rotor is, you need to apply a certain current to move the motor. A three-phase brushless-DC motor requires a minimum of six different states. If you don’t believe me, take a look at Figure 3. This is as simple as you can get to commutate a brushless-DC motor using a control method called “trapezoidal,” “six-step” or “120 degrees.”

Figure 3

Sensored trapezoidal commutation

Sensored trapezoidal commutation

Alternately, you can apply a smoother current waveform to the motor; this is called “sinusoidal” control or “180 degrees.” When used with the right motor, this control method can result in higher efficiency and lower audible noise versus trapezoidal control. The downside is the increased complexity in order to implement a smooth voltage and current profile, usually requiring highly accurate pulse-width modulation (PWM) timers.

Figure 4

Sinusoidal commutation

Sinusoidal commutation

In writing this brief article, I quickly realized how broad this subject really is. There are details like field-oriented control, motor startup, inner rotor versus outer rotor, pole count, delta versus wye winding, and many others that I don’t have time to cover here. However, I do hope that I have been able to make some of you a little bit more dangerous in your knowledge of motors.

2 comments on “Brushless-DC motor systems for the uninitiated

  1. j.sinnett
    March 13, 2019

    Another form of “sensorless”, brushless motor is the Stepper Motor.  These are extremely useful for certain specific applications that don't require high speed.  Typically a stepper motor is used in situations where the motor is required to rotate for a specific angle and then stop, and where they can be called upon to run in either direction.  They are used, for example, in desktop 3D printers, where the precise location of the actuators is controlled “open loop” by keeping track of the steps that have been issued to the motors.  This presumes, of course, that the mechanical load on the motor never exceeds the driving force that the motor can create.

  2. mshein11
    March 13, 2019

    Hello Mr. Sinnett,

    After having spent several years dealing with stepper motors, I have decided that they are the Pandora's box of motors. If you want a simple, low-power motor for position control, a stepper motor is easily the best choice. 

    However, the second you dive into the details on how a stepper motor functions physically, and attempt to deal with conceptually simple topics like stall detection, you open a can of worms. A brushed-DC or brushless-DC motor has a stall current, a stepper motor does not.

    Whenever I describe the different motor types, I always seem to say (after describing brushed and brushless) “and this is a stepper motor, which is just weird…”



Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.