How to Choose the Right MOSFET for Motor Drive

    Automotive OEMs are migrating to BLDC to maximize efficiency and reliability. This article looks at the important parameters engineers should consider during the design process to achieve these goals.

The discovery of electromagnetic induction (EMI) changed the world and heralded a new era. Today, it touches every sector, market and industry. In many ways, the ability to generate electricity at will and transform it into motion with precise control and regularity is the hallmark of a developed society.

    Generators and electric motors are by far the most common and widely deployed implementations of EMI. With the exception of solar, most of the available electricity is generated this way, either as large turbines in power stations or as smaller generators in renewable energy solutions such as wind or wave.

    In response to this abundance of energy, the generator’s counterpart, the electric motor, has successfully and inextricably replaced purely mechanical forms of power. The internal combustion engine is perhaps the latest step in this journey, as electric vehicles begin to become commonplace on our roads.

    However, there is an intermediate step in the automotive industry's transition to electricity, which is to replace mechanical equipment with electric motors.

    Add Application

    From a consumer perspective, the most obvious use of electric motors in cars is probably in driving power windows and seats. Central locking is another application that could be cited. Under the hood, more changes are taking place. Electric motors are increasingly being specified as purely mechanical options for functions such as fans, pumps (water, oil, fuel), power steering and anti-lock brakes, and automatic transmissions.

    The reasons are clear; electric motors offer better control, higher efficiency and greater reliability than mechanical alternatives. When the transition first began, OEMs turned to stepper and brush-commutated motors, but more recently the automotive industry – like many others – has moved to brushless DC motors (BLDC), and for good reason.

    BLDC offers higher levels of efficiency, better control over a wider dynamic range, and greater torque. Because the technology is brushless—effectively contactless from an electrical perspective—it eliminates all electrical interference common to brushed DC motors. This helps reduce electromagnetic interference, which can cause problems for more sensitive components in the engine control unit (ECU). It also avoids the arcing and subsequent wear common to brushed commutation, which can lead to performance degradation and eventual failure of brushed DC motors.

    Of course, replacing a mechanical motor with an electrical alternative does require additional control electronics. In the case of BLDCs, this is arguably exacerbated by the lack of electrical contacts. BLDCs are sometimes controlled by using a Hall effect switch, which provides the necessary feedback for the control loop. More recently, however, sensorless BLDCs have become popular, as removing the sensor further reduces the bill of materials.

    The control algorithms developed to drive BLDCs (sensored and sensorless) are handled by a microcontroller (MCU), which offers the added benefit of providing relatively simple integration into vehicle networks using CAN or LIN. MCUs designed for motor drive in automotive applications are also equipped with a pre-driver stage to control the MOSFETs required to deliver high drive currents through the motor coils. This final stage is critical in defining the efficiency of the overall motor drive solution, as described below.

    Improved drive

    The driver circuit of a BLDC usually includes MOSFETs to generate and destroy the electromagnetic field generated by the stator coils, which rotate around the rotor formed by permanent magnets. Detecting the position of the stator is essential to generate the correct excitation field in the coils. In sensored BLDCs, it is the magnetic field that is detected, while in sensorless versions, the control circuit measures the back EMF to determine the stator position.

    In either case, the coil is powered through MOSFETs arranged in a bridge topology. The choice of MOSFET is a major factor affecting the overall efficiency and performance of the BLDC. The data provided in the datasheet is used under specific conditions and may or may not be consistent with the operating conditions of the actual application. Therefore, it is important to understand the application before selecting the most appropriate MOSFET.

    Likewise, the operating parameters of the selected MOSFET will have a direct and significant impact on the overall solution. Careful consideration of these parameters will ensure that the MOSFET selected best meets the requirements.

    In general, there are three main areas that should be considered: reliability, efficiency, and design. Reliability is concerned with the limits of the device and ensuring that these limits are never tested during normal operation. Specifically, this involves selecting a device with a breakdown voltage that provides adequate protection against transients that may be introduced through other design choices. For example, for a BLDC running from a 12V supply, a breakdown voltage of 40V is sufficient. Similarly, in a 24V system, a MOSFET with a breakdown voltage of 60V will provide adequate protection. It is also important to consider the drain-source current rating, especially under surge or pulse conditions. In BLDC applications, the startup or stall current may exceed three times the full load current,

    In the case of MOSFETs, efficiency generally refers to the device's ability to manage heat dissipation, especially at the junction. Good thermal design is always necessary, especially in environments with high ambient temperatures such as automotive, but there are several parameters that should be considered when selecting a MOSFET. These include on-resistance, Rds(on), and gate charge (Qg). These two parameters are interrelated; a larger MOSFET can produce a lower on-resistance, but will also result in a higher gate charge. This can have a significant impact on switching applications such as BLDC drives.

    Temperature coefficient

    Driving a BLDC with three phases (coils) is typically accomplished with MCU-generated PWM (Pulse Width Modulation) signals to power each phase. Figure 1 shows a typical bridge circuit for a BLDC phase. If both MOSFETs are turned on at the same time, shoot-through will result, which can have catastrophic effects. To address this, a period, called the dead time, is designed into the PWM signal to ensure that only the intended MOSFET is on at any given time. The switching time of the MOSFET will affect the required dead time length, a parameter that is also affected by the gate charge of the device. During the dead time, the body diode of the MOSFET provides a commutation path, which is again not ideal due to the high power losses when the diode is on.

    Every MOSFET exhibits a dynamic capacitance (Crss in Figure 1); this is a parameter that can cause breakdown. This parameter, combined with Rg, can cause the gate charge of the low-side MOSFET to rise to a level sufficient to turn it on during switching.

    Figure 1. Typical bridge circuit for driving BLDC motor phases.

    Another important parameter that should be considered for switching applications such as BLDC drives is the zero temperature coefficient (ZTC) point. This is a point on the transfer curve (drain current, [ID], versus gate-source voltage, [VGS]) as shown in Figure 2. Operating the device below this point results in a positive temperature coefficient for the drain current, while operating the device above this point results in a negative temperature coefficient for the drain current. Figure 2a shows the transfer characteristics of a low-density planar MOSFET (ZXM61N03F) and Figure 2b shows the transfer characteristics of a high-density planar MOSFET (ZXMN3A01E6). Generally, it is recommended to operate the device in the negative temperature coefficient region. The device in Figure 2b utilizes a greater trench density to increase the number of vertical current flow paths in the channel. This has the positive effect of reducing Rds(on), although it also results in a higher ZTC point.

    Figure 2a (left). Low-density planar MOSFET ZXM61N03F

    Figure 2b (right). High-density trench MOSFET ZXMN3A01E6

    For a given size, N-channel MOSFETs typically have half the Rds(on) of an equivalent P-channel device, so N-channel MOSFETs are often specified in motor drive applications. Figure 3 shows the five stages of a full-bridge motor drive circuit using N-channel MOSFETs. It is also important to note that such circuits are subject to reverse recovery currents due to the body diode of the MOSFET. A PWM algorithm that minimizes dead time can reduce these effects, and it is also recommended to specify MOSFETs with fast recovery parallel diodes.

    Figure 3. Circuit showing commutation sequence and body diode recovery related breakdown.

    in conclusion

    Automotive OEMs are increasingly specifying brushless DC motors. They offer higher efficiency, improved reliability, and control for more functions, including replacing mechanical pumps and fans.

    Driving a BLDC requires combining an advanced MCU for control with properly specified MOSFETs to provide power. Thermal management is at the heart of good design, and this extends to understanding how to best meet the unique requirements of a BLDC drive circuit using the correct MOSFET design.

    By understanding and evaluating the relevant parameters, engineers can select the right MOSFET for the task, ensuring maximum reliability and efficiency even in the harshest environments.