How to drive BLDC motors effectively with MOSFET tech

It has been estimated that electric motors consume around 70% of all electrical power used in industrial applications. The need to reduce energy consumption is, however, leading many companies across the automotive and industrial sectors to adopt more efficient solutions, the most popular being brushless DC BLDC motors.

Besides being more energy efficient, BLDC motors are more reliable as there are no brushes to wear, which also eliminates electrical noise as there is no arcing. Overall, these motors are smaller and lighter, with a better power-to-weight ratio and increased torque. As automation increases, they are playing a key role in improving functionality and reliability while reducing cost.

So far so good. However, driving a BLDC motor is somewhat more complex than for a brushed motor and involves a combination of devices including MOSFETs, IGBTs, gate drivers, power management and magnetic Hall effect sensors. In BLDC motors, the rotor is a permanent magnet and the stator applies a rotating electromagnetic field to cause the rotor to spin.

The combination of the rotor position and the timing of the current in the stator is critical for control, which makes a BLDC motor somewhat more complex than a brushed motor. Often a microcontroller (MCU) is used to handle the necessary algorithm and to interface to the rest of the system.

The MCU uses the Hall sensors to determine the rotor’s position and energise the stator coils at the correct time. Systems that do not need position sensing can dispense with the Hall sensors and use Field Oriented Control (FOC).

In BLDC drive schemes, either MOSFETs or IGBTs are used to switch the drive current to the stator coils. In all cases the position of the rotor must be known for timing - if Hall effect sensors are used then the magnetic field is detected and in sensorless FOC systems, back EMF is detected.

Normally, the switching elements (MOSFETs/IGBTs) will be configured in a half-bridge arrangement, although the devices themselves will have the largest impact on the system efficiency and performance. To make an optimum selection, designers have to understand both the application and the relevant device parameters.

Important factors

The principal areas for consideration are efficiency, reliability and parameters that may influence the design. To ensure reliability, the designer should be aware of the device’s limits and ensure that these are selected correctly for the application. For example, transient voltages may cause a breakdown during operation. BLDC operating voltages are commonly 12, 24 or 48V so a MOSFET with a breakdown voltage of 40, 60 or 100V would be selected to give protection against transients.

Current ratings are another important parameter and should be understood for continuous operation as well as pulses and fault conditions. With a BLDC there can be a surge of current when starting and also if the motor should stall - these can be three times the full load current and a MOSFET that can endure these levels must be selected.

The higher the power level, the higher the on-resistance (RDS(on) of the MOSFET will be. This can impact efficiency through losses and reliability through elevated operating temperatures. Generally speaking, selecting a lower RDS(on) is the best choice from a performance perspective.

However, MOSFETs with low RDS(on) are more expensive so some consideration has to be given to system cost by selecting the optimum (and not the lowest) RDS(on)).

The optimum RDS(on) can be determined by ensuring that the losses do not create enough thermal energy to raise the MOSFET above its maximum operating temperature (typically +150 or +175°C). If the temperature is likely to exceed these limits then thermal management such as a fan or heatsink will be required, which would increase cost, size and weight.

Ideally, a dissipation of less than 1.5W in each half bridge is recommended so, for a 400W (0.5HP) motor powered by 12VDC draws over 30A, a MOSFET with an RDS(on) below 2mΩ (such as Diodes Incorporated’s 40V DMTH41M- 8SPS) would be a good choice. At 24V, the same motor draws around 16A, so a MOSFET with an RDS(on) below 8mΩ (such as the 60V DMTH6004SPS) would be the best choice.

While RDS(on) defines the static losses, the dynamic losses (those due to switching) are governed by the gate charge, QG, which has to be replenished each time the device switches. Devices with low RDS(on) tend to have higher QG and vice-versa, so this is always a trade-off to obtain the optimum solution.

The most common drive scheme is for the MCU to generate a pulse-width-modulated (PWM) signal for each of the three motor phases/coils. However, if the timing is not correct, then both MOSFETs in a phase can be turned on, creating a direct connection between the supply and ground.

This condition is known as ‘shoot through’ and almost always results in catastrophic MOSFET failure. The usual way of avoiding this is to ensure the PWM signals cannot overlap by introducing ‘dead time’ so that only one MOSFET can conduct at a time. When determining the length of the dead time, designers must take into account the switching time of the selected MOSFET and QG as they will both have an effect.

Other considerations

Also to be considered is the level of gate-source voltage (VGS) necessary to fully turn the device on – and how this relates to the zero temperature coefficient point (ZTC). Logic-drive MOSFETs require a VGS of 5V while standard MOSFETS usually require 10V. Providing insufficient VGS drive can mean the MOSFET is only partially turned on, which can lead to a much higher than expected RDS(on) and therefore higher losses and a greater temperature rise. Under extreme circumstances, if VGS falls below the ZTC then a hot spot can be formed, leading to thermal runaway and catastrophic failure – this can normally be checked from manufacturer’s data sheets.

Normally N-channel MOSFETs are used in motor drive applications as, for a given size, the N-channel device will have an RDS(on) value around half that of a comparable P-channel device. Full bridge circuits can be susceptible to reverse current flow due to the MOSFET’s body diode, although minimising dead time and selecting MOSFETs with a low Vf fast recovery parallel diode can both contribute to reducing this issue.

As the trend towards automation continues, demand for electric motors is growing and BLDC motors are proving increasingly popular in industrial and automotive applications. Compared with brushed motors, BLDC motors are more efficient and more reliable as well as producing more torque and weighing less – all desirable attributes. Driving BLDC motors can seem complex although the key to a successful solution lies in specifying the optimum MOSFET and performing good thermal design.

Through a thorough comprehension of the pertinent parameters and the application, designers are able to select the right components and deliver reliable and efficient solutions – even in harsh environments.