Not all electric motors are the same - Different versions and how they are controlled

02/07/2019 Know-How

Electro mobility is one of the hottest topics in the automotive industry right now due to the diesel emissions scandal. However, the idea of using clean electricity from renewable sources instead of internal combustion engines is not new: Streetcars, for example, were electrified at the end of the 19th century. In contrast, electro mobility is just beginning its triumphant march on our highways and streets: Since July 2016, the German government has been subsidizing electric cars; with the target of bringing one million electric cars onto German roads by 2020. But not all electric motors are the same, and electric automobiles therefore differ greatly. There are many different drive concepts and microcontrollers ideally suited for each of these concepts. An overview:

Direct current motor

Of all electric motors, the brush-commutated DC motor is the oldest and best-known one. In principle, this motor consists of a stator which generates a homogeneous magnetic field by means of a magnet. The rotor consists of a coil of wire through which a current flows, thus creating a dipole field like that of a bar magnet. Due to the differing polarity, the rotor starts to rotate until the different poles are opposite each other and the movement comes to a standstill. In order for the rotor to rotate permanently, the commutator reverses the current direction every 180 degrees in the rotor coil. As a result, the polarity of the rotor magnet changes, meaning the magnetic polarities are now the same. The polarities therefore repel each other and the rotor continues to revolve. Steady turning of the rotor is thus achieved.

The minimal motor described here has just one pair of poles in the rotor, whereas the motors used in real-life scenarios have several pairs of poles. This distributes the stator field around the entire motor, ensuring smooth running of the rotor. This also has the advantage that the magnetic flux is higher, which leads to a higher torque that is also more "ripple-free" than the basic design with just one pair of poles.

Controlling the motor

Since the actual motor commutates entirely mechanically, it is very easy to control via the applied voltage, as the torque of the motor depends solely on the magnetic flux and this in turn depends on the current flowing through the rotor. This is best achieved through a PWM signal, which is output by a microcontroller that actuates a power transistor, for example. Ideally, this is done via an H-bridge consisting of four transistors, as this also allows the direction of rotation to be changed.

As stated at the beginning of this article, the stator field is generated by means of a magnet. Of course, it is also possible that this field is generated by coils that are supplied with power from a separate circuit by means of slip rings. This can, however, lead to higher flux densities and a higher torque.

Synchronous motor

The three-phase synchronous motor consists of three pairs of coils offset by 120 degrees in the stator and a permanently magnetized rotor. This can be achieved using permanent magnets or through electromagnetic independent excitation by means of a field coil on the rotor with current supplied via slip rings. After switching on the motor, the field in the stator immediately starts to rotate in relation to the applied three-phase voltage. However, since the rotor cannot instantaneously follow the rotational field due to its mechanical inertia, a starting aid is required that is implemented as a squirrel cage in the rotor. The motor therefore behaves as an asynchronous machine until rotating at synchronous speed. The excitation field in the rotor is now switched on and the motor runs synchronously with the stator rotational field.

Normally there is no slip (i.e. the difference between the speed of the stator field and the speed of the rotor) with a synchronous motor. However, the rotor magnetic field follows the stator field under load. If the torque angle reaches 90 degrees, the rotor will pull out of synchronism and stop. The lack of brushes in this type of motor is a major advantage compared to a DC motor. This makes the motor much more resistant to wear, and EMC problems do not occur due to the lack of "brush sparking".

Controlling the motor

It is necessary to know the position of the rotor in order to adjust the motor speed when the load changes and in order to control a setpoint change in the speed. An easy way to do this is to attach Hall sensors to the motor coils which transfer the position to the controlling microcontroller. Recording the position without sensors is much more complex as only the currents in the individual phases and the back EMF, which are measured with fast A/D converters, are available as sensor elements. These are transferred to a mathematical model of the motor from which a microcontroller generates the PWM control signals for the six-fold motor bridge.

Since the motor expects sinusoidal control, but the motor bridge can only output pulse-width modulated signals, the sinus is simulated using a corresponding sequence of PWM pulses. A good value is 100 PWM pulses per cycle, as the low-pass effect of the motor coils smooths the pulses.

In order for the motor to run synchronously, the microcontroller must generate three PWM sequences offset by 120 degrees that simulate a sine wave, which requires high computing power of the microcontroller and a sophisticated timer unit for load and setpoint speed changes.

Asynchronous motor

The asynchronous motor has a very simple mechanical design: As with the synchronous motor, the stator contains three pairs of coils offset by 120 degrees. However, the rotor is a so-called squirrel-cage rotor, i.e. it consists of short-circuited conductors excited by the magnetic field supplied by the stator coils. As long as the rotor does not move synchronously with the stator field, voltage is not induced in the cage and there is no torque present.

When the rotor revolves slower than the stator field, the flux changes, inducing a voltage that leads to a current in the squirrel-cage rotor, which in turn produces a torque proportional to this so-called slip. Due to their simple design, these motors are widely used in drive technology.

If the speed of the motor is to be controlled digitally using a frequency converter, the principle is the same as with the synchronous motor, where the rotor position is determined via the phase currents and the back EMF. The control sequences for the motor output stage are then generated using a mathematical model that is calculated in the microcontroller.

BLDC motor

Although the motor is called a "brushless DC motor", it is not a DC motor but a three-phase synchronous machine, as it also contains three pairs of coils in the stator and the rotor is constructed with permanent magnets. As with the synchronous and asynchronous motors, the motor output stage is achieved through six transistors which are controlled by a microcontroller that generates the sequence.

This so-called commutation sequence consists of six blocks per revolution. Only two bridges are active at any one time, while the two transistors are open at the third bridge. For a better understanding, the individual commutation patterns are graphically illustrated.

If there are Hall sensors in the stator coils, as shown above, commutation is relatively simple, since the position of the rotor is known at all times. If no Hall sensors are available, the back EMF generated in the stator coils is normally evaluated. However, this does not work at very low speeds, as the induced voltage is too low and the pulse too wide to clearly identify the rotor position.

For some time now, methods have been known that are based on sending short test pulses that are too weak to move the rotor into the stator coils. Since the magnetic field of the rotor influences the inductance of the stator coils, the induced voltage (back EMF) can be traced back to the position of the rotor.

Implementation of the drive:

The following components are now required to drive the motor: A powerful microcontroller that, in conjunction with a three-phase PWM timer unit, generates the appropriate pulse patterns for the output stage. This output stage in turn consists of six power transistors and the corresponding driver modules. Currently, most suppliers offer corresponding hardware for digital control. Examples of modules from Renesas, STMicro, and Infineon are presented here:

Renesas' Synergy microcontroller family provides innovative modules that offer all the necessary features. The derivatives S7 and S5, two controllers based on CORTEX M4, which run at 120MHz and 240MHz respectively, deserve a special mention here. The modules have two 12bit A/D converters running at 2.5MHz for measuring phase currents and back EMF. The timer unit can trigger the A/D converters directly via the so-called event link controller, so that no interrupt routine has to be processed. The event link controller is a state machine that relieves the CPU of peripheral interactions. In addition to the triggering of the A/D converter mentioned above, the timers and the DMA can also be activated, and interrupts can be additionally sent to the CPU.

By far the most important function for digital motor control is a powerful timer unit. With the GPT32E, the S7 and S5 offer a high-resolution module consisting of four independent timers with a resolution of at least 8.3ns and a three-phase PWM generator for controlling the bridge. As already mentioned, the A/D converter can be triggered directly via the ELC unit. In addition, a simplified 32bit or 16bit timer is available if additional motors are to be controlled with the same controller. Synergy from Renesas is a concept that also provides comprehensive software for all the most important motors, thereby enabling rapid implementation of the drive.

The timer modules of the STM32 family from STMicro are similar in terms of the functionality. The latest derivate, the STM32H7, contains a so-called high resolution timer unit (HRTIM1), which runs at a clock rate of 400MHz and reaches a very high resolution of PWM. The module consists of six timers, one master, and five slaves, enabling ten high-resolution outputs for bridge control. This module is particularly interesting for very fast rotating BLDC motors, such as drives running at about 30,000rpm and outputting more than 100 PWM pulses per cycle. The TIM1 and TIM8 advanced control timers are certainly sufficient for standard drives. They provide a PWM unit featuring six outputs for the motor bridge and the dead time generator for the half bridges. Like Renesas, STM also offers software modules for the various motors.

Infineon's XMC microcontrollers, which are optimized for drives, also feature the CCU8 module, a very flexible unit consisting of four submodules. For instance, the meters can be operated "center aligned" or "edge aligned" and output individual pulses, also known as "single shots". It is therefore relatively easy to generate the PWM signal, including dead time for half bridges.

In short: Depending on the drive concept, different suppliers offer different solutions; the choice of microcontroller therefore depends very specifically on the actual application. This is where qualified field application engineers can give customers advice on the most suitable microcontroller.

Find components at <link _blank external-link-new-window "open internal link"></link>.

Subscribe to our newsletter and stay updated: <link _blank external-link-new-window "open internal link"></link>.