Microchip senior marketing Engineer-II

Nelson Alexander

The advanced motor control system based on permanent magnet synchronous motor (PMSM) with sensorless field oriented control (FOC) is rapidly popularized. There are two main driving factors behind this phenomenon: improving energy efficiency and strengthening product differentiation. Although there is evidence that PMSM with sensorless FOC can achieve these two goals, it needs a design ecosystem that can provide an overall implementation method to succeed. Using the overall ecosystem, designers can overcome various challenges that hinder the adoption of the system in the implementation process.

Why PMSM?

PMSM motor is a brushless motor using electronic commutation. It is often confused with brushless DC motor (BLDC), which is another member of the brushless motor series and also uses electronic commutation, but it is slightly different in structure. The structure of PMSM can be optimized for FOC, while BLDC motor can use 6-step commutation technology after optimization. After optimization, PMSM can obtain sine wave back EMF, while BLDC motor can obtain trapezoidal wave back EMF.

These motors use different rotor position sensors. PMSM usually operates with a position encoder, while BLDC motor operates with three Hall sensors. If considering the cost, designers can consider implementing sensorless technology to save the cost of magnets, sensors, connectors and wiring. Removing sensors also helps to improve reliability because it reduces the number of components in the system that may fail. When comparing sensorless PMSM with sensorless BLDC, sensorless PMSM using FOC algorithm can provide better performance, and the implementation cost using similar hardware design is equivalent.

The biggest beneficiaries of switching to PMSM are those applications that are currently using brushless DC (BDC) or AC induction motor (ACIM). The main benefits of switching include lower power consumption, higher speed, smoother torque, lower audible noise, longer service life and smaller size, so as to make the application more competitive. However, in order to realize these benefits of using PMSM, developers need to implement more complex FOC control technology and other application specific algorithms to meet the system requirements. Although PMSM is more expensive than BDC or ACIM, it has more advantages.

Challenges in implementation

pYYBAGG2wGuAVsL9AACmLHLIQdg843. png

Figure 1: three phase sensorless PMSM control system using three-phase voltage source inverter

However, to realize the advantages of using PMSM, we need to understand the inherent hardware complexity of realizing advanced FOC motor control technology, and master the professional knowledge in this field. Figure 1 shows a three-phase sensorless PMSM control system using a three-phase voltage source inverter. Controlling the inverter requires three pairs of interrelated high-resolution PWM signals and a large number of analog feedback signals that need signal conditioning. The system also needs hardware protection function to realize fault tolerance, and uses high-speed analog comparator to realize fast response. These additional analog elements required to realize sensing, control and protection increase the cost of the solution, which are not required for typical BDC motor design or simple ACIM voltage per Hertz (V / F) control.

In addition, there is the development time required to define component specifications and verify for PMSM motor control applications. To meet these challenges, designers can choose a suitable single chip microcomputer to achieve high analog integration with device specifications tailored for PMSM motor control. This will reduce the number of external components required and optimize the bill of materials (BOM). The highly integrated motor controller now has high-resolution PWM, which can simplify the advanced control algorithm, high-speed analog peripherals for precision measurement and signal conditioning, hardware peripherals required for functional safety, and the implementation of serial interface for communication and debugging.

pYYBAGG2wHyAZZRCAAETaYvPLyc953. png

Figure 2: block diagram of standard sensorless foc

In addition, there is a big challenge, that is, the interaction between motor control software and electromechanical behavior of motor. Figure 2 shows a standard sensorless FOC block diagram. To transform it from concept to practical design, we need to understand the controller architecture and digital signal processor (DSP) instructions to realize the time critical control loop with intensive mathematical calculation.

In order to achieve reliable performance, the control loop must be executed in one PWM cycle. The time of the control loop must be optimized for the following three reasons:

1) Limit: use PWM switching frequency not less than 20 kHz (duration 50 μ s) To suppress the noise from the inverter switch.

2) In order to realize the control system with higher bandwidth, the control loop must be executed in one PWM cycle.

3) In order to support other background tasks (such as system monitoring, application specific functions and communication), the control ring needs to run at a faster speed. Therefore, the goal of FOC algorithm should be in 10 μ Execute within s.

Many manufacturers provide examples of FOC software that uses sensorless estimators to estimate rotor position. However, before the motor starts to rotate, the FOC algorithm must configure various parameters to match the motor and hardware. The control parameters and coefficients must be further optimized to meet the required speed and efficiency objectives. This can be achieved by combining the following methods: 1) obtain parameters using the motor data book; 2) Repeat the test. The motor data book can not always accurately characterize the motor parameters, or the designer cannot obtain high-precision measurement equipment. In this case, the developer will have to use the method of repeated test. This manual adjustment process requires time and experience.

PMSM motors are used in many different applications, run in different environments, or have different design limitations. For example, in an automobile radiator fan, when the motor is about to start, the fan blades may rotate freely in the opposite direction due to the action of the wind. In this case, starting PMSM motor using sensorless algorithm is a challenge and may damage the inverter. One solution is to detect the rotation direction and rotor position, and use this information to decelerate the motor to a stationary state through active braking before starting the motor. Similarly, it may be necessary to implement additional algorithms, such as maximum torque per ampere (MTPA), torque compensation and field weakening [1]. These types of application specific additional algorithms are essential for developing practical solutions, but they also prolong development time and complicate software verification, thereby increasing design complexity.

pYYBAGG2wIyAC2eHAAGWrXwK1Xk301. png

Figure 3: application framework of FOC

One solution to reduce complexity is for designers to create a modular software architecture that adds application specific algorithms to FOC algorithms without affecting time critical execution. Figure 3 shows the software architecture of a typical real-time motor control application. The core of this framework is the FOC function, which provides hard timing constraints and many application specific additional functions. The state machine in the framework connects these control functions with the main application. This architecture requires a well-defined interface between software function blocks to make it modular and simplify code maintenance. The modular framework supports the integration of different application specific algorithms with other system monitoring, protection and functional security programs.

Another advantage of the modular architecture is to separate the peripheral interface layer (or hardware abstraction layer) from the motor control software, which is convenient for designers to seamlessly migrate their IP from one motor controller to another when the application function and performance requirements change.

Demand for a complete ecosystem

Meeting these challenges requires a motor control ecosystem tailored for sensorless FOC. The motor controller, hardware, software and development environment shall work together to simplify the process of implementing advanced motor control algorithms. To achieve this goal, the ecosystem should have the following characteristics:

1. An advanced tool for automatically executing motor parameter measurement, designing control loop and generating source code, which enables designers without domain expertise to realize FOC motor control, and write and debug very time-consuming complex time-critical code

2. Application framework applicable to FOC and different application specific additional algorithms to shorten development and test time

3. Motor controller with deterministic response and integrated analog peripherals that can realize signal conditioning and system protection in a single chip are used to reduce the total cost of the solution

poYBAGG2wJiAJP-cAAJZ7q8XZLc956. png

Figure 4: Microchip motor control ecosystem architecture

Figure 4 shows an example of a motor control ecosystem architecture, including an application framework and a development kit for a high-performance dsPIC33 motor control digital signal controller (DSC). This development kit is built on the basis of FOC software development tool based on GUI, which can measure key motor parameters and automatically adjust feedback control gain. In addition, it can also generate the required source code for projects created in the development environment using the motor control application framework (MCAF). The core of the solution protocol stack is the motor control library, which can realize the time critical control loop function of the application and interact with the motor control peripherals of dsPIC33 DSC. This GUI can be used with several available motor control development boards to support motor parameter extraction and generate FOC codes for various low-voltage and high-voltage motors.

The demand for high energy efficiency and product differentiation has promoted the transformation to brushless motor. A comprehensive motor control ecosystem can provide an overall method to simplify the implementation of sensorless FOC based on PMSM. This method should include special motor controller, rapid prototyping board and easy-to-use FOC development software that can automatically generate code.

reference material

[1] tb3220 – use angle tracking PLL estimator to realize Sensorless Field Oriented Control of permanent magnet synchronous motor (surface mount and built-in) for household appliances:http://www.microchip.com.cn/newcommunity//Uploads/202003/5e65d169337d8.pdf

[2] motorBench ® Development Kit:https://www.microchip.com/design-centers/motor-control-and-drive/motorbench-development-auto-tuning

[3] motor control design resources:https://www.microchip.com/design-centers/motor-control-and-drive

[4] motor control library:https://www.microchip.com/design-centers/motor-control-and-drive/motor-control-library

Leave a Reply

Your email address will not be published. Required fields are marked *