IMPLEMENTATION OF DSP BASED SENSORLESS CONTROL WITH
DIRECT BACK-EMF DETECTION METHOD FOR BLDC MOTOR

  IMPLEMENTATION OF DSP BASED SENSORLESS CONTROL WITH DIRECT BACK-EMF DETECTION ........pdf (Size: 663.66 KB / Downloads: 0)
ABSTRACT
This thesis presents a back EMF sensing scheme, direct back EMF detection, for
sensorless Brushless DC (BLDC) motor drives with a DSP based controller. Using this
scheme, the motor neutral voltage is not needed to measure the back EMF. Instead the
method utilizes the high-speed A/D converter channels, which are available on a DSP
controller. The true back EMF of the floating motor winding is detected during the off
time of PWM because the terminal voltage of the motor is directly proportional to the
phase back EMF during this interval. Also, the back EMF voltage is referenced to
ground without any common mode noise. Therefore, the developed back EMF sensing
method is immune to switching noise and common mode voltage. As a result,
attenuation and filtering is not necessary for the back EMF sensing.
The simulation of the BLDC motor drive system is implemented in SIMULINK
MATLAB software. The simulation of the system for important characteristics such as
speed, torque, phase current, terminal voltage, and Back EMF (BEMF) are monitored.
The simulation modeling involves solving many simultaneous differential equations,
each depending upon the inputs to the motor and the simulation constants.
A mathematical model of the drive system is also developed to analyze the
performance of the proposed driver.
The system is implemented developing the hardware, using a digital signal
processor (dsPIC30F), which is programmed with sensorless control for BLDC motor.
The implementation through assembly language programming of DSP has resulted in
reduced hardware and fast response of the controller. The high performance of digital
signal processors (DSPs) minimizes the control loop delays. Also further modifications
xvi
in control structure are easily possible by changing the software. The implemented
hardware can support speed range up to 3500 rpm of the BLDC motor, with reduced
back EMF noise. The validity of the proposed BLDC motor drive system is verified
through simulation and hardware results such as phase current, back EMF signal
waveforms and speed. The experimental results on a 3-phase, 24 V, 120 W BLDC
motor using dsPIC30F (DSP) based digital controller closely agree with the simulation
results.
1
CHAPTER I
INTRODUCTION TO BRUSHLESS DIRECT CURRENT (BLDC) MOTOR
1.0 Research Objective
The objective of this thesis is to present the design of a 3-phase sensorless
brushless dc (BLDC) motor control with back-EMF (electromotive force) zero-crossing
sensing using an AD converter. It is based on Microchip dsPIC30F3010 DSP which is
dedicated for motor control applications. The system is designed as a motor drive system
for 3-phase BLDC motors and is targeted for applications in both industrial and appliance
fields. The reference design incorporates both hardware and software parts of the system
including details of hardware layout. This thesis also includes the basic motor theory,
simulation implementation concept, hardware implementation and software design.
1.1 Research Methodology
The research methodology of this thesis involves a number of different tasks that
are needed to lead towards completion. The first task is to define the objective of the
research in which the target specification of end product is defined. This followed by the
literature review where all the theoretical information regarding the research is gathered
and a comparison of previous similar research is discussed. A brief description on the
BLDC motor theory and performance is then presented. The advantage of the proposed
back-EMF detection scheme for sensor-less control is compared with the conventional
back-EMF detection schemes. Next in this thesis, the simulation of the targeted controller
implementation for the drive system using MATLAB SIMULINK software tools is discussed.
The simulation waveforms of voltages, phase currents and speed response are obtained to
compare with the results from proto-type hardware drive system. The next task is to design
the hardware for the target controller, based on the application target. The component
2
ratings and type were selected. Once the hardware design is completed the software
implementation is carried out. The software code matching the hardware design is
developed in this stage. The next step is to integrate the software code and the hardware
to debug any failures. This task is implemented with ICD2 [15] debugging software. The
following task is to analyze the test results obtained with the controller and motor to
determine the performance, and also the waveforms of critical parameters captured during
this stage. The final stage is to conclude the research findings and the thesis write up.
Figure 1.1 shows the flow chart of the research methodology of this thesis.
Figure1.1: Flow chart showing the research methodology.
Software
Implementation
System testing and
results analysis
Research Conclusion
and thesis write up
Define
Objective
Literature
Review
Simulation
Implementation
Hardware
Design
Hardware and Software
Integration
3
1.2 Literature Review
The 3-phase BLDC motors are well adoptable for industrial applications that require
medium and very high speeds. Two important characteristics, low inertia and high peak
torque, result in the motor capable of quick accelerations and decelerations. Sensor and
sensorless control are two methods of control of BLDC motors [1-5]. In sensor control Hall
sensors are normally used which need maintenance. An approach to position sensorless
BLDC motor drive [6], a new algorithm for sensorless operation [7] and sensorless control
without signal injection [8] are reported. Two types of sensorless control techniques of PM
BLDC motors are discussed [12]. The first type is the position sensing using back EMF of
the motor, and the second one is position estimation using motor parameters. The position
estimation scheme usually needs complicated computation, and the cost of the system is
relatively high. The back EMF sensing scheme is the most commonly used method, which
is adopted in this thesis. The advantages of the position sensing using back EMF are:
• It is suitable to be used on a wide range of motors and the method is easily
implemented on both Y and Δ connected 3-phase motors.
• It requires no detailed knowledge of motor properties.
• It is relatively insensitive to motor manufacturing tolerance variations.
• It will work for either voltage or current control.
In a 3-phase BLDC motor, only two out of three phases are excited at any time,
leaving the third phase winding floating. The back EMF voltage in the floating winding can
be measured to establish a switching sequence for commutation of power devices in the
3-phase inverter. The conventional method of sensing back EMF is to build a virtual
neutral point that will, in theory, be at the same potential as the center of a Y wound motor
and then to sense the difference between the virtual neutral and the voltage at the floating
4
terminal. However, when using a chopping drive, the neutral is not a standstill point. The
neutral potential jumps from zero up to near dc bus voltage, creating a large common
mode voltage since the neutral is the reference point [9].
Meanwhile, the PWM signal is superimposed on the neutral voltage as well,
inducing a large amount of electrical noise on the sensed signal. Proper sensing of the
back EMF requires a lot of attenuation and filtering. The attenuation is required to bring the
signal down to the allowable common mode range of the sensing circuit, and the low pass
filtering is to smooth the high switching frequency noise. The result is a poor signal to
noise ratio of a very small signal, especially at start-up where it is needed most.
Consequently, this method tends to have a narrow speed range and poor start up
characteristics. To reduce the switching noise, the back EMF integration [13], third
harmonic voltage integration [10] and flux estimation [7] were introduced.
The integration approach has the advantage of reduced switching noise sensitivity.
However, it still has the problem of high common mode voltage in the neutral. The flux
estimation method has estimation error at low speeds. An indirect sensing of zero crossing
of phase back EMF by detecting conducting state of free-wheeling diodes in the unexcited
phase was also approached [6]. The implementation of this method is complicated and
costly, while its low speed operation is still a problem.
In this thesis a back EMF detection method, which does not require the motor
neutral voltage is implemented. The back EMF can be detected directly from the terminal
voltage by properly choosing the PWM and sensing strategy. The resulting feedback signal
is not attenuated, providing a signal with a very good signal/noise ratio. As a result the
5
proposed sensorless BLDC motor drive provides a much wider speed range up to 4000
rpm, from start-up to full speed, than the conventional approaches mentioned above.
The report of this thesis conducts the theoretical analysis of the concept of the
direct back EMF detection scheme, providing detailed understanding of the method. In the
past, several integrated circuits based on neutral voltage construction have been
commercialized [16, 17]. Unfortunately, all these ICs are all analog devices, which lack
flexibility in applications, regardless of poor performance at low speed. Use of 8-bit
microcontrollers have been the mainstay of embedded-control systems for a long time [9].
However, the computational power and command execution speed of these controllers is
lower compared to a Digital Signal Processor (DSP).
One single-chip architectural platform that is ideal for BLDC motor control is the 16-
bit Digital Signal Controller (DSC). The DSPs can apply very complicated control theory
and speed estimation for the sensorless BLDC motor control. The DSP devices are
available for a low cost; and the instructions sets are easy to use. Low system cost and
high flexibility are good motivations to design a new DSP based controller which is
dedicated to sensorless BLDC drive [18]. The flexibility mentioned here are the further
modifications in control structure easily accomplished by changing the software
programming. As a result, a low cost DSP based controller is developed, implementing the
proposed back EMF sensing scheme.
1.3 Brushless Direct Current (BLDC) Motor Background
Brushless Direct Current (BLDC) motors are one of the motor types which currently
becoming popular. BLDC motors are utilized in wide range of industries such as consumer
electronics, medical, automotive, industrial automation equipment and aerospace. The
6
commutation of BLDC motors are not the same as the brush type DC motor. The
mechanical commutator of the brush dc motor is replaced by electronic switches, which
supply current to the motor windings as a function of the rotor position. This kind of ac
motor is called a brushless dc motor, since its performance is similar to the traditional dc
motor with commutators. BLDC motors have many advantages compared to brush type
DC motors and induction motors, listed as follows [1-5]:
• Better speed versus torque characteristics.
• High dynamic response.
• High efficiency.
• Long operating life.
• Noiseless operation.
• Higher speed ranges.
In addition, the ratio of torque delivered to the size of the motor is higher, making it
useful in applications where space and weight are critical factors. Over the years of
advanced technology development in power semiconductors, embedded systems,
adjustable speed drives (ASDs) control schemes and permanent-magnet brushless
electric motor production have contributed for reliable and cost-effective solution for
adjustable speed applications. Household appliances are expected to be one of fastestgrowing
end product market for electronic motor drives (EMDs) over the following next few
years [19]. The major appliances include clothes washers, room air-conditioners,
refrigerators, vacuum cleaners, freezers, etc. The market volume is predicted to be a 26%
compound annual growth rate over the five years from 2000 to 2005, as shown in
Figure 1.2.
7
Figure1.2: Market trend for electronic motor drives in household appliances [19].
The automotive industry will also see the explosive growth ahead for BLDC type
electronically controlled motor system owing to the compact design and high efficiency of
BLDC motor. The appliances and devices use the electric motors to convert electrical
energy into useful mechanical energy required by the load. Consumers now demand for
lower energy costs, better performance, reduced acoustic noise, and more convenience
features. In recent years, proposals have been made for new higher energy-efficiency
standards for appliance industry, which will be legalized in near future [20]. These energy
standards proposals present new challenges for appliance designers. The continuous
global demand for higher efficiency and better performance, enable the transition of
industries to switch over to ASDs. The BLDC motor drive system which is cost effective
with high performance will be the sought after system.