Electronic Control of Torque Ripple in Brushless Motors
`
`by
`
`Peter Franz Kocybik
`
`A thesis submitted to the University of Plymouth in partial fulfilment for the degree of
`
`Doctor of Philosophy
`
`School of Electronic, Communication and Electrical Engineering
`
`Faculty of Technology
`
`In collaboration with
`
`Automotive Motion Technology Ltd
`
`July 2000
`
`11
`
`

`
`This copy of the thesis has been supplied on condition that anyone who consults it is
`
`understood to recognise that its copyright rests with its author and that no quotation from
`
`the thesis and no information deriv~d from it may be published without the author's prior
`
`consent.
`
`1
`
`

`
`Peter Franz Kocybik
`
`Electronic Control of Torque Ripple in Brushless Motors
`
`Abstract
`
`Brushless motors are increasingly popular because of their high power density, torque to
`
`inertia ratio and high efficiency. However an operational characteristic is the occurrence of
`
`torque ripple at low speeds. For demanding direct drive applications like machine tools,
`
`robot arms or aerospace applications it is necessary to reduce the level of torque ripple.
`
`This thesis presents an in depth investigation into the production and nature of torque
`
`ripple in brushless machines. Different torque ripple reduction strategies are evaluated and
`
`one reduction strategy using Park's transform as a tool is identified as the promising
`
`strategy. The unified machine theory is checked to clarify the theory behind Park's
`
`transform; in particular assumptions made and general validity of the theory. This torque
`
`ripple reduction strategy based on Park's transform is extended to include the effect of
`
`armature reaction. A novel adaptive torque ripple reduction algorithm is designed. The
`
`ineffectiveness of the conventional approach is demonstrated. Further a novel torque ripple
`
`reduction strategy using direct measurements of the torque ripple is suggested, reducing
`
`implementation time and allowing higher accuracies for torque ripple reduction. Extensive
`
`measurements from the experimental system show the validity of the novel torque ripple
`
`reduction strategies. The experimental results allow derivation of a formula for all load
`
`111
`
`

`
`situations. This formula makes it possible to further increase the reduction accuracy and
`
`enables improved real time implementation of the torque ripple reduction algorithm.
`
`The work presented here makes a substantial contribution towards understanding the nature
`
`of torque ripple in brushless motors and solving the associated problems. The novel
`
`reduction strategies form the basis for the development of intelligent dynamometers for
`
`motor test beds. Further the torque ripple reduction method presented here can be used to
`
`overcome manufacturing imperfections in brushless machines thus removing the cost for
`
`precise manufacturing tools. Future designs of controllers can "build" their own correction
`
`formula during set-up runs, providing a motor specific torque ripple correction.
`
`IV
`
`

`
`3.1
`
`Structure of Thesis
`
`3.2
`
`Objectives
`
`Chapter 4 Literature Review
`
`4.1
`
`General Review
`
`4.2
`
`4.3
`
`Design Measures to Reduce Torque Ripple
`
`Supply Side Measures to Reduce Torque Ripple
`
`4.3.1
`
`Current Profiling for Sinusoidal Machines
`
`4.3.2 Current Profiling for Trapezoidal Machines
`
`4.3.3 Current Profiling for Both Sinusoidal and Trapezoidal
`
`Machines
`
`4.3.4 Direct Torque Control
`
`Chapter 5 Theoretical Analysis of Torque Ripple Production
`
`5.1
`
`Cogging Torque
`
`5.2
`
`Energised Torque Ripple
`
`5.2.1 Analytical Descriptions of Energised Torque Ripple
`
`5.2.2 Sinusoidal Excitation
`
`5.2.3 Excitation with Multiples of Two
`
`5.2.4 Excitation with Multiples of Three
`
`5.2.5 Excitation with Other Harmonics
`
`5.2.6 Conclusions from Simulations
`
`5.2.7
`
`Influence of Speed and Current Loading on Energised
`
`Torque Ripple
`
`5.2.8 Typical Values and Measurement Method for Energised
`
`Torque Ripple
`
`Chapter 6 Conventional Torque Ripple Reduction Strategies
`
`VI
`
`27
`
`28
`
`30
`
`30
`
`31
`
`36
`
`36
`
`41
`
`42
`
`52
`
`54
`
`54
`
`60
`
`61
`
`65
`
`67
`
`69
`
`71
`
`71
`
`74
`
`74
`
`76
`
`

`
`6.1
`
`Reduction Strategies for Cogging Torque
`
`6.2
`
`Reduction Strategies for Energised Torque Ripple
`
`6.2.1 Park's Transform
`
`6.2.2 Park's Transform for Ideal Sinusoidal Machines
`
`6.2.3 Park's Transform for Machines with Non-Sinusoidal Back
`
`Emfs
`
`6.2.4 Validity, Restrictions and Implementation Quantities for
`
`Park's Transform
`
`6.2.5 Torque Ripple Reduction using Park's Transform versus
`
`other Torque Ripple Reduction Strategies
`
`Chapter 7 New Adaptive Torque Ripple Reduction Strategy
`
`7.l
`
`Armature Reaction and its Effect on Torque Ripple
`
`7.2
`
`Novel Adaptive Torque Ripple Reduction Strategy.
`
`Implementation and Advantages
`
`Chapter 8 New Improved Torque Ripple Reduction Strategy
`
`Chapter 9 Experimental System
`
`9.l
`
`System Requirements
`
`9.2
`
`Hardware
`
`9.2.1 Motor
`
`9.2.2 Drive
`
`9.2.3 Control
`
`9.2.4 Load
`
`9.3
`
`Software
`
`9.3.1 Software System and Language
`
`9.3.2 Software Strategy
`
`VB
`
`76
`
`82
`
`84
`
`86
`
`91
`
`104
`
`106
`
`110
`
`110
`
`123
`
`131
`
`137
`
`137
`
`138
`
`138
`
`139
`
`140
`
`142
`
`143
`
`143
`
`144
`
`

`
`Chapter 10 Experimental Results
`
`10.1 Measurements
`
`10.2 Cogging Torque
`
`10.3 Electromagnetic Torque Ripple
`
`10.3.1 Torque Ripple Reduction at Light Loads
`
`10.3.2 Torque Ripple Reduction at Medium Loads
`
`10.3.3 Torque Ripple Reduction at Full Load
`
`10.3.4 Torque Ripple Reduction Using the Conventional Torque
`
`Ripple Reduction Strategy
`
`Chapter 11
`
`Formula for Reduction of Electromagnetic Torque
`
`Ripple
`
`Chapter 12
`
`Conclusions
`
`12.1 Achievements
`
`12.2 Further Work
`
`Appendices
`
`A 1 Motor Data Sheets
`
`A2 Resolver Circuit
`
`A3 Torque Transducer
`
`A3.1 Calibration Certificate for Torque Transducer
`
`A3.2 Torque Transducer Circuit
`
`A4 Full Set of Experimental Results
`
`List of References
`
`PcrM Paper
`
`VIll
`
`147
`
`147
`
`148
`
`154
`
`155
`
`158
`
`162
`
`166
`
`169
`
`178
`
`178
`
`180
`
`182
`
`183
`
`186
`
`189
`
`190
`
`192
`
`194
`
`213
`
`222
`
`

`
`List of Figures
`
`Figure 1.1 Motion system and its components
`
`Figure 2.1 Magnet flux distribution in airgap with rotor position
`
`Figure 2.2 Sectional view of typical brushless motor
`
`Figure 5.1 Modification of airgap flux distribution due to slot
`
`openIng
`
`Figure 5.2 Magnet edge aligned with stator tooth to preserve lower
`
`reluctance path
`
`Figure 6.1 Cogging torque from literature
`
`Figure 6.2 Cogging torque with rotational position for sample
`
`motor
`
`Figure 6.3 Stator and rotor reference frames for brushless motor
`
`Figure 6.4 Measured back emf waveforms from test motor
`
`Figure 6.5 Harmonic components of back emf waveform
`
`Figure 6.6 Resulting torque ripple for machine with non-sinusoidal
`
`back emfs fed with ideal sinusoidal phase currents
`
`Figure 6.7 Torque ripple harnlonics for machine with non(cid:173)
`
`sinusoidal back emfs supplied with sinusoidal phase currents
`
`Figure 6.8 Transformed non-sinusoidal back emf waveforms in dq(cid:173)
`
`plane
`
`Figure 6.9 Required iq current for torque ripple compensation
`
`Figure 6.10 Required phase currents for torque ripple cancellation
`
`IX
`
`3
`
`15
`
`16
`
`55
`
`57
`
`80
`
`81
`
`84
`
`92
`
`93
`
`94
`
`95
`
`96
`
`99
`
`100
`
`

`
`Figure 6.11 Harmonics of required phase currents for torque ripple
`
`compensation
`
`Figure 6.12 Phase torques and resulting torque for proposed phase
`
`currents
`
`Figure 7.1 Flux distribution without (dotted) and with excitation
`
`currents (solid)
`
`Figure 7.2 No load back emf waveforms for test motor
`
`Figure 7.3 Phase to star voltages for maximum current load
`
`Figure 7.4 Full load back emf waveforms after scaling and phase
`
`shifting
`
`Figure 7.5 Harmonic components of back emf waveform for no
`
`load case
`
`Figure 7.6 Harmonic components of back emf for full load
`
`Figure 7.7 Phase torque and resulting torque for no load (light load)
`
`case. Excitation with sinusoidal currents.
`
`Figure 7.8 Phase torque and resulting torque for load case.
`
`Excitation with sinusoidal currents.
`
`Figure 7.9 Torque ripple harmonics for no load case
`
`Figure 7.10 Torque ripple harmonics for full load case
`
`Figure 7.11 Phase torque and resulting torque for full load case.
`
`Excitation with corrections currents derived from no load case.
`
`Figure 7.12 Torque ripple harmonics for full load case. Excitation
`
`with correction currents derived from no load case.
`
`Figure 7.13 Correction currents derived from no load quantities
`
`Figure 7.14 Correction currents derived from full load quantities
`
`x
`
`101
`
`102
`
`111
`
`112
`
`113
`
`114
`
`115
`
`116
`
`117
`
`118
`
`119
`
`119
`
`121
`
`122
`
`124
`
`125
`
`

`
`Figure 7.15 Spectral components of correction currents derived
`
`from no load quaptities
`
`Figure 7.16 Spectral components of new correction currents
`
`derived from load quantities
`
`Figure 7.17 Phase torques and resulting torque for currents derived
`
`from full load quantities
`
`Figure 9.1 Complete system
`
`Figure 10.1 Cogging torque for test motor
`
`Figure 10.2 Cogging torque ripple for test motor
`
`Figure 10.3 Resulting output torque after cogging torque correction
`
`Figure 10.4 Resulting cogging torque ripple after cogging torque
`
`correction
`
`Figure 10.5 Phase current for correction of cogging torque
`
`Figure 10.6 Resulting output torque at light load with sinusoidal
`
`excitation
`
`Figure 10.7 Resulting torque ripple at light load with sinusoidal
`
`excitation
`
`Figure 10.8 Remaining electromagnetic torque ripple at light load
`
`after implementation of cogging torque reduction
`
`Figure 10.9 Resulting output torque at light load after cogging
`
`torque and electromagnetic torque ripple reduction
`
`Figure 10.10 Remaining torque ripple at light load after cogging
`
`torque and electromagnetic torque ripple reduction
`
`Figure 10.11 Required phase current for cogging and
`
`electromagnetic torque ripple reduction at light loads
`
`Xl
`
`126
`
`126
`
`128
`
`141
`
`149
`
`150
`
`151
`
`152
`
`153
`
`155
`
`156
`
`156
`
`157
`
`157
`
`158
`
`

`
`Figure 10.12 Resulting output torque at medium load with
`
`sinusoidal excitation
`
`Figure 10.13 Resulting torque ripple at medium load with
`
`sinusoidal excitation
`
`Figure 10.14 Remaining electromagnetic torque ripple at medium
`
`load after implementation of cogging torque reduction
`
`Figure 10.15 Resulting output torque at medium load after cogging
`
`torque and electromagnetic torque ripple reduction
`
`Figure 10.16 Remaining torque ripple at medium load after cogging
`
`torque and electromagnetic torque ripple reduction
`
`Figure 10.17 Required phase current for cogging and
`
`electromagnetic torque ripple reduction at medium loads
`
`Figure 10.18 Resulting output torque at full load with sinusoidal
`
`excitation
`
`Figure 10.19 Resulting torque ripple at full load with sinusoidal
`
`excitation
`
`Figure 10.20 Remaining electromagnetic torque ripple at full load
`
`after implementation of cogging torque reduction
`
`Figure 10.21 Resulting output torque at full load after cogging
`
`torque and electromagnetic torque ripple reduction
`
`Figure 10.22 Remaining torque ripple at full load after cogging
`
`torque and electromagnetic torque ripple reduction
`
`Figure 10.23 Required phase current for cogging and
`
`electromagnetic torque ripple reduction at full load
`
`xu
`
`159
`
`159
`
`160
`
`161
`
`161
`
`162
`
`163
`
`163
`
`164
`
`165
`
`165
`
`166
`
`

`
`Figure 10.24 Output torque at full load for conventional torque
`
`ripple reduction strategy
`
`Figure 10.25 Remaining torque ripple at full load for conventional
`
`torque ripple reduction strategy
`
`Figure 11.1 Amplitude of peak-to-peak torque ripple for individual
`
`torque ripple harmonics over load range
`
`Figure 11.2 Amplitude of correction current harmonics over load
`
`range
`
`Figure 11.3 Torque output with sinusoidal excitation for load
`
`current of 4.5 Arms
`
`Figure 11.4 Torque ripple after cogging torque correction for 4.5
`
`Arms
`
`Figure 11.5 Resulting output torque with new formula for 4.5 Arms
`
`Figure 11.6 Remaining torque ripple after reduction with new
`
`formula
`
`167
`
`167
`
`170
`
`172
`
`174
`
`175
`
`175
`
`176
`
`XUl
`
`

`
`List of Tables
`
`Table 5.1
`
`Torque ripple frequencies and amplitudes through
`
`sinusoidal excitation
`
`Table 5.2
`
`Torque ripple frequencies and amplitudes through
`
`excitation by 2nd harmonic
`
`Table 5.3
`
`Torque ripple frequencies and amplitudes through
`
`excitation by 3rd harmonic
`
`Table 5.4
`
`Torque ripple frequencies
`
`Table 5.5
`
`Torque ripple frequencies for star configuration
`
`Table 7.1
`
`Amplitude of torque ripple harmonics in percent for
`
`no load and load case
`
`Table 7.2
`
`Amplitude of torque ripple harmonics in percent for
`
`no load and load case for different excitation
`
`Table 7.3
`
`Phase current harmonics in percent derived from no
`
`load and load case quantities
`
`Table 11.1
`
`Matrix T, Torque ripple over output torque level
`
`Table 11.2
`
`Matrix A, Correction current harmonics over load
`
`range
`
`XIV
`
`66
`
`68
`
`70
`
`72
`
`73
`
`120
`
`122
`
`127
`
`170
`
`171
`
`

`
`Acknowledgement
`
`This study was made possible through a studentship granted by the University of
`
`Plymouth. I would like to thank my supervisor Dr Peter White for his constant support and
`
`encouragement and for many fruitful discussions. He was always available and spent many
`
`hours advancing the progress of this project. I would also like to thank Automotive Motion
`
`Technology, formerly Norcroft Dynamics, for supporting this research project. Especially
`
`the hardware support and the opportunity to learn about the world of "real motors" by
`
`spending some time at Norcroft Dynamics was invaluable for the success of the project. In
`
`particular I would like to thank my second supervisor Dr Mike Werson for providing the
`
`idea for this research project and for his constant input during the course of the research.
`
`Many interesting discussions helped me to understand the problems more deeply and
`
`provided me with ideas for possible solutions. I would also like to thank Mr Sandro
`
`Murelli, who taught me a lot about the design and operation of brushless machines. I am
`
`also thankful to Dr Bob Beaven for interesting discussions concerning the nature of torque
`
`ripple in brushless machines. Thanks also goes to Mr Franz Fuchs, my fellow research
`
`student, with whom I spent many hours in helpful discussions. Finally I like to thank my
`
`parents for their encouragement and support.
`
`xv
`
`

`
`Author's Declaration
`
`At no time during the registration for the degree of Doctor of Philosophy has the author
`
`been registered for any other University degree.
`
`This studied was financed with the aid of a studentship from the Higher Education Funding
`
`Council and was carried out in collaboration with Automotive Motion Technology Ltd,
`
`Andover. The collaboration was in the form of hardware support and regular visits for
`
`progress meetings.
`
`Relevant scientific seminars, conferences and exhibitions were regularly attended.
`
`A paper was presented at the Power Conversion and Intelligent Motion conference, PCIM,
`
`1997 in Nuremberg.
`
`Signed
`
`Date
`
`l~'~
`'2 c! 7--/2 oaQ
`
`XVI
`
`

`
`Chapter 1 Introduction
`
`1.1 Motion Systems
`
`Modem motion systems comprise three principal components; the motor, the drive and the
`
`control. The motor is the essential component of every motion system converting electrical
`
`into mechanical energy. The drive provides the power management to the motor. The
`
`control unit serves to supervise the drive.
`
`When electrical motors were first utilised they were directly interfaced to the supply
`
`system. This made for ease of use but restricted their capabilities. Most applications were
`
`straight forward, relatively slow, highly repetitive processes. Motors had to be reliable and
`
`simple to use. To use motors for more demanding applications it became necessary to
`
`invent additional mechanical and electrical devices to extend their capabilities. This
`
`included devices to facilitate start up and braking for instance. Often these devices were
`
`cumbersome and not very reliable to use.
`
`New applications demanded higher motor capabilities such as assembly processes
`
`requiring different speeds and quicker response times. To meet these higher requirements
`
`drives were developed. Drives allow the operation of motors at different speeds and load
`
`levels by adjusting voltage and current levels. A well-matched matched drive system
`
`makes it possible to use a motor up to its electrical and mechanical limits. Older drive
`
`systems were very limited in their capabilities and only allowed fairly simple operations to
`
`1
`
`

`
`be performed. However during the last 20 years power electronics has evolved a great deal
`
`[1], [2] and [3]. Drives have a much faster response these days and will accommodate
`
`much higher power levels.
`
`The improvements in drive technology made it necessary to use control systems to
`
`supervise the drive [4]. Whereas a simple passive rectifier could convert an ac into a dc
`
`voltage to supply a fixed speed dc motor, an active power converter will drive a dc motor
`
`at variable speeds. However the active power control usually requires some form of
`
`control. As capabilities of and demands on drives increased demands on control grew in
`
`parallel. From simple analogue to extremely sophisticated digital or combined analogue
`
`and digital control systems there exists a wide range of control systems in use with specific
`
`drive systems.
`
`Although some motors are still operated without drive and control systems there is an
`
`increasing trend to use more and more sophisticated drive and control systems to facilitate
`
`higher user demands. It is not a simple matter of choosing the right motor for a particular
`
`application, it is the overall system performance, which counts. Only a well-matched and
`
`fine tuned motion system will achieve the highest possible performance. Therefore it
`
`becomes more important to design good interfaces between the individual components of a
`
`motion system. Increasingly drive and control components will be integrated with the
`
`motor to form a compact motion system.
`
`Figure 1.1 shows a motion system with its components. It illustrates the interface
`
`connections between the components and shows possible feedback loops between them.
`
`2
`
`

`
`Control signals
`
`Supply current
`
`Control
`
`....
`....
`..
`
`JI'"
`
`Drive
`
`...
`.....
`
`JI'"
`
`.... Motor
`....
`...
`
`....
`-.....
`
`Current and/or voltage feedback
`
`Position and/or speed feedback
`
`Figure 1.1 Motion system and its components
`
`In the following section the individual components of a motion system: motor, drive and
`
`control will be looked at in more detail.
`
`1.2 Different Motor Types
`
`Many different varieties of motors exist including special design motors for particular
`
`applications. The more important motors are described in [5], [6] and [7]. In the following
`
`the main motor types will be briefly looked at.
`
`The dc motor is probably the best known motor type. It is operated using dc voltages and
`
`dc currents. Stator and rotor carry windings, which are both excited using dc current. One
`
`of them constitutes the excitation field and the other one the torque producing field.
`
`Alternatively permanent magnets can be used on the stator replacing the stator winding and
`
`providing the excitation field. Conventional dc motors are used in a variety of standard
`
`3
`
`

`
`applications usually running at fixed speed. They can be supplied either from a dc voltage
`
`source, e.g. battery, or from an ac supply, e.g. mains, using a rectifier. No sophisticated
`
`power management or control is required. DC motors can also be used as variable speed
`
`drives when powered by drives capable of varying supply voltage and current. These can
`
`then be used in applications where different speeds are required, e.g. modern production
`
`processes. DC motors are simple to use and to control; which makes them very popular.
`
`Speed can be varied by simply changing the supply voltage, and torque production can be
`
`adjusted by varying supply current. Both are straight forward linear relationships. However
`
`most dc motors are more expensive to manufacture than other motors because they have
`
`windings on both stator and rotor.
`
`The induction motor is the most popular motor type in industry. About 80% of all
`
`industrially used motors are induction motors. Induction motors consist of a stator
`
`containing the windings and normally a laminated rotor containing a solid cage instead of
`
`windings, e.g. squirrel cage. These motors can be connected to an ac supply voltage, e.g.
`
`the 3 phase mains, to run at a fixed speed. This is a simple way of driving them and proves
`
`sufficient for many standard applications. For more demanding applications induction
`
`motors are used with inverters. Inverters allow variation of the supply frequency and
`
`voltage to the induction motor. Thus the induction motor becomes a variable speed drive.
`
`This is useful for production processes where different speeds are required. Inverters are
`
`off the shelf items, available from many manufacturers. They can be controlled using
`
`standard control strategies like PWM control. However the induction motor exhibits a few
`
`undesirable features when driven using standard control strategies. The start up torque is
`
`low and there are torque resonances in the lower frequency range. A considerable amount
`
`of research is directed towards addressing these problems and generally improving the
`
`4
`
`

`
`overall characteristics, like response times, of induction motor systems. Extremely
`
`sophisticated motor and drive models have been developed. The research is on going.
`
`Brushless motors or so-called permanent magnet motors are relatively new motor types.
`
`All brushless motors have magnets, hence the name permanent magnet motors. These are
`
`normally mounted on the rotor. The stator contains the windings. The stator configuration
`
`may be similar to the induction motor. Brushless motors exhibit the same control features
`
`as traditional dc motors. The motor speed can be varied by changing the supply voltage,
`
`and the torque output can be manipulated by controlling the supply current. Both are
`
`described through linear relationships. However the rotor position needs to be known to
`
`drive the brushless motors. For this purpose feedback devices like Hall effect devices,
`
`encoders or resolvers are used. A considerable amount of research is directed towards
`
`replacing these feedback devices by indirect methods like measuring supply currents and
`
`voltages to determine rotor position. Some so-called sensorless brushless drive packages
`
`are already available. As the price of magnets has come down over the last 15 years
`
`brushless motors have become cheaper to manufacture and are now seen as the main
`
`contender to the induction motor. The main advantage being the simpler control principle
`
`compared to the induction motor. However it is clear from the fact that the rotor position
`
`needs to be known that all brushless motors require some form of control and associated
`
`power stage. It is not possible to simply run them from the mains. This means that they are
`
`more suited for more demanding applications where the extra cost for the power stage and
`
`the control is justified.
`
`The switched reluctance motor is also a fairly new motor type. The stator contains the
`
`windings, whereas the rotor is a simple stack of metal laminations formed to provide
`
`5
`
`

`
`reluctance torque. This makes for a motor, which is easy to manufacture. However the
`
`control has the task of driving the nlotor in a way to ensure constant torque output. This
`
`requires sophisticated control mechanisms. Torque ripple and noise posed severe problems
`
`for manufacturers of switched reluctance motors up to a few years ago. Only recently have
`
`switched reluctance motors been used in higher volume products. Research to improve
`
`drive system characteristics is ongoing.
`
`Besides the main motor types more specialised motors or motors for a more restricted
`
`range of applications exist. The best known of these is the stepper motor, which is often
`
`used for lower power positioning tasks. Other motors are hybrids between standard motor
`
`types.
`
`From the above it can be concluded that the motor user faces a bewildering choice of motor
`
`types for most applications. Normally it is not very meaningful to compare one motor type
`
`to another. A well-designed motor will be comparable in terms of power to weight ratio
`
`and efficiency to most other well-designed motors of a different type. However from a
`
`scientific point of view there are considerable differences between the individual motor
`
`types. All motor types have inherent characteristics. Depending on the specific application
`
`these can be intolerable; forcing the choice of a different motor type to avoid them. The
`
`ideal motor doesn't exist. Later it will be seen how the design of a sophisticated control
`
`improves motor performance to overcome these problems.
`
`This study considers in detail the brushless motor. The main reason being that brushless
`
`motors have become cheaper because the price of magnets has come down over the last 15
`
`years. They are now more competitive and starting to take market share from induction
`
`6
`
`

`
`motors. Furthermore they are judged to be potentially very interesting for new application
`
`areas like hybrid cars, which use both combustion and electrical motors for traction
`
`purposes. Also a lot of research has been directed towards induction machines and
`
`switched reluctance machines. Progress there has been very fast over the last 5 years.
`
`Whereas in the area of brushless motors most research has been directed towards replacing
`
`the rotor feedback systems. The brushless motor has therefore been found to be both a
`
`promising motor type and interesting for further scientific examination. Hardware in the
`
`form of a suitable brushless motor was made available for this study by the supporting
`
`company.
`
`It has been mentioned before that it is necessary to judge the whole motion system rather
`
`than the motor itself when a particular motion task needs solving. A sophisticated control
`
`strategy allows improvement of the operating characteristics of the motor and can extend
`
`its performance envelope.
`
`In the following section the developlnent and importance of drive and control systems will
`
`be emphasised. Drives and controls are described together as they form a unit and cannot
`
`be chosen independent from each other.
`
`1.3 Trends in Drive and Control System Development
`
`Traditionally many industrial processes utilise fixed speed machines. Induction motors for
`
`instance are switched directly on line (dol) and run up to near their synchronous speed at
`
`which they are then operated. This is a simple and reliable way to use electrical machines.
`
`7
`
`

`
`However this mode of operation does not allow for different working speeds. Therefore the
`
`use of gear boxes is required to achieve different running speeds. Gear boxes introduce
`
`additional mechanical components with
`
`the associated problems. They
`
`require
`
`maintenance, wear out and they might introduce undesired resonances into the system. But
`
`most importantly they contribute to system losses. Further, the fact that the motor is not
`
`always operated at optimum speed, also leads to higher energy consumption. As energy
`
`consumption is under close scrutiny due to recent legislation and higher environmental
`
`awareness it becomes necessary to search for more energy efficient ways of motor
`
`operation.
`
`The second major disadvantage of fixed speed machines using gear boxes is their poor
`
`dynamic behaviour. Quick changes in production speeds are hard to accommodate. Many
`
`production processes require fast ramping up to full speed and numerous braking and
`
`reaccelerating cycles. Fixed speed machines are not ideally suited to this purpose.
`
`Finally many other applications require a continuous speed variation. Again fixed speed
`
`machines are not able to provide this feature.
`
`In conclusion it can be said that energy consumption, dynamic requirements and advanced
`
`application requirements necessitate the use of variable speed instead of fixed speed
`
`machines.
`
`Recent developments are reviewed in [8], [9] and [10].
`
`8
`
`

`
`1.4 Variable Speed Drives and Controls
`
`Most electrical machines can be used as variable speed machines if power electronics are
`
`used to supply the machine. Induction motors for instance can be operated at varying
`
`speeds if frequency converters are used for their supply.
`
`There are three major motor types, which are normally used for variable speed applications
`
`[11], [12] and [13].
`
`The first one is the conventional brushed dc motor. The major advantage here is the ease of
`
`speed control. As speed is directly proportional to applied voltage only a simple power
`
`electronics configuration is required. However the major disadvantage is the use of
`
`mechanical parts, that is brushes and commutators. This leads to mechanical wear and
`
`makes regular maintenance necessary, especially if a high number of speed changes occurs.
`
`Furthermore the occurrence of sparks at commutation instances makes the dc motor
`
`unsuitable for hazardous environments. Therefore there are considerable disadvantages in
`
`utilising the brushed dc motor for variable speed applications.
`
`The second motor type is the induction motor. The induction motor is the most widely used
`
`motor type today. Its simple construction makes it cheap to manufacture. Because it is used
`
`in a wide variety of applications many motor engineers are familiar with it and prefer its
`
`use. The major drawback of the induction motor is the non-linear operating characteristic.
`
`Neither speed nor torque output can be controlled in a simple manner. Advanced control
`
`strategies like vector control need to be employed to make use of the motor potential.
`
`Furthermore torque ripple occurs at low speed which makes fairly sophisticated control
`
`approaches necessary to combat this problem. Recently highly refined motor models have
`
`9
`
`

`
`been developed which take into account a large number of motor parameters and operating
`
`conditions. These models demand high computing power in order to implement advanced
`
`control techniques. However it is very likely that the induction motor will continue to take
`
`a major share of the motor market.
`
`The third motor topology considered here is the so-called brushless motor. This is the
`
`motor type chosen for the purpose of this project. This motor type has in the last few years
`
`emerged as a main competitor to the induction motor. As mentioned before the brushless
`
`motor needs a drive and the associated control to be operational. It is therefore only natural
`
`to utilise the variable speed capabilities of the brushless motor to justify the extra cost
`
`incurred by using the drive and the control. Chapter 2 will be looking at the operating
`
`principles of the brushless motor and list some application areas for its use as a variable
`
`speed machine.
`
`In the following section the terminology used throughout this thesis will be defined.
`
`1.5 Terminology
`
`It is necessary to define a few technical terms as the use In the literature vanes
`
`considerably.
`
`First the terms motor and machine will be used interchangeably in the following. The
`
`reason being, that each electrical n1achine can be operated as a motor or a generator.
`
`Usually the brushless machine will be operated in motoring mode.
`
`10
`
`

`
`Secondly the tem1 brushless machine itself requires clarification. Originally it applied to
`
`the so-called brushless dc motor which replaced the conventional brushed commutation
`
`through electronic commutation. Therefore the new machine was considered brushless as
`
`opposed to being brushed. However the induction motor for instance is of course also
`
`brushless.
`
`The following is an attempt to define brushless machines.
`
`Definition: A brushless motor is a machine, which uses permanent magnets as the
`
`excitation source and commutates its phases electronically.
`
`All brushless machines use permanent magnets, they are therefore often called permanent
`
`magnet machines. However conventional dc motors can also employ permanent magnets,
`
`as can synchronous machines. It is therefore slightly ambiguous to call the brushless
`
`machine a permanent magnet machine. What makes it discernible from other permanent
`
`magnet machines is the fact that the commutation is done electronically. In the permanent
`
`magnet brushed dc motor the comn1utation is still done mechanically. In the permanent
`
`magnet synchronous motor
`
`the
`
`three-phase