A New Approach of Sensorless Control Methodology

for Achieving Ideal Characteristics of Brushless DC

Motor Using MATLAB/Simulink
Santanu Mondal Arunabha Mitra Debjyoti Chowdhury Madhurima Chattopadhyay
Elec. & Instru. Engg. App. Elec. & Instru. Engg. App. Elec. & Instru. Engg. App. Elec. & Instru. Engg.
Techno India Heritage Institute of Heritage Institute of Heritage Institute of
Salt Lake , Kolkata Technology, Kolkata Technology, Kolkata Technology, Kolkata
santanu_aec1984@yahoo.co.in arunabha.official djbabai.debjyoti@gmail.com madhurima.chattopadhyay
@gmail.com @heritageit.edu

Abstract—This paper describes a new mathematical model of crossing detector [7]. The output of the zero crossing detector
permanent magnet brushless dc motor (PMBLDC) with (ZCD) is a square wave pulse and it is generated every time
sensorless commutated drive on MATLAB/Simulink based the trapezoidal back-EMF changes its phase.
platform. The main challenges in sensorless BLDC commutation
techniques compared with sensored drive lies in identification of Principally, four types of sensorless control methods can
first commutation point and minimization of torque ripple. In be found [8]. They are (i) Back-EMF Sensing method, (ii)
this work, an ideal back-EMF generation methodology has been Back-EMF Integration method, (iii) Flux Linkage-Based
designed and implemented in order to incline the system with the
ideal one. As a result, a significant reduction in torque ripple is method, and (iv) Detection of freewheeling diode conduction
achieved. The characteristics of this model give satisfactory method.
outputs over a wide range of controlled speed variation from 700
to 14000 rpm. Moreover, the torque and speed characteristic
Among these four techniques, we have adapted back-EMF
provide a high torque even at low speed. The strength of the sensing method owing to ease of implementation. In this
sensorless commutation technique used here is established by sensorless commutation technique, three zero crossing
performance analysis of the simulated design. detectors (ZCD) are employed with simple ON-OFF logical
operation. During each commutation cycle, two out of the
Keyword—Brushless dc motor; back-EMF; ZCD; electronic three phases remain conducting while the third phase is used
commutation technique; Sensorless drive
to determine the back emf through ZCD. This back-EMF
signal of the floating winding is used to determine the
commutation sequence of power devices in the three phase
I. INTRODUCTION
inverter [9, 10].
The use of permanent magnet BLDC motor is expanding
day by day in different applications especially in computer, This paper introduces the sensorless drive of BLDC motor
aerospace, military, automotive, industrial and household in Section I followed by mathematical modelling in Section II.
appliances as it holds high efficiency, torque, compactness Section III deals with implementation of simulation model in
over intact operating range and can be easily controlled due to MATLAB/Simulink. The output results of the simulated
proportional variation of torque with input voltage [1]. model are discussed in Section IV. Finally Section V
concludes the work.
The electronic commutation circuitry in BLDC motor
requires a three phase inverter, rotor position sensors (for II. DESCRIPTION AND MODELLING OF THE DRIVE SYSTEM
starting and providing the proper commutation sequence to
stator windings) [2]. The main limitation of such sensored In this model, a three phase, four pole, star connected
drive circuit not only increases the cost and size of the motor trapezoidal back-EMF type BLDC motor and three phase
but also needs special mechanical arrangement for mounting inverter is considered. Fig.1 shows the basic block diagram of
the position sensors (Hall sensors) [3]. Moreover, these Hall Sensorless drive of BLDC motor [11]. Figure 2 shows six-
sensors are temperature sensitive as they lose sensing switch inverter with equivalent circuit of BLDC Motor [10].
capability at the temperature beyond 120 ºC. To overcome The gating signals in the inverter circuit, given to the
this, sensorless drive of BLDC motor is introduced [4-6]. In MOSFET are sequenced at 60º intervals which is explained by
this sensorless commutation technique, both the hardware Table I where each MOSFET conducts for a duration of 120º
complexity due to position sensors as well as energy and space [12, 13].
consumption of associated circuitry are reduced.
In sensorless BLDC motor, the trapezoidal back-EMF is used During modeling, we assume that there are no power losses in
to detect the instantaneous position of rotor. The back-EMF the inverter and 3-phase motor winding.
waveforms generated from the stator go through a zero

978-1-4799-4445-3/15/$31.00 ©2015 IEEE

Fig.1. Basic Block diagram of Sensorless drive of BLDC Motor by back-EMF
detection method. Fig.2. Equivalent circuit of BLDC Motor fed six-switch inverter [6]

Applying Kirchhoff’s voltage law for the three phase stator 0 0

winding circuit and we get: 0 0
0 0
(1)
(7)
(2)
The stator self-inductances and mutual inductances are
(3) independent of the rotor position. Hence, assuming all the self-
inductances to be equal, we represent them by L (i.e., La = Lb
where Va, Vb and Vc are the respective three phase voltages of = Lc = L). Similarly, all mutual inductances of three different
the stator winding, Ra, Rb and Rc are the stator resistances, ia , phases can be considered to be equal and be denoted by M. As
ib and ic are the stator phase currents, La , Lb and Lc are the we have considered a balanced three phase system, all the
self-inductances, Mab, Mac, Mba, Mbc, Mca and Mcb are the phase resistances are equal and thus, are represented by R.
mutual inductances and ea, eb and ec are the three phase
back-EMFs. Rearranging equation (7) we obtain:

In a 3-phase BLDC motor, the back-EMF is a function of 0 0

the rotor position. The back-EMF of each phase has a 0 0
difference of 120º, thus equation of each phase should be as 0 0
follows:
(8)
. . (4)

. . (5) The electromagnetic torque equation is given as:

2
. . (6) . . . .
3
where k is back-EMF constant, is angular speed of rotor, . . (9)
and is the rotor position.
The equation of motion for simple system is given as:
From equations (1), (2) and (3), a complete BLDC model (10)
can be written as:

TABLE I. SIX STEP SWITCHING SEQUENCE

III. IMPLEMENTATION OF BLDC MOTOR IN
Rotor Position Switch Phase Currents MATLAB/SIMULINK
(Degree) Closed Ia Ib Ic According to the mathematical model, the complete motor
0-60 Q1 Q5 + - Off drive is simulated in MATLAB/ Simulink environment. The
entire drive is divided into several functional blocks such as
60-120 Q1 Q6 + Off -
BLDC motor, PI controller and inverter with commutation
120-180 Q2 Q6 Off + - logic. By the logical combination of these blocks, the
180-240 Q2 Q4 - + Off simulation model of BLDC motor drive is implemented. The
240-300 Q3 Q4 - Off + block representing BLDC motor is the main part of
300-360 Q3 Q5 Off - + simulation, which generates simulated outputs corresponding
to rotor speed and electromechanical torque. The main
attribute of this model is the implementation of a true back-
 EMF generator block that incorporates one dimensional TABLE II. BLDC MOTOR PARAMETERS USED FOR SIMULATION
interpolation of input values using a specific table. The
piecewise linear method is selected to produce the trapezoidal Motor parameters Values
back-EMF waveform, which divides one period of 0°-360°
No. of poles 4
into six commutation stages. Fig. 3 and Fig.4 show back-EMF
zero crossing detector and the entire model of BLDC motor DC source voltage 54 V
implemented in MATLAB/Simulink platform respectively. Moment of inertia 0.0002 Kg-m2
Stator Resistance 0.7 Ω
IV. SIMULATION RESULTS Self-Inductance 0.0272 H
The proposed simulation in this present work is established Rated Speed 3000 rpm
in the MATLAB / Simulink environment. The parameters of Damping Constant 0.2
BLDC motor for simulation study are shown in Table II. The Load Torque 4 N-m
output results like rotating speed, back-EMFs, three-phase
current of stator winding and electromagnetic torque are
shown in Fig. 5, Fig. 6, Fig. 7 and Fig. 8 correspondingly. V. SIMULATION RESULTS
These figures assure swiftness in start-up of the designed The proposed simulation in this present work is established
motor. Fig. 5 illustrates variation of rotor speed for different in the MATLAB / Simulink environment. The parameters of
preferred rpm which confirms a smooth achievement of BLDC motor for simulation study are shown in Table II. The
desired rpm with a settling time less than 0.02 sec. The output results like rotating speed, back-EMFs, three-phase
simulated model generates a smooth response over a wide current of stator winding and electromagnetic torque are
variation (700 - 14000 rpm) of speed. It has been observed shown in Fig. 5, Fig. 6, Fig. 7 and Fig. 8 correspondingly.
from Fig. 5 that in low speed (<1500rpm) the output achieves These figures assure swiftness in start-up of the designed
the desired speed within few milliseconds. motor. Fig. 5 illustrates variation of rotor speed for different
preferred rpm which confirms a smooth achievement of
desired rpm with a settling time less than 0.02 sec. The
simulated model generates a smooth response over a wide
variation (700 - 14000 rpm) of speed. It has been observed
from Fig. 5 that in low speed (<1500rpm) the output achieves
the desired speed within few milliseconds.

As desired rpm increases settling time also gets increased

upto 0.02 sec at 14000 rpm. The trapezoidal back-EMF
required for ideal commutation of the sensorless drive as
achieved by our proposed model is shown in Fig. 6.
The three phase stator current is represented by Fig. 7
having a phase shift of 120°. Fig. 8 shows the dynamic
Fig.3. Simulink model of zero crossing detector . response of electromagnetic torque which also settles within

Fig.4. Simulink model of BLDC motor.

 VI. CONC
CLUSION
14000
This paper proposes a neew mathematical model and
12000 simulation analysis of BLDC motor
m for sensorless operation in
MATLAB/Simulink platform. The zero crossing detector has
Rotor speed (rpm)

10000 been used to determine the rotoor position by replacing the hall
8000 sensors of sensored drive. Thee main advantage of this model
lies with precise commutation by
b introducing the methodology
6000 of true back-EMF generator. The
T output results of simulation
4000 give a satisfactory rotor speed over a wide variation of 700 -
14000 rpm. This model gives usu the flexibility to vary the time
2000 period of ideal back-EMF geenerated here according to the
0
required speed. The addition of
o the ideal back-EMF generator
0 0.02
0.04 0.06 0.08 0.1 also enhances the overall perforrmance of the designed motor.
Time (s)
Fig.5.Simulated output speed of BLDC motor.
REFER
RENCES
[1] M. R. Feyzi, M. Shafiei, S. A. KH. Mozaffari Niapour, Position
Sensorless Direct Torque Controol of Brushless DC Motor Drives Based
150 on Sliding Mode Observver Using NSGA-II Algorithm
Optimization.2011, 2nd Poweer Electronics, Drive Systems and
100 Technologies Conference.
Back emf (V)

50 [2] Cheng-Hu Chen and Ming-Yaang Cheng, A New Cost Effective

Sensorless Commutation Methood for Brushless DC Motors without
0 Phase Shift Circuit and Neutral Voltage. IEEE Transactions on power
-50 Electronics, Vol. 22, No. 2, Marcch 2007.
[3] Tae-Hyung Kimand Mehrdad Ehhsani. Sensorless Control of the BLDC
-100 Motors From Near-Zero to Highh Speeds. IEEE Transactions on Power
-150 Electronics, Vol. 19, no. 6, Novem mber 2004.
[4] Salaheddin A. Zabalawi and Addel Nasiri, State Space Modeling and
0
0.04 0.02 0.06 0.08 0.1 Simulation of Sensorless Conttrol of Brushless DC Motors Using
Time (s) Instantaneous Rotor Position Traccking. IEEE-2007. pp.90-94.
Fig.6. Waveform of Back-EMF generated for Phase A, Phase
P B and Phase C. [5] Cunshan Zhang, Dunxin Biaan. APWM Control Algorithm for
Eliminating Torque Ripple Caussed by Stator Magnetic Field Jump of
Brushless DC Motors. Proceeddings of the 7thWorld Congress on
80 Intelligent Control and Automattion, June 25 - 27, 2008, Chongqing,
China.
60 [6] Debjyoti Chowdhury, Madhurrima Chattopadhyay, Priyanka Roy,
“Modelling and Simulation of Cost Effective Sensorless Drive for
Current (mA)

40
Brushless DC Motor”, Internaational Conference on Computational
20 Intelligence: Modelling,Techniquues and Applications (CIMTA-2013),
Procedia Technology 10 (2013), 279-286.
2
0 [7] J. H. Zhou, Z. W. Zhong, M. Luoo & C. Shao. Wavelet-based correlation
-20 modelling for health assessment of fluid dynamic bearings in brushless
DC motors. Int J Adv Manuf Technol (2009) 41:421–429.
0 0.02 0.04 0.06 0.08 0.1 [8] Sharda Jaiswal, Debjyoti Choowdhury, Madhurima Chattopadhyay.
Performance Analysis of Sensoored and Sensorless Drive of BLDC
Time (s) Motor Using Different Tyyps of DC/DC Converters In
Fig.7. Stator Current for Three Phases. MATLAB/Simulink Platform. Proceedings of ITR International
Conference, 06th April-2014, Bhuubaneswar, ISBN: 978-93-84209-02-5.
[9] R. A. Gupta, Rajesh Kumar &Ajaay Kumar Bansal. Artificial intelligence
100 applications in Permanent Magneet Brushless DC motor drives. Published
online: 25 December 2009© Sppringer Science+Business Media B.V.
80 2009.
Torque (N-m)

60 [10] C. Gencer and M Gedikpinar, Modeling and simulation of BLDCM

using MATLAB/Simulink. Journnal of Applied sciences 6(3): 688-691,
40 2006. ISSN 1812-5654@20006 Asian Network for Scientific
Information.
20
[11] Xiaonan Xin, Hexiu Xu, Yangruui Luo, Huijuan Zhang. A New Method
0 of Brushless DC Motor Conttrol System Simulation. IEEE 2011.
PP.3667-3670.
0 0.02 0.04 0.06 0.08 0.1 [12] Jong Sun Ko, Young Seok Jung and Myung Joong Youn. MRAC load
torque observer for position coontrol of a brushless DC motor.Int. j.
Time (s) electronics, 1996, vol. 80, no. 2, 201-
2 209.
Fig.8.Electromagnetic Torque. [13] S.A.KH. Mozaffari Niapour, M.Tabarraie,
M M.R.Feyzi. A new robust
speed-sensorless control strategyy for high-performance brushless DC
0.02 seconds and the fluctuation of electrom
magnetic torque is motor drives with reduced torquue ripple. Control Engineering Practice
24(2014)42–54.
within the tolerance limit.