Design and Development of Switched

Reluctance Motor for Electric Vehicle

Application
Shubham R Patel* , Nikita Gandhi** , Naga Chaithanya** , Dr. B. N. Chaudhari* and Akshay Nirgude*
*
College of Engineering Pune
**
Tata Motors Pimpri

Abstract—Wide torque-speed range and high reliability of variable reluctance. The stator and rotor both have
are the primary need of the traction motor in electric vehi- salient poles which make SRM a doubly salient machine.
cles. Switched Reluctance machine (SRM) offers numerous The stator poles consist of winding, which when excited
advantages such as simple and robust construction, high
speed and high durability, low cost, and fault tolerance allows rotor poles to align with the stator poles. The
control capabilities, high reliability as each phase works torque is produced by continuous switching of stator
independently. In view of this, the paper describes the windings [2]. SRM provides advantages over other ma-
design methodology of 8/6, 7.5 kW, 4000 RPM ’Saturated chines such as high temperature loading capability in
SRM’, which allows high energy conversion ratio for temperatures range of 160◦ , reduced cost, simple and
the converter. The design step verification of the motor
winding inductance at different rotor positions from un- rugged construction. Despite this applications, SRM have
aligned to aligned position and static flux-linkage charac- some challenges too. SRM, due to its doubly salient
teristics is carried out. Design verification and performance structure possesses high torque ripple and acoustic noise.
calculation are carried out using finite element analysis Even though acoustic noise is not a major issue for the
method using Ansys Maxwell Software. Flux distribution electric vehicle but torque ripples are to be reduced [3].
is managed efficiently to minimize losses. The mechanical
design of motor is done considering the stresses and forces The paper elaborates the design methodology, challenges
exerted on the motor structure, which involved the drive faced during design and their solutions. The results of
arrangement used. In order to make the design process FEA and combined simulation approach using Maxwell-
more efficient Ansys Simplorer software is used to analyze Simplorer link has been discussed. The paper also deals
the performance verification of motor for converter control with multiphysics analysis for motor design.
system. The same design procedure was carried out to
design another low voltage 3.5 kW SRM. The paper II. SRM D ESIGN P ROCEDURE
presents the comparison of proposed machines with the
Induction motors of the same outer geometrical dimensions. A. Selection of Dimensions
Keywords—switched reluctance motor, FEA, maxwell,
electric vehicle, simplorer, saturated SRM, multiphysics. The design procedure of SRM is started with the
similar outer geometric dimensions of an Induction motor
I. I NTRODUCTION which was to be replaced by SRM. The stator outer
diameter Ds = 190mm and stack length L = 80mm.
The electric motor is an important component in an The DC bus voltage was chosen to be 320V.
electric vehicle because it affects the performance and An increasing number of phases reduces torque ripple
efficiency of the vehicle. Induction motor, permanent but requires more switches [4], [5]. A trade-off is made
magnet synchronous motor (PMSM) and brushless DC between torque ripples and no. of switches, so the
(BLDC) machines are widely used for this application. proposed number of phases is m = 4. The number of
PMSM BLDC have higher efficiency power density. stator and rotor poles is given as Ns /Nr = 8/6. The
However limited availability and a high cost of rare earth initial motor dimensions were calculated from design
materials used for PMSM and BLDC have become a procedure proposed by R Krishnan in [2] and Pavol et
limitation for their use in EV. Considering mass manu- al in [6].
facturing of EVs in future, it is very important to seek A number of turns of the winding is the important
for a low cost as well as high power density motor. factor which affects the shape of torque-speed charac-
Switched Reluctance Motor (SRM) has emerged as major teristics. With a larger number of turns, comes high
contender for this application [1]. torque at low speed with lesser speed range. While with
The switched reluctance motor works on the principle lesser turns higher speed range is achieved but the torque
978-1-4673-8888-7/16/$31.00 c 2016 IEEE decreases.[7] Fig. 1a shows comparative analysis of

Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on July 16,2024 at 05:39:35 UTC from IEEE Xplore. Restrictions apply.
 (a) Torque vs Speed for different no. of turns. (b) Flux linkage vs current characteristics for proposed
SRM.

Fig. 1

torque profile at a different number of turns. The choice

in selection of a number of turns depends on available
slot area and conductor size. For high current range, the
conductor size required is high which limits the allowable
number of turns. An optimal solution is drawn between
a number of turns and current rating considering the
total allowable ampere turns. From initial dimensions,
iterations are performed using analytical calculations and
FEA tool which led to first design.
Ansoft Maxwell Electromagnetic suit software is used
for finite element analysis of machine. The flux-linkages (a) Flux density plot at unaligned position.
versus current characteristics for aligned and unaligned
positions are drawn as given in Fig. 1b. The flux dis-
tribution and flux density are important parameters in
SRM design. FEM helps to analyze the flux density in
the machine at different operating regions and conditions.
The B-H curve and core loss curves of material which
will be used for laminations of SRM core is modelled
in the Maxwell. This way the magnetic non-linearity has
been taken into consideration for the analysis.
B. Saturated SRM (b) Flux density plot at completely aligned position.
It is typical for SR machines that they operate in a Fig. 2: Flux density plots for proposed SRM.
lot higher saturation in comparison to other motor types.
Magnetic fields above 2T are very common in SR motors,
while it is normal that the magnetic data about material
doesn’t cover the whole range. Higher saturation leads to loss magnetic materials (Super Core) for stack lamina-
higher energy conversion ratio. For same power rating of tions.[10] These materials have high cost and also limited
SRM, highly saturated machine will have comparatively by availiability at the manufacturer. The other technique
lesser converter kVA requirement. This decreases size of is used to mitigate this issue. It is observed that direction
both converter and machine [8], [9]. of flux and frequency of change in flux is lesser in
stator pole compared to other part of machine. Hence,
High power density SRM will have higher flux den- designing a machine to accomodate higher flux densities
sities, this will lead to increase in core losses. This at poles (specially stator poles) will not increase core
issue can be accomodated by use of high efficient, low losses with comparative increase in power density.

Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on July 16,2024 at 05:39:35 UTC from IEEE Xplore. Restrictions apply.
 TABLE I: Specifications of proposed SRM.
Sr.No. Parameter Symbol Value Unit
1 Stator outer diameter Ds 190 mm
2 Rotor diameter Dr 109.6 mm
3 Stack length L 80 mm
4 Air Gap g 0.3 mm
5 No.of Stator poles Ps 8 -
6 No.of rotor poles Pr 6 -
7 No. of phases m 4 -
(a) Flux density plot at unaligned position.
8 Stator pole arc βs 21 degree
9 Rotor pole arc βr 23 degree
10 Stator yoke thickness Ys 14 mm
11 Rotor pole thickness Yr 14 mm
12 Shaft Diameter Dsh 30 mm
13 No. of turns per pole Tph 50 Turns
14 Phase Voltage V 320 Volts

(b) Flux density plot at completely aligned position.

Fig. 3: Flux density plot for 48 V SRM.

Fig. 2 and 3 shows the flux density plot of respective

machines for rotor at unaligned and aligned position.
Fig. 4 shows the output power vs speed and torque
vs speed characteristics of proposed machine at rated
current. It is interesting to see that the characteristics are
obtained from RmXprt are for constant lead angle and
pulse width in analysis setup (i.e. constant turn on and
turn off angle). While during control of actual machine
we change the lead angle and pulse width accordingly Fig. 4: Output power and Torque vs speed characteristics
to obtain required torque profile [11]. The fall in power for 320 V machine.
and efficiency indicates the falling torque region, during
which in actual practice advancing angle is added to the
excitation which incrases the power upto falling power 100
region as shown in Fig. 5. The Fig. 4 hence shows
the approximate characteristics of machine that can be 80
obtained by controlling the phase excitation angles, the
graph is obtained by performing simulations for few set
% Efficiency

60
of angle througout the working area. Detailed analysis for
optimum lead angle and pulse width for given machine
can also be done, which helps the control designer to set 40
the angle of advance and turn on/turn off angle. Such
analysis will be done in future. The specifications of the 20
proposed SRM are given in table I.
A comparison is done between proposed SRM and 0
the previously designed Induction motor having approxi-
mately same outer geometrical dimension. Table II shows 0 0.2 0.4 0.6 0.8 1
the comparison. Speed (Rpm) ·104

C. Combined Simulation using Ansys Workbench Fig. 5: Efficiency vs. Speed for proposed 320V SRM
A transient-transient link simulation was incorporated
between Maxwell and Simplorer. The converter and

Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on July 16,2024 at 05:39:35 UTC from IEEE Xplore. Restrictions apply.
 TABLE II: Comparison between Induction motor and Switched reluctance motor.
Flat effi-
Max.
Overall Axial Maximum Max. ciency
Power Torque current Type of
Machine diameter length speed Effi- over
(kW) (Nm) density cooling
(mm) (mm) (Rpm) ciency wide
(A/mm2 )
range
Air
SRM (320 V) 190 80 7.5 22 9000 91 85 5
cooled
Liquid
IM (320 V) 190 80 8.5 25 7000 90 87 9
cooled
Air
SRM (48 V) 185 75 3.5 9 8000 91 89 3.75
cooled
air
IM (48 V) 185 75 3 11 7000 90 86 4.75
cooled

Power
Power
Torque 15
3

10
Power

1 5

Fig. 7: Core loss waveform of 320 V SRM at 4000 Rpm

0 0
0 2,000 4,000 6,000 8,000
Speed (Rpm)
Fig. 6: Output Power and Torque vs. Speed for proposed
48 V SRM

power supply was modeled. The block diagram in Fig. 9 Fig. 8: Copper loss waveform of 320 V SRM at 4000
shows combined simulation approach. The phase excita- Rpm
tion current and output torque for a transient simulation
at 4000 Rpm are shown in fig 10a and fig. 10b. The
thermal and structural analysis are performed in order
III. M ECHANICAL C ONSIDERATION WHILE
to improve the design, Ansys Workbench is used for
FABRICATION
this purpose. The losses calculated by Maxwell 2D is
imported to Ansys Workbench and the heat generation The final design of the machine is prepared using 3D
is used to solve the steady state thermal analysis of 320 V drawing tool CATIA V5, layout sheets were prepared for
SRM at 3000 Rpm and 18 Nm torque output. Fig. 7 and 8 manufacturing.
show the core loss and copper loss curve for the proposed • Selection of shaft diameter and shaft material is
320 V SRM. The temperature analysis is further used to done through considering the maximum torque and
calculate structural deformations and stresses acting on speed range of the machine.
the machine. The simulation is performed for different • Bearing selection requires consideration of dynamic
heat transfer coefficients and results are analyses to find load carrying capacity and radial load acting on
best suitable cooling method to be used.The table III shaft which depends on drive arrangement.
shows the motor maximum temperatures at different • SRM requires both speed sensing and position sens-
convection methods. [12], [13] ing of the rotor pole with respect to stator in order

Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on July 16,2024 at 05:39:35 UTC from IEEE Xplore. Restrictions apply.
 (a) Phase currents in Simplorer transient analysis. (b) Torque output in Simplorer transient analysis.

Fig. 10: Combined simulation using Simplorer.

IV. F UTURE W ORK

The design is to be manufactured. Standard test pro-
cedures and test reports are to be prepared. Static and
dynamic test are to be done on manufatured machine
to validate the simulation results. The test will be done
to analyze the effect of saturation in SRM. Vibrational
analysis could also be performed on stator structure in
order to tune the control system for noise mitigation [14].
V. C ONCLUSION
A optimized SRM design has been worked out for
provided outer geometric dimensions with consideration
of the performance criteria. The two SRM are designed,
one is saturated and other low voltage unsaturated. The
proposed saturated SRM seems to have comparable per-
Fig. 9: combined simulation using Ansys Workbench. formance with respect to a water cooled IM with ap-
proximately same outer dimensions while SRM competes
Motor
with IM in some areas like wide speed range, maximum
Convection Method Temperature efficiency, rugged construction.
(◦ C) Co-simulation approach makes it easy for designing
Free convection 200
a machine for a particular application since we can
Forced convection; low speed flow of air
130 analyze both machine and control simultaneously. A
over a surface
Forced convection; moderate speed flow of
80
detailed control analysis will be done further with the
air over a surface use of simplorer. The multi physics simulation approach
Forced convection; with the use of thermal
conductive epoxy material
75 would be a very useful tool for designing motor for EV
application. Use of further multi domain analysis like
TABLE III: Typical values of motor tempurature under vibrational, acoustic, etc could help to minimize the noise
steady state condition obtained from thermal analysis. of the machine.
ACKNOWLEDGMENT
to control the machine, Angle position sensor(Hall This work was supported by Engineering and Research
sensor) with bearing sensors will be used for this Center, Tata Motors Pune and Department of Electrical
purpose. Engineering CoEP.
• Winding temperature sensors are to be placed in
order to sense the overload fault in winding. R EFERENCES
• Bearing temperature sensors are required to detect [1] J. De Santiago, H. Bernhoff, B. Ekergård, S. Eriksson, S. Ferha-
rise in temperature of bearing in order to avoid tovic, R. Waters, and M. Leijon, “Electrical motor drivelines in
commercial all-electric vehicles: a review,” Vehicular Technology,
bearing failure. IEEE Transactions on, vol. 61, no. 2, pp. 475–484, 2012.

Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on July 16,2024 at 05:39:35 UTC from IEEE Xplore. Restrictions apply.
 [2] R. Krishnan, Switched reluctance motor drives: modeling, simu-
lation, analysis, design, and applications. CRC press, 2001.
[3] K. M. Rahman, B. Fahimi, G. Suresh, A. V. Rajarathnam,
and M. Ehsani, “Advantages of switched reluctance motor ap-
plications to ev and hev: design and control issues,” Industry
Applications, IEEE Transactions on, vol. 36, no. 1, pp. 111–121,
2000.
[4] T. Kakishima, K. Kiyota, S. Nakano, and A. Chiba, “Pole
selection and vibration reduction of switched reluctance motor for
hybrid electric vehicles,” in Transportation Electrification Asia-
Pacific (ITEC Asia-Pacific), 2014 IEEE Conference and Expo,
pp. 1–4, IEEE, 2014.
[5] M. Sanada, S. Morimoto, Y. Takeda, and N. Matsui, “Novel rotor
pole design of switched reluctance motors to reduce the acoustic
noise,” in Industry Applications Conference, 2000. Conference
Record of the 2000 IEEE, vol. 1, pp. 107–113, IEEE, 2000.
[6] P. Rafajdus, A. Peniak, D. Peter, P. Makys, and L. Szabo,
“Optimization of switched reluctance motor design procedure for
electrical vehicles,” in Optimization of Electrical and Electronic
Equipment (OPTIM), 2014 International Conference on, pp. 397–
404, IEEE, 2014.
[7] S. Faid, P. Debal, and S. Bervoets, “Development of a switched
reluctance motor for automotive traction applications,” in 25th
World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium,
Shenshen, China, 2010.
[8] T. Miller, “Electronic control of switched reluctance motors,” ser.
Newnes Power Engineering Series. Oxford, UK: Newnes, 2001.
[9] A. E. Fitzgerald, C. Kingsley, S. D. Umans, A. E. Fitzgerald,
A. E. Fitzgerald, and E. Engineer, Electric machinery, vol. 5.
McGraw-Hill New York, 2003.
[10] A. Chiba, Y. Takano, M. Takeno, T. Imakawa, N. Hoshi, M. Take-
moto, and S. Ogasawara, “Torque density and efficiency improve-
ments of a switched reluctance motor without rare-earth material
for hybrid vehicles,” Industry Applications, IEEE Transactions
on, vol. 47, no. 3, pp. 1240–1246, 2011.
[11] K. Kiyota and A. Chiba, “Design of switched reluctance motor
competitive to 60-kw ipmsm in third-generation hybrid electric
vehicle,” Industry Applications, IEEE Transactions on, vol. 48,
no. 6, pp. 2303–2309, 2012.
[12] M. V. J. R. S. Cassiano A. Cezario, Briam C. Bork, “Robust elec-
tric machine design through multiphysics,” in Ansys Advantage
Volume VIII Issue 2, pp. 11–15, Ansys Inc., 2014.
[13] C. Kokkinos, “Magnetic appeal,” in Ansys Advantage Volume VIII
Issue 2, pp. 15–18, Ansys Inc., 2014.
[14] W. Cai, P. Pillay, Z. Tang, and A. M. Omekanda, “Low-vibration
design of switched reluctance motors for automotive applications
using modal analysis,” Industry Applications, IEEE Transactions
on, vol. 39, no. 4, pp. 971–977, 2003.

Authorized licensed use limited to: National Institute of Technology- Delhi. Downloaded on July 16,2024 at 05:39:35 UTC from IEEE Xplore. Restrictions apply.