IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 2, NO.

2, JUNE 2014

329

A Sizing Methodology of the Synchronous

Reluctance Motor for Traction Applications
Seyedmorteza Taghavi, Student Member, IEEE, and Pragasen Pillay, Fellow, IEEE

Abstract Synchronous reluctance machines (SynRM) have

been used widely in industries for instance, in ABBs new VSD
product package based on SynRM technology. It is due to their
unique merits such as high efficiency, fast dynamic response,
and low cost. However, considering the major requirements for
traction applications such as high torque and power density, low
torque ripple, wide speed range, proper size, and capability of
meeting a specific torque envelope, this machine is still under
investigation to be developed for traction applications. Since the
choice of motor for traction is generally determined by manufacturers with respect to three dominant factors: cost, weight, and
size, the SynRM can be considered a strong alternative due to its
high efficiency and lower cost. Hence, the machines proper size
estimation is a major step of the design process before attempting
the rotor geometry design. This is crucial in passenger vehicles in
which compactness is a requirement and the size and weight are
indeed the design limitations. This paper presents a methodology
for sizing a SynRM. The electric and magnetic parameters of the
proposed machine in conjunction with the core dimensions are
calculated. Then, the proposed methods validity and evaluation
are done using FE analysis.
Index Terms Electric vehicle, sizing method, synchronous
reluctance, synchronous reluctance machine (SynRM), traction
application.

N OMENCLATURE
Te
b
Po
Vo
r
Tep
Teb
Tem
Con
Ns

Electromagnetic torque.
Base speed.
Number of pole pairs.
DC-bus voltage.
Maximum speed.
Peak torque in continuous mode.
Continuous torque at base speed.
Maximum torque developed over (0 - b ).
Stator connection (, Y ).
Number of stator slots.

Manuscript received July 15, 2013; revised October 16, 2013; accepted
December 17, 2013. Date of publication January 9, 2014; date of current
version April 4, 2014. This work was supported in part by Concordia
University, in part by the Natural Sciences and Engineering Research Council
of Canada, in part by Hydro-Qubec, and in part by Connect Canada.
Recommended for publication by Associate Editor J. O. Ojo.
S. M. Taghavi is with the Department of Electrical and Computer Engineering, Concordia University, Montreal, QC H3G 1M8 Canada (e-mail:
s_tagh@encs.concordia.ca).
P. Pillay is with the Department of Electrical and Computer Engineering,
Concordia University, Montreal, QC H3G 1M8 Canada, and also with the
University of Cape Town, Cape Town 7701, South Africa (e-mail: pillay@encs.concordia.ca).
Color versions of one or more of the figures in this paper are available
online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/JESTPE.2014.2299235

ns
q
Vph
Kvo
B1gm
B1dm
B1qm
o

Dro
L s
Ld
Lq

Rs

co
ir
lf
lc
s
Kfill
J
Aco
Pco
Wco
Wir
Pir
Pmec
Dri
Dso
Dsi
L

s
As
y
g

Kc
Ks
Kdm1
Kqm1
Kw1
Ky

Number of conductor turns per slot.

Number of slots / pole / phase.
Phase voltage.
Converter factor.
Maximum air gap flux density,
amplitude of the fundamental.
d-axis component of B1gm .
q-axis component of B1gm .
Air permeability.
Saliency ratio.
Rotor outer diameter.
Leakage inductance.
D-axis inductance.
Q-axis inductance.
Efficiency.
Stator phase resistance.
Copper resistivity.
Copper weight density.
Iron weight density.
End winding length.
Coil length.
Slots permeance.
Fill factor.
Current density.
Copper conductor cross section area.
Copper loss.
Copper weight.
Iron weight.
Total iron losses.
Mechanical losses.
Rotor inner diameter.
Stator outer diameter.
Stator inner diameter.
Stake length.
Pole pitch.
Stator slot pitch.
Useful area of slot.
Chorded coil pitch.
Air gap length.
Stack aspect ratio (L/ ).
Pole pitch to air gap ratio ( /g).
Carter factor.
Saturation factor.
D-axis magnetic coefficient corresponds to B1dm .
Q-axis magnetic coefficient corresponds to B1qm .
Winding factor.
Chording factor.

2168-6777 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

330

Ldm
Lqm
Lm
q
t
is
Pmo
Idm
Iqm
P

e
Em
Rcr
Rcs
Vd
Vq
Id
Iq
dm
qm
d

IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 2, NO. 2, JUNE 2014

Magnetizing inductance in d-axis.

Magnetizing inductance in q-axis.
Magnetizing inductance.
Flux in q-axis.
Time.
Stator phase current.
Continuous output power.
Magnetizing current in d-axis.
Magnetizing current in q-axis.
Number of poles.
Shaft speed.
Electrical speed.
Electromagnetic motive force (emf).
Equivalent resistance of rotor core losses.
Equivalent resistance of stator core losses.
D-axis voltage component.
Q-axis voltage component.
D-axis current component.
Q-axis voltage component.
Magnetizing flux in d-axis.
Magnetizing flux in q-axis.
Flux in d-axis.
I. I NTRODUCTION

HE WORLDWIDE trend in reducing fossil fuel consumption and air pollution refers to more applications
of electric motor drives in automotive industry, particularly
ac drives as traction. Recent advantages of high quality magnetic materials, power electronics, and microcontrollers have
contributed to new energy efficient and high performance electric drives that use new electric motor technologies. In general,
electric motors in power train applications need to meet major
requirements that can be summarized as follows [1][3]:
1) high torque and power density;
2) high torque at low and high power at high speed;
3) wide speed range;
4) fast dynamic response;
5) high efficiency;
6) reliability and robustness.
The synchronous reluctance solution represents a possible
alternative to ac drives [4]. This machine can be applied
to many applications where permanent magnet synchronous machines (PMSM) and induction machines (IM) are
used. Some of its characteristics such as high efficiency
and fast dynamic response are also dominant in traction
applications.
In principle, the synchronous reluctance machine (SynRM)
is similar to the traditional salient pole synchronous motor
but does not have an excitation winding in the rotor. In this
machine only the rotor is constructed by barriers and segments
(Fig. 1). The main advantage of the SynRM relies on the
absence of rotor copper losses that permits continuous torque
that can be higher than an IM of the same size. The other
important features of the SynRM are as follows [5], [6]:
1) the rotor is potentially less expensive than both PMSM
and IM due to cancelling cage, winding, and magnets
from its structure;

Fig. 1.

Structure of the TLA, type rotor of the SynRM.

2) the torque per ampere is acceptable and unlike the

PMSM and IM it does not depend on the rotor
temperature;
3) the control system is simpler than that of the field
oriented IM drives. However, rotor position information
is necessary.
Over the past decade, developments in machine design
methodology, high quality magnetic materials, and advanced
power electronics allow the machine designers to enhance
the performance of the SynRM. Many investigations reported
in the literature are mostly focused on performance optimization by rotor geometry [7][12], control strategy
[11], [13], [14], and comparison of SynRM with different machines [8], [15], [16]. Considering todays compact
passenger vehicles in which the size and weight are the
design limitations, the size estimation that leads to the other
electrical, magnetic, and mechanical parameters is crucial.
This paper presents a sizing method of the SynRM for
traction applications. The method criteria are based on the
machines mathematical model, equivalent circuit, initial data,
and assigned parameters. A specific torque-speed envelope,
which would be provided by manufacturers driving cycle
data supports the whole process dominantly. The electric and
magnetic parameters of the proposed machine, in conjunction
with the core dimensions, are calculated. Finally, the proposed
methods validity and evaluation are done using FE analysis.
II. M ETHOD A LGORITHM
The sizing process of the SynRM refers to electric,
magnetic, and thermal aspects. In traction applications, to
develop the required torque in critical times such as startup and maximum power demand at maximum speed, the
machine needs to be designed for high current density and
high frequency operations. Hence, the heat caused by copper
or iron losses is a concern that cannot be transferred by
air cooled system. Since, the cooling system of the electric
machine is supplied through the vehicles cooling system; the
thermal issue depends on both machine and vehicle cooling
system capacity. The stator outer surface area and volume are
two parameters that lead to the machine cooling capacity at a
certain operating temperature (120 C150 C). Practically, by
selecting the proper number of cooling jacket around the outer
surface of the stator, the cooling system would be capable of
cooling down the machine in continuous or transient mode of

TAGHAVI AND PILLAY: SIZING METHODOLOGY OF THE SYNCHRONOUS RELUCTANCE MOTOR

331

TABLE I
I NITIAL D ATA

Fig. 2. Flow diagram of the sizing procedure consists of initial data, assigned
parameters, torque profile, and FEA.

operations properly. Therefore, in this design procedure after

meeting the torque and power capability, the vehicle cooling
system is supposed to cool the machine properly. On the
other hand, the electromagnetic design strongly depends on
the application. Therefore, it is necessary to perform the whole
procedure of sizing with taking into account the maximum
output power, torque-speed profile, initial data, and assigned
parameters. The proposed method algorithm may be explained
by the flow diagram of Fig. 2.
The sizing process of the proposed machine begins with
preparing initial data and assigning the key parameters. These
will be used in calculations of geometric, magnetic, and
electric parameters in conjunction with the analytical model
of the machine. In this step, the predesign output parameters
will be the stator geometry and rotor outer diameter. A new
iteration with updated assigned parameters will be done if the
geometries such as stack length or stator outer diameter cannot
satisfy the design requirements. Otherwise, the electrical,
magnetic, and geometrical parameters, i.e., stator geometry,
winding specifications, inductances, saliency ratio, maximum
torque, etc., are characterized. Finite element software is called
to analyze the machines performance related to the output
functions such as maximum developed torque, torque quality,
and power as well as magnetic characteristics, e.g., flux density
and magnetic vector potential. The process ends if the finite
element results satisfy the design requirements. Otherwise, the
process is repeated by updating the assigned parameters such
as magnetic and current loading, stack aspect ratio, and pole
pitch to air gap ratio to obtain the proper size.
The sequence of how the flow diagram is executed for
the specific design example (reference machine, Table I) is

Fig. 3.

Torque envelope of the power train.

TABLE II
A SSIGNED PARAMETERS

described in different sections of this paper, which can be

summarized as follows:
1) In Section III, the design requirements are set, which
includes the torque envelope (Fig. 3), the initial data
correspond to the reference machine (Table I), and the
assigned parameters (Table II). These are loaded to the
software.
2) In Sections IV and V, calculations are performed by
machines equations that have been already coded, using
the initial data and preliminary assigned parameter to
target the desired size. After several iterations and updating the parameters such as stack aspect ratio (with final
value of 0.9), pole pitch to air gap ratio (with final value
of 345), etc., the machines pole pitch, slot pitch, rotor

332

IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 2, NO. 2, JUNE 2014

TABLE III
D ESIGNED S PECIFICATIONS

Fig. 4.

Desired performance characteristic of the motor.

A. Initial Data

outer diameter, stator geometry, stack length as well as

its corresponding assigned parameters are realized (refer
to Table III).
3) In Section VI from VI-A to VI-H, the size is satisfied,
then the algorithm goes to the next step that is to identify
the machines whole electrical and magnetic parameters
using the geometric data from the previous step and
machines equations. The results are shown in Table III
as well.
4) In Section VII, the designed machines performance
needs to be evaluated corresponding to the parameters
that are characterized analytically. Hence, the finite
element analysis (FEA) process is used and the FE
results evaluate the design (Figs. 816). Therefore,
the sizing algorithm can be ended at this step. In this
simulation, every iteration takes about 25 min after
updating parameters and usually the entire sizing
procedure ends after 1015 iterations.
III. A NALYTICAL P RESIZING AND C ALCULATIONS
As per the method algorithm shown in Fig. 2, the first step of
the sizing process is to analytically determine the solution
of the proposed machine. This step leads to the identification
of the stator geometry, rotor outer diameter, machines electrical, and magnetic parameters, i.e., inductances, resistances,
winding specifications as well as the output functions such
as maximum developed torque, power factor, and efficiency.
The proposed approach uses the machines analytical model,
i.e., the equivalent circuit, vector diagram, and mathematical
equations in conjunction with initial data, desired torque
envelope, and assigned parameters. The torque envelope of
the power train (Fig. 3) and its corresponding electric motor
desired performance characteristic (Fig. 4) are shown in which
the base and maximum speed have important effects on the
machines parameters [17].

The Initial data is provided based on the torque-speed

envelope (Fig. 3) and requirements. In general, only the
continuous torque-speed mode of operation (Fig. 3) is imposed
on the machine design, but in traction applications the motor
needs to take over the high current demand in the intermediate
mode of operation up to base speed, in which the maximum
torque needs to be developed during a critical time such as
start up time (about 30 s). Table I shows the initial data in
which the key parameters will be used in further calculations.
B. Assigned Parameters
Some parameters must be arranged during calculations.
Some of them can be modified during the design process [17].
1) Converter Factor (Kvo ): is due to the limitation of
available maximum voltage through the VSI. In general,
Kvo is adjusted between 0.9 > K vo < 1. In this paper,
Kvo = 0.98.
2) Magnetic Loading (B1gm ): is selected based on the core
type to provide enough flux density without unwanted
saturation. In this paper, the core type is M15 and
B1gm = 1 T.
3) Pole Pitch to Air-Gap Ratio (/g): this ratio has a strong
effect on the magnetizing coefficients (Kdm1 , Kqm1 ). It is
obtained during sizing algorithm for a sufficient value
of saliency ratio to satisfy the geometrical and torque
envelope requirements. In this paper, = 10 and the
obtained value of /g is (345).
Equations (1)(3) show the relations as follows:
L dm = L m K dm

(1)

L qm = L m K qm
L dm + L s
Ld
Then; =
=
Lq
L qm + L s

(2)
(3)

where d-axis is the direct and q-axis is the quadrature

axis of rotor reference frame and Ldm and Lqm are their
corresponding magnetizing inductances, respectively.
4) Stack Aspect Ratio ( = L/ , 0.6 < < 3): this ratio
leads to the machine stack length. The higher ratio refers

TAGHAVI AND PILLAY: SIZING METHODOLOGY OF THE SYNCHRONOUS RELUCTANCE MOTOR

to longer stack length and shorter pole pitch that is

suitable for higher speed applications. In this paper,
= 0.9.
5) Current Density (J): It can be selected from
3.5 to 8 A/mm2 for axial or axial-radial air cooled
machines [18]. However, in traction applications, by
employing an advanced cooling system, a higher value
would be applicable. This system is composed of
cooling jacket located around the surface of the stator,
coolant such as water-glycol or oil, cooling pump,
and temperature control devices that is supplied via
vehicles cooling system. According to this, the stator
outer surface area and volume are two parameters
that lead to the machine cooling capacity at certain
operating temperature (120 C150 C). In this paper,
J = 10 A/mm2 is assigned for continuous mode of
operation. It is increased to twice in peak torque.
A summary of the assigned data is shown in Table II.
IV. ROTOR C ORE S IZE
To obtain the rotor outer diameter machines pole pitch is
the leading parameter. Equations (4)(6) show this approach
[17][19]
B1gm
(4)
B1dm = 


K qm1 2 L d
1 + K dm1 L q

 


K dm1 o g L

3



(5)
= T
em
K
Ld
B1dm 2 Po 2 1 K qm1
K
+
K
(1
)
c
s
Lq
dm1

Dro = 2Po

L =
(6)
where the parameters are defined in Table II.
In the above-mentioned equations, it is illustrated that the two
key parameters are Kdm1 and Kqm1 . Since the flux leakage
is unknown, it may be necessary to assume the leakage
inductance (L ) is approximately equal to the q-axis magnetizing inductance (Lqm ). This assumption helps to simplify the
solution process. It means that even if the actual Lqm is higher
than the calculated one the torque and machines performance
will remain approximately the same [17]. Therefore, it is still
on the safe side of the design.
Considering this assumption along with substituting the
assigned saliency ratio ( = 10) in (3) gives K dm1 /K qm1 = 19
and then Kdm1 = 0.95. These two key parameters can be
substituted again into (4)(6) that consequently give the slot
pitch and the rotor outer diameter. On the other hand, the
machine pole pitch, pole pitch to air gap ratio, and stack aspect
ratio give the air gap and stack length, respectively.
The reason behind assigning the saliency ratio of 10 is
that, from the machine performance point of view, this value
is at the boundary of the two rotor types [axially laminated
anisotropic (ALA), transversally laminated anisotropic (TLA)]
(Fig. 1). Previously, it was reported in [19] and [20] that the
saliency ratio higher than 10 can be mostly achieved by ALA
rotor, whereas the TLA rotor presents 10. However, the

333

TLA type is preferable in many applications due to its simple

manufacturing and lower iron losses. In the TLA type rotor
(Fig. 1), the laminations are identical and punched or cut in
the traditional way. Some thin connections which are called
rib connect the ends of the segments to each other axially and
transversally. These connections maintain enough mechanical
integrity in the rotor structure against rotational forces in high
speed operations.
In this paper, achieving the desired torque profile while
considering the assigned saliency ratio, proper stack length,
stator outer diameter, and converter current limitations is the
key in selecting the rotor type and designing the machines
geometry.
V. S TATOR G EOMETRY
Electromotive forces for all coils, which are distributed
along the pole pitch are not the same. Therefore, the torque
includes all odd harmonics. In general, chorded pitch winding
is a solution to improve the torque quality. However, it has
been reported in [21] that, chording of the stator winding
in SynRM is not necessary to reduce the amplitude of the
lower order MMF space harmonics when the round rotor with
uniformly internal flux barriers is used because the lower order
harmonic generated fluxes are attenuated by the flux barrier
rotor. Furthermore, it reduces the power rating by 5%10%.
Therefore, chorded winding is neglected and this eliminates
the chording factor, K y , in (7), which gives the machines
winding factor

   
Ns

. (7)
K w1 = sin
q sin
K y Where; q =
6
6q
3(2P o )
A. Ampere-Turns Per Slots
The flux in machines magnetic circuit is deemed to be
proportional to the resultant ampere-turns. The magnetic
loading is arranged such that the machine operates in the
linear portion of the magnetization characteristic. Therefore,
the major part of the exciting ampere-turns is used to maintain
the proper flux density across the air gap. Equations (8)(10)
show approaches to obtain the d, q, and total ampere-turn/slot,
respectively, [17]
B1dm g K c (1 + K s )

3 2o q K w1 K dm1

Ld
n s Iqm = (n s Idm )
Lq

n s Im = (n s Idm )2 + (n s Iqm )2

n s Idm =

(8)
(9)
(10)

where the parameters are defined in Table II.

B. Stator Slot Dimensions and Core Size
To design the stator core geometry, all dimensions of the
stator slot need to be obtained, Fig. 5(a) and (b) show the
details.

334

IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 2, NO. 2, JUNE 2014

characterization of all the required parameters as a function

of the number of conductor turns per slot (n s ) [17], [18].
Therefore, (n s ) is one of the key parameters in the design.
A. Leakage Inductance
The leakage inductance can be expressed as

(n s q Po )2 
s + z + f L
Where;
(19)
Po q
h s1
2h s
2h s2
s
+
(slot permence)
+
=
3(bs1 + bs2 ) bs1 + bss
bss
5g
(air gap coefficient)
z =
5bss + 4g
0.34q
f =
(l f 0.64 ) (end winding length coefficient)
L

(stator end winding length).

(20)
lf =
2
Substituting all data into (19) to (20) yields the leakage
inductance as
L s = 2o

Fig. 5. Typical geometry and dimensions of the stator slot (a) and wedge
area (b).

The ampere-turns per slot and pole pitch allow calculating

the useful area and slot pitch through
n s Im
J K fill

.
s =
3q

As =

(11)
(12)

Equations (13)(15) yield the dimensions of each slot as

follows:
B1gm B1gm
bt s =
=
s
(13)
Bt s
2B 1dm
Dro + 2 (h s1 + h s2 )
bs1 =
bt s
(14)
3 2Po q
(15)
h s = As /bs1.
To reduce the air gap flux pulsations, it is quite necessary
to select minimum possible dimensions for the slot opening.
Therefore, in this paper, bss = 3, hs1 = 1, and hs2 = 2 mm.
On the other hand, bs1 and bs2 can be identical if the flat
magnetic wire (wire with rectangular cross section) is selected.
Therefore, the fill factor and heat transferring capability from
winding to the stators body will be increased. According
to the wire gage standards, if the calculated gage of the flat
wire is not available, some modifications might be applied
to the slots dimensions and the closest gage can be selected.
In this condition, the new slots area has to be very close to the
calculated useful area to maintain the assign current density.
Stator back or yoke width can be obtained through (16)
in which the yoke flux density, Bss, is considered to be 70%
higher than the air gap flux density
h ss =

B1gm
.
Bss

(16)

Now, all the parameters that are required to calculate the stator
geometry are found (Fig. 5). Therefore. the stator inner and
outer diameter can be found using
Dsi = Dro + 2g
Dso = Dsi + 2(h s1 + h s2 + h s + h ss ).

(17)
(18)

VI. M ACHINE PARAMETERS

To analyze performance of the proposed machine by FE,
its electric and magnetic parameters need to be identified.
According to the approaches outlined below, the machines
dimensions, initial data, and assigned parameters lead to the

L s = 8.5518 106 (n s )2 (H)

(21)

B. Stator Phase Resistance

Equation (22) shows the phase resistance as a function of
ns in which = 2.3 108  m is the copper resistivity
Rs = lc Po q(n s )2

J
n s Im

(22)



where lc = 2 L + l f . Therefore, this can be expressed as
phase resistance as
Rs = 3.292 103 (n s )2 ().

(23)

C. Machine Inductances
For uniform air gap, the magnetizing inductance can be
expressed as
L m = 6o L

(Po q K w1 n s )2
.
2 Po g K c (1 + K s )

(24)

Therefore, the magnetizing inductance and its dq components

can be defined as a function of ns as follows:
L m = 7.3211 104 (n s )2 (H)
L qm = 3.6605 105 (n s )2 (H)
L dm = 6.955 104 (n s )2 (H).

(25)

To simplify the solution process, the q-axis magnetizing inductance is considered equal to the leakage inductance. However,
it is always smaller. Therefore, the dq axis inductances of
the machine can be expressed as
L dm + L qm
L d = L dm + L s =
L q = L qm + L s
= 2L qm
L d = 7.3211 104 (n s )2 (H)
L q = 7.3211 105 (n s )2 (H).

(26)

TAGHAVI AND PILLAY: SIZING METHODOLOGY OF THE SYNCHRONOUS RELUCTANCE MOTOR

335

E. Electromagnetic Torque and Speed Range

Fig. 6.

Single phase equivalent circuit of the SynRM includes iron losses.

The maximum torque developed by the electric machine is

a major requirement in traction applications. This parameter is
basically defined in torque envelope of the power train (Fig. 3).
Hence, it is expected to be developed by electric machine
and remain constant over the full speed range. This can be
expressed as


Tem = 1.5Po L d L q Idm Iqm .

Fig. 7.

Vector diagram of the SynRM at base speed.

D. Number of Turns Per Slot

The number of turns per slot depends on the available
voltage across the stator phases. It is varied based on the stator
connection and the converter type. Since in this paper, the
machine has a delta connection and is assumed to be supplied
by a SVPWM power converter, the supply voltage can be
obtained as

6
K vo Vo .
(27)
Vph =

To identify ns , the machine equivalent circuit and its

relevant vector diagram at base speed are implied to define
the machines mathematical model through voltage equations.
Figs. 6 and 7 show the equivalent circuit including the iron
losses and the vector diagram of the SynRM, respectively.
Considering the machines vector diagram, the voltage equations can be expressed as

Vph =

Vd2

Using the required data from Table III yields Tem =

324 Nm. Since the proposed machine is supposed to be
designed to develop 300 Nm in intermediate mode of operation
at base speed and over the speed range as well, the difference
between the calculated and assigned torque can be adjusted
by supply voltage and phase current. On the other hand, it
may be necessary to make another adjustment on the supply
voltage with regards to the approximation on calculated n s .
Therefore, the new phase voltage and current that can be used
in further calculations are

10 300

Vph = 384.3
= 462.2 (V )
9.54
 324
300
= 53.4 (A).
Im = 66.75
324
The maximum speed range, the corresponding phase current,
and its d q components can be expressed as follows (the
results are included in Table III):




Vrph 2 3Po L d L q
r
=
(30)
b
be
2L d L q Tem

Tem L q
b


Idph =
3Po L d L q L d r
Ld
Iqph = Idph
Lq

Iph = Idph 2 + Iqph 2 .

(31)

F. Power Factor and Efficiency

The product of power factor and efficiency is a leading
parameter of the SynRM to justify the machines performance
with respect to speed variations

Vd = be L q Iqm + Rs Idm
Vq = be L d Idm + Rs Iqm
2
(Electrical Speed)
be = b Po
60

(29)

(28)

Vq2 .

Substituting the values of d-q ampere-turns from (8) and

(9) into (28) yields the number of turns of the conductor
per stator slot, which is our key in identifying all machines
required parameters that were previously characterized as a
function of n s ; n s = 9.54 turns, that can be approximated
as n s = 10 turns. Table III shows a summary of the design
specifications of the proposed machine that were obtained
through the design process based on the sizing method.

cos =

Te i
.
3Po Vrph Im

(32)

As is illustrated in (32), the efficiency and power factor

depend on each other. Their product remains constant at any
operating speed (i ) and its corresponding electromagnetic
torque (Te ). This means improving one can reduce the other.
Therefore, optimum values for both parameters can optimize
the machine performance with regards to output power and
losses and lead the designer to select the machine control
strategy of maximum power factor or maximum efficiency
techniques accordingly [14], [22], [23].

336

IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 2, NO. 2, JUNE 2014

G. Copper Loss and Total Copper Weight

The copper loss is maximized at base speed because the
maximum current is available for the machine. The copper loss
and the total copper weight of the machine can be calculated,
respectively, (considering = 8900 kg/m3 as the weight
density of copper) as follows (the results are included in
Table III):
Pco = 3Rs (Im )2
Wco = 3lc Aco Po n s qco .

TABLE IV
PARAMETERS C OMPARISON

(33)

H. Stator Iron Weight and Losses

The weight of stator teeth (Wts ) and yoke (Wys ) can
be calculated as in (34), respectively, (Fig. 5) (considering
ir = 7600 kg/m3 as the weight density of iron)


bs1 + bs2
Wt s = Ns [bt s h s + h s2 s
+ h s1 (s bss )]Lir
2
2

Dsi
+ h s1 + h s2 + h s + h ss
W ys = [
2

2
Dsi
+ h s1 + h s2 + h s Lir

(34)
2
Wir = Wt s + W ys stator total iron weight.
The iron losses can be calculated as follows:
Pir = Wt s (P2 )60H z + W ys (P1.7 )60H z


B max 2 f 2
Where: PB max
. (35)
= (P1.5 )60H z
1.5
60
Since, the base speed is 1800 r/m at 60 Hz, considering the assigned data, Bts = 2B1gm and Bss = 1.7B1gm
and M15 specifications as well (P1.5 )60H z = 7.063 W/kg
(P2 )60Hz = 12.56 W/kg and (P1.7 )60Hz = 9.07 W/kg, which
leads to the machine iron losses at base speed (the results are
included in Table III).
VII. P ROPOSED M ETHOD VALIDITY AND E VALUATION
The dimensions, electrical, and magnetic parameters of a
45 kW SynRM were identified based on the proposed sizing
methodology. To validate and confirm the designed machines
size as well as evaluation of its performance a comparison
with a previously examined machine with the same characteristics as well as the 2-D FEA using designed parameters are
performed, as described below.
A. Comparison of the Designed and Examined Machines
The performance of a 45-kW SynRM with the same characteristics, e.g., rated power, maximum torque, pole pairs,
and number of stator slots has been examined and reported
previously in [21]. Table IV shows the tested machines
specifications (reference machine) compared with the parameters of the machine, which is designed using analytical
and FE method trough the sizing algorithm. The volume ratio
of two machines, Vratio = 15.7/15.9, indicates the size of
the designed machine is extremely close to the reference
machine (Vratio = 0.987). Furthermore, 2-D, i.e., stator outer

diameter and stack length that may be the major limitations

of the machine size in todays compact EVs and HEVs, meet
the requirements of our reference machine sufficiently, this
confirms the validity of our proposed method.
The optimum geometries identified through this method
are not unique and the proposed method is quite flexible
to adapt to proper initial data, e.g., stack aspect ratio and
pole pitch to air gap ratio, to introduce the machine with
longer stack length or larger bore diameter while the stack
volume remains constant. Besides geometric limitations, these
two parameters are important in traction applications due to
their effects on the machines electric and dynamic response.
The longer length refers to faster dynamic response and lower
starting torque whereas the larger bore diameter exhibits higher
starting torque and slower dynamic response. A considerable
difference is indicated between the rotors outer diameters of
two machines that distinguish the design for traction from
industrial applications. This is due to higher current density
(10 A/mm2 ) and magnetic loading (1T) assigned for the
proposed machine. These lead to shorter teeth and smaller
yoke size and consequently, larger rotor diameter and higher
torque production at the same phase current. Currently, electric
power trains equipped with advanced cooling systems, allow
electric machines to operate with heavy current loading to
maintain the maximum torque and power requirements in
different modes of operation.
B. FE Analysis and Results
The FEA is necessary to verify and evaluate the proposed
machines performances toward different modes of operation.
The geometry of the machine, which is designed through the
proposed method has been modified properly to be used in
developed FE software. Since the rotor optimization techniques are not subjected to this paper, a proper geometry of
the TLA-type rotor [7], [10], [12], [24][28] with four layers
was adapted. The geometry of the rotor lamination has been
modified several times to meet the maximum possible saliency
ratio and output torque and then, the proposed machine with
proper rotor geometry was examined by FE to evaluate its
performance.
The simulations are done using current source operation to
realize the machine performance at different loads. As opposed
to voltage source operation, this eliminates the effects of

TAGHAVI AND PILLAY: SIZING METHODOLOGY OF THE SYNCHRONOUS RELUCTANCE MOTOR

Fig. 8.

337

Circuit diagram of the current source simulation.

Fig. 11. Torque-current profile of the proposed SynRM with different air
gap lengths.

Fig. 9. Torque-angle characteristics of the proposed SynRM in intermediate

mode.

Fig. 10. D-axis flux density and distribution at base speed (continuous mode).

current harmonics on the output torque results in a torque

ripple caused by the ununiformed air gap that is only dictated
by the rotor geometry. The circuit diagram of the simulation
is shown in Fig. 8 in which (I1, I2, I3) are the current
source inverter supplies, (R1, R2, R3) are the end winding
resistors, (L1, L2, L3) are the end winding inductances, and
(B1, B2, B3) are the magnetizing coils including resistance
and inductance.
Fig. 9 shows the torque profile versus rotor position of
the proposed SynRM and Fig. 10 shows its d-axis flux
distribution. The machine is simulated under direct current
control while the DC-bus voltage is 600 V and the phase
current and angle (i in Fig. 7) are adjusted at 220 A and
32 degree electrical, respectively. Refer to Fig. 7, the relation
between ( ), (i), and the rotor angle () can be described as

(/2 + = + i). This gives ( ) and leads to the torque characteristic while considering the rotor angle, which can
be obtained through FEA ( = 10 when i = 32 electrical).
If the ( ) is required, it can be calculated as: = /232+10
therefore, = 68. Compared with ( ), (i) is easier to be
controlled during the FEA simulation because, information of
the d- and q-axis parameters are not required. As it is shown in
Fig. 9, the machine equipped with the proper rotor is capable
of developing assigned torque (300 nm) at the base speed
with low ripple (<12%). This situation is compatible with
the intermediate mode of operation in required torque profile
(Fig. 3). Since the machine is supposed to operate at low
speed with maximum torque, the dominant loss would be the
stator copper loss that imposes a high temperature shock to the
stator winding. In traction applications, an advanced cooling
system, which is composed of a coolant such as water-glycol
can transfer heat from stator body sufficiently and regulate the
temperature at its rated.
These results confirm that the SynRM, which is characterized by the proposed algorithm is capable of meeting the
torque profile.
Fig. 11 shows that the performance of the SynRM is
sensitive to air gap length. To achieve a higher T/A (torque per
ampere), a shorter air gap length is required. However, higher
air gap length is limited by the converter rating because the
machine needs to draw more current at the same torque. In this
figure, the designed SynRM is able to satisfy the desired torque
envelope (Fig. 3) with air gap length of 0.6 mm.
As it was previously mentioned, from traction application
point of view, the maximum developed torque at base speed
and the maximum output power at maximum speed are two
crucial operating points.
Fig. 12 shows the machine input power characteristic
beyond the maximum speed (1800 1.85 = 3300 rpm,
b = 1800 rpm). According to (30), the speed range is
1.85 within which a constant power operation using vector control technique is expected. Low speed range of the
SynRM compared with PMSM, which is due to reduction of
both saliency ratio and power factor beyond designed maximum speed (3330 rpm), causes considerable power reduction.
Therefore, to deliver output power of 45 kW at maximum

338

Fig. 12.

IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 2, NO. 2, JUNE 2014

Input power profile of the proposed SynRM.

Fig. 14. Magnetic characteristics of the proposed SynRM characterized with
three different rotor positions.

Fig. 13.

Torque-current angle (i) characteristic of the proposed SynRM.

speed (5000 rpm) in continuous mode the machine must be

designed with higher power rating (63 kW). This is another
reason why the SynRM would be larger than the PMSM with
the same power rating.
Fig. 13 shows the torque current angle profile of the SynRM
at maximum phase current (220 A) during intermediate mode
of operation in which the developed torque is maximized when
the current angle (i ) is at 32 electrical degrees. The current
control technique is necessary to control the machine from
zero up to base speed when it operates as traction. Therefore,
the knowledge of the optimum current angle is necessary to
maximize the torque magnitude. The optimum current angle is
not a unique parameter. It is current dependent and varies with
respect to the magnitude of the phase current. This is because
for a specific rotor, the saliency ratio is not only sensitive to
rotor position, but it also depends on phase current.
Fig. 14 shows the characteristics of the flux versus phase
current for different rotor positions (64, 0, 54, mechanical
in stationary reference frame) which is also called magnetic
curves. To evaluate the machine performance in continuous
mode of operation, the phase current information is necessary.
Refer to Fig. 3, continuous mode of operation, and Fig. 11,
the SynRM with 0.6-mm air gap needs to draw (90125 A) in
phase to satisfy the torque requirement (90160 Nm). It is

Fig. 15. d-q inductances characteristic versus phase current at i = 32 and

= 68 electrical.

illustrated in Fig.14 that, when the rotor angle is at 54

the machine can operate around linear part of magnetic curve
that refers to less saturation and higher saliency ratio. As
opposed to this, when the rotor position is at 0 or 64 the
machine operates in the nonlinear region and it is expected to
have lower saliency ratio caused by saturation. Therefore, the
designed machine is capable of satisfying the desired torque
envelope (Fig. 3) when the rotor position is adjusted at 54.
It highlights the importance of the rotor position in SynRM
operation.
Figs. 15 and 16 show the d-q axis inductances and the
saliency ratio versus current ratings of the proposed SynRM,
respectively. The importance of the results in Fig. 16 are
that the saliency ratio of 10 can be achieved for current
rating in continuous mode (100 A) while the rotor position
is aligned at the proper angle (54 mechanical degrees in
stationary frame). This yields maximum power at maximum
speed in continuous mode of operation, which is compatible
with vehicle highway driving scheme (cruising). It has to
be noted that the previously mentioned operating point and
saliency ratio are two leading parameters (Fig. 3, Table II)
in the proposed sizing method. On the other hand, the rotor
alignment, which can be practically defined by the current

TAGHAVI AND PILLAY: SIZING METHODOLOGY OF THE SYNCHRONOUS RELUCTANCE MOTOR

Fig. 16.
Saliency ratio of the proposed SynRM versus phase current
characterized at two different rotor positions.

angle is the key parameter in achieving the proper saliency

ratio in different current ratings or mode of operations.
VIII. C ONCLUSION
The proposed sizing method algorithm was evaluated
successfully by FE analysis and test data comparison. This
method can be used as a fast and reliable procedure to
identify the most important parameters of the SynRM based
on the design specifications. This can be prior to FE analysis
and design of the rotor geometry and its optimization. The
geometries that would be characterized through this method
are not unique. They can be varied by considering the design
limitations, e.g., maximum stack length, outer diameter, and
converter current rating.
ACKNOWLEDGMENT
The authors acknowledge the support of the Natural
Sciences and Engineering Research Council of Canada,
Hydro-Qubec, and Connect Canada for this paper.
R EFERENCES
[1] M. Zeraoulia, M. E. H. Benbouzid, and D. Diallo, Electric motor drive
selection issues for HEV propulsion systems: A comparative study,
IEEE Trans. Veh. Technol., vol. 55, no. 6, pp. 17561764, Nov. 2006.
[2] C. C. Chan, The state of the art of electric and hybrid vehicles, Proc.
IEEE, vol. 90, no. 2, pp. 247275, Feb. 2002.
[3] M. Ehsani, Y. Gao, S. E. Gay, and A. Emadi, Modern Electric, Hybrid
Electric, and Fuel Cell Vehicles, vol. 1. Boca Raton, FL, USA: CRC
Press, 2005.
[4] A. Vagati, The synchronous reluctance solution: A new alternative in
AC drives, in Proc. 20th Int. Conf. Ind. Electron., Control Instrum.,
vol. 1. 1994, pp. 113.
[5] T. J. E. Miller, A. Hutton, C. Cossar, and D. A. Staton, Design of
a synchronous reluctance motor drive, IEEE Trans. Ind. Appl., vol. 27,
no. 4, pp. 741749, Jul./Aug. 1991.
[6] G. Pellegrino, E. Armando, P. Guglielmi, and A. Vagati, A 250kW
transverse-laminated Synchronous Reluctance motor, in Proc. 13th Eur.
Conf. Power Electron. Appl., Sep. 2009, pp. 110.
[7] A. Vagati, M. Pastorelli, G. Francheschini, and S. C. Petrache, Design
of low-torque ripple synchronous reluctance motors, IEEE Trans. Ind.
Appl., vol. 34, no. 4, pp. 758765, Jul./Aug. 1998.
[8] A. Vagati, A. Fratta, G. Franceschini, and P. M. Rosso, AC motors for
high-performance drives: A design-based comparison, in Proc. Conf.
Rec. IEEE Ind. Appl. Conf., 30th IAS Annu. Meeting, vol. 1. Oct. 1995,
pp. 725733.
[9] P. Jung-Min, K. Sung, H. Jung-Pyo, and L. Jung-Ho, Rotor design
on torque ripple reduction for a synchronous reluctance motor with
concentrated winding using response surface methodology, IEEE Trans.
Magn., vol. 42, no. 10, pp. 34793481, Oct. 2006.

339

[10] N. Bianchi, S. Bolognani, D. Bon, and M. D. Pre, Rotor fluxbarrier design for torque ripple reduction in synchronous reluctance and
PM-assisted synchronous reluctance motors, IEEE Trans. Ind. Appl.,
vol. 45, no. 3, pp. 921928, May/Jun. 2009.
[11] M. J. Kamper, F. S. Van der Merwe, and S. Williamson, Direct finite
element design optimisation of the cageless reluctance synchronous
machine, IEEE Trans. Energy Convers., vol. 11, no. 3, pp. 547555,
Sep. 1996.
[12] R. R. Moghaddam, F. Magnussen, and C. Sadarangani, Novel rotor
design optimization of synchronous reluctance machine for high torque
density, in Proc. 6th IET Int. Conf. PEMD, Mar. 2012, pp. 14.
[13] T. Matsuo and T. A. Lipo, Rotor position detection scheme for
synchronous reluctance motor based on current measurements, IEEE
Trans. Ind. Appl., vol. 31, no. 4, pp. 860868, Jul./Aug. 1995.
[14] T. Matsuo, A. El-Antably, and T. A. Lipo, A new control strategy for
optimum efficiency operation of a synchronous reluctance motor, in
Proc. Conf. Rec. IEEE, Ind. Appl. Conf. 31st IAS Annu. Meeting, vol. 1.
Oct. 1996, pp. 109116.
[15] T. Finken, M. Felden, and K. Hameyer, Comparison and design of
different electrical machine types regarding their applicability in hybrid
electrical vehicles, in Proc. 18th Int. Conf. Electr. Mach., Sep. 2008,
pp. 15.
[16] J. J. Germishuizen, F. S. Van der Merwe, K. Van der Westhuizen, and
M. J. Kamper, Performance comparison of reluctance synchronous and
induction traction drives for electrical multiple units, in Proc. Conf.
Rec. IEEE, Ind. Appl. Conf., vol. 1. Oct. 2000, pp. 316323.
[17] I. Boldea, Reluctance Synchronous Machines and Drives, vol. 1, 1st ed.
New York, NY, USA: Oxford, 1996.
[18] I. Boldea and S. A. Nasar, The Induction machine Design Handbook,
2nd ed. New York, NY, USA: CRC Press, 2009.
[19] I. Boldea, Z. X. Fu, and S. A. Nasar, Performance evaluation of axiallylaminated anisotropic (ALA) rotor reluctance synchronous motors, in
Proc. Conf. Rec. IEEE Ind. Appl. Soc. Annu. Meeting, Oct. 1992,
pp. 212218.
[20] D. Platt, Reluctance motor with strong rotor anisotropy, IEEE Trans.
Ind. Appl., vol. 28, no. 3, pp. 652658, May/Jun. 1992.
[21] X. B. Bomela and M. J. Kamper, Effect of machine design on
performance of reluctance synchronous machine, in Proc. Conf. Rec.
IEEE Ind. Appl. Conf., Oct. 2000, pp. 515522.
[22] T. Matsuo and T. A. Lipo, Field oriented control of synchronous
reluctance machine, in Proc. 24th Annu. IEEE, PESC Rec., Jun. 1993,
pp. 425431.
[23] K. Sungmin, S. Seung-Ki, K. Ide, and S. Morimoto, Maximum
efficiency operation of Synchronous Reluctance Machine using signal
injection, in Proc. IPEC, Jun. 2010, pp. 20002004.
[24] A. Fratta, G. P. Troglia, A. Vagati, and F. Villata, Evaluation of torque
ripple in high performance synchronous reluctance machines, in Proc.
Conf. Rec. IEEE Ind. Appl. Soc. Annu. Meeting, Oct. 1993, pp. 163170.
[25] T. Matsuo and T. A. Lipo, Rotor design optimization of synchronous
reluctance machine, IEEE Trans. Energy Convers., vol. 9, no. 2,
pp. 359365, Jun. 1994.
[26] D. A. Staton, T. J. E. Miller, and S. E. Wood, Maximising the saliency
ratio of the synchronous reluctance motor, IEE Proc. B Electr. Power
Appl., vol. 140, no. 4, pp. 249259, Jul. 1993.
[27] A. Vagati, G. Franceschini, I. Marongiu, and G. P. Troglia, Design
criteria of high performance synchronous reluctance motors, in Proc.
Conf. Rec. IEEE Ind. Appl. Soc. Annu. Meeting, Oct. 1992, pp. 6673.
[28] M. J. Kamper and A. F. Volsdhenk, Effect of rotor dimensions and
cross magnetization on Ld and Lq inductances of reluctance synchronous
machine with cageless flux barrier rotor, IEE Proc. Electr. Power Appl.,
vol. 141, no. 4, pp. 213220, Jul. 1994.

Seyedmorteza Taghavi (S10) received the B.Eng.

degree from the University of Science and Technology, Iran, Tehran, in 1986, and the M.Eng. degree
from Concordia University, Montreal, QC, Canada,
in 2011, where he is currently pursuing the Ph.D.
degree in electrical engineering.
His current research interests include electrical
machine design in automotive and industrial applications, renewable energy, and power electronics in
power systems.

340

IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS, VOL. 2, NO. 2, JUNE 2014

Pragasen Pillay (F05) received the bachelors and

masters degrees from the University of Kwa-Zulu
Natal, Durban, South Africa, in 1981 and 1983,
respectively, and the Ph.D. degree from Virginia
Polytechnic Institute and State University, Blacksburg, VA, USA, in 1987.
He is currently a Professor with the Department
of Electrical and Computer Engineering, Concordia University, Montreal, QC, Canada, where he
holds the NSERC/Hydro Quebec Industrial Research
Chair. From 1988 to 1990, he was with the University of Newcastle-upon-Tyne, Newcastle-upon-Tyne, U.K. From 1990 to 1995,
he was with the University of New Orleans, New Orleans, LA, USA. From
1995 to 2007, he was with Clarkson University, Potsdam, NY, USA, where he
was the Jean Newell Distinguished Professor of engineering. He is an Adjunct
Professor with the University of Cape Town, Cape Town, South Africa. His
current research interests include modeling, design, and control of electric
motors and drives for industrial and alternate energy applications.
Dr. Pillay is a member of the IEEE Power Engineering, the IEEE Industry
Applications (IAS), the IEEE Industrial Electronics, and the IEEE Power
Electronics Societies. He is a member of the Electric Machines Committee,
and a Past Chairman of the IEEE Industrial Drives Committee of the IAS,
the Induction Machinery Subcommittee of the IEEE Power Engineering
Society, and the Awards Committee of the IAS Industrial Power Conversion
Department. He has organized and taught short courses in electric drives at
IAS Annual Meetings. He is a fellow of the Institution of Electrical Engineers
and Technologists, U.K., and a Chartered Electrical Engineer in the U.K. He
is a member of the Academy of Science of South Africa. He was a recipient
of the Fulbright Scholarship. He received the Order of Mapungubwe from
the President of South Africa in 2008 for contributions to South Africa in the
area of energy conservation.