IJAEA-1-5-8(57-63)30.01.20141391065744

8/10/2019 IJAEA-1-5-8(57-63)30.01.20141391065744

1/7

Thermal And Vibration Characterization Of High Speed Switched Reluctance

Motor With Soft Magnetic Composite Material

K. Vijayakumar1, R. Karthikeyan1and R. Arumugam2

1Research Scholar, Department Of Electrical Engineering, Anna University, Chennai, India1Research Scholar, Department Of Electrical Engineering, Anna University, Chennai, India2Retired Professor, Department Of Electrical Engineering, Anna University, Chennai, India

[email protected]

Abstract

In aircraft electric motor driven fuel delivery system, the motor must be characterized by high power density, reliability,

less size, less weight and high speed. The high speed operation, fault tolerance, high power density makes switched

reluctance motor an ideal candidate for high speed aerospace applications. This paper investigates the application

potential of soft magnetic composite material (SOMALOY 500) in high-speed switched reluctance motor. Two

configurations viz (1) All sheet metal and (2) All Soft magnetic composite have been studied extensively using finiteelement analysis to obtain their thermal and vibration characteristics with two different type of stators such as stator with

smooth and ribbed frame. The study reveals that the all soft magnetic composite configuration albeit suffers from poor

average torque posses better thermal capability, lesser vibration of stator that may result in low acoustic noise level a

desirable feature in high speed aerospace applications.

Keywords: Finite element analysis, Soft magnetic composite (SMC), Switched reluctance motor (SRM),Thermal characterization, Mode shapes.

1.0 INTRODUCTION

The low cost, high speed, high power density, fault

tolerance and reliability of switched reluctance motor

make it a viable alternative in aircraft electric-motor-

driven fuel pump system [3]. Recently soft magnetic

composite materials find rampant application inelectrical machines [1]-[2] and these materials are

characterized by three dimensional isotropic

ferromagnetic behavior, very low eddy current loss,

flexible machine design and assembly and a prospect for

greatly reduced production cost. This paper investigates

the suitability of soft magnetic composite material in

switched reluctance motor designed for high-speed

aerospace application from a study of its thermal and

vibration behavior. Two configurations viz (i) All

laminated sheet steels (ii) All Soft Magnetic Composites

(SMC) have been extensively studied using finite

element analysis based software tool. The study

concludes that all SMC configuration albeit its poor

average torque in comparison with conventional motor,promises to be suitable for high-speed aerospace

application with better thermal behavior and lesser

acoustic noise.

2.0 SWITCHED RELUCTANCE MOTOR

A switched reluctance motor is an electrical machine in

which the torque is developed by the tendency of the

rotor to occupy a position so as to minimize the

reluctance of the magnetic path of the excited stator

phase winding. The switched reluctance motor is a

doubly salient but singly excited machine wherein the

stator carries the winding while the rotor is simply made

of stacked silicon steel laminations. This lends to asimpler geometry for switched reluctance motors as

evidenced from the two-dimensional Computer Aided

Design (CAD) model of a 6/4 switched reluctance motor

shown in Fig. 1.

Fig. 1: CAD Model of 6/4 (3phase) SRM

3.0 SOFT MAGNETIC COMPOSITE MATERIALS

IN ROTATING ELECTRICAL MACHINES

Electrical steel lamination is the most commonly used

core material in electrical machines. Electrical steels are

typically classified into grain-oriented electrical steels

IJAEA, Volume 1, Issue 5, pp.57-63 (2008)

Fragrance 57

8/10/2019 IJAEA-1-5-8(57-63)30.01.20141391065744

2/7

and non-oriented electrical steels. Typical applications

for grain-oriented steels are power transformer cores

whereas non-oriented steels are broadly used in different

kinds of rotating electrical machines. Electromechanical

steels currently used in the manufacture of electrical

machines posses high induction of magnetic saturation(Bs~2T), low coercive force (Hc< 100A/m), and they are

characterized by low total losses [2] , [3]. Electrical sheet

steels have been the dominant choice for the soft iron

components in electrical machines subject to time

varying magnetic fields. The new soft iron powder

metallurgy materials can be considered as an alternative

for magnetic core of these electrical machines. The basis

for soft magnetic composites is bonded iron powder as

shown in Fig.2.

Fig. 2: Iron Powder Particle

The powder is coated, pressed into a solid material using

a die, and heat-treated to anneal and cure the bond. The

B-H characteristics of M19 silicon steel and SMC

material SOMALOY 500, shown in Fig. 3, reveals that

although the SMC has inferior relative permeability

when compared with lamination steel it still posses the

following desirable characteristics [4].

Fig. 3: B-H Characteristics Of Laminated Steel And SMC

a. Reduced copper volume as a result of increased fill

factor and reduced end winding length and reduced

copper loss as a result of the reduced copper

volume,

b. Potential for reduced air gap length as a result of the

tight tolerances maintained in manufacturing SMC

material,

c. Modular construction allows the possibility of easy

removal of an individual modular unit for quick

repair or replacement,

d. Stator is easily recyclable since the stator can again

be compressed back into powered form.

4.0 LOSS CALCULATION

In order to estimate the thermal behavior of the machine

accurately meticulous attention must be paid to the

calculation of the losses. The loss components of the

machine can be summarized as:

Ploss.m=Pcu+PFe+Pmech (1)where Pcu, PFe, Pmecare copper losses, core (iron) losses

and mechanical loss respectively.

4.1 Copper Losses

The first step in calculating the copper losses involves

the calculation of the resistance of each phase winding of

the SRM. The mean length of a winding turn is given as,

lm= {2L+4W

t+ 2Dsin{

s/2}}*10

-3m (2)

The resistance of a single phase is calculated as,

Rs= 0.0177* lmTph/acohm (3)

The copper losses at rated current is given by

Pcu = i2pRsW (4)

4.2 Core Loss

In a high-speed motor, the core loss is much greater than

copper loss. Fig. 4 shows the core loss difference

between SMC and M19 material.

Fig . 4: Core loss Vs frequency plot of laminated steel and

SMC

The core loss can be calculated by

Pfe= Kha fBmajh +Kea (fBmaj)

2+Kaa

(fBmaj)1.5

+KhafBminh+Kea(fBmin)2+Kaa(fBmin)

1.5 (5)

where Kha, h, Kea and Kaa are coefficients deduced

from the data provided by the manufacturer, or measured

by the standard alternating tester,Bmaj andBmin are the

major and minor axes of the elliptical B locus during

360electrical degrees, and f is the frequency. The coreloss is computed based on nonlinear time-stepping FEA.

One electrical period is divided into 8 steps. The

meshing of stator and rotor is kept the same in each step.

Core loss in each element is calculated, and then the total

core loss of the motor is obtaining by summing up the

losses of all the elements


Fragrance 58

8/10/2019 IJAEA-1-5-8(57-63)30.01.20141391065744

3/7

4.3 Mechanical Losses

The equation for mechanical losses, comprising cooling

fan windage loss and motor bearings friction loss is

given by

Pmech= K (n/10)2

D4s (6)

The designer must therefore make every effort to

minimize the windage loss and to maximize the machine

cooling.

5.0 THERMAL ANALYSIS

For thermal analysis, there are two basic methods: the

thermal network method and the finite element method.

In general, the geometrical complexity of an electrical

machine requires a large thermal network. The whole

motor is modeled into many small solid elements, which

are of block shape and can be modeled by six thermal

resistances and seven nodes. However, the core loss of

each element in thermal network cannot be obtained

easily from the result of magnetic filed calculation.Mostly, the average is used. Because core loss

distribution is quite different in different position of the

stator core accuracy cannot be improved by increasing

the number of elements in thermal network. In this

paper, a three dimensional (3D) as well as two

dimensional finite element analysis (FEA) is used to

obtain the accurate temperature distributions for two

motor configurations by keeping the same mesh

structure.

6.0 THERMAL FIELD ANALYSIS

The heat transferred by radiation is ignored. The partial

differential equation of the heat conduction and

convection is expressed as

where [Kg/m ] is the mass density, c [J/(KgK)] the

specific heat, T [K] the temperature, t [s] the time, {V}

the velocity vector for mass transport of heat, [D] the

Thermal conductivity matrix, Kxx Kyy Kzz and K [W/

(mK)] are the thermal conductivity alone the x, y and z

directions, respectively, and q [W/m ] is the heat

generation rate per unit volume.

6.1 Boundary Conditions

The boundary conditions can be of Dirichlet type, where

the temperature T* on the boundary is specified. This

boundary condition can be applied to the surface

between the frame and outer air. A Neumann type of

boundary condition specifies the heat flux flow qnthrough the boundary. The heat flux flow through the

boundaries to the surrounding is described with the heat

transfer coefficient h and the ambient temperature Tamb.The governing equation [5] is given by

qn n (kT) h(T Tamb (7)

The finite element thermal analysis model with the

boundary conditions applied is shown in Fig .5.

Fig. 5: Finite Element Thermal Model With BoundaryConditions Applied

The convective heat transfer to the surroundings is

dependent on the geometry and the cooling conditions.

Air gap convection coefficient is determined by two main

quantities: the ruggedness of the rotor and stator surfaces

and the peripheral speed of the rotor surface. By assuming

smooth surface, the convection coefficient can be

calculated by experiential formula [6]

h28(1+ ) (W/ (m2K)) (8)

(0.5vr)2 va

2(m/s)}

1/2(9)

where vrand va are the peripheral and axial speeds of the

rotor surface, respectively.

7.0 RESULTS OF SIMULATION

The results obtained by the simulated thermal analysis

are presented in this section. The foot-notes in the Fig. 6

through 11 are (colour plot shown horizontally) are the

simulation descriptions of the respective plots.

a)

b)


Fragrance 59

8/10/2019 IJAEA-1-5-8(57-63)30.01.20141391065744

4/7

c)

d)

Fig. 6: Nodal Temperature Distributions In Different Parts

Of SRM With Soft Magnetic Composite Material. (A)Temperature Distribution With Smooth Frame. B)

Temperature Distribution With Ribbed Frame (Number Of

Ribs 40). C) Temperature Distribution With Ribbed Frame(Number Of Ribs 36). D) Temperature Distribution With

Ribbed Frame (Number Of Ribs 32)

a)

b)

c)

d)

Fig. 7: Nodal Temperature Distributions In Different Parts

Of SRM With Laminated Sheet Steel. (A) TemperatureDistribution With Smooth Frame. B) Temperature

Distribution With Ribbed Frame (Number Of Ribs 40). C)

Temperature Distribution With Ribbed Frame (Number Of

Ribs 36). D) Temperature Distribution With Ribbed Frame(Number Of Ribs 32)

Fig. 8: 3D Nodal Temperature Distributions In Different

Parts Of SRM With SMC

Fig. 9: 3D Nodal Temperature Distributions In Different

Parts Of SRM With M19


Fragrance 60

8/10/2019 IJAEA-1-5-8(57-63)30.01.20141391065744

5/7

Fig. 10: Transient Temperature-Time Curve Of Stator

Body For The Continues Duty Cycle For SMC

Fig. 11: Transient Temperature-Time Curve Of Stator

Body For The Continues Duty Cycle For M19

TABLE I: NODAL TEMPERATURE DIFFERENCES BETWEEN

TWO CONFIGURATIONS WITH SMOOTH STATOR FRAME

Material Minimumtemperature

Maximumtemperature

SMC 316.34 319.04

M19 319.126 321.499

% difference in

Temperature

27.86 24.59

TABLE II: NODAL TEMPERATURE DIFFERENCES BETWEEN

TWO CONFIGURATIONS WITH RIBBED STATOR FRAME

SMC M19

Min

temp

Max

temp

Min

temp

Max

temp

No.Of

Ribs

%difference

in max

Temperature

310.105 312.846 311.334 313.925 40 10.79

310.563 313.383 312.445 314.275 36 8.92

310.744 313.495 313.205 315.381 32 18.86

TABLE III: COMPARISON BETWEEN 3D AND 2D THERMAL

ANALYSIS

2D Thermal analysis 3D Thermal analysis

M19 SMC M19 SMC

Min Max Min Max Min Max Min Max

319.3

4321.499 316.34 319.04 311.09 312.191 309.602 311

% difference

in temp is

24.59

% difference in

temp is 11.91

* Temperatures are in Kelvin

The superior heat evacuation capability of soft magnetic

composite material (SMC) compared with M19 has been

evidenced from Table I, II and III.

8.0 VIBRATION AND ACOUSTIC NOISE

SOURCES OF SWITCHED RELUCTANCEMOTOR

Two disadvantages of the machine are its torque ripple

and acoustic noise [11].The acoustic noise sources are

listed as follows:

1) SRM has a doubly salient structure. The rotor poles

behave like blades, which cause acoustic noise due to

windage;

2) Non-uniform characteristics of materials produce

mechanically dynamic unbalance of the rotor and a non-

uniform distribution of magnetic flux, which cause both

magnetic and mechanical forces on the rotor and results

in acoustic noise;

3) Manufacturing asymmetries of the rotor and stator,

especially a non-uniform air gap, excites asymmetricforces on the rotor;

9.0 VIBRATION ANALYSIS OF SWITCHED

RELUCTANCE MOTOR

Usually, the main stages of the analysis and calculations

are:

9.1 Calculation And Verification Of The System Natural

Frequency: The subassemblies (wounding stator, rotor

stack with shaft etc.) model can be built and the natural

frequencies and mode shapes can be calculated.

9.2 Electromagnetic Model And Force Calculation: All

the related dimensions, phase winding turns and the

properties of all materials are modeled in a software

package, for instance, in ANSYS, to calculate the fluxlinkage curves as a function of rotor position and phase

current. Based on the current and rotor position, both

radial and tangential forces are calculated.

10.0 VIBRATION MODES OF ALL

LAMINATIONS (M19) AND ALL SMC

CONFIGURATIONS

The vibration modes and resonant frequencies has been

calculated with the aid of ANSYS commercial finite

element package. The following assumptions are made in

predicting the frequencies: (1) the stator yoke is a round

rigid body; (2) the teeth and windings have no rigidity,

so that their mass is attached to the yoke; (3) the periodic

force waves withrth orderare symmetrically exerted on

the stator yoke ring; (4) the only effects of all defects and

notches is a reduction in the mass.

The following formulas are used to predict the natural

vibration frequencies for the force wave with rth order,

where m is the equivalent mass per square meter (kg/m2)

on the cylindrical surface at the average yoke radius,

which can be expressed by the following formula:


Fragrance 61

8/10/2019 IJAEA-1-5-8(57-63)30.01.20141391065744

6/7

(10)

(11)and the equivalent spring stiffness coefficient per square

meter kr (N/m3) on the cylindrical surface at average

yoke radius.

(12)

My is the yoke mass including stator windings and poles

(kg); hy is the yoke height in the radial direction (m);

Ry(av) is the average radius of yoke (m); Ly(eff) is the

effective length of yoke ; kD is the spring stiffness

coefficient of the shock absorber under the motor. The

resonant frequencies calculated for the two

configurations viz (i) All laminated sheet steels (ii) AllSoft Magnetic Composites (SMC) using structural finite

element analysis are listed in Table IV.

TABLE IV: COMPARISON OF SMOOTH FRAME AND RIBBED

FRAME WITH TWO MOTOR CONFIGURATIONS

This paper also considers the effects of the frame on the

mode shapes and frequencies. A tight friction coupling

between the laminations and frame is assumed. Fig. 12

and 13 shows the effects of smooth frame and ribbed

frame to the all laminations (M19) stator. Given the

influence of the smooth frame and frame with ribs on the

natural frequency, it is therefore of interest to examine

the effects of the smooth frame and frame with ribs on

the natural frequencies. Fig. 14 and 15, summarizes the

results for the case of a z-constrained vibration of a

smooth frame and frame with ribs for all SMC

configuration, the result showing that a gradual decrease

(almost 45%) in the mode shape frequencies of frame

with ribs. The overall vibration mode shape frequencies

shows that the structural capability of configuration (ii)

is better when compared with configuration (i) in the

high-speed aerospace applications regime.

11.0 CONCLUSION

The application potential of soft magnetic composite

material in high-speed switched reluctance motor has

been analyzed. The results were presented by

considering the excitation of one phase. The key points

in determining the steady-state and transient thermal

characteristics of SRM are no-load temperature

distributions at the different parts of the machine,

temperature-time curves of stator body with SMC and

M19 configurations, smooth stator frame, ribbed stator

frame, and finally differences between 3D and 2D finite

element analysis. It has been observed that the steady-

S.

No.

Motor

configuration

Mode

shape

frequencie

s of Stator

lamination

s with

smooth

frame (Hz)

Mode shape

frequencies

of Stator

laminations

with ribbed

frame (Hz)

Differences

in

frequencies

(Hz)

a)700 a)371 329

b)1786 b)1170 6161 All SMC

c)3470 c)1512 1958

a) 901 a) 484 417

b) 2298 b) 1527 771

2

All M19

c) 4454c) 1972 2482

a)Mode at 901.147HZ b)Mode at 2298 HZ c)Mode at

4454HZFig .12:Vibration Modes Of The Stator Laminations With

Smooth Frame

a)Mode at 484HZ b)Mode at 1527HZ c)Mode at 1972HZ

Fig.13:Vibration Modes Of The Stator Laminations WithRibbed Frame.

a)Mode at700 b)Mode at 1786 c) Mode at3470Fig.14:Vibration Modes Of The SMC Stator With Smooth

Frame

a)Mode at 371 b)Mode at 1170 c) Mode at 1512Fig.15:Vibration Modes Of The SMC Stator With Ribbed

Frame


Fragrance 62

8/10/2019 IJAEA-1-5-8(57-63)30.01.20141391065744

7/7

state temperature is 319 K for configuration-2 and 321.5

K for configuration-1 considering copper and core loss.

The final study has led to the SMC based switched

reluctance motor configuration with ribs provision has

enhanced the heat dissipation. Also an in-depth analysis

of vibration modes of the stator of switched reluctancemachine with two configurations has been discussed.

The study started with the lamination and SMC with a

smooth frame, and then extended as well as to a ribbed

frame. The obtained result shows that a gradual decrease

(almost 45%) in the mode shape frequencies of frame

with ribs. The considerable torque reduction in

configuration-2 is due to the poor permeability of soft

magnetic composite material. At the same time the

structural capability of configuration-2 is better when

compared with configuration-1 in the high-speed

aerospace applications regime. The finite element

analysis aided in the estimation of mode shape

frequencies for the two motor configurations and in their

visualization.

REFERENCES

[1] M. Persson, P. Jansson, A. G. Jack, B. C. Mecrow,

Soft Magnetic Composite Materials Use for Electrical

Machines; 7th

International Conference on Electrical

Machines and Driveat Durham, England, Sep. 1995.

2] A.G. Jack, Experience with the Use of Soft Magnetic

Composites in Electrical Machines, International

Conference on Electrical Machines, Istanbul, Turkey,

1998, pp. 1441-1448.

[3] A. V. Radun, High-power density switched

reluctance motor drive for aerospace applications, IEEE

Trans. Ind. Applicat., vol. 28, no. 1, pp. 113-119,Jan/Feb. 1992.

[4]The Latest Development in Soft magnetic

Composite Technology from Hoganas Metal Powders,

Hoganas Manual, 1999-2003.

[5] K. N. Srinivas and R. Arumugam, Thermal

characterization through finite element anlaysis of the

switched reluctance motor, in Proc. IEEE Region 10Electrical and Electronics Technology Conference,

2001, pp. 819-824.

[6] Y. K. Chin, E. Nordlund and D. A. Station, Thermal

analysis lumped-circuit model and finite elementanalysis, The 6th International Power Electronics

Conference (IPEC), Singapore, November 2003, pp.

952-957.

[7] R.S.Colby, F.Mottier, T.J.E.Miller, Vibration modes

and acoustic noise in a 4-phase switched reluctance

motor, Conference Record of the 1995 IEEE Industrial

Application Society, 30th IAS Annual Meeting, Vol.1,

Orlando, USA, Oct. 8-12, 1995, pp.441~447.

[8] C.Y.Wu, C.Pollock, Analysis and reduction ofacoustic noise and vibration in the switched reluctance

drive, IEEE Trans. on Industry Applications, Vol.31,

No.1, January/February, 1995, pp.91~98.

[9] P.Pillay and W.Cai, An Investigation in switched

reluctance motors,IEEE Trans. On Ind.

Appl.,Vol,35.no.3.pp.589-596,Mar./Apr.1999.

[10]C.Yongxiao and W.Jianhua, Analytical calculations

of natural frequencies of stator of stator of switched

reluctance motor, in Proc.8th

Int.conf. Electric machines

and Drives 1997.pp.81-85.

[11] R. Rabinovici, Torque ripple, vibrations, and

acoustic noise in switched reluctance motors, HAIT

Journal of Science and Engineering B, Volume 2, Issues5-6, pp. 776-786, July 2005


Fragrance 63

Documents

IJAEA-1-5-8(57-63)30.01.20141391065744