J Electr Eng Technol Vol. 8, No. ?: 742-?, 2013 http://dx.doi.org/10.5370/JEET.2013.8.?.742

742

The Study of the Stray Load Loss and Mechanical Loss of Three Phase Induction Motor considering Experimental Results

Dong-JunKim*, Jae-Hak Choi*, Yon-Do Chun*, Dae-Hyun Koo* and Pil-Wan Han†

Abstract – The accurate determination of induction motor efficiency depends on the estimation of the five losses of stator and rotor copper loss, iron loss, mechanical loss and stray load loss. As the mechanical and stray load losses are not calculated by electro-magnetic analysis, the values of these two losses are very important in induction motor design. In this paper, the values of mechanical loss and stray load loss are proposed through investigating testing data from commercial products of three phase induction motors under 37kW. If the values of this paper are applied to motor design, the accuracy of design and analysis can be improved. The losses of motors are obtained by using load and no-load test results following IEC 60034-2-1 standard.

Keywords : Stray load loss, Mechanical loss, Efficiency, Induction motor

1. Introduction

The representative standards for efficiency test of three phase induction motor are IEEE std. 112, IEC 60034-2-1 and CSA C 390, where the stray load loss (PS) is determined by the means of output power. These standards also determine the PS from assigned values in pre-defined curve, which depend on motor rated output power [1]. The PS is determined by subtracting the conventional losses from the apparent total loss. The dependence of the PS on motor rating is often stated in literature but the analytical calculation of PS is difficult and historical test data have often been relied upon [2].

In squirrel-cage induction motors, the mechanical losses are produced by friction losses in bearings, windage losses of outside cooling fan, friction air losses of rotor and windage losses of internal fans of rotor rings [3]. In IEC 60034-2-1 and IEEE 112B standard, the mechanical loss is determined from no-load operation of the motor at variable voltage [4].

In this paper, the values of mechanical loss and stray load loss are investigated through testing data from 196 commercial products of three phase induction motors under 37kW

2. Efficiency Test by IEC 60034-2-1

2.1 D2L sizing equation IEC 60034-2-1 was approved to replace IEC 60034-2 in

2007. One method of this standard followed the IEEE 112B procedure of determining the SLL through test measurements. This standard also provided for assigning the value of stray loss as a percentage of input power which is dependent on motor output power [2].

2.1 Procedure of efficiency test

The test procedure for efficiency and losses is listed in

Table 1. The load test is applied at six different load points. The

first four load points should be chosen to be approximately equally spaced between not less than 25% and up to and including the 100% load. The remaining two load points should be suitably approximately equally spaced above 100%. In no-load test, test motor is uncoupled from the loading device and operated at a minimum number of 7 values of voltage ranging from 125% of the rated voltage to 20% [4].

2.2 Efficiency and mechanical loss calculation

Motor efficiency, η is defined as a ratio of output

mechanical power to the input electrical power

(1)

† Corresponding Author: Electric Motor Research Center, Korea Electrotechnology Institute (pwhan@keri.re.kr)

* Electric Motor Research Center, Korea Electrotechnology Institute (pwhan@keri.re.kr)

Received: February 25, 2013; Accepted: July 19, 2013

ISSN(Print) 1975-0102ISSN(Online) 2093-7423

Table 1. Efficiency test procedure

Order Procedure 1 Check of motor specification 2 Measurement of motor resistance 3 Rated load thermal test 4 Load test 5 No load test at variable voltage 6 Calculation of losses and efficiency

w(Plomopundy

wof

(Plo1 fr3)vo

2

stcoappo

wlorefu(sisu

where Ploss is tPc1) and rotoross (Pm) and

mechanical loperation. The nder load teynamometer. The stator

winding. The rf the slip (s) a

In Eq. (3),

Pc1_0) from nooss (Pk) which

shows the plrom no-load t). Extrapolatioltage axis int

.3 Stray load In the IEEE

tray load losonventional lopparent total ower and outp

where PL is theoad points segression metunction of the slop) and B (os removed to osed to calcula

Fig. 1

D

the total losser copper loss

stray load oss are detstator, rotor a

ests, whereby

loss is measrotor copper land the airgap

subtracting to-load input

h is the sum oflot between Ptest results of ng a straight tercept is the m

_

d loss calcula

standard 112ss (Ps) is deosses from the

loss is the put power at th

e residual losshall be smthod based osquare of the

offset) are conobtain the cor

ate the stray lo

1. Plot to deter

Dong-JunKim, J

es in the moto(Pc2), iron losloss (Ps). Th

termined undand stray loss

y the motor

sured as I2R loss is determpower in Eq.

×

the no-load spower (Pin_0) f mechanical a

Pk and the volf 37kW induct

line to zero mechanical lo

_

ation

2B and the IEetermined bye apparent totdifference be

he load point

ss. The residumoothed by uon expressinge load torque instant coefficirrect stray loadoad loss using

rmine mechan

Jae-Hak Choi, Y

or including sss(Pi), mechanhe core loss der the no-ses are determ

is coupled t

loss in the smined as a pro

(5).

×s

stator copper gives a cons

and core loss. ltage squared tion motor (Tvoltage, the

oss (Pm) [4].

EC 60034-2-1y subtracting tal loss (Papp). etween the iof interest

al loss data ausing the li

g the losses in Eq. (5) wheents. The offsd loss. The lopEq. (6) [4, 5].

nical loss

Yon-Do Chun, D

743

tator nical

and -load

mined to a

tator oduct

(2)

loss stant Fig. (V2)

Table zero

(3)

, the the

The input

(4)

at six inear as a

ere A set B pe is .

Fregr

Tof 3the

2.4

Tvalumotin Estd. cannsimoutl

Fig

Tab

Our

ToICSp

TemwV

Tab

In

Dae-Hyun Koo

Fig. 4 shows tression line (Y

Tables 1 and 337kW 4-pole losses.

Assigning va

The IEC 6003ue for the straytor rating and Eq. (7) [2, 4]. T

112 [5], whicnot be comp

milar to assignline.

sP

g. 2. Plot to d

ble 2. Load tes

utput power/rated [%] 1

orque[Nm] 24nput[kW] 4

Current[A] 8peed[rpm] 1mperature of inding[°C] 10

Voltage[V] 38

ble 3. No-load

nput voltage/ rated [%]

Voltage[V] current[A] Input[kW]

and Pil-Wan H

Y

the residual loY) and PS of op show the loainduction mo

alues of stray

4-2-1 standary load loss. Tis between 0.

Table 4 showsch are the perpared with Ied values of

10.025P fo=

etermine stray

st results of 37

125 110

49.9 219.449.9 43.7 85.6 75.9 767 1772

01.6 101.8

80.9 379.8

d test results o

125 100

475 380 345.0 24.8 12.21 0.88 0

Han

osses at six loptimum mode

ad test and no-tor respective

y load loss

rd also allows This value is d.5% and 2.5%s the assignedrcentage of ouIEC 60034-2IEC 60034-2-

1nr P kW≤

y load loss (op

7kW 4polemo

100 75

199.1 148.8 39.6 29.6 69.3 54.2 1775 1780

101.6 99.6

380.4 379.6

f 37kW 4pole

80 60 5

304 228 1918.0 12.8 100.64 0.49 0.

((

oad points (PLel. -load test resuely to determi

for assigningependent on t

% of input powd values of IEEutput power an-1 exactly b-1 of Eq. (7)

ptimum mode

otor

50 25

98.8 49.219.9 10.340.4 29.31788 1793

98.3 96.7

380.5 380.3

emotor

50 35 20

90 133 760.5 7.4 5.045 0.38 0.3

(5) (6)

L),

ults ne

g a the

wer EE nd

but in

l)

2 3 3 3

7

3

0

6 01

3

ofAmefR

T

The Stud

0

s

s

P

P

⎡= ⎢⎣

= 3. New Assi

.1. Stray loa Figs. 3, 4, an

f induction mAll test motorsmade by motofficiency in te

Research Instit

Table 4. Assign

Rated outp

31851 a

Fig. 3

Fig. 4

dy of the Stray L

1

0.025 0.005

1.005

nfor kW PP for P

−

<

igned ValueMechan

d loss

nd 5 show themotors under 3s have 380V oor manufactuesting laboratotute. The data

ned value for

put power (Pn) [k1 ~ 90

91 ~ 375 376 ~ 1850 and greater than

. Stray load lo

. Stray load lo

Load Loss and M

2log101

1000010000

n

n

PkW

P kWP kW

⎛ ⎞⎜ ⎟⎝ ⎠

<

≥

es of Stray Lnical Loss

e distribution 7kW tested b

or 460V/60Hzurers and tesory of Korea Ea set is compr

stray load los

kW]

oss distributio

oss distributio

Mechanical Lo

1P

W

⎤⎞⎥⎟⎠⎦

Load Loss an

of stray load by IEC 60034-z rating whichted for verifElectrotechnorised of 52 2-

s (IEEE std. 1

Ps/Pn [%] 1.8 1.5 1.2 0.9

n (2Pole)

n (4Pole)

ss of Three Pha

744

(7)

nd

loss -2-1. h are fying ology -pole

motIn

averpow600lowblacaverthe grapby t

3.2.

Fof in

Inhighhighare % aare

Faver2.2kvalu

112)

ase Induction M

tors, 85 4-polen these graprage value of

wer and triang034-2-1 calculwer than assigck line is therage value in assigned valu

phs, it is knowthe number of

. Mechanical

Figs. 7, 8, and nduction moton Fig. 7, the her than thoshest speed in tabout 3% of

above 10KW rproposed as E

Fig. 8 shows thrage values kW rating andue is expresse

Fig. 5. S

Fig. 6. M

Motor considerin

e motors and 5phs, red linef stray load logle-green line ated by Eq. (7

gned value of e value propo

this paper whue of IEC 60wn that the stf poles.

l loss

9 are the distrors under 37kWmechanical

e of 4 and 6 the line start inoutput powerrating. In this Eq. (8) using the mechanicalare about 1.5d 1.0% above d by Eq. (9)

Stray load loss

Mechanical los

ng Experimenta

59 6-pole mote (Average) oss (PS) accois the assigne

7). Most of thef IEC 60034-2osed by consihich is moved

0034-2-1. As tray load loss

ribution of meW of section 3losses of 2 ppole motors

nduction. Ther under 5.5kWpaper, the me

test and averagl losses of 4 p5% of outpu3.7kW rating

s distribution (

s distribution

al Results

tors. represents t

ording to outped value of IEe test values a2-1. The circlidering test and by 1.0% froshown in the

s is not affect

echanical loss3.1.

pole motors abecause of t

e average valuW rating and 1echanical lossge values. ole motors. T

ut power undg. The propos

(6Pole)

(2Pole)

the put EC are le-nd

om ese ted

ses

are the ues 1.5 ses

he der ed

av5va

4

4

Fig. 9 showsverage values.5kW rating aalue is expres

0

0

0

0

0

0

4. Effects of to M

.1. Stray loa Table 5 sho

Fig. 7.

Fig. 8.

D

s the mechanics are about and 0.5% abosed by Eq. (10

.03 , 50.03 0.002

5.5.015 ,

.015 ,0.015 0.00 2.2.01 , 3

.01 , 50.005 0.

.005 ,

Stray Load Motor Design

d loss

ows the comp

Mechanical l

.Mechanical lo

Dong-JunKim, J

cal losses of 61.0% of out

ove 11kW rati0)

5.5 73 5

117.5

2.2 33 2

3.73.7

5.5 000909

, 5.511

Loss and Mn and Perfor

parison of the

loss distributio

oss distributio

Jae-Hak Choi, Y

6 pole motors.tput power uing. The prop

.5 ,

.2 ,

5.511

Mechanical Lrmance

e stray load

on (4Pole)

on (6Pole)

Yon-Do Chun, D

745

The under osed

(8)

(9)

(10)

Loss

loss,

efficcomlowconstracloswhivaluincrto rshowsligthe mot

4.2.

Taffedecrredumalthisfan mot

Tandresuto tappimpmot

Tab

(stA

Dae-Hyun Koo

ciency, volummmercial motower stray loantributes to loway load loss ofse to the assile those of mue (0.91%) ofreases for higheduce the coswn in Table 5htly higher thnew assigned

tor developme

. Mechanical

The mechanicected mostly brease to reduuction of coollfunction due reason, it is by consideri

tor design. The two coolid these are forults are shownthe same moropriate selec

prove a mototor using fan

ble 5. Relation(15kW 4

Items

Stray load loss(PS/Input poweTest efficiency

VolumeActivematerial c

teel+copper+alumAssigned value of

load loss

Fig.

and Pil-Wan H

me and active ors of manufad loss by 6wer volume af motor A is 1.igned value motor B is 0f this paper. Aher efficiencyst in design an5, the assignehan test resultd value is exent.

l loss

cal loss of liby cooling fanuce mechanicling fan size rto the rise in important to

ing the mech

ing fans of Fr 3.7kW 2-pon in Table 6 wotor. In this rction of coolinor efficiency. II is increased

n between stra4pole motor)

Unit

PS) [W] er [%] y [%]

[%] cost

minum) [k￦]

f stray

9. Cooling fa

Han

material cost acturer A and 60% than mand active ma.48% of input (1.91%) of I.61% close t

As the materiay, it becomes mnd manufactured value of IEts of commerc

xpected to be

ine start indun. The coolingcal loss, whilesults in motothe operating design or sel

hanical loss

ig. 9 have thle motor. The

whose cooling results, we c

ng fan make aThe temper

d by 5 degree

ay load loss an

Motor (A)

M

250 1.48 91.1 100

] 244

91% (IEC 6000.91% (Propos

an (3.7kW 2po

between 15kB. Motor B h

motor B and aterial cost. T

power whichIEC 60034-2-to the proposal cost of motmore significaring process. AEC 60034-2-1 cial motor thumore useful

uction motor g fan size has le the arbitraor performanctemperature.

lect the coolinand cooling

he different sie efficiency tefans are appli

can see that a contribution rature rising e, which can b

nd material co

Motor (B) Remark

100 60%↓

0.61 -90.8 -82 18%↓

210 13.9%↓

034-2-1) sed value)

ole)

kW has

it he

h is -1, ed tor ant As is

us, in

is to

ary ces In ng in

ize est ed an to of be

ost

k

%

imm

camdipalo

loasIEovthcore

ardesp

[1

[2

[3

[4

[5

T

F

The Stud

mproved by mmechanical los

As the mec

alculated by manufacturing

ifficult to expaper, assignedoss are propos

As shown inoad losses distssigned valueEC 60034-2-1ver-sized due his reason, thould be a guieduce the mate

In case of thre used in moeveloping hipecification.

1] A. T. de A

Angers, “IEC 34-2losses inmotors”, 608-614,

2] E.B. Agastray loadInd. Appl

3] T.A. LipWisconsiUniversit

4] IEC 6003losses anfor tractio

5] IEEE Std

Table 6. Motofan (3

Fan Eff. [%]

Cur[A

I 85.2 8.

II 87.8 7.

dy of the Stray L

making fan lss or reducing

5. Con

chanical loss electro-magnprocess or m

pect these losd values of msed under 37kWn test results otribute varioues of IEC 6001 are used in

to stray load he new assignideline for merial costs. he mechanicalotor design, iigh efficienc

Refe

Almeida, F. T“Comparative

2 efficiency tesn low voltag

IEEE Trans. Mar./Apr. 200

amloh, “An Evd loss from col., vol. 46, no.o, “Introduct

in Power Ety of Wisconsi34-2-1, “Stan

nd efficiency fon d. 112, “IEEE

r efficiency te.7kW 2pole m

rrent A]

Temp. rising ΔT Ir

.04 30 1

.87 35 1

Load Loss and M

larger considcopper loss an

nclusion

and stray lonetic analysis mechanical strsses in designmechanical loW by using te

of commercialsly but these a034-2-1. If asmotor designloss bigger th

ned values omotor manufac

l losses, if thet is expected

cy motor an

erences

. E. Ferreira, Je analysis of sting standard

ge three-phaseInd. Appl., v

02. valuation of i

ollated test res 6, pp. 2311-2

tion to AC MElectronics Rin, 2004, pp 3

ndard methodsfrom tests (ex

E Standard T

est results accmotor)

Los

ron Stator copper

Rocop

73 137 9

69 133 7

Mechanical Lo

dering appropnd iron loss.

oad loss are and affected

ructure, it is n process. In

oss and stray est results. l motors, the sare lower thanssigned valuen, motor coulhan in practicof stray load cturers who tr

e proposed vato be helpfu

nd selecting

J. F. Busch, anIEEE 112-B

ds using stray e cage inducvol. 38, no. 2

nduction macsults”, IEEE T2318, Nov. 20Machine DesiResearch Ce02-304. s for determi

xcluding mach

Test Procedure

cording to coo

ss[W]

otor pper

Mech.(Pm/Pn)

Sl

98 204 (5.5%)

74 98 (2.6%)

ss of Three Pha

746

priate

not d by very this

load

stray n the es of d be e. In loss

ry to

alues l for

fan

nd P. and

load ction , pp.

chine Trans. 10. ign”, enter,

ining hines

e for

[6]

[7]

[8]

seni

seni

EngFromtechchieof E

oling

Strayload

29

41

ase Induction M

Polyphase I A. Bogliett

Stray-Load on Ind. App

K. BradleyInduction Mand CaloriAppl., vol.

A. BoglieInduction MAnalysis oIEEE Trans1301, 2004

ior Engineer o

ior researcher

gineering at Wm 2004 to 2hnology Reseef researcher, Electric Motor

Motor considerin

Induction Motti, “Impact ofLosses in Ind

pl., vol. 46, noy, “EvaluatioMotors with a imetric Metho21, no. 3, pp.

etti, “InternaMotor Efficie

of the Stray-s. on Ind. App.

Dong-Jundegree in2004 fromreceived engineerinNational UKorea EInstitute (

of Electric Mo

Jae-Hak M.S., andEngineerinin 1999, 2From 200electronicsat Korea Institute (

r of Electric M

Yon-Do CM.S. and Engineerinin 1996, 1From 200Japan SocScience (Jwith the

Waseda Univ2012, he hasearch Institute

Principal Resr Research Ce

ng Experimenta

tors and Genef the Supply duction Motoro. 4, pp. 1374-on of Stray

Comparison ods”, IEEE 682-689, 200

ational Standency EvaluatiLoad Loss Dpl., vol. 40, n

n Kim He n electrical m Kyungnam

M.S degreeng in 2013 frUniversity. HeElectrotechnol(KERI). He otor Research

Choi He recd Ph.D. degreng from Hany2001 and 2005 to 2007, hes. Since 2008Electrotechno

(KERI). He Motor Research

Chun He recPh.D. degre

ng from Hany1998 and 20001 to 2003, ciety for theJSPS) fellows

Department versity as a vs worked at e (KERI). Hesearcher and tenter, KERI.

al Results

erators,” 2004Voltage on t

rs”, IEEE Tran-1380, 2010.Load Loss of Input-OutpTrans. on In

06. dards for tion: A CriticDeterminationno. 5, pp. 129

received Bengineering University. H

e in electricrom Changwoe has worked ogy Researis currently Center, KERI

ceived the B.Ses in Electricyang Universi05 respectivee worked at L, he has workology Researis currently

h Center, KER

eived the B.Ses in electricyang Universi01, respective

he received Promotion hip and he w

of Electricvisiting scholKorea Electr

e is currentlytechnical lead

. the ns.

in put nd.

the cal n”, 94-

B.S in

He cal on at

rch a

I.

S., cal ity ly.

LG ked rch

a RI.

S., cal ity ly.

a of

was cal ar. ro-

y a der

Dong-JunKim, Jae-Hak Choi, Yon-Do Chun, Dae-Hyun Koo and Pil-Wan Han

747

Dae-Hyun Koo He received the B.S. and M.S. degrees in Electrical Engin-eering from Hanyang University in 1989 and 1991, respectively. From 1991, he has worked at Korea Electro-technology Research Institute (KERI). In 2002, he received Ph. D. degree from Dong-A University. He is

currently a director of Electric Motors Research Center, KERI.

Pil-Wan Han He received the B.S., M.S. and Ph.D. degrees in Electrical Engineering from Hanyang University in 1998, 2000 and 2013 respectively. From 2000 to 2005, he worked at LG electronics. Since 2005, he has worked at Korea Electrotechnology Research Institute (KERI). He is currently a

senior researcher of Electric Motor Research Center, KERI.