..tt..* . » <

........ :....._...,_.^.^>

w

^

4-

HI. ^A-

f

• +4

^ > -r

,1

4- ^ -*

1- #

^ ^

IN

# -f >i

4 f +

f -^t ^ ^

f - - * T-

1" 4

^ ^ ' ^ 4 t-t- i

4

f ^ ^ ^ ^t I. ^ ^ ^-

IN,

> t 4

4

4

i-r f

•A*.T

4

T

ir

>

4 T

T- m-i

1.r

•

4 1- 44 It-

44^T 4 T

T

T 4

4 IT

if. •c,*

T

T

t

4,

r -f

4.

4- a.

4. 4 T

44 t

^-

4 t

r 4'-

4^ 1*

4- ^4

4

4- ^- 4-

THE JEFFECT OF VOLTAaE UNBALANCEON THE OUTPUT OF A TWO-PHA8E

INDUCT^^N MOTOR

BY

Ross Elmer CullingsAND

Charles Eugene McCormack

THESIS

FOR THK

DEGREE OE BACHELOR OE SCIENCE

IN

ELECTRICAL ENGINEERING

COLLEGE OF ENGINEERING

UNrVERSITV OF TI.r^rNOTS

PRESENTED JUNE. 1910

V

UNIVERSITY OF ILLINOIS

June 1

THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION BY

ROSS. ELIJEE CULL.III.G.S _..an.a CEAEL'^IS EU.GElffi liCCOHLIACZ..

ENTITLED TM l!:J^5^CT..._Q£...VQLTAG^ UMALAIICE QM. TEZ QUWl OF A..

_T..7Q.rPHA3,K lilDIJCTI.QIJ IffiT.QE...

IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS FOR THE

DEGREE OF Bachelor oi ijcience in j^lectrical jiiiiglneering

APPROVED:

HEAD OF DEPARTMENT OF ii.LECTRIGAL EKGIIIEEE.IKG..

167155

TABLE 0? CONTENTS

Pa^e

Introduction ^— /

Theory of Hotcr Operation j

Description of Apparatus : 3

Tests

Rating of D.C. Generator ^

i^ethod of obtaining Unbalance G

Discus sion 7

Conclusions d

Connections

Data i3

Curve s zs

IKTRODDCTION

Polyphase induction iriotors have rr.ar.y advantages over motors of

other types and so have coire to be v:idely used as a source of mechan-

ical pov;er. They are very simple in construction as they have no slid-

ing contacts and no v/earing parts except the bearings. As there is no

commutator or brushes, there is absolutely no trouble from sparking.

These motors can therefore be used amongst inflammable material, in

dirty places etc., vfith perfect safety and Viithout any trouble of

operation. These motors are almost indestructible. They are simple of

construction and of operation and cost very little for attendence and

maint ai nance. They run at nearly constant speed for all loads, have a

high efficiency and a good pov;er factor. Thus it can be seen that in-

duction motors are suitable for practically universal use.

These motors are mostly polyphase, and so take poner from polyuhase

circuits. The different phases of these circuits may vary in term;inal

volt8.ge by a quite considerable percentage due to the fact that one

phase may be m.ore heavily loaded than the others and therefore have a

greater line drop. The effect of this difference in phase voltage on

the operation of tvvo phase motors, especially as to heating effect,

is the subject of this thesis investigation.

THEORY OF MOTOR OPERATION

An induction motor consists of a primary, or stator winding and

a secondary, or rotor, winding. Polyjihase alternating currents are im-

pressed upon the primary Vfiinding, producing a rotating state of magnet-

ism; in the lam.inated iron of the stator. This rotating wheel acts upon

the short circuited rotor conductors, producing currents in then;. The

action of this field upon the currents in the rotor produces a torque

which tends to speed up the rotor to synchronous speed. The rcotor

never quite reaches synchronous speed, however, for at thaf speed ther

would be no lines of force cut by the rotor, no currents produce in

the secondary, and therefore no torque to keep the r.achine running.

Therefore the rotor lags behind the rotating field, or slips an air.ount

varying vvith the load on the motor. The synchronous speed of a motor

in revolutions per second is n = 2f/v, v.'here f equals the frequency

in cycles per second and p equals the number of poles. In the rotor

under consideration p equals 6, f equals CO, therefore n equals 20

R. P. S. , or 1200 B'. P. U. The full load rated speed is 1120 E. P. M.

or a slip of 6.66 percent.

At starting, the currents induced in the rotor cf an induction

motor are very large due to the larje induced 5. I«. on account of

the rapid cutting of flux by the conductors. Therafore, for starting,

the urirrary 3. U. ?. should be cut doicn co a value belov; normal, say

60 percent, in order to keep the c' rrent 'caben by the motor from being

excessive. There are several methods of doing this, such as auto start

ers, starting compensators, etc, but any means of lovjering the supply

voltage will do. For instance, in operating this particular motor the

voltage of the alternator supplying the pcver was cut down by means

of the field rheostat. This was an exceptional case, however, and it

cannot ordinarily be done.

DESCRIPTIOK OF .i^PPARATUS.

The motor tested in this work ttss a Kestinghouse 7. p E. P. , two

phase induction motor, type CCL, taking 10 aznperes per phase at 60

cycle, 400 volts, and running 1120 R. P. Iv!. It is characterized by

strength of parts, high starting torque, large overload capacity, low

operating temperature, nearly constant speed, high efficiency and good

power factor. The rotor was of the squirrel cage type.

The motor was driven by power taken fron; a substation set. This

set consisted of a 49 K. W. , 220 volt D. C. si:u:it iBotor using current

from the po7;er Silant mains and driving a 4p K. 'A. tv7o phase, 440 volt

alternator. By means of the field rheostat the voltage on the alter-

nator could be dropped to a lov; value to be used when starting the

induction raotor in place of an auto' starter.

Load was applied to the induotion :rotor by nie?*ns of 2 belt connect

ion to a 110 volt, ^2 ampere, 1020 R.P.y'i. separately excited direct

current Northern Generator. The generator v;as loaded by ineans of lamp

banks connected in parallel across its terminals.

Two transformers, 10 K. W. , 440 - 220/110 volts were used as auto

transformers in order to obtain unbalanced conditions. Also two ssiall-

er transformers of one Z.Ti. capacity, 440 - 550/47p volts were used

for the same purpose. The method of using these transformers will be

be explained ir.ore in detail later. The instruments used in this test

were of the portable type, Weston make.

Tvio sets of tests v.ere run on the induction rrotor. During one the

tsraperature rise was kept constant at its maxiinuii; point, 50° C. rise

in the windings, and the rraxiraurj load which the acotor could carry was

deteriained. In the second test the output was kept constant and the

temperature rise and operation of the motor noticed. Tn both sets the

unbalance was varied over as large a range as possible.

The connections used in these tests are shown on page |0 . Po'.-rer

was taken fro- the depart nfient substatior and delivered at the irotor

in an unbalanced condition. The rcotor drove a D.C. generator as load.

The generator itself was loaded by aeans of lamp banks. For each run,

readings of current input, voltage and TJatts per phase into the A. G.

niotor Tfere taken, and the load and temperature rise observed. The

limit of temperature rise was found to be in the windings in each

case. The allovfable rise was 59° ^- above a standard 25° 2. root." LeiTir;-

erature. For each degree variation of room teiTiperat ur e froni the stand-

ard 25°, a correction of one-half of one percent per degree was iiiade

in the observed rise. Thus for a 3p° room temperature, the observed

rise was decreased by five percent. Ther moaet er s v;eve used to detern;-

ine the temperature rises, placing one on the fratne, one on the lamin-

ations, and tvjo on the '.bindings. After shutin-| down one was placed on

the rotor and one in the bearings. V;hile running a test, readings were

taken every 15 minites until the t err:per at ui'^e s became constant. The

limiting rise was found to be in the windings, as stated above. The

highest of the tv.'o t her n'Oinet er readings og the winding ten:perature

was taken as the correct one.

For constant temperature rise, the unbalance nas varied fvoz

to 41 percsnt. For constant output the unbalance could only be carried

to 27 percent owing to the limiting temperature rise.

RATIKG 0? D.C. GSITERATOR. i

The D.G. Generator supplying the load for the .-T^otor had to be

rated for different conditions of operation as to speed and excitation,

in order to find the losses in the machine itself. To rate the ^ener-

ator, it i-jas run as a separately excited niotor aith an adjustable I

resistance in both the field and the armature circuits. The ilux vjas|

kept constant by keeping the fi-^ld current constant, and the notor

mas run light at different speeds by shifting, the arir.ature rheostat.^

The input to the motor was taken by Trsans of an acmster and voltmeter

M

in the armature circuit. The product EI ras equal to the losses, i,

y I'

stray poTJer, and a small j. ^; loss. The arr^ature resistance Tfas meas-

ured by the fall of potential method. The stray po'.fer losses, vrhich

varied as the generated E.M.?., 7;ere plotted as shown on curve sheet II.'

Different curves were plotted for different values of field current.

An I-^R loss curve ^vas also plotted as shown on curve sheet I. As the

'i

D.C. machine was rcn as a separately excited generator, no notice 1

v?as taken of the field losses at all. '

To detertnine the output of the induction irotor, then, fet any

load, all that was necessary was to read field and armature currents,

arciature terininal voltage and speed of the D.C. generator. iVror the

curves, for this speed and niagnet izat ion, the stray power losses could

r nbe read directly. The ' "" ''ossss could be read froir I R curve. The

suit: of these losses, plus the output of the generatoif, t'lus three

percent allowed for belt losses, gave the output of the induction irotor.

As an example of this, let us say that the generator was deliver-

ing 49 amperes at 110 volts at a spe3d of 1020 R.?.}.L., T?ith an exoita

tion of 1.75 aicperes field current. Then froE the curves the I'^R loss

are o30 'kvatts, the stray po7>'er losses are 43O ivatts.

Output = 40 X 110 = 4400 ;7att3

Input to generator = 4400 + 450 + 53O = 5380 watts.

Allovjing three percent for belt losses the output of the inducti

motor is

1.05 X 5530 = 5550 viatts or 7.45

If S,;^ = E3 = 400 volts, Ij^ = I3 = 9.5 amperes, Ka^^ = XgFiE = 5.

Total X.?i. =6.2.

CoiriDeroial efficiency 4= 5550/620O = 39- 5;1.

Total volt amperes - 2 x 400 '< 9. = 7600.

Apparent efficiency = 5550/7600 = 75;'i.

2.(ETH0D 0? OBTAINING UNBALANCE.

The obtainin,^ of the unbalanced conditions '.vas one of the prob-

lem-s of this thesis v-ork. The method decided upon vjas to use trans-

formers to step up and dov.'n the voltage on one phase by ceans of an

auto-transformer connection. As can be seen in description of appar-

atus, there vjere two different sized t cansf orniers used, giving diff-

erent ratios of transformation. These transformers were connected up

and combined in different manners as is shovm on connection sheets,

pages I I toJ5, to give the different unbalances. At first sight, it

Tsould look as if the sTnallearft transformer, of one K.Vi. normal caoaci-

ty would not carry such an overload as was put upon it, but this is

explained by the fact that in the auto connection the greater part of

the current only goes bhri^ugh the lov; voltage coil and so does not

cause much heating effect in the transformer.

As an exarafle of the method of connscting of connecting up these

transformers, take the case of 22 percent unbalance, where the voltage

of one phase is reduced from 400 volts to 512 volts. One end of the lovi

tension coil is connected directly to one end of the high tension

coil; the free end of this lovj tension coil and a viire froni the other

end of the hi^h tension coil run directly to the induction motor. This

connection reduces the voltage in this phase as the S.M.P. generated in

the lov? tension coil opposes that of the high tension coil. The other

phase is taken to the i?.otor directly iron the 400 vol I. bus bars. Other

conneotionB r:ere irade as shov.'n on diagram sheet.

DISCU3cICr CF CUFV??.

Plata?) I and II have to deal v?ith the rating of the B.C. generator

vihich supplied the load for the induction ir.otor. Plate i sho'.vs the i R

loss in ??atts for any current in the armature, and Plate II shows the

stray po^fer losses at any speed and magnetization. These curves shov.'

that the stray power increases ?iith an increase of speed, vjith field

current constant.

Plates III to VII show the effect of unbalancing upon the motor

vfhen run at constant full load output. Plate III shows hov; the temp-

erature rises v;ith an increase of unbalancing. It is noticed that for

a 30 percent unbalance the temperature rises about 100 percent above

normal running temperature. This motor operates under low tempera-

ture. In other motors such a rise vfould be exceedingly dangerous.

The curves on Plate IV shoivs hor: the efficiency drops off with

increase of unbalance. There is about a 10 percent drop in both true

and apparent efficiencies for 28 percent unbalance.

Plate V indicates how the T;atts in one phase, the high one, in-

crease, vjhile those taken by the lov; phase decrease as the differer.ce

in voltages of the phases increases.

Plate VI shov,'s that the speed remains nearly constant. Also the

power factor, total ivatts divided by total volt-anperes is nearly cons-

tant, as is indicated din curve sheet 7«

Curves VIII and IX show the operation of the cotor with a constant

limiting temperature rise. They shov; clearly that the output and the

efficiencies of the motor drop off quite considerably v?ith increasing

unbalance. The output varies inversely with percentage unbalance.

COKCLUSIOKS.

The results of this thesis investigation can be sumned up as

f ollo^Ts:

For a constant output of the motor, an increase of unbalance re-

sults in a iTiore than proportional increase in temperature rise, and also

in a decrease of efficiency. It has practically no effect on the speed

however.

For a constant t err^per at ure rise an increase of unbalance raeans a

decrease in output, in a straight line ration, The efficiencies also

fall off in value with increase of unbalance.

Although these tests vrere run only on an induction irotor, it can

be seen that the results vrould be socerrhat the sane for a synchronous

motor also, and for any polyphase machine as ?;ell as a tvfo phase irotor.

In actual practise such a large unbalance is seldom if ever Eet with.

The vailtage variation is generally limited to IC or Ip percent, ^here-

fore for a motor giving constant output, as is generally the case, the

increased tem.perature rise kouIc not be dangerous. The greatest effect

of an unbalance -nould be the loss in efficiency resulting therefrom.

NesflnQhouse Tvvo Fhase Induction Nofor:

Typz CCL

DIFIGHfiMS OF Tl^m^FOI^MLR CONNECTIONS

'a

< 3/£ V. >

To Nlotor

foo V

To Motor

22%Unbaianca

i-oo V.

230 V

76 Motor

'^OO V.

To Motor

30 Unbalance

400 /

—i

^wmmmm

. 2^(^V. ^

^1% Unbalance

hnmnn

^% Unbalance

Dc^ffcf for flp/tinqof

jO.O. Gcnerctto r.

i^c,/US/V'-av F?3u/e.r

Z3S // oo

^.3S /Z 3o

£.3S / OB.o 3,8 760 i- /o

/ Z7.0 -i-.o SOQ/30.S 3. So

/33 3,S3r /

o

/^^ / OSO 3'<a^

/ss / /

/ ZjT 3. BO //so—

3rfo

/ 7:r /Z3 3. <^ o /o -fa

/ 7S / /a.3r 3.3rjr S 60 iZo/ 7S

1

/OS. 3. S3S 36Z

is.

1

1;

On

00

K

CD

5?

is

0:

H

i

vo

<3i

VO-

CO

K

CO

vo

vo

VQ

vS

vo

K

«0

vo

CO

O

^0

VO

M

\

CD

CD

CO

VO

o

v^

Vo

?^NO

VVo

?;

«^N

vo

>

A)

>

a:- o

CO

Hi^\

"V.

\

^1

\1

\1

\1

o O

H1 K

"<

I

1

/feacf/'/ys ^or Consfar?/Oot-/>oi~ af /fo/J/y^^a/a/ice

limeiLanynanons

YYindii

Afo/7fS

\— —

'

rramc

7:00 3/X

37^

7: So

^/:/

So1

- -

Uegrees C&ntigrade

77/??e. LGm/nat/onSram

No /

^//

70:50 so y/

¥/j

//: 00 S3.

6'^.

T/me /Vo/ /VcZ /Yaam.

7-30 ^/ 32.

^7 37^0

r:/6' dZ.l

f: 30 ^3.f

/feac/zn^sj/-or Con^^onf Oof-/>of af'Z/.Z% l//7/l?a/a/7cc

JJc^re.e.s Centigi'cicJer

T/me Lam/'nafi'on /y'a/7?is /Yao/77

J7 0-7

^'7f rS.r

dr.7

/o:oo 07s- ^13

/y / ff w//an^/

/ Jj99 ^/ / *rT ^

7 :

sc>.& s c.S^ Za,3 S '. OO

sa,s SI. s a: /jT

^S. 7 jre,o SB.

Sa.o SS5 Z7. 9

1

SB. 1 sa.3 X7.S d:oo

JJegrce^s Centigrade.

^/

-to. o 6"^ 6 0.3- Z 7.3- ^. /-^

JTa. o C3.S J>:3d)

1 Lo X^7 4

70. 7^. <^7 Z 7 3- 7£> '. o o

7 0. yo-.yjr-

7 0.3 73. ^3. Z 7.y /o:3(?

S3.3 7 0.3 73. GS. /o:4S

/ s

0)

0)

^ ^ o o^ ^ ^ ^ ^ OV V V >S I >k

^ o^ > \ 00 K)

^ rv

5:^

^ ^< ^< ^N Q\

^ S

CO

Kb

^ 00 (Q

<0

X N

I

^1

^ ^ ^

o

CO

CO

CQ

CO

5

CO

00

0>

«0

kS

;^\Q

t QJ

^^ p

\

v\ ^

o

KVs \

\ \

\J

\ \

<31

\

]

1

Tim (2. Lamination jVo.l1

•\ —

Frame7:^00 31 36.3 za/

jr.

7:56 67(;

Zc}-73^

^7:r

f:fS'

(^((^ 7/. zos^;/^' 6Z 3 70

^: 30 a 70

(Keadin^s f-or Constant Temjoeratare /T/JC

at /7o Unba/onc e. JJesr^cs Cen/ioradeL

f\eQi^/njc>'^'orConstant Tie.mJoemfc/re /F/je

at "^70 Unbalance,

JJe^rees Cen-ti^rade

1

LaminaXionHindi

No./ Frame /faom

^/ 367

fioor

^0 37 ^^.^^

r: /^^;7a-^

?: 30 ;i3. /

r: ;i3 f

</: ^7: ^7

9:/6"

9: 30 J 3.x

' /o:dd1 1

7<^'^'^

1

^7 ^3.Z

IXeadin^s for Constant Temjoerpi-urc /j/'5e

ot /"^.s^o Unbalance.

Ue^rees Cenh^rac/e

Time ijjminotionsWrndi 1^5

Frame ootn.

7:00 of

^0 7o ^fl7; 30 ^/ yr5

^/ 7^'

3

7"/. ^r.

^:3o 7f.6 7.^1. r

/headings ior Constant Tem/oerat/jre /f/seat % Unha/ance.

JJegrees C<°nfigrac/e

Laminaf/Qn No I JSoZ Frame /foon^

j'f. 70

9; so Sf. 7^

Sf. 73

/o:oo•

73

/Yecid/n_^5^or Constant Jen^jherat/jre /f/'se

Qt 3o fo Unha/ance,

Time_ lam/nation5\ /y^J^'^^^' /foo m.

3/

^/;? 33.f

str 7/. 2

^:3o 7/-^, 7^ ^^7

/re <5 /-or Consfant Jeiv/^^erafure /T/se

at i'/ ^oUnha/ance.

Time Lcinvnafions Wo. 2. ffocnn

70.

9:

r7J~

7/,¥' 7^: y^

KUCCNC DIETZGCN CO.. CHICAGO.

'm

4-'

ir-'

4> f'

' 4- ^ ^

' 4I- 4-.

f f|

I*it 4 ^. 4 ^ -i-

I J. ^

f f ^ f

^ -r V-

ii*' ^ , 4j'4' ^ f

4^

>i(^ 4 T- il^^

> >''^^

< ^ i '

'

....

'if--