150
HOUSEL Behavior of a Rotor in a Two-Phase Field Electrical Engineering B. S. 4 " J* : ' >- 1901 UNJVEKi'ITV OF TU-. UBRAM

Behavior of a rotor in a two-phase field

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

HOUSEL

Behavior of a Rotor

in a Two-Phase Field

Electrical Engineering

B. S. 4 " J* :

'

>-

1901

UNJVEKi'ITVOF

TU-. UBRAM

+ ****** I § I* * * * * * * * * * mmm * * * *\s-<Wimmm lIKf * * * * # ^ * * # *****,************

jf| * * * "SiHr 1toW* *' *

#* * * * * * * * 1 1 I

;

I . I I I |p |i

* i al # * lllifl 4 1 I 1 i I H* * ^ # * * * f + * # * * + + • * ^

^mmmm^m^^m* \ 1 MiSilBH 1 #; ;

* IBf F # # # f

t ^ T * * f"

I §

* * * *

* # # #* f«

air- . j£

****** *##*.#'^ .a #• # # 4. 4: •# 4* #

ir w w 1framing anb $Tabor

LIBRARYOF THE

# University of Illinois.|

^CI,A|. BOOK. VOLUME.

TF ~ 3F

'

4.' * 4 1 A\:W W W IF

*r ^ W

****** yao'l =»h*\at Jt A tr r

TP "T T-

# # ih f*

^ ^ ^# #= *r # 1r ^ * ^ ^ ^* Jk^:

*6 >Xl *A at JL ^ -de Jl ' & d

f* ^ * ^ ^ ^ ^ $ ^ ^ ^ ^ ^ ^ ^ ^

I

Accession No.

^. j!

^ jt 4. ^

.* * * * -1* * * * 4r 4 f************** * * * ^ T

J^f '^i^l

*

mm

# ^ ^

#i

4

V J

-- ;v

'

:

- =^ '

<*, j

* : T IF--- . f

-

• W' * '^li^"!/^r"-"-.:.' /-^r' • ^ T

Ilb

if

^* * * #

Digitized by the Internet Archive

in 2013

http://archive.org/details/behaviorofrotoriOOhous

BEHAVIOR CP A ROTOR IN A TWO-PHASE FIELD.

By 0. L. Housel.

THESIS FOR THE DEGREE OF BACHELOR OF SCIENCE IN

ELK CTR I CAL ENG 1 1 IEER IN

G

in the

COLLEGE OF ENGINEERING

UNIVERSITY OF ILLINOIS.

Presented JUNE, 1901.

tf'SA

UNIVERSITY OF ILLINOIS

May 31, 1901 190

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

Oboar Lloyd PIousel

Behavior of a Rotor in a Two-Phase Field

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

Bachelor of Science in Electrical Engineering.

>

PREFACE.

In deciding upon the subject for this thesis there seemed

to "be many reasons for selecting the one chosen. The use of

•induction motors is advancing in all lines of the industrial

applications of electricity. Among such fields of application

may be named: pumps, fans, blowers; printing presses; all

classes of machine tools as milling,boring, shaping machines,

lathes and drills ; elevators , hoists and cranes. The outlook

for its future is very promising, especially along street

railway development.

The writer being especially interested in the transmission

of power in mills and factories, selected the induction

machine as a representi tive motor for this field. The

facilities in the Electrical Engineering Laboratory also

being provided for this work the experiments were carried out

as herein indicated.

Tests were made on a V/estinghouse , type C, two-phase, 60oo;

two horse-power induction motor. These experiments were

carried on at different voltages as 105, 110, and 115 volts.

The torque was varied in each case from zero to a certain

maximum depending on the voltage.

The original intention was to make similar tests at 210,

220, £30, and 420,440,460 volts. For lack of time the latter

series had to be omitted. It was expected that tests would

also be made upon a V.'es tinghouse type P, two-phase, 60c>o,

three horse-power, variable speed induction motor. The

manufacturers were unable to ship this motor in time to "begin

the experiments.

In conclusion it may be said of the thesis investigations

•carried out as shown in the following pages:

(1.) Extremely light weight, 250 pounds rated, for maximum

power which it was found capable of developing,

3.4 Kilo Watts at 115 volts.

(2.) The slip was, at normal rated load, about 6 1/2

per cent.

(3.) Under all conditions of the trials with various

impressed voltages the starting torque was in excess

of that usually allowed for machines of this size.

(4. ) The real efficiency showed a high maintained value

of about 80 per cent, in general for a wide range of

load, from about 50 per cent, underload to about

25 per cent overload.

(5.^ The power factor showed a satisfactory maintained

value of about 85 per cent, from about full load to

maximum overload.

iii.

SYLLABUS.

INDUCTION MOTORS.

I. ^DEVELOPMENT OF THE INDUCTION MOTOR.

1 . -Production of rotary or rotating magnetic fields; 1, 2,

or ^-phase.

2. -Interaction of magnetic field upon its induced current in

conductor.

Spinning disks Bahbage, Arago, Faraday ; wattmeter disk.

Eddy currents in spinning disks of copper and aluminium

Eddy currents in copper or aluminium cylinder,

continuous or split, as the rotor of the induction

motor, of squirrel-cage type.

3. -Early types of induction motors:l, 2, and 3-phase.

4. -Example of principles of induction motor in wattmeters.

II. -THKORY AND PRINCIPLES.

1. -General principles underlying action of induction motor.

2. -Relation of induction motor,general and as two-phase :

-

(a.) to the alternating current transformer,

(b) to the direct current shunt motor.

3. -Theory and principles of special types of induction motors.

Single-phase :Brown,l.'agner ; Two-phase , Ferraris ,Tesla;

Three-phase, General Electric.

III. -CLASSIFICATION.

1. -General: Rotary induction apparatus which consists of

primary and secondary windings , rotating with respect

to each other, comprising:

(a) Induction motors.

(b) Induction generators.

(c) Frequency changers.

(d) Rotary phase converters.

-Induction motors with variable resistance in armature or

rotor.

-Induction motors with short-circuited armatures, squirrel-

cage rotors.

IV. -ELEMENTS OF DESIGN.

-Allowances, constants and coefficients.

(a) Leakage, leakage factor and air gap.

(b) Inductance of s tat or and rotor.

(c) Core losses, hystersis and eddy currents.

f R(d) Copper losses: Equivalent resistance — —'— for

calculating copper losses: value of this method:

dependance on voltage, load, frequency, etc.

Electric and magnetic circuits of machine.

Mechanical and structural details.

V. -CONSTRUCTION.

Electrical details, windings, diagrams, insulation, etc.

Magnetic considerations, lamination of cores, balance of

rotor.

Mechanical and structural details; static and dynamic

balance of rotor.

Description of induction motors in commercial use.

V.

5 . -Accesoriee

:

Phase-splitting devices for single phase motors;

Starting and switching-in arrangements, compensators.

Variable speed control, multi-speed motors.

Use of resistance, induction and capacity in starting

and in control.

VI. - INSPECT I OK.

1 . -Mechanical , electrical, and magnetic.

VII. -OPERATION.

1 . -Starting : effect and value of starting devices ; torque

,

current.

2 . -Functional working; normal and overload conditions.

3. -Effect of varying the impressed E. L. F. , on starting and

operation loaded.

4. -Cacpac ty of 2- and 3-phase machines operated single-phase.

5. -Reversal Of 2- and 3-phase motors.

6 . -Faulty working and remedies.

7. -Influence of frequency , speed and slip ; variable speed

performance.

8. -Regulation, control of vol tage ; effect on lighting circuits.

VIII. -PERFORMANCE AND TESTING.

1 . -Principles and methods employed : electrical , magnetic , and

mechanical.

2. -Errors,precautions , precision of measurements.

3. -Effect of variables on performance : voltage , frequency

,

phase shifting.

Functional performance : characteristics , losses , efficiency.

Starting performance , start ing and crossing devices:

current, torque, etc.

Electrical characteristics of performance: current ,E.L. F. ,

power and power factor, true and apparent efficiency,

torque efficiency, phase difference , idle or "wattless"

energy.(a) Losses , determination and separation of.

(b) Impressed E. M. P. - dependence upon of speed,

maximum torque, slip, power, power factor , efficiencies

and various losses.

(c) Frequency,- dependence upon, of speed, maximum

torque, slip, power, power factor, efficiency and

various losses.

(d) Current values: magnetizing current, idle or

"wattless" current , starting cutrent , induced rotor

currents.

Magnetic determinations: indue t ion, leakage , leakage factor,

air gap, core losses.

Performance,general characteristics of,- electrical,

magnetic, mechanical.

(a) Deductions and conclusion from behavior performance

and test.

(b) Do results of test corroborate principles of

action and constants used in designs?

vii.

IX. -SPECIAL STUDIES OF INDUCTION MOTORS.

1. -Analysis of action of induction motor, application of

diagrams , etc.

2. -Establishment of laws of induction motor, electrical,

magnetic, and mechanical.

3. -Phase splitting and phase splitting devices.

4. -Induct ion motor or an induction generator.

5 . -Concatenation of induction motors.

6. -Induction motors as frequency changing transformer.

7. -Influence of wave form upon iron losses.

vii i.

BEHAVIOR OP A ROTOR IN A TWO-PHASE FIELD.

TABLE OF CONTENTS.

CHAPTER I*

• DEVELOPMENT OF THE TWO-PHASE INDUCTION MOTOR.

Article. Page,

1. Production of a two-phase rotating magnetic field. -1.

2. Interaction of a magnetic field upon its

induced current in the rotor. -4.

3. Early types of induction motors. -7.

CHAPTER II.

THEORY AND PRINCIPLES.

1. General principles underlying the action

of the induction motor. -10.

2. Relation of an induction motor to an

alternating current transformer. -\7,

3. Relation of the induction motor to a direct

current motor. -14.

CHAPTER III.

CLASSIFICATION.

1. Classification of rotary induction apparatus. -16.

CHAPTER IV.

CONSTRUCT! ON.

1. Electrical and mechanical measurements. -17.

ix

CHAPTER V.

PERFORMANCE AND TESTING.

Article. Page.

1. Methods employed. -21.

2. Errors and precautions. -24.

3. Effect of variables upon the performance -27*

of the induction motor.

BEHAVIOR OF A ROTOR IN A TWO-PHASE FIELD

WESTINGHOUSE TWO-PHASE, 2 H. P. 60c*J, 110, 220, 440 VOLTS,

INDUCTION MOTOR.

General Performance and Efficiency Determinations.

Experiment 1. Performance at 115 volts.

Experiment 2. Performance at 110 volts.

Experiment 3. Performance at 105 volts.

xi.

BEHAVIOR OF A ROTOR IN A TWO- PHASE :FIELD.

INDEX OF TABLES.

Table I. Experiment 1. Voltage 115. Sheet Ari. Page,

,

31.

Table II. Experiment. 1

.

Voltage 115. Sheet B. Page »O

Table III. Experiment 1. Voltage 1] 5. Sheet C. Pacre 9

•7 <7.

i . <..«

Table IV. Experiment 1. Voltage 115. Sheet D. Pace >o 34.

Table V. Experiment 2. Voltage 110. Sheet A. Page >

Table VI. Experiment 2. Voltage 110. Sheet B. Pace . 3fl.

Table VII. Experiment C. Vol tage 110. Sheet C. Pace , 37.

Table VIII. Experiment 2. Voltage 110. Sheet B. Pace . 38.

Table IX. Experiment 3. Voltage 105. Sheet A. Pace . 39.

Table X. Experiment 3. Vol tage 105. Sheet B. Pace . 40.

Table XI. Experiment 3. Voltage 105. Sheet C. Page , 41.

Table. XII. Experiment 3. Voltage 105. Sheet D. Page , 42.

XI 1.

BEHAVIOR CF A ROTOR IK A TWO-PHASE FIELL.

INDEX OF PLATES.

Plate. I.

Plate. II.

Plate. III.

Plate. IV.

Plate. V. .

Plate. VI.

Plate. VII.

Plate. VIII.

Plate. IX.

Plate. X.

Plate. XI.

Plate. XII.

Wiring diagram of machine switchboard,

showing connections for 110, 220, and 440 volts.

Page, 20.

Brake wheel. Page, 22.

Photograph of rope brake. Page, 23.

Wiring diagram of table. Page, 25.

Photograph showing arrangement

of instruments. Page, 26.

Experiment I. Voltage 115.

General performance, Page, 43.

Experiment I. Voltage '15.

Analysis of one branch, Page, 44.

Experiment I. Voltage 115.

Analysis and vector relations., Page, 45.

Experiment I. Voltage 115.

Equivalent resistance ., analysis , Page, 46.

Experiment II. Voltage 110.

General performance, Page, 4?.

Experiment II. Voltage 110.

Analysis of one branch, Page, 48.

Experiment III. Voltage 105.

General performance, Page, 49.

Plate. XIII. Experiment III. Voltage 105.

Analysis of one "branch, Page

y.iv.

EXTENSION OP THE DEWEY DECIMAL CLASSIFICATION TO

INDUCTION MOTORS.

537. Electrical engineering.

537.8 Application of electricity.

537.83 Dynamic electric machinery.

537.838 Rotary, induction apparatus.

.8381 General types and descriptions.

.8382 Theory of this type of machine.

.8383 INDUCTION MOTORS.

301 Materials used in induction rotors

3011 General materials.

3012 Conductors, vires, and cables.

3013 Conducting materials.

3014 Insulating materials.

3016 High resistance materials.

309 History.

3001 General statistics. Data.

30011 Statistical information.

12 Statistical tables.

13 Statistical charts.

3002 Quantities , costs , economics.

21 General economics.

23 Costs, charges, rates.

3003 Contracts and specifications.

XV.

537.83830037 Guarantees.

3004 Designs, drawings, calculations.

41 General instructions, etc.

42 Formulae, principles.

43 Calculations, designing.

44 Methods of designing.

45 Electrical designing.o o

46 Magnetic designing.

47 Mechanical designing.

49 Drawings , examples applications of design.

3005 Executive and constructive.

51 • General productive work.

55 Inspection.

551 General inspection.

552 Mechanical inspection.

553 Electrical inspection.

554 Magnetic inspection.

30056 Constructive details.

561 General features of construction.

562 Electrical details. (Windings.)

563 Magnetic details.

564 Mechanical details.

57 Manufacturing , workshop productions.

58 Erect ing , ins tal lat i ons , etc.

3006 Working, maintenance, operation.

61 General performance.

xvi.

537.83830062 Manipulating, handling and starting.

63 Functional working.

64 Operation, operative work, running.

641 General operation.

648 Operation in parallel.

643 Operation in series. Concatenation.

65 Regulation, control, etc.

651 General considerations of operation and control.

65? Use of non-inductive resistance.

653 Use of inductive resistance.

654 Use of condensers.

66 Faulty working and remedies.

661 Break down.

68 Maintenance.

69 Repairs and renewals.

3007 Measurements, tests, testing, research.

71 General instructions, rules, etc.

73 Errors , precauti ons , precision of measurements.

74 ! lethods , electrical , mechanical , magnetic.

75 Determination of coefficients and constants.

76 Characteristics of performance.

7tl General characteristics.

762 Characteristics of magnetization, indue tion.

7 621 General.

7 622 Armature induction, reaction. Rotor induced

currents.

xvii.

537.8383007 6222 Character of magnetic field.

7 624 Electromotive force curves. ..ave forms.

7 6241 General features. Influence upon iron losses.

763 Economic characteristics.

766 Mechanical characteristics , slip, moment of

inertia, torque, etc.

767 Power characteristics ,horse-power curves.

7 68 Regulation characteristics.

7 69 Resistance characteristics and inductance

measurements.

30077 Loss measurements , efficiency , and duty.

771 General.

772 Stray power.

7721 General stray power determinations.

Frictional and mechanical losses.

7723 Iron (Core) magnetic losses. Hysteresis.

7724 Eddy current losses.

7725 Copper losses.

773 Efficiency determinations.

3007735 Mechanical measurements of efficiency and

losses.

77351 General efficiency considerations.

77352 Absorption dynamometers, pronv brake,

rope brake.

30078 Results , tables , curves , conclusions of tests.

79 Reports of tests , experiments , research,

XVI 1 1.

investigations.

537 . 8383008 Installations , accessories , ind ividuaJ

,

aggregate.

81 General equipment of plants.

82 Starting devices.

821 General considerations.

822 Starting devices for single-phase motors.

8221 Phase-splitting.

82212 Shunt phase-splitters.

822121 Resistance and reactance.

822122 Condensance and reactance.

82213 Series phase-splitters.

822131 Resistance and reactance. •

822132 Condensance and reactance.

82214 quadrature winding. (See A. E. Oct. 1900.)

822141 Resistance and reactance.

822142 Condensance and reactance.

85 Protective devices and safety devices.

86 Regulation and controling devices.

861 Regulation of speed. General considerations.

862 Variable impressed voltage.

863 Variable non-inductive resistance in rotor.

864 Commutatec field. Series parallel arrange-

ment of stator coils.

XIX.

537.83830087 Measuring instruments,

88 Listridut ing arrangements , switchboards , etc.

89 Miscellaneous.

30007? National electrical code. Installation of

induction motors.

.83831 Single-phase induction motors.

32 Tv/o-phase induction motors.

33 Three-phase induction motors.

XX

BEHAVIOR OF A. ROTOR IN A TWO-PHASE FIELD.

BIBLIOGRAPHY.

POLYPHASE ELECTRICAL CURRENTS AND ALTERNATING CURRENT

MOTORS, Ey SILVANUS P. THOMPSON, London, 1900.

ALTERNATING CURRENTS AMD ALTERNATING-CURRENT MACHINERY,

By D. C. and J. P. JACKSON, New York, 1896.

POWER DISTRIBUTION FOR ELECTRIC RAILROADS, By Louis BELL.

Third Edition. New York, 1900.

ALTERNATING CURRENT PHENOMINA,

By C. P. STEINMETZ. Third Edition, New York, 1900.

STANDARD POLYPHASE APPARATUS AND SYSTEMS

,

By M. A. OUDIN, New York, 1899.

1.

CHAPTER I.

DEVELOPMENT OF THE TWO-PHASE INDUCTION MOTOR.

Aft. 1. PRODUCTION OF A TWO-PHASE ROTATING MAGNETIC 'FIELD.-

It is a well known fact that a current traversing a coil

of wire causes a field of force to be set up at right angles

to the conductor. Suppose that this current is an alternating

current ,periodically varying from a positive maximum to a

negative maximum. The field of force set up in the coil Will

also vary in a like periodic manner and be in quadrature with

the electromotive force of the current. The electromotive

force and the current would be in phase if it were not

influenced by eddy currents' in the iron core and secondary

currents induced in the adjoining coils which oppose the

primary action and cause the current to lag in phase behind

the electromotive force.

Nov/ consider two similar coils placed so that their

planes are at right angles to each other, as in fig. 1." Send

through each coil a direct current.

A field will be induced at right

angles to the plane of each coil.

These fields can be represented by

a vector. Since the coils are at

right angles these vectors will be

also at right angles. From these

'"'"The Rotary Magnetic Field," American electrician. Vol. XIII.

Number 3, March, 1901, p. 132.

vectors "by completing the parallelogram the resultant can be

found, see fig. 2. Next take the case of two alternating

currents impressed upon the coils, the currents being 90°

apart, or in other words, one is a maximum when the other is

zero, and vice versa. Suppose that the current in coil E is

maximum when the current in coil A is zero, which will be the

case when in quadrature. Then the field induced by rt,

is a maximum and in quadrature with the current, it can

therefore be similarly represented by a vector. "..hen the

magnetic field of coil A is zero that of coil E will be the

resultant field. In' a certain length of time this vector

will decrease and the vector representing the flux in coil A

will increase. Completing the parallelogram, as in fig.?,

and drawing the diagonal the direction and magnitude of the

resultant flux can be found. This resultant makes an angle 9

with the former resultant flux; and as the flux in coil A

increases the flux in coil B will decrease, the resultant will

then move over toward the vector representing the flux in

coil A, and when the flux in coilBis zero the vector will

represent the flux in coil A and also the resultant flux.

The direction of rotation of this resultant vector

depends upon the initial relation of the flux in coil A to

B

(neglecting f or the time, eddy currents and self-induction,)

the flux in coil B. In fig. 4, two waves of electromotive

force are shown in quadrature; and since the flux is similarly

in quadrature it would vary in like manner. For the flux in

Fi 3 . f \* x y x / \ * x\^^^yS S X^

coil A there are two conditions , when the flux in coil B is a

maximum, at the point a, the flux in coil B is zero. As this

flux decreases the flux in coil A must increase toward either

a negative maximum or a positive maximum, it is, therefore,

due to this feature that the direction of rotation is deter-

mined, since negative values of the flux are represented by

vectors toward the left and positive values toward the right.

Considering again, the flux in coil B increases hut in

the opposite direction, while in coil A the flux decreases.

Complete the parallelogram and draw the diagonal for the

resultant as before. The resultant will be found to have

moved on past the horizontal position of the vector of coil A.

As the flux in coil B increases the flux in coil A decreases

and the resultant vector moves on until it equals and coin-

cides with the vector of coil B, reaching again the vertical

position but in the negative direction. The flux in coil A

now increases in the opposite direction and the resultant

vector R moves on until it equals and coincides with the

vector of the flux in coil A. The flux in coil B now begins

4.

to increase in the first direction and the flux in coil A

decreases. When the flux in coil A reaches zero the flux in

coil B is maximum and the resultant R has reached the point

of starting. Hence for each cycle of the two fluxes in

quadrature there is one revolution of the flux vector R. If

the current varies according to a sine wave the relation

between the two instantaneous values of the corresponding

flux is such as to always give the same magnitude to the

resultant vector representing the resultant flux.

Art. 2. INTERATION OF A MAGNETIC FIELD UPON ITS INDUCED

CURRENT IN THE CONDUCTOR.

About 1824, Gambey discovered that a magnetic needle

when set in motion, came to rest sooner if the bottom of the

box was made of copper or some good conductor than if it were

made of wood or some poor conductor. Arago demonstrated

before the Academie de Science , in 1824, by several experi-

ments, the' same phenomenon. Re suspended inside copper and

inside wood rings a magnetic needle:' this was pushed aside

from its stationary nosition to an angle of 45° and then

allowed to vibrate and the number of oscillations counted

until the needle came to rest. He found that in the wood

ring there were 144 oscillations, in the thin -copper ring

there were 66, in the heavy copper ring there were only 33.

This showed the damping action caused by the copper ring. He

-"Polyphase Klectric Currents and Alternating Current Machinery"By S. P. Thompson.

decided that there was a force present which existed only

when there was motion and he called this "Magnetism of

Rotation."

In 1825, Babbage and Herschel, showed that if a magnet

placed under a copper disk which is pivoted and able to move

freely, upon being "revolved caused, the copper disk to revolve

also and with a speed somewhat proportional to the speed of

the magnetf They also noticed that if several radial slits

were cut .in the disk the tendency to rotate was grea.tly

decreased.

It was not until 1831, that a theory was founded which

satisfactorly exlained these phenomina. At this time

Faraday showed that when a disk was revolved in a magnetic

field there were currents set up in the disk and that these

currents, called eddy currents, set up a field of force which

opposed, the field of the magnet and hence caused the disk to

tend to come to rest, according to Lentz law.*" He also

showed, that if a copper disk was placed edgewise between the

poles of a strong magnet, and a copper brush was placed on

the edge and one on the shaft, a current would flow through

the wire connecting the brushes, when the dick is revolved.

Therefore, a conductor moving so as to cut the lines of force

of any field will have induced in it an P. I.'.. F.- which will

cause a current to flow, which current in turn has a field of

Thompson Ibid.

Electricity and Magnetism. By S.P.Thompson. Page

force surrounding it which tends to repel the first field and

in turn is repelled by it; hence, the notion is opposed as

stated in Lentz law. A copper disk pivoted, and centered

oyer a revolving magnet, rotates in the sane direction as. the

magnet. This is explained as follows; when the nagnet re-

volves the lines of force are also revolved and cut the disk

thereby setting up eddy currents in the disk, these eddy

currents in turn set up a field of force which opposes the

field due to the magnet. Since the magnet is revolving and

the field caused by the eddy

currents is reacting against it and

cannot stop it the disk must

necessarily revolve with almost the

sane speed.

The revolving magnet produces

nothing nore nor less than a

revolving magnetic field. Hence, if the revolving magnet is

replaced by a revolving field the same effect will be produced.

The revolving field is stronger and therefore considerable

torque can be produced, as in the induction motor.

If a copper cylinder is now placed inthe rotating field

eddy currents are set up in it which cause it to rotate in

the came manner as the disk. By cutting slots in the cylinder

it is made equivalent to a "squirrel ca.ge" so called, in

which the bars serve as conductors for the eddy currents and

\

7.

and the field set up "by them is stronger and therefore the

torque is greater.

Art.?. EARLY TYPES OF INDUCTION MOTORS.

The first induction motor was made by ..alter Baily, in

1879. There were four electromagnets set horizontally and

close together. Over these and close to them was pivoted a

copper disk. By means of a commutator, which was turned by

hand, a current was sent from the batteries and alternately

reversed in the coils of the magnets, which were connected so

that the diagonal coils were in series with each other. By

this means the lines of force were cut by the disk and eddy

currents were induced in it, hence it rotated. If the

direction of the current in one coil is reversed the

direction of rotation is reversed and if the relative time of

reversals of the current in the two circuits is changed the

direction of rotation is also changed.

In 1885, Professor Ferraris made the first motor which

had a rotating field due to two alternating currents which

differed in phase by 00°. He used two pairs of electro-

magnets connected in series around which was a yoke made of

iron wire. The difference of phase was made by placing in

one circuit a large non-inductive resistance, a, in fig. 6,

and in the other circuit a large inductive resistance ,b , in

fig. 6, and connecting both circuits to a single phase

supply. A closed copper cylinder pivoted between the magnets

8.

was made to rotate and the direction of rotation was changed

by reversing the relative relations of of the spilt single

phase current, as it enters the motor practically as a two.

phase current in quadrature. He

later made a large motor the

windings of the magnets being

circular and entirely surrounding

the rotor. These magnets were

divided into two pairs and the

pairs were placed at right angles

to each other. Ferraris called

the difference in angular veloc-

ity of the rotating field and the

mechanical rotation of the rotor the slip , and also stated

that the current in the rotor was proportional to the slip,

and that the power of the motor is proportional to the slip

and the speed of the rotor. ,

The next step was taken in 1887, when Nikola Tesla

developed his two phase induction motor, the rotating field

being more nearly uniform. A ring of iron was divided into

four sections and around this in each section was wrapped

copper wire. These coils were then connected in series and

to the slip rings of the generator. The armature of the gen-

erator had two coils placed at right angles to each ether.

9.

Therefore, the E. M. P. of the two coils of the generator would

"be in quadrature. When the generator was delivering a two-

phase current a rotating field was consequently set up in .the

field of the ring forming the stator, and the iron disk

forming the rotor was caused to rotate by it.

CHAPTER II.

THEORY AND PRINCIPLES.

Art . 1 GENERAL • PRINCIPLES UNDERLYING THE ACTION OF THE

INDUCTION MOTOR.

Wheal the current is flowing in the s tat or of an induc-

tion motor, the rotor being at rest, a high E. M. F. is induced

in the short-circuited windings of the rotor, which causes

large currents to flow in these rotor windings. These

currents set up fields of force which oppose the revolving

field. Since the rotor is free to turn it tends to follow

the revolving field. If the rotor were mechanically driven

at the same speed as the revolving field, that is, at synchro-

nous speed, the lines of force would n't be cut by its con-

ductors and therefore no counter E. M. F. would be induced in

the windings. Hence, the magnitude of the induced E. I... F. in

rotor windings depends on the relative difference of speed

between the revolving field and the rotor. Now, as the rotor

speeds up, due to the inducer1 rotor currents, the relative

difference of speed is decreased and the induced rotor current

are therefore decreased. The difference of speed, or the

slip, will be such that the current induced in the rotor will

be just large enough to produce a torque sufficient to over-

come the mechanical resistance or work. If a load is belted

to the shaft of the rotor the slip will increase thus increas-

ing the induced current and in turn increasing the torque to

li-

the amount necessary to carry the load.

The speed of an induction motor can he varied in one of

three ways,viz S*

(a) Rheostatic control in the secondary or rotor,

(h) Varying the impressed voltage,

(c) Altering the number of poles.

The first is probably the most expensive yet the most desir-

able for all ranges of speed. ' The windings of the rotor,

similar to a three phase star, are brought to three collecting

rings and the current is lead through a variable non-inductive

resistance with controller, where the energy is dissipated in

heating the resistance. For a certain load there is a certain

current required in the rotor windings to produce the required

torque. If in the external circuit of these windings there

is put a non-inductive resistance which will consume some of

the energy of the current and dissipated it in heat, then the

current, to carry the fixed load, must be increased by the

amount dissipated in heating the resistance. To increase the

current in the rotor windings the slip must be increased and

therefore the speed of the rotor must be decreased. Hence,

the introduction of resistance tends to reduce the speed of

the rotor. The introduction of the resistance also tends to

decrease the angle of lag between the impressed E. M. F. and

current which raises the power factor of the motor.

"Alternating Current Systems" by H.S.Meyer. London Engineering

Vol.71, April 19, 1901, Page 195.

The second method requires a transformer or compensator

having a variable ratio, Which is placed in the external

circuit thus raising and lowering the impressed voltage.

These accessories are very inefficient, therefore this. method

is not a practical one.

The third method is the cheapest as well, as the most

efficient but much more complicated in its mechanical require-

ments. It can only be used where the speed is to be varied

by a series of steps in the ratio of two to one or some .

multiple of this. For every rotation of revolving field, if

it were a bipolar machine, there would be one revolution of

the rotor minus the slip. Thus, a bipolar machine running on

60 cycles and with 5% slip would have a s.peed of 34^0, which

would be excessive. By placing four coils in series in eachpol arity

leg, so connected that they are of alternate Ait will require

two cycles for any N pole to move one complete revolution

around the stator. Hence by this method the speed of the

rotor is just halved. If this division of the poles is

carried still farther, making eight poles the speed will be

one quarter of the speed of the bipolar field; etc.

Synchronous spend is always understood to be the speed

without slip and is therefore, the speed corresponding to the

frequency.

Multi-Speed Induction Motor Changes:

Poles. Frequency. R. P. M. Examples.

2. 60 3600

4. 60 1800 Thesis, 2 II. P. V.'es t inghouse Motor.

8 • 60 900 E. E. Lab. 40 H. P. " " .

16. 60 450

*o 60 225

Art. 2. RELATION OF AN INDUCTION MOTOR TO AN ALTERNATING

CURRENT TRANSFORMER.

If the rotor of an induction motor is blocked, the

rotating field will induce in the short-circuited windings an

alternating current. Around each conductor of the rotor

there will be a field of force, which field, through its

counter E.M. F. , will oppose the primary field just the same

as the primary field of a single-phase transformer is

affected by the current which would be induced in the short-

secondary. If these rotor windings were not all short-cir-

cuited but were open or closed through an external resistance,

the case would be similar to a transformer on open-circuited

secondary or with loaded secondary, respectively. " In the

former case the current flowing in the stator would be the

magnetizing current and that due to hystersis and eddy current

losses and therefore a constant quantity for a given frequency,

"The Polypha.se Induction Motor" by R. L.Mershon. E. W. $ E.

Vol. 35, Feb. 3 & 10, 1900, page 160.'

14

and the copper losses would be very small. In the second

case, as the external resistance is increased the current

flowing is increased and therefore the copper losses are in-

creased and the total losses at any load is the sum of the

losses on open circuit and the copper losses due to the

resistance.

When the rotor coils are short circuited a heavy induced

current flows through them and corresponding energy must be

taken from the line by the s tat or in addition to the magne-

tizing current. The slip in the induction motor is analogous

to the drop in voltage in the alternating current transformer.

Art. 3. RELATION OF THE INDUCTION MOTOR TO A DIRECT-

CURRENT LOT OR.

Upon starting a direct-current motor there is a heavy

rush of current in the armature. This is generally dissi-

pated in the usual starting box resistances which are put in

the armature external circuit. There is a similar result in

starting an induction motor, and it is taken care of in a like

manner if there is external resistance in the rotor circuit.

In the armature of the shunt motor there is a counter E. M. F.

generated due to the conductors cutting the lines of force of

the shunt field. The mechanical work of the motor is against

this counter E. I.I. F. In an induction motor there likewise is

a similar action. A counter E. M. F. is generated in the

primary windings due to the induced currents in the revolving

secondary windings. This counter E. M. P* is similar to the

counter E.M. F. of the shunt motor,- so that in both motors

the speed falls off as the load is increased and no other

change of operating conditions is made;1"

Mershon,I*bid.

CHAPTER III.

CLASSIFICATION?

Rotary induction apparatus which consists of primary and

secondary windings rotating with resect to each other, can

"be divided as follows:

(a) Induction I.Iotors.

(b) Induction Generators.

( c ) Frequency Changers.

(d) Rotary Phase Transformers.

The subject of this thesis comes under (a) Induction

Motors. This class can further be divided into,

(1) Induction motors with variable resistance in rotor.

(2) Induction motors with short-circuited rotor windings,

or "squirrel cage" rotors.

The motor on Which the thesis tests were made is in the

second class having a short-circuited or squirrel cage rotor.

A further classification is usually made into motors of

variable and constant speed, and therefore corresponding to

the above classification.

Standardization Rules. American Institute of Electrical

Engineers. Volume XVI. 1899.

17

CIIAPTKR IV..

CONSTRUCTION.

Manufacturers type G.

Serial number 135698

Type of machine Two-phase

Number of poles 4

Capacity in H. P. 2.

Speed R. P. M. Ho load 1800

Speed R. P. 1.1. Full load 1700

Volts, normal. 110

Current, Full load 10 amperes

Current, No load 5.5 ampere

ROTOR. .

Peripheral velocity - feet per second 70

Diameter over all 8 inches

piameter of core, internal 6.5 inches

Length over all 6.75 inches

Width of slot, at bottom 3/8 inch

Width of slot, at surface 1/16 inch

Depth of slot 3/8 inch

Insulation Press Board

Width of tooth, at root•

c/16 inch

Width of tooth, at surface 9/16 inch

Size of conductor 5/16 x 5/16

Number of conductors per slot

Air gap

Length of tooth

STATOR.

Length of core

Number of slots

Depth of slots

Width of slots

Insulation

.031 inch

4 11/16 inches

4 3/4 inches

72

1 inch.

1/8 inch.

Tape and Shellac

W 4 r\ + Vi nf + n n + ViII ill Ull Ul uuutii 1/8 i nch.

360°

Length of polar arc 90 u

Coils per slot 1

Coils per pole 9

Coils in series

Resistance Coil A, - Az 2.63 ohms

2.59 it

tf A it

" A7 - A B it

" B,- B^ 2.10 ti

" B„- B« a . 8a it

" Bj-- B 6 2.55 ii

Br- BS 2 . 62 it

The above measurements are taken from the V/estinghouse

Two-Phase 2 H. P. Induction Motor upon .which the experiments

for this thesis were made.

The stator winding is distributed and each pole consists

of nine (?) coils which are connected in series. The ';•

.terminals are brought to a switchboard on top of the machine

where connections can be made for operating at 110, 22.0, or

44-0 volts. A diagram is given on Plate Number 1, showing

these connections.

The bearings are in the casings which close the sides of

the stator. They are self oiling. The rotor is well balanced

and turns very freely in the bearings. It has been found

from three or four experiments, that it requires, on the

average, eleven minutes for the rotor to come to standstill

after the current is shut off, the motor running light at

that instant.

Goqn&cliGfti 'for I/O vajfs.

AZ

bb

A

S

3

A8,

1LQ_

T7Q"

&7

£>

J"

c onneci,o n3 for ZZOvolti.

A3

A5

/A

6

A7

A8

9.

C onne.z1iot) S for *f iO Zoit^,

6

"51

Plate No.2.

Wiring D Ldjra/?] of Machine Switch hoard.

CHAPTER V.

PERFORMANCE AMD TESTING.

Art.l. METHODS EMPLOYED.

The motor was fastened securely to a frame foundation.

On one end of the shaft was a brake wheel shown on Plate 2 .

It was so arranged that water could he applied to the rim to

dissipate the heat. Suspended from a spring balance was a

one quarter inch hemp rope, so placed as to he tangent to the

brake wheel rim. The rope made one turn around the brake

wheel and passed down and around a smooth pulley and then up

to a second spring balance as shown in Plate ?•

The difference in readings of the two balances gives the pull

acting at a radius equal to the radius of the pulley plus

the diameter of the rope. The output in watts is given from

the following formulae,

2tT x 746 x R. P. M. x (T,- Tz ) x effective radius in feet

33,000

Output V/atts.

Speed was measured by a Weston Magneto Speed Indicator.

This was belted to the shaft at the opposite end from the

bybrake wheel. It was calibrated Ameans of speed taken with a

speed counter and timed by the laboratory stop watch. The

deflection of the voltmeter is noted and the speed for any

other voltage is proportional to the observed speed and voltn

age, as shown by frequent calibrations.

Ph te /Vo. J

The electrical input was' measured by a Weston Wattmeter

and Voltmeter and a Thompson Ammeter in each leg. A table was

wired permanently and the instruments set on it as shown in

Plate 4, and in Plate 5.

The current was taken from the University Plant where it

is generated at a voltage of 440. It was transformed down to

110 in two V.estinghouse 0. D. Transformers. There was a

West inghouse potential regulator attached to the transformers

and by this the voltage could be maintained constant.

Art. 2. ERRORS AND PRECAUTIONS.

Having started a test on a certain voltage as the new

load was added it was necessary to allow it to run a minute

or two and so reach a steady value. Errors also might arise

from not having the instruments all read at the same time but

this was overcome as far as possible by having all readings

taken a signal by the person adjusting the load. Before each

reading the voltage was observed and kept constant by the

potential regulator. All instruments were calibrated by

comparison with the West on Laboratory Standards and the read-

ings corrected before calculations were made therefrom. In

all cases single readings alone were taken for each condition

of the experiment.

-pi a r

Plate No. f.

£6.

Art. 3. EFFECT OF VARIABLES UPON THE PERFORMANCE OF THE

INDUCTION MOTOR*

The speed is practically independant of the voltage at no

load, until the voltage becomes very small when the speed

drops off very rapidly. If the motor is loaded the output

will vary proportionally to the square of the voltage. This

can be shown by the following readings of output and! taken at

about the same speed, from the calculated results of the

experiments performed.

Exp. No. 1, Sheet C. Volts 115, Output Watts 1310, Speed 170?

Exp. Mo. 2, Sheet C. Volts 110, Output '..atts 118°, Speed 1710

Exp. No. 3,Sheet C. Volts 105, Output Watts 1220, Speed 1705

115A : 1310 :: 110*: 1189 :: 105*: 1220

V/hich proportion holds very nearly true, since the speed

varies somewhat and personal errors also enter.

Change in frequency varies the speed of the rotating

field and the speed of the motor depends on this. The output

depends directly on the speed and therefore it depends on the

frequency. The frequency in the rotor at starting is very

high but as the slip decreases the frequency of the induced

current decreases.

This motor requires a starting current of about fifty

amperes at 110 volts, found from a number of experiments.

The speed curves are shown on Plates 6,10,12, for 115, 110,

and 105 volts respectively. The lower the voltage the more

rapidly the speed drops off, implying therefore, a greater

28 •

slip for the same load with a lower voltage. The current

being proportional to the slip, should be a steeper curve for

lower voltages, as is shown on Plates 7 ,H, and 1?.

The break-down point of the motor could not be deter-

mined, on account of the excessive heating in the stator

coils due to the large current flowing at both 110 and 440

volts. The manufacturers of this machine, the Wes tinghous

e

Electrical Manufacturing Company, claim," with regard to' this

type and make there will be: "No damage if overloaded and

stopped for several seconds." Evidently the copper allowance

for the stator windings have been made quite small, and the

carrying capacity is not sufficient to carry the current as

a transformer with a short-circuited secondary and an

impressed voltage, of 110.

The output in all cases was carried to a maximum and

then as the torque was increased the speed fell off so

rapidly that the output was smaller. The maximum output for

115 volts was the greatest obtained. The efficiency for this

voltage reached a maximum at about full- load and held up

very well to about 100^ overload. The efficiency for 110 volts,

and 105 volts, respectively reached a maximum somewhat before

full load and dropped off rapidly afterward.

The power factor does not vary much from its main

characteristic, With the voltage and. is of the same general

appearance in all three experiments.

The wattless current increases more rapidly for the~>-Cir, cular No. lU^P'.'Polyphase Induction fAotors:" July lyoo.

29.

lower voltages.

On Plate 8, the inductive 'volts drop or E cos <p are

platted on a true input watts "base, they are a maximum' at no

load. and. decrease to a minimum and then increase slightly.

When platted on (p as a base a true cosine v/ave is formed.

Wattless watts and wattless current platted, on true

input watts base, Plate 8, increase from a minimum as the

input is increased.

On Plate °, there is platted, the equivalent resistance on

true watts input base. The equivalent resistance is found by

dividing E cos cp by the current, or in other n, ords, it is

applying Ohms law to the alternating current. It rises from

zero to a maximum and then gradually drops off. The other

figure on Plate 9, has its base line 1'aid off to represent the

impressed voltage, llo volts. Thr~e cases are taken, obser-

vations number 1, 5, and 19 from Experiment No.l, the values

of E cos <p , and E sin <p, columns 8 and 9. The triangle of

electromotive forces is made in each case and the points

C, D, and E were found to lie exactly on a semi-circle con-

structed on the impressed voltage.

Prom the foregoing tests the statements of Mr. H. G.'Meyer

in the London Engineering"* , earlier noted, have been more or

less verified. He sets forth the following' advantages of the

indution motor.

1. Possibility of starting under heavy load, which is an

absolute necessity in some cases.Jf- " ... — — ... I I

'

I .. ~ —

2. The primary circuit serves a double purpose, furnishing

both the exciting and load current, thereby doing away with a

direct current exciting source.

3. Great simplicity in mechanical construction, more partic-

ularly in the revolving parts where frequently, by the use of

a "squirrsl-cage" winding all current collecting devices are

avoided.

4. The voltage of the rotor can be made very low, thus simp-

lifying insulation.

5. The motor can be worked at different speeds.

BRAKE TEST OF

Y/ESTINGHOUSE 2 PHASE, 60eo, 2 H. P. INDUCTION LOTOR.

IMPRESSED VOLTAGE 115. EXPERIMENT No.l. SHEET A.

APRIL 11, 1901. ORIGINAL DATA. OBSERVED.

Table No.l.

No,

UPO-

i »• i • i /

WestonTob

to JO.

Balanceo to 15.

Voltmeter

WestonWZB8Z

A YTi tn sXtt.

Thompson.

WaJ.tmefa

Weston\lo\t meter

Weston

^29<5 J-

Thompson^11-0 IO

Wattmeter

Wzstson.v/500

1.o

a. O. 6

.

7-Qo

.

Qy

,

I o Q i/SZ Si 7 130. I'Sf S.G S6

z. 98.

8

T--/7 o-/o 11+ s 6. JO // S o 6 25 3/0

J. 37. & 7-/Z 1 - I.S 1/7- o 7. OS S*Z l/f-S 6.85 *6o

7. 36. Z 16 - S i -/is 1/3. S 7-8 b60. /M-o 7. SS S90

dr. Pis. o 1 Z - 6 z-7. n7.s 8. So 76o IIS. QJ6~

6. 3*z /f -6 z -iz l/S.O 9- IS dso l/S-.S 7 7a

7. $3.8 /6- 6 3 -6.J lld'Z 9.30 <9So l/S. 8 87o

8 92.7 IS-/Z J - J3 l/S-0 ,0.70 / OJD ltd'* 1 Q.%0 3 7o

3. '91.9 Jlo -M -f - 8 tH.9 11-60 1 1 So IISO 1 I.OO / o /o

IO. S -6 l/<?.7 IZ.7Q IZSO //i-S //.$ 1 2oo

II. 9o.o 2S-9 6-7- Hi- 7 13-10 13^0 1 l<T.O 1 z.s / Z 7o

{Z. 88-9 z/-// 7 — o.s 7/S 3 I J.Oo J4SS i/S-S 1 3.7 13 80

13. 8 8-0 7- S //SO IS-2 /f60 l/S.Z 17.7 IfBo

f*. 8 8.0 31- 8 8 -iz /is:/ IS-

1

/6-f llS-8 17.1 Id'60

IS* 8 8.o 33- /o 3-3 1/7 8 M.«f 1770 /ISO IS. 6 /yoa

/6. 87 Z 3S~- 2 9 -73 //SO 17 3 /S8o /ISO / b.S 18/0

1 7. 86.Z 36-jr 9 ~/3.J US- / 18 -O IS70 l/i-8 173 1900

/ 8. 8*3 38-/ 3 -t3 <* l/SZ IS . Z Z/oo IMJ IB.S 2 loo

id. 8Z.o 38 3 ~/S 1 IS. z Z0.3 l/S.Z 13.7 Zl 60

zo, 8 I. S 7I--Z 9 -ISA IIS o ZZ.O IIS. 8 Z/.2 Z3Z0

Z I eo.o to '/ 1/7-9 ZJ.O li S.o z z.z Z7Z0

2.Z. 78.Z fj-o /o-/.S n-1.7 27. Z Z6fo 1 ISO 23.3 Z5~6o

Z3. 76 S 4<? -O to-t 1/7-3 ZG.o Z QZC //* 8 ZS.o Z72o

%f. 7i.S 7-S~6 I o -o.s l/s.o Z7.Z 3 ooo 1/ S. o &6.1 ZQoo

BRAKE TEST OF

V/ESTINGHOUSE 2 PHASE, 60«#, 2 H.P. INDUCTION MOTOR.

IMP&SSEB VOLTAGE 115. EXPERIMENT No. 1. SHEET' B.

APRIL 11, 1901* ORIGINAL DATA. CORRECTED.

Table No. 2.

No.

n

ft P/VWeston

To hbalance

£ oOO

ft fi>tto«»

balanceto 10

WesionA0O0Z

Am mp.leyr 1 1 r/ 1/ f \s ( \_ /

7/70/77/5JO/7

^/fooe w 10 03

Vo)fi7?e1tr

*Zd3 /.

Wattmeter

^1600

/ 2 <*J4.1. u A 8

;. o 1/6~.Z if. 8/ / 90 l/6-.3~ S36 So

z. 98.8 O./ZS 6.68 -flo us.o 6.S3 3/0

a.cJ7-2 7-/67 O. S93 IMO 7.3-0 S63 IM.S 7.ZO

* 36.

z

9- 7Sa I.ZIQ II 3.S S.Z6 ego 7-9S S.90

SS.o it.S/z / ISO l/f-S 9-OZ 781 115.0 8.SS 6.86

6. / 3.8/2 z.zsra //SO S1Z II5.S 3.IS 770

93.Q /6.00a Z.%0% i/sz /O-SI 980 US. 8 9.9o 8 7o

8. /Q.I81 3 3/2 71 so i).3S / 080 ns.i lo.7Z 97oa. 9 / 9 Z0.3/Z ^.00 U1-9 IZ.3Z ;/Q3 1 ISO 1 1. ss / a 7c

/o. ZZ.6ZS 9-87S //?.* Z3./8 /2Q3 1/9-8 /z.so /Zoo

/ /. £5. 000 S 7X0 //i- 7 1 3.90 1378 //so I3./Z /z 70

i z 89.9 27. /ZS &.S8Z US3 !f-7S 1195 IIS. 3 11. /o /380

13. Z9.37S 7-006 //SO' 1 6. cs /660 JlSZ is. IS ii-80

88.o 8.Z6'o t/S. / /6.QO /68Z list IS. 85 /SOO/<5". 89.0 6-687 //?. 8 /8.0-S IQ/o I/So /7.Z0 / 700

/6. 87-^ 36 60S 3-8/

Z

//S. 1'9. IS /9Z8 1 iS.3 /Q.Zo /Qio

1 7- 3 7/87 Q.8/Z i/s. / 19. 80 ZOZO /ISO /9.os /So

1 8. 818 38-937 8-89S iisz Za.so l/SZ Z0.00 Z/o c

19, gzo B.937 IISZ ZZ.Zo ZZ9Z j/s.8 Z2.<?o z/60

2 0. 8 is f2. 000 8.97o //3-.0 li-ZO Z-93S 1 /So Z3.Zo ZJZo

-3 /. &0.0 4ZJ30 9-o6z I /<?.<? ZS.16 2o~60 //So Zi-30 ZC

7&Z 13.87S 9- 093 1/17 Z6Q0 z. 700 //-f-8 Z0.O6 Z.S60

£ 3. 76.S IS 873 9.06Z 1113 ZQ.Qo ZQSO /MJ 2 7.<?o Z7Zo

Hs 9.°y/ 1 1 f.O 30- 7o 30 70 JISO Z9. /c 2900

BRAKE TEST OF

WESTINGHOUSE 2 PHASE , KOco , 2 H. P. INDUCTION MDTOR.

IL1PRESSID VOLTAGE 115. EXPERIMENT No. 1. SHEBT C.

APRIL 11, 1901. CALCULATED DATA.

Table No. 5.

K 1

No.Pi,))f U 1 L

in.

Jot&

I

U ol f> ut.

To-i a \

WaUsIII pvt.

~Vy u t

EffLciei/tj

i <<j .

4. <-.

L.

1. I&so O. O 190

z. 1778 jje>7 Z.000 t9S 7SO

J. I7SS (o.S3(o 3. Z 8 7&S 1 023 7S.(o

f. I 7-70 8.S3(a 9 9o 72 80 77.9

6". I 7^0 / O 6<°Z 1 ISZ 78-8

6. 17'oz l/.d'^Z s.sz 1376 7S-d

7 i^ds /3.0>$Z /-780 I8S0 9>0.f

8. M.87J 7. to HoGS Z03-3 8/.S

9. /66C )(o.3IZ 7-78 /8 0S &2S3 80.1

/©. 18.187 8.7* 1380 Z<?30 80.0

// I6&Q 19-ZSo 9- /8 &088 79.0

I609 3.78 ZZoo Z8 7J 76.8

/ i3. /S~9o Z 2.3/S IO.&7 so 80 77.0

/f /S9 /ZS //.«*"/ zsec 79.o

IS, 1590 Z,J~.8lZ iz.61- £740 S6~70 778IS. IS73~ Z7./87 iz.95 zeso J7S8 7d:i

11 IS~S~8 z 8.37S I3.S-? Z31o ?92 7SO18. t$36 J0.O9Z li-ZS 30 7S <fZJ8 72.0

19. I^QJ 30. 18 7 H.^o 30 So tfSZ 6 9.3

ZQ. /«* 7Z 3ZPdl 13-30 -3MS 1-778 66.0

Zf. J3.G87 /6.0S (>i-7

Z2. IflZ 3f. 786 I6.3~7 3Z78 SZ60 6Z 3

Z 3. 13 BZ J6.8/£ /7S8 S3 9c S670 60.3

Zf. 1? fd 37.Z/8 /7-7<° 333~6 S97J 5~G./

BRAKE TEST OP

Y.'ESTIMGHOUSE 2 PHASE, 60co, 2 H. P. INDUCTION MOTO&;

IMPRESSED VOLTAGE 115. EXPERIMENT No. 1. SHEET L.

APRIL 11, 1901. CALCULATED DATA.

Table No. 4.

No.

Obi.

Apparent

Watts.

PoWe Y

F q r'f.o y

<DTbin <p J <5tV? <P 1 Cog f

Wattles?,

TJ&tt 5 .

L cos y

1 2XL)

.

I/.-4/

< A 7 8 a

/ .978 S.68 1.6 8 635 Zi.7 nz. s 1. If

Z. 7G7 ,56 7 dTf'JJ .8 2 3 d~.<5o 3.79 63/ 6fO 91.Z 9.71

c5. QJO . 660 Jd~

9

^. 68 1.87 6f6 71 z 86.6 9.88

93& . 73 6 .67 6 5-57 6.07 6 35 8 3.6 76.7 /o./z

j: /63f . 755 *o J 8 .655 5. 90 6.86 677 8 6.6 7S.0 9-6Z

6. II / 7 . 782 38 33 .623 6.06 7.60 635 90.

Q

77 7 9- 28

7. IZ/Z 36 a .se 7 0.17 8.5Z 7 IZ 93.Z 677 8.86

8. /304 .8ZS 3*° .559 6.31 9.1Z 7ZQ 963 61.3 8.40

9. . 838 •33 1 .615 6.7Z IQ.3Z 79S 96.<? 6Z.6 78Z

/a 15/

o

.850 3' 17 .5Z6 6.9<? ll-ZO 795 97S 6 0.3 710

//. 1533 . 86Z 3ca

Z7' ,50 6 7.04 //80 8ZZ 39.0 S8J 7. iz

I 7Zo . 8^8 29 46 19 6 7-3Z /Z.80 851 100.Z S7.Z 68013. 7817 .863 Z^

Q3S' .191 7-93 /3.9o 9/3 J6Q. O JT6.8 6.Z1

ii- I93Z &70 Z9°3Z' .19Z 8.Z6 74.6

1

96Z /CO.

3

S6.6 f.98

2o?8 • 8 7Z Z9'/Q' .489 8.81 /6.70 /ooo /oc.Z 56-3 6.66

lb. ZZoo .Q76 Z8°5d' ,1Q% 9.Z3 76.75 /03-7 /oo. 8 ss.i 6~.Z6

17. ZZ80 .866 Z7'37' .^63 S.I7 J7-6Z J63~Z /oZ.* 63.4 5. /5IS. XlOZ .888 Z7°ZZ' .13-9 9.65 78-45 J/oo 702.3 6Z.8 i.9Z

19. Zf8o .890 Z7° 7' .165 10.01 1 9-75 // 7

1

JoZ.S 5Z.4 1.6 S

zo. 27 7

S

.8S>7 Z7" 3d' .1.6/ IZ78 I OZ.O 63.0 -f.zz

zi. Z9/5 .883 Z7°f4 .169 //. 9o 2Z.1I 76 7Z /o/.S 538ZZ 307° .8Q0 .171 IZ. 70 Z3. 60 713~Z /oo. 9 51.3 3 78

za .876 ZQ°68 .481 13.91 25. Zo 1595 1 00. 55.1 3.4833-30 87o Z9°3Z 19Z /SO® 26.7f .17 3Z /oo.Z J-6. 6 J.Z7

BRAKE TEST OP

Y/ESTINGHOUSE 2 PHASE, 60©D, 2 H. P. INDUCTION MOTOR.

IMPRESSED VOLTAGE 110. EXPERIMENT Ho. 2. SHEET A.

APRIL 18, 1901. ORIGINAL DATA. OBSERVED.

Table No. 5.

No. R PMf i , / / ITo k

to jo to /S.

We si.on^2Q8Z

Weston*/3~oS.

VolfmiAtY

We sion

*ZS3I *MD/0

W&tlflKitK

We s ion.

/.oso

,

<J.•ro 6. 7 8

.

Q\J

i /06 8 O // a. 6 o.Z IZ2. //o. S.3S

z. /0J.9 / -/Z -9.S // fi Z70 1//.0 s.ss I80

foj.9 J'// Q -/J. I/O A S'S 3 7£> l//).o T.8S Z80I QS3 S'-'o / -&jf

/O.9. 8 6.33 ffo l/O.S 6.20 366

<5. 1 O/.S 7-0 / ^ 6.S loss 7.0o d~30 ///*>.-? 6.70 4 606. 39-2 / -/3. 103.6 77S 6f-6 //o.o 7.^0 3TS0

7- 979 // -8 /0 9.8 8.S<o 7J~&. 1/6.3" epo8. 96.Z /j^/Z 3-/.S ) 10.0 9.zc ///-/] S.7J 7S09. 9 S.i I6-Z 3 -/o l/G.O 1 o.zo 9so II0.S 9.6c 880to. 9?o /8-Z I0 9.Q 1 I. lo 1 od'o l/o.z / 0.30 9 80// 92.° S'SJ Jto.C //. 3o use //4.S /O.30 10 80

IZ. 900 6 ~3.S /0,9.B IZ -60 /Z70 //o.Z //.¥o /ZOO

io &S-S 6-/Z )7 o.o 1 3.60J 3 SO I/0.8 /Z.60 /3/0

If. 89.9 2 6-// 6-/S.J !/ C.6 /4.<?G 60 //6.<7 /3.JC 7^00)

13 8 9.0 Z& J 7-/.S /o9.8 IS', oc /SSO //o.Z /4.00 /<?60

16- ft7A JO -J 7-i.s 103.7 1 6Z0 16 SO //o. / IS~.3o I60O

n H7.Z 3/ -a 7-6 l/o.Z 1760 //4d /6.60 770O

18. 8J0 733-/3 7-7.S / /n.O ie.Qo /9 60 //o.S /8.00 I900

19. SS.o 3Z ~6 6 -/Z //CO 19.80 ZO 70 s/o.z J9Z0 Zooozo 6 -I3.S

/09-X ZI.ZO Z Z%o //6.0 ZO.70 Z78036-JL 6 - /f /OSS' ZZ.36 Z3SO 1/0.6 %\.?o ZZ60

ZZ. 78-6 37-

8

t> -/J.J" //o. Z3.80 zsoo l/A- S Z3.oo %ioo

23. 76.0 JS-8 6 -/JJ /OS, 8 ZSZo Z 620///). ZS60

Z f TS.o 30-

7

6 -13.0 Z740 I/O. Z d~.S0 Z6602S. 7o.3 11 ~7 b ~/JS / /o.Z Z7TO ZQZo //d. d~ Z7.44 ZQ60

BRAKE TEST OF

iSTIITGHOUSE 2 PHASE, 60co, 2 H. ?. INDUCTION MOTOR.

IMPRESSED VOLTAGE 110. EXPERIMENT No. 2. SHEET E.

APRIL 13, 1901. ORIGINAL DATA. CORRECTED.

Table No. 6.

No.

U Pwi

n-npf.VVe sW/

la h

1c JO

f\ Hi1^ Oil Ot/7

Oio IS.

Volt/77 eit/

^8 8Z

AM hi fi "f-p r/ \/ff //ye ttri.

Tho/?]pS<J/j

*/<?QQ&

Wdii1.tn(>1er

VVe si 0/1

* ISoSWe if0/7

t\ tft hi c fpi

Th a mj^^on

W&1t hit let

*/S00/

<i 77jt L

. 7 0.

/. /C6.fi o 770,0 <7T.3T3 770.0 S.6Z

Z. 103.9 Z.6ZS 0.Q9 3 7/77.0 S.9 6 £Q2 ///.a S.93- /so

3. /03.9 ^.6'6Z O. B/Z 777). 6 6.ZS 390 l/o.o 0./3~ Z804 /033 6SOO /.037 10 9.9, 6./0 ?• 70 7/0.S <o.S~0 JOO3. 10 l-S 7- 8 7s 0. 90s 70 9.S 7.-9S~ iSSO / / 0.4 7. ojr 41 60

6. 99. Z /o.Zso 7-3/2, / 0.9.4 <d.Z<? 6 70 7/oo 78 3'80

7. 979 7Z.37S 1-782 lo 98 9.03 78o 7/ 0.s 8.JJ 6 7o

8. 86Z /1.623 Z.S93 J10.6 9- 78 8^8 7/7. C 9.Zo 780

9. 9SI / 7. 000 3. IZS ) (0.0 7 O SS 3 SO 7/O.S / o./o 88Olo. /9.000 S. 62S JO$.fi 7/.RO I 080 7/o.X, I0.8Z 9®o

II. 9Z.0 Z/-87J dT.093 J/ 0.0 / Z.30 II 81 7/0,6' 71. SO 7 08O

IZ. 9co Zf.Ooo S.697 /09.fi 7^7-.0O 1 -370 //O.Z /Z.bO / JcOO

id. 8RJ Z6-/ZS 6.ZSO 770.Q 7<t9o /4ZO l/*8 74.00 /3 /O

if 899 Z7-S6Z 6.168 1/0.6 7S.SO IS-oo 7/0.4 /4-70 7^700

16. 83.° 29.062 6.S93 76 9.fi 7 6.S~0 73~9Q 7/O.Z (S.?0 7*9 60

/6. 87.3 J/-06Z 6783 709.7 77-80 1720 I/O. / 76.90 7 60O

it 87.Z 3Z?7S 6.873* 7/o.Z 7 8-^0 l 8OZ I/O.8 / 7.6O 7 7oo

18. 6<s~o 31.687 6.970 7/0.6 ZO.70 2004 l/o.s ie.Qo 7S0O

19. ~33.ZS6 6-ZSo 7/AO z/ 80 Z/ZO I/O. 2 zo.90 ZOOoZo. 8 Z.I 6. 394 709.7 z>3.30 ZZ7Z / (0.0 ZZBo 2/802L 803 6.3/S 70 9.S 240Z //o. Z3.$o ZZ60

2 2. 78.6 3Q37S 6.394 7/0.0 Z6.ZO Z<T6d t/0,S ZS.3Q Z400Z J. 76.o 39.37S 6.391 109.8 Z7 70 2680 )/o. Z6.70 ZS60

24. 7J-.0 40. 312 6.37S 7o9.<? £8.90 Z8?o 7/6. 28. 00 2660zs. 76-S 4 Z 6,39* 7/6.Z BO.fo Z3 90 7/0.S Z9.7o Z860

BRAKE TEST OF

NESTING-HOUSE 2 PHASE , 60co, 2 II.?. INDUCTION MOTOR

IMPRESSES) VOLTAGE 110. EXPERIMENT No. 2. SHEET

APRIL 13, 1901. CALCULATED DATA.

Table No. 7.

/V o.

ObS.

Pel)

f~~ouii d S.

l yv e

Uujpo T.

To1* I

True Y&fts1 rut

2, o • /• 6.

1 1830 O 77Z

z. /78o Z.S3Z 7.Z60 JO 6J~.0

3. 77$o 2. /SO sos 6pO 7S.-9

4. 77? Q 2.S60 6<?Z §3Q 77.<7

S. 6.968 3. 338 70/ QO.O

6. 888f <f.Z9o 7Zd~0 Q/-Z

7- t7fo /o.-izs 0.0/0 f'Q9 7<?SO 82.6

8. 7680 S-ZyO 76*78

9. I 66~0 / 3. 6/<f o&ZO )3'Z1 7 860 8/-

10. 7b/Z I6~.3?f 7 6J~4 Z06O 80. ^

II. I6'76 76. 7Q2. 8.0ZO 7760 7 7* $

U. 7S7Z I8.Z83 8-77 ZS/O 7s, O

/ J AT/

6

/3.&7S $.*79o Zoox Z7JO 73.SIf Z/093 /0.0& 2L09S Z90O 7 2>- 3

i j. /<z / XjA. ./aJ / o./z z ZZd jO Q 7 Z. a

7<fSQ Zi-783 I/.S8 Z 360 33Z6 77. Z

n 7<?S0 /z./s Z<?68 JSdZ 70. s/&. /3.1Z Z6ZO 396*7 67Z19. /<?/f 28.ooo J3.36~ Z63Z ^/ZO 6t%a 1366 Z8-77S J3.JQ Z68O <?<7SZ 60.

Z

z/. l~317 2 0.68

;

Ji.ez Z760 466Z d~9.

zz. /30Q JZ.Q3/ 7S.Z% Z800 -7960 S6.S23. IZbf 3303/ /s.ys £780 SZ<7 J-J.

Z + 7Z<7% 3"?, aaa 76. ZZ ZSZo SSOO S7- 3

25. //?2 3S. 968 1 7. 7f ZQOO SQSO 18-0

BRAKE TEST OF

WESTINGHOUSE 2 PHASE, KQco , 2 H. P. INDUCTION MOTOR.

IMPRESSED VOLTAGE 110. EXPERIMENT No. 2. SHEET' £>.

APRIL IS, 1901. CALCULATED DATA.

Table No. 8.

No.1/1/ . * A ,

Po We v? dip

f. 7 sin f fcof f

/. 2. 3. S. 6.

/. 2/>8 77*J 6 97 J~ S.43

Z. 6 J~7 43.0 64°29' 90.Z S~.36 2.f6

J. 688 J-6.8 Sf°j7' 82.2 £~./4 3. ^4

f- /\J7 6d.& 76.9 6~/4 f.Z7

S. 1-7 °/e' 73.9 S.02

6. 74.Z 42° s' Q>7. tf~.S2 6.76

r 78. 6 38 II 61.

8

6-.3~8 7. '0

8. Bo. 8 SS-8 7.31

9. / 133 SZ./ 3 * v e' S70 6./S 8. 9//o. 1 Z9S 83. 6 £-<?& 6.48 3.89

II. tsidr- 86. o J~7.o 6.33

IZ. 86. 8 Z9°4<>' 1-9.6 6.86 //9813. / 610 86.7 29* S3' 49.

8

74

Z

/2.90

14 1738 Q6.47 36° 7' J~0. / 7-9/ J 3.68

/830 Q7.0 Z9"3Z' 49.2 a// /4- ~33~

8S-0 28°Zi' «?74 /3-.6S

/a 88.

7

Z7°30' 46./ 8-48 / 6. 36

18 ZZ7Q 88.Z 28 6 47J 9.7s 78.ZS

7 3. 2,4/S &8.o 28°2/' /o.38 J9-4S

87-8 <?7.8 //,// 20-4ZZ

Z6-Q0 Q 7.S Zd°S7' 4S.4- //.e>z Z/.4Q

A3. Z982 87/ 2 9%y 4-9. / JZ.8S ZZ.dO

&3. 86.3 29°39' 4 3-4 13.68 24. 00

3.4. 3/68 Q6.o joy/ f/. o /4-7Z Z4-86-

26. 333~6 8^8 JT/. 3 /6~. OO ZG.Q2

BRAKE TEST OF

V/ESTINGHOUSE g PHASE, 60c* , 2 H. P. INDUCTION MOTOR.

IMPRESSED VOLTAGE 105. EXPERIMENT No. 3. SHEET A.

APRIL 25, 1901. ORIGINAL DATA. OBSERVED.

Table No. 9.

No.

Up 6.

ft. ''

We 3"* 0/7

VW. J.9J 7

iu i

to JO

P>o'itotn

Q to IS.

We 5to a?

l\ ni nip4p v

v7*oo&

W&it i7ieftr

yi7e s ft>/7.

*/5Q 5

rt) XT frlzTtY

^£93 1

Vi H./7H.KT

We ifo/y.

6~Qo

/

/. Z. O.4-T fr 6. 7 8 9

1. / J. O 1 oS.O 706 /oj:6 JO

Z. 7 - 6 / -8 76<?-3~ frdZ Z^O 70S.Z 6~.3<7 1 60

3 /Of. s f -// / ~/jT /dS./ b- 70 ^700 10 6.0 6.00 3^0

<Z /o/- 7-Z, 3 -7.3 706~-0 6-80 TflO 7oS6~ 6.60 410

S. j oq. 6 8 -AS 3 ~7S 7.33 its. a 7. 70 J-70

6. 99,

o

// - O 3 -7.S /di<7 Q.ZO 7oo /Ocf.O 7 33 6Z0

7. 97.9 /~3 — 704.Q 8-90 8.SS 730

8. 9 b./ I65~. O 1 O.oo Soo /od:d~ 630

9. 9JA fa-*? /dj'0 77-7 O / OZQ 70d\<fi 0.9 300

lo. Zo-*7 S~-7J 70&.Z 17-SO IIZO 77.S-C 706O

//• 93 ^/ -// 6 - <? /04~ IZ.7 IZIO 7d 6~ 6~ 7Z- 7<J 7726

I*. 9/7 z<7~7 6 -/^ ' 7dtT. J 71.0 7J?0 70<f. T' U.6 7300

7 J. 90.0 Z6 -/z7-.3~ 70S-Z 7^.0 76'6 /oS.Q 7?Z0 /<?3~0

Z3-8 7- 6 70 6~ 76-7 J66C 70>j:3 7J~S0 73~3~A

IS. 7/ -O 7 70 S.O /9.Q 7870 7'66'. J 73- 30 7360/6. 7-7 JO S, ZO 06 706'

3

7S-Jo 79 00

H 33~6 > -A* 70^.3 Z/. 6 ZO&C 76-4~.0 Zo*?o ZOO*?6

1 8. Ji-6 7-f.tr 709 Z 22-0 Z7 ac 70 6~0 Zi.-7° Xi 30

19 7-b / o$:o JZ3.0 Z3Z0 IQ<f.<? Z2. 66 ZZS6zo. yt.s Z7-/Z 7-7* 749*4 Zb-3 26Z0 76^8 Zf.&o Z-740

2 2. 6QZ 33 - 7-7 7<34~. 2786 / 6d~-0 Z7.zo 27?oZZ G2.o Q-70 7d 3. /o6~, ZO.60

BRAKE TEST OF

mSTINGHOUSE 2 PHASE, Mm, 2 H. P. INDUCTION MOTOR.

IMPRESSED VOLTAGE 105. EXPERIMENT No. 3. SHExT B.

APRIL 25, 1901. ORIGINAL DATA. CORRECTED.

Table No. 10.

l 1

No.

UPS.

CD O tyf

n. r. n.Weston

To/>

fo JO

foatto/n

Balanceo to/S

i w( r i rtc. [tf

We ii<i nAm/ij-ctt)',

"ffaotv/jsdn

"Woo 8

r»oxt/ //e'er ionme f<r

We i ton

J/

/\>riniGi€)' Y&fffy &tr

We ifo // .

#156

1o<c

.

O T 0. D

,

O. QJ.

/. /oj.a 7oSO S.30 70S /OS* 6 S 76 JO2. /OZ.O Z.zso /0<f. S S.70 208 /as. 2 S.62 / 60

J. /OA S o.i-J/ /OS. / 6f78 ^20 /6 6.0 6.30 3<?o

/OA S-OQQ 0.370 /0S.6 7.22. S36 /OSS 6.9s 440

s. 16 6.6 9. 62 S /.S93 I6 6~. 7. 86 <36o /os. 8 7*7 S~/0

G 39-Q //. 87S 7 970 /o<?.7> &.?Q ?z</ /OS. 8.Z3 6267 37 9 /3.Q7S Z.3/S /oS.O 3.7S 826 /AS. S 9.0 6 730

8- 9 6V 76 Jis 2-873' /oS.d /0.62 936 /6S.S /0.0s

3 .2 /9. /2 S 3.782 /oS6 //8a / oS~6 /as. s 7o.9t 36o

/o. 9f. Q 2/.US -f.3/ 2 /dS2 /Z.2? 7/S-4 /as. 6 /060

// 33.0 2Z.S6Z 1-782 / 0S.0 7J.?6 /24& /dS.S /1.76 //20

U. 9/7 &j:j/z <Tf 68 /ocT. 3 /<?.8o /3?8 /OS? 7<?.dd 7300/J. Qo.Q 27. 625' s.sso /dS. 2 76*72 /S~<?6 /oS. S /S~.70 /<?s~o

/+ 884 £9. J7S 6. OOQ 70S.0 /7 7o 76-72 /dS.3 /690 /SsfO

IS. 8<7.o J/.87S S. 87S /d^o 2o. So /92a /dS. 3 2o. /0 786o/6. Q/.Z SJ./2S S<f62 /fiS. Z2.do 26¥g / 6 S3 2/-ZS /960

17 86-6 3f-ZS6 <rS93 l&f, s 33./

6

Z/Z8 /as c 22 Zofo/e. 7ZP> zS2fo S.782 /o^.z 2230 /a>s. Z3.J~S 2/30

/B >6. 6 36.6ZS S~.87S 7dS~.0 ZS.Jo 237S- /os.f 2i.8S 2280ZO. y/.s~ 38.626- d:97o 'o<?.0 %9JO Z684 /o<9>g 28.3o Z?ooZ 1. 58.Z <7d. 24 f d~937 /aS.6 30.80 / (ISO 29-96 £7<?o

zz 6. 2.0 4/. S66 f./ZS* 3J. So ^6 26 /dS.C J2.9a 2360

BRAKE TEST OF

WESTINGHOUSE 2 PHASE, 60c©, 2 H. P. INDUCTION MOTOR

IMPRESSED VOLTAGE 105. EXPERIMENT No. 3. SHEET

APRIL 25, 1901. CALCULATED DATA.

Table No. 11.

No-

Obs 1

ro 1

1

inFooncL 5,

~Tyl>P

Writs. fro*

1. Z. T. s. 6

.

/. /834 o o O

Z, /SoQ /J-/ 7bo

J. 77 88 4-JZf 7> 970 -f96 Q7o 6S.<?

6.63/ 2.868 7Z<7 i / to 74-7

s. 7730 7.0 3/ SJS I3<?4 7J-.3

6. / 7/f 2.97

Z

f.Z70 / Q3~4 733~6 78.6

7 /o.3~oo cT.dlo /zzo I78Q 78-6

8 / 67O 72.2.30 J3 9Z ZO/ 78-^

9. 76<?o 7f.SfO 6SfO 76ZO ZZ7<? 80.6

/o. 70 Jd /JT.87Z 77 68 1368 83.S// /6Z2 76.780 Q.06O / 8J~0 Z6 78 78-1

IZ. /6of 78.8ff 9.600 ZOoz Z99 7S.0

13. Jjr/f Z/.68X /O.66~0 zzso J/9Z 73.SIf. 74-98 70. 680 2.3*7

3~3780 73.6

13. 7<?<?9 / /. 73~o Z<?9 8 <70o8 63.S/6. 792<t Z6.S61 7Z.6/Q Z6~3~Q i/6g 63.6

11 /J86 Z7-637 734Z o H6Z S 4-360 63.

18. 7JJ~-? I3.4~S6 2~6>3Z 466'3~ 60S(9. 7Z7Z /^.zoo Z63Z £7720. 72/

7

37.637 Z688 3~3~6o S3.Z

Z 1. 7/ of 33.3/ % /d'906 Z7 0o SQ7o21. 70QO 33.373 73~.93~o JLf60 S9Qo

BRAKE TEST OF

V/ESTINGHOUSE 2 PHASE, 60co, 2 H. ?. INDUCTION MOTOR.

IMPRESSED VOLTAGE 105. EXPERIMENT No.?. SHEET E,

APRIL 25, 1901. CALCULATED DATA.

Table No. IS.

No.

OhS.

P

Y

? bin f J 3 in f Tcos f

/. z. 3. s. 6.

1. S~3~7 6./R3~ 73°Z3' 6. .953 s~<z z o.$dZ 3~96 0.fd~6 63°<7S' 6. 8Q2 Z.3~6

J. 683 0.666 f^V/' 0.71S f.fiZ 4-3Z73'8 0.699, 44*76' a. 7/6 oT/6 3~.0<?

s. QZ6 0.731 <?Z°S6' 7). 08/ 3~.3/ ifJO

6. 9/3 0. 7.93 rV\lZ' 0.609 3~.Z8 6.88

7 993 /\.<f<T7 J~.ZO 7878, //2>7 O.SZ6 V'8 0S63 7T.97 Q.80

9. /Z39 0,6?9 3/°d'3' 6.SZ f\ 6.fZ /o-oz

10. /ZQ7 16° 6' 07733 s.ss 77.00

II. 7f/Z 17° 0.f63~ 6.ZS /'.So

/3~d~8 n.&Sdr 2?°<74' 6,<77J~ 6.86 73. 7 6

Id. /73C a 890 17° 7' 7.47 /<?.60

/<? 7360 Q.8W 27°4</' 7>.463~ 8-2f sj:63~

IS, Z/ 30 n.Qfio Z3°Z7' 0.47f 3.9Z /8<?o

lb. 3.338 6.ft7?\ 0.4?

A

76. 60 /9.S0

11 2??o 7).9,72 0.4JQ 77.38 2/./ 7/8. ZS~6Z 7). ®7o 23°& 7)f9Z 72. 08 2/.<70

19. 2730 o.aon 29°76' a. <796 72,88 ZZ.J3-

zo. 3/ /& 0.86 3 36°Z\6 0.3~6S J3-.<7Z 26*74

z / VZ/O o.8s.9 D. S77 76.02 26.98zz Jd~6o o.s<r<r 37*W 0.^78 /7-7f Z3.r0

We si//j<j/ia use. JL Phase, 6 Q oo, Z>H.R Itjduntie // fi/o to r.

EUGENE DIETZGEN CO.. CHICAGO.

Af>ri) IJ, 19%

I

EUGENE DIETZGEN CO., CHICAGO.

Pl&te No. JO,

fira/ielest. Gcene.r&\ ferfor/nonce.

We s-titfj /jOi/se Z Phase , 60 00;

Z> H.P. Induct/a/j f^otaf.

Ex/=er//7?e/7 T Ac. 2. Jnprzssed Vo)taje 110.

A/>ri) I8} /30 J.

EUGENE DIETZGEN CO.. CHICAGO.

EUGENE DIETZGEN CO., CHICAGO.

EUGENE DIETZGEN CO., CHICAGO.

EUGENE DIETZGEN CO., CHICAGO.

wm * - * .* * r + , * mr- f5 ~ * if

w^M^ W/fflr^-- /F^f^A IvfaR^i^H^\ ij^fv^

4.

f ,

-#-

# - 4 #

Ufa

I

*,

+ * +

5 s(fc

~ ~ * ..'

*

# # #. #; #' ^ ^ #

^ I * I ^ ^ ^ I_^ If I -f 4 # I I I

^ +

mJ- JL. ,JL^

:

^ *

IP -i^P #

^> - * # # m5i lw/ .JK^Ti \i/frMxC\\uy.

^ f # * f #

* m'4 : * f f

"4

# * .# .+ * *- ^ * * *

+ - * * + + + + 4 * • * t + * *"!

i

^ ^ JL

liirfiiiiiiiiii

J< f-i-4 ' * # + * fr.:;:

i& ^ # * * #

Hi

# #

* * * T ^r-

T- IP

^f1 1p : ;'

7:^^>^ I;,,

ij/kj ' ^ j; ^.

i ^ '

^ ^ ^I # # # ^ #t * * * #

& ^ ^ ^ ^dk' *t ' ' JL

# #| # # ^ # ^ * #

Jr . Jt;

^jL. .- JL

J» -a A Jk Jit JL ^ JL Jk-f r- t1- f t- *• 5^ t

/ ^^^^1W^^su^wi )^i|^7 ^^J^' -yk .

;' '^J^ 1^

- JL- +* +

# # 4- -f- f ^ ^ * f +f- * # f # # * * * * * *?

:

M, 7

# I * ^ *

p ^ ^r- r^ ^ i ' ^* T )T T

:* * i f • # #