and
Book
bY
Central of the Westinghouse
EAST PITTSBURGH,
Station
Engineers Electric Corporation
PENNSYLVANIA
Copyright
1964 Electric Fourth Corporation, Printing East Pittsburgh, Pennsylvania
by Westinghouse Fourth Edition: Printed
in the United States of A,merica
_(
_:
‘,
and Distribution
‘,,,’
?
This book is dedicated to the memory of ROBERT D. EVANS
who contributed so greatly to the transmission and distribution of electric power and to the preparation of the original edition of this book
Preface to fourth
edition
Some thirty years ago a wellknown electrical engineer was ordered by his physician to take a complete rest. During this period, as a diversion, he began to study transmissionline calculations. Out of that came, in 1922, a book that was quickly recognized as a classic on the subject because it was simple, practical, useful. The man was William Nesbit; the book, “Electrical Characteristics of Transmission Circuits.” In the two succeeding decades powertransmission systems grew tremendously in complexity. Voltages were doubled, longer lines were built, interconnections became more extensive, knowledge of how to protect against lightning was greatly increased, and new methods of calculating performances were adopted. With all this grew the need for a new book on transmission lines, one of broader scope that would meet the new conditions, but retain the entirely practical viewpoint of its predecessor. Fourteen men, all connected with the Central Station Engineering Group of the Westinghouse Electric Corporation, undertook to produce such a book. All of these men worked daily on actual problems such as are considered here. With this background of experience and with the reputation of the Nesbit book as inspiration, they presented in January, 1942 the first edition of a book which they hoped would be useful to all concerned with electricpower transmission as a practical reference book, helpful in solving everyday problems. In 1943 a second edition was brought out in which two chapters that discussed the general features of the electrical distribution problem were added at the end of the book. The third edition differed from the second edition only in that the two chapters were introduced just before the appendix. A fourth and completely rewritten .edition is presented herewith. It contains essentially the material of the previous three editions, sometimes with new authors, and three new chaptersExcitation Systems, Application of Capacitors to Power Systems, and Power Line Carrier Application. As before, all of the authors are from the
Central Station Section or are closely associated with it. As was the case with previous editions, this one also bears the imprint of two outstanding engineers, who contributed so much to the transmission of power, Dr. Charles L. Fortescue and Mr. Robert D. Evans. The latter, before his recent death, was one of the active participants in the previous editions. The name or names of the original authors and the revising authors appear at the head of each chapter. To conform to the original standards regarding the sign of reactive power, the authors in the first edition of this book found it necessary to change the curves and discussions from what they had used in their previous publications. With the recent change in the standards, the sign has again been changed so that the curves and discussions now use lagging kvar as positive. The material presented here is naturally the results of research and investigations by many engineers. It is not feasible to list here the names of the companies and individuals whose work has been summarized. These acknowledgments are given in the individual chapters. Much of the material used has been the result of cooperative studies of mutual problems with engineers of electricpower companies, the conductor and cable manufacturers, and the communication companies. The authors gratefully acknowledge the hearty cooperation of those engineers whose work has assisted in the preparation of this book. Thetitle page photograph is reproduced by permission of the Bureau of Reclamation, Grand Coulee, Washington. The acknowledgments would be incomplete without giving recognition to the fine cooperation of the editorial staff of the Westinghouse ENGINEER, in reviewing the material and making many helpful suggestions to the authors and to Mr. Raymond W. Ferguson, who assisted in editing the material.
A. C. MONTEITH
,
Vice President in Charge of Engineering
C. F. WAGNER
Consulting Engineer September 1, 1950
Contents
Original Aulhor m and Revising Author
CHAPTER 1
General Considerations of Transmission
C. A. Powel
n
. . . . . . . . . . . . .
. . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
page
1
C. A. Powel
Symmetrical Components
J. E. Hobson
n
.
. .
. .
page 12 page 32 puge 64 page 96
page 145
D. L. Whitehead
Characteristics of Aerial Lines
Sherwin H. Wright and C. F. Hall
n
D. F. Shankle and R. L. Tremaine
4 5 6
Electrical Characteristics of Cables .
H. N. Muller, Jr.
n
J. S. Williams
Power Transformers and Reactors .
J. E. Hobson and R. L. Witzke
n
R. L. Witzke and J. S. Williams
Machine Characteristics
C. F. Wagner
n
. .
. .
. .
. .
C. F. Wagner
Excitation Application
Systems
.
pdge 195
page 233
J. E. Barkle, Jr.
of Capacitors to Power Systems .
A. A. Johnson
9 10
Regulation and Losses of Transmission Lines .
G. D. McCann n R. F. Lawrence
page 265
SteadyState Performance of Systems Including Methods of Network Solution . _ . . . . . . . . . . .
E. L. Harder
n
. . .
page 290 page 342 page 401
E. L. Harder
11 12 13
Relay and Circuit Breaker Application. PowerLine Carrier Application
R. C. Cheek
. . .
. .
. .
. .
. .
. .
E. L. Harder and J. C. Cunningham 8 E. L. Harder and J. C. Cunningham
.
.
PowerSystem StabilityBasic Application. . . . . .
Elements of Theory and . . . . . . . .
.
.
page 433
R. D. Evans and H. N. Muller, Jr. m J. E. Barkle, Jr. and R. L. Tremaine
14
PowerSystem Voltages and Currents During Abnormal Conditions . . . . . . . . . . . . . .
R. L. Witzke = R. L. Witzke
.
.
page 496
Original Author
CHAPTER 15
n
and Revising Author
Wave Propagation on Transmission Lines .
C. F. Wagner and G. D. McCann a C. F. Wagner
. . . . . .
. . . . . .
. . . . . .
. . . . . ,
. . . . . . . , . . .
. . . . . . . . . . .
page 523 page 542 page 5% page 610 page 643 page 666 page 689 page i19 page 741
page
16 17
18
Lightning Phenomena
.
.
.
.
. . . . .
. . . . .
C. F. Wagner and G. D. McCann m C. F. Wagner and J. M. Clayton
Line Design Based on Direct Strokes Insulation Coordination . .
n
A. C. Nonteith m E. L. Harder and J. M. Clayton
.
.
A. C. Monteith and H. R. Vaughan
A. A. Johnson
19 20 21 22 23
Grounding of PowerSystem Neutrals
8. B. Griscom
n
S. B. Griscom
Distribution
Systems .
.
.
n
.
.
John S. Parsons and H. G. Barnett
John S. Parsons and II. G. Barnett
Primary and Secondary Network Distribution Lamp Flicker on Power Systems
S. B. Griscom m S. B. Griscom
Systems . . . . . . . . . .
John S. Parsons and H. G. Barnett m John S. Parsons and H. G. Barnett
.
.
.
Coordination
R. D. Evans
n
of Power and Communication
R. L. Witzke
Systems. . . . . . .
Appendix Index .
. .
. .
. .
. .
. .
. .
. .
. .
. .
784
page 813
CHAPTER
1
GENERAL
Original Author:
CONSIDERATIONS
OF TRANSMISSION
Revised by:
C. A. Powel
HROLGH discovcry, invention, and engineering application, the engineer has made electricity of continually grcnter 11seto mankind. The invention of the dynamo first mntle engine power many tiines more effective in relieving the toil and increasing the opportunities and comforts not only of industry but also of the home. Its scope, hoxever, was limited to relatively short distances from the powr station because of the low voltage of the distribution circuits. This limitation, for economic reasons, kept, the general IN: of electricity confined to city areas \\.herc a number of customers could be served from the same power station. The nest step in the development of the presentday electric systems was the invention of the transformer. This invention was revolutionary in its effect on the electric industry because it made high voltage and long transmission distances possible, thus placing the engine po\\er, through the medium of the nlternatingcurrent generator, at the doorstep of practically everyone. The first alternating current system in America using transformers was put in operation at Great Barrington in Massachusetts in 1886. Mr. William Stanley, Westinghouse electrical espert IIho was responsible for the installation, gives an account of the plant, part of which reads:
C. A. Powel
Barrington to its present size involving as it does a cnpitalization in the privatelyowned powcr companies of some 17 billion dollars with an annual revenue of 4 billion dollars. The growth since the beginning of this century in installed generating capacity of all electric polver plants
T
(a
(b)
“Before leaving Pittsburgh I designed several induction coils,
or transformers as we now call them, for parallel connection. The original was designed in the early summer of 1853 and wound for 500 volts primary and 100 volts secondary emf. Several other coils were constructed for experimental purposes. “iit the north end of the village of Great Barrington was an old
Fig. l(a) Gaulard and Gibbs transformer for which George Westinghouse had secured all rights in the United States. (b) First transformer designed by William Stanley. The prototype of all transformers since built, it definitely established the commercial feasibility of the alternatingcurrent system, 18841886.
desertedr&her mill which I leasedfor a trifling sum and erected
in it a 2.5 hp boiler and engine that I purchased for the purpose. After what seemed an interminable delay I at last installed the Siemens alternator that 1\Ir. Westinghouse had imported from London. It was wound to furnish 12 amperes of current with a
maximum of ,500volts. In the meantime I had started the construction of a number of transformers in the laboratory and engaged a young man to canvass the town of Great Barrington for light customers. Ke built in all at Great Barrington 26 trnnsformers, 10 of which were sent to Pittsburgh to be used in a demonstration plant between the Union Switch and Signal Company’s factory* and East Liberty. “We installed in the town plant at Great Barrington two 50light and four 25light transformers, the remainder being used in the laboratory for experimental work. The transformers in the village lit 13 stores, 2 hotels, 2 doctors’ ofices, one barber shop, and the telephone and post offices. The length of the line from
contributing to the public supply has been from about 13 million kilowatts to 55 million kilowatts in 1948. Of this 55 million kilowatts the privatelyowned utilities accounted for 44 million kilowatts and governmentowned ut,ilities for 11 million kilowatts divided equally between the federal government and local governments. Thus, 80 percent of the generating capacity of the country is privately owned and 20 per cent government owned. With this 55 million kilowatts of generating capacity, 282 billion kilowatthours, divided 228 billion kilowatthours by privatelyowned generation and 54 billion public, were generated in 1948. The average use of the installed capacity for the country as a whole was, therefore, 282 000 p5130 hours, and the capacity factor for the 55 5130 country as a whole 87G0= 58.5 percent. This capacity factor of 58.5 percent is generally conceded as being too high. It does not allow sufficient margin to provide adequate spare capacity for maintenance and repairs. Fig. 2 illustrates how the spare and reserve capacity has shrunk in the past few years. 11 ratio of installed capacity to peak load of 1.15 to 1.20 is considered necessary to provide a safe margin for emergencies. Such
1
the laboratory to the center of the town was about
4000
feet.”
Our centralstation industry today is, for all practical purposes, entirely alternating current. It can, therefore, be said to have grojvn from the small beginning at Great
*About two miles.
General Considerations of Transmission
Chapter 1
The average cost of all electricity used for residential service has shown a steady downward trend since 1925 from 7 cents per kilowatthour to 3 cents in 1948. This is all the more remarkable as since 1939 all other items making up the costofliving indes have shown increases ranging from 10 percent (for rents) to 121 percent (for food), the average increase of all items being 69 percent. The revenue from sales to residential customers accounts for about 36 percent of the total utility revenue; to large power customers about 29 percent; t,o small light ant1 power customers 27 percent, and to miscellaneous customers (railroads, street lighting, etc.) 8 percent.
1. Sources of Energy
The sources of energy for largescale generation of elect,ricity are: 1. Steam, from (a) coal, (b) oil, or (c) natural gas 2. Water (hydroelectric) 3. Diesel power from oil Other possible sources of energy are direct solar heat, windpower, tidal power, shale oil, and atomic energy, but none of these as yet has gone beyond the pilotplant stage, for the reason that coal and petroleum are still abundantly available. But as fossil fuels become scarcer and more OF KILOWATTS expensive, there is every reason to believe that all of these, OF KILOWATTS as well as petroleum manufactured from vegetable matter, may become useful and economical supplementary sources II/IIIlIIl ^ ot energy. The estimated reserves of coal and lignite in the United 1940 1910 1920 1930 1950 ls6’ States are about 3000 billion tons. This constitutes almost YEAR 99 percent of the mineral fuel energy reserves of the Fig. 2Trend in production of electricity, installed capacity, country; oil shale, petroleum and natural gas amounting and sum of peak demands. to little more than 1 percent.’ a margin in 1948 would have given a capacityfactor of By far the greater part of the electric energy generated about 53 percent, instead of 58.5 percent. in this country is obtained from fuel, the 55 million kilo
I I I I I I
TABLE
~PREFERRED
STANDARDS FOR LARGE 3600RPM ~PHASE 60CYCLE CONDENSING STEAM TURBINEGENERATORS

1 1irCooled
Generator
HydrogenCooled Generators Rated for 0.5 Psig Hydrogen Pressure 15000
Turhinegenerator rating, kw
Turhine capability, kw Generator rating, kva power factor shortcircuit ratio Throttle pressure, psig Throttle temperature, F Reheat temperature, F Sumher of extraction openings Saturation temperatures at, Is1 t 2nc1 openings at “turbinegenerator rating” with all ex3rc1 traction openings in serv4tt 1 ice, F 5t11 Eshaust pressure, inches Hg abs Generator capability at 0.85 power factor am.I 15 psig hydrogen pressure, kvs Generator capability at 0.85 power factor ant 30 psig hydrogen pressure, kva 1
11500
12650 13529 0.85 0.8 600 825 4 175 235 285 350
16 500 17647
0.85 0.8 850
20000 22000 23529 0.85 0.8 850
900
30000 33000 35294 0.85 0.8 850
900
40 000
44 000
60000
66 000
no ooo*
99 000 I
900
4 175 235 285 350 1.5
20 394
47058 0.85 0.8 /850\orj1250\ \9ooj \ 950/ 5 175 235 285 350 410 1.5 64117
4 175 235 285 350
1 .5 8i 058
1.5
5 175 235 285 350 410 1.5 40588
105882 0.85 0.85 0.8 0.8 /85O\orj 1250 \ Fl45Ojorjl4~50\ ** \900( \ 95oj :1oooj \lOOO( 1000 5 5 5 175 180 Ii.5 235 245 240 28.5 305 300 350 380 370 410 440 440 1.5 1.5 I.5 81176 131764
70588
*A 10 percent pressure drop is assumed between the high pressure turbine exhaust and low pressure turbine inlet for the reheat machine. **Them nre two different units; the first for regenerative cycle operation, and the second B machine for reheat cycle operation.
Chapter 1
General Considerations sf Transmission
Fig. 3The first centralstation turboalternator installation in the United Statesa 2000kw turbine coupled to a 60cycle generator, 2000 kw, 2400 volts, twophase, 1200 rpmat the Hartford Electric Light Company, Hartford, Connecticut, 1900. This turbine was about four times as large as any one built before that time and caused much comment the world over.
watts of installed capacity being made up of approximately 35 million kilowatts of steam turbines and one million kilowatts of diesel engines. .1pprosimately 16 million kilowatts of the installed capacity are in hydroelectric stations. Of the 282 billion kilowatthours generated by all means in 1948, roughly 200 billion came from fuel; 76 percent from coal, 14 percent from natural gas, and 10 percent from oil.
2. Development of Steam Power
The modern steamelectric station can be dated from the installation by the Hartford Electric Company in 1900 of a 2000kw unit (Fig. 3) which at that time was a large machine. Progress in design and efficiency from then on has been continuous and rapid. In 1925 the public utilities consumed in their fuelburning plants an average of 2 pounds of coal (or coal equivalent) per kilowatthour, whereas today the corresponding figure is 1.3 pounds per kilowatthour. This average figure has not changed materially in the last 10 years. It would appear that the coal consumption curve is approaching an asymptote and that a much better overall performance is not to be expected, even though the best baseload stations generate power for less than one pound of coal per kilowatthour. The very high efficiency in the best baseload stations is obtained at a considerable increase in investment. It cannot be economically carried over to the system as a whole for the reason that there must be some idle or partly idle capacity on the system to allow for peaks (seasonal and daily), cleaning, adjustments, overhaul, and repairs. How much one can afford to spend for the improvement of station efficiency above “normal” depends on the shape of the system load curve, the role of the station in that curve, and the cost of fuel. Most of the credit for the improvement in steam consumption goes to the boiler and turbine manufacturers who through continuous betterment of designs and materials have been able to raise steam pressures and temperatures. Between 1925 and 1942 the maximum throttle pressure was raised from 1000 psi to 2400 psi and the average from 350 to 1000 psi. In the same period the throttle temperature was raised from 725 to 1000 degrees F. and the
AGE OF CAST
STEEL+L~;EoO&i~ AGE OF SUPER ALLOYS
Fig. 4Progress
in turbine
generator
design.
average from 675 to 910 degrees. Generator losses in the meantime have been greatly reduced from about 6 percent in 1900 to 2 percent today, but these losses never did form a large part of the total, and their influence on the overall performance of the station has been minor. The increase in maximum size of 60cycle, twoand fourpole generating units over the years since 1900 is shown in Fig. 4. The remarkable increase has been due to improved materials and designs, particularly in large forgings, turbine blading, and generator ventilation. In 1945 the American Society of Mechanical Engineers and the American Institute of Electrical Engineers adopted standard ratings for turbinegenerator units. These were revised in November 1950 to include the 90 000 kw unit and are listed in Table 1. The machines are designed to meet their rating with 0.5 psi hydrogen pressure, but experience has shown that between 0.5 and 15 psi the output of the generator can be increased one percent for each pound increase in the gas pressure without exceeding the temperature rise guarantee at atmospheric pressure. In many locations operation at more than 15 psi gas pressure
4
General Considerations of Transmission
Chapter 1
may be difficult because of codes regulating operation of “unfired pressure vessels” at greater pressures, but serious consideration is being given to operation at 30 lhs. For a hydrogenair mixture to be explosive, the percentage of hydrogen must lie between 5 and 75 percent. The control equipment is designed to operate an alarm if the purity of the hydrogen drops below 95 percent. The density meter and alarm system is in principle a small constantspeed fan circulating a sample of the mixture. If the density varies, the drop of pressure across the fan varies and registers on the meter.
is preferred, in which a single combination guide and thrust bearing is located below the rotor (Fig. 1, Chapter 6). Where the axial length of the machines is too great an additional guide bearing must be provided. In this case the combination thrust and guide bearing is usually located above the rotor and the additional guide bearing below the rotor. The advantages of the umbrella design are (a) reduction in overhead room to assemble and dismantle the unit during erection and overhaul, and (b) simplicity of the single bearing from the standpoint of cooling and mini
3. Development
of Water Power
The great transmission systems of this country received their impetus as a result of hydroelectric developments. Forty years ago conditions favored such developments, and in the early years of this century waterpower plants costing $150 per kilowatt or less were common. Steam stations were relatively high in first cost and coal consumption per kilowatt hour was three times as much as today, and finally fuel oil was not readily available. As undeveloped waterpower sites became economically less desirable, steam stations less costly and their efficiency higher, and as oil fuel and natural gas became more generally available through pipe lines, steam stations rapidly outgrew hydroelectric stations in number and capacity. Today very few waterpower sites can be developed at such low cost as to be competitive with steam stations in economic energy production. For this reason hydroelectric developments of recent years have almost all been undertaken by Government agencies, which are in a position to include in the projects other considerations, such as, navigation, flood control, irrigation, conservation of resources, giving them great social value. As the waterpower developments within easy reach of the load centers were utilized and it became necessary to reach to greater distances for water power, only large developments could be considered, and stations of less than 100 000 kw became the exception rather than the rule, as witness Conowingo with 252 000 kw, Diablo with 135 000 kw, Fifteen Mile Falls with 140 000 kw, Osage with 200 000 kw, and many others. The developments of recent years undertaken by various government agencies have reached gigantic proportions, as for example Hoover Dam with 1000 000 and Grand Coulee with 2 000 000 kw installed capacity. A natural corollary to the increase in station capacity has been a gradual increase in the size of the individual generator units, the growth of which is shown in Fig. 5, culminating in the Grand Coulee generators of 120 000 kw at 120 rpm with an overall diameter of 45 feet. Most of the multipurpose hydraulic developments call for large, slowspeed machines. For such conditions vertical units are used to obtain maximum energy from the water passing through the turbine. The rotating parts are supported by a thrust bearing which is an integral part of the generator. TWO general types of generator design are used as distinguished by the arrangement of the guide and thrust bearings. Where the axial length of the generator is short in relation to its diameter, the “umbrella” design
YEAR
Fig. JTrend
in maximum
waterwheel
generator
ratings.
mum amount of piping. The design also lends itself readily to a totallyenclosed recirculating system of ventilation, which keeps dirt out of the machine and facilitates the use of fireextinguishing equipment. It also reduces heat and noise in the power house.
4. Combination
of Water and Steam Power
There are very few locations today where an important market can be supplied entirely from water power because of seasonal variations in river flow, but in most cases a saving will be realized from combining water power and steam. The saving results from the combination of low’ operating cost of waterpower plants with low investment cost of steam stations. Moreover, hydroelectric units in themselves have certain valuable advantages when used in combination with steam units. They start more quickly than steamdriven units, providing a high degree of standby readiness in emergency.
Chapter 1
General Considerations of Transmission
5
They are well adapted to maintenance of frequency, and also to providing wattless energy at times of low water flow. And finally, hydropondage can be drawn upon to relieve steam plants of shorttime peaks to save banking estra boilers. To what extent a waterpower site can be developed economically involves a thorough investigation of individual cases. An economic balance must be struck between the steam and water power to give maximum economy. One might install enough generating capacity to take care of the maximum flow of the river during a short period. The cost per kilowatt installed would be low but the use made of the equipment (capacity factor) would also be low. Or one might put in only enough generating capacity to use the minimum river flow. In this case the cost of the development per kilowatt installed would be high, but the capacity factor would be high
mission). The latter group in this particular study nas about $70 per kilowatt. Curve A gives the total cost, of energy per kilowatt, hour for a modern steam plant costing $95 per kilo\vatjt with fixed charges at 12 percent and coal at 8l a ton. Curve B gives the total cost of energy from the wnterpower plant having the capital cost indicated in Curve C. To obtain such a curve it is necessary to determine the amount of energy available at the various capacity factors, the assumption being made that all hydro capacity installed is firm capacityi, that is, that the system load can absorb all of the energy generated. Curve B shows the typically high cost of hydroelectric energy as compared with steam at high capacity factors and its low cost at low capacity factors.
5. Transmission
Liability
In a hydroelectric development the transmission becomes a large factor of expense and in comparing such developments with equivalent steam plants, it is necessary to include the transmission as a charge against the hydroelectric plant. Figures of cost published on the Hoover DamLos Angeles 287kv line indicate that this transmission costs over $90 a kilowatt, and other lines contemplated will probably show higher costs. Under certain conditions it may be more costly to transmit electrical energy over wires than to transport the equivalent fuel to the steam station. It has been shown3 that the cost of electric transmission for optimum load and voltages can be expressed as a linear function of power and distance, as follows: 0.61 X miles For 5Oyc load factor: mills/kwhr = 0.54+
100
For 90% load factor: mills/kwhr
=0.30+
0.35 X miles
100
It was also shown that fuel transportation can be expressed as a linear function of energy and distance, thus:
IO 20 30 40 50 60 70 80 SO 100 CAPACITY FACTORPERCENT
Fig. 6Cost
of energy at various and hydroelectric
capacity plants.
factors
of steam
also. Obviously between these two extremes lies an optimum value. The ratio of installed waterpower capacity to the peak load of the system that gives the minimum annual cost of power supply has been referred to as the “economic hydro ratio,” and it can be determined without great difficulty for any particular set of conditions. In a paper2 presented before the American Society of Mechanical Engineers, Irwin and Justin discussed in an interesting and graphical manner the importance of incremental costs on the economics of any proposed development. Fig. 6, taken from their paper, shows in Curve C the capital cost per kilowatt of installation for various capacity factors. The costs were segregated in items that would be the same regardless of installation (land, water rights, dams) and those that vary with the amount of installation (power house, machinery, trans
Railroad rates on coal $1.20+5+ mills per mile Pipeline rates on crude oil $5.00+4 cents per mile per 100 barrels For pipeline rates on natural gas two curves were given for estimated minimum and maximum interruptible contract rates $0 +12 cents per mile per million cubic feet $50+12 cents per mile per million cubic feet The authors point out that a comparison between transmission costs alone for gas, oil, and coal are likely to be misleading because there is a wide difference in the costs of the fuels at their source. There is also a considerable variation in the transportation costs above and below the average.
t“Firm Capacity” or “Firm Power” in the case of an individual station is the capacity intended to be always available even under emergency conditions. “Hydro Firm Capacity” in the case of combined steam and hydro is the part of the installed capacity that is capable of doing the same work on that part of the load curve to which it is assigned as could be performend by an alternative steam plant.
6
General Considerations of Transmission The equivalence between the fuels is given as: 1 tonof coal . . . . . .._........ 25000000BTU I barre1ofoi1...___..._,.......... G250000BTU 1OOOcubicfeetofgas . . . . . . . . . . . . . . . 1OOOOOOBTU
Chapter 1
6. Purpose of Transmission
Transmission lines are essential for three purposes. a. To transmit power from a materpower site to a market. These may be very long and justified becauseof the subsidy aspect connected with the project. b. For bulk supply of power to load centers from outlying
steam stations. These are likely to be relatively short. c. For interconnection purposes, that is, for transfer of energy from one system to another in case of emergency or in response to diversity in system peaks.
Longdistance directcurrent transmission has also been considered. It offers advantages that look attractive, but present limitations in conversion and inversion equipment make the prospect of any application in the near future unlikely. In many industrial applications, particularly in the machinetool industry, 60 cycles does not permit a high enough speed, and frequencies up to 2000 cycles may be necessary. Steps are being taken to standardize frequencies of more than 60 cycles.
8. Choice of Voltage
Transmission of alternatingcurrent power over several miles dates from 1886 when a line was built at Cerchi, Italy, to transmit 150 hp 17 miles at 2000 volts. The voltage has progressively increased, as shown in Fig. 7, until in 1936 the Hoover DamLos Angeles line was put in service at 287 kv. This is still the highest operating voltage in use in the United States today, but considerntion is being given to higher values. An investigation was begun in 1948 at the Tidd Station of the Ohio PO\\er Company on an experimental line with voltages up to 500 kv. The cost of transformers, switches, and circuit breakers increases rapidly with increasing voltage in the upper ranges of transmission voltages. In any investigation involving voltages above 230 000 volts, therefore, t,he unit cost of power transmitted is subject to the law of diminishing returns. Furthermore, the increase of the reactance of the terminal transformers also tends to counteract the gain obtained in the transmission line from the higher voltage. There is, therefore, some value of voltage in the range being investigated beyond which, under esisting circumstances, it is uneconomical to go and it may be more profitable to give consideration to line compensation by means of capacitors to increase the economic limit of
Frequent attempts have been made to set up definitions of ‘Yransmission lines, ” “distribution circuits” and “substations.” None has proved entirely satisfactory or universally applicable, but for the purposes of accounting the Federal Power Commission and various state commissions have set up definitions that in essence read:
A transmission system includes all land, conversion structures
and equipment at a primary source of supply; lines, switching and conversion stations between a generating or receiving point
sntl the entrance to a distribution center or wholesale point, all
lines and equipment whose primary purpose is to augment, integrate or tie together sourcesof power supply.
7. Choice of Frequency
The standard frequency in Xorth America is 60 cycles per second. In most foreign countries it is 50 cycles. As a generalpurpose distribution frequency 60 cycles has an economic advantage over 50 cycles in that it permits a maximum speed of 3600 rpm as against 3000 rpm. Where a large number of distribution transformers are used a considerable economic gain is obtained in that the saving in materials of 60cycle transformers over 50cycle transformers may amount to 10 to 15 percent. This is because in a transformer the induced voltage is proportional to the total fluxlinkage and the frequency. The higher the frequency, therefore, the smaller the crosssectional area of the core, and the smaller the core the shorter the length of the coils. There is a saving, therefore, in both iron and copper. The only condition under which any frequency other than 50 to 60 cycles might be considered for a new project \vould be the case of a long transmission of, say, 500 OI 600 miles. Such long transmission has been discussed in connection with remote hydroelectric developments at home and abroad, and for these a frequency less than 60 cycles might be interesting because as the frequency is decreased the inductive reactance of the line, 2rfL, decreases and the capacitive reactance, increases, 27r$’ resulting in higher load limits, transmission efficiency, and better regulation. Full advantage of low frequency can be realized, however, only where the utilization is at low frequency. If the low transmission frequency must be converted to 60 cycles for utilization, most of the advantage is lost because of limitations of terminal conversion equipment.
1
YEAR Fig. 7Trend in transmission voltages in 60 years.
Chapter 1
General Considerations of Transmission
7
TABLE ~FORM OF TABULATIONFOR DETERMININGVOLTAGES AND CONDUCTOR SIZES Based on the Transmission of 10 000 Kva for 10 Miles at 80 Percent Power Factor Lagging, GOCycle, 3Phnsc
COSDCCTORS . Total I=R Loss 2 300 I Kva , / VOLT.kGE DROP AT FULI, TOAD ANNU.~l. OPERr\TING COST
10 000 Iiva
.
7 40.5 .; 1 2.5 1 RIO001 4, Ill= 11.0 tLl 100 $09 600 I 0 ,500 $1 600 $1 ZOO $120 300 24600 60600 0 500 1 600 1 non 106 800 45 2 8RO 001 7 2 I.? 4 14.0 ‘) 720 0.0 11.5 I4 n 17.;1 15500 69600 0500 1000 1.500 !)7 700 , I l10 I 14.3 I 71 /4520001 ____ ~~ .4.0 6.41 7.0 19.500 71600 9700 2400 2200 IE;;; ii 8.1 68105 9 700 7 I 600 8 700 2 400 2 200 .r Li 12.9 7 0 14.S 0000 ilOO 9700 2 400 a 200 91 900 ______ ~~~~4.5 3 8 5.R 0700 74400 10000 3400 3000 101 400 !I 1 4.0 !I.7 4 son 74 400 10900 3400 3 000 ‘Jfi 300 $7 580 6 410 5 8Fo fi 330 $5740 5 520 Ii I)!)0 5 700 $12 A30 sin 100 S31i 310 10 680 28 fin0 45 R!W 9 7iO 4.5 200, GO 8:iO IO ,540 !I .xjo !I 190 10 140 !I GO ifi 100 32 200 5 1 200 18 000 30 100 32 970 4i .Tan Ii3 !1 IO :%A230 31 .%I)
TABLE
~@JICK~~TI!~ATIN~ DATA ON THE LOAD CARRYING CAPACITY OF TRANSMISSION LINEst

Kw Which
Dclivcrcd I,inc Voltage

Can Be Delivered Based on SO;0Regulation and 90°j0 Power Factor Distance in ;lMes
13.2 kv3foot spacing Stranded Copper 4 2 4/o 33 kvlfoot spacing Stranded Copper 1
I
T
5 950 1 loo 3 000 10 ?5 000 6 700 8 350 11500
10 490 TOO 1 500 20 2 500 3 350 4 180 5 750 ~~ 40
1d 330 470 1 000 30 1 TOO 2 200 2 800 3 800 60
I
20 245 350 750 40 1 250 1 TOO 2 100 2 900 80 3 140 3 990 4 590
2/o
d/O 300 000 66 kv&foot spacing Stranded Copper

2/o
4/O 300 000
6 250 8 000 9 150
4 180 5 320 6 120

KIT Which Can Be Delivered Based on 107, Loss and Equal Voltage at Sending and Receiving Ends Distance in Miles 132 kv16foot spacing Stranded Copper 4/O 300 000 500 000 220 kv2bfoot spacing Hollow Copper 500 000 ACSRX5 000
L
40 1
80
120
160 30 100 44 800 77 100 320 119 000 118 500
tData
obtained from Figs. 19 and 22 of Chap. 9.
power transmission than increase the voltage much above present practice. The basic principles underlying system operation as regards voltages have been set forth in a report” \vhich lists the voltages in common use, the recommended limits of voltage spread, and the equipment voltage ratings intended to fulfill the voltage requirements of the level for which the equipment is designed. The report shoultl be carefully studied before any plans are made involving the adoption of or change in a system voltage. In selecting the transmission voltage, consideration should be given to the present and probable future voltage of other lines in the vicinity. The advantages of being able to tie together adjoining power districts at a common voltage frequently outweighs a choice of voltage based on lowest immediate cost. If the contemplated transmission is remote from any existing system, the choice of voltage should result from a complete study of all factors involved. Attempts have been made to determine by mathematical expression, based on the wellknown Kelvin’s Law, the most economical transmission voltage with all factors evaluated, but these are so numerous that such an expression becomes complicated, difficult, and unsatisfactory. The only satisfactory way to determine the voltage is to make a complete study of the initial and operating costs corresponding to various assumed transmission voltages and to various sizes of conductors. For the purposes of the complete study, it is usually unnecessary to choose more than three voltages, because a fairly good guess as to the probable one is possible without knowing more than the length of the circuit. For this preliminary guess, the quickestimating Table 3 is useful. This table assumes that the magnitude of ponel transmitted in the case of voltages 13.2, 33, and 66 kv is based on a regulation of 5 percent and a load power factor uf 90 percent,. In the case of 132 and 220 kv, the table is based on a loss of 10 percent and equal voltages at, the sending and receiving ends of the line. The reason for this and the bases of the calculations are given in Chapter 9. A representative study is given in Table 2. It is assumed
8
General Considerations of Transmission
Chapter 1
that it is desired to transmit over a singlecircuit ten miles long 8000 kw (10 000 kva) at 80 percent powerfactor lagging for 10 hours a day followed by 2000 kw (2500 kva) at 80 percent powerfactor for 14 hours. The preliminary guess indicates that 23, 34.5, or 46 kv are probably the conductor economical nominal voltages. Equivalent spacing and the number of insulators are as given in Table 4. Conductors of harddrawn stranded copper are
TABLE 4 CONSTRUCTION FEATURES OF TRANSMISSION LINES IN THE UNITED STATES* Line Voltage Length in Milts Max. 13.8 34.5 60 115 138 230 Equivalent L 9 pacing Type** Av. Av. Sumber of Insulators
35 40 40 133
25 25 25 4.5
100 100 140 260
8
10 15 F 8 1l 11 12 20
loss, leakage over insulators and the escape of energy through the air between the conductors (known as “corona’‘see Chap. 3) appear. In addition to these two losses, the charging current, which increases as the transmission voltage goes higher, may either increase or decrease the current in the circuit depending upon the powerfactor of the load current and the relative amount of the leading and lagging components of the current in the circuit. Any change in the current of the circuit will consequently be accompanied by a corresponding change in the 12R loss. In fact, these sources of additional losses may, in some cases of long circuits or extensive systems, materially contribute toward limiting the transmission voltage. The weight of copper conductors, from which their cost can be calculated, is given in Chap. 3. As an insurance against breakdown, important lines frequently are built with circuits in duplicate. In such cases the cost of conductors for two circuits should not be overlooked.
10. Choice of Spacing
Conductor spacing depends upon the economic consideration given to performance against lightning surges. If maximum reliability is sought, the spacing loses its relation to the operating voltage and then a medium voltage line assumes most of the cost of a highvoltage transmission without the corresponding economy. (See Chap. 17) In general a compromise is adopted whereby the spacing is based on the dynamic voltage conditions with some allowance for reasonable performance against lightning surges. Table 4 shows typical features of transmission lines in the United States including their “equivalent spacing” and the number of suspension insulators used. By equivalent spacing is understood the spacing that would give the same reactance and capacitance as if an equilateral triangular arrangement of conductors had been used. It is usually impractical to use an equilateral triangular arrangement for design reasons. The equivalent spacing is obtained from the formula D= +ABC where A, B, and C are the actual distances between conductors.
employed, the resistance being taken at 25 degrees C. The stepup and stepdown transformers are assumed as 2.5x 10 000 kva,l2 500 kva at either end, and highvoltage circuitbreakers are used in anticipation of future additional circuits. The costs of the pole line, rightofway, building, and real estate are not included as they will be practically the same for the range of voltages studied. Assuming that the cost figures in the table are correct, a 34 500volt line with No. 00 copper conductor is the most economical. The transmission loss will be 5 percent and the regulation 7 percent at full load, which is deemed satisfactory. The voltage is sufficiently high for use as a subtransmission voltage if and when the territory develops and additional load is created. The likelihood of early growth of a load district is an important factor in selection of the higher voltage and larger conductor where the annual operating costs do not vary too widely.
9. Choice of Conductors
The preliminary choice of the conductor size can also be limited to two or three, although the method of selecting will differ with the length of transmission and the choice of voltage. In the lower voltages up to, say, 30 kv, ‘for a given percentage energy loss in transmission, the cross section and consequently the weight of the conductors required to transmit a given block of power varies inversely as the square of the voltage. Thus, if the voltage is doubled, the weight of the conductors will be reduced to onefourth with approximately a corresponding reduction in their cost. This saving in conducting material for a given energy loss in transmission becomes less as the higher voltages are reached, becoming increasingly less as voltages go higher. This is for the reason that for the higher voltages at least two other sources of
*This table is based on information published in Electrical World and in Electrical Engineering. While it does not include all lines, it is probably representative of general practice in the U.S.A. **SCW Singlecircuit wood. SCSTSinglecircuit steel.
11. Choice of Supply Circuits
The choice of the electrical layout of the proposed power station is based on the conditions prevailing locally. It should take into consideration the character of the load and the necessity for maintaining continuity of service. It should be as simple in arrangement as practicable to secure the desired flexibility in operation and to provide the proper facilities for inspection of the apparatus. A review of existing installations shows that the apparent combinations are innumerable, but an analysis indicates that in general they are combinations of a limited number of fundamental schemes. The arrangements vary from the simplest singlecircuit layout to the involved duplicate systems installed for metropolitan service where the importance of maintaining continuity of service justifies a high capital expenditure. The scheme selected for stations distributing power at bus voltage differs radically from the layout that would be desirable for a station designed for bulk transmission.
or circuits. buses. Sometimes both are required. their use seems to be on the decline. This type of construction should be used only where interruptions to service are relatively unimportant because outages must exist to all feeders simultaneously when the bus. gFundamental schemes generator of connections voltage. The generating capacity connected to a bus may be so tNELA Publications Nos. Bustie reactors are also shown that. 164 and 278’20Elec. The disconnect also enables the breaker to be isolated from the bus for inspection and maintenance on the breaker. 8. The disconnect switch serves as additional backup protection for personnel. One breaker from each generator or feeder can be removed from service for maintenance with complete protection for maintenance personnel and without disrupting service to any feeder. Sketch (a) shows several feeders connected to a common bus fed by only one generator. However. as the improvement in performance over the conventional adjacent phase grouping is not sufficiently better to justify the estra cost. Each feeder has a circuit breaker and a disconnect switch. with the breaker open.Chapter 1 General Considerations of Transmission 9 In some metropolitan developments supplying underground cable systems segregatedphase layouts have been and are still employed to secure the maximum of reliability in operation. keep the shortcircuit currents within the interrupting ability of the breakers. one complete bus section can be removed from service for cleaning and maintenance or for adding an additional feeder without interfering with the normal supply to other feeder circuits. The circuit breaker provides protection against short circuits on the feeder and enables the feeder to be removed from service nhile it is carrying load if necessary. with all generators in service. App. These bustie reactors are important I (b) t (cl (4 Q MAIN BUS \ Q 2 A (4 SYNCHRONIZING BUS Fig. Sketch (e) shows a double bus commonly used where reactors are in series with each generator and each feeder. There are many intermediate schemes that can be utilized that give a lesser degree of flexibility. It is often questionable whether the expense of such an arrangement is justified and it should be used only where the importance of the service warrants it. With more than one generator complete flexibility is obtained by using duplicate bus and switching equipment as shown in (b). If the bus is supplied by more than one generator. for supply at . the reliability of supply to the feeders using this type of layout is considerably increased. generator or power source is out of service for any reason. There are also several connections differing in degree of duplication that are intermediate to the three layouts indicated. generator breaker. Several fundamental schemes for bus layouts supplying feeders at generator voltage are shown in Fig. during maintenance or repair work on the feeder. Also. large that it is necessary to use currentlimiting reactors in series with the generator leads or in series with each feeder. particularly in view of the continuing improvement of protective equipment and the more reliable schemes of relaying available today for removing faulty equipment. Comm. These vary from the simplest form of supply for a small industrial plant as shown in (a) to a reliable type of layout for centralstation supply to important load areas shown in (4 and (0 t. give a number of station and substation layouts. an example of which is shown in (c). An analysis of the connections in any station layout usually shows that they are built up from parts of the fundamental schemes depending upon the flexibility and reliability required. as for instance in (21). Quite frequently disconnect switches are arranged so that when opened the blade can be connected to a grounded clip for protection.
The separate compartments within the station should be locked and made as tight as possible for protection against accidental contact by operating personnel either physically or through the medium of a wire or any conducting material. circuit breakConmetalclad bus structures. bus work and feeders. The phase angle between bus sections becomes important when a station is supplying a network system and should be kept to a minimum to prevent circulating currents through the network. In Fig. A selector breaker is similar in every respect to the feeder breaker and serves as backup protection in case the feeder breaker does not function properly when it should open on a feeder fault. Unfortunately. and whether all the generatetl power is supplied in bulk over transmission lines or whether some must also be supplied at generator voltage. however. Fire walls are generally provided between bus sections or between each group of two bus sections to provide against the possibility of a fire in one section spreading to the adjacent sections. Sketch (e) shows only three bus sections. Also a highpotential test bus is necessary to test circuit breakers. . For a network supply at least four bus sections are generally used so that the network can still be supplied in case one bus section should trip out on a fault. and dusttight tinuous supply to all feeders is provided through reactor ties to a synchronizing bus should a generator fail. For any type of generatingstation design proper current and potential transformers must be provided to supply the various types of relays to protect all electrical parts of the station against any type of fault. transformers and transmission circuits. transformer and transmission circuit is of the same capacity and can be treated as a single entity. The synchronizing bus also serves as a point nhere tie feeders from other stations can be connected and be available for symmetrical power supply to all feeder buses through the reactors. 9. The station auxiliaries that go with each unit are usually supplied (0) t (b) Lu (4 Fig. The power system must be such that a whole unit comprising generator. The simplest layout is obtained when each generator.10 General Considerations of Transmission Chapter 1 because they not only limit the current on short circuit but also serve as a source of supply to the feeders on a bus section if the generator on that bus section fails. Bustie circuit breakers are provided to tie solidly adjacent bus sections for operation with one or more generators out of service. This is not the case for station design shown in (e) where a tie feeder must be brought in to a particular bus section. Each feeder can be connected to either the main or auxiliary bus through what is called a selector breaker. this is seldom the case because the number of generators do not equal the number of outgoing circuits. some simplification is possible if the transformers are selected of the same capacity as the generators. Likewise. Even here. Stations of this type would be expected to have four to six or more bus sections especially if the station supplies netnork loads. A ground bus must be provided for grounding each feeder when it is out of service for safety to personnel. (a) shows the “unit scheme” of supply. before being reconnected to the station. the main and auxiliary buses serve as one bus for the feeders connected to that section. current and voltage conditions must be obtained from’ current and potential transformers through the proper metering equipment to enable the operating forces to put into service or remove any equipment without impairing the operation of the remainder of the station. With stations supplying transmission systems the scheme of connections depends largely on the relative capacities of the individual generators. Sketch (f) shows a more modern design for central stations with the feeder reactors next to the bus structure. so that the combination becomes the equivalent of a highvoltage generator with all the switching on the highvoltage side of the transformer. following an outage for repairs or maintenance. in contrast with (e) where the reactors are on the feeder side of the breaker. transformer and transmission line can be dropped without loss of customer’s load. Stray animals have caused considerable trouble by electrocuting themselves in accessible bus structures. The bustie breakers can be used when one or more generators are out of service to prevent voltage and phaseangle differences between bus sections that would exist with the supply to a bus section through a reactor. ers. 9Fundamental schemes of supply at higher than generated voltage. This arrangement is possible because of the proven reliability of reactors.
a fault on any section of circuit may also be cleared without loss of (a) (b) load. No. Sketch (b) shows the loopedin method of connection. perhaps. and (c) shows the other extreme where the feeders and lines are expected to be in service continuously. 2. loFundamental schemes of transmission.E. 13. (c) Bussed supply. and the type of load. Vol. Sketch (d) shows an arrangement which is frequently applicable and which provides a considerable flexibility Jvith the fewest breakers. EEI Publication No. which is permissible only where feeders and lines can be taken in and out of service at will.) On the other hand. Furthermore. Sketch (c) is in effect an extension of the buses from station to station. and as in (a) any section of the circuit can be removed from service without reducini power output. but where the capacity of the transmission lines is such as to give an economical transformer size.S. simplification and reduction in the cost of switching is Pierce and E. E. by R. Briefing the Record. Sketch (b) shows the case where conditions do not permit of the transformers being associated directly with the generators because. A. February 1948. Economic Balance Between Steam and Hydro Capacity. of outgoing feeders at gencrater voltage.. 10891094. (b) but can be justified only where a temporary outage of the transmission is unimportant. (a) Fullysectionalized supply. dated May 1949. Sketch (b) shows the extreme of simplicity.M. the importance of which depends on two factors. Any section of either transmission circuit can be taken out for maintenance without the loss of generating capacity. E. 1932. (b) Loopedin supply. however. Some industrial loads are of such a nature that the relatively small risk of an outage does not justify duplication of buses and switching. Relaying Only a few fundamental ideas have been presented on is somewhat more difficult than with (a). but not unduly the possible layout of station buses and the switching so.E. EEINEMA Preferred Voltage Ratings for AC Systems and tion as regards insulation and spacing. 8 and 9 include fundamental layouts from which almost any combination can be made to meet local conditions. If. considerably on local conditions and the expenditure considered percheaper than that in (a) and slightly less than that in missible for the conditions prevailing. 117. Jaklitsch. George. the multiplicity of sources of supply. or all of the generating capacity may be lost. Chap. Sketch (a) is a fullysectionalized scheme giving the ultimate in flexibility and reliability. Figure 10 shows an assumed transmission of 100 miles with two intermediate stations at 33 miles from either end. a second line trips out. sectionalized scheme. 1945. Most line outages originate from lightning and a 3. pp. if a line is of poor construc4. steps are required to clear a fault. Figs. Vol. part putting in an elaborate switching and relaying scheme. of course. Flexibility on the lowvoltage side is retained as in arrangements of transmission circuits. The possible combinations are almost infinite in number and will depend (a). J. p. itself. Economics of LongDistance Energy Transmission. Mechanical Engineerequalize the currents so that several distinct relaying ing. The same argument applies to the transmission line itself. Fewer breakers are required than for the fully Fig. R6. Relaying in (c) is REFERENCES complicated by the fact that ties between buses tend to 1. TransA proper balance must be kept between the reliability actions A. good engineering to attempt to compensate for this by . 67. 55. 3. (See lions.E. Ebasco Services. NEMO Publication No.s of Transmission through a station transformer connected directly to the generator terminals. The scheme is.I. Here it may be desirable to include the transformer bank as an integral part of the line and perform all switching operations on the lowvoltage side. Transacpermissible if the circuit is built lightning proof. The choice depends on the requirements of service continuity.Chapter 1 General Consideration. except within that part of the transmission where one section is temporarily out of service. an independent supply being provided for the initial startup and for subsequent emergency restarts. Also Electrical War& August of the switching scheme used and the design of the line 30. Inc. it would not be Equipment.. 147. edited by J.
all of the same type. uses a positivenot appreciated until several years later when the papers sequence filter and a zerosequence filter for the detection by Evans. and has been used applications. a recent addition unbalances. D. are all equal. Out of the concept of symmetrical components have plication to system fault calculations and stability calculations focused the attention of the industry on the simplifisprung. Hobson D.. taken phase solution. was in terms of symmetrical components. three phase systems are most frequently encountered. three unsymthe fact that it permits an unbalanced load or fault to be metrical and unbalanced vectors of voltage or current in represented by an impedance in shunt with the singlea threephase system) can be resolved into three sets of phase representation of the balanced system. Although metrical. L. an unbalanced group of n HE analysis of a threephase circuit in which phase voltages and currents are balanced (of equal mag. J. Thus three vectors of more difficult. this discussion will be generally adopted for calculating such circuits.one set are symmetrically located if t. or non. anced faults or unbalanced loading.associated vectors. Peters. and others in fault protection grew from an understanding of the symdeveloping the sequence netnorks and extending the ap. Each set ance. “Method of Symmetrical Coreadily and accurately. Fortescue in his 1918 paper.been greatly augmented and rapidly developed since its introduction. is relatively simple since to each other. particularly the effect of connections and the pheis a “symmetrical component” of the original unbalanced nomena associated with threephase coreform units has vectors. and stability limits on section. 12 . Symmetrical components is the method now analysis of any multiphase system. however. regulators to respond to voltage changes in all three phases The method was recognized immediately by a few engi.I. one of the longest ever presented before the during unbalanced shortcircuit or unbalanced load conditions.E. and in which all circuit elements in each phase are of equal length and symmetrically located \Gth respect are balanced and symmetrical. The three vectors of each set are of equal magniThe understanding of threephase transformer performtude and spaced either zero or 120 degrees apart. since presented to the engineering profession by Dr. unbalanced loads. and the simplification made possible by the to the list of devices originating in minds trained to think use of symmetrical components in such calculations. It was limited to a consideration of threephase systems. and the connection of inneers as being very useful for the analysis of unbalanced strument transformers to segregate zerosequence quanticonditions on symmetrical machines. Wagner. This method of analysis makes possible the prediction. Its more general ties for the prompt detection of ground faults are interestapplication to the calculation of power system faults and ing examples. The cation and clarification symmetrical components offered in negativesequence relay for the detection of system faults. The same concept of resolution can be applied to been clarified by symmetrical components.metrical component methods. The n vectors of each set nitude in the three phases and displaced 120” from each other). but the subsequent work of R.rather than in one phase alone. Wagner. The use’ of faults within a protected line section and for initiating of symmetrical components in the calculation of unbalthe high speed tripping of breakers to isolate the faulted anced faults. Charles L.metrically located if the angles between the vectors. as the result of unbalanced loads.the positivesequence filter for causing generator voltage balanced conditions. Modern concepts of protective relaying and mained the tool of the specialist. unbalanced faults or shortcircuits that are not symmetrical in the the method of symmetrical components is applicable to the three phases. of the behavior of a power system ordinates Applied to the Solution of Polyphase Networks.” This paper.he angle between adjacent vectors is either zero or 120 degrees. The analysis by Kirchoff’s laws is much in sequential order. F. E. is that an unbalanced analyses. extensively since that time in power system stability The fundamental principle of symmetrical components. C.E. threephase power systems now overshadows the other Symmetrical components as a tool in stability calculations was recognized in 19241926. Evans. such as voltages or currents. vectors. The HCB pilot wire relay. many electrical devices. F. when the circuit is not sym. Its value for such calculations lies principally in group of three related vectors (for example. as have been rotating vectors. almost fullborn. can be resolved into n sets of balanced vectors. Whitehead T Stated in more general terms. A set of vectors is considered to be symthe treatment of a singlephase leads directly to the three. as applied to threephase circuits.the physical concepts and the mathematical analysis of rotating machine performance under conditions of unbalrotating vector operators such as impedances or admittances. and others were published.CHAPTER 2 SYMMETRICAL Original Author: COMPONENTS Revised by: J. the calculation of power system performance under un. For several years symmetrical components re. The engineer’s knowledge of such phenomena has A. is now recognized as a classic in engineering literature.
a2. etc. since each sequence network may be set up independently as a singlephase system. but rotated 120 degrees forward from the vector v. For example. protection. has been furthered by the development of symmetrical components. form a balanced. displaced equal angles from each other. V’=aV is a vector having the same length as the vector V. a2. since the vectors are of equal length. and a (taken in this order) v:ov Fig. This is unfortunate since the mathematical manipulations attendant with the method are quite simple. and cross the reference line in the order 1. but also.+j+~oo (~+a) (l+a2)=l+jO=@ (1a) (la2)=3+jO=3d0 l+a Ifa2=a= la = la* (l+a)Z=a= ~+j.or for the determination of shortcircuit.jP40 2 2 . A vector operated upon by a is not changed in magnitude but is simply rotated in Position 120 degrees in the forward direction. but also. The square of the vector a is another unit vector oriented 120 degrees in a negative (clockwise) direction from the reference axis. 1. This relationship is shown in Fig. The method of symmetrical components is responsible for an entirely different manner of approach to predicting and analyzing powersystem performance. 1+a2= ~. by providing new and simpler concepts the understanding of power system behavior has been clarified.. THE VECTOR OPERATOR “a” For convenience in notation and manipulation a vector operator is introduced. relays. or oriented 240 degrees forward in a positive direction.Chapter 2 Symmetrical Components a2= (~j1?O)(cj120)=~j240= i j"". as pointed out above. In this connnection it is of interest to note that the network calculator has become an indispensable tool in the analysis of power system performance and in power syst. currents and voltages and for the application of circuit breakers. It stands somewhat unique among mathematical tools in that it has been used not only to explain existing conditions. Sot only has the method been an exceedingly valuable tool in system analyses. set of vectors of positivephasesequence rotation. L4s shown in Fig. grounding transformers. the resultant of a2 operating on a vector V is the vector Y” having the same length as I. requiring only a knowledge of complex vector notation. the physical concepts arising from a knowledge of the basic principles have led to the development of new equipment and new schemes for power system operation. 13 (2) The extensive use of the network calculat. Symmetrical components early earned a reputation of being complex. and so it is with this important tool.i300 (l+a2)Z=a2= !j~=. A miniature network of an extensive power system. protector tubes. etc. and a (following the usual convention of counterclockwise rotation for the TABLE ~PROPERTIES OF THE VECTOROPERATOR“~" l=l+joEe'O : : r I. but located 120 degrees in a clockwise direction from V. 1. Things men come to know lose their mystery. symmetrical. set up with threephase voltages. lRotation of a vector by the operator 0.?!$~~i120 $+jG2=$*0 a=~j+=. Inasmuch as the theory and applications of symmetrical components are fully discussed elsewhere (see references) the intention here is only to summarize the important equations and to provide a convenient reference for those who are already somewhat familiar with the subject. The three vectors l+jO.em design. and mutual impedances between phases would indeed be so large and cumbersome to handle as to be prohibitive. Through usage it has come to be known as the vector a and is defined as (1) This indicates that the vector a has unit length and is oriented 120 degrees in a positive (counterclockwise) direction from the reference axis. separate impedances for each phase.
and it is understood that EO..O Ed. E. lvritten as Eaz.l is 120 degrees ahead of EM and 120 degrees behind E. without writing Eaa. Ebl. and negativesequence components of E.. t Eb. 4. so that E. is equal to aE. since the vectors do not cross the reference line in the order named.‘a’E.=aE. are the zero.= E. 5. written as Eal.El. for example.:. EaI. .. the three positivesequence vectors E..+Eb+Ec) El = 3 (E. 3.l. E. RESOLUTION AND COMBINATION VECTOR COMPONENTS OF ES is the negativesequence component of E. Similarly the three negativesequence vectors are defined by ES.. EC1 metrical threephase set of vectors of positive phase sequence since the vector E. 2. Ebl =a2E1.14 Sgmmetricul Components Chapter vector diagram).~. set of vectors of negativephasesequence. Fig. Eb?. + UEb+ UZE. = E. and a2Ea2.. are. El. since an angle and a magnitude are necessary to define each vector. All three of the zerosequencecomponent vectors are defined by E. The positivesequence component of E. and E?. symEcl.+a2E.. Thus all nine component vectors of the three original unbalanced vectors are completely defined by E. and is likewise the zerosequence component of Eb and E... Resolution of Unbalanced ThreePhase Voltages A threephase set of unbalanced voltagevectors is shown in Fig. and E. El is the positivesequence component of E. 3. keeping the same Fig. 17. El. The three original unbalanced vectors possess six degrees of freedom.? taken in order form a symmetrical set of negativesequence vectors as in Fig. is equal to a2E. Fig. Ez=+(E. Any three unbalanced vectors such as those in Fig. Fig.l. but the third named follows the first. are defined by the angular position and magnitude of E. convention of counterclockwise rotation. 5Positivesequence components of the vectors in Fig. and Ecl. a. II. etc. The positivesequence component of Ek. since each of the three sets of component vectors is described by one angle and one magnitude. The nine component vectors also possess six degrees of freedom.l.etc. 4Zerosequence components Eoo’Ebo’Eco’Eo of the vectors in Fig.. and a2 (taken in this order) form a balanced. The vectors 1. positivesequence vectors are defined by El..) 3=. 3 can be resolved into three balanced or symmetrical sets of vectors by the use of the following equations: El= w. 6Negativesequence EC2 =02Ep components of the vectors in Fig. respectively..l. since Eao= EbO=Ed. aE. as shown in Fig. the three Ebe’a% 1. 3Unbalanced vectors. form a balanced.. Fundamental properties of the vector a are given in Table 1.. symmetrical. Likewise. is the zerosequence component of E. Ebl. 3.. and Ecl=aEl. and are shown on the vector diagram of Fig. 6.so that Eav. positive. The negativesequence components of Eb and E.+aE. since Eal = El. 2Properties of the vector operator u. This set of threephase vectors is = shown in Fig.) (3) Fig.
. since it presupposes no zerosequence component.+E. Referring to Fig. The negativesequence set of vectors cloes not rotate in a direction opposite to the positivesequence set. or only a negativesequence component. component vectors in Fig.=E. the order in which the maximum occur with time..00 356 I 2@300 4 ’ I 0. c. 9Threephase line currents. sequence components. The above are. 0. that is. and l&D=+ (Eab+Ebc+EcJ =O...=E.00 0.+E. phasesequence of the positivesequence set. but the phasesequence. 7Combination of the three symmetrical sets of vectors to obtain the original unbalanced Fig. The angle.84 0. 2. The subscript “D” is used to denote components of delta voltages or currents flowing in delta windings. SDetermination of unbalance factor. threewire systems. Eh=Et~+Ebl+Eb~=Eo+a*E~+aE~ Ec=Eti+EE. of the negativesequence set is a. if the vectors are balanced and symmetricalof equal length and displaced 120 degrees from each otherthere will be only a positivesequence component. there will be no zerosequence component for a set of threephase linetoline voltages. a. 8. As an example the chart can be used to determine the unbalance in phase voltages permissible on induction motors without excessive heating. positive. The chart is applicable only to threephase. 3. respectively. 3 will have zero.to positivesequence amplitudes and the phase angle between them.?= Eo+aEl+a2Ez (4) The combination of the sequence component vectors to form the original unbalanced vectors is shown in Fig. depending upon the order of phase sequence for the original vectors. b. This limit has usually been expressed as a permissible negative sequence voltage whereas the phase voltages are of course more readily measured. however. Equations (3) can be used to resolve either linetoneut’ral voltages or linetoline voltages into their components.8OL 0. . b.20 R Fig. and therefore opposite to the a. The unbalanced vectors can be expressed as functions of the three components just defined: E.12 I I 1.04 I I 1. by which En2 leads Eal can be obtained also from the same chart. positive. and negative 0. Resolution of Unbalanced ThreePhase Currents Three line currents can be resolved into three sets of symmetrical component vectors in a manner analogous to that just given for the resolution of voltages.Chapter 2 Symmetrical Components 15 Note that all three sets of component vectors have the same counterclockwise direction of rotation as was assumed for the original unbalanced vectors. since three delta or linetoline voltages must form a closed triangle. The only data needed to use the chart is the scalar magnitudes of the three line voltages. and negativesequence components of I. In many cases it is desirable to know the ratio of the negatives. 7.88 I I 0. the current in the reference phase. c.+E. This ratio is commonly called the unbalance factor and can be conveniently obtained from the chart given in Fig.l+E.96 . the zero.Eb % I I 1. a. 9: b  Ib C AL Fig. In general a set of three unbalanced vectors such as those in Fig. Hanever.+E. Inherently.
. resolved into symmetrical components.. =O. then anced load. and negativesequence current flowing through the un3. the referencephase through the proper angle. the impedances components of voltage.=Zb =Z. is detailed in a following section.reve&l.g) =I”z.+aE. cannot (b) have a zerosequence component since 1..+a*E. Mutual Impedances between phases can also be reas are threephase voltages and currents. and Z.g+a2Et. although the El. the sequences are independent. ciple that there is mutual coupling between sequences and throughout the analysis. If the biguity possible where stardelta transformations of voltage and current are involved. has been chosen as the reference phase current. and not rotating vectors drops across the impedances in the three phases. as reference. and E..Z. or power are symmetrical.+I. Resolution of Unbalanced Impedances and Ad..+112.=4(E. ~=&+aI. or power for the other phases are easily found by El = 112. as shown in Fig. zc C’ Cor into a deltaconnected transformer winding. I.=~(Zn+a*Zb+aZ. where Eo. both positive. which form an unbal. Consider the three starimpedances of Fig. and Ez are components of E.g+Ebg+EcY) =1oz.+I.+1b+Z.solved into components.) =IoZ.) Z.Z. 11(b). It is customary in symmetrical component notation to denote the reference phase as when the circuit constants are not symmetrical.= IoZO pedance. but once selected. Consider .) Zmo= 4(2mb. The voltages and currents over an entire sys. = 2. and the components of voltage. Their sequence components are:  EC z. must obviously be equal to zero. this phase must be kept as the The above equations illustrate the fundamental prinreference for voltages and currents throughout the system.z. 11(a).balanced load.Zmbcof Fig.= 2.=0 but EO=IIZZ+12Z1 so that there is a zerosequence voltage. 11(a). created by positiveages and currents. If Z. current.+I?z. Another way of stating this fact is that zerosequence current cannot flow into a delta.+aEb. into a starconnected load cannot have a zerosequence component unless the neutral mire is returned or the neutral The sequence components of current through the impoint is connected to ground. 11Three unbalanced self impedances. Eb. as expected. or where the components of neutral point is not grounded in Fig.=3(E. (9) The choice of which phase to use as reference is entirely E. all current (as well as zerosequence current) create a zeroreferred to the components of the reference phase. arbitrary.pedances. Likewise the currents flowing Fig. neutral wire.dition. loThreephase delta currents.+I~Z. For this confound by analysis are directly the components of the refer. The amand. mittances Equations (8) and (9) also hold for unsymmetrical Self ImpedancesUnbalanced impedances can be series line impedances. 10.+12z. imE.and negativesequence tern are then expressed in terms of their components. equations . 7iY!iL3 Fig. 2. Fig.+Zrm+Zmnb) (8) (11) . current. and the sequence components of the line voltconnected load or transformer bank.. = &z*+aZb+a*Z. ence phase.ta21e Three line currents flowing into a deltaconnected load. As the (‘phase a”. the voltage impedances are vector operators..=g(&+zb+zc> 2. can be resolved into components : (7) Where 1. El=~(E. The sequence voltage drop.16 Symmetrical Components a Ea9 b Eb9 Chapter 2 I a=I()+1 42 Three delta currents. impedance. nor can zerosequence ages impressed across them are interrelated by the folcurrent flow into a starconnected load or transformer bank lowing equations: unless the neutral is grounded or connected to a return E. star voltages and currents are to be related to delta voltrepresenting a neutral voltage shift. 12(a). (10) rotating the positiveor negativesequence components of Ex = I.
unless Z.1_yb z... etc. In Fig. (t)) IJnbalnnced self and mutual impedances.zZ. IC ZC a Yb=l Zb’ zmbc J Ye=+.. no zerosequence current can flow from the line to the load. Admittances can be resolved into symmetrical components.. J c’ c e then Y.. or starconnected loads. in all three phases.) =ro(z.+EzY.IOZ. both self and mutual.. 13Stardelta impedance (b) conversions. 4..+Ebb. Z1..... is not the reciprocal of Z. and Z. 13(a). if both self and mutual impedances are symmetrical.=$(Y. as in Fig. that Y. are used more frequently than Z. however. 12(b). is given by . 12 Cn) Three unbnlancecl mutual impedances.) Note. the equivalent star load must be left with neutral ungrounded.(Z.. Note. Fortunately.)+I.I&T.. the symmetrical components of the three voltage drops across the section are: Eo= W.Y.b+Zb.+2~... etc. the impedance to positivesequence currents...b+Zbc+Zcn zbcxzcca Z.. = II( 2. and the components used to find the sequence components of the currents through a threephase set of line impedances. 11(a). that this is not strictly correct and that 21. respectively.) (13) .) =lo(Z.) = .~+aEbb*+a*E. Z1. in other words.)=I.~+Ebb’+Ecc’) =2IoZ..+‘2Z.I&o+21*2..+azEbg+aE.> =IoZ. For this condition positivesequence currents produce only a positivesequence voltage drop. L and Z2 the shorter expression “zerosequence impedance” is usually used to refer to Z.*)+~2(~~~. 13(b). Z1.Z. the impedance 10 zero. ~.)+I~(Z*+2~. is not the reciprocal of Z. except for &symmetrical loads. Since Zo. the components of admittance are not reciprocals of the corresponding components of impedance unless the three impedances (and admittances) under consideration are equal.1 . let Ya=$.*) Es= f(E.+EzY. Y.) (1% (b) Fig.+a*Yb+aY.. unsymmetrical transformer connections. = +(E. E.I*Z. the following equations are applicable..~. Where Z.+u*Ecc’)= I~z.z. = ~(E. When the delta impedances form a threephase load.z. and Z2 are. Iz=EoYz+ElYl+EzYo UG) The components of the threephase line currents and the components of the threephase voltage drops created I)y the mutual impedances will be interrelated by the following equations: E.. positive. as in Fig... and positivesequence currents produce only posi zbc (a) Fig. positivesequence. both self and mutual impedances are present in a section of a threephase circuit. For a circuit that has only symmetrical impedances. Eo= Io(Z. and negativesequence impedances. (17) Again. and Y... 8. is to be converted to an equivalent star shown in Fig.+a*E.=+(Y. ..~)+I1(zo~. to the equivalent delta Fig. should not be confused with 21.1 yc _ zabxzbc Z.Z. 11=&Y. the threephase systems usually encountered are symmetrical (or balanced) and the sequences are independent.=Zb =Z.Z.+Z. (14) E. the sequences are independent of each other.. etc.... Z.. 13(b).+E. StarDelta Conversion Equations If a delta arrangement of impedances..+aE.Y.. as functions of the symmetrical components of the voltage drops across the impedances.=I. from the star impedances of Fig.oI1Z. and negativesequence. the positive sequence component of self impedances.(z.= E.+Ecc. and I.) = II& E.= :(E.+EIY..=+(Ya+Yb+Yc) Y.)+12(z.. is not the reciprocal of Z. 13(a).. however.)+~*(~.+aYb+u*YJ Y. and 22 are commonly referred to as the zerosequence..~) =Io(z. hence.~ ~l=~(E. The reverse transformation.~b...z+211Z.+az.* (12) E.Chapter 2 Symmetrical Components 17 tivesequence voltage drops..+UE. as defined in Eq. If.z. rather than 2.
. and E. since xerosequence current can flow from the line to the star load and return in the ground. = (aa?)E? END=E.m a2a =E~D = la2 TEln E”D E. a21 E. =z&30 1 a= = TE2D Fig. =73ci160 = al 3E.. vectors. EOD=O under all conditions.. The choice is arbitrary. are a positivesequence set of linetoneutral vectors in Fig.'z.ivcSequence LinetoLine Voltage As a Funct.b = Z.. EPD E? =xoj15~ CB AC = al END 3 a2 .D E.. The relationship between the two sets of threephase vectors is shown in Fig..D=E. as illustrated in Table 3. TABLE 3 NegativeSequence LinetoLine Voltage As a Function of Negative Sequence LinetoSeutral Voltage END= Eab = d/:jE2c~= = (1 .\/zE2d150= (a1)E2 END= E. The equations expressing EID as a function of El. 14(a).. 14(b). vz““0 = END 3 aa2 E2 =J~$ = 3Em E2D E..t’30=(1aa’)E.. and Eo is an indeterminate function of E. Again. =@.a)E: E?D = Ebe =j&?E.b=~/:j~. = t/3E2tiao = (1 a’) E. The linetoline voltages will also form a positivesequence set of c4 BA CB AC If E. RELATIONSHIP BETWEEN SEQUENCE COMPONENTS OF LINETOLINE AND LINETONEUTRAL VOLTAGES Assume that E:. as functions of ElD and E2D..=zi=G3E2~ la Certain authors have arbitrarily adopted phase CB as reference.b = jd\/3E. =j$ = ~EID E El =D.1 J&J E. and lineto BC CA BA (b) Positivesequence relationships.D = Eh = jdZE. form a negativesequence set of vectors.A (cl Since the linetoline voltages cannot have a zerosequence component. (c) Negativesequence relationships.. Although END(the positivesequence component of the linetoline voltages) will be numerically equal to ~~E1E. respectively. is the positivesequence component of the linetoneutral voltages (which is equal in this case to E. = j$ El =$.I)..+&+z$ z.SZbf y%... the negativesequence component of the linetoline voltages.D. E.150=s.. for various linetoline phases selected as reference.D E E.ionof Positive Sequence LinetoScutral Voltage E. Using this convention: . = ~~E~ri’~O= (azl)E? END=Eba= . and E. Reference Phase AB BC c4 BA CB AC (a) t y Ebc (bl :::. Table 2 gives the relation between E1~ and E. E. TABLE 4 Reference Phase AB E la E. the vector diagram of Fig.18 Symmetrical Components Z. can be solved to express El and E. but cannot flow from the line to any delta arrangement. and E2D. 14(c) illustrates the relation between E2= E..=~SEI~~30=(1a)E. z..). EbR. zcz. threephase load with neutral grounded ca.d160 = (a . the nlgebraic relation expressing E2D as a function of E2 will depend upon the linetoline phase selected for reference. EID=E.. = 1/3E.D=E.=zc+z. 14Relationships neutral between linetoline components of voltage.D 3 d/3 E aa2 E. = (a2u)El EID = E.unot be found.b=jy’~El=(a~2)E~ E. since the relations between the linetoline and linetoneutral components are easily remembered and the angular shift of 90 degrees is easy to carry in computations. and ENDas a function of Ez.??. HI.130 = E.+~ An equivalent delta for a starconnected. Eba... the angular relationship between El and E1~ depends upon the linetoline voltage taken as reference.zb 0 TABLE 2 Chapter 2 (18) I3 ReferencePhase LinetoLine Voltages AB BC Posit.D=Ebn=~/3Ele1’50=(a?l)E.D 3 43 E. Refer to Table 4 for the relationships. = (a* a)Ez END= E.
SEQUENCE COMPONENTS OF LINE DELTA CURRENTS AND ZZ 1. is taken as reference for the line currents. CC) Negativesequence relationships. To illustrate the sequence transformations. IV. to use in analysis.=$ V. and the relation esisting for the negativesequence components of the currents are given in Figs. 16(b). but that the angular shift for negativesequence is opposite to that for positivesequence. =j&E. 15Relationships (b) Positivesequence relationships. also. If the transformer windings are symmetrical in the three phases. the relation are easily remembered. there will be no interaction between sequences. regardless of the delta phase selected for reference.0 degrees in the transformation for negativesequence voltage and current. Refer to Table 5..) is taken as reference. both positivesequence current and voltage will be shifted in the same direction by the same angle. of phase and Fig. and each sequence component of voltage or current is transformed independently. 16(a). I.j&?Ez EoD=O El= j$ Symmetrical Components 19 upon the phase selected for reference. In Fig. Also.  The relation existing between the positivesequence component of the delta currents and the positivesequence component of the line currents flowing into a delta load or deltaconnected transformer winding. ZZ L 1d I. (4. a connection of power or regulating transformers giving a shift of 0 degrees in the transformation for positivesequence voltage and current will give a shift of .Chapter 2 El. 15(b) and 15(c). STARDELTA TRANSFORMATIONS VOLTAGE AND CURRENT (20) OF IO 0‘f IC c az \ 1. Delta Reference Currerl t . From the vector diagrams El’ = NElej30 (b) 1 Iz’ = 12cj30 N between components delta currents. the winding ratio is 12and the overall transformation ratio is N = $. the j operator is convenien. (cl. Linetoline or linetoneutral b (0) voltages on the delta side will be N times the corresponding voltages on the star side of the transformer (neglecting impedance drop). Although the components of line currents are dz times the delta phase selected for reference. Ib Each sequence component of voltage and current must be followed separately through the transformer.E2D (19) E2 =‘72 Eo is not a function of EO~ The equations and vector diagrams illustrate the interesting fact that the numerical relation between the linetoline and linetoneutral positivesequence components is the same as for negativesequence. and the angular shift of the sequence will depend upon the input and output phases arbitrarily selected for reference.. and (e) give the relationships for the three phases with each component of voltage and current considered separately. the angular relationship depends If the current (1. phases a and a’ have been selected as reference phases in the two circuits. Regardless of the phases selected for reference. END= . Negativesequence current and voltage will also be shifted the same angle in . Figs. I.
and II. etc. one direction. Only positivesequence power is developed by the generators.= I. each being determined by conditions in its respective circuit. VII. and the sequence impedance characteristics. In (h) Relationship of positivesequence linetoneutral and linetoline voltages. .20 Symmetrical Components Chapter 2 E’cg . where 190 the angle between E” and 10. (c) Relationship of positivesequence currents. 16Transformation rent and voltage of the sequence components of curin a stardelta transformer bank.or negativesequence quantities are transformed. (d) Relationship of negativesequence linetoneutral and linetoline voltages. (e) Relationship of negativesequence currents. but the power in one phase of an unbalanced circuit is not onethird of the above expression. unbalanced load. ‘TO.I. I. or other dissymmetry in the circuit. and a current will circu Fig. I. cos e. This power “between sequences” is generated in one phase and absorbed by the others. and I!.= Ib = I. The equation shows that the total power is the sum of the three components of power. VI. for common connections of power and regulating transformers are given in Chap.= l.. The unbalanced fault.= TO<. there will be no zerosequence component of I. E.= I.. This power is converted to negativesequence and zerosequence power by circuit dissymmetry such as occurs from a single linetoground or a linetoline fault. The transformation characteristics for the three sequence currents and voltages.I. and does not appear in the expression for total threephase power. and the negativesequence angular shift will be eq~lal to the positivesequence shift but in the opposite direction. The action of a transformer hank in the transformation of zerosequence currents must be given particular attention. and from the delta side the bank acts as an open circuit to zerosequence. cos 81+E. P=3(E.‘. since each phase will contain components of polrer resulting from zerosequence voltage and positivesequence current. thus acts as the “generator” for negativesequence and zerosequence power. and still others transform zerosequence quantities in the same manner as positive. may have a zerosequence component. this is a general rule for all connections of power and regulating transformers.I.hc bank without transformation. and l?. & the angle between E. = I1 The zerosequence linetoneutral voltages.’ are entirely independent. some consideration should be given to conjugates of the symmetricalcomponent sets of vectors. A system of positivesequence vectors are drawn in Fig. Since zerosequence current cannot flow from the delta winding. CONJUGATE SETS OF VECTORS Since power in an alternatingcurrent circuit is defined as EP (the vector E times the conjugate of the vector I). From the star side the transformer bank acts as a return path for zerosequence current (if the neutral is grounded). For zerosequence current alone. 444 late around the delta such that I. others permit it to pass through t.) (22> (b) Ebc (d) (9) Fig. If the star winding is grounded. As previously stated. 17Conjugates of a positivesequence set of vectors. wherever phase shift is involved in the transformation. e1 the angle is between E. cos 8. THREEPHASE POWER The total threephase power of a circuit can be expressed in terms of the symmetrical components of the line currents and the symmetrical components of the linctoneutral voltages.+E. 5. since certain connections do not permit zerosequence current to flow. 17(a).
are included in the zerosequence network “as viewed from the fault. 10. 19Conjugates of a zerosequence set of vectors.site direction to the given vector. the increment alone is the fault current total.). accordance with the definition that the conjugate of a given vector is a vector of the same magnitude but displaced the same angle from the reference axis in the oppo. Another approach is to imagine in each network a voltage impressed across the terminals of the network. The entire sequence network can often be reduced by simple manipulation to a single voltage and a single impedance. The voltages appearing in the negative. which is an equivalent network for the balanced power system Imder an imagined operating condition such that only one sequence component of voltages and currents is present in t. t ivesequence set. ~tc. equal mutual impedances between phases. The type of unbalance or dissymmetry in the circuit can be represented by an interconnection between the equivalent sequence networks. dealing with each sequence separately. As shown above for the case of balanced loads (and it can be readily shown in general) currents of one sequence will create voltage drops of that sequence only. General Considerations One of the most useful concepts arising from symmctricnl components is that of the sequence network. rotating machines symmetrical in all three phases. and Rotating Machinery The impedance of any unit of the systemsuch as a generator.and zerosequence networks will be generated by the unbalance. Similarly. Furthermore. Only parts of the system over which zerosequence current will flow.” The two terminals for each network correspond to the two points in the threephase system on either side of the unbalance.work. Transformers. since currents and voltages of only one sequence are present.” by imagining current of the particular sequence to be circulated through the network from the fault point. It is particularly necessary when setting up the zerosequence network to start at the fault point. or a section of lineto be in t (b) \ ~oo=$. Setting Up the Sequence Networks The equivalent circuits for each sequence are set up “as viewed from the fault. a transformer. Sequence Impedances of Lines. the remainder of the power system remains balanced and an equivalent sequence network can be ob . VIII. the threephase system can be represented by an equivalent singlephase diagram. faults. and will appear as voltages impressed on the networks at the point of fault. the normal currents in any branch must be added to the calculated fault current in the same branch to get the total current in any branch after the fault occurs. 18. as the result of a zerosequence voltage impressed at the unbalanced point.hc system. one terminal of each network will be the fault point in the threephase system. The positivesequence network is the only one of the three that will contain generated voltages. For a series unbalance. the positivesequence network represents the system operating under normal balanced conditions. of the system. as in Fig.nbe assumed to generate only positivesequence voltages. if :L power system is balanced (equal series impedances in all three phases. IS(b). Section 11). The conjugate of a zerosequence set of \xxtors is another zerosequence set of vectors. 7. unbalanced open circuits.o=~co Fig. the conjugates of the positivesequence set of vectors are shown in Fig. such as an open conductor. t. For the case of shunt faults between conductors and ground. of vectors. unbalanced load. the conjugates to a negativesequence set of vectors form a posiyo=Ibo= I. Since the fault current equals zero before the fault. Sote that the conj rlgntes to a positivesequence set of vectors form a negativesequence set. (al tained for the balanced part. The advantage of the sequence network is that. Even under such emergency unbalanced conditions.Chapter 2 A Symmetrical Components 21 Ibz Fig. which usually occur at only one point in the system. Kearly all power systems can be assumed to be balanced except for emergency conditions such as shortcircuits. or unsymmetrical conditions arising in rotating machines. the other terminal will be ground or neutral at that point..he two terminals will correspond to the two points in the threephase system immediately adjacent to the unbalance. since alternators c%. There will be no interaction between sequences and the sequences are independent. However. l&conjugates of a negativesequence set of vectors. SEQUENCE NETWORKS 5. since zerosequence currents might not flow over the entire system. or point of unbalance. investigating the path of current flow and the impedance of each section of the network to currents of that sequence. 6. For shortcircuit) studies the internal voltages are shorted and the positive sequence netxork is driven by the voltage appearing at the fault before the fault occurred according to the theory of Superposition and the Compensation Theorems (see Chapter 10. all hanks of transformers symmetrical in all three phases. and to follow the path of current flow through the net. This gives exactly the increments or changes in system quantities over the system. see Fig.
These points will be clarified by detailed consideration of a specific example at the end of this chapter.I222 E. and writing the equation for the voltage drop. 20(a) is drawn for a three Nj@pF$J EQUIVALENT SYSTEM IN L%~. and zerosequence between neutral and the point in question are 21. Regardless of the type of fault. The impedance to negativesequence currents for all static nonrotating apparatus will be equal to the impedance for positivesequence currents. (a) Threephase short circuit on generator. negative. (23) gives the sequence voltages at the fault.I tfb (4) VECTOR DIAGRAM SHOWING VOLTAGES AND CURRENTS DURING FAULT (3) SHORTHAND REPRESENTATiON OF POSITIVE. since the zerosequence currents flowing in the three phases. if the impedances to positive. it is convenient to find distribution factors for each sequence current by circulating unit sequence current in the terminals of each network. so that it is to he expected that the sequence networks will have mutual coupling. NEGATIVE. After the currents flowing in each network have been determined.l is the generated positivesequence voltage. an equivalent impedance equal to three times the ohmic neutral impedance will appear in the zerosequence network. the positivesequence network being the only one of the three having a generated voltage between neutral and the point for which voltages are to be found. This convention of assumed current flow must be carefully followed to avoid ambiguity or error even though some of the currents are negative. any unbalance or dissymmetry in the system will result in mutual action between the sequences. = I&.& E. Equations can be written for the conditions existing at the point of unbalance that show the coupling or connections necessarily existing between the sequence networks at that point. 20(a) through 20(e) show such an equivalent system with the more common types of faults applied. 22. 3 of this chapter. = . 8. respectively. then Eq. since the threephase currents of either sequence add to zero at the neutral. Assumed Direction of Current Flow Ry convention. the sequence voltages at the point will be distribute through each network in accordance with the distribution factors found for unit current. the positive direction of current flow in each sequence network is taken as being outward at the faulted or unbalanced point. AND ZEROSEQUENCE NETWORKS (a) Fig. if Z.. letting it flow through the network and finding how this current distributes in various branches. This in effect means that the system is reduced to a single generator with a fault applied at its terminals. IX. however. An impedance in the neutral will not appear in either the positivesequence network or the negativesequence netnork. 5. . 10 add directly to give a neutral current of 310. CONNECTIONS BETWEEN SEQUENCE NETWORKS THE As discussed in Part II.22 Symmetrical Components Chapter 2 scrted in a sequence network is obtained by imagining unit current of that sequence to be circulated through the apparatus or line in all three phases. and 6. or by actually measuring the voltage drop when crwrent of the one sequence being investigated is circulated through the three phases of the apparatus. 20. For example Fig. between them at the point of unbalance. For example. (23) nhere E. = E:. it is usually sufficiently accurate to reduce a given system to an equivalent source and single reactance to the point of fault. The sequence impedance characteristics of the component parts of a power system have been investigated in detail and are discussed in Chaps. thus the sequence currents are assumed to flow in the same direction in all three sequence networks. and &. The impedance to zerosequence currents for all apparatus will in general be different from either the impeclunce to positivesequence or the impedance to negativesequence. 5. or possibly direct connections. and the magnitude of sequence current at the fault. 3. taking the neutral point of the network as the point of zero voltage. the sequence voltage at any point in the network can be found by subtracting the impedance drops of that sequence from the generated voltages. 22 and Z0 are the total equivalent impedances of the networks to the point of fault. In particular. The impedance to negativesequence currents for rotating apparatus will in general be different from the impedance to positive sequence. Sec.l I. As pointed out in Sec. This follows from the fact that within any one of the three networks the currents and voltages of that sequence are entirely independent of the other two sequences.&?L k!F 1 N2 x0 rl Fo E. Figs. 4. Distribution FactorsIf several types of unbalance are to be investigated for one point in the system. the current will ‘+.
Chapter 2 Symmetrical Components EQUIVALENT SYSTEM EQUIVALENT SYSTEM EOI EOIF \ E al \ (b) POSITIVE. Cd) SHORTHAND REPRESENTATION OF (b) EQUIVALENT SYSTEM EQUIVALENT SYSTEM \ &IF POSITIVE. fault on ungrounded generator. fault on generator grounded through . ECF (d) VECTOR DIAGRAM SHOWING VOLTAGES AND CURRENTS DURING FAULT . AND ZEROSEQUENCE DIAGRAMS FOR SYSTEM(a) (“a” PHASE) Eaf GROUND Go (b) POSITIVE. NEGATIVE.AND ZEROSEQUENCE DIAGRAMS FOR SYSTEM (0) (“a” PHASE) EC1 EC.NEGATIVE. (b) Singlelinetoground (c) Singlelinetoground a neutral reactor. SHORTHAND REPRESENTATION OF (b) (d) VECTOR DIAGRAM SHOWING VOLTAGES AND CURRENTS DURING FAULT SHORTHAND’REPRESENTATION OF (b) w Fig. NEGATIVE.AND ZEROSEOUENCE DIAGRMS FOR SYSTEM(a)(‘b”PHASE) POSITIVE. NEGATIVE. 20 (e) (d) Linetoline fault on grounded or ungrounded generator. (e) Doublelinetoground fault on generator grounded through a neutral reactor. AND ZEROSEQUENCE DIAGRAMS FOR SYSTEM (a) (‘a” PHASE) Ebi VOLTAGE (d) VECTOR DIAGRAM SHORTHAND REPRESENTATION OF (b) (b) ECI (d) VECTOR DIAGRAM SHOWING VOLTAGES AND CURRENTS DURING FAULT ICl *.
2b 26 (nl ONE LlNE OPEN WlfH IMPEDANCE Id$THER LINES 01 *> TWO LlNES OPEN TWO LINES OPEN. points in the system adjacent to the unbalance are represented by X and Y.. 1) . N.. I 1) )I 2 UNSA. (j 7 F I G( FAULT s INGLE LINETOGROUND FAULT WEEPHASETOGROUNO TT HREEPHASETOGRWNO RU GROUND.2Y 5... negative..v ./.. UNBALANCED STAR LOAD (II ! 01 ! (.. 2b ..C&NCED DELTeiLo*0 UNGROUNDED (mL II bl :I X.~ Tl... IMPEDPINCE. For series unbalances. and Zrefer to the positive. x Y Yl b) Y z.. P. respectively. N is again the neutral. SYSTEM THRIYJGH IMPErJ&cE (b) SINGLE LINETOCWUND FAULT LINETOLINE F&“LT THROUGH IMPEDANCE THREEPN~SETOGROUND FIULT WITH IMPEDINCE IN PnnSE 0 Ii1 DOUBLE LINETOCROOND (h) (b) FAULT DOUBLE LINETOGROUND FAULT THR‘WC” IMPEDANCE ..lRD LlNE NOLINES OPEN.) :)x Y N T./.2b .IMPEDANtES WA0 LINEB ~E~TRIL RETURN x x..Z.24 ! ( Symmetrical Components Chapter 2 a1 F 5 II ii ..2b IMPEDANCE IN ONE LINE “NEPUAL SERIES IMPEDINCES lMPEOANCES IN ONE LINE AN0 NEUTRAL RET”URN (Ul EQUAL IMPEDANCES TWO LINES IV) IN EQUAL IMPEDANCES IN TWO LINES WITH IYPECANCE IN NEUTRAL RETURN “NEWAL SERIES IMPEDANCE Fig.Y SYSTEM ONE LINE OPEN ($1 . .... 21Connection of the sequence networks to represent shunt and series unbalanced conditions.... .lb) I:C) xh . and zerosequence networks. For shunt unbalances the faulted point in the system is represented by F and neutral by N.~. Corresponding points are represented in the sequence networks by the letter with a sequence subscript.
+z*z. Jl 0 7 J2 ‘0 (35) (36) (37) (38) (39) IOFZ z. Z. Part (1) shows the equivalent system (2) the corresponding positive. and E.~Jl~Z~=Z Ea1(Z..l ZZ&al ZZ. EZF= 12. = E. = .z. = .s of the linetoneutral voltages at the point of unbalance. In these equations EIF..GtZo) z+Z.Zz.IzFZ.z.and zerosequence diagrams.and negativesequence currents add to zero at the system neutral so that the terms ‘neutral” and “ground” arc synonymous.l Z.. LinetoLine (27) cw IF = I.IlFZl= Z1Z.I. Double LinetoGround E.+ZZ.+Z.+Zo) I1F = z~zz+zlz~+z~z13 ZlZ? + ZoE. (25) through (29) the zerosequence impedance. Ed I 1F 3 12.and negativesequence quantities are balanced with respect to neutral.7 1 2 JO 15.FZ.+Z2Zo El.negative.l.+z~zo ZrZoE:..Z”+%2Z. EOF . Fig.I‘. I~F.FZ. z.= z ff.‘:ll zLzo+ ZSZ” 2122+ Z2Zo E:. Two Lines OpenFig.zo+zzzzo I2F= IOF=  (51) (52) (53) (54) E&o Z1fZ*fZo 20(d) : ZZoE. z.l z.Zo (29) FaultFig. and IOFare components of the fault current IF..I~FZ~ = = Z1Z~+Z.. They are the same only when no impedance exists between the neutral and ground. . G. = 31OF ElxE~y=E.FZl= Em = .EOy= I.* z.1(ZZo+ZZz+3ZoZz) I1F = zzlzo+zz1z. The networks as shown are for the “a” or reference phase only. and E.Eoy= .z.y+zE. E. In Eqs.1 Z. Part (4) is a vector diagram showing graphically the relationship between the various voltages and currents. 21(s) E..= 11.IzpZx= z1+z2+zo EIF = E&l. For example.+z. is infinite for the case of Fig.= .z”FZ2= .I(Z.z.ZzE.Z.z”+z?zi ZzZoE. IL(Z.Zo+ZZ~Z.1.. In parts (3) of Fig.I~F.. N.# 1 (33) 12.z. since by their definition positive.i”. and 2. I1F = I2F 21(p) Ed = IOF = z1+z2+20 (W (‘47) (48) (W (50) 9.+3z.l = EoF= .zy$ZlZO+z2z” Z:ZoE.“.I.I.FZ. is the linetoneutral positivesequence generated voltage.+z. are impedances of the system (as viewed from the unbalanced terminals) to the flow of the sequence currents.z.+z..l. Z&E.+3z. = E.Eqy= .pZo = (43) (JJ) (45) 14. In the zerosequence diagrams of (2) and (3) a distinction is made between “neutral”. One Line OpenFig. E?r. Single LinetoGround FaultFig.F are comp0nent. 20(aj all portions of the network within the boxes are balanced and only the terminals at the point of unbalance are brought out.I~FZ~ E.+=z*+z.and negativesequence networks no such distinction is necessary. by their nature zerosequence currents require a neutral or ground return path.$=z1+z2 (30) .IlFZl= ??kT!z+z. 20(e). 2 25 (31) (32) phase fault on the system.. 20(b) and includes 3Xo in the case of Fig.E.. I2F= &IIF=z 2 IlF z2 z 7 FaultFig.Z.z.Chapter 2 SymmetricalComponents IF = VBIld EIF= E. it is necessary that one be specific when speaking of linetoneutral and linetoground zerosequence voltages.+Z~) 20(e) (34 =%122+zlzo+zJ!o JO = 7 lJ2 . 20(b) (25) (. E1F= E.z. ~ EBx Es.+Zo EalZz Es. In the positive. and (3) the shorthand representation of the sequence diagrams. 21 gives a summary of the connections required to represent the more common types of faults encountered in power system work.zo +zzzo (W (11) (42) 197= 1°F = .Z2+Z1Z. ThreePhase FaultFig. 21(n) 13.Z?E. 20(a) 1 IIF=IF+ 10. Therefore.l IlF= z. Thus. In many cases impedance exists between neutral and ground and when zerosequence currents flow a voltage drop exists between neutral and ground. all positive. Zl.z. Zerosequence quantities however. are not balanced with respect to neutral.& Jl 2 0 Eo.+Zd +z +z 2 1 0 Ezx.. ZzZoE:. Impedance in One LineFig. and “ground”.+Z.“r. Equations for calculating the sequence quantities at the point of unbalance are given below for the unbalanced conditions that occur frequently.+z.
. ((1) SIMULTANEOUS SINGLE LINETOGROUND FAULT AND OPEN CONDUCTOR ON PHASE A.r%.z?+.?%~z~zo+%%~%~ Tf two or more Imbalances occur simultaneously. (56) l..l ~/. the connections will have to be made through phaseshifting transformers.=  Symmetrical Components ZZzZnE.ancous faults is consitlcrably more complicated than for single unbalances..+%%1%::+3%. (b) SIMULTANEOUS SINGLE LINETOGROUND THROUGH IMPEDANCE AND OPEN CONDUCTOR ON PHASE A.~‘. LOCATIONS. (f) FAULT ON PHASE B AND OPEN CONDUCTOR ON PHASE A.f%(j= %%.. 21 that should not permit the application of the same principles to simultaneous faults on multiple unbalances.Za+%Z?Zo (55) Chapter 2 j7.%._~~_~~_~~~~~!~~. = _ .+%%.%..+%z.. (c) SIMULTANEOUS SINGLE LINETOGROUND SIMULTANEOUS SINGLE LINETOGROUND TRANSFORMERS SIMULTANEOUS SINGLE LINETOGROUND FAULTS ON PHASES A AND B AT DIFFEREN.26 EEQ E. mutual coupling or connect ions will occur between the sequence nctnarks at each point of unbalance. le) Fig. So assumptions were made in the derivation of the reprcsentation of the shunt and series unbalances of Fig. In fact various cases of single unbalance can be combined to TRANSFORMERS SIMULTANEOUS SINGLE LINETOGROUND ON PHASE A AND LINETOLINE BETWEEN PHASES B AND G.. z&Connections between the sequence networks for typical cases of multiple unbalances. ?‘he analysis in the cases of simult. Cdl FAULT ON PHASE C AND OPEN CONDUCTOR ON PHISE A. and if the unbalances arc not symmetricsal with respect to the same phase..
% X’d’ 25% x”d= I7 96 x2= 17% xc= 5% Fig.iple unbalances. (4) That a voltage. 1.000 is an assumed voltage of J 1/3 volts between line ‘(a” and neutral. (c) Tabulation of generatorconstants. (6) find the sequence components of voltage and current at the desired locations in the system.. Assumptions (1) That the fault currents are to be calculated using transient reactances. to ‘lc” phase a. positivesequence. GENERATOR G. use synchronous reactances and the voltage back of synchronous reactance.000 kva transformer. (4) make the proper connection between the networks at the fault point to represent the unbalanced condition.and negativesequence networks in parallel. F ig. normal bus voltage can be used to find steadystate conditions and the machine reactance in the positivesequence network taken as being zero. This . the representation for a simultaneous single linetoground fault on phase ‘~a” and a linetoline fault on phases “6” and ‘<c” can be derived by satisfying the terminal connections of Figs. 21(f) shows the positive. 17. For example. of fault. in step (5) depend upon when the fault currents and voltages are desired. which also gives one illustration of simultaneous faults at different points in a system with one fault not symmetrical with respect to phase a. a. and linetoline voltages at the terminals of G’. and linetoline voltages at the breaker adjacent to the fault. a2. (5) solve the resulting singlephase circuit for the sequence currents at the fault.= 21 % x”. negative. 1. EXAMPLE 16.000 kva for the calculations. 21(d) dictates that the three netlvorks be connected in series. 22(a). (g) Line currents. 23(b) and the generator constants in Fig. in general. linetoground voltages. 1. (b) Negativesequence reactance to the point of fault.$:. Xc= 6% (c) 37.’ 12”/ X2’ 12. The fault involving phase “a” is usually taken as the rcfcrence and all others are shifted by the proper amount ))efore making the terminal connections rcquircd to satisfy that particular type of fault. ALL GROUND WIRES 0. (3) reduce the network to as simple a circuit as possible. = 100% x’r. use subtransient reactances and the voltage back of subtransient reactance immediately preceding the fault. X.‘i. 1.k. CONDUCTOR 4/O CU. Calculate the following: (a) Positivesequence reactance to the point. x. use transient reactances and the voltage back of transient reactance immediately before the fault. as viewed from the fault of j 100% will be used for reference.. (a) System singleline diagram. 21(d) and 21(f).500 KVA NEUTRAL UNGROUNDED xd=b. To summarize. while Fig. The line construction is given in Fig. 23Typical system assumed for fault calculation. (d) Fault current. and zerosequence shifts. Simultaneous faults that are not symmetrical to the reference phase can be represented by similar connections using ideal transformers or phase shifters to shift the sequence voltages and currents originating in all of the unbalances except the first or reference condition.Chapter 2 Symmetrical Components 25000 KVA q x=10% 20000 fb KVA X=9% 27 form the proper restraints or terminal connections to represent mult. (2) A base of 50. (f) Line currents. (e) Line currents. The positive. Problem OF FAULT CALCULATION Let us assume the typical transmission system shown in Fig. The positivesequence voltage to be used. “b” phase a?.000 KVA GROUNDED THROUGH 4% REACT. A few multiple unbalances that may occur at one point in a system simultaneously are given in Fig. (5) That the reference phases on either side of the stardelta transformers are chosen such that positivesequence voltage on the high side is advanced 30” in phase position from the positivesequence voltage on the low side of the transformer.66. if a few cycles after the fault occurs. and linetoline voltages at the 110 kv breaker adjacent to the 25.375 EBB STEEL (b) GENERATOR G. .5’X o14’ i45’ 14’4 x+x$ FAULT (0) ” 5000 KVA X=5% $b 15000 KVA x=6% t 15’ Cd0 445’ IO’0 66 KV LINE 110 KV LINE CONDUCTOR 410 CU. 23(c). 23(a) to have a single linetoground fault on one end of the 66 kv line as shown. linetoground voltages. if immediately after the fault occurs. the procedure in finding voltages and currents throughout a system during fault conditions is: (1) set up each sequence network as viewed from the fault. (2) find the distribution factors for each sequence current throughout its network. If regulators are used.. 22. and if steadystate conditions are desired. (c) Zerosequence reactance to the point of fault. linetoground voltages. Both of these reqllirements can be met simultaneously as shown in Fig. @if. (3) That all resistances can be neglected. and the machine impedances. respectively for an unbalance that is symmetrical to phase “a” are 1. . NEUTRAL 50. (b) Line construction.:~ 44 25000 KVA x=6% 14.
xl=xz=x.497+2.35 feet+xd for 23.+Ig’zocg.06 feet+xd for 12.89s(0.f.348) = 2. E0= zerosequence voltage of circuit a Ego=zerosequence voltage of circuit g =O.89= 11.i for 14 feetfxd for 28 feet) = 0.344) = 1. and between the two sets of ground wires. Line Reactances (Refer to Chap..smm. unit IK is 3/2 amperes in each of the two ground wires with three amperes returning in the ground. since the is assumed to be continuously current current current current of of of of circuit circuit circuit circuit a g a’ g’ It should be remembered that unit I0 is one ampere in each of the three line conductors with three amperes re . xd=+(xd for IO feet+&! for 10 feet+xd for 20 feet) = +(0. I0 = zerosequence I.. = +(0.892(0.$(xd for 10 feet+xd for 10 feetfxd for 20 feet) =0.’ is one ampere in each of the three line conductors with three amperes returning in the ground.+x...893(0. minus three times the average of the xd’s for all possible distances between conductors of the two circuits.348 = 0. =3(2. In each case the zerosequence mutual reactance between two circuits is equal to x. and the single ground wire for the 66kv line.845 ohms per mile..404) =0.” Although not strictly correct. =x.320+0.498) = 1.77 ohms per mile.?&rg) and Zo(. between each line and the t1vosets of ground wires.307 ohms per mile.+x.+xd=0. zd=+(zd for 14 feet+& for 14 feet+xd for 28 feet).320+0.>.59 ohms per mile.+x. form the zerosequence circuit denoted “a”‘.) +IKZo(a’R) fIOIZOb’) +I$ZO(a’n’) Eg’o=Iozo(ag~~ +Igzocgg.. .497+0. =O where zOca) zerosequence self reactance of the n circuit = =x.497+2. ‘x.’ is three amperes in the ground wire with three amperes returning in the ground.wire grounded.$(xd for 14 feet+x.411 ohms per mile. Then let: .40 ohms per mile.nj +Ii’zo(m’) =0 Ed= Iozo(aa. turning in ground.35 feet+xd for 12.w.q’= 0. with ground return.804 ohms per mile.279+0. j /  0) (< 0 0 a’ J _ _______ 0 ) 2 Fig.* 0 0 . .307) = 2.5 feet) =$(2..=0. the zerosequence mutual reactances between lines. the two ground conductors for this line. the two ground wires (g). EO’= zerosequence E. zO(ap*) zerosequence mutual reactance between the a = and g’ circuits.$(xd for 75 feet+xd for 62 feet+% for 48 feet) =2.79)+2. the 66 kv line (a’).5 feet fxd for 23.3303) = 1.893(0. x1=x2=x.69 ohms per mile.z~(~) zerosequence self reactance of the g circuit = =$x.he currents carried by the two ground wires of circuit “g” are equal.497 ohms per mile.893(0.493) = 1. denoted by “a” in Fig. with ground return.324) = 6. we assume t.497+0. since the ground wires grounded.348 ohms per mile.5 feet) =2. forms the zerosequence circuit denoted “g’..$(xd for 12. = 0.+&. with ground return.z~(~‘) zerosequence self reactance of the g’ circuit = =3x.o=Iozocag.. These quantities are interrelated as follows: Eo=Iozo(a) +Iczo<aa)+Idzow +Ihzow Positive.and NegativeSequence the 66 kv Line. ‘b. z~(~‘$) zerosequence mutual reactance between the a’ = and Q’ circuits.03 feet) =2.86 ohms per mile.) +Io’zc.28 Symmetrical Components Chapter 2 18. 24.307=0.’ = zerosequence are assumed to be continuously voltage of circuit a’ voltage of circuit .06 feet+xd for 12. with ground return. E.____. The zerosequence self reactance of the llOkv line in the presence of all zerosequence circuits is obtained by ZeroSequence Reactance+Since zerosequence currents flowing in either the llO. +IRzo(gj +IOI. ~~(~~1) zerosequence mutual reactance between the a = and a’ circuits Exe$(xd for 60 feetfxd for 50 feet$xd for 70 feet +xd for 46 feet+&i for 36 feet+xd for 56 feet +xd for 74 feet+xd for 64 feet+zd for 84 feet) =2. must be evaluated as well as the zerosequence self reactances. the zerosequence self reactance of either the 110or the GGkv line will be affected by the mutual coupling existing with all of the ground wires. For 4/O copper conductors x.$(xd for 14.‘= zerosequence I. The three conductors of the llOkv line..893(0.79) +2.89 2(0.‘o= zerosequence ground .+x. are assumed to form one zerosequence circuit. Indeed.90 ohms per mile.(.____’: _(0 .$(xd for 15 feettxd for 18.or the BGkv line will induce a zerosequence voltage in the other line and in all three ground wires. form the zerosequence circuit denoted “g”. = zerosequence I.279+0. zg(~‘) zerosequence self reactance of the a’ circuit = =&..and NegativeSequence Reactances of the 110 kv Line. 3) Positive. and the single ground wire (g’).497 ohms per mile.03 feet+xd for 18.364) =0.=0.* 09’ r . Reactances of x. unit I. and unit I. the three conductors for the 66kv line. zo(ap) zerosequence mutual reactance between the a = and g circuits =x.26 ohms per mile. Similar definitions apply for .. 24Z&osequence circuits formed by the 110 kv line (a).
it will be foind that EO li. obtained by finding the distribution of one am19. Each network is finally reduced to one equivalent imThe sequence networks are shown in Figs. 26.845) (40) (50 000) = 14y 0 (110) (110) (10) Zerosequence mutual reactance between the 66.3% (37 500) Positivesequence reactance of the 66kv line= (0. Illustrative examples of expressing these reactantes in percent on a 50 OOOkvabase follow: Positivesequence reactance of Gz= (25)(j0=33. it will be found that EX $with 1: = 0) = $(with lo = 0) = 0.= 2.r ‘_ ? i56.pedance as viewed from the fault. Z.‘? :: z*= j2LOX Fig. When this is done. . The Sequence Networks pere taken as flowing out at the fault. and 27. It IO will be found that F = 2.87 ohms per mile.. The distribution factors are shown on each sequence network. 26Reduction of the negativesequence negative sequence distribution network factors. 25. be zero in the above equations and solving for E” T. and the 1.87) (30) (50 000) = l8% (W w9 (10) circuits is obtained by letting IO’ be zero and solving for Ed F.0% (Pe .0 j7.5X j21. j2OX jl4X j22.  ‘. 25Reduction of the positivesequence positivesequence distribution network factors.05 ohms per mile.and the IlOkv line for the 30 mile section= (0. Carrying out this rather tedious process. 0 The zerosequence self reactance of the 66kv line in the presence of all zerosequence circuits is obtained by Eo letting lo be zero in the equations and solving for 7.5% 29 letting I. (66) (66) (10) .25 ohms per mile.5% PI kva base and the networks set up as viewed from the fault.804) (40) (50 000) =36 (& .Chapter 2 Symmetrical Components .. and the Positivesequence reactance of the 1lokv line = (0.4% Fig.= j26. 0 The zerosequence mutual reactance between the 66 and the llOkv line in the presence of all zerosequence . with all reactances expressed in percent on a 50 QOO.
258 p.637) = 1.~~+2.8% Symmetrical Components j22.637)(437.c.FO i70.5 amperes.and negativesequence quantities.637) = 1.“” j77.231 p.. r . Voltages and Currents at the Breaker Adjacent to Generator GI r. E.637)(2090)~j30=2340 amperes =2030j1170. i 76. = 1.956 .u. or normal.0388 j20.=&.1022 jl5.Vh! 1 il6X I i77. .8 =2090 amperes.5) =715 amperes.s=Eo+a2El+aEz=30 ZOOj12 900 = 32 800 volts.5% 1?Ecv The sequence voltages at the fault: E. 43X66 lo=I1 =I..1~c.7% The base. 21.=2145 amperes. E2= 12Zz= j(1.911 p. Voltages and Currents at the Breaker Adjacent to the Fault i55.b=EagEbg= 30 200fj12 900=32 800 volts.684)(1.~~+2. or 7960 volts linetoneutral.141 jl6% j82.sq ‘. I >j6% j52..7% Chapter 2 20.E. j .Eag= 30 ZOOj12 900=32 800 volts.=61.4% Y WV’ LjlSX j77. jl6% j97.407 p.0~~+13.637) = 1.7% Fig.859) (1.5 amperes. .G37)(26.=Eo+El+Eg=O E. I0 = (0.2% w .071 j50% .5% i46. 27Reduction of the zerosequence network zerosequence distribution factors.1oj. Z.~~=26. 50 000 The base. =540 amperes..4% j13.30 j20% & ‘Xgy. Eo= I. current at this point is &X 13.=j100~0j(1.&= j(1. =550 amperes. >j6% j25.7)%= j22. Iz= (0.u. there will be a phase shift in positive.5% 1.93% j16X j82.637 p.12% w .752)(1.5% 3 126.596 ’ j33. . E.x’ POq No<‘? S . Iz = (0.4%.J)~=j56.2X Using the distribution factors in the sequence networks at this point: II= (0.. = (1.637)(13.637)(21)%= j34.5 amperes.9% =j21 700 volts. I.5%= jS 600 volts. 22. Since a stardelta transformation is involved. Ecg= Eo+aEl+a2E2= 30 200j12 900 = 32 800 volts. The total reactance of the resulting singlephase netlvork is 2.725) (1. = 60 400 volts.u. >i. E. Voltages and Currents at the Fault The sequence networks are connected in series to represent a single linetoground fault.~~Z.768)(1.039 jl6% e4/x&. 361.4~~+21...l ~. .= EC. and the .u.. II= (0.u.6=70.1% Since normal current for the GGkv circuit (for a base kva of 50 000) 50 000 =p = 437.=j13. Ebc = Ebg .= j13 100 volts.OX j16% @ .=Io+aI~+a21~ =70j&6=70.. or normal. I. L’ ISl . voltage at this point is 13 800 volts linetoline. The linetoground and linetoline voltages at this point are equal to those calculated for the fault.5% +vv = J .2% .=lO+Il+lo= 1705 amperes..7~.. The total fault current = 1~+11+1~=4. Then : jlOO% IOF=I1F=I2F=. = 615 amperes.637) (2090)@O = 2480 amperes = 215O+j1240.0 . ~~=IO+a21~+aI~=70+j8..
684X21X1.= 30.=+j8580=8580 volts..35=41. or 63 500 volts linetoneutral.5+j9.3y0 = j825 volts.637)(262) . or normal.637) (262) = 16.4 amperes. The base.637~o)~~30= =3045+j5270 volts.7~o =j50 000 volts. Wagner and R. Ea=(j0.7 amperes.=E.I. December 1937. Transactions. Evans.Chapter 2 Symmetrical Components _ Ih=IO+a21~+a1~=40.637)(262) at this point are: current at this point is is 110 000 50000 4$x110 = 29.043)(1.35=41. 10271140.=jlOO%j(O.2 amperes. The sequence voltages at this point are: a276.637)(20)%= jl2.= . Method of Symmetrical Coordinates Applied to the Solution of Polyphase Networks. Transactions. by Edith Clarke.= Zo+aZl+a212 = j140 = 140 amperes. Eba= a2E1+aEz= 3610 j4290 = 5600 volts. 1918.5j9. Ez= j(O. Eo= j(O. by C.E. Sequence Network Connections for Unbalanced Load and Fault Conditions.9 amperes. E.= 50300j62750= 80 400 volts. amperes. 1. 37. Symme@icaZComponents (a book). V.j0. pp. The Electric Journal.E. Ebs=Eo+a2E1+aE2=50300j21750=54800 volts. Part II. Ih=IO+a211+alz= 418O+jiO=4180 amperes. E. 23. 5. I*= (0. A.=lo+ll+lz=4180+j70=4180 amperes..637’%)~j30=a14. = El+ Ez = 3610fj4290 = 5600 volts. voltage at this point volts linetoline. 50. 4.5% E.+E~+E. or normal. by C.63i)(20)~o=ji8. 2.=1. 3. = (0. 1937. 34.068)(1.. The sequence currents I. E.068)(l.039) (1. 1933..=10+nlTl+n?lz=N.8’% = j8130 volts.6 The sequence voltages at this point amperes.9=30. pp.I. McGrawHill Book Company. = 18. I. V.+1.. D. Lyon. j(0. E.10 830j4290 = 11 650 volts. pp.=j41000=41000volts. Eo=O. Harder. Fortescue.E.681)(1. 10 = (0.6 I.=aEl+a2E2=7220=7220 volts. EC.. are: 31 Io=O. Applications of Symmetrical Components (a book) by W.637) (20)% = j1.E.63’7)(21)% E1=(~j100a/.043)(1. A. F. Eab= 503OO+j62750= 80400 volts. E. 481488. Es. V. March 1931.637)(12)0j..725X12X1. E.2~0 = 565j980 volts. V. McGrawHill Book Company. L. Simultaneous Faults on ThreePhase Systems.+1.=Eo+aE1+a~E2=50 300j21750=54800volts.O39) (1. Voltages and Currents at the IlOkv Breaker Adjacent to the 25 000 kva Transformer The base. L. by E. 919941. = 262 amperes. REFERENCES 1. I. j(0.c= 10 830 j4290 = 11 650 volts.i25)(1.. Ebo= 100 600 = 100 600 volts.
Wright and C. such as paper. In Fig. dimensions. typical of both copper and steel conductors in the larger sizes. and voltages under fault conditions. General formulas are presented with their derivation to sho\v the basis of the tabulated values and as a guide in calculating data for other conductors of similar shapes. I Courlesy of Gmeral Cable Corporation Fig. 2. Other types of conductors such as Copperweld and CopperweldCopper conductors are also used for transmission and distribution lines. and ACSR (aluminum cable. This type of con32 Cow1e. By the use of a filler. steel reinforced). TYPES OF CONDUCTORS Fig. 1. L. This chapter presents a description of the common types of conductors along with tabulations of their important electrical and physical characteristics. between the outer aluminum strands and the inner steel strands. ductor is known as “expanded” ACSR and is shown in Fig. and steel for currentcarrying conductors on rural lines. and also for long river crossings. operation. Also included are the more commonly used symmetricalcomponentsequence impedance equations that are applicable to the solution of power system problems involving voltage regulation. 2A typical Courlesy OJ Allrminum company of America ACSR conductor. load flow. bronze.s~ of . A stranded conductor is easier to handle and is more flexible than a solid conductor. Additional formulas are given to permit calculation of approximate currentcarrying capacity of conductors taking into account such factors as convection and radiation losses as influenced by ambient temperature. stability. Shankle and R. 3A typical “expanded” ACSR conductor. a conductor of large diameter can be obtained for use in high voItage lines. 3. I. aluminum strands are wound about a core +f stranded steel.CHAPTER 3 CHARACTERISTICS Originul Authors: OF AERIAL Revised by: LINES D. F. 1A typical stranded conductor. Varying relationships between tensile strength and currentcarrying capacity as well as overall size of conductor can be obtained by varying the proportions of steel and aluminum. is shown in Fig. as buried counterpoises at the base of transmission towers. In the electricpower field the following types of conductors are generally used for highvoltage power transmission lines: stranded copper conductors. 4A typical Anaconda Hollow Copper Conductor. copper bronze. hollow copper conductors. composition and operating conditions. 4 is shown a representative Anaconda Hollow Copper Conductor. In this type of conductor. A typical ACSR conductor is illustrated in Fig. A stranded conductor. as overhead ground wires for transmission lines. system currents. Use is made of Copperweld. and expansion of electrical power systems it is necessary to know electrical and physical characteristics of conductors used in the construction of aerial distribution and transmission lines. or other system problems where the electrical characteristics of aerial lines are involved. It consists of a twisted copper “I” . wind velocity. Hall S the design. particularly in the larger sizes. and permissible temperature rise. Tremaine Sherwin H. F.4lnmirrum Company OJ America Fig. Courtesy of Anaconda Wire and Cable Company Fig. (bare copper).
copper strands are used with in which case copper resistance increases 17 percent. reference 1. that of the inner layer of strands. negative. Different relationships between currentcarrying capacity._ “_ . A discussion of large small conductor sizes the frequency has a negligible effect diameter conductors and their characteristics is given in on resistance in the dc to 60cycle range. Known as the General Cable Type HH hollow cop=241.and NegativeSequence Resistance Courtesg of Copperweld Steel Compnnu Fig.5 for hard drawn 97. While this variation is not strictly linear for an extremely wide range of temperatures. Thevariations of resistance with temperature are usually Copperweld conductors consist of different numbers of unimportant because the actual ambient temperature is coppercoated steel strands. Another form of hollow copper conductor is shown in =234. II. . and tensile strength can be obtained by varying the number and size of the Coppermeld and copper strands.. and cannot be used to give ac resistproduces a reduction in skin effect as well as inductance as ance variations unless skin effect can be neglected.. Corona losses are therefore smaller. the following general formula may be used. When the dc resistance of a conductor at a given temperature is known and it is desired to find the dc resistance at some other temperature. outside diameter. it is made up of segmental sections of copcopper. The constants developed are particularly useful in the application of the principles of symmetrical components to the solution of powersystem problems involving positive. beam as a core about which strands of copper wire are Rtl= dc resistance at any other temperature t1 degree C.5 for ann‘taled 100 percent conductivity copper.. inductive reactnnces and shunt capacitive reactances of the various types of conductors and some general equations showing how these quantities are used. Temperature Ejfect on ResistanceThe resistance of copper and aluminum conductors varies almost clirectly with temperature..1 for aluminum. When high currentcarrying capacities are desired as perature varies from winter to summer over a range of 0 degree C to 40 degrees C (32 degrees F to 104 degrees F) well as high tensile strength. .3 percent conductivity pey’conductor.. for practical purposes it can be considered linear over the range of temperatures normally encountered. Rt2 lQ+tz =(1) Rt1 M+tl where Rtz = dc resistance at any temperature tz degree C. 6A typical Copperweld conductor._ typical ” General Cable Type HH. particularly those required for analysis of powersystem problems. For compared with stranded conductors._. lmllow cylinder. current density. Copperweld strands to form CopperweldCopper conduct (a) Upper photographType V (b) Lower photographType F . 5A . frequency. Hollow copper conductors result in conThe above formula is useful for evaluating changes in ductors of large diameter for a given cross section of copper. The “I” beam is twisted in a direction opposite to M=a constant for any one type of conductor material. This construction also dc resistance only. 7Typical CopperweldCopper conductors The resistance of an aerial conductor is affected by the three factors: temperature.~ . Strength is provided by the core of steel indefinite as well as variable along a transmission line. The basic quantities needed are the positive.______. a typical conductor being illustrated in Fig.l . Courtesy of Copperweld Steel Company Fig. wound.. The following discussion is primarily concerned with the development of electrical characteristics and constants of aerial conductors.. and zerosequence impedances of transmission and distribution lines. = inferred absolute zero temperature._.Chapter 3 Characteristics of Aerial Lines 33 ors as shown in Fig. Positive. r_____. 7. Fiw 5.. ELECTRICAL AERIAL Courtesy of General Cable Corporation CHARACTERISTICS CONDUCTORS OE Fig. G. Practical formulas and methods will now be given to take into account these factors. and zerosequence resistances. An illustration of percentage change in resistance is when temand protection by the outer coating of copper. 1. negative. per mortised into each other to form a selfsupporting = 228. This is generally true for conductor sizes up to 2/O.
063598 f =frequency in cycles per second.34 Characteristics of Aerial Lines Chapter 3 Skin Eflect in Straight Round Wires. 1933. K is given as a function of X. P = permeability = 1. However. Table 5 (skin effect table) is carried in the Bureau of Standards Bulletin No. rdc = dc resistance at any known temperature. This is due to skin effect. 50. ZeroSequence ResistanceThe zerosequence resistance of aerial conductors is discussed in detail in the section on zerosequence resistance and inductive reactante given later in the chapter since the resistance and in + 103 slO+ y. For other frequencies the following formula should be used. In Table 5. the following procedure is followed. For these reasons the effect of current magnitude will not be analyzed in detail. The presence of this additional factor complicates the determination of resistance of magnetic conductors as well as any tabulation of such data. K = value given in Table 5.and negativesequence inductive reactance of threephase aerial lines it is first necessary to develop a few concepts that greatly simplify the problem.O for nonmagnetic materials. Fig. 10l 10Z n Combined Skin Eflect and Temperature Eflect on Resistance of Straight Round WiresWhen both temperature and skin effect are considered in determining conductor resistance. Evans. (2). determine Ii. and 60 cycles. These tabulated resistances are generally values obtained by tests. F. D.210g z 10a B lo" x 10n 2 lo'3 1 IOH E' 10K p lo'6 0 10n 0 IO 20 30 40 Amperes prCablp 50 16 15 i: 12 II IO 9 . Then using Eq.and NegativeSequence actance Inductive Re To develop the positive. Skin effect is due to the current flowing nearer the outer surface of the conductor as a result of nonuniform flux distribution in the conductor.or tables give the resistance at commercial frequencies of 25.E IO6 :10' 5108 . Positive. to values of X= 100. r mire= dc resistance of the conductor in ohms per mile. McGrawHill Book Company. 6 : : 60 Fig. The conduct. . To facilitate interpolation over a small range of the table. First. Current magnitude determines the flus and therefore the iron or magnetic losses inside magnetic conductors. where X= . the total inductive reactance of a conductor carrying current will be considered as the sum of two components: *This figure has been taken from Symmetrical C@mponen& (a book) by C. 169 on pages 2268. This increases the resistance of the conductor by reducing the effective cross section of the conductor through which the current flows. Copperweld.The resistance of mnmagnetic conductors varies not only with temperature but also with frequency. TY= Krdc ohms per mile (2) where TY= the ac resistance at the desired frequency (cycles per second). Having calculated X. First calculate the dc resistance at the new temperature using Eq. 2. and the tables on magnetic conductors such as Copperweldcopper. using the new dc resistance for rdc and the value of K obtained from Table 5. and ACSR conductors include resistance tabulations at two current carrying levels to show this effect. SElectrical Characteristics of Steel Ground Wires* ductive reactance presented to zerosequence currents is influenced by the distribution of the zerosequence current in the earth return path. 8 gives the resistance of steel conductors as a function of current. Wagner and R. it is accurate as well as convenient to plot a curve of the values of K vs. 10S . calculate the new ac resistance rf. Then substitute this new value of dc resistance and the desired frequency in the equation defining X. (1). from Table 5. values of X. E#ect of Current on ResistanceThe resistance of magnetic conductors varies with current magnitude as well as with the factors that affect nonmagnetic conductors (temperature and frequency).
7J Examining the terms in the first bracket. (2) The inductive reactance due to the flux external to a radius of one foot and out to some finite distance. 2lnl= inductance due to the flux outside the conductor ’ to a radius of one foot. Dr2 and r must be expressed in the same units for the above equation to be valid. 9A two conductor single phase circuit . in the twoconductor case. Lewis. r = radius of conductor.) a and radius 6 where p = permeability of conductor material. 4). radi. Grouping the terms in Eq. 10). (4) keeping in mind the significance of the general term 21n t. Conductor 2 also produces similar lines of flux. In cable circuits. per conductor.(inductance) . (4) Fig. per conL = %+dj+2hT ductor (G)  L = ~+21n~+21nD~ ductor abhenries per cm.Chapter 3 Characteristics of Aerial fines (1) The inductive reactance due to the flux within a radius of one foot from the conductor center. 9 it can be seen that it is unnecessary to include the flux beyond the return conductor 2 because this flux does not link any net current and therefore does not affect the inductance of conductor 1. Equation (6) can be written again as follows: Fig. This term is not dependent upon the conductor characteristics and is dependent only upon conductor spacing. however. The term in the second bracket of Eq. The classic inductance formula for a single round straight wire in the twoconductor singlephase circuit is: L=i+2 In 4 abhenries per cm. (6) is an expression for inductance due to flux external to a radius of one foot and out to a distance of D12.‘8 It can be shown most easily by considering a twoconductor singlephase circuit with the current flowing out in one conductor and returning in the other. This concept was first given in Wagner and Evans book on Symmetrical Components2 and was suggested by W. (5) we have: D12 abhenries per cm. loInductance due to flux between radius (2 In k abhenries per cm. radius out to Drzft. Rewriting Eq. From Fig. where i= inductance due to the flux inside the conductor. per con(5) Ldueto L d. From derivation formulas a general term such as 21n i represents the flux and associated inductance between circles of radius a and radius Z.which. (See Fig.ue flux out to flux to a one external ft. surrounding a conductor carrying current. In Fig. 9 such a circuit is shown with only the flux produced by conductor 1 for simplicity.to a 1 us ft. 21nDy = inductance due to the flux esternal to a one foot radius out to D12feet where D12is the distance between conductor 1 and conductor 2. is the distance between conductor 1 and conductor 2. the distance between conductors is less than one foot and the inch is a more common unit (see Chap. including the flux inside the conductor. it is evident that this expression is the sum of the flux both inside the conductor I* and that external to the conductor out to 02 Furthermore this expression a radius of one foot 2lnl r>’ ( contains terms that are strictly a function of the conductor characteristics of permeability and radius. Dn=distance between conductor 1 and conductor 2. A. For practical purposes one foot is used as the unit of length since most distances between aerial conductors are in feet.
809 a 54threelayer..756 0........0...i72 91........ it is called the “internal reactance......... GMR is expressed in feet....C......+zd ohms per conductor per mile (9) where xa= inductive reactance due to both the internal flux and that external to conductor 1 to a radius of one foot...iBSa a a a.......2 0.....2794 Hollow stranded conductors and A.......0 OUTSIDE Fig.....O...........S10 a Single layer A...4 ’ RATIO a* I/2 0. If we let the first term be called zn and the second term zd....................... GMR is a mathematical radius assigned to a solid conductor (or other configuration such as stranded conductors)........ ........................ Sometimes a tabulated or experimental reactance with’ 1 foot spacing is known..00 0....... and from this it is desired to calculate the conductor GMR... 12A Threeconductor threephase spacing)....... 50.. (11) Antilog.................... Converting Eq...........779a ..... GMR = conductor geometric mean radius in feet.......... ......r=2(xD+xJ ohms per mile of circuit 00) since the circuit has two conductors. 0..S.........2235(a+S) CIRCULAR TUBE 1.72Ga a .77~~ 127............... _ ~ xd = inductive reactance due to the flux surrounding conductor 1 from a radius of one foot out to a radius of D12feet.... 2 = 0. (7) to practical units of inductive reactance............... In other words..2794f log... ............... When reactance is known not to a onefoot radius but out to the conductor surface..... which describes in one term the inductance of the conductor due to both its internal flux z and the external flux out to a one foot radius 21nl ..........” The formula for calculating the GMR from the “internal reactance” is: GMR = Antilogi.. then z=z.. (8) 1 G.... r1 0 ( GMR therefore makes it possible to replace the two terms which is entirely dependent ‘upon the condu‘ctor characteristics...................... physical radius “Internal Reactance” 0........36 Characteristics of Aerial Lines Solid round conductor....90 2 z W 0..........6 INNER RADIUS OUTER RADIUS DIAMETER 0.o Reactancewith 1 ft spacing (60 cycles) 0...O........ It can be defined as the radius of a tubular conductor with an infinitesimally thin wall that has the same external flux out to a radius of one foot as the internal and external flux of solid conductor 1.. a Point outside circle to circle..... ......... circuit (symmetrical The values of GMR at 60 cycles and xs at 25............O... and 60 cycles for each type of conductor are given in the tables of electrical characteristics of conductors.............95 ~ 0. or both a “go” and “return” conductor..... distance to center of circle Rectangular section of sides a and p .. Full stranding 7 .....85 0.... 4 60 ohms per conductor per mile (8) where f =frequency in cps....... singlephase circuit... 19............ out to a radius of one foot....82Ga 26twolayer...8 1.............. .R..........0........758 38. &+0...R.. For the twoconductor...S..... &=distance between conductors 1 and 2 in feet.................2794 (60 cycles) feet (12) ////////////////‘////////////////// Fig... til........................ the total inductive reactance is ..... They are given .0..........80 0..776 Chapter 3 0...........\/IR = feet....0... ..........35a07Oa Point within circle to circle ..............C. 11Geometris PHASE o Mean Radii and Distances.......0.0............... . By derivation from Eq..... GMR in the first term is the condlictor “geometric mean radius”............ then...... (neglecting steel strands) 30two layer.27946 log.....0.
which practically eliminates mutual effects between the conductors and earth. This factor is denendent on distance between conductors only. Figure 12 shows a threeconductor. Eb. This table is taken from the Wagner and Evans book Symmetrical Components. With all three currents flowing simultaneously. the magnetic field external to the conductors is not zero. produced by line to ground voltages E. positive. and xd.+Ib+ l..Ib. Assume that the threephase voltages E. Since the log of unity is zero. due to the unsymmetrical conductor spacing. Also.’ = laxaa+l. Considering only Ib flowing in conductor b and returning by the same remote path j feet away.) since xd(ba)=xd(ca)=xd(bc)=xd. To reduce this effect to a minimum. flowing in conductor c and returning by the same remote path g feet away.=inductive reactance for conductor a due to the flux out to one foot.. I.. (16) where xac is the inductive reactance associated with the flux produced by I.bZd(t. threephase circuit carrying phase currents I. Eb.‘=Icr(xa+xd). assuming the currents to be equal and balanced. E. In terms of x.. It. divided by the current per phase results in the positive.’ = Ia% IbZd(ba) Iczd(. xnb=xd(bk) Zd(bn) (15) ivhere x& is the inductive reactance associated with the flux ^^ . and E.2794 &log?. the resultant effect is the sum of the effects protluced by each current flowing alone.. inductive reactance spacing factor. and 60 cycles. particularly telephone circuits. E.or negativesequence currents. When the conductors are unsymmetrically spaced.or negativesequence voltages E.. singlephase circuit. Ib.. Dividing x1=x2= the equation by I. %a= xa+xd(ak) (14) where only I.=O.= IbI.xa~. In the following solution. First. consider the case Ivhcre the three conductors are symmetrically spaced in a triangular configuration so that no transpositions are recluired to maintain equal voltage drops in each phase along the line.or negativesequence inductive reactance per phase for the threephase circuit. flowing simultaneously..>.= I. and the currents I. Having developed 2. can be considered as positive.’ =z~fxd I. produced by Ib that links conductor a out to the return path f ieet away. Icx.Chapter 3 Characteristics of Aerial Lines 37 in these tables because they are a function of conductor characteristics of radius and permeability. and is not associated with the conductor characteristics in any way. 50.. and Eq. considering only I.hey may be either positive. then the ratios zs and ‘p (ck) (ck) approach unity. In addition to the GMR given in the conductor characteristics tables. Eb. Values of xd for various spacings are given in separate tables in this Chapter for 25. xac = mutual inductive reactance between conductor a and conductor c. the voltage drop for each conductor is different. that links conductor a out to the return path g feet away. Also assume the currents I. Since I. Therefore no return current flows in the earth...or negativesequence inductive reactance per phase for a threephase circuit with equi_.or negativesequence currents are assumed to flow producing voltage drops in each conductor. that may result in telephone interference.EE..page 138. %d(bn)) (17 Expanding and regrouping the terms we have: Eaa’=~&.E. Zd(ck)) Using the definition of xd.~in terms of a twoconductor. Values of xB for various conductors are given in the tables of electrical characteristics of conductors later in the chapter. I.. TO simplify the problem further. can be written thisexpression and dcbk) the remote to Assuming the distances d(+. Thgefore..a) lcxd(c@ +[Irrxd(ak)+IbZd(bk)+leZd(ck)]. x=b = mutual inductive reactance between conductor a. (20) (21) E.. the positive. Taking phase a. it is sometimes necessary to determine this quantity for other conductor configurations. are balanced (equal in magnitude and 120” apart) so that t. are applied to the conductors and corresponding positive. is flowing and returning by a remote path e feet away.. threephase circuit. path approach infinity. . (18) reduces to Ea E.. E. we have in terms of xs and xd factors: &a=~d(ck)xd(ea) EaEc+’ = Ia(&+Zd(sk)) +rchd(ck) +I&d(bk) xd(ca)). thereby causing induced voltages in adjacent electrical circuits. With all three currents I. the conductors are transposed so that each conductor occupies successively the . consider only one current Howing at a time. and conductor b. assumed to be the point k. written ~ahd(ak) W3) the terms in the bracket may be +lbhd(bk) xd(ck)) .. 0. the voltage drop is: Finally. and the values of xd are given in the tabIes of inductive reactance spacing factors for various conductor spacings. the two terms in the bracket are zero. The voltage drop per phase. I. (19) and I. lateral spacmg is the same as for one conductor of a singlephasecircuit as previously derived. ohms per phase per mile E. 1. are also balanced so that I. where x. .and negativesequence inductive reactance of a threeconductor..=self inductive reactance of conductor a...c + \vhcre (13) xX3. deck). xd=inductive reactance corresponding to the flux external to a onefoot radius from conductor a out to the center of conductor b or conductor c since the spacing between conductors is symmetrical. these quantities can be used to determine the positive.E. Ib. Figure 11 is given for convenience in determining such values of G&fR. and 2. I.or negativesequence voltages.
(IbfIc) Zd(12) +zd(23) +xd(31) 1.0 NO. 14Quick reference curves for 60cycle inductive reactance of threephase lines (per phase) using hard drawn copper conductors. and xd is taken NO. E a’ = In& Ibxd(12) lc.27Y4&)(log d<l?)+log d<?J)+log d& xd = + 0.or negativesequence inductive reactance per phase x1=x2= (&+xd) ohms per phase per mile where xd = gxd(l2) +~d:2~) hd(313) ohms per phase E. 13A Threeconductor threephase rical spacing). E.xd(13).zd~3d31 feet (23) and use the inductivereactance spacing factor for this distance.6 5 t:: L Since Is+Ib+I.31.lerd(m. In the third line section where I.000 500.2794& log GMD where GMD (geometrical mean distance) = *a and is mathematically defined as the nth root of an nfold product. Referring to Fig.1 is the inductivereactance spacing factor for the GhID (geometric mean distance) of the three conductor separations.I. we have the positive.g~d(32).” EL” = I&a . (19) for the first line section where I. which is similar to the symmetrically spaced case except .000 In the second line section where I. the total voltage drop for each conductor is the same..5 0.’ I I Ed* I I I 1 Fig.. the inductive reactance voltage drop of phase a in each of the three line sections is obtained..+ 0. First. Taking the average voltage drop per line section.4 2 3 EQUIVALENT 4 5 7 CONDUCTOR IO 20 SPACINGFEET 30 40 per mile. is flowing in conductor 1.x.El)+@.3 1.=O. 13 and using Eq. xn is taken from the tables of electrical characteristics of conductors presented later in the chapter. called a “barrel of transposition”. 3 Dividing by I. . circuit (unsymmet same positions as the other two conductors in two successive line sections. For xd. is flowing in conductor 3.5 NO..000 150.. then. (22) Fig.000 l.I.\3/d. E: E:’ = Inxa.xd(z). we therefore have an expression for the positive. For three such transposed line sections.7 NO. we have ~a”)+(%“%“‘) 3 +1d.= .. _ I b(xd(l?) 3 _Ic(Zd~12)+Zd(23)+Zd(31)) 3 x‘l8 x $ 0.=I. 1.(x. For a threephase circuit where the conductors are not symmetrically spaced. 5d==:(o.&(31) E 3Yg =(E. we can take the average of the three indllctivereactance spacing factors xd = +(xd(12) +rd(23) +xd(31)) ohms per phase per mile IO A Ib k 1 2 3 %’ I I I Eo ///////////////N////N/lllllllNllllllllll//// E..38 10 IE Characteristics of Aerial Lines Ib Chapter 3 Expressed in general terms. Adding these together and dividing by three gives the average inductive reactance voltage drop for a line section.x. I.000. in the case of unsymmetrical conductor spacing.is flowing in conductor 2.I 2fo 410 300. In the following derivation use is made of the general equations developed for the case of symmetrically spaced conductors. This latter procedure is perhaps the easier of the two methods. See Eqs.279& I st SECTION 2nd SECTION A IC log d1zdz3d31 3rd lb  SECTION I f sd=0. or we can calculate the GMD of the three spacings GMD = . (10) and (21). fxd(31)) =.or negativesequence inductive reactance.2794@ log Vtll2rE?3c/31 xd= 0. E. For total reactance of singlephase lines multiply these values by two. and any electrical circuit parallel to the three transposed sections has a net voltage of very low magnitude induced in it due to normal line currents.
or dissymmetry is to be considered.Chapter 3 Characteristics of Aerial Lines 226.590. (ll).” Since most threephase lines or circuits do not have conductors symmetrically spaced.500 1. nents cannot be used. (15). 16Quick reference curves for 60cycle inductive reactance of threephase lines (per phase) using CopperweldCopper conductors. (10) and (21). thus preventing the use of general symmetricalcomponents methods. See Eqs. in which case the general symmetrical components methods can be used. . curves for 60cycle inductive react(per phase) using ACSR conductors.” For quick reference the curves of Figs. IsQuick reference ance of threephase lines For total reactance of values by two.or negativesequence inductive reactance is generally used. assumes that the circuit is transposed. from the tables of inductivereactance spacing factors. (IG). l7Quick reference curves for 60cycle inductive reactance of threephase lines (per phase) using Copperweld conductors. (IO) and (21).1 Fig.000 795.4 I 2 3 EQUIVALENT 45 7 IO 20 CONDUCTOR SPACINGFEET 30 40 Fig.000 I 2 3 4 5 7 io EOUIVALENT CONDUCTOR SPACING  20 FEET 30 40 Fig. and (17) have been plotted giving the reactance (x~+z~) for different conductor sizes and “equivalent conductor spacings. the above formula for positive. 0. When a singlecircuit line or doublecircuit line is not transposed. Geometric mean distance (GMD) is sometimes referred to as “equivalent conductor spacing. Some error results from this method but in general it is small as compared with the laborious calculations that must be made when the method of symmetrical compo. 94/2D EQUIVALENT CONDUCTOR SPACINGFEET I.400 47 1. either the dissymmetry is to be ignored in the calculations. In considering this dissymmetry. (10) and (21).192. For total reactance of singlephase lines multiply these values by two. For total reactance of singlephase lines multiply these values by two. This formula. unequal currents and voltages are calculated for the three phases even when terminal conditions are balanced.600 336. singlephase lines. multiply these See Eqs.000 1. however. See Eqs. In most cases of dissymmetry it is most practical to treat the circuit as transposed and use the equations for x1 and x2 derived for an unsymmetricallyspaced transposed circuit.
By referring to Fig. ZeroSequence actance Resistance and Inductive Re b b’ 00 t b0 _ OC t/_ V’ 5 Ob’ cd FIRST 2 00’ SECTION CO 00 Ob’ 00’ b0 SECOND OC’ SECTION b0 00’ co 00 THIRD OC’ ob’ SECTION The development of zerosequence resistance and inductive reactance of aerial lines will be considered simultaneously as they are related quantities. As contrasted with the usual conductor arrangement as shown in Fig..R 3.or negativesequence reactance of the paralleled circuit is: ~dn. b. The formula also assumes symmetry about the vertical axis but not necessarily about the horizontal axis. A second conductor. or any combination of these. 19 might be used. first consider a singlephase circuit consisting of a single conductor groundecl at its far end with the earth acting as a return conductor to complete the circuit.and negativesequence inductive reactance. the arrangement of conductors shown in Fig. The first term in the above equation is the positive. they flow out through the phase conductors and return by a neutral path consisting of the earth alone. their development is considered jointly. this arrangement of con Fig.or negativesequence reactance for the combined circuits. neutral conductor alone.dx&n f 21=x2 = 0. The formula assumes transposition of the conductor as shown in Fig. the positive.i:l:~~. the effect of mutual inductance betjveen the two circuits is not entirely eliminated by transpositions.and NegativeSequence Reactance of two parallel threephase cirParallel CircuitsWhen cuits are close together. 18. The zerosequence resistance and inductive reactance of this circuit are dependent upon the resistivity of the earth and the distribution of the current returning in the earth.2794% 3 logro G~~IRconductor [ A 1oglo (li.40 Charactekics of Aerial Lines 00 00’ Chapter 3 Positive. Since the return path often consists of the earth alone. grounded at its far end with the return path for the current consisting of the earth.)” b0 Ob’ CO Fig. it is necessary to use a method that takes into account the resistivity of the earth as well as the current distribution in the earth. This problem has been analyzed by Rudenberg. particularly on the same tower. 18 showing two transposed circuits on a single tower. 18.)i~:ii. Figure 20 shows a singlephase circuit consisting of a single outgoing conductor a.. phase circuit with earth . This has been demonstrated in several references. 18Two parallel threephase circuits showing transpositions. Fig. is shown to illustrate the mutual effects produced by current flowing in the singlephase circuit. Since both the zerosequence resistance and inductivereactance of threephase circuits are affected by these two factors. or the earth in parallel with some other path such as overhead ground wires. Mayr. The second term represents the correction factor due to the ohms per phase per mile. (24) in which the distances are those between conductors in the first section of transposition. However. 1 ductors results in five to seven percent greater inductive reactance than the usual arrangement of conductors. 20A single conductor single return. on a single tower mutual reactance between the two circuits and may reduce the reactance three to five percent. l9Arrangement materially oc which of conductors on a single tower increases the inductance per phase. As with the positive. Since zerosequence currents for threephase systems are in phase and equal in magnitude. This permits the development of some useful concepts for calculating the zerosequence resistance and inductive reactance of threephase circuits.db~. overhead ground wires.
of conductor a with (~rth return (the voltage between CL earth for unit curand I~cnt in conductor CL). Therefore the corresponding zerosequence self and mutual impedances per phase are three times the values given in Carson’s simplified equations.OO4657. &R (30) rc = resistance of conductor a per mile. which was assumed to have uniform resistivity and to be of infinite extent. singlephase circuit with earth return now considered as a singlephase.f loglo ohms per mile (26) ohms per mile. Equating the logarithmic expressions of the two equations. therefore equals the resistance of one conductor for a threephase circuit. This expression is similar to the inductivereactance as given in (‘arson’s simplified equation for self impedance..b=distance from the equivalent conductor to a parallel conductor.0 @& (W log. (30) the zerosequence self impedance of two ground wires with earth return can be derived. considered the return current to return through the earth. is sllbstituted for D12.OO4657f log.o&~=j0. GMR = geometric mean radius of conductor a in feet. Eq. D. D. CMR. Equation (31) gives the zerosequence mutual impedance between two overhead ground wires.2794%f loglo sR twowire circuit. Also an inspection of Carson’s simplified equations show that the self and mutual impedances contain a resistance component O. equivalent depth of return. The more commonly used method is that of Carson.01397”/” log. ohms per phase per mile (31) where f =frequency in cps.=r.+0. .. t)(. The solution of the problem is in two parts: (1) the determination of the self impedance z. (27) 4 f This defines D.+0. (30) gives the zerosequence self impedance.tnecn conductors n and b with common earth return (the voltage between b and earth for unit current in a and earth return). f =frequency in cps.and (2). replace the three conductors by a single equivalent conductor in which three amperes flow for every ampere of zerosequence current. I (25) z.. the mutual impedance zp.Qrn)=0. A useful physical concept for analyzing earthreturn circuits is that of concentrating the current returning through the earth in a fictitious conductor at some considerable depth below the outgoing conductor a.004657flog~oGMR or D. z.\s a result of Carson’s formulas. Considering three conductors for a threephase system..004657flog. d. = 0. d nb Rewriting Carson’s equations in terms of equivalent depth of return.01397$ ohms per phase per mile.OO159~+jO. and Carson and Campbell in this country. .004657f ohms per mile 2160 J .= resistance of a single conductor equivalent to the two ground mires in parallel.+O.. __ GlMR ohms per phase per mile (30) where T. have: we 20=37.O0159f. the selfinductive reactance is given by the previously de(See Eq..=resistance of a conductor equivalent to the three conductors in parallel. (r. This is different than the GMR for a single conductor and is derived subsequently as GMRclrcujt. For the singleconductor. (28) and (29). unit zerosequence current consists of one ampere in each phnsc conductor and three amperes in the earth return circuit. p=earth resistivity in ohms per meter cube. For the case of a single overhead ground wire. GMR=geometric mean radius for the group of phase conductors.. like Pollnczek.z. or some other equivalent conductor if the mutual impedance between two parallel threephase circuits is being considered.01397j log. To convert the above equations to zerosequence qunntities the following considerations must be made. the distance between conductor a and the fictitious return conductor in the earth. and frequency.Chapter 3 Characterzstics of Aerial fines 41 and pollaczek in Europe. therefore becomes 2 where r.. 2 ab log.00477f+j0. which is a function of frequency. I j0. r. sR where D.+0. p. and using average ]l(:ights of conductors nhovc ground. . twowire circuit. who. 20= 37. This ccluivnlent depth of the fictitious return conductor is represented as D.. Subsequently the GMR of a group of conductors are derived for use in the above equations. f.b = distance between conductors a and b in feet. the following fundamental simplified equations may be written: . Calling the zero sequence impedances zo and zOm.00159f+j0. 3r. d. and shows that it is a function of earth resistivity. Zerosequence self impedance of two ground wires with earth return Using Eq.=21GO f? feet.. or jO.00477f+j0.OO4657f ohms per mile.004G5ijlog~0~ log. (8)) for a singlephase.00159~ij0. (29) Al) These equations can be applied to multipleconductor circuits if r. the GMR and d&brefer to the conductors as a group.=r. To use Eqs.00477f+j0.00159f +jO. is the resistance of one of the two ground wires). z~~=0. rivedj0.
Substituting the values of self and mutual impedances given by Eqs. 2 ~(GMII) D.004i7~+j0. is the distance between the two conductors 5 and y. b. 21 is For conductor b: and for conductor c: The equation for zerosequence self impedance of n ground wires with earth return can therefore be obtained by substituting : for rC and Eq.etor or q/(G$IR) (dxy) dxy2 where d.DE! GMR ( log. GMR is the geometric mean radius of the n ground wires as a group.) Substituting ‘. )I ohms per mile. I Fig..+9(0. in ohms per phase per mile. The expression for self impedance is then converted to zerosequence self impedance in a manner analogous to the case of single conductors with earth return. The more common case is that of threephase conductors in a threephase circuit which can be considered to be in parallel when zerosequence currents are considered.) for GMR inEq.. the current in each conductor is one third. which are not transposed. (33) for GMR in Eq. =5 nhere r.OO477j+jO.00159f) +jO. and c as shown in Fig. (30). is the n resistance of one of the n ground wires. transpositions.O1397j ohms per mile per phase. These expressions were expanded to include the case of multiple overhead ground wires.+Zbb+z~o+2z~b+2Zbc+22. (: f+F+Y in which z. the average drop per conductor is ~(z. z..“g.)d(g.gl)d(6ng.)d(szp3)d(gzen) (d(. (dxy) (32) Zerosequence self impedance of n ground wires with earth return Again using Eq.. In order to derive the zerosequence self impedance of threephase circuits it is first necessary to derive the self impedance of threephase circuits taking into account $2 log. This factor was not considered in the preceeding cases for multiple overhead ground wires..+O. zbb.01397~ ohms per mile per phase... the zerosequence self impedance of n ground wires with earth return can be developed.. is the resistance of a single conductor equivalent to n ground wires in parallel. 21Self impedance of parallel return. voltage drop in conductor a for the position inin Fig.. (GJIR therefore becomes *(GMR)2 cOnd.. (28) and (29) in this expression. ds+2 ab ohms per mile.42 Characteristics of Aerial Lines Chapter 3 GMR= geometric mean radius for the two ground wires. GMR= ‘q(GMR)” conductor (d~slsr)d~p. log.3.a.l) feet..d(gag?)d(gJg.= f 3r. and zCC are the self impedances of the’ three conductors with ground return and z. ze= 3r. Consider three phase conductors a.. for r. 21. (30)..r)d~ple. log. Since conductor a takes each of the three conductor positions successively for a transposed line. and ~(GMR)(d. then r.. This expression can also be written in terms of all possible pairs of distances as follows. Also the three conductors in a threephase circuit are generally transposed.)) (d(. .O04657f 3 log. With the conductors transposed the current (3O). gR b (30) Since r.. the zeroseyuencc self impedance of two ground wires with earth return becomes zo= ~+0. which may be written as follows in terms of all possible distances.) (d(s?s.b.. conductors with earth divides current The dicated equally between the conductors so that for a total of unity. and zac are the mutual impedances between the conductors. ?+2 log10 bc Self impedance of parallel conductors with earth return In the preceeding discussion the self and mutual impedances between single cylindrical conductors with earth return were derived from which the zerosequence self and mutual reactances were obtained.g. ??bc.
+O. d(GLIR) dab2&c2dCa2 feet. 22. GiCIDSeparatlo..bdb.00477f +jo. (31) results in an equation for the zerosequence mutual impedance between one circuit and n ground wires.(GMD)?. mRelrcult= +(Gi\/lR)aCOndUCtOr Zerosequence self impedance of two idel2tical parallel circuits with earth return For the special case where the two parallel threephase circuits are identical. Therefore GMRClrcu. By previous derivation (See Eq.. D.(GMD)2..) 2 feet.01397j ohms per phase per mile log.) compute the zerosequence self impedance.. 21.Chapter 3 Characteristics of Aerial Lines 43 The ninth root in the denominator of the logarithmic term is the GMR of the circuit and is equal to an infinitely thin tube which would have the same inductance as the threeconductor system with earth return shown in Fig... (38) and Eq. 2 nb Zerosequence mutual impedance between two circuits with earth return Using a similar method of derivation the zerosequence mutual impedance between 2 threephase circuits with common earth return is found to be ~O. 04.o1397j log. then the GMD between the two groups is ohms per phase per mile GMD = 3i/dagldbgldc&ldagndb~~d~~~ feet Substituting this quantity for d.(GMD)‘~eparation n ground return) and threephase circuit (with earth wires (with earth return) ohms per phase per mile (37) dnb2dbc2dca2 where ~GMR.t= YGMR conductor(~d..d. c. Note the similarity between Eq.~).~. For nonidentical circuits it is better to compute the mutual and self impedance for the individual circuits. (23)).. +.~. is the zerosequence self impedance of one circuit by equation (37) and ~0~~)is the zerosequence mutual impedance between two circuits as given by Eq.+. SD (3% \+re GMD is the geometric mean distance between the 2 threephase circuits or the ninth root of the product of the nine possible distances between conductors in one group and conductors in the other group. +‘“‘oo4657j log” ~(G~IR). (31) ohms per phase per mile (31) where dabis the distance between the two conductors.. Therefore lXq. b..rn) 0. Zerosequence mutual impedance between one circuit (with earth return) and n ground wires (with earth return) Figure 22 shows a threephase circuit with n ground Ql l Q: Q: 0 a B = r. following the same method of derivation 20= ~+0.. is the geometric mean distance between the two sets of conductors or the ninth root of the product of the nine possible distances between conductors in one circuit and conductors in the other circuit.. This equation is the same as ~(z~+z~~. and the phase conductors a.. .. Since zerosequence current consists of unit current in each conductor or a total of three times unit current l’or the group of three conductors..00477j+~0....t. In Fig. 20 in which GMR is the geometric mean radius of one set of conductors.oo477j+jo. ohms per mile..b is replaced by the GMD or geometric mean distance between the two groups. z... This .b in Eq. (G. if the ground wires are considered as one group of conductors....~.L~~~I~pedance derivccl equation in (35) or ~(GMR)3conductor wires. (dabdbcdca)’ feet.*. rC is the resistance per mile of one phase conductor... 22A threeconductor lJ/GMR D. GMRclrcult= ~(G~IR)R..\ID) (39) G&IRc. (38).. This equation can be applied to two groups of conductors if d.. and using $(~o+zo~. (36) must be multiplied by three to obtain the zerosequence self impedance of three parallel conductors with cxth return..d.J where z. (35) Substituting GMR.. (~(GR/lR)conductor(GMD)2separation and GMD ).O0477f Fig.(. Equation (31) gives the zero sequence mutual imbetween two conductors: Z()(m) =o.t. Zerosequence self impedance of three parallel conductors with earth return Equation (36) gives the self impedance of three parallel conductors with earth return and was derived for a total current of unity divided equally among the three conductors.t= l’(G~IR)COndllCtOT(G~ID)fseparntlon rcet.Daratlon istheGMR..=~+o.) is z...o0159j from equation (35) in equation D.d.o1397j = log.. =SG&Zii feet.rcu. Therefore. are considered as the second group of conductors. the voltage drop for zerosequence currents is three times as great..m=w log10 eOndllCtOt. (36) In equations (34) and (36).
by I. (30) and the GMD between the two circuits for d. As stated before. The following derived equations are those most commonly used in the analysis of power system problems. Two circuits. Part X. Writing the equations for E. Equation (43) can be expanded to give the zerosequence impedance of a threephase circuit with n ground wires (and earth) return..as a ground wire. consider the singleconductor.006985f log. First.a+IbZm (41) Eb = I&n+IbZbb. in meter ohms for 25. (46) It is now possible to write the previously derived equations for zerosequence self and mutual impedances in terms of ra. .. divide E. are necessary that result from the use of the earth as a return path for zerosequence currents. 20 the zerosequence self impedance of a single conductor.O0477f ohms per phase per mile. ‘X. and Eb. singlephase circuit lvith earth return and one ground wire with earth return. zO(n) zerosequence self impedance of the three= phase circuit. ZeroSequence Reactances. zo2w 20= ZO(Il) (44) zwa Where zo= zerosequence impedance of one circuit with n ground wires (and earth) return. Referring to Fig. zocapj zerosequence mutual impedance between the = phase conductors as one group of conductors and the ground wire(s) as the other conductor group. These equations can be further simplified to make use of the already familiar quantities r. 2. (41). and the result is 2 Za=Zaa~ Zbb Fig. Practical Calculation ance of Aerial LinesIn of ZeroSequence Imped . general method must be used to obtain the zerosequence impedance of a particular circuit in such arrangements. The quantities r..44 Characteristics of Aerial Lines Chapter 3 General Method for ZeroSequence Calculations The preceding sections have derived the zerosequence self and mutual impedances for the more common circuit arrangements both with and without ground wires.. xd. 4. The quantities r. zd are given in . (43) when zerosequence self impedances of singleconductor. p. singlephase circuit with one ground wire (and earth) return is therefore defined by Eq. Equation (44) results in the equivalent circuit of Fig. we have: E. Solving these simultaneous equations for F of the desired circuit gives the zerosequence impedaice of that circuit in the presence of all the other zerosequence circuits. This general method is shown in detail in Chap. (42) for Ib and substituting this value of lb in Eq. 20 conductor a is considered as the single conductor of the singlephase circuit and conductor b will be used as the ground wire. 50. For more complex circuit and ground nire arrangements a ohms per phase per mile. singlephase circuits are substituted for zlla and %bb and the zerosequence mutual impedance between the two conductors is substituted for z. These values are given in Eqs. zotej=zerosequence self impedance of n ground wires. (48 If we assume conductor Z. = Ia&. and the zerosequence mutual impedance between a single conductor and another single conductor with the same earth return path was derived. these equations can be applied to multiconductor circuits by substituting the circuit GMR for the conductor GMR in Eq. r.. (43) The zerosequence impedance of a singleconductor. Referring to Fig.. and 2. za. 23Equivalent circuit for zerosequence impedance of one circuit (with earth return) and n ground wires (with earth return).=O. the preceding discussion a number of equations have been derived for zerosequence self and mutual impedances of transmission lines taking into account overhead ground wires.the tables of Electrical Characteristics of Conductors and Inductive Reactance Spacing Factors.b in Eq. and x8 are given in Table 7 as functions of earth resistivity. Therefore solving Eq.6655 X 106Eohms per phase per f mile. and xd.. (30) and (31). and x. r. (40) Zerosequence impedance of one circuit with n ground wires (and earth) return. 23 for determining the zerosequence impedance of one circuit with n ground wires (and earth) return. To do this two additional quantities.. The general method consists of writing the voltage drop for each conductor or each group of conductors in terms of zerosequence self and mutual impedances with all conductors or groups of conductors present. ze = 0. and 60 cycles per second. Ground wire conductors or groups of conductors have their voltage drops equal to zero.. xn. (31). then & = 0 since both ends of this conductor are connected to ground. They are derived from Carson’s formulas and can be defined as follows: r. one with two overhead ground wires and one with a single overhead ground wire are used to show the details of this more general method. To obtain z.
llt~cll zerosequence impedance hetween two circuits elrrth retlrrn) but without ground wires Z. 4.)+Zd~bb.)+Zd(nc))+2d.ndt..665X lo@ 1 log” (G~R)~.013!~i”f g(Gh’IR) COnd”Ct”T(G~ID)*sepnrat..+j~(0.. =xd (with z0(gj = %.+jO. (40) 20~~~) 1 IUg:. z..cd) n .j+.Darat..OO6985f loglo 4.=r.6G5X 106p = I jO.4nwlJ zocgj =3$+r.~ circuit with n ground wires %:I =~+r. Zerosequence mutual impedance between one circuit (with earth return) and n ground wires (with earth return) %crosequence self impedanceone rdurn) ground wire (with earth L^..) 0. dxy.) al*@&9 ZO=ZO(n).665GX 10” : h4) (44) oilms per phase per mile (50) \Vllere Q=xd from Table wires. (49) G).00477f (30) D. at/e..+*“o..(...00477~+j0.on ohms per phase per mile..+j(~~+x.+j0. ~0~~~~zerosequence mutual impedance between the = threephase circuit as one group of conductors and the ground wire(s) as the other conductor group..t.. ____ +jO.)+Zd(cn.+r.&w&r .3 +$“.) from Table 6 for spacing czto h.)+5d(cb.6G5GX 106: +jo.)(..) = ohms per phase per mile.O06985j log. zoCg) =zerosequence self impedance of n ground wires.+j(x. Zerosequence impedunceOne (and earth return.. etc.00477f+j0.OO6985j log.. 6 for spacing between ground where zOCn) =zerosequence self impedance of the threephase circuit.+r.$ =37..+j(z.00477”r+j0... ($Tagldb&cF1..) +Zd(bc.dngnd.01397~ ohms per phase per mile..+j(z.kd) ohms per phase per mile (52) where Zd=(Zd(nal)+xd(bgl)+2d(cgl) 1 3n =w = ~+0.OOG9SJ” log..O06985f log.4.4kJ .\[.6656X lo@ f +“‘oo6g85f 1og10(Gh~R)2.du.+r. GiLID* ohms per phase per mile %ll) = r.01397~ ohms loglo D... per phase per mile..dg2gn) (dg3gdg3q2 ....+jO..O139if loglo 3n V &.2x.+3x. 4.IK) = 0.ba.gndcgn)’ Zocae)=r..nd...~U.dp3gn) (4mdgn$!z. ~/(G~IR).o sequence self impedancetwo return) ground wires (with earth = r.UUUYO. GMR =n&GNIR)n c”r.00477~+j0.2794i log.oJ ohms per phase per mile (47) f ~. ohms per phase per mile.duetor(dalrr’L(~glgR (b&~R .d..2i94)$ .+0.j(x. = 3.zO(..004i7J” +j0.0069S5~loglo 4.Chapter 3 zerosequence impedanceone but without grolrnd wires circuit (with Characteristics earth return) of Aerial Lines self impedancen ground wires (with 45 earth Zerosequence return) z.ijx.(.01397~ = ohms per phase per mile log10 & (30) where r0=T” ohms per phase per mile..ijO. a~(~) 3r.. = rc+0. a... GMD...+0.d”~t.) (43) or xd= ~ 2 (sum of xd’s for all possible distances n(n1) between all possible pairs of ground wires).1) \vhcre :tncl l’d = +(xd(uh) +Zd<hc)+Zd(ea)) :rCICCri. 4...? 20~~) 3r.. . z.) r.2ig4)~ f log.6656x lo6 p f jO.= r.J log10& conductor . ~ D..2794& 1 log10  1 GMR COlld”COX j2(0.)+2d(cc’)) n n ohms per mile per phase (33) where xd= (51) sum of 2dsSfor all possible distances between all ground wires.01397jIoglo G&D ohms per = phase per mile.iuctordxy (32)  +xd(nCn)+2d(bsn)+xd(cyn) 1. (37) J log.3zJ Where zd is +(xd(aal) fZd(ah.+r..
Applying Eq.. (54) and (55) when both charges ql and q2 are present. regulation. . (62) The derivation of shuntcapacitive reactance formulas brings about terms quite analogous to those derived for inductive reactance. and zerosequence shunt capacitive reactance for the more common transmission line configurations. and Zerosequence Shunt Capacitive Reactance The capacitance of transmission lines is generally a negligible factor at the lower voltages under normal operating conditions. IJse of capacitance in determining the performance of long high voltage lines is covered in detail in Chap.” Capacitance effects of transmission lines are also useful in studying such problems as inductive interference.+ * megohms per conductor per mile 031) where D12 and r are in feet and j is cycles per second. Also if the charges on the two conductors are equal and their sum is zero. (62). 24 assuming conductor 1 alone to have a charge 41. these terms can be resolved into components as shown in Eq. power factor. For these reasons formulas are given for the positive.=x:+x. 2*4*5.. megohms per conductor per mile. corona. However. mile. (61) may be written x.46 Characteristics of Aerial Lines Chapter 3 4.the voltage between conductors 1 and 2 is VIZ = 18X lo9 q1 In *$ volts.. This can be written as X en=0.+0. C.=O. and transient’s on power systems such as those that occur during faults. (57) VI2 = 36 X 10gql In 5 The capacitance between conductors 1 and 2 is the ratio of the charge to the voltage or 1 +c. (54) log. and as in the case of inductive reactance. (53) and the principle of superposition to the twoconductor. Therefore VI2 is the sum of Eqs. singlephase circuit of Fig. 9. With both charges q1 and q2 present. on the conductor surface is assumed to be uniform. (55) 12 This equation shows the work done in moving a unit charge from the outer radius of conductor 2 to conductsor 1 a distance D12 meters away through the electric field produced by q2.= 18X log q In s volts (53) where q is the conductor charge in coulombs per meter. the voltage between any two points of distances x and 2/meters radially from the conductor can be defined as the work done in moving a unit charge of one coulomb from point PZ to point P1 through the electric field produced by the charge on the conductor.= (59) 18X log In *f The shuntcapacitive 1 ductor) is xcn= 2rjc reactance to neutral (or per conor in more practical units megohms per conductor per (60) CONDUCTOR I CONDUCTOR 2 x.O6837 log. (58) 12 36X log In *f The capacitance to neutral is twice that given in Eq. “Regulation and Losses of Transmission Lines. 41~0. the capacitances of systems of parallel conductors can be determined. by the principle of superposition the voltage VIZ is the sum of t$he voltages resulting from q1 and q2 existing one at a time.. Now assuming only conductor 2. This is true because the spacing between conductors in the usual transmission circuit is large and therefore the charges on surrounding conductors produce negligible distortion in the charge distribution on a particular conductor. ql+q2=0 Substituting or q2= Ql ql for qz in equation (56) r volts. VI2 = 18 X lo9 q1 ln h+qz r ln $ 12 VIZ= 18X lo9 @ In $ volts.) This is given This equation shows the work done in moving a unit charge from conductor 2 a distance D12meters to the surface of conductor 1 through the electric field produced by ql. For a more detailed analysis of the capacitance problem a number of references are available. 24A two conductor single phase circuit (capacitance). Positive. Negative.in the case of a single isolated charged conductor. lightning performance of lines. 1 farads per meter. By use of this equation and the principle of superposition. the voltage between concluctors 1 ancl 2 is volts. The case of a twoconductor. In deriving capacitance formulas the distribution of a charge. Eq.. The term xa’ accounts for the electrostatic flux within a one foot radius and is the term . and voltage distribution under normal operating conditions. 24. q. (58) because the voltage to neutral is half of VIZ. singlephase circuit is considered to show some of the fundamentals used to obtain these formulas.= farads per meter.0683: Vx. negative. it becomes an appreciable effect for higher voltage lines and must be taken into consideration when determining efficiency. having a charge q~.. (See Fig.0683? log. *f Fig.
(67) (d) ZeroSequence xc of two ground wires (and earth).0683 y log..’ is given in the tables of Electrical Characteristics of c*ontluctors. 5.) +xd’. ShuntCapacitive Reactance.1 =3x:(. stability and other factors may determine the choice of a conductor for a given line. = (e) Zero Sequence xc of 11ground wires (and earth). 3n +xdl(bqn)+x. c) xc. =&L+x.. .’ taking into account the . zd’ is given in Table 8.c+ zdbc). The maximum continuous current rating is necessary because it is determined by the masimum operating temperature of the conductor. (W xd’=value given in Eq. For more than one mile of line. 63) xd’= x~‘(~I~?) xd’ for distance between ground wires. xd’ and x:d. (64) 1 xJ=$sum of all three xd’s for distances between all possible pairs).) =x:+?x:“(“l) xd megohms per conductor per 71 n mile (69) . . a. f =frequency in cps. the maximum continuous currentcarrying capacity may be important in selecting the proper conductor. b. and x...0683 s log. (g) ZeroSequence xc of one circuit with n ground wires xO)m2 megohms per conductor per mile. xd’ taking into account the electrostatic flux external to a radius of c.. Q’(~.l megohms per mile. and xl is given in Table 9. (61). (Conductors a. x’ = x1 ++x...=x. See Table (8) (65) (1)) ZeroSequence xe of one circuit (and earth). (71) xd = XII’(S) XdW Shunt Capacitive Reactance.. voltage regulation..clectrostatic flux within a onefoot radius. (72) xd’=xd’ for spacing between conductors. in Eq. in capacitance calculations the only conductor radius used is the actual physical radius of the conductor in feet. _ 12. of ThreePhase (f) ZeroSequence xc between one circuit (and earth) and n ground wires (and earth) X6(. 47 0..)+x./) megohms per mile of circuit. xi = xi =x:+x: megohms. The following equations have been derived in a manner similar to those for the twoconductor.(cgn) 1.) = zd . (61).+~J(~) xd’ megohms per conductor per 2 2 mire.and transmissionline design the temperature rise of conductors above ambient while carrying current is important. (70) 1 xd’= (Zd)(na. (Conductors a and b) xc. For short tie lines or lines that must carry excessive loads under emergency conditions.I~.= x:+x. the shuntcapacitive reactance as given by the above equations should be divided by the number of miles of line. ZeroSequence ShuntCapacitive Reactance Factor. wind velocity.l(.30 log. ShuntCapacitive l&ctance Spacing Factor.~+x.) mile. singlephase case.. The following discussion presents the Schurig and FrickC formulas for calculating the approximate currentcarrying capacity of conductors under known conditions of ambient temperature.(.. rractance.‘.. It is a function of the conductor outside radius only. (65). This temperature affects the sag between towers or poles and determines the loss of conductor tensile strength due to annealing.’ taking into account th(: flux external to a radius of one foot and is a function of the spacing to the image conductor. Table (9) gives z:.. (‘73) 04 In using the equations it should be remembered that the shunt capacitive reactance in megohms for more than one mile decreases because the capacitance increases. The term zd’ accounts for the electric flux between a one foot radius and the distance D. it is sometimes necessary to consider the maximum continuous current carrying capacity of a conductor. Circuits (a) Positive (and negative) sequence xc.‘256 megohms per conductor per mile.)~~~+2xd) megohms per mile of circuit. x.) +xd’(i. and limiting temperature rise. x’=x. 3 3 2 o’(..? to the other conductor and is the term 0..‘(. of SinglePhase Circuits (h) xc of singlephase circuit of two identical conductors x’=~(x:+x.g. making use of the terms x:. Note that unlike inductivereactance r \vhcre the conductor geometric mean radius (GMR) is .. They are summarized in the following tabulation.32: megohms per conductor per mile. 2h megohms per mile per X0 f conductor (63) \vhcrc h = conductor height above ground.:.Chapter 3 Characteristics of Aerial Lines where sum of all xd’s for all possible distances % xd’= n(n1) between all possible pairs of ground mires) orxd’=l (sum of all xd’s for all possible distances n(n1) between all ground wires).scd. megohms per conductor per x:(..per conductor’permile.. divided into compocents 2. Zerosequence capacitive reactance is. of one conductor and earth. The basis of this method is that the heat developed in the conductor by 12R loss is dissipated (1) by convection (j) x.. ‘. like inductive. Conductor Temperature Rise and CurrentCarrying Capacity In distribution. Dl2 in Eq. (c) ZeroSequence xc of one ground wire (and earth).nc foot out to a radius D feet. While power loss. therefore. * * * +r. (i) xc of singlephase circuit of two nonidentical conductors a and b.
AMBIENT 30’ 40’ TEMPERATURE 50’ ‘C 60’ Fig.5 ft jsec). The above general method can be used when test data is not available.” The tables of Electrical Characteristics of Conductors include tabulations of the approximate maximum current lOOOr 900 000 700 I EL’ ” : I 60. it should be used. (Copper Conductors at 75 “C.). 50 \ Y I . Tests have shown that aluminum conductors dissipate heat at about the same rate as copper conductors of the same outside diameter when the temperature rise is the same. To= (degrees Kelvin) absolute temperature of surroundings. R = conductor resistance per foot.0128~/pn c 7’.ion. The value of R to use is the ac resistance at the conductor temperature (ambient temperature plus temperature rise) taking into account skin effect as discussed previously in the section on positive. d = outside diameter of conductor in inches. 25Copper conductor current carrying capacity in Amperes VS.+ W. 26Aluminum conductor current carrying capacity in Amperes VS.d watts.” or 0. wind velocity at 2 fps. The watts per square inch dissipated by radiation. The effect of the sun upon conductor temperature rise is generally neglected. 4. in general. Ambient Temperature in ‘C. I I I I I I I I / I I I I I I I I I I Chapter 3 I I I I I I in the surrounding air. WC. TV.or surface area in sqllare inches per foot of length. T. The lvntts per sqllare inch dissipated by convection.and negativesequence resistances. IV. This small effect is less important under conditions of high temperature rise above ambient.). Ambient Temperature in “C. = watts per square inch dissipated by radiat. 0’ IO’ 20. v = velocity in feet per second. (75) \vhere I = conductor current in amperes. being some 3” to 8°C. = Irntts per square inch dissipated by convwtion. or to check test results. and (2) radiation to surrounding objects. can be determined from the following equation: w ~0.3 inch to 5 inches or more when the velocity of air is higher than free convection air currents (0.0 for atmospheric pressure). At = (degrees C) tcmpcrature rise. applicable to both copper and aluminum conductors.= (degrees Kelvin) avcragc of absolute temperatures of conductor and air.. (Aluminum Conductors at 75”C. wind velocity at 2 fps).. and R. By calculating (W.0 for “black body.=3+(&~(&)P] watts per square inch 0’ IO’ 20’ AMBIENT 30” 40’ “C 50’ 60’ TEMPERATURE Fig. This can he expressed as follows: I?R= (W. (75).+W.12”& At watts per square inch (76) \\here p=pressure in atmospheres (p= 1. Where test data is available on conductors. T = (degrees Kelvin) absolute temperature of conductor.5 for average oxidized copper). (77) nhere E = relative emissivity of conductor surface (E= 1. A = conduct. .20.. can be determined from the following equation: . It’. This method is. This formula is an approximation applicable to conductor diameters ranging from 0.).48 Characteristics of Aerial Lines 1000 900 800 700 I II I I I I I I .0. it is then possible to determine I from Eq.
710 .1038 ~0. 2.09470.0630 Oli30’0.350 0.lH930.09.406 0.5).374 1.503 1.017500.10~00.18450.14210.07520. hollow copper.0977 0 241 0.019020. The conductor data (r.01113 0.0343 71.214 0 2Si 0.7 TABLES OF CONDUCTOR CHARACTERISTICS The following tables contain data on copper.75 90 0 00417 3.303 0.00488 2.) Resistance per Conductor per Mile 50°C.071 0.08420.Chapter 3 Characteristics of Aerial Lines 49 TABLE ~CHARACTERISTICS Pizc Of Conductor reakirlr trc!nRtl ‘OWKIS OF COPPER CONDUCTORS. permit the determination of positive.607 0.256 0.128 5 at 73”C.955 0.08070.19530.376 .I2 1 706 706 891 801 076 076 610 610 540 540 480 490 480 420 2 2 1 1 736 17c 720 364 420 360 310 270 0 019870.” 000 1.029 :70.440 0.144:10.) 50 cycles =I ycles lnductivc Resctnncr Ohms per Conductor Per AIile At 1 Ft.339 .276 0. tarnished surI’xc (E=O.555 0.414 7 0.11880.018130 234 0 234 0. .llfi?O 814 !J 0.1004 0 241 0.607 . weight.18310.144 L! 0. 170 020 .. OOII .2?& 0.1281 0 142: O.214 214 :256 .1821 0.083 .0. These thermal limitations arc based on continuous loading of the conductors. Spacing 25 cycles ( 50 cycles 1 60 cycles 0. 16220.it 1 Ft Soacinv 25 cycles 0. The crosssectional inserts in the tables are for ease in III .19530 19610.39 1. 1487 0 :320 1 0 2.11750. r.945 1.21 1. and an air velocity of 2 feet per second.. 26 have been calculated to show how currentcarrying capacity of copper and aluminum conductors varies with ambient temperature assuming a conductor temperature of 75.077: 0.440 .381 .440 0.08590.01.0208 0 19300.106l 0.15601.9 0 12SiO (i2!1 20.19610. This gives a conductor total temperature range of 35°C to tOO”C.699 0.384 0.18450.16700. 2 433 2 350 2 430 1 879 970 505 591 205 1 1 1 1 858 8JO 841 674 667 534 s29 424 420 333 264 1 280 1 030 826 wind 120 0. JAI Shunt Capacitive Rcactnnce Mrqohms per COIldUCt.18790.014040.302 0.102 1..422 0.0975’0.hc conductor manufacturers.204 30.13030.184 .458 0.110 001) 51.081~ 0. and zerosequence impedances of conductors for use in the solution of powersystem problems.54 0.277 0.606 0.0821 . Copperweldcopper.277 0.349 10.08i.349 0.12050.16660.03G8 0.360 .c.108' 1283 0.355 0.895 2.117.0. which along with the previously derived equations.067 .11880.L .1005 0.882 3001 6701 0401 2301 300 220 130 090 040 940 840 810 0.0731~0.278 0.400 0.41 3.4!)2 710 15480.235 0. Such data on copper conductors has been presented rather thoroughly in the technical literature.06200.17220.06480.303 .328 / 30.349 0.382 0.349 0.235 0.” 000 .765 0.3. I 2.481 0.303 .22!1 :30.590.303 .f.1961 0.11320.50. HARD DRAWN.368 iO.692 0.'C ) I20 .914 1.!I 0.278 0. the conductor manufacturer should be consulted for test data or other more accurate information as to conductor temperature limitations.100f 0.733 2.10810.22rj '!I i o.12060..251 .io.214 0.445 0.0583’0.19180.0934 0..349 0.382 .518 170 150 140 130 0. For design purposes copper or ACSR conductor total temperature is usually assumed to be 75°C as use of this value has given good conductor performance from an annealing standpoint.278 0. :.(i37 .11000.0256 0.05940. li For conductor air at 25°C.19500.O 12430. and currentcarrying capacity as limited by heating.085: 0. i26 .143 :1603O.129 14260.20.119~ de 0. ACSR.58 7 0 08li70 260 .417 0.4. negative.0641 0.11730. ::. 2.!I” IA80 k2 Ii00 4.07870.015700. For those lines where a higher conductor tem perature may be obtained that approaches lOO’C.119~ 0. 97.!~0.00590 0.276 0. 1.357 0.245 0 24!f 0.o!i35i%: 1 1 1 0. LI 12 0.‘) along with inductive and shuntcapacitive reactance spacing factors (zd.TriI 2 0.481 0.069. I lli7 l)OO 000 000 000 Ol1” OIIO 000 000 0.’ The ambient air temIwature is generally assumed to be 25°C to 40°C whereas 1ho temperature rise is assumed to be 10°C to GO”C.381 0.256 0.“I.1281 0.13030.09970.1381O.18310.11700.lciOl 0.257 0 303 0..51 0.0.277 0. These conditions were used after discussion and agreement with t.w7 I I Ii0 13750 9lX li 0.11800. y.344 08880.381 .235 0..0.0260 0.10710.360 7 0 09740 292 3 0.276 0.699 0.162 0.18 110 0. These values are conservative and can be used as a guide in normal line d esign.371 .54 0. 9 (122°F. and Copperweld conductors.125:30.277 0.07600.06850.083610.213 0.382 0.964 0.234 0.555 0.0801 0.00992O.15390.07800.235 ...349 0.‘J 0. Using Schurig and Frick’s formulas.07180.18450.1644 1 151 170.52 . Fig.08370.0. the limit being about 100°C where annealing of copper and aluminum begins.14701. .348 .11830.811 I i !)0.1020 .0901 0.5.lO'J30.13250.091 0. The technical literature shows little variation from these conditions as line design limits.iO Z!.302 E"8? 0.56 366 926 752 804 620 04s 913 003 lOi.11450.d.600 9 i ? 2 11 27 ?2 21 19 17 15 15 :..1?240.4 miles per hour carrying capacity based on 50°C rise above an ambient of WC.220 0.12280.09810.17590.184 .0319 0.58 .‘) permit easy substitution in the previously derived equations for determining the symmetrical component sequence impedances of aerial circuits.0329 0.234 0.406 .14430.00663 0.699 0.08000.II L’. zd’) and zerosequence resistance.I.555 0.770 .016680. .277 0.17080.00638 0.01 3.11210.11700.” IO0 :i.196~ 0. 4/c I 7 0 173'30 52% 800 3.08610. 25 and Fig.091F 0.00526 2. 0.161!10.3 PERCENT CONDUCTIVITY ‘a Ohms 25°C. inductive and shuntcapacitive reactance factors (r.00568 1.tiSO 0..jo 21.213 0.0741 . 2.0308 0 0285 0.l37!~0.440 0.X!ll 17 ll.285 1 0.o L’.757 0.256 .1044 389 ii % 360 11 130 9 9 9 7 7 5 4 3 3 3 2 3 617 483 1.07390.182. z.il 0.092 17 0.510 590 750 3fiO 590 140 7 li 5 5 4 4 4 4 336 .46l 7!O.413 0.121i90.235 0.303 0.iO 528 Iill" l/C . 3 0.09200.432 0:143 0.216 1.06R80.18530.10950 17990.460 ) 2. 0.54 5.C and wind velocity of 2 feet per second.277 0. 14.333 . fai!? .0349 0. 12'9 :12830. 1.254 1 .229 0.369 .2. Also tabulated are such conductor characteristics as size.Or Per Mile .80 frequency=60 cycles.19430 .1147O.0.19090. (75°C total conductor temperature). I273O.06820.47 (2 it/xc).226 1 50' cycles 1 60 cyclen ~ Ii 0. 5.481 .750 1.17390.012520.087 0.350 0.lGl . (77°F.10.204 1.216 0.
1006 0 12480.112t .1057 0.38 . 3190 0329 6160 0335 2130 0351 lOQO.16180.M)U 6100.17000 17200. .415 0 421 0 430 0.1327 1 6 10.1908 0 1883 0 1936 380 385 390 393 395 401 399 393 407 405 399 412 148 0.21i .24 24 ii 0.1730 O.55 .0039 l Based on copper t For conductor at 1 “Current Approx.12 .09350 09570.li51 0.14520.1836 0.642 806 .14420 .95 2.0.1790 O. .11511 03fi 0 1659 i0 12901 051 0 1.24 .Ol .0493 1600.1728 0 li54 0.773 0 974 1.li66 0.5 50026 71550030 66660054 3 0.0834 4 0 118223 2 1.E .184 18 2 1 75 840 WO 49 278 11 340 (1) Electrical Characteriitics not available until laboratory (1) measurements are completed.4 luiles per hour. rise) with TABLE ~BCHARACTERISTICSOF “EXPANDED” (Aluminum ALUMINUM Company of ‘America) CABLE STEEL REINFORCED Circular Mils ___  Ohms per Conductor per Mile 50°C.4 miles per hour (2 ft/sec).24 .0244 7740.1718 0.901 70 10510. air at 25% wind 1.41 .861 895 .08G21 293 130 1384 70 13841.16720 502 10.230 234 23i 235 244 242 250 248 254 252 0 11490 095i 0 116iO 09i3 0.41 .78 .11820 355 210 l/O I .10 3. Reastanco Ohms per Conductor 25°C.07 .41 .30 00030 00026 00030 50026 50030 2 2 ? i 2 2 0 13fi2 0.1551 0.lM) 4 0.12880.552 .177i 154 0.219 .246 3 0 1329 70 1329 1 196 3 0 1291 70 1291 1 162 3 0 12i3 70 12731.108 0.1353 24 2 1 55 i24 M)o 41 900 9 070 1338 000 66 2 0 1350 19 O. .1291 0 1355 0 1281 0 123fi 0 1151 70 13620 953 70 12910.12190 1015 0 12080.0368 iioo. (25’C.09240. 75% Capacity” is 75% of the “Approx.47 3. i6C.1255100.12880.15410. and is approximately the current which will produce 50°C.12580.56 352 445 560 706 888 12 41 .21i .224 .82 .) Current Approx.168 0.24 .0230 4730.465 228 0 11400 0950 227 0 11350 0946 230 0 11490 095i 232 0 11590 0965 .0321 3910 0327 0390.1664 0 lil5 0 1726 0.485 0.1182 70.12 .146 3 0 1214 i0 12141.081 0 1111 70.13270.11111. (122’F.0355 8650 0372 5x0. (i7”F.18i2 0.214 216 .0068 1 5420 0081 1 2230 0060 9700 0051 769O.11510 806 350 000 314 500 300 000 300 000 250 000 25OMM 40 3i 34 32 31 28 31 38 26 28 34 24 23 25 31 22 24 22 200 100 200 300 400 500 200 400 300 100 600 500 600 000 500 500 100 400 7 7 6 6 5 5 5 6 4 5 5 4 4 4 5 4 4 4 4 4 3 3 2 3 2 2 2 2 1  7770 0520 23io 0507 6990.1693 0.424 0 441 0 435 0 451 0 445 0 458 0 452 0. frequency=6O.55 1.1568 0 0 0 0 0 0 0 0 0 0 0 0 299 302 304 307 310 314 317 321 325 328 3?9 334 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 345 344 339 348 346 350 346 351 358 353 367 362 376 3il 382 3i7 387 359 362 365 369 3i2 3i6 5 0 0 9 9 8 8 111300054 103350061 9.0256 1780.211 .398 10.1039 0.147 .187X0.13080. Current Carrying Capacity m Amps.0479 6210 0465 082 0.82 56 351 444 55s 704 887 .14420. 97 percent.0241 9360 0217 For OXR”t APpro% 75% Capacit) t 380 340 3GQ 250 200 160 110 060 010 970 950 900 900 910 830 840 840 800 I587 IGIE 1652 1691 Ii.12610 883 iO.41 .e 25 des 0588 0619 0653 0692 0735 0784 084C 0905 098C 104 107 118 117 117 131 131 131 140 147 147 147 154 154 168 730 690 670 ii: 600 530 530 490 500 460 168 187 196 196 235 235 278 278 311 311 350 117 117 131 131 131 140 147 147 147 155 154 168 168 187 196 196 S1 306 306 .168i 0.13580 137t .16180 16180 16180 1611 155 0.11450 11550 11750 118: .591 062 065. (122°F.147 .1269 0.78 .0290 9330 0304 8850 0265 2770 0278 4420.033.14900.414 0 412 0 406 0 Iii 0 415 0 420 0. 3/O 9 8 6 5 4 3 645 420 675 345 280 480 __1 8020.0912 10850 0904 11190 0932 11140 0928 11040.50 000 54 2 0. 069.1495 0 1508 0.56: i0 592 69’7 0 723 .1812 0.10780.38534 ooO35371 7 200 1 1.17150 li550.0435 0190..093 2 2 3 2 2 3 0 Ii49 70 13G01.563 10.223 .10880 11180.07030 633 10.12380 1032 0.12 .16780 1681 0. 757” Capacity ~____A 85000054 2 0.514 .612 0.1624 0 1639 0 1646 O.12880 12880.09610 783 iO.27 70 511 .I1780 11880 12180 1221 .386 0.55 1.0041 4840 0041 5660 0050 3840 0043 3040 0043 3560 0045 2410 0041 1910.0920 1132 0 0943 .0420 4iQO 0403 1120 0391 9400 0386 3990. . SPdCillK oflAmerica) j I /right ounda per Mile Ceometric !&an Kadius at GO c&Cl.24 .0450 5440.11860 0988 0 11760 0980 0.1697 0.41 .) Small Currents .858 70.221 226 0 0 0 0 0 0 0 0 0 0 0 0 10400 10530 10680 10780 10830 11000 086i 08i8 0890 0898 0903 0917 10950.78 . 75% capacity: XII Inductive Reartanrc Ohms per (‘onductor per . conductor temp.1628190 097i 1.23 1. 25% air temp.1855 0.18 98 2GFROO 6 IO 2109 4/O G 10 lRi8 3/o 6 10 1672 2/O 6 10 1490 l/O 6 1 0.o35 51iO 03j3 8590 0349 1930. aluminum 61 percent conductivity.\llle at 1 Ft.342 0.442 557 702 885 ..155 0.000 700 000 650 000 600 000 566 000 550 000 500 000 500 000 500 000 450 000 450 000 450 000 419 000 400 OWJ 400 000 400 000 380 500 380 500 350 000 556 5w 4ii 477 397 397 500.1660 0 1637 0.430 .18490.185~ 168 18i 196 196 0.92 .14420.12880.Characteristics of Aerial Lines TABLE ~ACHARACTERISTICSOF ALUMINUM (Aluminum Company Chapter 3 CABLE STEEL REINFORCED ra I I 2” Shunt Capacitive Rcactarlce Negohms per Conductor per !vlilr at 1 I:t.140 0 1151 70.1436190.447 10.1739 0.09GI .cycles.47 2.74 17w 208 .385 50 cles 0.17200 li2( 168 0.1409 19 0 0921 4 0.) Current Approx.” wind 1.18690.16180 16180 16180 1611 ::: 0.511 .148 F .15710 0 1585 0 1603 0.220 . 073 0781 084’ 090 098 104 107 118 per Mile 50°C.0313 5880 0328 1220 0311 4620.18590. Sparmy Ail (‘urrcnts 25 cycles .14420 14420 .1049 ~ 27 200 24 400 19 430 23 300 16 190 19 980       Single Layer Conductors ~ 460 340 3w 270 230 200 180 180 160 140 140 120 100 351 441 556 702 885 12 351 .342 0.154 0.1522 0 1536 0.16380.16950.54 00054 90000054 Si45005I 79500054 i950002G 795 00030 715 50054 il.lGiO 0.69 65 .544190 092Gl.
.020 0.293 12.3. 25 .12590.919 0..0891~0. Shunt Capacitive Resctance Alegohrns per Conductor per Mite at 1 Foot Cable Corporation) gy.) 150°C.53 0 3f3.126* 0.123* 0.&Oe..pare cycles cycles ..s i).395 0.W .1401 0.OL’!J2 205 .0406 0. (77°F.5t 60 cycles 2: F.0.307 0.389 0.0638 0.368~0 1140 D 27.45.0408 0.0387 0.0974.x .0752 0.2fi8 1 100 1 020 1.683 0.359 0.036 1.59 0.500 0.0360 0.820 l.: Geometrent ric CsrryMean ing Radius CaparFeet ) .1576’0.o!l4t 0 (J9Xt 12 920 12 760 I2 120 310 10 500 9 692 8 884 8 270 8 8 8 8 7 7 6 6 5 5 4 4 4 4 4 3 07G 074 068 063 268 286 460 458 653 S50 845 843 037 036 034 4.1124. 1033~0.50 400 300 _0.218 1. 0.1214 0.O.0742 10.260 0.317 1.217 1.13.478 1. Shunt Capacitive Krnctnnce ~IrRotlms per Conductor .llG1~0 O!JSS 10.0962!0.P y2iF’).0376 0 0373 0 0353 0. miles p6 :r h 1requency=60 TABLE QBCHARACTERISTICS OF GENERAL CABLE TYPE HH HOLLOW (General APBreaking Strength _ “. Resistance Ohms per Conductor per Mile 25°C.203 0.x. air nt 25°C.650 1 15. wind 1.i I 0 500 450 500 non 000 no0 000 000 000 000 __28 42 50 50 50 50 18 21 21 22 21 6 22 18 13 12 18 15 1% 18 15 14 16 15 12 15 14 13 14 15 12 I8: pi.500 0.0518 0 O182 0.388 3.5 ii.sg 3.12600.332 0.XJO 300 500 .o.280 0.266 13X5 0. 0.i:o on0 .1575’0.i 7YO O.5 ‘0 1047’0 0872 ‘0 109810 0915 ~0.0.34 0968 0995 0.0961 0.119* n.197G 0.930 0.0165 XI.302 0.1049 3.725 0.0233 0.0494 0..014 1.1576’0.1833 0.z 0.131t 0.5 IO.1353 0.335 i0.423 0.1259 0.324 10 1864 0.1%7’0 1O.282 0.338 iO.1057 O.354 0.096* 0 091t 0 OYl* 0 08fit o.38 0.238 0.19690.281 0.280 0.606 0 59.371 0.0937 .400 .1100 0.38’0.0. average tarnished surface.325 126.3* 0.ti50 1.238 0.X .5.50 000 000 ti. 1375 0 1146 0 276 0.1131 0 2iL 0.1260.389 0.50 000 000 300 300 300 500 600 600 9.0243 0. 0 210 000 000 000 on0 on0 000 no0 000 000 000 000 000 000 000 on0 .0583 0.0355 7YO 0.30 1630 0.0907 0.207 0.0879’0.o.5 0 0992~0.404 0.4x 0 44: 0.l88!J 0. Inductive Rcactsnce Ohms per Conductor prr AIile at 1 Ft.262 ‘0. tThickness uniform throughout.0428 1390 133. 0.x0 .223 0.131GI0.259 0.0 .0373 0. .10890.400 nan :m :wo :1no 50 L’.0665 0.1400 0 1187 and .44c 0. lnR5 X2.650 0.57 IO.1734 0.48’ 1412 1017 1621 lfi14 16B3 ID81 16.1787 0.j I:! 000 .1357!0.1810 0.5 0.0932 0 0777 0.281 0.1558 0.0612 0.393 0.13780.21.13100 1091 i).074 1.21Y 0.0!)70 0.1152 0.12600 1258 0.32G 0.404 0.500 0.50 0 3. 1285NO. .1748 0. 75°C.007 1.0478 0.0551 0.287 14G6 0. (122°F. Ii30 0 7*33 0 608 0.232 0.400 1.0722 0.12380.0311 0.126 1.1127 0.1575’0.0950 0.lOOt 0. 1574 0. 100.003 0.5 1329 0.J15* 0.24.76fi 0.X 0.1943 0.323 0 324 0.277 1351 I 0 270 i370 0.282 0.iG 3 252 . 0.lY28 0.19640.1876 0.‘920 0 839 0. (1) Conductors of smaller diameter for given crosswct. 0.&J 330 321 300 300 300 250 230 230 .530 0.0347 0.57 0.281 0.: 0.1053 0.378 0.0615 0.014 1.0 1.50 fion 5. 0.474 0 481 0.380 0.gj no0 000 000 000 000 000 000 000 000 0.0181 0.o 230 “1t I 0 :i.3% 0.lOR* O.750 0.274 1304 0.375 0. diameter expansion is possible.0722 0.306 0.239 3.122~ 0 1022 2.50 R.) ra cycles x9n 7.“Ctor C.08ZZ7 0.914 0.‘CT 1Iile nt I Ft.21’3 0.430 0.128 1. in the naught sizes.46f 0.263 0.686 1 610 1. 198. I 250c. . I009 0.232 J.39. .0384 0.4.333 0.0179 0.4. cycles cycles .663 0.5 1.252 10.O84.232 0 L’Y3 .1113 :X271 0.307 0.102* 0 096t 0.230 0.0913 0.0945lO.282 0.0191 0.1782 0.391 23 / I 30 cycles I GO cycles 400 000 .1757 an0 no0 790 “00 750 000 700 000 fid0 no0 I ix10 000 .0764 0.0996 0.115* 0 lost 0..188 1.5 978 937 9 1.0184 0.1444 0.314 0.5 ‘0.34.0394.0917 0.1805 ~ 0.320 0.1.lll* 0.130t 0 144t O.4 miles per hour (2 ft/aec).730 1.1005 0.1507 0.306 10.nrl? dS COPPER CONDUCTORS I 2.103 1.0939 0.1097 0..0867’0.0836 0 0936 1.0811 0.ional area also available.24~) ‘0 1~40 1038 J.000 1.0180 0.515 0.“.0861 0.1891 0.0420 0.0393 950 0 0399 950 0.353 0.0.0802 0. 1241’0 ~O.332 10 itJo 0.0.1200 I0 12fiOlO ’ 12390.137* 0.400 0.434 0.34 34 32 30 28 2.1149 0 276 . 12580. Spacing HOLLOW Company) c heaking strain Pounds Weight 1 Pounds P‘er Llile .1368 0.“.39 YlO 864 838 790 764 709 687 ti26 606 694 524 110 0.Chapter 3 Characteristics of Aerial Lines TABLE QACHARACTERISTICS OF ANACONDA (hnaconde Wire & Cable 51 COPPER CONDUCTORS 2.0357 790 0 0Xl.470 wind 1 38 34 29 27 33 22 ?I 19 17 16 IS 13 16 13 13 13 13 10 11 11 9 9 9 7 7 7 5 6 6 5 5 5 000 200 500 .339 0 388 0. frequency=60 cycles.0398 6 6 5 5 5 5 4 4 4 4 4 4 3 3 3 2 642 331 813 7i6 739 343 9x4 Y53 9. some additional (2) For conductor at.282 1437 0.352 10. 1278iO.a13780 1149 O.202 0.1710 0.0903 0.0788 0.310 0.818 0 76fi 0.o.1‘400l0 1575 0.0214 0.2Y4 0.5 0 560 0.119G 0 1059 0.0.239 0.2..1234.0953 0 0794 0.2Y3 0.1834 0.382 0.329 0.399 0.1879 0 3.1097 3 208 0 13.0443 110 0.59 0.482 0.1947 0.736 0.8GO 0.386 0.700 0.376 Notes: *Thickness at cd&es of interlocked segments.0646 0 0691 0.342 0.OY14 0.2.1338 0.1227 0 1211 0.:.390 I . 0.367 0.i CJ.0885 0.367 0.1194~0 0..248 1.1183.1010 i:: 4/O 3/O 310 $1: 30 2io 125 600 121 300 119 400 at ‘0’261 0 216 0 260 IO.1123 0 0880 0.20!) 0.362 0.237 I.itycZ’ Ohms ra Resistance per Conductor per Mile Xfl Inductive Reactance Ohms per Conductor per Mile .5 1.269 0. air at 25°C.252 0.237 0.558 1.z 0 2..1113 0 1205 0.1109 0.080 I.336 0.0340 0.797 0.soo 200 700 400 300 200 300 100 400 100 850 100 200 050 IJ.1833 0.1012iO.218 0 225 ‘0 1014lO.io 430 .260 10.1400 1400 0.0289 1395 llG0 I rl60 1020 960 YlO 900 850 810 780 750 2: 700 670 660 650 600 590 580 530 520 520 460 460 4 50 370 370 370 360 350 340 cycles.247 0.il’ o..18J5 O. “‘.1793 0.1701 0.130*11 0.o 1028 5.387 0. Spacing I 5.215 0.3!)6 0.329 !0.0843 0.440 0.0307 0.505 0.253 ~0.12Go:o 1390~0.1898 O.237 0.:391 0..0732 0.2Y3 0.1401.2. 1071 0.319 /0.5 23 22 21 21 21 21 ID 19 17 I7 15 15 12 12 IO 10 IO !I 550 120 390 230 070 910 750 110 590 590 590 590 430 430 270 270 0.378 0.279 1410 0.5 0.1924 0.242 n.410 0.1212 0.0201 0.1097 O. 1 dc .37 155 148 133 521 510 510 785 0.3.308 ‘0 308 10:307 IO.230 0.
434 0.378 0 3i7 0. 1542” 3Y 1080” lx.1089” 6x.12i5 o 1551 0.388 0.006 1 016 0. 1294” 2x 3x 4x 5x 6x.1256 0.310 0. solid circles.451 0.0039’ 0 0045 0 00561 0 0060 0 346 0.962 0.x. 0 233 0.1207 0.704 O.G.988 0.). 1381” Ox. 1092” lx.366 0.87 3. 1752” 2x 1332” 5x 1242” 9x 1600” 2x.420 1.434 0. 1026” lx ax 3x 4x.00 1.(77”F.261 0 253 0 249 0 247 n 245 0 230 0 274 0 26i 0 261 0 255 0. 2x.385 1.1143” 3x.533 0.51 3.200 1.51 3.416 0. 1546” 4x. and an ambient of 25°C.86 3 86 3.0123: 0.2i4 0547 06% 0 540 0 64f 0.457 0.651 0.565 4 92 4 92 4 93 5 37 5 39 5 42 5 42 cycles.400 1.0044 471 0 0035 392 0. 1454” 4x.15i6” .468 0.1479 0. 1068” 2x.1590 0 1325 .954 1.3i2 0.1531 0.1241 0 1529 0.825 0 249 0 498 0.441 0 560 0 559 0 559 0.67! 306 0.545 0 518 0.980 0.436 0.1462 *Based on a conductor temperature of 75’C.07 0.896 0.1136 0 1206 0 1224 0. Sparmq AwaKe Currcllts 6’ cyrl?n enrtaoce Megohms per Conductor per Mile ox! ft.413 1. rise is 757r.2RO 0.136 1 133 1.474 0.44 2.326 0.359 0 36i 0 365 0 350 0. total temperature. 1221” 2. and crosshatched circles represent copper. 0.1.1446 0. 1438” 2x.398 1399 1.146i” 5x.466 0.1171 0.9i9 1.561 0.389 1.86 0285 0.x.16550.52 3. 1851” 2x.952 1.498 0 493 0 4R9 0 479 0 0 0 0 0 54i 534 522 509 505 0.&32 0.517 9 311 ll 0% 3 193 4 942 2 5S5 2 143 4 022 2 754 3 256 2 233 1 362 1 743 240 0 0035: 944 0.OS3p 1x.432 ‘0.616 1 til0 1542 1.1164 0.0050 0. 960 I 730 I 920 I 5 292 I 0 01859 ’ 840 4 669 0.498 0.20 2.21 279 2.428 0 423 0 386 0. 1225” Ix.268 0.462 0 459 0. 1.290 0.1137 0.413 0.504 0. 13Rl” Ix 1621” Tn.973 0.487 0.1663 0.344 0.44 2. 5x.221 0 206 0 210 0.20 2.511 1.49 4 91 0. 2x.625 0.0040 495 0.15060.612 0.82 3.12X 0 1489 0.901 0.239 1. 137i” 31 1257” 4.885 1.024 1.Olll< 0.308 0 411 0 3fiR 0 336 0 311 0. 1.525 1.754 3.W98( 0..lOlB” 3x. 1513” 6x.273 1.107 0 755 0.273 0.1239 0 1520 0.351 0 447 0.722 0 755 0.593 1.4ij 0. 1576” Sr.446 0.233 1. 13i3” 6x.599 0 391 0 308 0. 1467” 2x.769 0.1294” 5x.643 0. I.878 1. 1046” 2x.691 0.1155 i 800 i 170 1 860 3 370 ! 220 ) 880 7 600 j 430 I510 3 846 3 094 I490 1 970 3 563 j 536 5 1 3 6 5 6 2 9 7 5 5 4 410 900 000 956 266 870 680 730 322 876 62fi 233 400 420 410 410 360 350 350 360 350 310 310 310 310 280 . 271 277 281 2Rl 285 2i9 285 289 294 281 268 293 298 302 281 289 296 301 298 306 310 290 298 304 309 .0808” 1 8 7 3 420 460 340 038 564 26i 191 853 1 414 1 423 1. 0 0072’ 0 OOi6: 0 ooiw 0 0087: 0 0044! 0 00501 0.200 1.4 miles per hour (2 ft/sec.671 0 655 0. of the “Approximate Current Carrying Capacity at 60 cycles.545 2 02 1939 1.51 3. The approximate magnitude of current a51.776 1. 1311” lx.813 0 783 0.0081: 0 OOYl> 0 0099t 0 0109! 0 00631 0 0072: 0. Characteristics of “EXPANDED” Aluminum Cable Steel Reinforced.570 0560 0.096 1. 1615” 2x.544 1.273 1267 1.475 0.435 0.1258 3x.1275 .0081: 0.645 0.84 3. 1412” 3r: 1307” 2x.698 0. 1780” 3x.598 1.522 0.1614 0.884 1 127 1 126 1 123 1. The authors wish to acknowledge the cooperation of the conductor manufacturers in supplying the information for compiling these tables.525 0 509 0 501 0 664 0.392 0 367 0.. . 1225” 5a.680 I 698 I 666 I 621 I 637 I.20 2.726 0.113 1 112 1. 1632” 4x.1310 Rx. 1200” lx 1281” 2x.O06i! 0. 1266” 2x.49 3.86 3.283 0. 4x.0115f 0 01254 0 0169.0091: 0. 1833” 2x 1091” 4x. 15R.82 3.78 1 785 1 i59 1.67: 0.885 0.05 3.328 0 348 0.651 0 68.1198 0.756 1.1091” 3x.356 0. 1833” 7x.1127” lx.377 0 37i 0.232 1. and aluminum conductors respectively.579 0.962 0.331 0.175410.022 0.00641 O.528 0.4i5 0 475 0.110 1.77 3.535 0.0037 298 0.967 1.377 0 349 0 36R 0 327 0.445 0.50 3.530 0. 1546” 1112” 1307” 1222” 1153” 0.219 0. Oi97” 2x.729 I.271 0.03 0 559 0 546 0 523 0. 1209” 2x 1281” lx 1068” lx.280 0.07 1296 1289 1.920 2.548 0 548 0.281 1275 1229 1.358 0.21 2.62$ 0.07 3.40 2. I. 1480” 7x.755 0.211 1.i47n~~ :x li51" “L. 1141” lx.1507 0. 1242” Ix.401 0.1648” 2x.1255 0 1489 0.Mile at 25”C.0895” lx.” 3iOF 2lnK 2/O .J 2/O G % l/OK l/OJ l/OG l/OF 1N 1X 1 .781 0.3i6 0.5” 3x. 3x la.0044’ 536 0 0047’ 514 0 0046 594 0.44 307 3.905 1.832 0 256 0 512 0 614 0 719 0.749 1.228 0 423 0.040 1. 1016” 1x. “t 25” C.648 0.735 I.. The double cross hatched area in the insert for Table ZB.650 0 654 0 62i 0.623 0. (122°F.216 0.20 2.360 0 356 0.475 10 463 0 454 0.464 0.941 1.40 2. 1540” 137i” 1257” llfi4” 1690” 1089” 102fi” 2x.986 0. 1944” 1x.326 0.1232 0. wind 1.634 0.1470” 9x.212 0 225 0.346 0.1311” 6x.985 1.79 2.388 1762 1.490 0 534 0 4Y4 0.0155E 0. 0 0156f 0.786 0.402 0.1292 o 1448 0. 1332” Ix.0226 7i0 .540 2.78 1909 1.748 221 2. 1731” 4x. I. plus 25% rise due to heating effect o current. 1585” 4r.351 0 362 0.691 0.0057.1345 318 0. 1347” lx lOS7” lx.234 0.705 0.200 1.323 0.450 0.599 0.1509 2 541 2 144 1 861 1 649 1483 48i 015 iO1 4i6 356 3oi li6 973 598 349 lil 075 0. 1087” 2x 1438” lx.iO2 0.508 0 541 0 531 0 4!15 0 504 0.42 3.982 0.306 .624 0. 0.446 0 0 0 0 573 570 568 559 0.808 0.“r _ utside hmeter uches ) 788 j.” necessary to produce the finding the appropriate table for a particular conductor.01911 I i i i ! 730 640 370 000 290 4 4 3 3 3 3 3 3 3 3 2 3 2 2 2 2 2 2 2 1 479 168 951 9:.1347” 6x.459 0.262 1.252 0.X32 0.22.444 0.237 0 231 0.514 1.1386 IIIIIiIIII I I I I I I I I ’I I 3. 13fil” Ix. 1459” Ix.78 2.40 3.77 2.456 0 .691 0.035 1 014 1.627 0 612 0 598 0 592 0 587 0.148i 1.524 0 520 0.2i3 0 262 0 258 0 536 0 523 0 510 0.20 2. 955 D umber and Diameter of Wires Copper ‘x . 1731” Ir.1152 14060 1171 1404 0 lli0 142i 0.102 1. fre uency=60 **Resistances at 50°C.1548 0.344 0 344 0 346 0 345 0.0028 0 285 0.0207 420 .51 3. 1.174 Komina Dcsignation 350 E 350 EK 350 v 300 E 300 EK 300 v 250 E 250 EK 250 v 4/O E 4/OC 4/O EK 4/o v 4/O F 3/O E 3/O .13i9 333 0.395 1.894 0 880 0 892 0.905 2.179 0.W.909 0. lx.098 1.562 0.1405 0 1438 0 1465 0 1488 0.397 0.199 0.1397 0 1423 0 1447 0.1290 .576 0.548 0.67( 0.784 0.924 1.78 0.1221” 5x.8il 0 869 0.lOlfi” 2x.1205 0.44 3.237 1.529 1.86 3.369 0. 1222” 1x.397 0 382 0.109 0.514 1.22 2 21 2.652 0.51 3.246 0.0608” lx.243 0.703 0.983 0.742 1.936 2. based on an ambient of 25°C.548 0.SR3 0.51 3.476 0.412 0.408 0.1221 0.813 :: 2.705 0.423 0.57.111 1.028 1010 1.361 0. 749 0.310 0.899 0.550 0 545 0 555 0 519 0 509 0. 7.712 0.03 3.1241 0 15280 1274 0.584 0 569 0 559 0 54.1572 0.05 0 820 0.497 0.716 0.0138t 0.466 0.440 0.902 0 899 0 R82 0.448 0.0039 320 0.871 0. Ppacmg 25 50 fi0 TIC3 cycles cycles / I 5o 1 6o 25 / 5o cycles cycles cycles / cyrlrs 420 ! 7 409 I 0 0220 850 6 536 0 0245 430 6 57s 0.62( lx 1540” 2x.356 0.222 0 243 0 23.906 0.098 1.1531 0.0053.121 1. 13il” 122R” 1120” 1036” 1513” 1.755 0.571 0 68: 0 268 0 535 0 64i 0 264 0.225 0. 1046” 1x.871 0 8il 0.622 0 407 0.452 0.399 .44 2.3i7 0 4i5 0.098 1.1469 0.306 298 .1153” fir.0050 676 0.388 0.21 2.489 0.1120 0.9i9 0.J 3/o G g”.2iO 0.lli2 0.753 2.875 0.442 0.555 1.871 1.50 3.599 0 599 0.. 261 265 266 264 269 270 268 2i3 274 2i3 27.) Small Currents rcz Resistnuce Ohms per Conductor per Mile at 5O’C.941 2.84 3. 1rrent rrying pacity it 60 Resistance Ohms per Conductor per .1266 o 1547 0.884 O. 9u 6x.554 0.386 0 403 0.434 0.573 0. 0808” 2x.7 2 :\ 2G 2F 3P 3N RK 35 3A 4P 4N iA” 5P 5D 5A fiD SA 6C 7D 7A I.511 1393 1.1118 0.382 1.230 0.0139: 0.896 0.020 1.1385 0 0 0 0 0 13550 1129 1383 0.443 0 412 0 420 0.519 0 570 0 555 0.886 0. average tarnished surface.29 0.x 1164” 2x 1699” 5u. represents stranded paper.301 314 0 0 0 0 0 1306 1324 1331 1322 1344 0 1088 0 1103 0 1109 0 1101 0.487 0.704 0.258 0 2x 0.351 /0.) Current Appror. For these figures open circles.1159 0.871 0.691 0.49 3. I.21 2.223 0.52 Characteristics of Aerial Lines Chapter 3 TABLE 4ACHARACTERISTICSOF COPPERWELDCOPPER (Copperweld Size of Conductor Steel Company) CONDUCTORS ra XII Cap&ve .348 0 352 0. of Capacity” 25 cyclev InductiveReactance Ohms per Cooductor per Mile One ft.200 1.1123 0.542 0.497 0.1472” fix. 750 552 732 305 134 154 9i4 411 960 622 502 359 i03 346 OiSl 8iOl 0 01711 0.0102: O.952 0 952 0 952 0 950 0 052 0. 1092” 2x. 1632” 3x.548 0.714 0.1406O.716 0.. 1851” 2x. 1459” 5. 1615” 1x.747 1. 1371” 5x 122R” kc 1120” 3r 1036” lx.554 1.00561 0 0064.805 0.098 1.1289 0.129 1.01521 0. WOX.44 2.225 1.719 0.781 0.606 0. .598 486 0 583 0 R15 0 239 0 478 0 5i3 0.01409 0 01903 0 0175f 0.787 0 818 0. 1373” 1x.353 0. 5x.50 3.42 2 42 2. 1780” 1648” 1542” 1167” 1454” :opper quivalent irculilr Ilk or .463 0 465 0.0088: O.462 0.456 0.382 1.742 2.390 0. I.I 699 0. steel. I I.499 0.84 3.755 2. 4x.1141” 2x.126 1.I 1G IF 2P 2N . 1018” 9x.1143 0 1136 0.lY44” 5x 1143” 9x.198 0.1240 0.434 0. 1893” 2x.558 0. 0 216 0.575 0 65i 0 641 0 626 0 611 0.385 1.518 0. 0 297 0 593 0 71: 1.419 0.: 3 586 3. 260 2io 250 240 240 230 240 230 230 220 210 210 200 210 190 180 190 180 160 160 160 140 140 130 0 1348 0 1341 0 1365 0 1372 0 1363 0.
485 0.46 8.91 3.15R90 1324 0.8 0.253 0 485 0 582 0 258 0 12X9 0 lOi 0 259 0.07816 1.978 1.628 0.310 0 596 0 i15 0 351 0 1754 0 1462  19 590 l Il:rscd on cotlductor temperature of 125°C.77 l.490 0.111% 0 OOil 0 386 0 396 0.549 0.56 0 664 0 301 0 1506 0 1255 0 301 0 56X 0 682 0 310 0 154i0 1289 0 307 0 580 0 696 0 318 0 1589 I) 1321 0 313 0 591 0 i10 0 326 0 lG“!l 0 135X 0 319 0 603 0 724 0 334 0 16il 0 1392 ‘I/W’ ] ! 2” 7.279 0. :I?‘.584 0 624 0.00931 690 610 530 470 410 470 410 350 310 270 230 200 250 220 190 160 140 120 90 0. 4 i No.267 0 505 0 fi06 0 241 0 12OG0 1005 0. 3 so.3 04 0. !I 7 so.00559 0 01175 0 01046 0.0420.229 0.445 1.261 0 501 0 601 0 269 0.2 1.6 2.06440 2. 1’ 3 No..1!)202 1 52879 1.22753 1.55 8 78 0.00067 1.588 1.042 1 050 1.8 0 2i7 3 No.19 2. 5.610 0 261 0 13Ofi 0 10% 1 030 1 290 1. R 7 No.470 0.9 K 1.4i2 0 241 0.297 0 5i2 0 687 0 334 0.87 6.09375 1. II 3 No.1 1.473 0 539 0 233 0 1165 0 09il 0 553 0 241 0 12060 I005 0 567 0.790 0.040 1.840 0 975 1 021 1.174 2IJ2” !I/lli” 5/R” 0 460 0.0 3.623 0 773 0.375 0 249 0 254 0.548 0 291 0.299 0.998 0.496 0 506 0 511 0.87 4 88 6.75 3.‘:<“” S/Ii.244 1 568 1.365 0. R I 9 No.17538 1.756 0.87 4.978 2 00 2.13069 1.449 0.584 1986 1.5 3.50 1762 2 21 2. 9 7 No.615 0 278 0 13880 1157 0. 6 7 No.53 8.580 0. 1’  1.35 2 41 1738 2. A 7 No.0 1..617 2 03 2 55 3 20 2 31 2 91 3 B6 4.4199!) 1. IG" :':i/:L?" ‘~.5 2.1010 0 613 0.645 0 286 0..00276 0 OOiO5 0 00628 0.363 0 419 0 4i6 0 499 0 261 0 493 0 592 0 233 0.00000 1./Iv :3/V’ I. 4 7 No.002 0.9 K 1.78 3.50 4.47 4 37 5 50 8.5 1 6 1.134iO 1122 0 267 0.524 0 629 0 286 0. No 7 :/ No x :I No !I :.1040 . 1’ 3 No.61 5 81 7 32 1 090 1 343 I 675 2 09 2 61 3 27 2 34 2 94 3 70 4 65 5 R5 7 36 1 0119 0 257 0 54.3 2.15207 1.994 1450 1.314 0.64051 9 0.lli” :3/S” ..291 1 584 1. R 11St.919 2.62 7.00212 1. No Ii :.00340 X 1.6 3.Chapter 3 TABLE Characteristics of Aerial Lines 4ISCHARACTERISTICS (Copperweld 53 OF COPPERWELD CONDUCTORS Steel Compnny) Resistance Ohms per Conductor per hide at WC.831 2.931 2.664 0..38504 1.640 0.714 1 833 2.90 3.406 0 411 0 45X 0 513 0 580 0 605 0.5 :.433 0 321 0.43 3 07 3 87 4. 7 I !I so. 710 356 Oi6 467 163 922.323 1 331 1.l.2 0.671 0 311 0 1553 0 1294 0 293 0 14650 0 301 0 15060 0 310 0 154iO 1221 1255 1289 ‘)/IV’ i 12” .721 0 642 0 572 0 613 0 546 0 486 0 133 0 3x5 0 343 0 306 0 392 0 349 0 311 0 277 0 247 0.455 1.14 1.5M) 0.573 0 247 0.600 0 605 0 688 0.492 0.986 1.928 1.34 3.843 0.861 1.01071 1.275 0 526 0.656 0.658 2 09 2.343 0.8 3.00013 1.. 9 7 No.) ra i/U" 13.43 3 06 3 86 4.11126 1.00000 1.87 6. and an ambient of 25°C.4 2.220 0.1 0. lli” :I so.73 2. SC> 10  0 827 0 835 0 843 0 847 0 981 1.680 0. s 3 No.170 0 780 0.736 0. I4 390 18 380 !3 390 !2 310 18 510 15 330 12 670 IO 460 R 616 i 121 8 3i3 6 934 5 732 4 730 3 898 3 221 2 236 .>/:gy" !I/lii" 3. 8 I !I No.51 4. Ii 3 No.” TABLE ~SKIN EFFECT TABLE K 1. 9 0 247 3 .44 4 33 5. 7 0 910 0 Xl0 0 i21 0 642 0 5i2 0 613 0 54fi 0 IXfi 0 433 0 355 0 343 0 306 0 392 0 349 0 311 /624 !900 313 ioo 248 800 292 200 231 700 lR3 800 145 700 115 fioo 91 650 72 680 99 310 iR i50 62 450 49 530 39 280 31 150 i5 5i0 66 910 9 344 19x ROO 15 830 J5 530 7 410 395 500 Ii 740 45 850 5 Rii I1 040 37 690 4 660 !5 500 30 610 3 696 !4 7x0 29 430 4 324 IO 470 21 650 3 429 IF 890 20 460 2 719 13 910 16 890 2 157 I1 440 13 890 1 710 9 393 11 280 1 356 7 758 9 196 1 076 9 262 11 860 1 46i 7 639 9 754 1 163 6 291 7 922 922 5 174 6 282 4 250 5 129 3 509 4 160 731.2 0 00621 0 00553 0.559 0 644 0 294 0 1471 0 1226 0. 5 7 so.47 2.: 580.2 2.496 0 595 0 266 0 1330 0.58i 0 261 0 13060 1088 0.549 0 391 0 236 0.783 0.798 1.938 2 44 3.006 1264 1.617 0.508 0 512 0. 10 3 NI.00124 1.41 5. 0 303 0 1512 II 1281) 0 311 0 592 0 ill 0 31ti 0 604 0 725 0 311 0 1553 0 1294 0 289 0 545 0 654 0 293 0 141350 1221 0 295 0 5. 7 i x0.2 3. 6 3 No. fi :I St. II t 9 No.66 4. .620 0. :x” ‘4” I 9 No. (i7”F..486 0.16 2.252 1.3 3. 9 3 No. 0 09758 0 OORi5 0.73 1.674 2.56587 1.. 7 7 No.4 731.710 0 737 0 2iR 0 517 0 ti?l 0 250 0 124X 0. ‘*Resistance at 75°C.11 2. 8 3 No.00391 0. 550 460. 6 : 9 No.335 1 560 1.4 3.44 3 07 3.00758 1.20056 1..250 0 12480.00519 1.1 2.00032 1 .821 2.uo.64 1926 2. R i No. rise due to heating effect of current.00492 0 00439 0.22 2 79 3.936 2..292 0 561 0 673 0 326 0.5 1.8 1.514 0. The approximate magnitude of current necessary to produce the 50% me is 75% of the “Approximate Current Carrying Capacity at GOCycles.8 2.606 1.00001 1.457 1.65 1.65 4.289 0.00004 1. . 9 7 No.15 2.03323 1.272 0 309 0.15 1 678 1967 2 48 2.31809 1. 11 0 220  0 00323 0 00288 0 00257 140 120 110 40% Conductivity 628 900 498 800 395 500 313 iO0 24R so0 292 200 231 ml 183 800 145 ioo 115 ii00 91 650 72 680 99 310 78 750 62 450 49 530 39 280 31 150 i0 240 I1 600 . !. 5 3 so. 9 344 7 410 5 877 4 660 3 696 4 324 429 i19 15.385 0 390 0.31 2. 1: 0 910 0 810 0.220 1.0 2.536 0 285 0.25620 1.537 0.00348 0.ii2 2.5 0 (i 0 7 0. 3 ?I<>.32 0.14 1666 2 10 2 64 1.853 0.1429 0 1191 0.590 0.1 3.00601 fi20 540 470 410 360 410 360 310 270 230 200 170 220 190 160 0 306 0 316 0 326 0 331 0.1 460 ( 289.371 0 396 0.778 0 824 0 370 0 8Ri 0 281 0 533 0.782 0.90 3.9 K X _1._.40 5.29 2 85 3 63 4 58 5 78 7 28 3 X0.66 4 61 7 33 1.61 7.260 1 576 1.577 0 643 0. 'I" 30% Conductivity 19 No. I.33 1 476 1530 1.01969 1.3 1.16290 1358 0.lli” 9.1671 0 1392 0.684 0 633 0 638 0 728 0 i99 0 Ri? 0 902 0 2% 0 529 0 635 0 254 0 1249 0 1074 0 793 0 798 0 917 0 995 1 075 1 10R 0 285 0 541 0 619 0 “lx 0 13300 1109 0 672 0 676 0. based on an ambient of 25°C. 5 3 No. 7 9 No.058 1 062 1 237 1315 1.4 1.9   .994 0 480 0. 0 668 0 X3 0 13’iX 0 1157 1 69i 0 299 0 5fi9 0 OR3 0 ?Sli 0 142!1 0 II!)1 2 12 2 Ii4 3 30 2 35 2 95 3 71 4 GA 5 86 7 38 0 305 0 581 0 697 0 29I 0 14il 0 1221.826 0..11 2 66 3.30 2.60314 1.658 0 303 0 1512 0 1260 0.546 0 608 0 672 0. plus 50°C. i 3 No.5 0 654 0 2G9 0 1347 0 II22 I 364 0 293 0 55.28644 X 3..01470 1.88 6.239 0.1109 0 255 0. 7 3 No.698 0.7 3.450 0. 5 7 No.489 0. total temperature..802 1.69 0 269 0.659 0 318 0.05240 1.35102 1.461 0.636 0.631 0 281 0.45570 1.0 0.13 1.02582 1.513 0.91 3 66 4 62 7. 5 9 No.31 2.826 2 30 2.07 3.271 1.273 0.7 1. 5 I 9 No.7 2.
0.5.0221 0.1931 0.: 11 1 1 0. .0291 0.1111 0.0566 0 0992 0.1267 0.0892 0.1334 .^ “. Feet ^ .1217 0.1314 0.ld15 t 0.1853 0.2037 0.5 26 27 28 29 30 31 27 3: . 2.1212 .2026 0. SEPARATION Inches 0.1906 0.1539 0..1720 0.0156 0.+.1918 0.2060 0.1.2015 0..1297 0._. TABLE ~ZEROSEQUENCE RESISTANCE FACTORS (re.0899 n nnro xd at 25 cvcles td .2594 10 11 __849 0.2049 0.1<539 3 0. .O047G4/ ze=0.0960 0. n 1% 4 0.1754 0.0410 0.0465 0.0517 0.1443 0. 1_ 17 18 19 20 21 22 23 24 2.1299 0. FUNDAMENTA EQUATIONS =z.1585 0.0857 0.006985/ where /=freqllency p=Reuistivity IORIO 4 G65 600.2511 .1607 . 0.0613 0.2071 0 0.1968 0.+j(x. : .1769 0.1242 0.2003 0.=r.1359 11 0.0782 0.1627 7 0.1657 ^ _^^_ V. 2 0.54 Characteristics of Aerial Lines Chapter 3 TABLE 6INDUCTIVE 25 CYCLES SEPARATION REACTANCE SPACING FACTOR(2.0410 0. 2.___ I_.5 xd x. .1337 0.1736 13 n 17.1893 0.1839 0.1826 0.59 :: lO:i% 16 .1702 11 12 0.1502 0.2328 lcrg.+ =r..0184 0.+r.1677 0.1053 0.1380 Xd at 50 cycles m=O.08850..07010.1166 0.xe) AND * INDUCTIVE REACTANCE Ohms per Conductor per Arile FREQUENCY 25 Cycles re All 0.0927 0.!=O. .1402 EE 0.^^_ \‘O 0. feet.1192 *From Formulas: re=O.1576 0.1880 0.1739 0.1644 0.lY.W+x.1082 0. li64 logto d d =separation.2804 ^ ..1693 0.1483 0.1461 0 0.0820 0.1463 0.!?I 1.0545 0.1708 0.1423 0.0701 0.1402 0.1256 50 CYCLES SEPARATION Inches 0l11213141~/~17/ .1666 0.1980 0. 1.1521 0..) OHMSPER CONDUCTOR MILE PER 5 16171819 ~ I 10 I .0088 0.1724 0.1991 60 CYCLES 10 a.2513 i:z$iFeet 0.0742 0.I3 14 0. 0.j4f.2157 0.1866 0.1139 0.1558 0.0291 0. ohm) tThia is an average value which may be rlsed in the absence of definite information.=O..1563 4 0.08921 nnvo Y.“““Y 0.2347 0.2794 at l?&e d 60 cycles d=separarlon.1661 0.1685 0.143Z .0081 0.0106 0..L 2J j g:.r..073 5 0.o cf d= xtioll. .25130. (meter.1023 0..1111 0.A” 11 1329 .1291 0.1489 O.0352 0.1192 0.0789 .0226 0.1631 _ _^__ 0.. 8/Q 0. feet.18120.1647 0.0658 0.2425 0.
1136 0.: “.127 0.472 :::FZ :8 20 2 40 % i: 1:: 12.0166 0.213: “1 0. 126 0.0103 0. 0.0429 0.129 0.0581 0..1116 0.0026 0.0405 0.418 0.06831 lo&o d ’ d=separation.0574 0.0821 0.1019 0.137 0.0206 0.131 0.0272 0.106 0.'at 0 1 2 0. f 2h ground.114t .0930 0.0326 o.108 0.I+x: x.836 0.0611 0. 192) 16 0.0955 0. 2:55 :<7 0.I0 0.274 48 0.07370.024 0.266 ‘43 0 267 .566 60 Cycles 0.0874 0.0711 0.0478 0.023 1. z”!.057: 0. 253 :I(.0532 0.0218 0.0515 0.0492 0.348 0. feet.0085 0.0978 0.109 50 Cycles 0.936 0.528 0.0567 0.1123 0.1821 .410 Ei 0.j 0. 0.u4$ve Feet 25 Cycles 0.091: 0.201’ .23: “1 0.0318 0.078: 0.0462 0.542 0.328 0. 2321 L17 0.078: 4 0.0262 0.0385 0.0501 0.088.0193 0.0823 0.0519 0.0591 0.0291 0.056 1.074( 0.138f g 0.0570 0.727 0.267 0.0841 0.0365 0. feet: 60 CYCLES SEPARATION Inches ‘eet 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 0 0 0.492 0.0783 0.0473 0..:j 0.I640 logosd d=separation. 0.122 0.069: 0.0989 0.1271 7 0.303 0.0494 :I 0.5 0.135 0.468 0.1102 0.1155 9 10 11 4 5 6 7 8 1 3 2 ~ ~ ~ 0.0350 0.275’ .00540.1181 0.0601 0.0251 0.0446 0.j 0.17% .272.320 0.0457 0.0761 0.039 0.1094 0. 260 10 0.277 25 cycles zd’=. 47 0.0803 0.049.42 0 .0683 0.I4 I.0334 0.1101 O.133 0.0548 0.0917 0.0510 0.0563 0.084 1.363 0.0544 0.1037 0. 0.j 0.=x:=x.0441 0.o2o6o.08198 logm d d=separation.0342 0. 555 0.0540 0.0423 0.0652 0.1079 0.040 0. 0.l(l 0.%39 :x.g.o120 0.1129 0.5: I!.64 .0452 0.1087 0.0941 0.102’ 0.0300 0.124 0..05913 0.104 0.437 0.063.0737 0.1142 0.257 :x3 0. 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 R4 35 36 :37 38 39 40 41 42 43 44 45 46 47 48 49 0 0.121 0.134 0.100’ 0.0399 0.0555 0.117: 0. feet. TABLE 9ZEROSEQuENcE Megohms SHUNT CAPACITIVE REACTANCE Ze’ per Conductor per Mile FREQUENCY FACTOR Conductor Hei.0229 0.2’26: I.125 0.0024 0.o.1071 0.‘=0.0497 0.0482 0.123 0.0903 0.138 xd’ at 50 cycles &.128 0.45 0. “62 41 0. 0.0523 0.164( 1.0357 0. y II.209 ‘2.137 0. 0.0467 0.0943 0.0136 0.0506 0.1028 0.234 “H 0 ..0577 0.5.00850.*I 0.0487 0.187! 1.0588 0.08.o16o o.136 0.0 where h=height above f = frequency.390 0.984 1.2491 34 0.098i .394 0.197‘ 17 0.0241 0.0282 0.lllN 0.0435 0.244 32 0 246 XI 0.0309 0.242 :I.0536 0. 0.0967 0.0180 0.30 ze’=log.1009 0. . 0.2161 L’L’ ” ““0 L’:j 0.0417 0.0604 0.0617 0.156: .133 0.0152 0.788 0.0608 0.0379 0.1481 9 0.1141 0.0889 0.0559 0.364 0.04110. FUNDAMENTAL EQUATIONS x. =x:+x:2x: SEPARATION Inches 0 . 0.1063 0.132 0.0120 0.0614 x: at 60 cycles r/=0.0584 0.1046 0. 0. 20.259’ :%!I II.0552 0.0411 0.0594 0.0066 0.271 .875 0.098 0.237: “!I 0.0527 0.1055 0.113 0.a 0. 269 .096.0372 0.0046 0.119’ 0.0999 0.511 0.132 1.) 0.0326 0.05320.0858 0.1161 0.170: 12 0.130 0.0392 0.1149 0.1109 0.Chapter 3 TABLE %SHUNT 25 CYCLES SEPARATION INCHES Feet ~ 0 0 Characteristics of Aerial Lines CAPACITIVE REACTANCE SPACING FACTOR (xi) MEGOHMS PER co~~uc~o~ PER MILE 55 x.251 35 0.
which is given by: & 17. For this reason. Under conditions where abnormally high voltages are present. feet . Table 10 gives barometric pressures as a function of altitude.1 Kv/cm rms. relative humidity.92 for segmental conductors) 6 =nir density factor (W where Z. for threephase voltages.500 0 1000 2000 3000 Pressure.47 29.9b 459+“F Pressurr.22 20. and scratches will increase corona. in turn. reduces the probability of flashover and improves system performance. m=surface factor (common values.58 16. die burrs. it is believed that these high losses were caused by sublimation or condensation of water vapor. Thus the relative corona characteristics of these two types of conductors might interchange between fair and foul weather. or the equivalent phase spacing. Die burrs and die grease on a new conductor.81 Altitude.84 for stranded.84 24. and 77°F) the air density factor equals 1. There are apparently other unknown factors found under desert conditions which can increase corona. 31. in. feet 4 000 5 000 6 000 8 000 LO 000 15 000 20 000 Chapter 3 With the increased use of highvoltage transmission lines and the probability of going to still higher operating voltages. On highvoltage lines grounded through a groundfault neutralizer. measurements of loss made under conditions of fog and hoarfrost might be unreliable unless the conductors were at operating temperatures. Rain is by far the most important aspect of weather in increasing corona. Care in handling conductors should be exercised. Factors Affecting Corona At a given voltage. the common aspects of corona (radio influence and corona loss) have become more important in the design of transmission lines.lg The effect of atmospheric pressure and temperature is generally considered to modify the critical disruptive voltage of a conductor directly. 15) and switching surges. The temperature to be used in the above equation is generally considered to be the conductor temperature. if it is desired to obtain the best corona performance from a conductor. l7 By increasing Jhe electrostatic coupling between the shield mire and phase conductors. 6. the inphase current due to corona loss can prevent extinction of the arc during a line to ground fault.1000 . and weather.59 23. Hg.02 30. In areas where RI must be considered. In .75 standard conditions (29. the higher the critical disruptive voltage assuming constant diameter.) T =radius of conductor in centimeters D = the distance in centimeters between conductors. For equal diameters.=critical gradient in kv per centimeter. (Ref. Any distortion of the surface of a conductor such as raised strands. line configuration. It can reduce the overvoltage on long opencircuited lines. 25. In recent years the RI (radio influence) aspect of corona has become more important. 10 is more accurate. The air density factor should be considered in the design of transmission lines to be built in areas of high altitude or extreme temperatures. Above this voltage corona effects increase very rapidly.56 IV CORONA Characteristics of Aerial Lines TABLE 10STANDARD BAROMETRIC PRESSURE AS A FUNCTION OF ALTITUDE Altitude.92 28. Corona in fair weather is negligible or moderate up to a voltage near the disruptive voltage for a particular conductor. 29 Chap. a stranded conductor is usually satisfactory for 80 to 85 percent of the voltage of a smooth conductor. Recent work indicates value given in Sec. Hg. in. Hoarfrost and fog have resulted in high values of corona loss on experimental test lines. particularly the segmental type. corona is determined by conductor diameter. atmospheric pressure. corona at times of lightning strokes to towers or shield wires reduces the voltage across the supporting string of insulators and thus. for singlephase. Corona increases somewhat more rapidly on smooth conductors than it does on stranded conductors.28 6. of Hg. largely because of the energy loss associated with it. or as the x power of the air density factor.00. StrigeP concluded that the material from which a conductor is made has no effect on its corona performance. but their effect is minor compared to that of rain.86 27. 0. However. corona can affect system behavior. temperature. In the early days of highvoltage transmission. and imperfections in the surface should be corrected. can appreciably increase corona effects when it is first placed in service. The equation for critical disruptive voltage is: (i93) Eo=go 6% T m log. type of conductor.=21. Also. corona was something which had to be avoided. D/r where: E. Falling snow generally causes only :a moderate increase in corona.9s 22. A high value of critical disruptive voltage is not the only criterion of satisfactory corona performance.88 13.92 in. condition of its surface. Under The more closely the surface of a conductor approaches a smooth cylinder. It will attenuate lightning voltage surges (see Sec. This condition improves with time. Consideration should also be given to the sensitivity of the conductor to foul weather. and the earth’s electric field can affect corona. taking some six months to become stable. this factor might establish the limit of acceptable corona performance. 10 and 16 use g.= barometric pressure in inches of mercury “F = temperature in degrees Fahrenheit. 0. which are conditions not likely to occur on an operating line because the conductor temperature would normally be above ambient. = critical disruptive voltage in kv to neutral g.82 26. The calculated disruptive voltage is an indicator of corona performance.
Chapter 3 Characteristics of Aerial Lines 57 4 0 9 Curve ll. phase mile. Curve 51. variations up to 10 to 1 in corona loss and radioinfluence factor have been recorded during fair weather. and a conversion made for the ated at various voltages.88. Corona LOSS used as an indicator of expected performance during fair Extensive work by a large number of investigators has weather. This conductor had a poor surface.88 to show operating condition in desert. 12. (Reference 28 gives some measurements made in an industrial area.00. This particular curve is plotted for 6 =0. 13. Measurements made on threephase test line. 700 feet long.00. CoronaLoss Curves for Smooth Air Density Factor. iO0 feet long.04 in. Corona loss test made in desert at a location where abnormally high corona loss is observed on the HooverLos Angeles 287. This work has lead to the devel. Con imIllst. has proved to be a reliable indicator of corona performance (see Sec. 23. Curve 41. Curve 2Same as curve 1. 23. which is strung with this conductor. Refcrcnce curve gives threephase loss. Ref. by multiplying the test (:orona is temporarily increased. For a particular design. but line dimensions are not given. 27FairWeather ductors. Fairweather coronaloss measurements made by a number of different investigators are shown in Figs. Measurement made in threephase test line.4 in. smooth aluminum. Ref. when judiciously used. Curve 71. All curves are plotted in terms of kilowatts per three*kc the former might be negligibly small. Ref. 9) for transmission voltages in use up to this time. Ref. 16foot sag. Comparison with curve 2 shows effect of desert conditions. The relation between fair .4 in. seriously reduce the corona perI‘ornmnce. Ref. ‘I’llus it is not practical to design a highvoltage line such tlmt it will never be in corona. Curve 91. 700 feet long. Ref.96 in. smooth aluminum. Curve 61. Plotted curve is for 45foot flat spacing. smooth.65 in. This effect is moderate voltage by 1/h2j3. The curves should be 7. Curve 31. in some cases. \vhicah the same loss is observed during fairweather. The CarrollRockwell and the Peterson formulas are considered the most accurate especially in the important 10~ loss region (below 5 kw per threephase mile).5~kv line. 23. 6=0. r 3 '("LL f Fig.design under consideration. 27. Corona loss test made in California. HH copper. 6. 30foot ground clearance. The data presented in these curves has been If a conductor is deenergized for more than about a day. This conductor was smooth and clean. The conversions were made on the basis of voltage gradient at the surface of each conductor. On a contl~~c:~. Some error might have been mtroduced (‘omparcd to that of rain.or energized at a voltage slightly above its fair weather coronastarting voltage.. 16foot sag. Dimensions of line not given. and 29. HH. using an estimated value for charging kva.rial areas. foreign material deposited on the concIuc:tor can.57 in.65 in. 22foot flat spacing. Reference curve obtained by converting perphase measurement to loss on threephase line. except converted to 6 = 1. 27. smooth. 30foot ground clearance. The presence of rain produces corona loss on :Lc*ontluctor at voltages as low as 65 percent of the voltage XI. HH copper. to give loss on a line having 45foot flat configuration. Ref. Recent work on corona loss has been directed toward the extrahighvoltage range and indicates that more recent information should be used for these voltages. This also precludes expressing a ratio between fair. sible. Ref.to the original publications.her curves are for 6 = 1. 16foot sag.in these curves because in most cases it was necessary to gizing a line during fair weather where such a choice is POS. 30foot spacing. 30foot clearance to ground. corrected for air density factor. Reference curve gives perphase values. reference should be made been done in determining corona loss on conductors oper. Plotted curve obtained by conversion of perphase measurements to threephase values. opment of three formulas(10~14~16) generally used in this country (Reference 18 gives a large number of formulas). All ot. 12.1 in. The Peterson formula. 6 = 1. 30foot flat spacing. 28. It can be mitigated by reener. Measurements made on threephase test line. In reference this conductor is referred to as having an infinite number of strands.) Corona is an extremely variable phenomenon.and foulweather corona. smooth cylinder. Measurements made on threephase test line. 19. Curve 81.convert the original data from perphase measurements.
Corona loss on a conductor is a function of the voltage gradient at its surface. Ref. Curve 21. d 0 IC )O K"LL I Fig. Ref. 30foot ground clearance. ACSR.6 to 81 kw per mile.S. based on calculations using Reference 16.0 in. iO0 feet long. Curve 6l . 28FairWeather ductors. 8. hollow aluminum. Conductor wveathercd by esposurc to air without continuous energization. Ref.9 foot flat.Characteristics of Aerial Lines Chapter 3 I . Operation at or above this point can result in uneconomical corona loss. Ref. 11. weighing the annual energy cost and possibly the maximum demand against reduced capitalized line cost. 6 = 1. CoronaLoss Air Density Curves for Stranded Factor. the annual corona loss will not be of much economic importance20. Curve 52. trunsmission line can reach high values during bad storm conditions. Test configuration: threephase linf>. spating. ACSR. 22 App. Curve 81. Radio Influence (RI) Radio influence is probably the factor limiting the choice of a satisfactory conductor for a given voltage. 30foot ground cleurnncc. 700 fret long. 30foot ground c. Curve 71. This results in corona being more prevalent on the middle conductor. hollow copper. 22. The measured loss on their esperimental line varied from 1.49 in. to give loss on n line having 45foot Aat configuration. 23. A. Curve 94. Plotted curve obtained by conversion of perphase measurements to threephnsc values. 14. 22foot Rat spacing. Ref. \Yashcd in same manner as for curve 1. Corona loss on a satisfactory line is primarily caused by rain. Test configurntion: threephase line. This is shown by the fairly high degree of correlation between total rainfall and integrated corona loss which has. Ref. A very careful analysis. 32.125 in. Curve 31. 30 for one conductor. 20foot flat spacing. TOOfeet long. Test configuration: threephase line.09 in. Where radioinfluence must be considered. 16foot. 16foot sag. and the maximum loss will not constitute a serious load. twen noted.5 kw per mile.25 in.A. 700 feet long. Thus the effect of reduced conductor spacing and lowered height is to increase the corona loss as a function of the increased gradient. Ref. with an average value of (5. However. steelaluminum. hollow copper. steelaluminum. 12. Conductor washed. Ref. 11. Borgquist and Vrethem expect only a variation from 1. 16foot sag. Con and foulweather corona loss and the variation which can be expected during fair weather is shown in Fig. a very rapid increase in loss. The RI performance of transmission lines has not been as thoroughly investigated as corona loss. Curve 41. flat spacing. RI plotted against voltage on linear graph paper is characterized by a gradual increase in RI up to a vol . Plotted curve obtained by estimating average of a number of fairweather perphase curves given in reference and converting to threephase loss for line having 32foot flat spacing. I i Curve ll. Reference curve is average fairweather corona loss obtained by converting perph:kse measurements to threephase values. such conditions are not likely to occur simultaneously all along a line. The knee of the fairweather loss curve is generally near the critical disruptive voltage. steelaluminum. 30foot flat spacing.4 in. JVashed in same mannel as for curve 1. On transmission lines using a flat conductor configuration. Conductor was washed with gasoline then soap and water. 30foot. 14.(2’j26r41) The corona loss at certain points on a. Corona loss is characterized on linear coordinates by a rather gradual increase in loss with increased voltage up to the socalled “knee” and above this voltage. The calculated fairweather corona loss common in the U.00 in.lrar:lnc*r.8 feet high. TOOfeet long. on their 380kv lines now under construction in Sweden. sag. This conductor used on 220kv lines in Swetlen which have above dimensions. Test configuration: threephase line. Hollow Copper. 30foot f1:lt.91 in. 20foot flat spacing. 50foot average height. A transmis sion line should be operated at a voltage well below the voltage at which the loss begins to increase rapidly under fairweather eonditions. for a line 22.G to 16 kw per mile. Test configuration: threephase line. the gradient at the surface of the middle phase conductor is higher than on the outer conductor. must be made if operation at a voltage near or above the knee of the fairweather loss curve is contemplated. spacing. Test configur:ktion: threcphase line. Ref. \Yashed in same manner as for curve 1.04 in. 24strand. is generally less than one kw per mile. Recent publications (see references) present most of the information available. 14. using an estimated value for charging kva.
7* (Stranded aluminumsteel) 6 not given. Air Density Factor. Ref. Ref. Data for curves 6 and 7 were taken at same time in order to show effect of subconductor separation.5foot average height. 6 not given. 39. Ref.Chapter 3 Characteristics of Aerial Lines Curve l4/0. police and a number of government services might also have to be considered. These are available signal intensities along the line.04/23. the coupling to lower voltage circuits. converted to threephase loss for 45foot Rat configuration. This conductor was selected for USCon the Swedish 380kv system.21 For extrahighvoltage doublecircuit highvoltage lines the tolerable limits of RI might be higher because the number of receivers affected. 25. 13. probably close to unity. Plotted curve is average of two singlephase fairweather curves.and foulweather corona performance is shown in Fig. number and type of receivers in vicinity of the line.7* (Smooth) Ref. satisfactory signaltonoise ratios. the magnitude of the RI factors for the entire frequency range of communication services likely to be encountered. and Fourconductor Bundles. 12. lines must be designed to operate below the voltage at which the rapid increase starts during fair \v&her. Also fewer lines are required for the same power handling ability.randed aluminumsteel). Latter was used in making the ronversion to threephase. should be known. See discussion of Curve 1. Three. but its effect on the various communication services which require protection. 26. Curve 62/1. A. Curve 24/0. Plotted curve is average of two singlephase fairweather curves. 32. proximity of particular receivers.8* (Smooth). 8 not given.7* (Stranded aluminumsteel). 13. marine. *Bundleconductor designationnumber of subconductors/outside diameter of each subconductor in inches/separation between adjacent subconductors in inches. 6 = 1.4 feet.5 feet average height.7* (Stranded aluminumsteel).his is wrong. App.01. For practical conductors. theoretical considerations of the radiation characteristics of a transmission line as spacing is reduced. shiptoshore SOS calls. voltage at line height is given. the general importance of particular communication services. Rcfcrcnre curve obtained by converting singlephase measurements to threephnsc values on the basis of surface gradient.00/17. Curve 52/1. 12. Curve 72/1. transfer of RI to lowervoltage circuits. Ref.or. effect of weather on the RI factors and on the importance of particular communication services.4foot flat configuration. An evaluation of these factors in terms of their effect on various communication services must take into consideration many things. Dimensions of line used in making conversion are not given. 39. was converted from perphase mcasuremmts.10. 6 = 1. Refcrcncc curve gives singlephase measurements versus linetoground voltage.985/11. the lcvcling off value is much too high to be acceptable. being higher for smooth conductors and largediameter contluctors. Xn evaluation of RI in the design of a highvoltage line must consider not only its magnitude. 29FairWeather CoronaLoss Curves for Two. but assumed 1. Reference curve gives perphase mrasurcmcnts versus gradient. which is average value for Germany. and wider rightofways are used which tend to reduce the RI problem. converted to threephase loss for 45foot. 25. In determining the RI performance of a proposed line. 6 = 1. Ref. 6 = 1. See Curve 7. and the coupling to receiver antennas is reduced.827/15. the increase in the RI is very rapid. Hat configuration. the magnitude of the RI field begins to level off. tage slightly below the minimum voltage at which measurable corona loss is detected. 6 not given.092.25/17. which could account for their closeness. indicate that the RI from a transmission line will not be seriously affected by reduced spacing. Curve 33/0.04/15. and \\hcre RI is a factor. and means for improvement of reception at indiand vidual receiver locations.092. \mplitudemodulated broadcasting and powerline carrier arc the most common services encountered but other services such as aviation. The rate of increase in RI is influenced by conductor surface and diameter. curve is approximately 15 percent low in voltage.7* (St. flat configuration. but it is not clear whether actual test voltngc or equivalent.985/15. Although RI increases very rapidly with increased gradient at the surface of a conduct. Above a certain voltage. Fig. Dimensions 42. which gives threephase corona loss. Converted to threephase corona loss on line of 42. Ref. Above t.ion than the writer did in plotting curve 4. Original author probably selected a worse fairweather condit.his voltage. Curve 42/1. Reference curve. If t.42 .00. Figures 32 and 33 are characteristic RI curves. The relation between fair. Convcrtetl to flat configuration of 45 feet.7* (stranded aluminumsteel).
Lateral attenuation is affected by local conditions. operating experience and current practice are the best guide. Typical lateral attenuation curves are shown in Fig. Very little is known about the radiation characteristics of transmission lines and caution should be exercised in applying data not taken on a line configuration closely approximating the design under consideration. Bundle conductors are also called duplex. 400 K”LL 500 60C) Fig. with the result that the meter indication is near the peak value of the RI field.) for discharging is used.6 sec. For quasipeak measurements. Considerable work on bundle conductors has been done by the engineers of SiemensSchuckertwerkez7 who concluded that bundle conductors were not economical at 220 kv. The average value is the amplitude of the RI field averaged continuously over s second. The peak value is the maximum instantaneous value during a given period. Frequencymodulated broadcast is inherently less sensitive to RI because of its type of modulation. Figure 31 is based on Peterson’s formula and indicates satisfactory conductors which can be used on highvoltage lines. a circuit having a short time constant (0. 30Corona Loss on 1. the number of circuits. with the lower values applying generally to highvoltage lines. or even less in the case of medium highvoltage lines. and the presence of ground wires affect the radiation from the line with a given RI voltage on the conductors.0010. and is used as one phase conductor. microwave relay. HOARFROST. the attenuation at 1000 kc varies from 0. 34. App. Selection of Conductor In the selection of a satisfactory conductor from the standpoint of its corona performance for voltages up to 230 kv. No correction made for air density factor. reported for lines on which the calculated corona loss is in excess of this value. Aural tests of radio reception indicate that quasipeak readings interpreted in terms of broadcaststation field strengths represent more accurately the “nuisance” value of the RI field. 22.. (television3’. The lateral attenuation of RI from a transmission line depends on the line dimensions and is independent of voltage.!i s : 2 2c)Q $ Y I!jOVERCAST. referring to the number of subconductors and are sometimes referred to as grouped or multiple conductors. quasipeak.09 Inch Stranded AluminumSteel Conductor under Different Weather Conditions. 4 S 9. are the best solution for overhead transmission. Experience in this country indicates that the corona performance of a transmission line will be satisfactory when a line is designed so that the fairweather corona loss according to Peterson’s formula. . The type of measurements made must be known before evaluating published RI information or misleading conclusions can be drawn. Rusck and Rathsman46 state that the increase in transmitting capacity justifies economically the use of twoconductor bundles on 220kv lines. Bundle Conductors A “bundle conductor” is a conductor made up of two or more ‘(subconductors”. 50foot average height. radar. For medium highvoltage lines (135 kv) considerably more margin below the one kw curve is necessary because of the increased probability of exposure of receivers to RI from the line. The lower signal strengths.3 db per foot. increase the signaltonoise ratio. This conductor is in use on the Swedish 220kv system.1 to 0. Ref. etc. Note variation in fairweather corona loss and the relation between fair. The RI field from a transmission line varies somewhat as the inverse of the radio frequency measured. Directional antennas which are generally used at these frequencies. and peak values of the RI field. etc. Plotted curves obtained by converting perphase measurements to threephase values for a line having 32foot flat spacing. but for rated voltages of 400 kv or more. conductors. Thus services in the higherfrequency bands. and wider bandwidths generally found in the highfrequency bands can alter this picture somewhat.. Because of the rapid attenuation of RI laterally from a line. Unsatisfactory corona performance in areas where RI must be considered has been.‘6 is less than one kw per threephase mile.) are less apt to be affected.60 Characteristics of Aerial Lines Chapter 3 35 3c l 25. A. on the average.’ FOG CLEAR I( Standard radionoise meters35J6can measure the average.01 sec. The conductor configuration.and foulweather corona loss. frequencymodulated broadcasting. 10. a change of a few hundred feet in the location of a rightofway can materially aid in protecting a communication service.3 to 0. triplex. At distances between 40 and 150 feet from the outer conductor.) for charging and a long time constant (0.1 kw should be used. and a design approaching 0.
except that surface factor. (79a).27*28 These advantages must be weighed :tgainst increased circuit cost. For a twoconductor bundle. This effect modifies the corona performance of a bundle conductor such that its corona starting point corresponds to the voltage that would be expected from calculations. increased charging kva if it cannot be utilized.23.\ssuming isolated conductors which are far apart comI):ucd to their diameter and have a voltage applied bet\vcen them. and such other considerations as the l:trgc amount of power which would be carried by one circuit. equivalent to a single conductor 11~to eight inches in diameter. 80 and 81 for practical calculations. Curves based on Peterson’s formula Carrol and Rockwell paper for comparison. This effect can be seen by comparing curve 6 of Fig. Cahen and Pelissier21*24concluded that the corona performance of a twoconductor bundle is more accurately indicated by the mean between the average g=e rlog. 6. and less rapid increase of corona loss and RI with increased voltage . For a twoconductor bundle.5 29 STRANDS 1272 NUMBER IACSRMCM 1590 ~SOLIO~  I2 STRANDS  6 STRANDS OF STRANDS OUTSIDE LAYER Fig.Chapter 3 Charscteristics of Aerial Lines 61 360 320 60  SEA LEVEL 40 CONDUCTOR . It is possible with a twoconductor bundle composed of conductors of practical size to obtain electrical charact&tics. . 31QuickEstimating CoronaLoss Curves. the equation for maximum gradient at the surface of a subconducto? is: 9= e(lSB/S) 2rlog& (80) where : S = separation between subconductors in centimeters. and air density factor. m. ret111c:od surge impedance and consequent higher power capaIjilitics. the sc’parntion is not very critical. to a minimum at the corresponding point on the inside surface. but the rate of increase of corona with increased voltage is less than for a single conductor.5 500 c I6 STRANDS  750 666 IN INCHES I COPPERTHOUSANDS 1000 OF CIRCULAR MILLS OR AWG OR AWG IN 4 I I/O 3/o 4/o 3 795 1033.D/r (79b) . This equation is the same as equation (79a). the gradient at the surface of a subconductor is not uniform. the gradient at the surface of one conductor is given by: where the symbols have the same meaning as used in Eq. and it is advantageous to iisc a larger separation than the optimum which balances the reduced corona performance and slightly increased circuit cost against the advantage of reduced reactance. excepting corona. 29.!!I 664 DIAMETER 2 6 1: Ii/! I/O 2 II:!! !I 3/04/o ! 250 ! 350 300 500 397. It varies in a cosinusoidal manner from a maximum at a point on the outside surface on the lineofcenters. 28 with curve 2 of Fig. These factors should be added to Eqs. ‘l’hcoretically there is an optimum subconductor separati~ii for bundle conductors that will give minimum crest gtndicnt on the surface of a subconductor and hence high(‘St disruptive voltage. have been omitted. Because of the effect of the subconductors on each other. with a few check points from the The advantages of bundle conductors are higher disrup1ivc voltage with conductors of reasonable dimensions.
21. France. RI values are quasipeak. Conductor height 47.=21. Line has equilateral spacing. of . Test frequency868 kc. which can be as low as 0. 26. 1. 2 I '0 I 40 HORIZONTAL 80 DISTANGE 1 I 120 160 200 FROM OUTSIDE CONDUCTORFEET Fig.62 Characteristics of Aerial Lines Chapter 3 700 02 600 $ 80 60 I f I I I I I 1 !. but dimensions not given. made Fig. kv per centi4 ( b301) meter rms can be substituted for g and the equations solved for e. 42.1 I+. and maximum gradient at the surface of a subconductor.. g.5 feet. and Germany. 0 X n 0 are plotted values which apply to this curve only. 27foot flat spacing. Curve 2Lateral Attenuation from the 220kv EguzonChaingy line in France. I I I I I I I I I II. 0 . 1 of the Bonneville Power Administration. Ref. Measurements made opposite midspan on the 230kv CovingtonGrand Coulee Line No. Distance measured from middle phase. Curve IAverage lateral attenuation for a number of transmission lines from 138. in kv rms. test frequency800 kc. 34Lateral Attenuation HighVoltage of Radio Influence in Vicinity Transmission Lines. 21. This value neglects air density Factor and surface factor. Ref.108 inch ACSR conductor.to 450kv. which is given bv: g=e(l+~lS) 2rlog& (81) If it is desired to determine the approximate disruptive voltage of a conductor. 33FairWeather RadioInfluence Field from a Transmission Line as a Function of Voltage. Ref. 380 kv Systems using bundle conductors are being built or under consideration in Sweden.I .I/ 1 /I I I 200 IO 8 6 8 6 too 0 2 . Curve 3Lateral Attenuation from 230kv MidwayColumbia Line of the Bonneville Power Administration. 41foot height. Test frequency 1000 kc.80 (consult references 10 and 16 for more accurate calculations).K VLL “: KILOVOLTS Fig. test frequency 830 kc. I I iI 300 /. 32Radio influence and corona loss measurements on an experimental test line. Ref.
Bull. June 1216. Foust. Cahen. Pelissier. Dwight. Trunsuctions. 458. 10. W. 1833. W. see :L~SO Appendix. Sot.Z.E. 1942. 1924.C. 410. 43. Influence. A. W. 25. by 0. H.l England. Fran~aise des Electriciens. S. A. F. C. A. Elektrotechnik und Mnschinenbau. Ijesert ~Ieasurements of Corona Loss on Conductors for Opera1ion above 230 KV. illonteith. 1949 (Published for one year trial use). F. Peterson. a Haute Tension (CICRE).~~~mnrc/ricol C~mpor~et~. .E. A.015 to 25 megacycles.sion of Power andSignals (a book). F. Carroll.E. 47140. A. des Grands Res.E. Philip Sporn. l)ic~lcctric Phenomena in HighVoltage Engineering (Book) F. . 1950. Cozzens. Trunsuclions Vol. i\l.Z.E. VIII. Part II. 51. Cozzens. AIEE Paper 493. 33. Woodruff. 3./s. P. D. Especially on Corona Behavior of Bundle Conductors. Vol..osscs from Conductors I.E. F. Recherches Esperimentales sur lc Comportemcnt des Conducteurs des Lignes a 400 KV. Electr. page 25. 1948. Electrical Engineering. 15. 1929.ine ConductorsTidd 500kv Test Lines. E. v. 1935. Vol.O. 9. Proposed American Standard Specification for a Radio Soise MeterO. 1948. 1949. Jones. 21. 1). Die Wirtschaftliche Bemessung von BundelleitcrLeitungen Elektrotechnik und . 1950. 15.\L%.1Iethod of Calculating Corona Loss from HighVoltage Transmission Lines. 13. 26. 46. New Techniques on the AnncomElectric Analog Computer.I. 1~. 12. Vol. \V.I. Slemon. Stickel.Z. Int. 67. July 26. %.\Iecnnisme de 1’Effet de Couronne sur les Lignes de Transport d’Energie en Courant Alternnt. A. Monteith. 1936. Corona Losses at 230 XV with One Conductor Grounded. Revue Generale de I’Eleclricite’. 28. 44. Prinz. 7’raUsnclio7~s Vol. l!LlO. page 809. Conf. Appendix C by S. 51827533. S.E. Cthen. Measurements Pertaining to the Coordination of Radio Reception with Power Apparatus and Systems. 6th Series. Fruncaise des Etectriciens. AI. Mnngoldt. by II. S.I. 1932.E. p l12.E.E. Tmnsactions.Z. 1947 pages 15717. G. and RllA. Borgquist. Browr.Z. 14. R. June 24 to July 3.I. Vol. A.pages 172733. 1950. General fllcclric Review Volume 33. Carroll. Pelissicr. C. Technical Information and Documents Unit 40. Trunsuctions Vol. J. Ailleret. Greene. Jr. 1937. Rathsman. A. R. F. Electrical Works. A. John Wiley and Sons. L. 2. 1!)34.. H. The Swedish 380 KV System. C. J. as presented AIEE Convention. Effect of Earthing on Corona Losses. 52. 41. Skagerlind.E. Vol. S. Veroff. 1933. Conductor Diameter rind Length of Insulator Strings. 1931. RadioNoise Influence of 230KV Lines. Rorden. CIGRE paper 412. V. A. A. 67. I<:i:in ncues Verlustgesetz der Wechselspannungskorona. J. L. F. pages 192201. Radio Influence from IIigh Voltage Corona. F. F. 19I!).E. B.G.Ifaschinenbuu. J. Conference Internationale des Grands Resesux Electriques a Haute Tension. SE. 1933. Development of Corona Loss Formula (discussion of reference 15).E. 37. pamphlet. January 1919. hlarkt. S. 7/8. Bundelleitungen. 13. Carroll. Rathsman.. 130 G. Corona Loss Xleasurcments for the Design of Transmission Lines to Operate at Voltages between 220KV and 330KV. by W. 1948. R. Transactions Vol. Inc. 26. Fran&se rles Electriciens. A. ‘1. 682. . Carroll. M. A. Series Capacitor and Double Conductors in the Swedish Trsnsmission System. 6th Series. Peterson. \\‘. Kromer. E. 35. Hickernell. 11. 500KV Experimental Station at Chevilly: Use of Bundle Conductors. 19. Results of Tests Carried out at the 500kv Experimental Station of Chevilly (France). The Transmission of Electric Power (a book). G. AIEE paper No. I. H. W. 53.E. 6. The Brown Boveri Review. pages 623. B. SiemensSchuckertwerke A. J. Television Interference Seldom Comes from Power Systems. Bull. ?‘runsaction. Lewis. Drehstromfernuebertragung mit Bundelleitern. Rcnnudin Bull. W. J.E. Radio Interference Suppression in Canada. 11. July/August. Simmons. E. 67.E.\IcGrawHill Book Co. F. Frick. pages 111825. Trunsaclions. 54. Vol. Pakala. Calif. Harder. D.if. Bo G. ~lecl~~ccrl Trunsmis. l l. 1934. 1~. ThreePhase MultipleConductor Circuits. Cahen. 47. 11.Z. by Edward \V. lfyTherm CopperAn Improved OverheadLine Conductor. L. Transactions Vol. atengele. pages 8487. AI. A. G.Corona on T\voConductor Transmission Lines (In German). A. Schurig and C. C. Peterson. Cahen. Carroll. Empirical . 16. 50. Scsbit. 31. page 410. Comparative Investigation of D. CodeNo.4inch Diameter. 52. page 1590. W. . 1948. Corona Effects. Effect of Radio Frequencies of a Power System in RadioReceiving Systems. 48. p 89. 34.E. L. Jan. 12. J. P. Dec. 1932. Philip Sporn. 16. A. 4960. H. page 293. !!‘runsactions Vol. T. Progress Report on 500KV Test Project of the American Gas and Electric CompanyCorona. TrunsaclionsVol. disc. !I.E. 1947.E. Calif. 1582. /&chid Chnractcristics of Trunsmissior~ Circuits. Ihating and Current Carrying Capacity of Bare Conductors for Outdoor Service. XXXV Nos. C. Jan. 17. Cahen.Vol.4.E. Part II. Simmons. 30. 6th Series. S.C. A. Oct. Sew York. B. 66. 1935. Discussion of Reference 32 by C. Bucrklin. . June 1216. I:varis. R. Part II.E. \Ingncr and R. E. M. L. &c. No. 1926. pages 6891.I. C. 1938. I”. Vrethem. Frick. 42. 8. Strigel. 32. A. F.1943. Pelissier.I. 29. D\vight. Vol. 400KV Transmission Lines with Special Reference to Multiple Conductor Lines (Bundelleitungen). 27. S. L’emploi de Conducteurs en Faisceaux pour L’Armement des Lignes a T&s Haute Tension. SiemensSchuckcrtwerke. paper No. page 558. A. Uiss. Temoshok. 13. Inc. pages 5562.62. by L.E.I. Inc. 33. 11cGrawIlill Book Company. Kimbark. Rockwell. Cozzens. W. (1948 Lithoprinted Edition) Illinois Institute of Technology. by E. 1). Part 2.(a book). AIEE Technical Paper 5085. Transmission of Electric Power at Extra High Voltages. AIEE Paper 47214. Eleclric Journu1. Dinapoli.E. 21 . Cadogan Square. A. John \Vilcy and Sons. S. Transactions Vol. F. I. Wagner. 58. March 1930. electric Power Trunsmission (a book). I~Iickernell. pages 27990. pages 559. *4..E. pages 15839. by L. XIX. Rusck. G. Pasadena. Feb. 23. 1tctlucing Inductance on Adjacent Transmission Circuits. R. A. Relative Surface Voltage Gradients of Grouped Conductors. 36. 40. 1950. 45.I. sur 1’Effet de Couronne. AIEE paper No. Vol. Aggers. IS So. Corona I. T. Busemann.Chapter 3 REFERENCES Characteristics of Aerial Lines 63 1. It. H.W. Elcclrical TVorZd. pages 1025g.pages28491. by Prof. Vol. pages 3643.\Ierrimnn. Electrical Engineering. Trunsacfions. Noycs. Clarke. Carleton. 1949. 39. pages 6778: disc. llethods of Measuring Radio Noise 1940A report of the Joint Coordination Committee on Radio Reception of EEI. 0.E. 62. The Swedish 380KV System. Rusck. Dec.E. D. . 7. 56. Peek. B. pages 16972. Pelissier. Item So. A. J. British Intelligence Objcctives Subcommittee. Corona Loss ~Ieasurements on a 220KV BOCycle ThreePhase l:spwimental Line. and Other Factors. pages 537. S. 18. Carroll. Blakeslee. Final Report So. A. BerlinSiemensstadt.A. L. (a book). Number 3. 66. Vol. Jan.E. as presented AIEE Convention Pasadena. by \\‘. Westinghouse Technical Night School Press. 79. F.disc. Snyder. Henning. by C. Crary. 1Vissenschuftliche Veroeftentlichungen Aus Den SiemensWerken. du Diametre et du Profil des Cables des Lignes Aeriennes a Tres Haute Tension. Sot. 99.and A. Markt. 1948 session. Radio Influence. S. Vol.E. 5. 38. Clearances. London S. Transactions V’ol. l!)lL). IIcsistancc and Rcactancc of Commercial Steel Conductors.
before the sheath is applied. Oilfilled paperinsulated cables are available in single.and threeconductor types. and sheathing materials. This cable is insulated and sheathed in the same manner as solid cables. Because cables for power transmission and distribution are cornposed of so many different types of insulation. The belted assembly consists of the three separately insulated conductors cabled together and wrapped with another layer of impregnated paper. l(a) and I(d) indicates.o a. Sample sections of paperinsulated singleconductor. Williams ABLES are classified according to their insulation as paper. (b) Threeconductorbelt&. wrapped with a metallic binder tape. and (c) respectively. (d) Singleconductoroilfilled. In the shielded construction each conductor is individually insulated and covered with a thin metallic nonmagnetic shielding tape. Singleconductor oilfilled cable consists of a concentric stranded conductor built around an open helical spring core. and decrease the thermal resistance. lPaperinsulated Courtesy of General cables. conductor in successive layers to achicvc a requiretl dielectric strength.ages from 1 to 69 kv. (b). and this is the insulation generally used for cables operating at 10 000 volts and higher. varnishcclcambric. Solid paperinsulated cables are built up of layers of paper tape wound onto the conductor and impregnated with a viscous oil. and . Paper insulation is impregnated in different ways. Threeconductor oilfilled cables are all of the shielded design. S. which serves as a channel for the flow of lowviscosity oil. and have three 64 (a) Singleconductorsoli~l. Fig. but the material shown here covers only single. Muller. Oilfilled cables are relatively new and their application has become widespread in a comparatively short time.andthreeconductorshielded cables are shown in Fig. compactsectorconductors. solid threeconductor cables are standard from 1 to 46 kv. (e) Threeconductoroilfilled. CHARACTERISTICS OF CABLES Rwised by: J. reduce corona formation. l(e). as a comparison of Figs. shielding tape only three mils in thickness is used. over which is applied a tightfitting. or asbestos. Reasonable estimates of clcctrical characteristics for cables not list4 can be obtained in most cases by reading from the table for a cable having similar physical dimensions. Multiconductor solid cables are also available. estrudcd lead sheath. l(a). (c) Threeconductorshicldcd. compactsector conductors. each c of these materials having unique characteristics \vhich render it suitable for particular applications. Pnpcr can be wound ont. Solid singleconductor cables arc standard for all volt. The oil used is only slightly more viscous than transformer oil. oilfilled. and accordingly cables so insulated can be subdivided into solid. Jr. hollowstrandedconductor. conductors. the three conductors are then cabled together. This construction is shown in Fig. rubber. compactsector conductors. Threeconductor cables are of either belted or shielded construction.~omp:~c~tIollrl(l conductor.or threeconductor designs.CHAPTER 4 ELECTRICAL Original Author: H. or belt. N. and sheathed with lead. the discussion here must be limited to those cable designs most commonly used. The purpose of the metallic shielding tatie around each insulated conductor is to control the electrostatic stress. Cable Corporatzon oil channels composed of helical springs that extend through the cable in spaces normally occupied by filler material.threeconductorbelted. or gasfilled types. To minimize circulating current under normal operating conditions and thus limit the power loss.
(% of grcatlv reduced insulation thickness for a given 1. in oilfilled pipetype cables a temporary lead sllwt II can be stripped from the cable as it is pulled into the StWI pipe. elimination of space between the conductor and the insulation. Fig. contained by a steel pipe into which three sin&conductor cables arc pulled..lcrLLtc pressure so that during load cycles oil may flow ljC1t .) (i!) Is. This construction is sketched in Fig. pipetype gasfilled have a similar cross Courtesy of the OlxnileCallender Cable Company Fig. Oilfilled pipetype cable may section. The . in gasfillet1 pipetype cables the lead sheath surI‘~)(lnding each conductor remains in place.. of either the oil. The principal advantages of such a conductor are: reduced overall diameter for a given copper crosssection. ll. Examples of oil. except that longitudinal flutes or other channels are provitlctl at the inner surface of the sheath to conduct rlit.W. The singleconductor type employs construction generally similar to that of solid ~::~t)lcs. 5. More recently a crushed stranding that results in a compacted sector has been developed and has found widespread use for conductor sizes of l/O .rogcn along the cable. ‘I‘tw physical and electrical characteristics are fairly well 1\11o\vn. The threeconductor design employs channels in the filler spaces among the conductors.. of :Ln~lsince operating records show that this cause accounts for :Lsignificant percentage of all highvoltage cable faults. thereby preventing the entrance of \\.llBvclops in the sheath. which results in higher Ilighpressure cables. with nitrogen illtr~Aucct1 both inside and outside the sheath SO that no ‘lift‘erential pressure develops across the sheath. Its use in smaller conductors is not practical.p~~r:~ting VoltAge. 2 and 3. This action prevents the c.cac*llrrcnce a fault caused by moisture in the insulation. but their specifications are not yet standardized. no highpressure oil or gas is introduced directly inside the lead sheaths.cr at.ic)n void formation in paper insulation permits of t I. the :w Ils~l:tl wnge for threeconductor oilfilled cables is from 23 1. Fig. 4Crosssectional sketch of compression cable.le an(1 its connected rcscrvoirs is maintained under m.~~~cnt.I~~(:s usetl for voltages ranging from 69 to 230 lw.:. and larger. 4.jI. are being used widely for the higher range of voltages.G.Chapter 4 Electrical Characteristics of Cables 65 remains fllGc1at all operating temperatures. Alost highpressure cables are designed SO that the oil or gas filler comes into direct contact with the conduct. Courtesy 01 Genernl Cable Corporation. the first departure from concentric round conductors was the adoption of sectorshaped conductors in threeconductor cables.. During recent years there has been a trend toward the modification of cable conductors to reduce cost and improve operating characteristics. . through any crack or opening which . The oil in the cLLt. and serves to fill per :tll cAlc voids and exclude moisture at faulty points in the slicath or joints. IIIIII:~ like those provided in oilfilled threeconductor (. particularly in multiconductor cables.and gasfilled pipetype cables are shown in Figs.vc>en cnblc and the reservoirs to prevent the develthe . Singleconductor oilfilled . ‘1’1~11sr~1 application calls for pressure of about 200 pounds I)~Ysclllare inch. The immediate presWOOof the iron pipe makes difficult the calculations of (‘iwuit impeclance.:I1)1(x The gas is normally maintained between 10 and 15 I)IHII~~IS square inch gauge pressure.. the dcfectivc point.. it is intlcctl a real advantage. particularly the zerosequence quantitiw.or ills\llation.nl(~~lt of voids or excessive pressure in the cable.l.4. 3Crosssection of highpressure cable. but voids within the solidtype insulation are prevented by pressure exerted externally on the sheaths.:tt.Lt oil {vi11seep orit. 2Highpressure pipetype oilfilled cable. Compression cable is another highpressure pipetype cable in which oil or nitrogen gas at high pressure is introduced within a steel pipe containing leadsheathed solidtype singleconductor cables.~ Another advantage of oilfilled cables is . Referring to Fig. ( iasfilled cables of the lowpressure type hare recently t)wome xtnndard up to 46 kv.or gasfilled vari(\ty.
Electrical Characteristics of Cables Chapter 4 concentricstrand type. 6 and 7. This is unlike overhead transmission line theory where dimensions are in feet. insulation resistance. are given in Figs. must have a smaller radius so (h) Photographs in this figure furnished by the OkoniteCallender Cable Company Fig. Many equations contain a factor for frequency. if a tubular conductor having an infinitely thin wall were substituted for the solid conductor. elimination of many longitudinal channels along which impregnating compound can migrate. charging current. particular emphasis being placed on quantities necessary for the application of symmetrical components. dielectric loss. its flux would necessarily all be external to the tube. singleconductor cables may or may not. (6) Rope stranded. One component of conductor reactance” is normally calculated by evaluating the integrated fluxlinkages both inside and outside the conductor within an overall twelveinch radius. or distance to equivalent earth return are espressed in inches in the equations in this chapter. While most singleconductor cables are of the . The symbols used in this chapter for various cable dimensions. capacitance. segmental. (11)ITollow core.‘. (b) Compact round. retention of the close stranding during bending. All the threeconductor cables inherently satisfy this condition by the nature of their construction. diameter. f. A theoretical tubular conductor. and for solid cables. (cl Annular stranded (rope core).2 A general rule is that regardless of the complexity of mutual inductive relations between component parts of individual phases. 1. the use of inches when dealing with cable construction seems appropriate.? Geometric Mean Radius (GMR)This factor is a property usually applied to the conductor alone. electrical breakdown. in order to be inductively equivalent to a solid conductor. low ac resistance due to minimizing of proximity effect. (cl Noncompact sector. (a) Standard concentric stranded. annularstranded. or hollowcore. which is the circuit operating frequency in cycles per second. they may also be compactround. some of the flus lines lie within the conductor and contribute to total fluxlinkages even though they link only a portion of the total conductor current. Several factors have come into universal use for defining the crosssection geometry of a cable circuit. (f) Segmental. Unsymmetrical spacing and change in permeability resulting from different phase currents when certain methods of eliminating sheath currents are used. ELECTRICAL CHARACTERISTICS (4 The electrical characteristics of cables have been discussed comprehensively in a series of articles’ upon which much of the material presented here has been based. This chapter is primarily concerned with the determination of the electrical constants most commonly needed for ponersystem calculations. Those physical characteristics that are of general interest in electrical application problems have been included along with electrical characteristics in the tables of this section. (4 Compact sector. and some of these are covered in the following paragraphs. 5Cable conductors. although usually the error is small in calculating shortcircuit currents. and depends on the material and stranding used in its construction. All linear dimensions of radius. both for singleconductor and threeconductor types. the method of symmetrical components can be applied rigorously whenever there is symmetry among phases. Considering a solid conductor. separation. and thermal resistance. may produce dissymmetry. I. Geometry of Cables The space relationship among sheaths and conductors in a cable circuit is a major factor in determining reactance.
(2) Fk YTypical sector shape of conductor ductor cables. Sometimes in calculations involving zerosequence renctnnces. where the subscript denotes reference to only a single actual conductor. t. which applies to the group as though it were one comples conductor. THREECONDUCTOR CABLE SINGLECONDUCTOR CABLES UNSYMMETRICALLY SPACED. (1) The component of circuit reactance caused by flux outside a twelve inch radius is widely identified as “reactance spacing factor” (zd) and can be calculated directly from the GMD: GM% f xd = 0. The factor by which actual radius must be multiplied to obtain GMRI. This equivalent radius is called the geonw~xic mean radius of the actual conductor.hat the fluslinkages present inside thesolid conductor but :tIwnt \vithin the tube will be replaced by additional link:~gc>s Ixt\vcen the tube surface and the limiting cylinder of t\\x!lvcinch radius. The subscript here denotes that this G>ID applies to separations among three conductors. The return path can in many cases bc a parallel group of wires.or negativesequence reactance of a threephase circuit depends on separation among phase conductors. but also in the more complex case where two circuits each composed of several conductors are separated by an equivalent GRID. 6.=S for an equilateral circuit. A solid copper conductor of radius d/2 (xn Ix: replaced by a theoretical tubular conductor whose r:ttlirls is O.chapter 4 . The positive. the reactance spacing factor is negative so as to subtract from the component of conductor reactance due to flux out to a twelveinch radius. is a term that can be used in the expression for external fluslinkages.2794 lo&. so that a geometric mean of all the separations between the conductor and each of its returns must be used in calculations. Fig. 11. and the other estending outward from this cylinder to the current return path beyond which there are no net fluslinknges.. the equivalent separation may be calculated by deriving a geometric mean distance from the cube root of three distance products3 (see Chap. can be expressed as a GMD.3 The fluslinkages per unit conductor current between the 12inch cylinder and the return path are a function of the separation between the conductor and its return.c.:tnc*e calculations without further reference to the shape or rnakcup of the conductor.. denoted here as When the equivalent separation is less than twelve inches.12 60 ohms per phase per mile. care . = YSa. . not only in the simple case of two adjacent conductors where it is equal to the distance between conductor centers. 3. the equivalent separation between cable conductors and the sheath of a nearby cable. (18). Geometric Mean Distance (GMD)Spncings among conductors. GMD3. or the equivalent separation between two nearby sheaths. S.. one extending inward from a cylinder of 1% inch radiusas discussed in the preceding paragraph. Because these and other versions2 of geometric mean distance may be used successively in a single problem. This quantity can be used in re:w~. The total fluslinkages surrounding a conductor can be divided into two components.LEAD SHEATH INSULATION Electrical Characteristics of Cables LEAD SHEATH 67 SINGLECONDUCTOR CABLES EQUILATERALLY SPACED PHASE C \ . as can occur in cable circuits. Also.i7'3 (l/2. Using the terminology in Fig. denoted herein I)y GAIII. 6Geometry of cables. therefore.. If the conductors are arranged other than equilaterally. S. simplification may result if the three conductors comprising a threephase circuit are considered cu a group and converted to a single equivalent conductor. are important in determining total circuit rcactnnce. A distance that represents the equivalent spacing between a conductor or a group of conductors and the enclosing sheath can be expressed as a GMD. 3): GMDx. or between conductors and sheaths. Geometric mean distance. varies with GMRB. but transposed along their length to produce a balanced circuit. If the conductors are equilaterally spaced the distance from one conductor center to another is equal to the GND among conductors for that circuit.. This procedllre is illustrated later in Eq. used in threecon stranding or hollo\vcore construction as shown in Chap. BUT PERFECTLY TRANSPOSED Fig. This requires the use of a new GMR. The zerosequence reactance of a threephase circuit may depend on spacing among conductors and sheath as well as among conductors.
inches 0. with the outer filaments of the conductor carrying more current than the filaments closer to the center. except for heat flow calculations.018 1.29  RATIO t Fig.15 1.17 1.. several geometric factors are required for complete definition..91 2.00 2. SGeometric threeconductor factor shielded for singleconductor cables having round cables.25 1. 1. proximity effect increases the apparent impedance of these cables appreciably.72 1.97 1. This results in a higher resistance to alternating current than to direct current. AT  Conductor Size (Circular Mils) 211 250 300 400 500 600 800 1000 1 500 2 000 2 500 3 000 600 000 000 000 000 000 000 000 000 000 000 000 Round ConcentricStranded  nner Diameter of Annular 1 !stranded Conductor.528 0.24 1. however.. are shown in Fig. ductors.04 1.01 1.. or conductors. This geometric factor is not applicable for heatflow calculations in shielded cables... This phenomenon is called proximity effect.00 1.50 Diameter inches 0.80 2.046 1. 9.12 1..68 Electrical Characteristics of Cables Chapter 4 must be taken to identify and distinguish among them during calculations.336 1. (3) Because of the various possible combinations of conductors and sheaths that can be taken in a threeconductor belted cable. Some skineffect ratios are tabulated in Table 1 for stranded and representative hollow conductors. The increase in resistance is negligible except in very large conductors.026 1.893 1.006 1. and is commonly called skin effect. but for larger conductors it becomes quite appreciable.439 .814 0.’ Proximity E#ectThe alternating magnetic flux in a conductor caused by the current flowing in a neighboring conductor gives rise to circulating currents.01 1. In small conductors this ratio is entirely negligible.630 0..16 1.. and heat transfer. leakage current.. . The directcurrent resistance in cables can be taken as the resistance of solid rod of the same length and crosssection. Positive. CL outside diameter of conductor. ( )uter I Xam. because these characteristics depend on a field or flow pattern between conductor and sheath. = Geometric factors for singleconductor cables can be ‘read from Fig. $ where : ri = inside radius of sheath. to those for singleconductor cables.825 1.75 1.239 1.152 1. become important under certain conditions of cable installation. Two of these factors.145 1.08 1. which cause an apparent increase in the resistance of a conductor.and NegativeSequence Resistance Skin EfiectIt is well known that the resistance of a conductor to alternating current is larger than its resistance to direct current. Booth.06 1. 5 and 6. Proximity effect can. dielectric loss. but increased two percent to take into account the effect of spiraling of the strands that compose the conductor. When cables are laid parallel to metal beams.36 .39 1.. charging current.03 1. NOTE: This is approximately correct for shielded sectorconductor cables if curve is entered with the dimensions of a roundconductor cable having identical conductor area and insulation thickness. The shielding layer establishes an equipotential surface surrounding each conductor just as a lead sheath does for singleconductor cables.20 1.412 1.02 1.” This factor is applicable to the calculation of such cable characteristics as capacitance. etc. The heat conductivity of the threemil shielding tape is not high enough to prevent a temperature differential from developing around the shield circumference during operation: this poses a more complex problem than can be solved by the simple geometric factors given here. Hutchings and Whitehead have made extensive tests on .303 log.012 1. and must be considered when figuring the 60cycle resistances of large conTABLE ~DIMENSIONS AND 60CYCLE SKINEFFECT RATIO OF STRANDED COPPER CONDUCTORS 65’C.63 1. as is frequently the case in buildings or ships.52 1.02 1. ( )uter I Zatio I1iam.005 1.728 0. The mathematical expression for geometric factor G in a single conductor cable is G=2. the ones applicable to positive. 8. The ratio of the two resistances is known as the skineffect ratio.28 1.068 1.and to zerosequence electrical calculations.031 1. .01 1.575 0.   I0.631 1.. walls.998 Ratio  I I Iatio 0. Geometric FactorThe relation in space between the cylinders formed by sheath internal surface and conductor external surface in a singleconductor leadsheathed cable can be expressed as a “geometric factor.09 1. When alternating current flows in the conductor there is an unequal distribution of current.. Geometric factors for threephase shielded cables having round conductors are identical. See Sets. 2.
enter the curve with the dimensions of a roundconductor cable having identical conductor area and insulation thicknesses.6 t/T D CABLE? ( 0. For cables having sector conductors.2 2.0 0. Multiply the resultant geometric factor by the sector correction factor given above. turn in sheath.2 0.8 .6 I 0 0.2 I.O 9 !i w m 3.0 I 1.4 0.4 RATIO + Fig. 5 and 6.4 I. with three conductors paralleled and re .6 11 1.) GO is calculated for singlephase operation.4 I (1 1.chapter 4 Electrical Characteristics of Cables I.2 I I 1.8 ’ 2x) 2. 9Geometric ‘OTE: factor for threeconductor belted cables having round or sector conductors.8 1. (G1 is calculated for threephase operation.6 0. FOR THREECQNDUCTOR BELTED CABLES pT! L 1. See Sets.6 c G.
The additional losses caused by placing a metal plate or other structural shape close to a cable circuit arise from both hysteresis and eddycurrent effects within the plate. Some practical passibilities.16 0. Eddycurrent losses depend on the magnetic field strength at the plate. are listed here: a. because flux developed by cable currents could concentrate within the lowreluctance plate.04 ’ 0. SPACING  BETWEEN BETWEEN SHEATHS SHEATHS REACTANCE RESISTANCE AT 50 CYCLES AT 50 CYCLES I I I I I III III 0. The magnetic plate can be shielded by an assembly of laminated punchings. REACTANCE RESISTANCE structural shape of the surface. it is seldom easy to theorize on which material or combination of ma z 0. illustrate forcefully that prosimity effect can be significantly large.08 0. and also upon the resistance of the paths available for flow within the plate. drawn from experience in the design of switchgear. and because the action of eddycurrents to counteract the incident flux would be comparatively small in a highresistance material. loGeometric factor for threeconductor shielded cables having sector conductors. transformers. terials will contribute lowest losses. SPACING BETWEEN BETWEEN BETWEEN AT 50 AT 50 SHEATHS SHEATHS SHEATHS CYCLES CYCLES Fig.02 iii ” 0. taken from this work. SPACING (b)3lN. .07 (ok6 IN SPACING (b)3IN. Figures 11 and 12. The laminations normally have low eddycurrent losses and they must be designed so that flux density is not excessive.00 RATIO 0. 12Increase in cable resistance and reactance caused by proximity to steel plate for threephase systems (cable sheaths are insulated). and generators. Il. in terms of insulation thickness T and mean periphery P. so that flux is diverted from the plate and into the laminations.09 0.70 Electrid Characteristics of Cables Chapter 4 the impedance and currentcarrying capacity of cables. as they are affected by proximity to flat plates of conducting and magnetic material. placed so that the magnetic field acts to build up or 0 ’ ’ 0. Because the factors that affect hysteresis loss and those that affect eddycurrent loss are interdependent.01 z “‘0 I 2 3 4 FROM 5 STEEL 6 7 PLATE8 9 IO DISTANCE INCHES Fig. b. placed between the cables and the plate. The magnetic plate can be shielded with a sheet of conducting material. Hysteresis losses are large if the flux density within the plate is high throughout a large proportion of the plate volume.12 + 0. that calculation and prediction is difficult. and by the w z 2 2 E 0 0. SPACING cc)I IN. The results in an actual installationof cables close to metal surfaces are inHuenced so greatly by the material involved.Increase in cable resistance and reactance caused by proximity to steel plate for single phase systems (cable sheaths are insulated).20 (a)61N. Although these tests were performed at 50 cycles it is believed that the results serve to indicate effects that would be experienced at 60 cycles. i\ material having high permeability and very high resistivity would promote hysteresis loss.0 I “‘0 I 2 3 DISTANCE 4 FROM 5 STEEL 6 7 PLATE  B 9 INCHES IO Fig.04 (r “0 DO3 l2 0. such as copper or aluminum.
however. rO= 1. and is the distance between 43 conductor center and sheath (9) center for threeconductor cables made up of round conductors. additional I*R losses develop flow ill 11~Sheath.+ ohms xm2rs XIII2 rs2 + per phase per mile. is sheath resistance from Eq. the resistance in.4 l’rosimity effect also has an important bearing on the c:lll. This subject is discussed in the Swl ion on currentcarrying capacity. GIMD~. The increm(:n~sin resistance and reactance do not. (9).ly with current density. For sectorshaped conductors an approximate figure can be had by using Eq.833 inches and the conductors are sector shaped.156)] = 0.998 inches. Ti = inner radius of lead sheath in inches. 1 &=&0. The sheath loss in a threeconductor cable is usuall> negligible except for very large cables and then it is important only when making quite accurate calculations. These curves cover only a for 1’~ sprcific cases. when is l:lrg:(: cables approach very close to steel. by a nonmagnetic steel. (1. s 0 I ohms per phase per mile.664 inches. 71 counteracting circulating currents within the conducting sheet: these currents considerably reduce the magnetic field strength at the plate.. (8). Eq. (7) where re is the alternatingcurrent resistance of the conductor alone including skin effect at the operating frcquency. (6). T = conductor insulation thickness.. 11 and 12 are t):ts(:(I on tests performed at approximately twothirds of m:1simllm current density for each cable used.t1:ln on resistance as Figs. t)rlt. (5) ohms per phase per mile.833 inches.~c somewhat lower. The common way to represent these losses is l)Y incwasing the resistance of the conductor involved. Example IFind the resistance at 60 cycles of a ‘750000 circularmil. (5).rontcarrying capacity of cables when installed near sl~(:l plates or structures. ‘l%(: effect of parallel metal on reactance is much larger I. Xonmagnetic: steel is of particular benefit when the structure near the cable circuit partially or entirely surrounds intlividual phase conductors. 44160(&)” ?= r (r +T. These fiKllrcs also show that the magnitude of the increase in imI1. either entirely or partially.nges in the sheaths. Actually. The curves of Figs. and give merely an indication of the import. is the resistance of tlw sheath in ohms per phase per mile. d = conductor diameter.2i9I & log. (4) x. Since the overall diameter is 2. change grwt. When desired the sheath loss in threeconductor cables can be calculated from the equivalent resistance. for lead sheath. IhiS increment in resistance can be calculated by the fol1°\\iw equation. For other arrangements the geometric mean distance among t. This variation :tg:bin is less than variations in materials and has not been :tcc:o~mted in Figs. so I where r.s become higher and the reactance increments be.:! GU2r8 r=ohms per phase per mile. For “ingl(!contlrlctor cables operating in threephase systems. is the mutual reactance between conductors and slle:lth in ohms per phase per mile.) to positive. and ri are sheath radii defined for Eq.. can be used instead of S with results sufficiently accurate for most practical purposes.? + T.2 lrwc . c. including the effect of sheath currents. The magnetic material can be interleaved with condrlcting bars that are bonded at the ends so that circulating currents develop to counteract the incident magnetic field as in (b). 11 and 12 indicate..)Z X 1O6 ohms per phase per mile. From Eq.r. is ra = r. In these largest cables the sheath losses are ahollt 3 to 5 percent of the conductor loss. The conducting sheet must have sufficient crosssectional area to accommodate the currents developed. The overall diameter of the cable is 2.Chapter 4 Electrical Characteristics of Cables and 0. (6) Thus the total resistance (r:. Nonmagnetic steel has low permeability and high resistivity when compared with conventional steel plate: these char:icteristics do not act in all respects to reduce losses. 1 &===(d+2T). From conductor tables (see Table 10) the diameter of an equivalent round conductor is 0. More detailed information can be found in the reference listed. 11 and 12.~(l:mcc independent of conductor size.Y). the derivation of which is given in refervII(!(:s:I. The magnetic plate can be replaced. and r. = outer radius of lead sheath in inches.84)+2(0. the net effect is often a loss reduction.:mceand magnitude of proximity effect. When the sheaths are contin110115 bonded together at their ends so that sheath cur:tntl 1’(‘11t s I~:LJ~ longitudinally.998(0. except that cl should he 82 to SG percent of the diameter of a round conductor having the same crosssectional area. the variation is only about I pwcent per 100 amperes. In threephase systems the mitltllc cable of the three is influenced less than the outer ones by the presence of the parallel steel.:r(xmrnt. T.or negativcsequence current flow in singleconductor cables. (7) applies rigorously only when the cables arc in an equilateral triangular configuration.hree conductors.417 inches (8) . Sheath Currents in CablesAlternating current in 1IWwntluctors of singleconductor cables induces alternatiw volt. These two quantities can be determined from the following equations: G = 0.200 rs=(To+TJ(ToTJ in which S = spacing between conductor centers in inches. and are of relatively littlc importance in most practical calculations. r. threeconductor belted cable having 156 mil conductor insulation and 133 mil lead sheath.
cables or x1=x2 = z. 3.ante is due to conductor skin effect. which takes into account the effect of sheath currents. r. The total positive. If.log. but actual measurements indicate that for all practical purposes these losses are negligible with present designs and can be ignored in most cases. the conductor resistance. ~GMRI. and these results are shown in Table 2. Because the calculations are complex. conductor proximity effect. The conductor configuration for these tests lens a triangular grouping. 13Positivesequence resistance in steel pipe (estimating of highvoltage curve).2794 f log. with the group lying at the bottom of the pipe. GMDs..=0.557C2. Three Conductors in Steel PipeTypical values for positive. (8).and negativesequence currents can be calculated quite simply as: X1=x2=&+xds1=22=0. (5) and (6). 60 ohms per phase per mile a 0 5 a GM&. 60 xm2 rs2 + ohms per phase per mile.. Theoretically some allowance should be made for the losses that occur in the metallic tape on the individual conductors of shielded cable.+zd ohms per phase per mile .). (C. = geometric mean distance among three conductors (see Eq.2794 f loglo GMR1.. From Eq.=geometric mean radius of one conductor. ThreeConductor CablesBecause negligible sheath current effects are present in threeconductor nonshielded cables. The increased resist.3 f x1=22=0. (11) m 9 The conductor component of reactance is 12 x.O04f9 oh& per phase per mile.664)ZX 106 ’ = 0. and an empirical calculating method has been proposed by Wiseman that checks the tests quite closely. Electrical Characteristics of Cables Chapter 4 0. Sheath currents obviously have little effect on the total alternatingcurrent resistance of this cable. If a maximum value is desired. some change in resistance would be expected.and negativesequence currents can be calculated from the following equation.and NegativeSequence Reactances CablesThe reactance of singleSingleConductor conductor leadsheathed cables to positive. (10) or xGZ ohms per phase per mile. 13. (12) 60 where GMR1. including skin effect is 0.005 = 0. MCM (14) (15) Fig. The separation component of reactance is GMDzc f xd = 0.2794 . In preparing Fig. in the presence of steel pipe. instead. and to I?R loss in the pipe itself.557 ohms per phase per mile.091+ . the pipe size assumed for each cable size was such that 60 percent of the internal pipe crosssectional area would have been unoccupied by cable material: choosing a nearest standard pipe size as a practical expedient does not affect the result appreciably. the conductors were to be laid in an approximately flat cradled arrangement.and negativesequence in shielded cable can be calculated as though the shields were not present. = 0.701)(0. Substituting in Eq. 1).133) =0. some tests giving higher losses and some lower than the triangular. 44160(0.091 ohms per phase per mile.701)” =O. only an estimating curve is presented here. The component to be subtracted’ because of the effect of sheath currents is composed of terms defined by Eqs. CONDUCTOR SIZE. The ratio of actual resistance as installed to the dc resistance of the conductor itself based on data obtained in laboratory tests is shonn in Fig.096 ohms per phase per mile. ~ (13) 12 where GMD3. an estimated increase of 15 percent above the resistance for triangular configuration can be used. Positive.and negativesequence resistance of large pipetype cables have been established by test?. the reactance to positive.284 inches. is found that r. GMR1. From Table 6 it.72 and Ti = 1. Actual tests on the flat arrangement produced variable results as conductor size was changed. xx. The resistance to positive.. Field tests have been made on lowvoltage circuits by Brieger14. 13.and negativesequence resistance is then.2794 6o tog.200 “= (2..
pg. .and negativesequence currents can be calculated as though the shields were not present. bare) Uncabledl Cable Sheath (Phase Conductors) Nonleaded Lead Lead Nonleaded Nonleaded Nonleaded Nonleaded Nonleaded Nonleaded Nonleaded Lead Nonleaded Nonleaded Nonleaded Conduits.s in singleconductor cables and proves to be negligible.229 0.125 0.264 0.137 0. 61.534 0.127 0. aa in note 2.189 0.\Iatwial 1938.) 0.101 0. Because the detailed .Chapter 4 Electrical Characteristics of Cables 73 TABLE ZIMPEDANCE OF THREEPHASE120/208 VOLT CABLE CIRCUITSIN FIBRE AND IN IRON CONDIJITS. Vol. and the remaining values are as defined in Eqs. except that conductors arecabled in position. 8s in note 5.603 0. of four conductors arranged rectangularly. except that conductors are cabled in position.729 0. 2.566 0.~ positiveandNegativeSequence Impedance. bare) 000 AWG (3 conductors.152 0. I Cabled6 I Secondary Mains .136 0.359 _500 MCM (1 conductor. Cabled6 _ ZeroSequenceImpedance. 0.676 0.143 0.538 0.236 0.328 0.144 0.349 0. bare) Uncabled* Cabled8 _ 0.135 0.539 0.120 0. with either one or two neutral is maintained only at the duct entrance: a random configuration may develop within the duct. ’ Asvernbly “‘m1 c(Migumtion ’ . conductors inside the phase conductor \Vhere: GMD3. but k. conductors after entering the duct. Simmons can be quoted as authority for the statement that the reactance is from five to ten percent less than for round conductors of the same area and insulation thickness. in NonMetallic by L.433 1.137 I Reactance (Ohms) 0.188 0.156 0.135 0.539 I Reactance (Ohms) 0.” Resistance (Ohms at 25°C. ’ Assernhly ‘l“lia ilrmnwment ’ Assembly taken from “Impedance of ThreePhase .471 0.144 0. Ohms Per Phase Per Mile at 60 Cycles. 6.793 0. (12) and (13).159 0.) 0.  Phase Conductor Size 500 MCM (I per phase) Neutral Conductor Size Conductor Assembly Duct Material (4 inch) Fibre Iron Iron Fibre Iron Fibre Iron Fibre Iron Fibre Iron Iron and Iron __ 0000 A WC (1 conductor. _ _ .015 0. l’or sectorshaped conductors no accurate data on change iI1 reactance because of conductor shape is available.475 0. Three Conductors in Steel PipeConductor skin effectandproximityeffectsinfluence theapparent reactance of highvoltage cables in steel pipe. = S = geometric mean distance among three conductors.169 0. Brieper. bare) 0000 AWG (2 conductors. Febmay assume All conductors a rsnare group.085 (I per phase) .707 0. For shielded threeconductor cables the reactance to positive.814 0.583 0.777 0.135 0. No.113 0. This is true because the effect on reactance of the circulating currents in the shielding tapes has been calculated by the method used for determining sheath effect. Ohms per Phase per Mile at 60 Cycles. in the sequence ABCABC.972 0. of six phase conductors arranged hexagonally.211 0. making it similar to belted threeconductor cable.inJenrbly ’ Assembly Wi~llVl in position. bare) rllllry ’ .079 0. while being Pulled into the duct. of three phase conductors arranged triangularly with three neutral conductors interposed in the spaces between phase conductors.693 _ Cabled’   I Uncabledj I Uncableds 500 MC&l (1 conductor.144 0.629 0. in the sequence (clockwise) ABCneutral. bare) 0000 AWG (2 per phase) 0000 AWG (I conductor.380 0. Phase Conductor Size 500 MCM Conductor Assembly Uncabledz Duct Material (4 inch) Fibre Iron Cabled’ Cabled’ 0000 AWG 12per phase) Uncabled& Fibre Iron Fibre Iron Fibre Iron Fibre Iron Cable Sheath (Phase Conductors) Nonleaded Lead Nonleaded Lead Nonleaded Nonleaded Nonleaded Nonleaded Nonleaded Nonleaded Lead Nonleaded Nonleaded 1 Resistance (Ohms at 25°C.497 EEI Bulletin.772 0.187 0.
when any of the various devices used to limit sheath current are employed. the variable factors in this problem has been presented by Del Mar’. =r.286 7. has an effect upon the zerosequence impedance of cables.286 5 000 3. or in the parallel combination of both ground and sheaths.lO 000 28 000 0.34 620 10 1.. and from selfinductance of the return paths.06~10’ 2. inches. it must return in either the ground.). 14Positivesequence reactance in steel pipe (estimating of highvoltage curve). none in sheath. Table 2 contains information14 useful in estimating the impedance of lowvoltage (120/208 volt) cables in iron conduit.31 1. is given in Eqs. ohms per mile. Each TABLE 3EQUIVALENT DEPTH OF EARTH RETURN (De). The zerosequence current flowing in any one phase encounters also the reactance arising from conductor selfinductance. (1’7) gj log. 3 All return current in ground.1 3. AND EARTH IMPEDANCE (r.22d) ohms per phase per mile.286 7. present in combined form some of the fundamental effects contributing to total zerosequence reactance. much or all of the return current flows in the earth. r. from mutual inductance to the ground and sheath return paths.76 50 2 000 0. cables . and the use of one GMR to represent a paralleled conductor group. therefore. ohms per mile.= distance to equivalent earth return path. D.36~10’ 2.286 1000 2. Cable sheaths are frequently bonded and grounded at several points.89 100 2 800 0. (16) or (17) in terms of impedance to zerosequence currents.40~10~ 3.47 880 0.+r. with the conductors lying side byside at the bottom of the pipe. the curve results should be increased by 15 percent. AT 60 CYCLES Equivalent Equivalent Equivalent Earth Depth of Earth Earth Resistivity Resistance Earth Return. a curve is supplied in Fig. On the other hand. The impedance of the group of three paralleled conductors.73  1 (( 3  where : r. 2 All return current in sheath. AND 2. INCHES Fig. An actual cable installation should approach one of these three theoretical conditions: 1 Return current in sheath and ground in parallel. The method of bonding and grounding. Reactance (meterohm: )re XC inches ohmspermile dims per mile) feet . A calculating method that accounts in detail for of these inductive effects cannot always be identified individually from the equations to be used for reactance calculations because the theory of earth return circuits3. considering the presence of the earth return but ignoring for the moment the presence of the sheath.2 As zerosequence current flows through each conductor it encounters the ac resistance of that conductor.18 500 3.=ac resistance of earth return (See Table 3). (See Table 3). The curve is drawn for triangular conductor grouping.60 20 000 0. ThreeConductor CablesActual and equivalent circuits for a singlecircuit threeconductor cable having a solidly bonded and grounded sheath are shown in Fig. 4. (16) f D.8382 or 2.44x103 5 0. ohms per phase per mile CENTER TO CENTER SPACING BETWEEN CABLES.286 3. which allows much of the zerosequence return current to flow in the sheath.. none in ground. from mutual inductance to the other t&o phase conductors. z.+Te+j(za+z. D. ZeroSequence Resistance and Reactance When zerosequence current flows along the phase conductors of a threephase cable circuit.286 3.06~10~ 8 800 0.286 3. or the sheaths. = T.286 2. with the group lying at the bottom of the pipe.74 Electrical Characteristics of Cables Chapter 4 calculation of these factors is complex. If the grouping is instead a flat cradled arrangement. 3 h I . 15 (a) and (c). ld that serves for estimating reactance within about five percent accuracy.=ac resistance of one conductor. and as it returns in the ground or sheaths it encounters the resistance of those paths.286 2. The resistance and reactance effects are interrelated so closely that they are best dealt with simultaneously.36~10~ .44x10’ 6 200 0.40~10~ 2.36~10~ 2. GMR1.+j0..05 280 0.
=0...//. =S= (df2T) for round conductors in three conductor cables. ohms per mile.=reactance of an individual phase conductor at twelve inch spacing.)(S)? for round conductors.8382 or z.=sheath J resistance. considering the presence of the earth return path which is common to both sheath and conductors. ohms per 2. 12 ohms per mile. (THREE EARTH ACTUAL CIRCUIT SINGLECONDUCTOR (b) I //T//////f/ CABLES) phase per mile. (18 GMR1...+Ig)=0 0 I //////I h.. = +(GMR. x. 0.. when both ground and sheath return paths exist.+Ibo+I. 15. =zO././///t77t/.//m+ . = reactance of sheath.. inches. inches.= reactance of earth return..tual threeconductor and equivalent zerosequence circuits for and singleconductor leadsheathed cables.chapter 4 Electrical Characteristics of Cables 75 GMR.~=r...+Ig )=o 7//. ohms per mile. (I..8382 $ log.+Ico)+(31.. (Refer to Table 3).= geometric mean radius of an individual conductor. s 0 I ohms per mile. ohms per phase per mile. GMDI. = outside radius of sheath. inches.2794 & log. inches. in zerosequence terms is .. Am7mm/ EARTH k3 (IMPEDANCES EQUIVALENT CIRCUIT EXPRESSED IN ZERO SEQUENCE (Cl TERMS) Ti = inside radius of sheath.. =3r. . ohms per mile.//////////N/////m//////h/~~ a EARTH ACTUAL CIRCUIT (ONETHREECONDUCTOR CABLE) (0) D. n/.=r.. .4. total zerosequence impedance is: TERMS) Fk. r..+.200 = (To+ri) (T. 1 go EARTH MODIFIED EOUIVALENT CIRCUIT (IMPEDANCES EXPRESSED IN ZEROSEOUENCE (4 The equivalent circuit in Fig. x. ohms per mile.. = geometric mean distance among conductor centers. (22) The mutual impedance between conductors and sheath.. or z..  20.8382 $ log./ LTV77 . (25) . (23) (24) rofri ohms per phase per mile.+r.+j(3x. is given in terms of impedance to zerosequence currents: 211. inches. . rofri (19) (IO +h+. =0. From this circuit. x..+r.+jO.../. considering the presence of the earth return path but ignoring for the moment the presence of the conductor group. ohms per phase per mile.+xe) mile.=geometric mean radius of the conducting path made up of the three actual conductors taken as a group..+j0.=3r..) & log. 15(d) is a conversion from the one just above it..ri) for lead sheaths..) + (I.+j(3xs+x. The impedance of the sheath. ohms per phase per (20)  IO THREE CONDUCTORS) (21) (REPRESENTS / where : zm : I I r. and combines the mutual impedance into a common series element.
resistance of one conductor.=O.=0. This would represent total zerosequence circuit impedance if all current returned in the ground. =zc ohms per phase per mile.68+j3.203 ohms per mile.338 inches.W.604 inches. Zerosequence impedance is often calculated for all return current in the sheath and none in the ground. The conductor component of impedance is (r.4 + (2. and overall cable diameter is 1. Assume D.to positivesequence resistance is 2. =0. = 0.)(0. ohms per mile. conductor (7 strands) with conductor diameter of 0..= reactance of earth return.87 ohms per mile phase per mile.27+j4. 15 also shows the actual and equivalent circuits for three singleconductor cables in a perfectly transposed threephase circuit.8382 (. r. 2 A.) for threeconductor cables having sector conductors is difficult to calculate accurately. for the tables of characteristics is of practical accuracy.987.870. where the sheaths are solidly bonded and grounded.=3X 1. = IY(GMR~(GMDJ” x.+j(xa+x.37 ohms per mile.+jO. No. = reactance of an individual phase conductor at twelveinch spacing. or zc=rc+~.8382 ro+ri f G log102(GR/IR3cj ohms per (28) If current returns in ground only with none in the sheath.) ohms (31) (30) D.286+j0.87 If all current returned the sheath. f xd= 0. The actual nature of a groundreturn circuit is usually indefinite.32. (0.146=0.987 ohms per mile at 60 cycles.39 ohms per mile. lead sheath thickness is 109 mils. For the sheath component of impedance: 0.74 = 4. r.2.623 =3. 3. GMR of three conductors is: GMR3.G. inches.18+1.27fj4. z0=1. where : rc= a. Conductor insulation thickness is 156 mils.31=4. for three sectorconductors is roughly 90 percent of the GMRsc for three round conductors having the same copper area and the same insulation thickness.2xd) per phase per mile.987 + jO. and also because lowresistance connections between sheath and earth are sometimes difficult to establish.286): 2800X 12 z.732 inches.89fj1.87 = 1.68+j3. = a. ohms per mile.87)2 zo= 1. since it may be mixed up with water pipes and other conducting materials.2794co log. .18+3.+j0. because the magnitude of the answer is usually close to that calculated considering a paralleled return. = 0.c.987+0.= ~(0.c.to positivesequence reactance is 9. = 2800 feet and resistance of one conductor=0.156=0. = distance to equivalent earth return path (see Table 3). Fig. the zerosequence impedance becomes: zo= (z.62312.8382 log.109) 2X2800X12 2. GMR of one conductor is (see Chap. (26) ==r. .286+j3.27+j4.183.0 1. o 338 = 1.726X0. The equivalent geometric mean radius (GMR.286+j3.. If return current may divide between the ground and sheath paths.+~.57j7.= geometric mean radius of the conducting oath made up of the three actual conductars taken asa group.200 ___ = 1. . Example 2Find the zerosequence impedance of a threeconductor belted cable. The impedance expressions applying to singleconductor cables differ in some respects from those for threephase cables: D. as would be the case with nonsheathed cables or with insulating sleeves at closely spaced intervals. = r.9. 11): GMRI.8382 & log.=r. and the ratio of zero.13 ohms per mile ‘“=(1.+j(z:. ohms per mile.106)(0. belt insulation is 78 mils. SingleConductor CablesFig.27+j4... The method used to calculate values of GMRS.76 Elect&d Characteristics of Cables Chapter 4 If current returns in the sheath only.)+z. f z.) ohms per phase per mile..+3r.292 inches. with none in the ground : &I= (2..38+j0.68+j3. ii ohms per mile. and none in the sheath.olog10 ~ GMRsoohms per phase per mile. (27) The mutual component of impedance is: z. and none in the ground.87 = 3.286+jO.44 ohms per mile.8382 log. Distance between conductor centers is: S=O. resistance of earth (see Table 3)) ohms per mile.2~ .18=4. (29) The zerosequence impedance of shielded cables can be calculated as though the shielding tapes were not present because the impedance is affected only slightly by circulating currents in the shields. The positivesequence impedance of this cable is: ZI = 0.zc+z. .+ 3r. x.=O. (See Table 3. inches. D.106 inches. As an alternate basis for estimations. Therefore the ratio of zero.Gn> =.22.31 = 2. it appears that the GMR.604)*=0.292+2XO.z..13+0. hut is not considered to be appropriate for explanation here. GMR3.
...+r.8382 or i. o~.2794 (% log.he sheath only.+j0.. CAI. . log.and negativesequence. inches.k ~ =I 1’ 2’ =0.5) 1OOOGl Io. fk ’ (43) (44) 1. directly in ohms per mile..5 ohms per phase per mile. then the curve in Fig. (25) ZS If current returns in t.5. but some rather complete results are available from field tests on installed lowvoltage cables.8382 log10Gi\/IR3. inches. and zerosequence shunt capxcitances for singleconductor metallicsheathed cables are all equal. ohms per mile 24 =0. total zerowqucnce impedance when both ground and sheath paths csist is: 2 z0= zc. = (GMD#.. = outside radius of sheath.79G megohms per phase per mile. 10...= . (2% Cables in Steel Pipes or ConduitsWhen cables are installed in iron conduits or steel pipes.+~. =r. .= Iz*= I. (41) c =O. ri=inside radius of sheath. ohms per phase per (35) where : GMD3c3s= geometric mean of all separations between sheabhs and conductors. shunt capacitance and charging current can’be derived from the curves of geometric factors shown in Figs. ohms per mile 0. the appropriate geometric factors are given in Fig. is recommended as giving more accurate values of geometric factor. 8 and 9.k 0 323f.. 9..k. (37) If current returns in the ground only: zo= (Z. (46) lOOOG.) ohms per phase per mile. with none in the ground: . applying to round conductors.8382 ro log. 8. _ ri) for k%d sheaths. Shunt Capacitive Reactance Shunt capacitive reactances of several types of cables are given in the Tables of Electrical Characteristics. "O'= f.!(F)(GMD. ohms per phase per mile. ’ s.+r.+O.phast? (36) =r. (4.. Ohms 1 per phase per mile ohms per phase f De in Table 2. and can be derived from the curves of Fig.79Go megohms per phase per mile. (40) (33) where : GMRss = geometric mean radius of the conducting path made up of the three sheaths in parallel =.Chapter 4 Electrical Characteristics of Cables 77 CMD3. (39) f. + 2. =z.=0.222d) per mile & =r. rs = resistance of one sheath.. . The positive.+j0. The geometric factor for threeconductor shielded cables having sectorshaped conductors is approximately equal to the geometric factor.. 8).. x9= reactance of one sheath. ___ GMRX. Asho .xZ. Threeconductor shielded cables having round conductors are similar to singleconductor cable in that each phase conductor is surrounded by a grounded metallic covering. From the equivalent circuit of Fig. HOKever. (42) 0 ___ r TO 0.97fkkv amperes per phase per mile.+j(~~+~. 1.” ohms per phase per mile cw G&IRS. The geometric factors given in these curves are identified by symmetricalcomponent terminology.f=zq'=megohms per phase per mile. ~ ro+ri’ ~~=r~+r~. zs= r. (32) 5.amperes per phase per mile. based on insulation thickness and mean periphery of insulation.O892k microfarads per phase per mile. IoOOG.?I= 2.k~ II. = geometric mean distance among conductor centers.200 = (To+ ri) (T.. as shown Threeconductor belted cables having no conductor shielding have zerosequence values which differ from the positive.. microfarads per phase per mile. 3c _ ohms 3s (34) per phase per mile. the zerosequence resistance and reactance are affected by the magnetic material because it closely surrounds the phase conductors and forms a likely return path for zerosequence current. For singleconductor and threeconductor shielded cables (see Fig. inches. In addition.1. microfarads per phase per mile. z~=r. = US. if the sector shape of a shielded cable is known.>:. G.&JfZ.+j(x. therefore the positive. for any cable whose dimensions are known. and zerosequence values are equal and are dependent upon the geometric factor relating a conductor to its own shielding layer. r.59761 g. negative.ij(~.22. Some special tests of the zerosequence impedance of highvoltage pipetype cable have been mndc but the results are not yet of a sufficiently ivide scope to be generally usable.323jkkv amperes per phase per mile. I 38) ~l~=~2’=~oo’=. negative.2~) mile. so method of calculating this zerosequence impedance is available.+r.
k. This condition arises since the shape of sectors varies and a rigorous calculation is not justified. (4% For threeconductor belted cables. X.015 0.030 GasFillccl (lo\\prcssurc~ 0. 9. taken from specifications of the Association of Edison Illuminating Companies15.09.0 3.0 4.0060 0. (50) In these equations. in accordance with Dr.0 k Typical k 3. is important to note that in converting shunt capacitive reactance from an ‘Lohms per phase per mile” basis to a total “ohms per phase” basis. length in miles’ ohms per phase.009 0. and zeroXl'.010~5 below tbevc values When threeconductor belted cables have sectorshaped ronductors. For singleconductor and threeconductor shielded cables. Insulation Resistance. The calculation of cable insulation resistance is difficult because the properties of the insulation are generally predictable only within a wide range.and negativesequence reactance for sectored cables has arbitrarily been taken 7. negative. included in these tables are other characteristics useful in cable work.0097 0. 7f.0 4. Where sectorshaped conductors are used. This correction factor is plotted in Fig. the zerosequence characteristics are calculated for the case of all return current in the sheath and none in the ground. lo6 ohms per phase per mile.597Gl f~lc~COS 4 I . Table 7 for shielded cables is similar in form to Table 6 and where sectored cables are listed the same approximations in GMR and reactance apply. and zerosequence shunt resistances. negative.0090 0. 1~ dielectric constant (see Table 4). the geomebric factor must be corrected from t.>' = 0. some approximations are necessary as pointed out previously. it is necessary to divide by the circuit length: x’. The equations presented below are therefore quite dependent upon an accurate knowledge of insulation power factor.0 6. It.0 3.04. Table 8 for threeconductor oilfilled cables is similar to both Tables 6 and 7 and the same considerations apply. Simmons’ recommendations.5 percent less than that of an equivalent roundconductor cable.r = Xc’ Solid Paper 0. ku = linetoline system voltage. Where ground must be considered or where there are paralleled threephase circuits. and rot are positive. Cl. are positive. the overall diameter of any particular cable can be calculated. ancl its use is explained below the curve. conductor diameters and GMR’s.=dielectric constant. In these tables for threeconductor cables. this is usually sufficiently accurate because of the indefinite nature of the ground return circuit.0075 0. and zeroseqlience charging currents.021 0. These standard values will very probably be several times larger than actual measured power factors on new cables.025 0.7 5. and the type of conductors normally used in any particular cable. or else constants are given from which these quantities can be calculated. and IO. lo6 ohms per phase f. From the quantities given in these tables of threeconductor cables. Q’ and ~0’ are positive.l~ of new cable is usually II. kilovolts. the impedance must be calculated as illustrated in the examples given.0 Solid Paper OilFilled GasFilled Varnished Rubber Cambric . according to the values in Table 4. .155(d+27’)+2(t+L) (51) I. In each case the positive. and zerosequence resistances and reactances are tabulated. I?.018 0. rind C0 are positive.009 0. (47) 6.013 0. C. II. sheath thicknesses and resistances.06. As pointed out in the discussion of zerosequence impedance. In Table 6 the positive. negative. TABLES OF ELECTRICAL CHARACTERISTICS The 60cycle electrical characteristics of the most usual sizes and voltage classes of paper insulated cable are contained in Tables 6 through 11. In the foregoing equations. The variation in sector shapes probably is greater than any error present in the approsimation given in the tables. negative. The equivalent GMR of three conductors in sectored cables is necessarily an approximation because the GSIR of one sector cannot be determined accurately. TABLE ~DIELECTRIC Insulation CONSTANTS OF CABLE INSULATION Range of 3. = cos +=power factor of insulation. 1.04. and zerosequence capacitances. Also. C?.5 3. In each of these tables the electrical characteristics have been calculated by the equations and curves presented in the foregoing pages. In Table 5 are listed maximum values of insulation power factor. (49) 1‘f= T.. The reactances calculated from these approximate GMR’s are sufficiently accurate for all practical calculations.he value \vhich applies to round conductors.027 by a wide ~nnrgin.022 0. in per unit.79G T1’ Tt’ = T”’= = . sequence capacitive renctances. o=2. negative.78 Electrical Characteristics of Cables TABLE SMAXIMUM Tcmperaturc of Cable (Deg. lo6 ohms per phase per mile.) 25 to 60 i0 80 85 90 *The power factor Chapter 4 POWER FACTORS* OF CABLE INSULATION OilFilled (lowprcssurr~ 0.79Go r”‘=j. such as typical weights per 1000 feet.7 3. rl#.cos 4 .k~COS 4 per mile.04.
53 2. 11.184 0.166 0.310 0.138 0.66 8.271 0.42 1. ?I) 1.987 0.297 0.417 0.113 0.392 0.366 0.141 0.191 0.106 0.73 3.142 0.780 7 0.410 E% 0.220 0.07 5.5 105 170 165 160 155 1R.04 1.74 2.151 0.j 55 55 55 85 80 75 75 75 75 :i: :i 75 75 7. D.211 0.572 0.184 0.216 3.561 0.230 0.19 1.51 2.137 0.39 2.567 0.174 :: E 0. ‘The following symbols are used to designate the cable types.00 0.77  100 .1 1.091 0.79 2.210 0.322 3.128 0.55 15.707 1.700 0.786 0.239 0.:"9.5 65 60 55 5.3 9.380 0.126 0.89 2.j 155 155 1.230 0.084 0.193 0.27i D.230 0.392 0.332 0.191 0.g 1Ok 110 :.08 5.622 0.269 F%. 120 125 130 135 .D6 5.129 0. 833 0.254 0.332 0.133 0.5 105 105 105 105 105 105 10.138 0.151 0.181 0.495 0.08 1.190 0.143 0.7li 1.171 0.180 0.780 0.184 0.54 2.44 t:.39 5.455 0.40 9.34 1.226 11 fflO 10 200 9 on0 8 100 7 !I00 5400 4 500 4 000 3 Ron %i:: 2900 2600 z :g I 12 500 11 200 9 800 9 200 8600 6700 5 100 4 cl00 4200 : 7 ' "! 3400 ".224 0.987 0.214 .171 0.778 0.SO 1.310 0.47 1.121 0.305 Ei Zi 0.13 1.32 2.39 1.712 10. & BELTED PAPERINSULATED ZER PE  I SHI<.373 0.298 ..987 0.245 i%: 0.295 g::g 0.265 0.76 1.s 90 90 85 85 85 85 2 83 85 8.193 0.332 0.215 0.06 5.34 1.254 0.230 0.139 0.366 0.41 3.190 0.263 0.067 0.210 0.s 55 55 53 55 5.199 0.091 2.131 0.436 0.20 3.3 100 ‘J.424 0.132 0.%i 2 500 1.572 0.123 0.31 3.17 2.72 4.64 2.987 0. 0.171 0.091 0.5 95 100 100 !I..50 1.786 0.249 0..179 :: :i‘i 0.t .899 0.58 1.539 0.575 0. CSCompact S&Or.245 0.46 3.3 !I 3 2 Ii’.455 0. .8’ 1 Ii!) 1.216 0. R!) 70 70 .152 0.495 0.23 4.246 EE 0.54 5.263 0.497 .58 0.212 0.885 O..439 0.310 0.126 0.987 0.220 0.387 0. .323 0.292 0.42 4.33 3.780 0.230 0.03 0.204 0.265 0.5 1.69 5.622 0.230 0.218 0.135 0. lli l.364 0.542 ".230 0.: k% Et: 0.68 1.623 0 192 0. SRStranded Round.7.185 0.673 105 100 9.126 0.132 0.221 0.14 1.784 0.69 1.135 0.249 0.415 0.620 0 538 ‘Ac resistance based upon 100% conductivity at 65% including 2% allowance for stranding.:: 3.520 0 3.126 0. ‘For dielectric constant=3.310 0.48 1 7:s 1 63 1.68 2.142 0.280 8%: 0.084 0.\TH 6 i 1 0 Oil: 0000 2cjO000 300000 350 000 400000 500000 6OnOOO 750000 70 in 1 io0 1910 2390 2 820 3 210 3 160 3 650 4x90 4900 5660 6310 7080 8310 9 800 11800 1 680 2030 2600 2930 3440 3 300 3 890 4530 5 160 5 810 6470 7240 8660 9 910 11920 2 150 2470 2900 3280 3660 3480 4080 4 720 5370 6050 6830 7480 8 890 10 300 12 340 2450 2900 3 280 3 560 4090 3870 4390 5 150 5830 6500 7160 7980 9430 10 680 12 740 4350 4640 4990 5600 6230 7 180 7840 7480 8340 9030 10 550 12 030 14 190 0.i 63 ii 2 2 1 0:.256 0.577 0.373 0.58 0.79 5.> .332 0.539 0.758 0.364 0.867 0.224 0.i98 0.495 0.323 0.520 EZ:: 0.194 3.780 2.392 0.39 6.220 0.297 EE 0.134 0.79 4.128 Z: :..373 0.700 0.249 0.572 0.327 0. ‘See Fig.37 1.256 0. 0.57 4.::i': 3.171 0.134 0.40 1 .134 0.190 0.263 0.125 0.135 0.123 0.149 8: ::2” 0 139 6300 5400 4700 4300 4000 2800 2300 2000 1800 1700 1500 1500 1300 1200 1100 6700 5800 5100 4700 4400 3500 2700 2400 2100 1900 1800 1700 1500 1400 1300 8500 7600 6100 5400 5000 3600 3200 2800 2600 2400 2200 2000 1800 1600 1500 9600 0.221 3.327 0.129 0.23 1.07 1.41 3.142 0.31 4.184 0.263 0.246 ::.633 0.800 0.476 0.: .091 2.084 0.373 0.513 0.364 0.28 4.265 0.03 1.178 0.467 0.i 40 40 :i 40 40 40 40 40 40 40 40 40 35 55 50 45 2.262 0.478 0.57 6.572 0.166 0.417 0.50 1.417 E.i 9. O.642 0.536 0.497 0 539 0. ‘CMR of sectorshaped conductors is an approximate figure close enough for most practical applications.151 0.82 3.125 0. none in ground.392 0.1s 1.200 0.297 0.342 iE:o' 0.606 0.700 0.190 0.00 0.74 4.611 0.292 0.88 3.166 0.455 0.3.43 1.129 0.321 ET2 0.148 0.364 0.X1 2.254 0.86 5.263 0.786 0.161 0.304 X:% ::.106 0.5 100 100 103 105 110 110 115 120 123 95 100 100 100 105 100 100 105 110 110 110 115 120 120 125 110 110 110 115 115 120 120 2.091 2.129 0.5 85 130 12.08 0 !XlR 0.539 0.171 0.166 :: Et! 0.5 0 7.j 155 l5.14 6.232 0.99 6.P.323 0.622 0.50 2. Ii4 1.128 0.780 0 292 0.497 0.642 0.206 0.225 0.231 0.191 0.138 0. 2.218 0.256 3.232 0.343 0.528 0.55 0.263 0.X 120 !I0 !JO 0.!llX 0 8.210 0.i !10 !)O 95 9.622 Xi 0.S 1% 165 8300 6800 6100 5700 4300 3800 3500 3200 2900 2700 2500 2200 2000 1800 8600 78M) 7100 6500 6000 5600 5300 5400 5100 4900 4600 4300 4000 I   1.67 Elm 3.‘i 5.106 0.5 5i 5.663 EE 0.700 0.67 3.58 0.259 D.493 0.58 0. % 6600 6200 5 600 5 200 4900 4300 3 900 3 500 15 000 13 800 12 800 12 000 11300 10 600 10 200 7 900 7200 G 900 6 200 5700 5 100 8.291 0..231 EIi 0.265 0.373 0.323 0.3 9. 000 0000 2.156 0.i 0.Chapter 4 TABLE Electrical Characteristics of Cables 79 CABLES 660CYCLE CHARACTERISTICS OF THREECONDUCTOR Grounded Neutral Service POSITIVE NEGATIVESEQ.409 0.327 0.758 0.41' 1. 2.144 0.517 n.202 0.62 .11 2. 'Baaed upon all return current in the sheath.190 0.978 0.147 0.134 0.II 70 70 70 7n 70 70 70 7.2 9.113 0.165 0.106 0.48 1.210 0.497 0.184 0.1s 1.218 0.622 0.067 0.113 0.27 1.43 1.131 0.126 0.122 0.786 0.5 115 110 110 105 10.25 4.470 0.428 Xii 0.353 0.220 0.142 0.232 0.151 0.366 0.:E:: 0.563 0.175 L? E 0.L’.249 0.217 0.185 0.366 0.214 0. 7.06 6.210 0.918 0.184 0.700 0.237 0.220 0.744 0.141 0.88 5. ii) 100 100 .210 0.265 0.50 000 3ooono 350000 400000 500000 600000 750000 6 l 1 0 0:: 0000 250000 300000 350 000 400000 5ooono 6Onooo 75nooo 2 2 1 0 0::: 0000 250000 300000 350000 400000 500000 600000 750000 2 1 0: 000 0000 250000 300000 350000 400000 fjooooo 600000 750000 75 8.142 0.11 4.329 0 302 0.495 0.323 0.171 0.680 0.237 3.232 0. 120 9.067 0.90 4.:.656 0.419 0.292 0.166 :: :i 0.250 0.227 0.151 0.877 0.0.527 0.21 1.292 0.642 0.5000 13 600 11300 10200 9 600 ii % 8300 7 800 7400 7000 6700 6200 5800 5400 16300 14 500 12 500 11400 10 700 .152 0.61 3. 0.89 3.246 D. 'r g 45 45 45 45 45 45 4.63 1.270 EZ 0.417 0.250 0.218 0.83 5.106 0.218 0.50 1.126 0.084 0.297 0.297 0. 110 11.642 0.46 4.6 0.64 1 .067 0.327 0.88 E ~~~~ 2.310 0.159 0.50 1.456 0.07 3. >I) 1.786 0.37 8.642 0.21 7.5.236 0.249 0.130 0.332 0.79 2.191 0.192 0.142 0.97 3.t 1.
66 1.01 1.780 0.98 1.642 0.261 0.408 0.249 0.623 0.4 times the round conductor diameter d..46 1.204 0.220 0.233 0.22 10 470 11290 12 280 13 030 14 760 16 420 18 860 500 000 600 000 750 000 ‘Ac resistance based on 100% conductivity at 65°C.622 0. 120 120 125 125 130 13.256 0.297 0.141 0.249 0. the derivation of the equations given with Table 10 becomes evident. $ ohms per phase per mile (13) and r.207 0.635 0.228 0. and r.585 0.545 ET: 0.170 0.323 0.297 0. necessary for the use of Table 10. according to Fig.19 1.250 0. Table 12 gives the one other quantity.373 0. CSCompact Sector.520 180 900 830 0.48 3.191 0. This equation refers to cables with round conductors.:.532 0.317 0.82 1.265 0.410 0.327 0..29 2.24 2.265 0.248 0.310 0.290 0.197 0.5 173 Ii.417 0.95 2.165 0.622 K% 0.35 1.302 0.364 0.lil 0.166 0.191 0.786 0.210 0. XFor dielectric constant=3.191 0.237 0.03 1.490 0.61 1.292 0.151 O.268 0.273 E!i 0.515 0.488 0.159 0.5 17.40 2.268 0.210 0.366 0.159 0.495 0.222 0.230 0.249 0.663 0..01 2.190 0.230 0.397 0.628 0.50 2.113 0.312 0.562 0.425 0.64 2.243 i%: 6000 5400 5200 4800 4400 4100 3800 3600 3400 3100 2900 2600 8300 7500 6800 6600 6000 5600 5200 4900 4600 4400 3900 3700 3400 9900 9100 8500 7200 6800 6400 6000 5700 5200 49M) 4500 110 105 110 110 115 115 120 120 :3: 135 115 115 120 :.158 0. in which.g 0. negativeand zerosequence characteristics can be quickly determined by using the equations given at the foot of the tabulation.26 2.447 0.737 0..775 0. are conductor and sheath resistances respectively.17 2.217 0.594 0. t = belt insulation thickness in inches (when present).15 4.2794 f log. For sectored cables there is no exact rule.44 3.559 0.418 0.167 0. none in ground.761 0.95 2.562 z:E 0.455 0.165 240 240 240 353 345 345 345 345 345 345 345 345 34.373 0.215 0.327 0.226 0.211 0.J 250 240 240 240 3 860 4 260 4 740 5 090 4 790 5 510 G 180 F 910 7610 8 480 9 170 10 710 12 230 14 380 5 590 5 860 6 440 6060 6 620 7 480 8 8 9 10 070 990 720 G50 0.767 0.633 5.5 345 500 000 600 000 750 000 08 OEEI 250 300 350 400 000 000 000 000 12 280 13 610 15 830 8 9 9 9 .257 0.166 0.249 0.897 Ei 0.290 0.178 0.91 3.80 1.5 175 175 i”.412 0.156 0.323 0.237 .113 0.987 0.530 0.178 0.220 0.188 0. d = diameter of individual conductor in inches.091 0.00 1.134 0.248 0.310 0.308 0.366 0.15 1.322 0. D = outside diameter in inches.7. ‘Based on all return current in the sheath.665 0. T = conductor insulation thickness in inches.16 2.410 ES 0.700 0.332 0.78 3.03 0.90 1.530 0.285 0.566 1.620 0.2794 k log.480 0.134 0.53 2.232 Z:2: 0. (1 l).05 3.275 0.177 0.141 0.151 0. 17..10 2.432 0.594 0.298 0.80 2.187 0.263 0.106 0.177 0.477 0.747 0.721 0. 60 ro+Ti ohms per phase per mile (22) zd=O.693 0.788 0. EThe following symbols are used to designate conductor types: SRStranded Round.265 0.684 0.154 0.366 9900 9100 8500 7200 6800 6400 6000 5700 5200 4900 4500 0.540 0.407 0. 7.879 2. A set of calculated constants is given in Table 10 for singleconductor cables.563 0.65 3.153 0.171 0.226 0.323 0.975 0.377 1.5 135 140 130 135 135 135 135 135 140 140 145 150 155 0. 17.365 0.201 0. but a close approximation can be obtained by using an equivalent cable with round conductors and calculating the diameter D by Eq.232 0.327 0.332 0. .636 0 681 0.249 0.80 TABLE 760CYCLE Electrical Characteristics of Cables CHARACTERISTICS OF THREECONDUCTOR Grounded Seutral Service POSITIVE NEGATIVE SEQUENCE & ZEROSEQUEXCE Chapter 4 SHIELDED PAPERINSULATED CABLES 2o. from which the positive.190 0.243 0.572 0.330 0.285 0. 6.851 0.600 0.561 0.=0. These reactance spacing factors are tabulated for equivalent cable spacings .585 0.280 0.548 0.183 0.246 0.890 0.786 0.285 0.773 0.263 0.166 0.56 0.58 0.230 0.087 0.191 0. including 2Yo allowance for stranding.45 2.73 1.364 0. and then subtracting 0.126 0.091 0.99 3.480 0.171 0.527 0. 60 GM&c ohms per phase per mile (12) 24 f x.441 0.364 0.=0.328 0.139 0.538 0. %d.263 Ei 0.819 0.166 0.252 0.600 0. These equations are derived directly from those given for the calculation of sequence impedances in the sections under Electrical Characteristics.239 0.298 0..532 0.288 0.181 X:%t 0. ZGMR of sectorshaped conductors is an approximate figure close enough for most practical applications.151 8200 6700 6000 5400 5200 4800 4400 4100 3800 Ei 3100 2900 2600 8300 7500 6800 6600 6000 5600 5200 4000 4600 4400 0.60 3.i 1!11) 183 180 17.2794 log.210 0.?92 0. %ee Fig.446 0.491 0.523 0. L = lead sheath thickness in inches.325 0.545 0.210 0.3 to 0.171 0. Since 12 x.468 0.566 0..450 0. depending upon the shape of the sector.5 173 17.
190 0.773 0. the voltage class listed in the first column refers specifically to groundedneutral operation.226 0.165 0.392 0.94 1.635 0.323 0.113 0. :lssuming all zerosequence return current to be in the sllcaths.635 0.151 0.447 0.310 0.168 0.210 0.5 140 140 140 150 150 155 1.73 3.191 0.503 0. Electrical Characteristics of Cables CHARACTERISTICSOF THREECONDUCTOR Grounded Neutral Service 81 PAPERINSULATED CABLES OILFILLED & SEC.0 inches. CSCompact Sector.1c resistance based on 100% conductivity at 65°C.265 0.310 0.147 0.366 0.33 2.227 0 223 0.151 0.160 GO30 5480 4840 4570 4200 3900 3G90 3400 3200 3070 6700 6100 5520 5180 4820 4490 4220 3870 3670 33.69 inches.94 1.191 0.364 0.854 3.327 0. but in cables rated 7000 volts and above.508 0.366 0. can be had by referring in each specific case to the next higher voltage class listed in the tables.5 11.180 0.06 1.455 0. 792 0 i29 0 036 O.  POSITIVE lEGATIVE ‘SHEATH 0:: 0000 250 000 300 350 400 500 000 000 000 000 5 6 6 7 9 9 9 11 590 150 860 680 090 180 900 .150 0. See Fig.247 0.08 1. a greater thickness of insulation is recommended for a givenvoltage class when cable is operated with an ungrounded neutral.543 0.570 0.i 0.533 EE 0.5 123 123 135 13.51 1.50 0.018 0.284 0.928 0 826 0.41 2. A good approximation of the electrical characteristics of these higher voltage cables when operated with other than a solidly grounded neutral.410 0.607 0.538 0.160 0.181 0.366 0.32 2.166 0.575 0.238 0.5.134 0.157 0.788 0.297 0.260 0.232 0.439 0.532 0.If.166 0.5 and 0.265 0.5 140 115 125 12.210 0.77 1.176 0.797 0.190 0.04 1. The constants calculated in this manner apply to one threephase circuit of singleconductor leadsheath cables.249 0.113 0. ‘The following symbols are used to designate the cable types: CRCompact Round.091 0.256 0.64% o.220 0.364 0.55 2 41 2.265 0.618 0.i88 O.623 0.164 0. OF CURRENT CARRYING CAPACITY One of the most common problems in cable calculations is that ef determining the maximum permissible amperes III.195 0.248 0.297 0.200 0. The constants of several typical cables calculated by the methods outlined are listed in Table 13.5 to 36.6“ 2.188 0.690 0.232 0.249 0.GO8 0.510 0.235 0.67 2.172 0. none in the ground.202 0.442 0.87 2.839 0 I342 0. ‘Ikued on all return current in sheath.263 0.447 0.820 0.327 0.168 0.539 0.597 0.OG 2.72 2 58 2 4I 2.56 0. Frequently cable systems are operated with other than a solidly grounded neutral.410 0.297 0.495 0.55 1. 7.478 0.550 0 323 0.249 0.172 0.495 0. In each of the tabulations.310 0.750 0. Singleconductor oilfilled cables have hollow conductors (the oil channel forms the core).220 0.171 0.672 0.272 0 285 0. TABLES .237 0. !I70 0.243 0.171 0.185 0.690 0.392 0.190 0.148 0.767 0.85 1.210 0.640 ::.28 2.274 0.460 0.5 113 125 12. For all spacings less than 12 inches.703 0.195 0.16 2.266 0.178 0.2.513 0.091 0.532 0.134 0.177 0. consequently Table 11 includes cables of the two most common inside diameters.497 0.5 125 125 135 135 135 140 140 135 139 13.44 1.171 0.iGl 0.chapter 4 TABLE 860CYCLE.646 0.“.495 0. none in ground. which should cover the range met in practice.468 0.230 0.220 0.691 0.490 0.208 0.02 0.376 0. This table is similar in form to Table 10 and the impedance characteristics are tlctermined in precisely the same way. including 2% allowance for stranding.210 0. 0.191 0. from 0.532 0.327 0.595 0.165 8330 7560 6840 6300 6030 5700 5430 5050 4740 4360 750 000 1 000 000 18 980 1.230 6030 5480 4840 4570 4200 3900 3G90 3400 3200 3070 6700 GlOO 5520 5180 4820 4490 4220 3870 3670 3350 8330 7560 6840 6500 6030 5ioo 5430 5050 4740 4360 11.217 0. 2d is ncgntive.190 0.5 13.542 0.364 0.539 0.134 0.56 3.213 0.185 0.417 0.30 3.548 0. The GOcycle characteristics of singleconductor oilfilled cables are given in Table 11.436 0.230 0. Representative sizes and types of cable have been chosen to cover as many types of calculation as possible.290 0.254 0. JFor dielectric constant=3.406 0.153 0. These typical cases are included to be used as a check on the general magnitude of cable constants when making calculations for a specific case.29 0 265 0.787 0.171 0.500 600 000 000 000 000 13 930 1B 040 8 240 8 830 9 660 10 330 11540 12 13 I4 16 230 040 880 320 0.566 0.242 0.263 0.241 0. ‘(iAIR of sectorshaped conductors is an approximate figure close enough for most practical applications.113 0.780 0.219 0.ioo 0. In lowvoltage cables the same insulation thickness is used for both grounded and ungrounded operation.718 0.672 0.327 0.230 0.263 0.729 0. Here again the sequence constants apply to one threephase circuit of three cables with zerosequence return current assumed to be all in the cable sheaths.166 0.74 1.490 0.658 0 639 0 603 0.399  GM) 000 750 000 0:: 0000 250 000 300 350 400 500 000 000 000 000 12 900 15 660 G 360 6 940 i %i 9 10 10 13 690 100 820 220 GO0 000 750 000 1 000 000 00 000 0000 250 000 300 000 350 400 .i 2.091 0.234 0.392 0.566 0.
417 0.63 1.175 0.134 0.323 0.412 0.234 0 230 0 222 0 210 0.92 2.: 130 130 130 130 130 130 130 130 130 140 1. Similarly.55 0. CSCompact Sector.42 4.410 0.151 :: .172 0.00 1.575 0 6.780 0. the calculations of currentcarrying capacity are usually based upon this limitation.320 0 310 0.430 0.922 0 883 0.332 0.258 0.398 0.80 4.763 0.898 0.850 0.263 :: :i: 0.289 0.364 0.06 1.146 0.900 0.s 1.900 0.715 0.604 0.366 0. In Tables 14 through 19 earth temperature is assumed to be uniform at 20 degrees Centigrade.354 0.520 0.570 0.297 0.847 0.%C resistance based on 100% conductivity at 65’C.286 0.366 0. CRCompact Round.488 0.09 1.00 2.137 0.165 0.292 0.193 0 302 0.2fi3 0.106 0 126 0.17 3 .17 1.364 0.08 2.218 0 210 0. These tables were taken from a publication*6 of the Insulated Power Cable Engineers Association and give maximum allowable amperes per conductor for representative cable types.210 0.95.43 2.622 0.jO 250 350 500 750 2 1 0 0:: 0000 000 000 000 000 3 800 4 320 4 4 4 5 6 7 9 12 010 440 970 620 180 530 540 900 0.41 3. For special heatresisting rubber this factor becomes 0.745 0.355 0.21 1.144 0.091 0.70 2.02 0. 00 000 0000 250 000 it% E 750 000 1 000 000 16 450 4 670 5 120 5 300 5 5 6 7 8 10 14 18 360 910 570 160 540 750 650 560 8 _  6 900 7 300 8 200 8 660 9 380 11200 12 790    X.01 1.69 2.191 0.495 0. 3For dielectric constsnt=3.445 0.135 0.185 0.545 0.224 0. per conductor for any given cable.64 2.5.215 0.284 0.52 4. Circuits are frequently installed with each duct containing three cables.292 0 332 0.49 3.175 0.786 0.95 1.281 0 270 0.207 0 24.310 0.82 3.155 0.310 0.171 0.587 0.158 0.263 0.326 0.222 0.141 0. Approximations can also be obtained for the currentcarrying capacities of other types of insulation by applying multipliers to the tables presented for paperinsulated cables.165 0.178 0.521 0.953 0.264 0.188 0.297 ::% 0.794 0.147 0.251 0 238 0.O’ g. However because temperature rise is most often the controlling factor.834 0.495 0.437 0.147 0.201 0.210 0.310 0.05 2.070 0. etc.151 0.82 TABLE 960CYCLE CHARACTERISTICS Electrical Characteristics of Cables OF THREECONDUCTOR (SHIELDED TYPE) Grounded Neutral Service Chapter 4 PAPERINSULATED CABLES GASFILLED   I POSITIVE N EGATIVE & SEQ I ZEROSEQLXNCE I SHEATH I z 3 .233 0.302 0.665 0.62 3.190 0.158 5100 4600 4500 4200 3800 3500 3200 2800 2400 2200 2000 6900 6300 6200 “5%: 4800 4500 4000 3500 2900 2600 8400 7900 7300 6700 6300 5600 4800 4200 3700 0. ~Haued on all return current in sheath: none in ground.392 0.210 0.163 0.707 0 685 0.31 3. efficiency.707 0.249 0. Unfortunately there is no simple correction factor or curve that can be used to translate the figure for cables in ducts .28 1.455 0..539 0.32 1.36 1.321 0. The number of overhead power cables is a small percentage of the number in ducts.191 0. Corrections for earth temperatures other than 20 degrees Centigrade are given within the tables.600 0.622 0.323 0.84 1.810 0.323 0.288 0.13 0. Sometimes regulation. may dictate the maximum permissible amperes. 2OhIH of sectorshaped conductors is an approximate figure close enough for most practical applications.642 0.211 0. including 2% allowance for stranding.86 4.171 0. and other factors not contemplated in making up the tabulated information.930 4.070 0.134 0. The value for varnished cambricinsulated cables can be obtained by multiplying the value given in the tables for paper insulation by 0.767 0.10j11 taking into account variable loading.05 0.816 0 i11 k% 0.409 0.971 0.34 3. 6The following symbols are used to designate conductor types: SRStranded Round.392 0.297 0.642 0.430 0.249 0.32 1. Special conditions may make it advisable to calculate a cable temperature problem in detail.165 0.780 0.131 0. but will be somewhat higher than the capacity of an equivalent shielded threeconductor cable of the same conductor size and voltage rating. “hot spots” along the cable route.:: 0.21 1.56 3.091 0. The limiting factor iq cable applications is not always the maximum permissible insulation temperature. and for this reason space does not permit inclusion of loading tables for cables in air.091 0.58 2.288 0.153 0. 7.786 0.622 EX 0.29 1.234 5100 4600 4500 4200 3800 3500 3200 2800 2400 2200 2000 fin00 6300 6200 5800 5300 4800 4500 4000 3500 3000 “GO0 8400 7900 7300 6700 6300 5600 4800 4200 3700 110 110 110 110 110 110 110 110 110 115 125 110 110 110 110 110 110 110 110 110 125 140 110 110 115 115 115 125 135 150 160 1..353 1 000 000 2 :.987 0.376 0.417 0 455 0 539 0.070 0.198 0.477 0.85 to the figure given for an equivalent paperinsulated cable.462 0.364 0.274 0 39 0.195 0.288 0. carrying capacities for rubber insulation can be determined with the same degree of accuracy by applying a factor of 0.760 0. %x Fir.91.271 0.156 0. the resulting figure being accurate to within five percent of the calculated value.281 0.5 0.249 0 240 0 230 0.30 1.7.730 0. economy.392 0. The current capacity of these circuits will be less than that tabulated here for one cable per duct.242 0.190 0.500 0.987 0.90 3.134 0.58 3.249 0.509 0.
622 0.546 0.575 0.17 3.70 2.126 0 141 0.05a 0. = (i8+xd).512 0.467 0.5390.Sl 2.39 1 24 1 15 0 9i 0 86 0 ifl 0.294 I 58 0 osi 0 776 2 2 2 2 0.681 0.628 0.+l(x.332 0373 0.31c 0. 0.50 000 500 000 750 000 1 000 000 1 500 000 ? 000 000 (110 I270 I510 I 740 I 930 2490 i 180 4380 5560 3000 1190 600 720 930 1040 0: 000 0000 250 350 500 i50 000 000 000 000 1170 I320 I570 I800 I 990 2 550 1340 4 5iO 5640 3090 3300 740 890 1 040 1160 1270 1520 1710 I8iO 2080 2 620 3 410 4650 0 lR4 0 232 0 292 0.412 0.632 0.470 0.417 0.4oc K 0. 5400 / 4920 .60: .987 0.62f I 0.152 1.151 0.159 0.200 0.263 O.484 I 0.786 0.155 0.431 0.542 0.814 0.23 4.622 0.263 0.152 0.316 0.371 0.49t .356 0.512 0.464 0.263 0.408 0.998 1.373 0.373 0.392 0.141 0.091 0 070 0.190 0.159 0 178 0 200 0.01 2.74 2.  r.50 1.495 0.632 0.310 0.445 0.QYi O.786 0.h Currents.524 i::.06i 0 OR4 0.275 0.445 0.319 0.57: I 0.632 0.403 0.392 0. 0 41i 0.464 I 0.61 0.528 0 575 0.221 0.484 0.6% 1.02 0.341 0.070 0.29t 0.573 0.200 0.091 0.232 0.489 0 4i5 0.50 1.10 2.412 1.141 0.05 2.and NegativeSequence Impedances: (a) Neglecting Sheath Currents.44: 0.484 0.385 0.+j(xa+xd).374 0. 0.05( 0.2Of 0.392 0.262 0.4! 1.528 0.400 0.458 0.159 EL! 0.29c .134 0.6: 2.263 O.285 0.221 0.632 0.221 0 262 0.33: 0.412 1.221 0.539 0.292 0.543 0.17f 0.Of3 O.311 0 297 : u 2 06 1 9x 1 51 1.134 0.262 0.392 0. a ( f : ? i B I c _2 50 1.528 0.464 X2 i 0.445 0 543 0.126 0 141 0.395 0.4i 3.374 I 0 35f b .2: 0.326 0.041 2.41: 1.374 0 356 0 412 0 4OR 0 400 0.431 0.336 0.40( 0.191 0.552 0. *Ac Resistance based on 1007” conductivity ‘For dielectric conskmt=3.470 0.41: 103iO 1.050 0.332 0.496 0.987 0.63:  7780 6660 .58 0 987 0.58 0 987 0.084 0.325 0 305 0.06 1.80 0.03 3.041 0.400 0.445 0.4Of 0.412 2830 1.126 0. 190 0 134 0.x.481 0.200 0 221 0 262 0.470 0.89 5.386 0.lQC 0. 1670I 1470I 1210.546 0.350 0.450 445 445 445 445 445 445  I 340 0.12E 0.512 0.49E .184 0.539 0.05c 0.89 1.osc 0.0x 0.106 0.385 0.313 0.418 0 470 0 528 0 575 0.134 0 091 0.681 0. 0.35f .316 0.470 0.31 2.54E 0 633 0.134 0.31: 0.385 0.75 0.ll 1. lifi 0.325 0.35e 0 331 0.262 0.41! 0 54: 0 662  650 650 650 650 / 650 650  1 500 000 2 000 000 3 680 1.2: 3.512 0.262 0.814 0.20 1.442 0.442 0.552 0.681 0.313 0.  0.417 0. 2350 L 2010 . not cmnpact round.633 0 602 i :a: I 0.070 0.91 2.lOF 0.6: 3.412 1.539 0 524 0 512 0.484 0.313 0. I Ez I 0 51: .221 0 262 0 313 0.417 0.2s 2.106 0.681 0.397 0 301 0 3% 0 3x0 0 37 0 h6fj 0.319 / 0.OO 0 901 “360 0 69’ 2070 1 1 1 1 51 48 46 43 9150 9420 761’0 6870 6410 5640 4940 4250 3750 3210 2830 100 105 105 110 0.442 0.450 60 60 60 60 60 60 60 60 60 60 60 60 60 75 75 is 75 2 2 75 75 75 i5 75 75 75 220 215 210 200 195 185 180 175 175 175 175 175 175 175 295 285 275 265 260 250 245 240 240 240 240 240 240 395 385 370 355 350 345 345 345 345 345 345 475 460 .57! 0.65 1..496 0 484 0 464 ii :i2 7 II 400 0.495 0.552 0.681 0.152 1.313 0.232 0.412 0.418 0. 1 Conductors are standnrd concentricstranded.200 0.liR 0. (b) Inclu$ng Shea.602 0 573 0.091 0.453 0.310 0.464 0.373 0.58 0.284 0.332 2 120 2250 2 530 2 740 2930 3 550 4300 5630 6910 9460 1790 2900 3 040 3 190 3380 3 590 4230 5040 6430 0.44: 0.369 0.134 0.9( 1.41C 0.61 3.352 0.50 1 58 0.091 0 07c o.178 0. zI=zZ=rc+ Note: Ei+j(xa+n&i) 12.310 0.546 0.26: 0.406 0. 1055 000 3910 00004080 250 350 500 750 000 000 000 000 4290 4990 5820 7 450 8680 1420 3910 6720 7810 9 420 0940 0.385 0.7( 2 5! 2.2( 1.37 1.52 3.484 0.302 0.464 0 442 0.628 0.25E .20[ 0.470 0.374 I 0.622 0.313 0 296 0 285 000 000 500 000 ! 000 000 lzo 115 110 110 I05 100 100 95 90 90 !I0 90 90 90 90 150 150 140 140 135 130 125 I20 I20 I15 I15 115 I15 II5 II5 0 02 0000 250 350 500 750 000 000 000 000 0 0:: 0000 250 350 500 750 000 000 000 000 i  1 000 000 1 500 000 2 000 000 0 0:: 0000 250 350 500 750 000 000 000 000 0.512 0.412 6320 1. at 65T.425 0.152 1.358 0.392 0.389 0 375 0.495 0.412 1.0: 0.3% 0.436 0.OiO 0.347 0.311 0.418 0.. _ : 0.633 0.681 0 814 0 998 1.050 0.356 0.87 4 47 4.221 0.418 0.305 0.8: 2.786 0. 3890 .232 0.385 0.464 0.Chapter 4 TABLE lo60CYCLE Electrical Characteristics of Cables 53 CHARACTERISTICS SINGLECONDUCTOR OF PAPERINSULATEDCABLES Grounded Ncutrnl  CONCENTRICSTRAND 2r  Service T 6 c 2 : ’  r.i9 3. 0.633 0 067 0.356 1 01 ?‘.ii: _ 1 000 000 1 500 000 2 000 000  2 1 0 02 0000 250 350 500 750 000 000 000 000 l 920 0.15: 1.392 0.19 1 16 1 05 0.998 1.447 0 439 0.374 0.414 0.663 0. 0.339 0 316 0 302 0.31c .622 0.091 0.65 3.403 0.98 2.415 0.385 0.2. including 2 7’0 alloaanrr for strandinr.381 0 383 0.134 0.814 0.134 0 091 0 Oil 0 051 0.  i i : .041 2.106 0.352 0.91 2.264 0.373 0.091 59 32 2I 14 0.9.428 0.07c 0.814 0.8.263 0.) in sheath.998 1.417 0.46: I 0 44: .632 ‘390 1010 1 150 1330 1 450 1590 1760 1960 2 2 3 4 250 730 530 790 0.iY6 0 622 0.05 0.90 0.190 0.633 0.332 0 372 0.81( 0.178 0.267 0.126 0 141 0.7.lOf 0.310 0 263 0 19c 0.445 0.274 / 0. 0 4%I 0.400 0 374 0. .134 0.495 0.496 0. 1.374 0. none in ground) Where x.263 0.15: 8 160 1.356 0.275 0.263 0.3x3 0 380 0 0 0 0 3ii 375 370 366 D.425 0.4Of I 0.041  1.292 0.528 0.329 I 0.441 0.814 0 998 1.+r.352 0.495 0.484 0.87 2.332 0 0:: 0000 250 350 500 750 000 000 000 000 1710 1940 2 100 2300 2500 3 110 3 940 5240 6350 8810 1080 0.40 1. 0.575 0.464 0.3! 1.344 0. 0.575 0.417 0.41i 0.392 0.31 4.575 0.341 0.34 3 23 2.041 3.632  1 000 000 1 500 000 2 000 000 5350 1.93’ 0.190 0.35c 0.5 140 145  1 000 000 1 500 000 2 000 000 350 500 750 I 000 000 000 000 000 1 000 000 1 5oa 000 2 000 000  6000 6250 0480 1’ .632 0.526 0.495 0.991 0.681 0.42E 0.573 0.M 0 43  Positive.292 2010 0.050 0.159 O.154 0.084 O. 4390 .55: .190 0.152 1. ZeroSequence Impedance: (Based on all return current zo=r. xd is obtained from Table .814 0.392 0.998 1.2. 3440 .lQC 0.524 0.342 0.418 0.292 0.417 0.528 0.442 0. ii I : I _560 670 RR0 990 0: 000 0000 250 000 3.524 0.  0.385 0.373 0.632 0.445 0 546 0.27 1.412 1.681 0.232 1 500 0 292 1610 0.524 0.628 0 602 0. 0.420 0.091 O.385 0.152 1.571 0.313 0.400 0.47( 0.54 0 0 0 0 ii 69 61 55 ERQO 105 105 8100 i5iO 6720 5950 5130 4610 3930 35?0 4590 i680 6700 6060 5250 4i10 105 110 115 120 120 130 135 120 125 130 13.622 0.681 0.53 1.786 0.R .106 0. 3.633 0 067 0 OR4 0.041 < 1. 0.302 0. 3020 I 2790 .331 0.341 0.496 0.814 0 998 1.352 0.539 .313 0.26 1 15 0.344 0.40( 0 374 0.310 0.411 0.184 0.361 0.998 0.400 / 0.325 0.539 0 524 0.80 5. 2 16 2 12 2 ‘2 1 1 1 1 1 1 05 02 35 i9 72 51 38 15 90 95 95 b : iooo 5350 I I 1 1  0 361 0.i< !250 !040 !840 1650 1530 1300 1090 885 800 645 555 1810 I020 i300 2990 36iO 2450 2210 3010 1860 1610 (340 LO60 980 805 685 jiO0 5540 4520 1100 3600 3140 2860 2480 2180 1890 1610 1350 1 t i .61 I 1 1 I 1 000 000 1 500 000 2 000 000 7 780 1 152 0420 1.47( 0.041 .12f 0 141 0.221 0 262 0. zl=z2=r.8: 1.06i O.141 0 159 O.52: 0.041 2.420 0.291 0.366 0.liB 0. O.445 0.85 3.313 0.456 0.3i8 0.496 0.633 0.573 0.
69 ISCFIES  $ . because in many circuits more than one cable per phase is installed in order to carry the total current. the extra losses arising from proximity to the plate may affect the currentcarrying capacity.84 Electrical Chnrncteristics of Cables OILFILLED Chapter 4 TABLE ll60CYCLE CHARACTERISTICS SINGLECONDUCTOR OF PAPERINSULATED CABLES Grounded Seutral Service 0.399 0 398 0. E.290 0 284 0 0 0 0 277 2io 260 251 ki a 2 22 02 Id ‘22 c: z:$ 12 CL n 392 0 310 0 263 0 IRR i I!..293 D.635 4450 1.145 9790 1060 3900 6610 7390 7610 8 170 9180 1.533 1 3 3 3 i60 i.i60 0.gs 2: ^ $z cu on00 250 000 .) in sheath.838 I.291 3.399 0.921 5 090 0 956 5 290 0 983 5 950 6700 8080 9440 1 050 I 145 1. The currentcarrying capacities of cables in air have recently been revised by the IPCEA and are now available in the cable manufacturers’ catalogs.039 0.55 for stranding. W.450 0.14( I1 770 l.550 0 612 0.44R 0.918 1.x37 0.vans.2 p z.038 0 0 0 0 310 263 188 132 4860 0.267 0.716 0 373 0 381 0.63! 0.408 0.?Rf 13 110 15 840 18840 __15360 16790 12990 1.05c 0 280 1.44f 0 50: 0.": 0.399 0. Also.266 0. 288 D.44l 0.918 1. x&lr: ‘21 ‘a +j to a reasonable figure for cables in air.37. page 138. 0. 33.3n ti.188 0 132 0.401 0..0% 7310 1. April 1930.089 0 067 0 047 0.805 3.305 I 303 1 298 3.307 0.067 0 047 0.550 0.5.257 0. including 270 allowance Tubular Conductors” by .259 0.639 0. 350 500 750 1000 000 000 000 000 5 IS0 0.261 0.404 0 596 i i.837 0.133 0.038 750 1 000 1 500 2 000 000 000 000 000  Cl 606 0 573 0 490 0.188 0. 2m 1+ 2 0 J?c s2am 0 439 0.3x4 0.758 0.265 3.255 0 246 0.592 0.+x.374 0.50 1.i86 0 285 0 281 0 276 0 0 0 0 269 263 253 245 0 649 0.763 0.274 0 268 0.and NegativeSequence Impedances: (a) Neglecting Sheath Currents.55( 0.238 0 233 0.4lf 1 63: 1.36( 0 341 0.089 0. 'Cf j240 ioio I900 I700 1470 1070 3620 3380 2920 25iO 3850 500 5090 5600 5040 4700 4110 3710 7410 7240 6820 6260 / 5fi80 .1WETEROFSPRINGCORE I 2s I  T INSIDE DIAMETER OF SPRIWG CORE re = 0. 5240.5 1.263 0.048 0.63: 6820 1.625 D 568 3.752 1.2X6 1.52 i29 669 2 0 089 0.. Positive.089 0.i6: 0.219 5: (0 0 837 0. 467C1 417c.551 0.550 0 612 0 692 0.028 1.956 0. G.28E 2010 1.295 0 258 0 276 0 266 0 295 0 294 0.918 1. Note: xd is obtained from Table 12.384 0.263 0 188 0.635 1 X35 0. : 2: 1z.83! 0.~).33 I.384 0. Ch. This reduction in carrying capacity is given by the curves of Fig.5 125 130 135 140 145 125 125 130 130 135 140 145 150 c : & .477 0.469 0 4?1 0 369 i: iii: 750 001 : %i K 2 000 001 250 001 > .374 0. 19 which are taken from the test values presented by Booth..276 3.390 0.55c % 0.279 0 273 0.505 0.372 0.: :h 22 sg ET .76:  =a : >a a. Ewan.310 0 263 0 188 0.660 0.42i z .547 1. and the system conditions being investigated.51( 0.4.793 3.374 0 356 0.270 1.384 0.255 ). .61: 0.61: 0. .068 0.315  1 000 ooc 1 500 ooc 2 000 ooc ‘i ‘Ar Resistance based on 1OO7o conductivity at tX°C.048 0.133 i: 2: 3.501 55C 631 ilf 0 0 0 0 431 427 421 418 I 333 1331 1 32X 1 325 I 320 1 312 I.355 0.068 0.  (HOLLOW CORE) i . (b) Including Sheath Currents: z1=z2=ro+ ZeroSequence Impedance: (Based on all return current z~=r~+r.33‘ “A 0 275 0.46C 0.180 1. I 1 I 99i 850 759 688 1450 1350 1230 IO00 I700 1410 1140 1750 2510 5950 5790 5540 5150 1770 4430 3920 3580 6590 6480 6180 5790 5320 4940 4460 4060 72la 6860 6430 5980 5540 4980 460C 761C 714c 596C  115 I15 115 I20 I20 125 130 135 140 125 125 125 130 135 135 145 150 130 130 130 135 i40 140 145 155 135 135 140 145 145 150 160 160 160 170 10 .248 O.304 0.760 0 410 0 399 0.05C 1.5 INCHES L.+jk+x. IX8 1 I.406 0.983 1. E 5.374 0 356 0 342 1 500 000 2 000 000 1500 000 2 000 ooc oooc 250 OOC 350 ooc 500 ooc 750 OOC : E ::I 2 000 ooc OOO( 250 OOC % ooil 0000 250 000 350 000 500 000 1970 1.i.807 0.460 0 483 0 516 0. CABLES IN PARALLEL The problem of current division among paralleled cables is frequently encountered. 282 0. mutual effects may develop between cable circuits which are adjacent throughout their length but which terminate on separate busses.396 0.039 i:E I.390 0.048 I 039 1310 I 263 1 188 1. 0 ok7 0 047 0.547 1.692 3.310 0.310 1.238 0.133 0.547 1.294 1. VII.048 3.xX.45c 0.009 1 905 I.i36 0 76X 0. 0 361 0.835 ~ 6 590 0 956 6 800 0 983 7 340 1 050 8320 1. Vol.089 0.038 0. 0660 1.342 0 283 0.440  ODOC 6 480 250 000 6 700 330 000 7 460 500 oat X310 750 1 000 I 500 2 000 ooc ooc ooc ooc 9800 11 270 13720 16080 __250 350 500 750 ooc ooc ooc ooc 7 600 3390 9270 0840 L2 340 15090 18 000 0.295 0. none in ground) Where x.369 0.391 798C 752c 69s 632C 588C 519a 471c  130 135 135 140 14.382 0.263 0. the problem may take any of several forms.2 :.61: 0 69: 0.4% 0.e so $ PI” .550 1. j 22 . SC&&ted for circulirr tube as given in S~mmelrica~ Components by Wagner Above values calculated from Set of Curves for Skin Effect in Isolated & l?. Review.$ $ .483 0 433 0.z so gg __ ooc oooc 250 000 350 500 750 1000 000 000 000 ooc $ z z 2 u ‘t $$ g”+ 2.760 0 807 0 837 0.346 0.601 0 545 0 519 0 462 0.1x0 8630 1.039 0.547 l. IV. !!I 750 001 1 000 001 1 500 001 2 000 001 750 001 1 000 001 2 000 00’ P   .658 0.3R2 0.281 1.145 1 601 I.241  0.345 0 356 0 373 0. = (x.132 0. 2 m Em ge 12 'E r: .283 3.46C 0 483 0. 00 ooa 39H0 I090 4 320 4650 O. In the discussion on proximity effect it was mentioned that where cables are installed parallel to steel plates.4IE 4450 1.983 9140 1.334 0 0 0 0 0 0 0 0 39F 39F 391 3% 374 36C 34t 336 0 320 0 3li 0 315 0 310 0.360 0.O : I& 3s: .711 0 418 0 110 0 399 0.730 3.310 1090 13 750 ~___ .985 ) 9i5 I.416 ooi I.302 1. Depending upon the type of circuit.T .180 1 310 1.44E 0 0 0 0 .j: 0.028 1 1X0 1 310 1.2i2 0.611 0.460 0 483 0 516 0.541 0.763 0 450 0.5oi 0.J.374 0.918 6 100 1.038 0.831 8560 0. the cable type and configuration.381 0. LZ 17" :rno.T : s.089 0.057 I.760 0 381 0.500 0 447 0.7 0.33.067 i ::A 0.83! 1. ZFor dielectric constant = 3.690 0.5 720 5 930 6 390 7480 8 950 10350 12 960 15530 1.281 I.+j(x.692 0.37.028 1. : >  0.649 I.356 0.067 0.410 0.585 0.492 0.263 ). 0.418 0 410 0.342 0 421 0 418 0. 0x9 1 068 1. 130 L.76: 0.334 0 398 0 396 0.360 0 346 0.381 0 4OF 0. zl=zz=r.39f 0 39c 0 38: 0.286 1 416 I.3i4 0 356 0 342 0 421 0. jd  INSIDE DI.746 0.532 0.2 jS 2’ Ilo 110 110 115 II5 I?0 I?0 I25 130 140 z 120 12.69: O.41( 1.551 0.51E 0 55c 0.246 8.I!% I 392 ) 310 I.
.061 0. SENDING BUS a bc 1 1 jLENGTH RECEIVING IN MILES 4 I31 P rt is difficult to anticipate in detail the problems met in prnctice.. .5 12. hlmost any problem involving paralleled cables can be rapresented by simultaneous equations of voltage drops caused by self and mutual impedances but such equations often become numerous and cumbersome.0 28. Xd 0.00 8..0 25.75 3.5 22.50 8.25 5.5/64 in..181 6 300 2 270 1... Sheath.. 0.0 11.114 0.0 34.: . Insulation.070 0.75 6.25 3..5 17.119 0.263 0..049 0.631 0.. 500 MC&l..2704 60 logto 5.115 0. 13/64 in. .014 0. .084 0...50 3.0 12. 0. Sheath 0.0 ...035 0.75 7. 6 in ..381 0.25 6.5 24.. .094  0.5 14.. and more complex..068 0...098 0...75 5..0 23. 3.0 14. OHMS PER MILE AT 60 CYCLES In.. TABLE b’ ” Ib’  E..0 0.095 0.0 16.0 29.50 0. Sectored.274 0... ...141 0.960 2.053 0..0 22. 7.0 36.00 zd 55 In.089 0. 27.056 0.5 16. Insulation s/64 in.53 2 10 1..5 20. Sheah..156 0. 16Equivalent circuited sheaths CABLES POSITIVERESIS il z82 ZjE 2. Eb Ec ////////////////////N///////////////////lll//N////// (ZERO CURRENT IN EARTH RETURN PATH) i I f I 1 I Fig.5 11...0 26.5 15.. .090 0.. Three Ciblbles spaced 13 in..112 0...5/64 in.75 10.234 0.25 2. when paralleled cable circuits connect a gcncrnting source to a balanced load.E fi  sin!&Conductor.0 32.005 0. Xd 2..217 0..046 0. h 1360CYCLE CONSTANTS OF TYPICAL r M in. SinRkConductor OilFilled 750 MCM.00 1.016 0..302 ... ...027 .074 0.1  2d In... 7/64 in.0 31.00 2.252 0. cases to be found in practice...081 0.231 0 226 0.... 225 ivIiL In s”lation. Sheath temperature 50°C.00 4.75 4.00 7.302 0.100 1..50 6.010 0.149 0.0 21. 650 mil Inslllntion: 9/64 in. where S is spacing in inches..00 9..5 18. Sheath ‘rhWe Shentha in contact and 4/O Copper Neutrirl Wire. 8... . 500 MCM.084 0..94 0. .123 0.045 0..... 0. . it is usually permissible to assume that the total current in each phase is composed only of the respective positivesequence component: this assumption is based on the unsymmetrical cablecircuit impedances being much smaller than the symmetrical load impedances.160 0.038 0.065 0.103 0.134 IN OHMS PER PHASE PER MILE ZEROSEQUENCE (ALL&yTUHFN IN 1 AND NEGATIVESEQUENCE REACTANCE DESCRIPTION CP%.107 0. Sheath. 500 MCM. .019 0.50 9...042 0.385 0....134 0.0 33.00 6.035 0.. 7/64 in.i5 2.. ITor example. 4/f% in.031 0.. 7. 10.062 0..00 0.134 0.0 35.065 0.0 13.. 6/64 in.135 0... Insulation.. ..082 0. horizontall: sinKlcConductor 250 MCM.5 19.25 Xd In.115 0.. since there are so many different..00 3..169 0..50 5.422 0.. Therefore in :tpproaching most problems it becomes desirable to search :tt)out for one or more simplifying assumptions so that the problem can be reduced to simpler terms. Conductor III~~ll~fwn.079 0.159 0. ’ Thr*Conductor OilFilled Type H.1 I!) 0.134 0.. inside diam.126 0..089 0. horizontally. 6/64 in.053 0.162 0.076 0.87 0.25 4. but the examples outlined here indicate methods .0 17. Sheath ‘hrre Cables spaced 3.50 0..190 0. r’lrepqonductor belted.50 4. .0 18.074 0.0... still without introducing errors large enough to invalidate the solution.25 9.111 0.011 0. Sectored..I[ solution that can be modified to fit actual circumstances. In.028 0...50 Electrical Characteristics of Cables SPACING FACTORS (Q)*.023 0.0 19.50 in.203 0.75 8. ‘l%rre Cshlrs spaced 4 in. ‘l“lrccCunductor Type H.5 21.0 30. .. 1000 AMCM. 30/64 in. Three outlined examples of calculations on paralleled wbles are included here.13. . .049 0.239 0. 0. . with openand no net groundreturn current (see Example 3).059 0..0 0.071 0. circuit for parallel cables.336 0. &It..5O 1..5 13.005 0. but they assist only by illustrating gcncrnl methods.75 9.133 0..75 1. insulation.133 0. :..025 0.. Sheath .Chapter 4 TABLE lZREACTANCE In... ‘Conductor temperature 65°C. Sectored 500 MCM....25 7.032 0.0 Xd In.221 0..25 8...226 2 440 6 300 2 270 4 670 3 400 3 870 ~~inulcCondt~ctor. 9/64 in. 15.126 0.079 0.039 0.133 s J *zd=O..106 5..042 0.. 20..179 0.0 xd In..022 0.057 0..070 0.5 23.130 0.
0.!lO’ 0. / 1 .3 24I 271 2!lli 317 Ill 79 101 113 130 148 167 18!1 ?Oc..07 nt 10c. a Multiply tabulated currents by these fsctors when earth temperature is other than stranded.x0 Volts 81 107 140 161 186 214 243 280 311 340 385 417 80 105 137 156 180 206 2. and all induced ac losses. 0. 0. 0.07 at 10°C :. 0. 0.4 128 143 163 177 19. ".68 a.83 at 4O'C. The cable conductors can be of different sizes.5 260 283 303 :: 78 103 134 153 177 202 230 264 203 326 359 388 74 06 125 142 163 186 211 241 266 296 323 348 113 128 ii: E 102 115 131 148 168 190 208 230 249 267 77 100 130 149 172 196 223 'jj 282 315 345 371 71 92 ll!) 136 155 177 200 229 252 279 305 317 82 103 1’0 136 1. b.92 nt 30°C. 0. 0.83 at 4O"C.02 at 30°C.52 172 1!J4 212 234 25.09 at 1O’C.S 369 308 76 100 131 149 172 196 214 2.  465 433 390 348 Ii17 430 383 480 5X5 541 482 427 :1. 0.73 at 5O'C)3 451 403 348 444 501 383 566 500 427 (I. Example dType of Circuit.39 321 476 443 396 352 540 444 499 393 11.72 at 5OT)3 422 312 267 369 291 469 408 343 529 455 323 381 (1. d.x1 nt ‘WC:. CScompactsector 2 Current rat&s are based on the folloainX conditions: a.55 283 316 346 373 72 94 122 138 156 181 206 235 2.ifi li6 l!)? 211 229 244 76 100 130 148 170 19.5 239 264 289 307 :A no 112 128 14. 0.5 xc.00 at 10°C.66 at 50%) 32s 269 370 358 294 420 471 309 326 (1. SRstandard round concentricstranded.5 2'13 23L 270 310 341 Xi R!) 90 117 133 I.3% "Ii4 2!11 324 357 383 75 98 128 146 168 192 219 249 273 304 334 359 81 108 139 130 184 211 242 276 . One cable per duct.90 at 30°C.X :a: 233 65 85 108 122 1. 7. 0.0% at 3O'C. Ratings include dielectric loss. all cables equnlly loaded and in outside ducts only.72 at 50°CXs 245 288 406 348 4. Ambient earth temperature= 20°C.. 0.70 at 4ooc.36 270 300 336 36'0 399 77 101 132 150 174 198 226 258 287 320 351 378 74 1% 143 165 188 214 243 269 300 328 353 .5 272 175 201 !?09 261 288 321 331 380 it 72 34 ::A 159 181 206 234 258 283 311 334 6.:: 153 175 199 228 251 278 305 327 220 244 266 285 123 139 158 179 196 215 235 251 123 140 Fig 112 !i. 0.5 58 74 !) . n.5 .00 at 3O’C.79 at 40°C. 227 248 "Ii4 s4 69 80 100 11.303 310 375 406 78 102 133 132 17.72 at 5OT)J 4. 1 The folloainr symbols arc used here to desianate conductor types: SSolid copper.. 40°C.53 174 198 217 210 261 2iR 57 73 93 105 120 135 152 172 188 207 224 230 148 166 188 213 235 259 282 303 476 454 429 399 534 508 443 479 607 576 540 497 (I.t:. 0.5 201 2?!1 260 288 32 1 353 380 96 73 124 l11 162 183 ’ 211 240 263 292 320 344 2 115 130 149 170 ID3 218 ?39 264 288 300 79 104 136 156 180 208 287 270 207 3x2 366 395 74 97 127 145 166 190 217 246 271 301 390 35s 115 130 149 170 193 218 230 264 288 309 134 1.217 ‘i19 276 307 339 365 161 183 210 240 266 295 324 349 306 438 494 (k7oi  428 399 3.72 at 5OW' 78 102 132 15 1 175 200 230 266 ?A0 300 3.83 nt 4OT.50 Ei 92 104 117 132 149 163 178 193 206 220 249 27.\XIPERES CONDUCTOR2 6 4 2 1 1% 143 164 18!) 218 2so 286 80 1Oli 13!) 161 184 211 ‘I? 270 305 340 376 406 78 109 134 153 177 "OR 2. 0.50 400 205 330 365 304 449 502 572 77 !I9 120 147 170 I!)4 223 257 284 317 34!) 377 429 470 543 12%5 142 103 187 214 245 271 301 332 357 406 450 510 ii”6 71 92 119 135 155 177 202 232 255 283 310 333 377 417 468 129 146 169 194 222 2.00 at 309C. 60 cycle slternatina current.08 at 10°C.07 nt 10°C. 0.op a t3 0.70 at 4vc.83 at 40% 0. O.: 74 9s 123 140 161 184 211 242 268 297 327 352 11.5 I 100 .39 158 180 !205 22.83 at 40~0.X 3!U 424 487 465 4R!) 408 450 r. 0.O Cables Chapter 4 BELTED PAPERINSULATED CABLES in Duct I Rank NINE I TKELYE C0ndW2tar Size A\VC IOOo”rCM ___~ ONE Condllct”T ty.00 at 1ooc.5 108 122 138 l.08 at 1O’C.83 :a 40°C. 0. I THREE / 7.5 267 2X6 166 189 .9 at 40°C 0 .07 at 1OT. c.5 3OT OYX 1 138 156 178 202 221 24. and having two paralleled cables per phase. 0.86 TABLE 14CURRENT Electrical Characteristics of Cables CARRYING CAPACITY OF THREECONDUCTOR Sumher of Equally I Per Cent 30 / SO 1 75 / 100 1 30 i 50 I 7. Circuit: Refer to Figure 16.08 at 10°C. 0 73 at 5OT)1 83°C 7.83 at 41l'C.92 at 30°C.e.?‘i 517 487 618 581 550 SO5 (I. 0.51 315 267 384 29." Copper Temperntore 64 . 0. and their spacings can be entirely unsymmetrical.5 84 108 122 l10 158 379 202 221 24.54 423 381 341 507 471 422 376 573 532 473 41.73 at 5.07 at 10°C. !I8 127 14. 0.0X at IfPC. 0.507 428 350 (1.!U al: 3'pgja 0. "!I6 417 :x7 251 462 393 .w at 30c.5 248 271 290 57 ii: 107 120 135 153 173 189 208 227 241 . 0. O.5 165 187 i?J 88 99 112 127 143 158 177 194 211 224 72 03 120 136 1.67 at 5"T13 319 414 396 280 459 409 351 306 520 458 391 341 (I. 0.83 at 40%.5 131 130 170 195 222 245 271 297 319 ii 106 120 Ei 126 144 . 433 378 320 273 480 416 3ao 2!)8 541 466 390 331 (1.w at 30°C.!12 *t 3O'C.53 281 316 348 375 .x 300 350 400 31fi 3.!l2 at 30°C.59 288 315 338 113 128 146 166 188 214 23.: A threephase GOcycle cable circuit connected between a sending and a receiving bus.? 173 I !I7 21.3 (1. 0. 0I. and its magnitude is known. 0. 0.73 at 5.72 at 50°C)~  440 392 298 340 491 436 375 327 555 418 363 489 (1.5 167 191 217 247 273 304 333 360 69 89 114 130 14!1 169 192 218 240 263 289 30!) 60 77 99 112 128 145 163 184 200 223 242 257 53 :t 98 111 125 141 159 174 190 206 220 79 104 136 156 180 206 236 270 300 33.04 at 10~~. 20°C. o. 0.73 a.323 273 5 1!I 439 330 302 1. .66 at 5OT)" 346 293 250 319 380 273 425 302 356 1ooc. Conditions: The current flowing into the sending bus and out of the receiving bus is nearly balanced threephase load current (positivesequence only). Problem: To find the division of load current among all conductors. ‘28.PC.5 I 100 1 30 I 50 PER Loaded SIX Load / 73 Factor I 100 I 30 1 50 1 73 I 100 I 30 1 r. 211 224 2. 128 146 166 189 21. using singleconductor unsheathed cables.
CScompactsector Cllrrcnt rntinse are hnsed on the following conditions: a.10 at 1OT.*.61 at 5O”CJJ The followinp symbols are used here to designate conductor types: Ssolid copper. C. all cables equally loaded and in outside ducts only.. O.54 2Yl 321 354 384 412 182 208 238 273 149 170 193 219 239 263 283 302 145 1 164 j 186 212 242 278 308 341 369 396 132 149 169 193 220 250 275 302 327 348 117 132 147 168 191 215 236 259 280 298  269 318 346 2!13 383 323 .oad Factor Cables 87 PAPERINSULATED CABLES TWELVE SHIELDED in Duct I Bank NINE OSE I THREE I I I YI 119 153 172 196 225 258 295 325 364 397 429 87 114 144 163 185 212 242 276 305 339 369 396 446 491 551 10x 4OT 81 104 13Y 149 It39 193 220 250 276 305 330 3s4 399 437 485 .i 442 487 548 (1.oI.‘e 134 124 173 lY6 222 2.79 at 40012.48 168 lS9 31.S 117 132 149 169 1YO 207 227 243 260 119 133 14Y 172 195 221 242 266 285 303 100 ( 112 i 126 i 141 162 182 19’) 21i 233 24i i 88.5 ’ 230 441 367 PYY 248 490 408 329 276 (1. 0. 0. SRstandard round concentricstranded..10 at 10°C..’ 0. Also if one ampere positive Sequence current is assumed to flow through the overall circuit.!.75 *t 4OT.10 *t 1ooc 0.58 *t 50%. 0.X? ’ li0 187 203 .:: :i AMPERES PER CONDUCTOR’ 75 ‘95 121 136 154 175 198 223 246 272 293 314 350 385 426 yzgjd 89 116 149 168 190 218 249 285 315 351 383 413 467 513 576 ‘($g 83 108 136 153 173 1’?8 225 2.. lillll 751.=aI.76 at 4O’T : 0.*=.Chapter 4 TABLE l&CURRENT I I Electrical Characteristics of Cables CARRYING CAPACITY OF THREECONDUCTOR Number of Equally I Per Cent Loaded SIX J.89 nt 3ooc..X 177 202 227 2.6s 3t 70% 164 185 209 238 262 288 311 334 142 161 182 205 222 244 263 28” 119’ 134 1 1.57 283 313 340 366 it 120 136 154 174 19s 224 245 271 293 313 350 384 423 66 85 107 121 137 IS6 174 196 215 236 2s 273 303 330 365 a7 113 144 162 183 211 241 275 303 337 366 3’34 444 488 545 (1. then I. lh. 0.3 2OT.09 at IODC.08 at O. 0. 302 257 329 279 364 306 0 90 at 3o*c.3 I:<!1 l.55 279 309 335 359 :o” 156 177 200 227 262 301 334 373 405 434 492 x3 6 16 ‘b% 1O..k lli 131 I.. therefore these voltages can be climinated by subtraction.66 *t SO”C3 Volts lY3 21Y 2.=a2lhS.Ol nt 3 ..50 I 170 192 220 25 1 289 3 13 349 370 4OY 143 I 162 134 152 172 IY7 223 251 277 306 331 356 192 221 2. Ratings include dielectric lous..\lllltlply tabulated currents by these factors when earth temperature is other than 393 337 281 238 430 367 304 259 481 409 337 284 (1.89 at 3OT. ‘ soocia Copper 149 170 193 219 240 264 28. 0.56 at SOT. at 4OT. and IcS. .. l!JY 211 465 ‘401 516 440 583 495 *t 10°C.33 258 283 305 323 *t $:801: *t tu1 77°C . Applying these assumptions leads to a set of three simultaneous equations relating three currents: Modified equations: Simplifying assumptions: It is apparent that E..=E.91 *t .79 at 4OT.67 *t 5OOCJJ 424 363 464 396 520 439 (1. = E.. 0..71 at 5 g:c”iz Tempera 107 12 1 136 156 175 lY7 216 232 255 273 84 105 118 134 150 16Y 184 202 217 232 Copper 140 159 178 202 230 264 290 320 347 374 125 140 158 181 206 2.z Eb”. and all induced 3c losses..79 *t 40% 0.88 *t 3OT.t 171 lY4 220 251 276 304 32Y 352  400 336 27..5 303 Temperature 129 145 165 186 202 221 238 254 111 126 141 160 174 l!)O 204 216 1’28 144 157 171 184 l!J. 0.I. 7. 21):s 13 “OC.67 at 5oT): 288 443 391 333 489 313 428 365 550 347 479 402 (1.il 83 1o. and E.X 178 203 232 26.97 lo!1 12. One cable per duct.3 stranded. 72 !)O 100 112 12i 14. 0.74 at 40°C.50 288 316 352 384 413 185 209 238 275 302 335 E. E.=l.5 2Yl 322 350 376 419 465 51. 0.09 at 1wc.i Yfi 123 138 1.00 at 300<’ 0.R’L *t 78 102 12!l 145 164 167 212 241 265 2!)3 818 340 381 416 464 IVC 40% 141 162 182 205 221 246 267 283 60 77 !I7 1OY 122 139 157 176 193 211 227 242 84 IOY 13!) 1. 372 313 236 2 13 406 340 27i 232 452 377 304 I 253 (1...g 410 450 502 10c 4OQC . : S’ UT)’ :. CS 483 534 602 b’..50 Zifi 301 320 368 3YO 432 lOT 4OT . 1). 60 cycle alternating current.ii 170 181.71 :o” at 30°C 5ov2i~ 23 000 Volts T 1. . 0. 51. Anibxnt earth temperature=20°C.” at 3OT.90 at 3”OC. ::. 0.1 162 177 1!)4 208 222 24i Xi!.00 *t 10°C 0.
and Z.88. Abovevaluesapply specifically to sector ductors multiply by 0..79 and prorconcables 757” losd factor assumed. 10 20 30 40 7...urcDeg. and to determine how zerosequence current divides between cables..70 210 240 26..*. Ratings include dielectric loss and extra imity loss. (B. because the effects of a EARTH EOUIVALENT EXPRESSED CIRCUIT. based on the respective conductor characteristics. Apparent conductor impedances: Using the currentdistribution factors for each conductor to solve the complete voltage drop equations. This assumption simplifies the determination of the various self and mutual impedances. the sequence networks are not interdependent and an impedance value of each sequence may be calculated and used independently. i7Actual and equivalent zerosequence circuit for two parallel threeconductor leadsheathed cables (see Example 4).. using the appropriate spacing for each. Therefore.ZI.. GMRl of conductor a’..2i94 log.08 1..00 0.88 Electrical Characteristics of Cables TABLE~GCURRENT CARRYINGCAPACITYFORTHREEPAPER INSULATEDCABLES OILFILLED (amperes per conductor)* Rated Line VoltageGrounded . or A.08 1. inches Z./1 (0) ACTUAL CIRCUIT EARTH Apparent impedance of phase b Apparent impedance of phase c I I (b) ///N////////////////////////////////7/// m Supplementary equations: The original assumption of positivesequence current flow through the circuit precludes the existence of any net ground return current.2794 log.5 295 320 342 380 412 440 455 ground return path may be ignored with very small error: 2%~ Z(rc+js.. 75 0 00 000 0000 250 000 300 000 350 000 400 000 500 000 600 000 700 000 750 000 Deg. WITH ALL QUANTITIES TERMS IN ZEROSEOUENCE Fig. C.0 ampere positivesequence current in the circuit. . Conditions: Each cable contains three conductors that. G. abc LENGTH IN MILES Factor for Various Earth Temps.5 ______ i!iO 210 240 265 295 320 342 382 417 415 460 168 190 210 240 265 295 320 342 382 417 445 460 Corrrction 1.) Copper Tcmpcritt.. = b(. ohms per mile. Problem: To find the zerosequence impedance of the entire cable circuit.99. The remaining mutual impedance are calculated similarly.08 1..5 Chapter 4 CONDUCTOR Circular 51ils.. z.or negativesequence currents will cause no zerosequence voltage drops.=&C..G.8 Example IType of circuit: A threephase GOcycle cable circuit connected between a sending and a receiving bus.) = where I = circuit length in miles. based on a total of 1.90 . . losses such as sheath shaped a duct 0. I. an (‘apparent” impedance for each phase of the circuit can be calculated../////1////.W.__ 34 500 1 46 000 / JIasirnurn Xcutral 69 000 C. multiply above values by ac.1.z. 12 =jo. conductors. are symmetrically transposed so that the flow of positive. 12 Z. are determined similarly.90 . 1.. 7. using two dissimilar threeconductor lead sheathed cables in parallel.00 0... resistance of conductor a’.. This apparent impedance is valid only for the particular current division calculated : Apparent impedance of phase a abc ING I/. For round for six loaded After substituting the proper self and mutual impedance values as defined later.79 1. A series of more complex examples of the above type of problem is described by Wagner and Muller. To obtain actual currents.r) where S. = reactance of conductor a’. is the axial spacing in inches between conductors a’ and a”.*= Z. bank.90 . the distribution factors must be multiplied by the actual load current in amperes.& S./.*. ohms per mile. by the nature of the cable construction. these equations can be solved by the method of determinants for current distribution.00 0. *Applies to three similar loaded cnbles in in B duct bank. Zb.Z.jO. to a twelve inch radius.
~~.!Jl at 3O’C.10 at 1ooc. 0.&. 310 200 375 346 450 418 x2 514 0.75 at 4ooc..Z.67 at 5(.z~+I.74 at 40°C.y~ .91 at 30% 4ov.~~ z3sfs. Then Eo= Io.. Rntinw include dielectric loss.) ]= + (Z.*) + (z.*. To find the zerosequence impedance of the entire circuit requires that one of the complete voltage drop equations be solved for Ea.. = 1.~.66 &It5 OWJ Temperature 135 152 173 195 116 132 148 168 34 300 Volts 203 230 262 302 194 220 250 289 166 187 212 240 148 168 191 215 187 212 240 276 169 191 216 246 147 166 1811 213 130 147 166 188 180 203 231 264 Copper 156 178 203 230 L I (1. all cables equally loaded and in outside ducts only. Ambient earth temperature20°C...~&~ ‘1’1~voltage drop E.:g 70°C 172 194 220 250 149 169 191 215 125 141 160 179 108 121 134 131 149 172 195 221 242 28.. Zo~fZo.80 at 4OT.Z. CScompactsector stranded.+I~~.90 at 31 OT .~~ This equation furnishes a solution for Z.58 305 363 439 1oYz 4OT 107 121 136 ::: 197 277 331 2% 401 495 Xl .Z.z2s’s*.z. which makes I”.*.z. 0.*) (z.=Eo =~=Eo..) ...Z<<. 60 cycle alternating current. from which ZCr follows directly.Z.Chapter 4 TABLE Electrical Characteristics of Cables BJRRENT CARRYING CAPACITY OF THREECONDUCTOR Number of Equally SIX Per Cent 30 I 50 / 73 / 100 1 30 1 50 1 75 1 100 1 30 / 50 Load Factor 1 100 PER I 30 1 50 / 75 1 100 I 30 1 50 / 75 j Loaded 89 PAPERINSULATED CABLES 171 GASFILLED Cables I NINE ONE I THREE I I TWEI ’ VE 1 75 100 APPROXIMATE AMPERFS CONDUCTOR= Copper 144 162 129 145 164 187 212 241 Tempernture 112 125 141 162 182 205 97 109 122 139 157 176 81°C 203 234 270 308 19.) + (Z.%c..zo.83 at 3O’C I 0. c...89 at 30% 0. ..*z. with the effect of earth as a return path included.~Z.~Z. + Z..yZ... 0..61 at 5O'C)3 I 266 235 316 280 376 333 457 402 (1.f.. .10 it 10% 0. ..91 at 30% 0.76 at 4OT.) (Z.76 at 40% I 206 233 240 277 328 285 340 397 0.~~. the following single equation results: Z.. (‘omplete voltage drop equations: E.y) (z.10 at loOc.58 at 506~ 27. 0 40°C.. .08 at 10°C.5 233 196 16..90 at 30%.~~Z.Z..f.70 at 51lW~ per Temperature 77°C 119 133 118 134 1:o I69 173 196 222 255 270 335 400 490 (..10 at 10°C.67 at 5O"C)3 308 275 369 327 443 391 550 479 (1. + Modified equation : Zo*[(Z.~~Z.) + + (Z& ...82 at 4O'C. .10 at 10% 0.90 a . O.. + zo.08 at loOc...*.... can be eliminated by subtraction and 11~ sheath currents can be solved in terms of cond&tor “urrents by using the last two equations only.0X *t IOT.~~z..82 8t 40c... d. z..RS nt 3oor.0 ohms.x2 at 40% 0..Z...Z..~Z.5 224 256 395 182 215 245 281 176 202 230 2fil 196 225 258 295 276 246 325 305 330 293 397 369 399 350 483 446 483 426 6O2 551 (1.~~ Ps..76 nt 40% 0.60 at 5OT) 290 252 212 183 346 300 214 250 413 354 295 250 430 505 354 298 (1....tt 5ww 303 265 224 103 366 31'8 267 227 444 381 318 269 545 464 383 323 (I..~~..~~ Z..) (Z.) (Z. b.~~ Z‘~~.. 0.*Z. 0. I.) (Z..) (Z...5 336 408 1O'C. Supplementary equations: The equations necessary to determine each impedance value are shown here: every impedance must be expressed in zerosequence terms.. .. 0.~~Z... 0 126 144 162 18’2 156 177 200 227 262 301 334 405 492 616 150 170 192 220 251 289 315 :z 583 143 162 183 210 238 271 298 134 152 172 197 223 251 149 170 192 221 254 291 141 160 182 208 238 273 I 130 145 166 189 216 246 117 133 149 170 193 219 145 164 186 212 242 278 132 149 169 193 220 250 117 132 147 168 191 215 236 280 333 402 '... and all impedances are to be exI~~WXY~ zerosequence terms considering the earth as a in rcllrrn path for each circuit. 0..~~.~~) + (Z.TiS 321 299 384 356 427 461 528 577 (I.71 at 5O'C)S 215 283 255 340 303 410 365 502 10°C...89 *t 30% 0..y .. 0.. 0... 1.~) (Z..*z.i.$I 125 140 158 181 206 233 2...y + I..~.3 327 276 228 193 390 329 269 226 475 396 320 268 (1.Zs.*z. Also.~.Z.Z.o . O.. it is convenient to assume that the total zerosequence current ‘lo\ving into the sending bus is one ampere.Z.71 at 5OT)S 183 211 241 275 327 341 397 417 487 513 606 643 (1. I..09 at 10% 0...67 a. The three actual conductors in each cable have been reduced to one equivalent concl~~c:t.91 at 3OT...~* ..~Z.) (z.. 270 239 318 283 379 335 465 407 0... 0. ‘lfter making these changes.. Circuit: Refer to Fig. ’ ~lultiply tahnlated currents by these factors when earth temperature is other than 20°C.~)(Z..09 at 1ooc 0.71 .+ IgZ<r 0 = Z”.60 at 50'03 304 271 363 323 439 386 540 469 (1.~Z.Z. + 0 = zo~z.Z..89 at 3O’C.~~+ Zo*~Zc~~.79 at 4OT $gjs :5” 216 184 217 255 257 30% 306 364 'LOOB t3 O°C..fZ.Z.~~ + I.56 at SOoC)Z ’ ‘l’hc following symbols are used here to designate conductor types: SRstandard round concentricstranded. 0. 17.) (A. 0. = Z”..*+ I. .. 0.yz.~. and all induced ac losses. .&*+ z....~* . 105 117 132 149 169 190 207 243 288 347 140 159 178 Ez 264 290 347 424 520 (.~yz. 0.. 0. One cable per duct... ’ Current ratings are based on the following conditions: u.orin this figure...Z..) + (Z.
st 4OT.07 at 1O’C.73 at 50°C)' Temperature. 0.08 110 :z 218 248 287 330 384 423 470 .50 800 1000 1250 IROil 17.500 Volts CONDUCTOR’ 0 Ii 1 0 Oil: 0000 250 300 3.5 438 401 537 . 0.80 180 206 239 275 315 360 396 438 481 5iil 597 663 725 754 782 889 1014 1118 1206 1293 169 193 222 263 290 332 365 404 442 478 546 603 :x: 706 797 904 993 1067 1137' 154 176 197 225 250 297 326 360 393 423 480 520 574 f% 692 781 855 911 967 Copper 140 159 182 205 233 265 290 319 347 373 423 466 503 527 %i 676 E 831 0.y MY 710 772 7Y7 825 939 1067 1176 1282 1368 low c WC ‘0. 1108 lli6 12”4 1300 1352 14’0 134ti I 1442 (1.08 81°C 5 15 000 Volts  : 0: 000 0000 250 300 350 400 600 2 1 113 14Y 195 226 256 2'37 34.07 at 1ooc.OE I &I“4.83 at 40°C. 157 17Y 203 232 267 305 334 369 401 432 488 537 581 602 622 697 784 856 919 870 0.!. fI.82 112 147 193 222 252 295 341 392 432 481 527 572 655 727 790 851 830 968 107 140 183 211 239 278 320 367 404 449 491 530 fjO5 668 726 7.6 at SOY 957 ?% 1053 1256 1130 1352 1213 (LO! 1 at 1O’C. 0.!U at 3ooc.58 181 208 238 270 307 336 371 403 434 490 542 587 609 630 705 184 208 236 269 2Y4 325 352 378 427 470 508 526 544 606 1 IF 134 202 234 270 311 356 412 4..80 Temperature.548 627 605 766 797 826 946 1080 1192 1296 13YO (l.O 23 000 Volts 2 1 0 0% 0000 280 $2 400 500 600 700 750 800 1000 1250 I.5 1190 986 839 (1. 108 142 18fi 214 242 283 327 374 9 2.O $J! 72i 761 0.82 1102 1000 1220 1105 1330 1198 1422 1274 (l. 105 138 180 208 231 271 31' 361 396 439 481 .5 759 828 886 935 2 *t 30 at 50~ iii 137 157 177 202 230 263 : / fiO0 700 750 800 1Oiml 1230 1500 1750 2000 452 488 505 522 581 650 707 755 795 0.57 827 860 892 1012 110 144 189 218 251 290 333 380 418 4B4 507 548 626 696 758 78Y 817 922 1039 1146 1240 1343 ~(y.i22 597 663 721 750 776 874 219 250 285 325 356 394 429 461 524 579 629 651 674 756 90 116 I. 0.08 at 10Z.X 2000 ."E 391 41Y 474 521 564 584 6C4 67.73 *t 5owr Copper 1068 841 673 ..77 4Y3 560 621 674 700 725 816 914 1000 1078 1162 212 241 275 314 344 380 403 445 504 557 604 627 2: 809 E.53 780 882 100 131 170 195 220 253 293 335 367 406 443 478 542 598 647 672 695 782 883 972 1042 1105 2 at 30' at 5ov 1:.O! 181 207 239 273 314 3fj2 396 444 488 525 600 663 729 :x: 898 172 197 227 25s 296 340 373 416 466 491 559 616 675 702 726 827 162 186 213 242 277 317 346 386 422 454 514 566 620 643 665 752 848 925 994 1058 0.82 110 144 189 218 247 287 333 383 422 470 514 556 636 705 766 795 823 933 104 136 177 204 230 265 306 352 387 429 468 506 577 637 691 716 741 832 Y6 125 lfil 185 209 239 274 315 345 382 416 447 507 557 604 625 646 724 816 892 Y58 1013 2 at 30' at 50z 87 114 146 167 188 214 245 280 306 338 367 395 445 488 528 54i 565 631 706 772 824 869 0. 0 at 301 4OT.: 1019 113 149 196 226 261 303 348 3YS 437 486 532 576 659 733 802 835 865 '380 107 140 183 210 242 278 319 364 400 442 483 . 0.80 Continue . f at 4O'C.x 512 561 fiO7 fXl% 772 84G 881 !I14 1037 113 149 I!)6 226 262 300 344 3Y. O.52" 592 656 712 736 762 86I 97d 1072 1162 1233 HF5% 100 131 170 195 220 254 "Yo 335 367 408 444 480 s13 fiO1 652 674 6Y6 785 864 Y66 IO14 1106 0.~~O 103 134 175 201 231 270 % 370 420 4.4 399 440 ‘l!lO 53!) 586 6ti9 746 810 840 869 WI 1130 12.50 1368 1464 (1.580 fX0 735 804 837 86fi 980 000 700 7.80 184 211 244 278 320 367 405 :z 536 615 684 744 779 808 921 175 200 230 263 302 345 380 422 461 408 570 632 680 717 743 842 162 185 213 243 2: 346 382 418 451 514 568 617 641 663 747 845 926 991 1053 0 at 30( at 5001 150 171 196 221 252 288 316 349 380 409 464 511 554 574 595 667 751 818 875 928 0.5 1 172 196 224 255 290 316 349 379 407 459 506 5r8 568 588 657 111 147 I!)2 222 256 2Y5 340 3!)0 427 474 518 560 641 714 779 810 840 950 104 13fi 178 204 234 268 308 352 386 428 466 502 571 632 688 714 740 832 '34 122 1..59 500 R10 611 fi79 741 771 7Y7 898 1012 1110 1204 1300 2 nt 30' at 5ov 103 135 175 201 232 262 300 345 370 423 460 496 501 621 677 702 725 816 914 1000 1080 1162 0. 178 203 234 267 307 351 386 428 468 507 580 645 703 732 759 860 980 1081 1162 1240 77°C 164 187 2lF 245 % 351 389 424 458 521 577 627 650 674 759 858 940 1007 1073 147 167 192 217 247 281 307 340 369 398 450 496 538 558 576 646 725 7Yl 843 893 D at 30' at 50°f 132 150 1020 035 1122 1025 1215 1106 1302 1180 1O’C. 1178 1032 731 1280 1103 919 783 138.500 1750 2000 186 214 247 283 326 376 412 463 508 .co: at 1063 941 1175 1037 1124 1278 1192 1360 (1.07 at IO’C ( at iOT. 794 1342 1166 992 851 1260 1442 1068 Y14 (1.542 618 685 744 772 i% 1026 1133 1230 1308 (1.82 101 132 172 197 223 257 296 340 372 413 450 485 551 608 659 684 707 7Y4 1:: 154 175 198 226 260 2Y8 .0 2 at 3ooc.83 115 132 109 230 266 30'3 356 408 4‘ln 499 546 393 679 7.90 TABLE I THREE Electrical Characteristics of Cables WCURRENT CARRYING CAPACITY OF Number SIX Per Cent 30 ) 50 j 75 / 100 1 30 / 50 ( 75 1 100 PER Chapter 4 SOLID PAPERINSULATED Rank I TWELVE SINGLECONDUCTOR of Equally Loaded I Load 1 Factor 30 1 30 / Cables in Duct NINE CABLES 75 / 100 1 30 ( 50 j is ( loo AMPERES 7. 501 . h 1104 981 730 1220 1078 22.50 400 500 109 142 1% 214 245 283 32L 371 409 4.83 at40%0 .
64 *t 4ooc. 0 197 224 256 201 321 356 388 417 474 526 574 595 617 698 782 856 919 975 1064 1 *t 300 at 5OOC 182 205 235 267 294 324 353 379 429 475 517 535 555 624 696 760 814 860 936 0. 0.56 1151 996 830 708 (1.32 580 602 624 706 790 865 92’3 985 1075 177 200 230 259 286 315 343 3G8 416 4. 698 610 535 763 664 580 818 711 618 868 750 651 942 811 700 1O’C.5 788 833 905 i at 300 at 50’< 204 232 252 280 304 326 365 406 441 457 472 528 589 638 682 718 778 0.~t~~ . 0.10 *t at 213 242 278 317 352 390 426 460 524 584 639 664 690 783 882 972 1048 1115 1225 1OT. 0.5 282 321 356 3!15 432 466 532 593 649 675 % 898 988 1066 1135 1248 I!)!) 226 259 2Y3 324 359 392 421 479 5.I.13 348 375 425 471 514 533 554 622 312 335 379 419 45.70 at 40% 0.5.32 at 5O”C)Z . 0. 0 187 211 242 274 303 334 364 390 442 491 535 554 574 646 722 788 845 893 973 9 at 300 I at 509c 169 1YO 217 246 270 298 324 347 302 433 470 486 503 564 628 682 730 770 834 0. 0. .76 215 24.84 at 30% 0. / 4ooc.87 at 30°C.84 at 30°( :.5 z: 547 279 300 337 371 403 417 430 481 365 394 447 497 542 563 582 660 332 358 405 448 489 506 523 589 293 31.65 4O’C.42 at 5O’C)~ 742 814 876 931 1013 I at 1O’C. 0. 40%.X 382 413 470 524 573 597 617 702  Copper Temperature.78 46 000 Volts Copper Tempernture.14 at IO’C.51 at 5O’C)2 69 000 Volts 395 428 489 54 5 599 623 644 736 832 018 984 1066 1163 (1. 60°C 441 490 536 556 575 652 495 514 531 599 % 387 418 477 532 582 605 626 713 446 496 .$& OF MCI1 THREE I SIX loo AMPERES CONDUCTOR’ 34 500 Volts Copper Temperature.5 491 562 629 690 718 747 852 967 1068 1156 1234 1367 (1. 0.66 %2 461 513 561 583 603 683 772 848 913 972 1060 (1.36 at 5O’C )* 741 489 812 529 873 563 927 592 1007 635 (1. 70°C 227 260 239 341 380 4”2 464 502 5i5 644 710 736 765 875 994 1098 1192 1275 1418 221 251 290 330 367 408 446 484 551 616 673 702 730 832 941 1036 1123 1197 1324 209 239 273 312 345 382 410 451 514 573 626 651 676 766 864 949 1023 1088 1186 197 224 256 201 322 355 389 419 476 528 577 598 620 701 786 859 925 981 1072 225 255 295 336 374 416 4.72 262 302 329 367 403 433 4Y2 553 605 629 652 740 241 276 301 335 366 394 444 497 542 562 .72 268 309 336 377 413 446 506 570 626 650 674 766 865 951 1028 1094 1205 249 287 313 349 :7: 464 520 569 590 612 691 777 850 915 970 1062 226 259 282 313 341 367 412 460 502 520 538 604 675 73.::: 543 562 582 660 333 358 406 450 490 508 525 592 664 724 776 822 892 5 at 30’ at 5ooc 305 328 370 409 444 460 475 533 595 647 692 732 791 0.Chapter 4 TABLE 18CURRENT Electrical Characteristics of Cables CARRYING CAPACITY OF SINGLECONDUCTOR (Continued) Number of Equally Loaded I Per Cent 30 / 50 I 75 1 100 1 30 ( 50 1 75 1 loo 1 PER Load 30 Cables NINE Factor / 50 ( 75 / 100 I 30 / 50 1 75 / in Duct Bank I TWELVE 91 SOLID PAPERINSULATED CABLES C’IINd.5 3. ( *t ‘IOT.59 500 518 iii 670 730 780 824 893 9 at 30’ at 509( 158 170 204 230 253 278 302 323 364 401 435 450 465 520 577 626 668 704 760 0. 0.85 nt 30% 0.1” at 102.59 1016 1115 240 276 E 365 393 444 496 542 561 583 657 738 805 867 918 1002 274 317 346 385 425 459 522 589 645 672 698 794 992 900 1074 1144 1265 259 299 326 364 398 % 546 598 622 645 731 %i 976 1035 1138 239 274 299 332 364 391 441 494 538 559 578 653 732 799 % 994 i at 300 at 5ooc 221 251 274 304 331 356 400 447 486 504 522 585 654 703 762 805 875 0. 0. 65°C 279 322 352 394 433 469 534 602 663 689 717 816 927 IO20 1110 1184 1314 270 312 340 380 417 451 512 577 633 658 683 776 879 968 1047 1115 1232 256 294 321 358 392 423 482 538 589 611 638 718 810 887 0.54 391 425 439 453 508 259 278 312 343 372 iii 442 792 734 672 872 804 733 942 865 788 924 1008 840 1096 1001 903 i nt IOT.X 205 234 268 304 337 374 408 440 500 556 608 631 654 741 833 914 984 1045 1144 10°C.67 *t 4ovz.76 220 249 288 328 364 405 443 478 547 610 669 696 723 823 930 1025 1109 1182 1305 (1.582 657 214 244 266 295 321 344 386 430 468 485 501 562 191 217 236 260 283 307 339 377 408 422 436 487 834 736 626 541 914 802 679 585 987 862 726 623 1048 913 766 6.
C.92 Electrical Characteristics of Cables %.00 0..0 8. or A..90 .G. 18. TABLE 19%CumENT CARRYING CONDUCTOR OILFILLED CAPACITY FOR SINGLEPAPERINSULATED CABLES by Halperin and Shanklin.oo 0. GMRs. = Z.08 1. where the terms are defined as for Eq. where the terms are defined as for Eq.& S. Fig.09 1.. Z. covering those cases where the cables are not necessarily bussed together.+xe)] ohms. 3). where I= circuit length inmiles.)(r r.+. 1 132 174 228 i 270 ’ 378 575 672 j 780 1110 1 1 xulation ‘hicknesz mils i8 78 04 141 203 266 375 46'3 688 lsulation hicknrss mils A more general version of the above type of problem. . and ar’ ranged in a duct bank as illustrated in Fig.90 1.= 0. is described by Cheek.)(NO..= l[r.+j(3z.. Z.77 .7 15 23 34.5 46 69 115 138 161 230 Basic Impulse Insulation Level for Equipment 30 45 60 75 110 150 200 250 350 550 650 750 1050 I SolidPaper Insulation ‘ithstand Voltage kv 94 04 113 16D 244 310 450 563 825 r: 1 OilFilled Paper Insulation ?thstand Voltage kv 1 j .79 ..2794 log. or with! return current only in the sheaths. Z.3xJ] ohms.91. is defined similarly.79 Factor for Various Earth Temps. . GMR of three conductors.00 0.= Z[:3r. (23). with sheaths and ground in parallel.. where Xd= 0...” Example SThe use of complex GMR’s and GMD’s will very often reduce a complicated problem to workable terms. = Z. .3 in inches.+x.79 0.t = Z[r.. 10 20 30 40 256 287 320 378 405 450 492 528 592 655 712 742 767 872 990 1 082 1 165 1 240 .+j(x.338 inches (from example 1).no sheath loss considered. *Applies to three similar loaded cables in a duct bank.oo 0... (19). The use and significance2 of these factors should be studied thoroughly before attempting a solution by this method (see Chap.77 75% load factor assumed.~(x.2 2. Correction 1.. p5s”~55”~5..89 . 2.08 1..+r..89 1.OF SHEATHS) 101 T .= l~r.le 110 145 190 225 315 480 560 648 925 *Based on recommendations (amperes per conductor) * Circular Rated Line VoltageGrounded 34500 ] 46000 1 69000 Neutral A\Iils. (B.5 5.. Ratings based on opencircuited sheath operation.... is determined similarly.)  1 115000 1 138000 Four paralleled cables similar to the threeCircuit: conductor belted cable described in Example 1.W. . and the other terms are defined as for Eq.*. Z. TABLE 20SUGGESTED WITHSTAND FOR CABLES WITH METALLIC Chapter 4 IMPULSE VOLTAGES COVERING* I ! Insulation Class kv .1. 347 365 402 438 470 530 585 635 667 685 775 875 957 1030 1 100 . (b) General equivalent circuit. 1. 282 300 367 390 430 470 502 568 628 688 715 740 845 955 1043 1 125 1 200 _ 70 0 00 000 0000 250000 300000 350000 400000 500000 600000 700000 750000 800000 1 000 000 1 250 000 1 500 000 1 750 000 2000000 Deg. Z.+r. for six loaded cablea in a duct bank... . is determined similarly.  C... using for S the centertocenter spacing between cables..08 1 .90 I 1 GROUND1 1 BRANCH 1 J nn 1. Problem: To find the overall zerosequence impedance’ of the circuit.e. r OF ONE CONDUCTOR ND. Z. 18Four threeconductor cables in a duct bank Example 5). i. 1 .+j(3~~+~. Ratings include dielectric loss and skin effect.08 1..600 (r 0 tr.)] ohms.oo 0. 335 352 392 427 460 522 578 630 658 680 762 852 935 1002 1070 (a) Cable configuration. OF CIRCUITS CONDUCTOR BRANCH i65 870 982 1 075 1 162 1 240  .i LIaximum 75 Copper TemperatureDeg. (26).2xd)] ohms... multiply above values by 0. 70 75 286 310 367 395 440 482 512 592 650 710 740 75 .
812)4(5)“(10)4(15)2 T. sity paper. CABLE MIDDLE CONDUCTORS / / 1 ( 9 . 1. IN.286+j3. oilfilled insu15 Foust and Scott. oilfilled in12 Foust and Scott. positive and negative waves.Chapter 4 Electrical Characteristics of Cables 93 G>~R of the four conductor groups. solid insulation.q =0. solid insulatibn (General Elect& 14 Foust and Scott.812 inches.y72\ in i n4114‘1 . combination fbd ‘nsu’at’on. LU\LIJ. p0 5% ti. medium density paper. Fig. paper. combination reg6 Faust and Scott.838 log :l” 4. From Fig. r”‘at’on (General Cable Corporation). high density paper. s *n unpublished test. regular density lation (Genera’ Cabli Corporation).““I\. solid insulation. paper.as) = 1~(0. EC1uivaIent spacing of three conductors to their sheath. Kay: test. OUTER CONDUCTORS 0. (. ’ An unpublished test high density paper.. 9 Foust and Scott. v:. paper. s. 0.848 ohms per mile. 400 MILS 500 600 impulse tests on paperinsulated cables piesented by Foust and Scottl3). 18(b).600 I” SPACING I I I / I/ / 9k o5: f: +3”SPAClNG 92 90 .330 =0. AAL.0 SO. capacity of flat config 0 0 100 200 300 INSULATION THICKNESS.L.5 x 40 positive wave. regular density paper.~.=‘$(0. average of five tests 1 x 10 positive wave ular and medium density paper.2 SO. ..” ~>lII(~. Oilregular and medium den10 Foust and Scott. regular density Co. resistance of the sheath branch.5 x 40 positive wave.n :. solid insulation.68’ 1. solid insulation. solid insulation.NG I / i I” ’ ’ ’ FROM (l.>ID among the conductors and the sheaths. Also fronl =0.12 1 x ‘0 negative wave.‘dv. 13 Foust and Scott. uration.Lcz. solid insulation (Simplex Wire and Cable Co).4’i9 inches. ’ STEEL ’ ’ ’ :0’2345678 DISTANCE Fig.r)r)v . Three phase system.~16. regular density’paper. 20Summary of some (based on information . solid insulation (The Okonite 8 An unpublished ’ Davis and Eddy.lT negative wave. Paper. solid insulation.260 ohms per mile.~. 1. ’ “e’d and Leichsenring positive wave solid insulation..\v.5 x 5 positive wave.286+jO.X/6”SPAC.lQfK\ %pedance of the groundbranch 2800X 12 = 0. 11 Foust and Scott.5 x 40 positive wave.A 29. CABLE. solid insulation.. 3 “e’d and Leichsenring. 0. solid insulation. 0.338)4(5)6(10)4(15)2=3.). . high density Company). IN. 19Effect singleconductor IO ’ OF CABLES PLATEINCHES of steel plates on currentcarrying cables. regular density 1 “e’d and L e ’ c h senring. GMRI.
28G+j3. No.E. and grade or compounding of insulation. 1933. C.itute of Technology. F. Vol. H. Specifications for Impregnated PaperInsulated LeadCover Cable: “Solid” Type (7th and 8th editions). Vol. by It. 9. Rosa and Glover.18 the withstand voltages for representative cables may be listed as in Table 20.94 Electrical Characteristics of Cables 6. V.848(0. by D. Vol. Test data from various sources is available. Lei senring. A. The Transmission of Electric Power. Neher. A. Impulse Strength of Cable Insulation by E.Z. by C. The Electric Journul.. 61. Impulse Strength of InsulatedPowerCable Circuits. 1. Trunsactions. A. 2. 1939. 10. by Andrew Ccmant and Joseph Stichcr. 1933. Vol. Symmetrical Compon. Proximity Effect in Cable Sheaths. Evans (a book).I. 47. p. Vol. W. October and November 1932. A. 3rd edition.848+(0. McGrawHill Book Corn pany. July 1940. 14091412.G. and S. October 1938. as suggested by Halperin and Shanklin. 63 1944.. Illinois Inst.. The Behavior of High Tension Cable Installations Under Effect of Voltage Impulses.E. A. 59. 44. 1916. 16. Underground Systems Reference Book. McGrawHill Book Company. W W.E. Electric Cables.E. by J. E1. 8. by H.I. wave shape and polarity of the test impulse voltage. 3.A. 1943.R. by Herm Halperin and G. July 13.275 ohms per phase per mile. September. Technical Paper 49134. 247+jo. Whitehead. October 1948. February 1 15. 35. S320. Some ImpulseVoltage Breakdown Tests on OilT Insulated Cables. Calculation of the Electrical Problems of Underground Cables. A.j3. M. SC Transacfions. M. or substantially the same as with the sheath and ground in parallel. 4. by L.260) ’ = 1. Muller. 20.095Sj0. 1190. Vols.oigi z”=0.Z. 62. 903.E.06 ohms per phase per mile. 67. Brieger. The Transmission of Electric Power. so that the spread of the test points is wider than might be obtained with uniformly controlled test conditions.Z. Several variables are inherent in the curves.ents.E. “OilFilled” Ty GasFilled” Type (1st edition) (4th edition). F.E.E. Lewis. 1935. Eddy. Transactions. by E. NELA Publication N 050. Determination of Cable Temperature by &leans of Reduced Scale Models. by C. Journul. A.I. JuneJuly 1939. AC Resistance of Large Size Conductors in Steel Pipe or Con. Wagner and R. Wagner and H. 1932. Foust and J. 4. Jr. 14. Impedance and Capacitance 60. 41. 18. Books 40. Vol. 46. by W. 497. 7. 67. Vol. Simmons. Westinghouse Technical Night School Press. duit. it is possible that cable insulation will be stressed by transient overvoltages caused by switching operations. 237. P29226. Shanklin. F. “LowPressure prepared by Association of Edison Illuminating Companies.E. W ruff. Booth. . REFERENCES 1.247fj0. 1938. Transactions. = (0. E. E. May to November. In this case the effect of high sheath resistance is minimized by the fact that four sheaths are paralleled. pp. the relation between actual voltage gradient within the insulation and the average gradient. N. December 1932. p. The Electric Journal. D. Pa. 1946. 51. Ev McGrawHill Book Company. 390. C. 1948. Even when circuits are totally underground.E. I and II.I. Simmons. 74. 5.E. F. For these reasons the impulse strength of cable insulation is information of some value for predicting cable performance in an actual installation. Electric Power Transmission and Distribution. Reactance of Large Cables in Steel Pipe or Conduit by W. J. by W Nesbit. Del Mar. Whitehead. by R.5482. 389. 1931. 17.12*13and some of these results for paperinsulated cables are shown in Fig. IMPULSE STRENGTH OF CABLES Powertransmission circuits are often made up of cables and overheadline sections connected in series. Held and H. The zerosequence impedance with sheath and ground in parallel. by C. 993. 0. Tran tions. Hutchings. 17451758. Formulas and Tables for the Calculation of Mutual and Se Inductance. Paris. Impedance of ThreePhase Secondary Alains in Nonmetallic Iron Conduits. by L. B. by W. Paper No. S169. July. 12. pp. (The first article in this series contains a comprehensive bibliography for 1932 and before. B.E.848 = 1. Calculation of the Electrical Problems of Underground Cab1 D. Electric Light and Porter. D. Bulletin. Using 1200 volts per mil average stress as a safe ivithstand impulse strength for paperinsulated cables. A. p.I. The factors not yet completely investigated include the effect of normal insulation aging.1 ohms per phase per mile. May.) 2. 13901396.I. 1918. Vols. Symmetrical Components by C. 43. Illinois Institute of Technology. The zerosequence impedance considering all return current in the sheath and none in the ground. The Temperature Rise of Cables in a Duct Bank. The Principles of Electric Power Transmission. 29. 337. 67.E. John Wiley & Sons.0797) f0. The absolute value of this impedance is 1. by W. camp The Insulated Power Cable Engineers Association.022+jO.E. Wiseman. 1931.E. Cheek. 1948. Vol. 207.0797 ohms per phase per mile. Vol.260) +. Transactions. N. by H. p. Current Carrying Capacity of Impregnated Paper. Del Mar. John Wile & Sons. 1926. 29. A. 63. Bureau of Standar Scientific Papers. Wagner and R. ZeroSequcncc Impedances of Parallel ThreeConductor Cables. by C. C. pp. Rubb Varnished Cambric Insulated Cables (1st edition). Waddic Chapman & Hall. Vol. 65. The absolute value of this impedance is 1. p.E. 83. April 1949.E.E.E.E. East burgh. E. p.Z. October 1938. and 527. 11. The Electric Journal. Pub1 No. pp. pp.Salter. Transactions. Lewis (a book). 59. 1924.E. Messrs. C. p. by J. M. Calculations of Inductance and Current Distribution in Voltage Connections to Electric Furnaces. Vol. Problems in the Measurement of AC Resistance and Reactance of Large Conductors. Unbalanced Currents in Cable Groups.286+. Vol. 283. 42. and this construction may impose lightningsurge voltages on the cable insulation. also So. 47. A. 1948. 45. I and II. Transactions. Electrical Characteristics of Transmission Circuits.Z.E. 1948. No industrywide standards have been established for cable impulse strength. Transactions. by C. Le . CurrentRating and Impedance of Cables in Buildings and Ships. Impregnated Paper Znsululion. H. Dwight. A.
H. and HighPressure GasFilled Cable. H.E. Thermal Transients and Oil Demands in Cables. E. CurrentRating and Impedance of Cables in Buildings and Ships. by E. E. 1950. Jr. A. Cable and Accessories.]. 1nsululion Under the 140.I. Bulletin. Vol. 67. 35.E. p. September.E.I. Trunsuctions.E. 69. A. I. II. edition. Buller and J. July 1940. by C. H.:mce Large Cables in Steel Pipe or Conduit. (Bibliography on M. Vol.E.E. 1949. Shanklin. 336. T. 1939.I. Underground Thermal Characteristics of a 120 kv HighPressure.I E. Vol. 68. B..~1 hlar. . Vol. Power Factor Measurements on Polyphase and Multiconductor Cable Using SinglePhase Bridges. 69. Part I. 1949.I. r. Vol. 151. G. July 1940. Boast. 58. VI. Paris. 541. by Herman Halperin. Part II. W. p. Cable Installation. Davis and W. Transactions. 180. Collins. B.E. Komives. and S. Systems Reference Book. 61.E. 237. Transactions. A.E.E.I. A.I. A. Gasfilled Cable Impregnated with ExtraHigh sulated Cables.I. W.E. 63. by C. 17. \V&b and 0. 601. Vol. E. 1166. 1944. by E.nbalanced Currents in Cable Groups. by F. p. p. Trunsuclions. p. 0.E. 83.fl. Wiseman.E.I. JuneJuly. Vol.E. 342. Association. 144. 1390. 5th l~o:rd Ratings of CableII.E. 66. September 1942. 1948).E. 59. C. 1531. by L. by J. [. 67. Resistance and Reactance of ThreeConductor Cables. by F. Thomas. Vol. F. Salter.I. Scale Models.I. Shanklin. Transactions. Wagner and H. p. NELA Publication 181.G. and S. Specifications for Wire and Cable with Rubber and Rubberl~:~~ouomicalLoading of HighVoltage Cables Installed in UnLike Insulations. Transuclions. 53. Part I. Wiseman. Muller.I. Transuc~ions. 1943. Rating of Cables in Relation to Voltage. Atkinson. Vol. Cable: “Solid” Type (7th and 8th editions.. A.I. by L. Atkinson and Simmons. Impulse Strength of InsulatedPowerCable man Halperin and G.E. by W. W. p. I)Y I).E. Sanderson. Vol. Hatcher. 154. TRZ~Sp. B.I. Church. pp. H. by I.I..E.. Transactions. 1948. ‘I”‘~mclions. p. 394.E. 395 and 423. by P. 68. Specifications for Impregnated PaperInsulated LeadCovered in Service (committee report). by E. Transactions. A. Vol. C. . A. Paper No.E.I. I. The Behavior of High Tension Cable Installations Effect of Voltage Impulses. A. Wollaston. A. Greebler and G. and R. Wollaston. Shanklin and Buller. by E. Neher. B. A. and Tmnsaclions. Part I. A. Vol. Journal.E. 72.. 70.E. (1st edition. ‘I’ranspositions and the Calculation of Inductance from GeoIII(~Iric: Mean Distances. by H. by Herman Halperin. R.E. The Electric Journal.I. December. Booth. p. Atkinson. by K. Vol. by H. p. l(. p. Simmons. HighPressure. Vol.E.t+oblcms in the Measurement of AC Resistance and Reactance _ .\C R&stance of Large Size Conductors in Steel Pipe or Con.I. Transient Temperature Phenomena of 3Conductor Cables. p. H. comnating Companies.E E.lriit. till. Vol. D. I !U:l. p 931. A. p. 157. 158. Vol. 54. 142. A.E. Church. A. actions. p. and f. 29.E. Journal. Faucett.I.E. 120 kv HighPressure GasFilled Cable. 1411.I. 120 kv CompressionType Cable. 67. Neher. Ylti. 1949. Trunsuclions. TransantI September 1932. Louis Meyerhoff. October 1938.I.E. M. by C. 1284. W. October 1942. Vol.E. M. Transactions. Atkinson. 38.I. H. 1942. Transactions. The Temperature Rise of Cables in a Duct Bank. Eager. 1948. Vol.o:rtl Rnt. 61. S. Determination of Cable Temperature by Means of Reduced by D. 153. October 1946. 497.E.E.E.~f Large Conductors. E. May 1932.E. McGrath. 487. Dielectrics) by D.I. M. M:rsimum Safe Operating Temperatures for 15 kv PaperIn156. Cheek. M. Biblioaruvhies l’ublication No. p. 1939. Simmons.E. p. 141. W. Miller and F. Vol. Some ImpulseVoltage Breakdown Tests on OilTreated PaperInsulated Cables. 114. Transactions. by Andrew Gemant and Joseph Sticher. by J.~in&ratures in Electric Power Cables Under Variable LoadA.E. and R. 680. %croSequence Impedances of Parallel ThreeConductor Cables. Vol.E.I. Transactions.E. “LowPressure Current Carrying Capacity of Impregnated Paper. by E. Impulse Strength of Cable Insulation by E. T. Whitehead. 68. Shanklin. Thomas. 61.E. p.I.E. Specifications for Varnished Cambric Insulated Cable. p. Booth. P.ings of Cable. 71.E. H. prepared by Insulated ~hmund Subway Systems. 67. 41. 152. Vol.E. 357. . Simmons.E. 58. Vol. Whitehead. 1947). W. impedance Measurements on Underground Cables. Part II.E.I. 207. 1946. Jr. Vol. Viscosity Oil.E. 1949. 83. A. 359. Vol. p. 1948. Jr. p. by Joseph Sticher. and R. . of 1). The Electric Journal. Hutchings. Hutchings. 1931. p. September 1942.&. 1421.E. Transucliens. W. A.I. prepared by Association of Edison Illumiam1 Varnished Cambric Insulated Cables (1st edition). 117. Electrical Characteristics of Cables 65. 1931. Franklin and E. The Temperature Rise of Buried Cables and Pipes. 1948. 52. p. MeyerhoIT and G. Circuits. JV.E. 115.E. by C. The Thermal Resistance Between Cables and a Surrounding Pipe or Duct Wall. 389. p.I. C. A. Part II.15. Low. 390. N. p. 1950. p. Trunsuctions.E.E. by R. November 1935. G.E.E. Transactions. Leichsenring. Vol. Foust and J. Ijo. 1922. Oilfilled A. 69. October 1939. A. October 1938.E. and 182. Transactions. p. 55.E. October 1938. L.~:Lc~t. 68. Electric Light and Power. Trunsaclions. A.E. p.I. Cable for Power Transmission Ships. A. September 1946. by C. Medium. Vol. p. October 1948. 497. 719. #J. W. Scott. by R. Vol. by W. Faucett. W. by R.E. p. 611. 1409. lrnpcd:mce of ThreePhase Secondary Mains in Nonmetallic and iron Conduits. H. 155. 116. 658. 98.E. Vol. Rubber. Vol. Salter. Transactions. Brieger.I. by Her143.E. [. A. cc/ions. A. t’:conomicnl Loading of Underground Cables. December. 61. T. AC Resistance of Segmental Cables in Steel Pipe. I. Part I. A.E. Greenfield. 1931. December 1934. Neher. A. P29226. J. GasFilled” Type Type (4th edition. Barnett. A. Trunsuctions. GasFilled No. 1581. Calculation of the Electrical Problems of Underground Cables. “OilFilled” September 1943. 73. H. E. A. ml~ti by The Insulated Power Cable Engineers Association. Vol. . Vol.E. 9. 050. A. A.E. 1950. A. A.Chapter 4 64. 1st edition. 475. A. August Komives. 63. 1190. 58. Transactions.fl. Held and H. 1946. L. A.. CurrentRating and Impedance of Cables in Buildings and and Distribution. by E. :I. 606. Transuclions. Transactions. prepared by Insulated Power Cable Engineers ‘I”‘~~~~~~ctions. October 1939. 59.I. W. 68.I. p. February 1938. 68. 67. The Electric Journal.E.E. 74. l. Calculation of the Electrical Problems of Underground Cables. R. T. April 1936. March I. Manz. H. 65.E. $. 1947). Vol.. Part I. Transactims. 1946. p. Vol. Vol.E. C. 29. Electric Light and Power. by W. 0. !182. Joseph Sticher. Y  113. Appendix II. Power Cable Engineers Association. 72. Transactions. by I. p. ing. p. 652. F. (:riidc for Wartime Conductor Temperatures for Power Cable 159. p. Collins.I. Transactions. 535. 1949.I.R. Heat Transfer Study on Power Cable Ducts and Duct Assemblies.E. A. Doan. 1931. by G. General Characteristics of Oilfilled Cable. 71. 1950. Tr~~wctions. Part I. Eddy. 556. R. p.
44fnzAB.. and can be neglected for the present. Hence.. It is expected that the listed references will be consulted for a more detailed consideration of each section. a voltage is induced in tf S winding. Hobson and R. and sets up an alternating flux about that winding. it is well to review open and let a sinusoidal voltage be impressed on the j briefly the fundamental theory of transformer action.former. w=27rf ep= . that flows in the P windin Two windings on a common magnetic core are pictured under this condition (I. (4).x 1OV volts p 42 = 4. therefore. S.2rfnl&. THEORY Thus it is evident that a sinusoidal flux linking a coil in duces in it a voltage which is also sinusoidal and whit: 1.44fnlAB.x 1O8 volts. links both windings. drop in the winding is also ignored. The current.CHAPTER 5 POWER TRANSFORMERS Original Authors: AND REACTORS Revised by: J. refer again to Fig. as made up of two component! one produced by the linkages resulting from the mutuj flux and the other produced by leakage flux. Let the number of turns in the P winding be nl. X 10a volts. all of the mutual flux which has jut been considered in connection with the P winding mu! also link the S winding. then the mutut flux must also be sinusoidal and the induced voltage += f&3X sin wt (1) given by Eq. The ratio of the leakage flux to th mutual flux depends on the relative reluctance of thei respective paths. whit consists of two parts: a mutual flux whose path is who11 in the core and which. n =cross sectional area of magnetic circuit il square centimeters (assumed uniform). Then the voltage induced in the P winding at any instant E.. cos wtX 1O8 volts where hence. and some of the physical characteristics of power transformers and reactors. Fundamental Cowiderations lags the flux by 90 electrical degrees. = 4. which in turn is a function of the saturr E ES tion of the core and the magnitude of the current. L. operating practices. if the small i. Similarly. B nlllx=maximum flux density in the core in line per square centimeter. .. by the flux is eP= 721*X 10e8 volts dt = nlw@. E. the voltage induced i and the number of turns in the X winding be Q.. Then. as assumed. Witzke and J. In the 01 dinary commercial transformer the leakage flux is sma Fig..44fnqAB. with close approl that there is a flux in the core which links both windings imation. by the flux linking it.44fnlAB. L.. pertinent application data. an a leakage flux whose path is partly in air and which link only the P winding. = 4. the standards of operation and present practices regarding distribution transformer application are not included in this chapter. E. 1. cos wt X IO* volts 96 I (2) By hypothesis. I. Assume the P winding by the mutual flux can. It : convenient to consider the voltage induced in the P wine ing. William1 and the rms value of this voltage is E =%y&n. If. Although the fundamental theory presented here holds also for distribution transformers. = 4.=O) is called the exciting curren in Fig.. winding. (3).. Grounding transformers are included since they are ordinarily associated with power systems. f =frequency in cycles per second. the latter is sinusoidal. the rms voltage induced in the 5’ winding by th flux is given by E. be set equal and opposite to the impressed voltagl and is a sinusoidal function of time. No attempt is made to give a complete exposition of the material. which is expressed by Eq. R.X 1O8 volts. Witzke N this chapter are included the fundamental theory. lTwowinding transformer. (4 I. 1 and consider the S winding a lyzing their effect on system operation. Before going into the various problems involved in the To apply the above principle to the operation of a tram application of transformers and the methods used in ana.X 1O8 volts (3 where.
(91 ‘I’hc leakage flux produced by I. Referring again to Fig. therefore. the total flux linking either winding can be divided into two components. and obtain stated. However. The ideal transformer is defined as having no losses. and. 2(a). with all impedances on the primary voltage base. respectively. with magnetizing current considered.gc flux about the winding. The equivalent circuit is shown in Fig. (5) rcmcmbering that the flux caused by I. the resultant flux in the core produced by t. 1 can be derived2. and the primary and secondary voltages and currents involved in transformer action. Consideration of the energies involved shows that an additional component. 1.=nJ. permits writing ~. such that the magnetomotive force acting on the magnetic circuit remains unchanged after 5 is closed.. they are based on a number of assumptions that. a current. losses. is opposite that co:mscd I. flows and. tends to Il.intiing is open. nJ. it is small as in the case of the P winding. in analyzing the operaI. sets up a mutual and leak:.hc combined action of the currents flowing in the P and 8 \vintlings must equal the mutual flux present when the S \\. a leakage flux whose path is wholly or partly in air and a mutual flus most of which lies in the iron core.. As previously di e. the classical equations for the (!()hpled circuits are IppLu 00 1s oa’ E. where N. the induced volt:~gcs. and to insulate the two windings. the exciting current is small in comparison to the normal load current I’. the mutual coupling between circuits must have an energy component to furnish the iron loss in the magnetic circuit. 2Equivalent circuits for twowinding transformer. I. The coefficients Lp. with the magnetizing branch neglected. and requiring no exciting current. and M are not constant but vary with the saturation of the magnetic circuit’. and considering instantaneous currents and voltages. The ratio of transformation for the ideal transformer is N. I. hence me can :Lssume total current. With the above considerations in mind the equivalent circuit representing the two coupled windings in Fig. the effective resistances of the primary and secondary windings: Lp and LS are the “(‘hinductances of the primary and secondary windings: :‘nd 111 the mutual inductance between the two windings. and leakage fluxes is therefore required.r&ced by the P winding and it.? . nJa. (e) Equivalent circuit in percent. is ‘rl~e positive direction of current flow in the two windings is taken such that the fluxes set up by the two currents ‘vilf be in opposition.. (d) Equivalent circuit in ohms. if neglected along with the resistance drop. (c) Equivalent circuit in ohms. with the leakage impedance referred to the secondary voltage base.ion of the transformer or of the system to which it is connccted. (b) Equivalent circuit in ohms.=Mp~sis~s’ dt I (e) Fig. fn other words. (a) Equivalent circuit in ohms. in the manner already described in connection with the P winding. cannot always be made. must be L. Furthermore. A more rigorous developmont that takes into consideration the effects of exciting “urrent. where the mathematical artifice of an idea1 transformer2 is introduced to preserve actual voltage and current relationships at the terminals. which accounts for the negative sign. In by :L w&designed transformer.ddcdto the current in the P winding before the S winding is closed.~llify the flux in the core. Therefore.chapter 5 Power Transformers and Reactors 97 If the circuit connected to the S winding is closed.hcrelation between the P and S voltages as (b) ‘I’hc seven equations developed above summarize the general relationships between the flux. induces a voltage in the S winding opposing that produced by the mutual flux. in the P winding to be equal the to l’r. IIowever. I. .. The direction of this current is such that the mutual flux produced by it opposes that . no impedance drop.% I EPX di dt \vhcre Rp and Rs are. Ls.‘.
98
Power Transformers and Reactors
Chapter 5
The shunt resistance branch in 2~ represents the iron losses and the shunt reactive branch path for the no load, or exciting current, of the transformer. The variation in M during the cycle of instantaneous current and voltage variation is ignored, and a mean value is used. The branches, Zr = Rpfjw +j and, 2s = Rs
are essentially constant, regardless of in
stantaneous current variations, since their corresponding leakage fluxes lie mostly in air. Zp and 2s are components of the leakage impedance between the P and S windings such that 1 zps = zp+zs N2 * (10)
Fig. 3Equivalent
LP
(b)
Zps is defined as the leakage impedance between the P and S windings, as measured in ohms on the P winding with the S winding shortcircuited. Actually it is not possible3 to segregate Zrs into two parts, Zr associated with the P winding and Zs associated with the S winding by any method of test; for example, Zr, the portion of Zrs associated with the primary winding, varies with excitation and load conditions. It is customary, in many calculations involving the equivalent circuit, to make zp=izg=4zp3 * (11)
circuit and corresponding for twowinding transformer.
vector diagram
The ideal transformer can be shifted to the right, as in Fig. 2(b), to get all branches of the circuit on the same voltage base. Since the impedance of the shunt branch is large compared to Zrs, it can be omitted for most calculations involving transformer regulation, and the equivalent circuit becomes that of Fig. 2(c). A notable exception to those cases where the shunt branch can be disregarded is the case of the threephase coreform transformer excited with zerosequence voltages. This will be discussed in detail later. The form of the equivalent circuit given in Fig. 2(c) can be changed to show the leakage impedance referred to the secondary voltage, by shifting the ideal transformerto the left, as in Fig. 2(d). For this condition Zsr, the leakage impedance between the P and S windings as measured in ohms on the S winding with the P winding shortcircuited, is related to Zps as follows:
and currents are identified there. The primary and secondary leakage impedances Zp and Zs are shown separately, and the primary and secondary resistances Rp and RS are also indicated. Ih and I, represent the coreloss component and the magnetizing component respectively of the exciting current I,. The vector diagram in Fig. 3(b) is drawn for a 1 :l ratio of transformation and for a load of lagging power factor. The powerfactor angles at the P winding terminals and the S winding terminals are designated in the diagram as BPand 0s respectively.
II. ELECTRICAL CHARACTERISTICS 3. Transformer Impedances
The turns ratio of a twowinding transformer determines the ratio between primary and secondary terminal voltages, when the transformer load current is zero. However, when load is applied to the transformer, the load current encounters an apparent impedance within the transformer which causes the ratio of terminal voltages to depart from the actual turns ratio. This internal impedance consists of two components: (1) a reactance derived from the effect of leakage flux in the windings, and (2) an equivalent resistance which represents all losses traceable to the flow of load current, such as conductor 12R loss and stray eddycurrent loss. Impedance drop is conveniently expressed in percent, and is the impedancedrop voltage expressed as a percentage of rated terminal voltage, when both voltages are referred to the same circuit; in threephase transformer banks, it is usually appropriate to refer both impedancedrop voltage and rated voltage to a linetoneutral basis. Percent impedance is also equal to measured ohmic impedance, expressed as a percentage of “normal” ohms. Normal ohms for a transformer circuit are defined as the rated current (per phase) divided into rated voltage (linetoneutral). Representative impedance values for distribution and power transformers are given in Table 1; for most purposes the impedances of power transformers may be considered
The equivalent circuit using percentage impedances, percentage voltages, and currents in per unit is given in Fig. 2(e). An ideal transformer to maintain transformation ratios is not required.
2. Transformer
Vector Diagram
The vector diagram illustrating the relationship between the terminal voltages, the internal induced voltages and the currents in the transformer of Fig. 1 can be drawn directly from the equivalent circuit for the transformer. This circuit is repeated in Fig. 3(a) and the various voltages
Chapter 5
TABLE 1 TRANSFORMER
Power Transjormers and Reactors
IMPEDANCES TABLE ~TRANSFORMER IMPEDANCES (Continued)
99
(a) Standard Rcactances and Impedances for Ratings 500 kva and below (for 60cycle transformers)

(b) Standard Range in Impedances for TwoWinding Power Trsnsformcrs Rated at 55 C Rise (Both 25 and GOcycle transformers) HighVoltage Winding Insulation Class
kV
RatedVoltage 15
1iver
Class in kv 25 Average Reactante % 4.4 4.8 4.9 5.0 5.1 Average Impedante % 5.2 5.2 5.2 5.2 I 5.2 Average Reactante % 69
2.5
Single

pt,ase Kf:1
I{atirrg*
Average
I&
_3 10 25 .jO 100 500
actante Cl,’ /O 1.1 1.5 2.0 2.1 3.1 4.i
Avcrage Impedante % 2.2 2.2 2.5 2.4 3.3 4.8
age Reactante
 Yo
0.8 1.3 1.7 2.1 2.9 4.9
Average Impedante %
Averwxe Impedante %
LowVoltage Winding Insulation Class
kV
T
Impedance Limit
in Percent Class FOA FOW Min. 6.75 8.25 9.0 9.75 9.75 10.5 10.5 12.0 11.25 12.75 12.0 13.5 15.0 12.75 14.25 15.75 13.5 15.75 17.25 15.0 17.25 18.75 16.5 18.75 21.0 12.0 12.0 13.5 13.5 15.0 15.0 16.5 15.75 18.75 18.0 21.0 23.25 19.5 22.5 25.5 21.0 24.0 27.0 22.5 25.5 28.5 24.0 27.0 30.0
Class OA ow OA/FA’ pOA* OA/Ff Min. hlax. 7.0 8.0 8.0 9.0 9.0 10.0 10.0 11.0 10.5 12.5 12.0 14.0 15.0 13.0 15.0 17.0 15.0 16.0 18.0 15.0 17.0 19.0 16.0 18.0 20.0
fillIll!
2.8 2.4 2.3 2.5 3.2 I 5.0 . *I?<,r threeptxm transtorrners use )5 ot me
with rated
15 25
15 15 15 25 25 34.5 34.5 46 34.5 69 34.5 69 92 34.5 69 115 46 92 138 46 92 161 46 92 161
4.5 5.5 6.0 6.5 6.5 7.0 7.0 8.0 7.5 8.5 8.0 9.0 10.0 8.5 9.5 10.5 9.5 10.5 11.5 10 11.5 12.5 11.0 12.5 14.0

6.3 6.3 6.4
6.5 6.5 6.5
34.5
linetoline voltages.
threephase kV% ratmg. and enter
46
:Lsequal to their reactances, because the resistance componcnt is so small. The standard tolerances by which the impedances may vary are f 73 percent of specified values )‘oI.t,wowinding transformers and f 10 percent for three\virltIing, auto, and other nonstandard transformers. The percent resistance of transformers is less consistent :tmong various designs than is the impedance, and though t.ho curves in Fig. 4 show definite values for transformer resistance, considerable deviation from these figures is possible. Transformers can be designed to have impedances within closer tolerances than mentioned above, or impedances outside the normal range, but usually at extra cost. A guide to the impedances of threewinding transformers is given below (this guide does not apply to autotransformcrs). (1) Select a kva base equal to the kva rating of the
69
92
115
138
161
196
230
OA/FA/FOA _ _ .  ^ Uehnrtlo” Of trsnstormer cI&sses:
*The impedances are expressed in percent on the selfcooled rating of 0. A/F.% arld
OWOilimmersed. watercooled. OAOilimmersed. self cooled OA/FAOilimmersed, selfcooled/forcedaircooled. OA/FA/FOAOilimmersed, selfcooled/forcedaircooled/forced oil cooled. FOAOilimmersed, forcedoilcooled with forced air cooler. FOWOilimmersed, forcedoilcooled with water cooler. Note: The through impedance of B twowinding autotransformer can be estimated knowing rated circuit voltages, by multiplying impedance obtained from this table by the factor (T).
500
1000
I
I 1000 2000
SINGLEPHASE THREEPHASE
I 2000
I
II 5000
TRANSFORMERS
I 10000
I
I 20000
KVA
I
II 50000 100000
4000
I0000
TRANSFORMERS
20000
40000
KVA
Fk
aPercent
resistance
of transformers, ratings.
based on OA kva
largest capacity winding, regardless of voltage rating. All impedances will be referred to this base. (2) Select a percent impedance between the mediumvoltage and the highvoltage circuits (ZM&), lying between the limits shown for twowinding transformers in Table 1. (3) The percent impedance between the mediumvoltage and lowvoltage circuits (ZMI,%) may lie between the limits of 0.35 (ZM,%) and 0.80 (ZM&~). Select a value of ZML% lying within this range.
100
Power Transformers and Reactors
Chapter 5
(4) Having established Znnye and Z~iLa/o, the percent impedance between the highvoltage and lowvoltage circuits (Zn,%) is determined as follows:
&IL% = l.lO(ZMH%+Z.\IL%)
(13)
at a specified power factor is reduced to zero, with the primary impressed terminal voltage maintained constant. Percent regulation can be calculated at any load and any power factor by an approximate formula: Regulation = where : “Regulation” is a percent quantity; r = percent resistance =load losses in klv, at rated kva x 1OO rated kva (14)
When impedances outside the above ranges are required, a suitable transformer can usually be supplied but probably at increased cost. 4. Regulation The full load regulation of a power transformer is the change in secondary voltage, expressed in percent of rated secondary voltage, which occurs when the rated kva output
r
DO
A PERCENT POWER FACTOR 95 90 8070 50’ ’
_,REGULATION CHART
1 IILL
.I
\ \ \ \ \ \ \ \ .3 .2
.’
2.0 2 I . ’ 2.2 . 2.3 2.4 2.5 . 2.8 2.7 2.8 2.9 3.0
DIRECTIONS:A STRAIGHT :5 ‘1 EDGE PLACED ACROSS THE I RESISTANCE AND REACTANCE : SCALES WILL GIVE INTERCEPTSON THE”A”AND”B” SCALES, THE . : 7: SUM OF WHICH WILL GIVE THE . PERCENT REGULATION FOR THE. 6 : RESPECTIVE POWER FACTORS. E.G.,AT 1.0% RESISTANCE AND. 10.0% REACTANCE,THE REGULAm TION WILL BE’ AT 100% P.F. 1.0+.5=1.5X 11 80%” u6.8+.3=7.I% 18 50% u II 9.2+.’ =9.3x _ . 7 . . 8 c . 9
L : : .+” : . 1 :’ : . : : . ‘I. 
 lo lo_ .!I I
,! *\ .I
.I
\ .3 ‘.‘\
O 1: I ’ L2 . 11 g1; . l3
,
‘\
.I .! I#( .2 ,.4 .3
\
1213. “ If, L314 I! I5 . 50 It
3.1 COPYRIGHT 1911 BY J.F. PETERS . 3.2 COPYRIGHT 1921 BY H.L. COLE DO
.I.: .6 i.4 F
. 8
: SO
I2 8070
PERCENT POWER FACTOR 95
ii
Fig. SChart
for calculating
regulation
of transformers.
Chapter 5 z = percent impedance =
Power Transformers and Reactors
101
impedance kva x 100 rated kva 2 = percent reactance = dzz _ r2 p=cos e g=sin e e=power factor angle of load (taken as positive when current lags voltage). The fullload regulation of a transformer can be determined for any power factor from the chart in Fig. 5; this chart is based on Eq. (14). Typical regulation for threephase transformers at full load and various power factors is shown in Table 2. These
TABLE ~APPROXIMATE REGULATION FOR 60CYCLE THREEPHASE TYPE OA TRANSFORMERS AT FULL LOAD Insulation Class kv 15
loss; that is, all losses incident to magnetization at full voltage with the secondary circuit open. Load losses are 12R loss caused by load current, eddycurrent loss induced by stray fluxes within the transformer structure, and similar losses varying with load current. Noload losses are measured at rated frequency and rated secondary voltage, and can be considered as independent of load. Load losses are measured at rated frequency and rated secondary current, but with the secondary shortcircuited and with reduced voltage applied to the primary. Load losses can be assumed to vary as the square of the load current.
6. Methods of Calculating Efficiency Conventional MethodThis method is illustrated
below for a transformer having 0.50 percent noload loss and 1.0 percent load loss at full load. Percent noload loss is determined by dividing the noload loss in watts by 10 times the kva rating of the transformer, and the percent load loss (total minus noload) is determined bv dividing the load loss in watts by 10 times the kva rating of the transformer. Note that the noload loss remains constant regardless of the load whereas the load loss varies directly as the square of the load. Percent load.. . 50.00 25.00 100.00 75.00 (1) Percent load loss.. .25 ,062 1.00 .562 (2) Percentnoloadloss.. .50 .50 .50 .50 (3)
Sum of (2) and (3). Sum of (1) and (4). Dividing 100 times (4) by (5). . Subtract (6) from 100 1.50 101.50 1.48 98.52 1.062 76.062 1.40 98.60 .75 50.75 1.48 98.52 .562 25.562 2.20 9i.80 (4) (5)
Lagging Power Factor Percent
80 90 100 80 90 100 80 90 100 80 90 100
_1000 kva 4.2 3.3 1.1 5.0 4.0 1.2 6.1 4.9 1.4 7.7 6.2 1.8
Percent Regulation 10 000 kva 3.9 3.1 0.7 4.8 37 0.8 5.7 4.4 0.9 7.2 5.6 1.2 9.7 7.6 1.7 5.5 4.2 0.6 7.0 5.4 0.9 9.4 7.3 1.3 :ansformers, at
100000

kva
34.5
69
(6)
(efficiency)
138
SlideRule
Percent Percent Percent Sum of Sum of
MethodThis
method is illustrated
75.00 .50 .562 1.062 76.062 50.00 .50 .25 .75 50.75 25.00 .50 .062 .562 25.562
for
(1)
the same transformer.
load ............... no load loss. ....... load loss. .......... (2) and (3). ........ (1) and (4). ........ loo.00 .50 1.00 1.50 101.50
(2)
(3) (4) (5)
80 90 100 Note: These fipures apply loadscorrespondingto their (
230
2
figures also apply, but less accurately, to transrormer banks made up of three singlephase transformers; in this casethe table should be entered with the threephase bank kvn rating. The regulation of threewinding transformers can be calculated directly from transformer equivalent circuits, if the impedance branches and loading for each circuit are known. The regulation of fourwinding transformers may also be calculated using formulas developed by R. D. Evans.4
At rule, 1. 2. 3.
this point the operations are continued on the slide and are described here for the full load point only: Set 1.5 (sum of noload and load losses) on D scale. Set 101.5 over this on the C scale. Now starting at the right end of scale D, read the first figure (i.e., 1) as 90, the next (i.e., 9) as 91, the next (i.e., 8) as 92, etc., until 98.52 is read under the left end (i.e., 1) of scale C. This 98.52 is the percent efkiency at full load.
5. Definition
of Efficiency
The efficiency of a transformer, expressed in per unit, is the ratio of real power output to power input; Output = 1 _ Losses Efficiency = ___ (15) Input’ Input Total losses are the sum of the noload losses and load losses. Noload losses are eddycurrent loss, hysteresis loss, 12R loss caused by exciting current, and dielectric
This procedure is repeated in a similar manner for other loads. NOTEIf the sum of the percent noload and load loss at full load is 1 percent or less, the first figure at the right end of D scale (i.e., 1) is read as 99 percent and the second figure (i.e., 9) is read as 99.1, the third figure (i.e., 8) is read as 99.2, etc. If the sum of the percent noload and load loss is greater than 1 percent as in the case illustrated above, the right end is read as 90 percent. In calculating the values for the other points, judgment will indicate whether 90 or 99 is to be used as the first figure on the right end of scale D.
102
Power Transformers and Reactors L= Fe=L ti Cl1 v%
Chapter 5
Chart MethodThe chart in Fig. 6 may be used to calculate transformer efficiency at various loads. The procedure is described in the caption below the chart. 7. Loss Ratio and Product
Maximum operating efficiency for a transformer results when the noload (constant) losses equal the load (variable) losses. This condition will likely occur at some load less than rated kva: CuXL2=Fe (16)
(17)
where: L=per unit kva load at which transformer operates most efficiently. Cu = load losses at rated load, kw. Fe = noload losses, kw. R = loss ratio =load loss at rated load noload loss *
EFFICIENCY CHART  .o
I114 .9 F .I .8 FULL LOAD *9 + .2 46 99.5L .3 .4 .4 .3 .2 : z= = r z 2 .9 .3 .2 .I r 1.1 98.0.9 y I.2 .8 97.77 97.8 .7a 2 0 .5 I Ii 2 5 .6tJ7 : 5 .7 5 5 6 cb .6 98.5.4 99.0.9 .8 .8 .I .9 .B .7 98.5 97.598.5 ,. .4 .3 .2 .I 98.0 .9 .6 98.5 98.0 97.8 .9 .4 .3 98.0 .2 .I _ 97.5 ,.9 .8 .7 .6 96.07 .998.5.8.6 1.0 .4 .3 .2 97o.I .?.6 s ii &SL 1 = _ .9 99.0 99.0 .9 .8 .7 98.0.I .2 .I .2 .I _ 99.0 *’ 98.53 0 J I’4 2 2 3 2.5.4 .3 .4 .3 .3 .2 1 .399.5 99.5 .B *? .6 3/4 .7 $6 99.5 ** 99.0.2$6 3 .B l/2 l/4 .9 .B , .7 99.5.I.Ol 
.? 
.8s
1.3
COPYRIGHT 1911 BY J.F. PETERS
Fig. 6Chart for calculation of efficiency. Directions: A straightedge the known full load copper loss and iron loss points will give intercepts scales for various loads.
placed between on the efficiency
Chapter 5
Power Transformers and Reactors
0 TRANSFORMER RATINGWA (a) SINGLEPHASE
.
A
3
4
5
4 6 IO 20 40 60 100 TRANSFORMER RATING MVA (b) THREEPHASE
Fig. 7Relation
between transformer loss ratio efficient loading.
and the most
‘1’1~ relation between loss ratio and most efficient transformer loading is shown in Fig. 7. The range through which loss ratio may vary in normal transformer designs is shown by Table 3. The product of percent noload and load losses is a quantity that has become standardized to the extent that it is predictable with fair accuracy for large power transformers.
TABLE 3 Normal Limits of Loss Ratio, R I Voltage Class kv ,I(i wtl below. Ii!) 10 138, incl.,
~\hc 138.................
Fig. STypical values of product of percent losses (percent fullload copperloss times percent iron loss). For OA/FA or OA/FA/FOA units use OA rating to evaluate product. For FOA and FOW units use 60 percent of rated kva to evaluate product.
Cu = kfi xv’%
100
‘*
kw
(19)
where: .
P =product
(G/Fe) FOA*+ FOP*
1.4
of the percent va:lues of noload and
Loss Ratio, OA, OW OA/FA” OA/FA/FOA*
l.i5 to 3.25 1.50 to 2.75 1.25 to 2.00
R =
load losses, (ygx(q). kva = transformer rating.
to 2.4
8. Typical Efficiency Values
Conventional transformer efficiency is given on the basis of losses calculated at (or corrected to) 75 degrees C and
TABLE 4APPROXIMATE VALUES OF EFFICIENCY FOR 60CYCLE, TWOWINDING, OA, THREEPHASE POWER TRANSFORMERS
.
1.2 to 2.0
1.0 to 1.8
*hcd on lossev at OA rating. **Hlixed on losses st 60 percent of FOA or FOW rating.
Fig. 8 shows typical values of the product of percent losses, as :L function of transformer size and voltage rating. TO
(Full load, unity power factor, at 75°C) kva
2000 10 000 50000
values of noload and load losses for a particular rating it is first necessary to select values Of IWS ratio R and loss product P from Table 3 and Fig. 8. ‘when the respective loss values, in kilowatts, are given h!lo\\:
"stirnate
~Unsformer
I
15 kv
98.97 99.23 I 34.5 kv 98.89 99.22 1 99.47
Voltacre Class
69 kv 98.83
138 kv
98.56
161 kv
98.47
99.17
1 QQ.45
99. 12
1 99.44 /
99.11
'.)9.4l
(18)
Note: These figuresapply also to OA/FA and loads corresponding to their OA ratings.
OA/FA/FOA
transformers.
at
104
Power Transformers and Reactors
Chapter 5
unity powerfactor load unless otherwise specified. Table 4 (rives approximate values for GOcycle power transformers it full load, unity powerfactor, and 75 degrees C.
III. TRANSFORMER 9. Forms of Construction.
CLASSIFICATIONS
Coreform construction for singlephase transformers consists of magnetic steel punchings arranged to provide a singlepath magnetic circuit. High and lowvoltage coils are grouped together on each main or vertical leg of the core, as shown in Fig. 9. In general, the mean length of turn for the winding is comparatively short in the coreform design, while the magnetic path is long.
Shellform construction for singlephase transformers consists of all windings formed into a single ring, with magnetic punchings assembled so as to encircle each side of the winding ring, as in Fig. 10. The mean length of turn is usually longer than for a comparable coreform design, while the iron path is shorter. In the design of a particular transformer many factors such as insulation stress, mechanical stress, heat distribution, weight and cost must be balanced and compromised5. It appears that, for wellbalanced design, both coreform and shellform units have their respective fields of applicability determined by kva and kv rating. In the larger sizes, shellform construction is cmite appropriate; the windings and magnetic iron can be assembled
\
/
.x
Fig. 9Coreform construction.
.

_ ^ ,,
,,
... 7zu!..“.
Fig. loShellform
construction.
Chapter 5
Power Transformers and Reactors for small and medium size transformers, portable substations can provide spare capacity on short notice. If transportation or rigging facilities should not be adequate to handle the required transformer capacity as a single unit, a definite reason of course develops for using three singlephase units.
11. Types of Cooling
Basic types of cooling are referred to by the following designations.6 OAOilImmersed SelfCooledIn this type of transformer the insuiating oil circulates by natural convection within a tank having either smooth sides, corrugated sides, integral tubular sides, or detachable radiators. Smooth tanks are used for small distribution transformers but because the losses increase more rapidly than the tank surface area as kva capacity goes up, a smooth tank transformer larger than 50 kva would have to be abnormally large to provide sufficient radiating surface. Integral tubulartype construction is used up to about 3000 kva and in some cases to larger capacities, though shipping restrictions usually limit this type of construction at the larger ratings. Above 3000 kva detachable radiators are usually supplied. Transformers rated 46 kv and below may also be filled with Inerteen fireproof insulating liquid, instead of with oil. The OA transformer is a basic type, and serves as a standard for rating and pricing other types.
1:~. 11Assembly of 1.5 000 kva threephase transformer, showing “formfit” tank being lowered into position.
()I1:t steel base structure, with laminations laid in horizonr,:llly to link and surround the windings. A closefitting I,:& member is then dropped over the core and coil assemhly and welded to the steel base, completing the tank assembly and also securing the core to the base member. ‘I’his “formfit” construction is shown in Fig. 11; it is more compact than can be achieved by assembling a core form unit, within a tank, and the flow of cooling oil can be (lilocted more uniformly throughout the interior of the coil assembly.
OAIFAOilImmersed
SelfCooled/ForcedAir
10. Comparison of SinglePhase and ThreePhase Units for ThreePhase Banks
.\ threephase power transformation can be accomplished 41 her by using a threephase transformer unit, or by intercLotmcctingthree singlephase units to form a threephase Ijlink. The threephase unit has advantages of greater c~flicicncy, smaller size, and less cost when compared with :LI):mk having equal kva capacity made up of three single\)ll:lse units. \l’hen three singlephase units are used in a bank, it is Possible to purchase and install a fourth unit at the same l(~(Uion as an emergency spare. This requires only 33 Wccnt additional investment to provide replacement caPicity, whereas 100 percent additional cost would be rerluircd to provide complete spare capacity for a threephase ‘Init. However, transformers have a proven reliability l1igl~erthan most other elements of a power system, and for this reason the provision of immediately available spare Wacity is now considered less important than it once was. ‘rl~rcephase units are quite generally used in the highest Of circuit ratings with no onthespot spare transformer ‘apacity provided. In these casesparallel or interconnected circuits of the system may provide emergency capacity, or,
CooledThis type of transformer is basically an OA unit with the addition of fans to increase the rate of heat transfer from the cooling surfaces, thereby increasing the permissible transformer output. The OA/FA transformer is applicable in situations that require shorttime peak loads to be carried recurrently, without affecting normal expected transformer life. This transformer may be purchased with fans already installed, or it may be purchased with the option of adding fans later.
Fig.
12Installation phase,
view of a 25 000 kva, 11512 kv, three60 cycle, OA/FA transformer.
106
Power Transformers and Reactors and a second rating based on forced circulation fans or blowers.
Chapter 5 of air by 1 I
11 The higher kva capacity attained by the use of fans is dependent upon the selfcooled rating of the transformer and may be calculated as follows: For 2500 kva (O4) and bclon: kva (FA) = 1.15xkva(OA). (20) For 2501 to 9999 k~a (Oh) singlephase or 11 999 kva (O4) threephase: kva (F’a) = 1.25Xkva (OX). (21) For 10 000 kva (Oil) singlephase and 12 000 kva (O4) threephase, and above: kva (FA) = 1.333 x kva (OA). (22) These ratings are standardized, and are based on a hottestspot copper temperabure of 65 degrees C above 30 degrees C average ambient.
IV. POLARITY AND TERMINAL 12. SinglePhase Transformers
MARKINGS
i j 9
.! ,: 1 2 1
Primary and secondary terminals of a singlephase transformer have the same polarity nhen, at a given instant of time, the current enters the primary terminal in question and leaves the secondary terminal. In Fig. 13 are illustrated
4 Hz HI Es
ES H2 ER x2 XI
OA/FOA/FOAOilImmersed SelfCooled,‘ForcedOil Forced  Air Cooled,‘Forced  Oil Forced  Air
CooledThe rat,ing of an oilimmersed transformer may be increased from its Od rating by the addition of some combination of fans and oil pumps. Such transformers are normally built in the range 10 000 kva (0.1) singlephase or 12 000 kva (OA) threephase, and above. Increased ratings are defined as two steps, 1.333 and l.GG7 times the OA rating respect’ively. Recognized variations of these triplerated transformers are the OA/FA/FX and the OA4/FA/FOA types. Automatic controls responsive to oil temperature are normally used to start the fans and pumps in a selected sequence as transformer loading increases. type of transformer is intended for use only when both oil pumps and fans are operating, under which condition any load up to full rated kva may be carried. Some designs arc capable of carrying excitation current with no fans or pumps in operation, but this is not universally true. Heat transfer from oil to air is accomplished in external oiltoair heat exchangers. OWOilImmersed WaterCooledIn this type of watercooled transformer, the cooling water runs through coils of pipe which are in contact with the insulating oil of the transformer. The oil flows around the outside of these pipe coils by natural convection, thereby effecting the desired heat transfer to the cooling water. This type has no selfcooled rating.
2
In
El
EP x2
XI
I,
XI X2
Xl
.I ‘; ;
“2
‘1 I
ES HI
FOAOilImmersed ForcedAir CoolerThis
ForcedOilCooled
With
(a)
SUBTRACTIVE POLARITY ADDITIVE
(b) Is
POLARITY
Fig. 13Standard
polarity
markings formers.
for twowinding
trans
FOWOilImmersed ForcedOilCooled Wi;h Forced Water CoolerExternal oiltowater heat exchangers are used in this type of unit to transfer heat from oil to cooling water; otherwise the transformer is similar to the FOA type. AADryType SelfCooledDrytype transformers, available at voltage ratings of 15 kv and below, contain no oil or other liquid to perform insulating and cooling functions. Air is the medium which surrounds the core and coils, and cooling must be accomplished primarily by air flow inside the transformer. The selfcooled type is arranged to permit circulation of air by natural convection. AFADryType ForcedAir CooledThis type of transformer has a single rating, based on forced circulation of air by fans or blowers.
singlephase transformers of additive and subtractive polarity. If voltage is applied to the primary of both transformers, and @ace& leads connected together, HI to Xl in Fig. 13(a) and HI to XB in Fig. 13(b), a voltmeter acrossthe other pair of terminals [Hz and X, in Fig. 13(a) and Hz and X1 in Fig. 13(b)] indicates a voltage greater than E, if the transformer is additive as Fig. 13(b), and less than E, if the transformer is subtractive as Fig. 13(a). Sdditive polarity is standard for all singlephase transformers 200 kva and smaller having highvoltage ratings 8660 volts (winding voltage) and below. Subtractiire polarity is standard for all other singlephase transformers.6
13. ThreePhase Transformers
The polarity of a threephase transformer is fixed by the connections between phases as well as by the relative locations of leads, and can be designated by a sketch showing lead marking and a vector diagram showing the electrical angular shift between terminals. The standard angular displacement between reference phases of a deltadelta bank, or a starstar bank is zero degrees. The standard angular displacement between reference phases of a stardelta bank, or a deltastar bank, is 30 degrees. The present American standard for such threephase banks is that the highvoltage reference phase is 30 degrees ahead of the reference phase on the low voltage, regardless of whether the bank connections are stardelta or deltastar.6 The standard terminal markings
AAIFADryType
This
SelfCooled/ForcedAir
Cooled
design has one rating based on natural convection
Chapter 5
“3
Power Transformers and Reactors
107
V. STANDARD
HIGH VOLTAGE
LOW
INSULATION Class
CLASSES
H2 HI
VOLTAGE
14. Choice of Insulation
DELTADELTA OR STARSTAR CONNECTION (b) LINETONEUTRAL I?ib. IdStandard
STARDELTA OR DELTASTAR CONNECTION VOLTAGE VECTOR DIAGRAMS for
polarity markings and vector diagrams threephase transformers.
for a threephase, twowinding transformer are illustrated in l;ig. 14. 41~0included are the vector diagrams for deltaclclta, starstar, stardelta and deltastar connected transI’ormers. The phase rotations are assumed to be HIH%H3 :1ntl x,,Y?x,. Icig. 15 summarizes the phase angles that can be obtained I)ct\vccn high and lorvvoltage sides of stardelta and delta
The standard insulation classes and dielectric tests for poxver transformers are given in Table 5. The insulation class of a transformer is determined by the dielectric tests lvhich the unit can Jvithstand, rather than by rated operating voltage. On a particular system, the insulation class of the connected power transformers may be determined by the ratings and characteristics of the protective devices installed to limit surge voltages across the transformer Jvindings. Ratings of the protective devices Jvill in turn depend upon the type of system, its grounding connections, and some related factors. For example, Ivhen the system neutral is solidly grounded so that a grounded neutral (SO percent) arrester can be used, an insulation level corrcsponding to the arrester rating may be chosen rather than an insulation level corresponding to the system operating voltage. Many transformer banks having a starconnected threephase xvinding, vGth the neutral permanently and solidly grounded, have an impulse strengt,h corresponding to a lower linetoline classification than indicated in Table 5 (See Chap. 18 for a more detailed discussion of this subject).
15. Dielectric
Tests
E og c Ii,
LAGS
EA6BBY 30’ x,c ob
Eag LEADS A ‘H3 Bo
EAG BY 30’ X,*0
The purpose of dielectric testing is to sholv that the design, workmanship, and insulation qualities of a transformer are such that the unit will actually meet standard or specified voltage test limits. Below is a description of the various dielectric tests Ivhich may be applied to polver transformers : (1) The standard impulse test consists of applying in succession, one reduced full wave, two chopped waves, and one full wave. (a) A full wave is a 1.5X20 microsecond wave, usually of negative polarity for oilimmersed transformers, or positive polarity for dry type, and of the magnitude given in Table 5. (b) A reducedfull wave is a 1.5X40 microsecond wave, having a crest value betmeen 50 and 70 percent of the full wave crest. (c) A choppedwaveis formed by connecting an air gap to cause voltage breakdown on the tail of the applied lvvave.The crest voltage and minimum time to flashover are specified in Table 5. (2) The standard appliedpotential test consists of applying a lowfrequency voltage between ground and the minding under test, Tvith all other windings grounded. The standard test voltage magnitude is listed in Table 5, and its specified duration is one minute. (3) The standard inducedpotential test in general consists of applying between the terminals of one winding a voltage equal to twice the normal operating voltage of that winding. A frequency of twice rated or more is used for this test, so that the transformer core will not be overexcited by the application of double voltage. The duration of the test is 7200 cycles of the test frequency, but not longer than one minute. Commonly used test frequencies
Eag
LAGS
E,,BY
150’
Eaq
LAGS
EAG BY 90’
Eag
LEADS
Eric
BY 90’
l%. IsAngular
phase displacements obtainable phase stardelta transformer units.
with
three
star, threephase transformers built with standard connections and terminal markings. In this Figure A, B, and C rcl)rcsent the three phases of the highvoltage system, \\herens a, b, and c represent the three phases of the lowvoltage system. Phase rotations ABC and abc are :lSSumed
108
Power Transformers and Reactors TABLE ~STANDARD INSULATION CLASSES AND DIELECTRIC TESTS FOR DISTRIBUTION AND POWER TRANSFORMERS (Taken from Table 11.030 AELI Standard C57.111948 for Transformers, Regulators and Reactors.) Rated Voltage Between Terminals of PonderTmnsformers (a) Low Frequency Tests
Chapter 5

Impulse Tests OilImmersed Transformers 500 kvn or Less Chopped Wave
FUll Bnve (e‘
Insuliltion Class
SinglePhase For YConnection on aPhase System (0) kv rms 0.69 “iii9 5.0 8.66 14.4 19.9 26.6 39.8 53.1 66.4 For DeltaConnection on 3Phase System kv rms 3Phase
Delta
)
OilImmersed Transformers Above 500 kva

kV
or YConnected kv rms (cl
OilImmersed ‘Ihe
Dry
‘YP~ (b:
Chopped Wave Min Time to Flashover hIicrosec. 1.5 1.6 1.8 2.0 3.0 3.0 30 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0
FUll
P v’nve
(c)
Kv Crest _kv rms 10 15 19 26 34 50
i0
kv rms 4 10 12 19 31
Min Time to Flashover Microsec. 1.0 1.25 1.5 1.6 1.8 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0
KV
Crest
Kv Crest
54 “ii 110 130 Ii5 230 290 400 520 630 750 865 1035 1210 1500 1785
values
applv
Kv Crest
45 i5 95
110 150 200 250 350 450 550 650 750 900 1050 1300 1550
1.2 2.5 5.0 8.66 15 25.0 34.5 46.0 69.0 92 115 138 161 196 230. 287 345
Notes: (a) (b)
0.09 (cl
2.89 (d 5.00 (d 15.0 25.0 34.5 46.0 69.0 92 115 138 161 196 230 287 345
1.2 2.5 5.0 8.66 15.0 25.0 34.5 46.0 69.0 92 115 138 161 196 230 287
3G
54 G9 88 110 175 230 290 400 520 630 750 865 1035 1210 1500 1785
30 45 GO 75 95 150 200 250 350 450 550 650
95 140 185 230 275 32.5 395 460 575 690
.. . .
i9.7 93.0
113 133 166 199
750 900
1050 1300 1550
.. .

Intermediate voltage ratings are placed in the next higher insulation class unless otherwise specified. Standard impulse tests have not been established for drytype distribution and power transformers. Presentday *re *s follows: 1.2 kv class. 10 kv: 2.5 class. 20 kv: 5.0 class. 25 kv: 8.66 kv class. 35 kv: 15 kv class. 50 kv. These values
WV&Y62tests.
for impulse to both
tests of such apparatu and full
choppedwave
(c) Yconnected transformers for operation with neutral solidly grounded or grounded through this reduced insulation is below the level required for delta operation, transformers cannot (d) These apparatus are insulated for the test voltages corresponding to the Y connection, so tions. The test voltages for such deltaconnected singlephase apparatus are therefore one (e) 1.5X40 microsecond WBYB.
an impedance may have reduced insulation at the neutrnl. When be operated deltaconnected. that a single line of apparatus serves for the Y and delta applicaStep higher than needed for their voltage rating.
are 120 cycles for 60cycle transformers, and 60 cycles f.or 25cycle transformers. Combinations and modifications of the tests described above are contained in transformer standard publications, for example ASA C57.221948, and these publications should be consulted for detailed information.
16. Insulation
Class of Transformer
Neutrals
Transformers designed for wye connection only with the neutral brought out may have a lower insulation level at the neutral than at the line end. The following rules are included as a guide in selecting the permissible neutral insulation level : (a) A solidly grounded transformer may have a minimum neutral insulation class in accordance with column 2 of Table 6. (b) A transformer grounded through a neutral impedance must have a neutral insulation class at least as high as the maximum dynamic voltage at the transformer neutral during system shortcircuit conditions. In no case
should the neutral class be lower than that given in Column 2, Table 6. (c) If the neutral of a transformer is connected to ground through the series winding of a regulating transformer, the neutral insulation class must be at least as high as the maximum raise or lower voltage (phase to neutral) of the regulating transformer. In no case should the neutral class be less than that given in Column 3 of Table 6. (d) A transformer grounded through the series winding of a regulating transformer and a separate neutral impedance shall have a neutral insulation class at least as high as. the sum of the maximum raise or lower voltage (line to neutral) of the regulating transformer and the maximum dynamic voltage across the neutral impedance during system shortcircuit conditions. In no case should the neutral insulation class be less than that given in Column 3 of Table 6. (e) If the neutral of a transformer is connected to ground through a ground fault neutralizer, or operated ungrounded but impulse protected, the minimum neutral
Chapter 5
TABLE
Power Transformers and Reactors
INSULATION CLASS
109
~MINIMUM
AT TRANSFORMER

NEUTRAL
(3) (4)
(1)
Winding Insulation Class at Line End
(2)
Grounded Solidly or Through Current Transformer
winding. The initial current is assumed to be completely displaced from zero. (b) The duration of the short circuit is limited to the following time periods. Intermediate values may be
Grounded Through Regulating Transformer
.
Grounded Through Ground Fault Neutralizer or Isolated but
11 mpulseProtected
determined by interpolation. Symmetrical Current
in Any Winding 25 times base current 20 times base current 16.6 times base current 14.3, or less, times base current
Time Period
in Seconds
2
_1.2
2.5 5.0 8.GG 15 25 34.5 48
3
4
5
Same as Line
8.66 8.66 8.66 15 15 15 15 15 15 15 15 15 15 8.66 8.66 8.66 15 15 25 25 34.5 34.5 46 46
E Ind
G9
02 115 138 161
8.66 15 25 34.5 46
69 69 92 92
115 138 161

“2. Where kva is mentioned in paragraph 3 the following is intended: When the windings have a selfcooled rating, the kva of the selfcooled rating shall be used. When the windings have no selfcooled ratings, the largest kva obtained from the ratings assigned for other means of cooling by the use of the following factors shall be used:
Multiplying Factor Type of Transformer 1.0 Watercooled (OW) 0.75 DryType ForcedAirCooled (AFA) 0.60 Forcedoilcooled (FOA or FOW)
19G
230 287
345

69 69
196
insulation class shall be in accordance with Column 4 of Table 6.
VI. TEMPERATURE AND SHORTCIRCUIT STANDARDS. 17. Temperature Standards
The rating of electrical apparatus is inherently determined by the allowable operating temperatures of insulation, or the temperature rise of the insulation above ambient temperature. For transformers and voltage regulators with Class A insulation, either air or oil cooled, t>hcrating is based on an observable temperature rise (by r&stance or thermometer) of 55 C above an ambient I’cmperature at no time in excess of 40 C, and the average (luring any 24hour period not exceeding 30 C. Transformers and other induction apparatus are designed to limit the hottestspot temperatures of the windings to not more than 10 C above their average temperatures under continuous rated conditions. The limits of observable temperature rise for aircooled transformers with Class B insulation is 80 C by resistance measurement.
“3. For multiwinding transformers: The base current of any winding provided with esternal terminals, or of any deltaconnected stabilizing winding without terminals, shall be determined from the rated kva of the winding or from not less than 35 percent of the rated kva of the largest winding of the transformer, whichever is larger. “In some cases, the shortcircuit current, as limited by transformer impedance alone, will exceed 25 times base current. It must be recognized that such cases can occur with transformers manufactured according to these standards and that the transformers built under these standards are not designed to withstand such shortcircuit current.” Under shortcircuit conditions the calculated copper temperatures for power and distribution transformers shall not exceed 250 C where Class A insulation is used assuming an initial copper temperature of 90 C, or 350 C where Class B insulation is used assuming an initial copper temperature of 125 C.
VII. TRANSFORMER TEMPERATURETIME CURVES 19. Constant Load
A “heat run” of a transformer on test is made to determine the temperature rise of the various parts at rated load. If the test were made by applying only rated load, with the transformer at room temperature, thirty hours or more would be required before stationary temperatures were reached. Such a process would be quite inefficient of time, energy, and in the use of testing facilities. Accelerated heat runs are made by closing radiator valves, etc., and applying loads in excess of rated load until the expected temperatures are reached. Radiation restrictions are then removed, the load reduced to normal, and the test continued until stable temperatures are reached.
18. ShortCircuit
Conditions
A proposed revision to American Standard C57.121948 (section 12.050) reads in part: “1. Transformers shall be capable of withstanding without injury short circuits on any external terminals, with rated line voltages maintained on all terminals intended for connection to sources of power, provided : (a) The magnitude of the symmetrical current in any winding of the transformer, resulting from the external short circuit, does not exceed25 times the basecurrent of the
110
Power Transformers and Reactors
Chapter 5
It is evident that the temperaturetime characteristics of a transformer cannot be obtained from the accelerated heatrun data. Information is secured from the heat run, however, which permits the temperatures to be calculated under assumed load conditions. Exact calculations are quite involved, but sufficiently accurate results can be obtained by the use of an approximate method due to S. 13. Griscom for estimating the temperatures reached under variable load conditions, changing ambient temperatures, etc. Certain simplifying assumptions can be made that permit a quick estimate of the expected temperatures. Let L = transformer load in kva. IV = total losses (in kw) at load L. TF = final temperature rise at load L in degrees C above the temperature at t = 0. M = thermal capacity in kw hours per degree C. k = radiation constant in kw per degree C. T = oil temperature rise in degrees C at time t above the temperature at t = 0. H = thermal time constant in hours. t = time in hours. If the heat radiated is directly proportional to the temperature rise of the transformer above the ambient, the radiation constant can be obtained from the heat run data for W and TF:
temperature rise, that is, T and TF are defined as before but refer to the topoil specifically. Further, the final topoil temperature rise TF is not directly proportional to the losses for all types of transformers as Eq. (23) would indicate, but is more correctly represented by the relation
TF= TF(II)
W
m
(28)
Total loss at full load >
where: m =0.8 for type OA transformers. =0.9 for type OA/FA transformers. = 1.0 for type FOB transformers. TF~I~) =final topoil temperature rise at full load in degrees C. The use of this relation when substituted in Eq. (23) indicates that for other than the type FOX transformer the radiation constant k :and the time constant H are not completely independent of load but vary according to a small fractional power of the total loss. However for convenience in calculations this variation in k and H is normally overlooked and the values obtained from Equations (23) and (27) for the full load condition are taken as constant. The error introduced by the procedure is not large compared to that normally expected in transient thermal calculations. To determine the temperature rise curve for any load L therefore, the radiation constant k under full load conditions is first determined from the heat run data using Eq. (23). The thermal capacity M is dependent on the thermal capacities of the various parts of the transformer. For convenience it can be assumed that the transformer parts can be separated into three elements: the core and coils, the case and fittings, and the oil. Although the core and coils are of copper, iron, and insulation the specific heats of those elements do not vary widely. Since, also, there is a reasonably constant proportion of these elements in different transformers, a single weighted coefficient of thermal capacity for the coils and core is warranted. The following relation is accordingly suggested:
k=;
(23)
where the temperature at t=O is taken as ambient. Since the total heat generated is equal to the heat radiated plus the heat stored (heat consumed in raising the temperatures of the various parts)
W=kT+Mz
This equation can be solved for T, giving T=$ *( or T=TF ( where H = F = the transformer time constant in hours. 1ei > l& >
(24)
(25)
(29
M= ko[0.06
(wt. of core and coils)
(27)
This derivation may be broadened to show that Eq. (26) is equally correct for the case where the oil temperature rises T and TF are those above the temperature at 1= 0, whether the value then is the ambient temperature or otherwise. The foregoing discussion has been based on the assumption that the temperature throughout all parts of the transformer is the same. This, of course, is not the case. When the transformer load is increased, the copper temperature is above that of the surrounding parts, and when the load is decreased, the copper tends to be more nearly the same temperature as the surrounding parts. Also, the top and bottom oil are at different temperatures. Eq. (26) is therefore commonly taken as referring to the topoil
+0.04 (wt. of case and fittings) +0.17 (wt. of oil)] (29) Here the coefficients of the last two terms are also weighted to make further allowance for the fact that all parts of the case and fittings and the oil are not at a uniform temperature. The values of k and M found as above may be substituted in Eq. (27) to obtain H. The value of TF for the desired load L is determined next by substitution of heat run data in Eq. (28). The quantity W for the load L may be evaluated by the relation
w=
K
L
1
2
Full load kva
x (full load copper loss)
(30) 1 The quantities H and TF may now be substituted in Eq. (26) from which the topoil temperaturerise curve may be plotted directly.
+ (noload loss)
Chapter 5
Power Transformers and Reactors 40
a 6000kva, threephase, selfcooled, For exampk, 24 OOO5040volt transformer has the following full load nerformance data as supplied by the manufacturer: Iron loss= 10 920 watts. Copper loss=43 540 watts. Total = 54 460 watts. Topoil rise = 40 C (from heatrun test data). LV copper rise=46.3 C. HV copper rise=43.3 C. Wt. of core and coils = 25 000 pounds. Wt. of case and fittings= 18 000 pounds. Wt. of oil= 17 400 pounds. From this information the time constant H may be evalu:II c:tl and the expression for T obtained for the load L equal to the rated load. W “‘=a= 54.46 = 1.36 kw per degree C. 40 000+0.04x18 000+0.17x17 4001
‘U = &o[0.06X25
= 5.18 km hours per degree C. I[=!f=.5.‘“=3 li 1.36
y= l’& 
’
81 hours
’
&/3.81)~
e1/H) =J;O(l
‘l’hc full load topoil temperature rise curve shown in Fig. 16 was calculated from this relation. To plot the topoil temperaturerise curve for halfload conditions for this transformer the same time constant H is used as found above. From Eq. (28) : T~=40 (0.5)2X43.54+ 54.46 10.92 y 1g 2 c . . >
Fig.
16Topoil
temperature rise versus transformer.
time,
for a typical
T=19.2 The curve represented by this equation also appears in I$. 16 The rise of the hottestspot copper temperature above h topoil temperature is known as the hottestspot copwr gradient and at full load may be estimated from the r&ttion Grrct,, =Gc(r,,+A. \vlicrc: Gn(h) = hottest spot copper gradient at full load in degrees C. Gccri, =apparent copper gradient at full load in degrees C. A= 10 C for type OA and OW transformers. = 10 C for type GA/FA transformers. =5 C for type FOA and FOW transformers with directed flow over coils. The apparent copper gradient at full load (Go& is the difference between the average copper temperature rise snd the topoil temperature rise, both of which are de(31)
termined during the heatrun. The average copper temperature rise above ambient at full load is required by standards not to exceed 55 C for class A insulation. The use of that value to obtain the apparent copper gradient will generally lead to overly pessimistic results since the actual value of the average copper temperature rise is normally below the limit. Therefore it is advisable to use the value measured on the heat run and obtained from the manufacturer. For any load L, the hottestspot copper gradient may be calculated from the relation
GH(L)=GHW)X
L full load kva
1.6
(32)
From the performance data of the transformer previously cited: Gccrl, =46.340= 6.3 C for the LV winding. GH([,) = 6.3+ 10 = 16.3 C. The hottestspot copper temperature for fullload is thus 16.3 C above the topoil temperature. For, say, halfload, Eq. (32) must be used to obtain GH(L)= 16.3X(O.5)‘6=5.4 C. It is not feasible in a study of this kind to keep track of short time variations of copper or hottestspot temperature, and it is suggested if it is desirable to show roughly how this varies, a time constant of 15 minutes be used.
112
Power Transformers and Reactors
Chapter 5
TEMPEkATURE
I!
t
I
1
6 NOON
6
IO
12
2
Fig. 17Stepbystep
graphical
calculation
of temperatures
under
changing 
load conditions.
20. Variable Load A stepbystep analysis using Eqs. (28) to (32) can be made to consider conditions of variable load, changing ambient temperatures, etc. The method of approach is based on the fact that the initial rate of change of temperature is the slope of a line joining the initial and final temperatures, the two temperatures being separated by a time interval equal to the thermal time const,ant of the transformer. As before TF is calculated from heat run data and the total loss W for each load condition through the use of Eq. (28). The loss W is obtained from Eq. (30). The final topoil temperature is then found by adding TF to the ambient temperature. Since the load is varying, the final temperature cannot be reached for each load condition and the stepbystep analysis must be employed to obtain the topoil temperature curve. Points on the hottestspot temperature time curve may then be obtained by adding the hottestspot copper gradient Gn for each load to the topoil temperature at the time corresponding to the load for which the gradient was calculated. Gn is obtained in the same manner as previously outlined. To illustrate the stepbystep method, the oil temperaturetime curve for the 6000kva transformer previously described will be calculated, starting with an oil temperature of 55 C for an assumed load cycle as tabulated in the adjacent column. Figure 17 illustrates the use of the calculated data in the graphical stepbystep process to plot the curve of topoil temperature with time and the manner in which the hottestspot gradients are added to obtain the hottestspot temperaturetime curve. The accuracy can be increased by using shorter time intervals.

I ha1 Oil Temp. ambient plus final rise 43.7c 43.7 42.7 46.2 53.9 62.9 73.0 75.0 ii.0 65.0 56.9 49.2 13.7 Hottestspot 3radient Eq. (32) 2.8C 2.8 2.8 5.4 8.5 12.2 16.3 16.3 16.3 12.2 8.5 5.4 2.8
Time
12 2 AM 4 6 8 10 12 2 PM 4 6 8 10 12
Ambieni
t .
[I
29c 29 28 27 29 31 33 35 37 34 32 30 29

Final tOil Rise Load Loss 1 (30) 1 zq. Eq. GW (ma) _.. 14.x 2 15.7 14.7 2 15.7 14.7 2 15.i 10.2 21.8 3 24.9 4 30.2 31.9 41.1 5 40.0 54.5 6 40.0 6 54.5 40.0 54.5 6 31.9 41.1 5 24.9 30.2 4 19.2 21.8 3 14.i 2  15.7 

VIII.
GUIDES FOR LOADING OILIMMERSED POWER TRANSFORMERS 21. General The rated kva output of a transformer is that load which it can deliver continuously at rated secondary voltage without exceeding a given temperature rise measured under prescribed test conditions. The actual test temperature rise may, in a practical case, be somewhat below the established limit because of design and manufacturing tolerances. The output which a transformer can deliver in service without undue deterioration of the insulation may be more or less than its rated output, depending upon the following
i 3 I 1 {
1 i
Chapter 5
Power Transformers and Reactors
design characteristics and operating conditions as they exist at a particular time6: (1) Ambient temperature. (2) Topoil rise over ambient temperature. (3) Hottestspot rise over topoil temperature (hottestspot copper gradient). (1) Transformer thermal time constant. (5) Ratio of load loss to noload loss.
160
22. Loading Based on Ambient
Temperature
;\ircooled oilimmersed transformers built to meet established standards will operate continuously with normal life expectancy at rated kva and secondary voltage, providing the ambient air temperature averages no more than 30 C throughout a 24hour period with maximum air ~(ampcrature never exceeding 40 C. Watercooled transformcrs are built to operate continuously at rated output \vith ambient water temperatures averaging 25 C and never exceeding 30 C. When the average temperature of the cooling medium is difTerent from the values above, a modification of the transformer loading may be made according to Table 7. In
TABLE ~PERCENT CHANGE IN KVA LOAD FOR EACH DEGREE CENTIGRADE CHANGE IN AVERAGE AMBIENT TEMPERATURE Air below 30 C nvg. Or or Water rtbove 25 C avg. Water below 25 C avg. 1.5L%,perdeg.C 1.5 1.o* 1.o*
rating.
Air above 30 C nvg.
0I
0
40
80
’
’
120
’
’
160
KVA
’
’
200
’
’
240
’
Type of Coolirlg Selfcooled \Vatercooled ForcedAirCooled ForwdOilCooled
*Based on forcedcooled
PERCENT
RATED
+l .O% per deg. C +1.0 +0.75* +0.75*
Fig. EHottestspot copper rise above topoil temperature as a function of load, for various values of full load copper rise. 2.2
cases where the difference between maximum air temperatllre and average air temperature exceeds 10 C, a new temperature that is 10 C below the maximum should be rlsed in place of the true average. The allowable difference between maximum and average temperature for watercooled transformers is 5 C.
El I1.2
I
I I
l\l\ I IN\
23. Loading Based on Measured Oil Temperatures
The temperature of the hottestspot within a power transformer winding influences to a large degree the detrrioration rate of insulation. For oilimmersed transform~1’st,he hottestspot temperature limits have been set at 105 C maximum and 95 C average through a 24 hour Wriod; normal life expectancy is based on these limits. The topoil temperature, together with a suitable temperature increment called either hottestspot copper rise Ouer topoil temperature or hottestspot copper gradient, is often used as an indication of hottestspot temperature. :lllolvable topoil temperature for a particular constant load may be determined by subtracting the hottestspot copper gradient for that load from 95 C. The hottestspot CoPPergradient must be known from design information for accurate results, though typical values may be assumed for estimating purposes. If the hottestspot copper gradient is known for one load condition, it may be estimated for other load conditions by reference to Fig. 18.
TOP
OIL TEMPERATURE
IN DEGREES C
Fig. 19Loading
guide
based on topoil
temperature.
CA) OA, OW, OA/FA types. (B) OA/FA/FOA, FOA, FOW types.
A conservative loading guide, based on topoil temperatures, is given in Fig. 19.
24. Loading Based on Capacity Factor
Transformer capacity factor (operating kva divided by rated kva) averaged throughout a 24hour period may be well below 100 percent, and when this is true some compensating increase in maximum transformer loading may be made. The percentage increase in maximum loading
81 1 58 1.26 1.21 1.4 is equal TABLE 9PERMISSIBLE DAILY SHORTTIME TRANSFORMER LOADING BASED ON NORMAL LIFE EXPECTANCY  Maximum SIaximum Load In Per Unit of Transformer Period of Increased Loading.18 1.67 1.54 1.50 1.0 8.09 2.79 1.37 1. Short time loads larger than those shown in Table 9 will cause a decrease in probable transformer life.5 1.33 1.66 1. BASED 1 10(a)PERMISSIBLE SHORTTIME Following TRANSFORMER ON REDUCED Following LIFE EXPECTANCY 50 percent or less of rated kva@ 100 percent of rated kvacb’ Type of Cooling Period of Increased Loading Hours 0.68 1.10 2.08/1.81 1.57 1.58 1.5 1.00 blaximum OA CG 0.25 / 0. These conditions were chosen to give results containing some probable margin.26 1.90/0.0511. 1.05 1.91 1.05 OA/FA@’ OA/FA/FOA’“’ FE:.00 1.261 1. OA/FAch’ Rating’“’ Percent Increase. or for a more detailed check on some particular unit.40 1.06/1.11 1.00 1.82 1. the hottestspot copper temperature can be calculated by the method shown in section 19.46 1.21 1.5 0.00 1.0 2.0!) 1.89 1.33 1.0 0.0 2.33 1.05 1.00 1.70 1.06~1.09 1. Shorttime loads which occur not more than once during any 24hour period may be in excess of the transformer rating without causing any predictable reduction in transformer life.59 1.50 1.00 1.33 1. OW Average’“’ 0.20 1.38 1.18 1.49 1.1.00 1.82 1.10 1.15.il 1.10 1 0.40 1.36 1.45 1.00 1.70 o. percentcapwlty factor cut n ?4hour period.53 1.41 1 28 1.56 1.4 0.64 1.18 1.27 1.39 1.31 1.51 1.38 1.32 1.90 0.32 1.10 2.05 1.47 1.0711. OA. following B topoil Based on the FA kva rating.59 1.39 1.68 1. In Per Unit of Transformer Rating 0.13 1.21 1.31 1.io 1.0 4.23 1. operatmg kva to rated kY3 X100.12 1.78 1.14 1.10 1.00 1.35 1.60 1.14 2.36 1.92 1.13 1.00 2.50 1.00 1. A/FOAsnd 3A units. I n.35 1.18 1.19 1. For special designs. as a function of capacity factor. based on a normal transformer life expectancy.20 / 0.0 8.11 1. The permissible load is a function of the average load previous to the period of aboverated loading. but the amount of the decrease is difficult to predict in general terms.18 1.28 1.15 1.17 Load In Per Unit Of Transformer 2.10 'Here. . Some estimate of the sacrifice in transformer life can be obtained from Table 10(a) which is based on the TABLE 26. following a topoil More husicntly.23 2.43 1.24 1.34 1.31 1.31 1.5 0.tl ~ 25 25 20 20 averaged through Hours .92 1.54 1.24 1.69 1.45 1. _.97 1.00 1. rise of 25 C for OA and ( rise of 45 C for OA and ( OA.39 1.10 Probable Sacrifice In Percent of Normal 1 0.48 1.21 1.0 4.0 24.25 1. is given in Table 8.09 1.00 2.14 2.82 1.62 1.73 1.30 1. 0.17 1. according to Table 9.0S~1.50 1.29 1.ed kva For each Percent By Which Capacity Factor Is Below 100 0.16 1.5 1.42 1.26 1.24 1.50 1.0801.50 / 1.76 1.50.20 1.00 / Life Caused By Each Overload 0.15 2.114 Power Transformers and Reactors Chapter 5 TABLE %PERMISSIBLE TRANSFORMER LOADING BASED ON AVERAGE PERCENT CAPACITY FACTORS* Percent Increase Type of Cooling Above Rat.0 0.19 1.09 1.69 1.90 0.25 Rating 1.50 / 1.26 1. when compared with most conventional transformer designs. Limiting of Load by Automatic Control The loading of a transformer can be supervised by control devices to insure that hottestspot copper temperatures LOADING.58 Gq 1.59 1.0 4.00 2.‘C’ FOA SelfCooled Watercooled ForcedAirCooled ForcedOilCoolf.50~0.0 24. (a) (h) CC) (d) More basically.15 2.__~ 0.90 1.50.0 8.48 1.77 1.07~/1. it is necessary to chose one method or the other.24 1. Loading Based on ShortTime Overloads theoretical conditions and limitations described in Table 10(b).15 1.0 24.60 1.22 2.51 1.32 1.92 1. Baaed on the FOA kvn rating.75 1. 1. Regardless of Capacity Factor 04/FA/FOA.70 II 1.06 temperatures ^ L^LI^ of 30C for air and 25C fur water are assun~ed through 25..90 1.37 1.0 2..00 1.94 l. 2 4 8 .38 1.5 1 Initial Load. and the probable sacrifice in transformer life can then be estimated from Table 11.0 2.53 1.20 rise for rise for 1.80 1.31 1.19 1.40 1.10 0.00 1.23 2.11 1. A/FOA and 3A units.72 1.05 1. The load increase based on capacity factor and the increase based on shorttime overloads cannot be applied concurrently.43 1.35 1.47 1. OA.
1 0.25 j 0.\Inximum oil temperature = 100c. Also. 334.. This control may be accomplished with a thermal relay responsive to both topoil temperature and to the direct heating effect of load current. contacts to close can start fans or pumps for auxiliary cooling.a Rlaximum hottestspot copper temperature = 150 C. The relay can be arranged to close several sets of contacts in succession as the copper temperature climbs with increasing load: the first. and inert gas above the oil. transformer. Lightning ProtectionCoordinated arresters are installed to protect both high.0 2.Chapter 5 Power Transformers and Reactors LIGHTNING 115 TABLE 10(b)CONDITIONSAND TRANSFORMER CHARACTERISTICS ASSUMEDINTHE PREPARATIONOFTABLE IO(a) 0. the transformer winding attains during operxtion.4 OW Hottestspot rise (C) Topoil rise (C) Time constant at full 6. and it. 20Singleline diagram of CSP power transformer.0 load (hours) Ratio of full load copperto iroo loss Ambient temperature=30 C. standard switching. Fig. it may be designated a CSP power transformer.0 4. metering.10 1 0.0 3.’ The thermal relay can be coordinated with each transformer design to _~I.00 hours 0. . as indicated in Fig. and voltage regulating functions may be included within a power transformer assembly..5 OA/FA/FOA FOA FOW 65 40 1.5 45 3.5 Temperature In DegreesCentigrade To SacrificeXot bIore Than The Above Perc$nt of Xormal Life 132 142 1. and it can inherently follow unpredictable factors that affect permissible safe loading for a particular installation.b IX. TABLETSPROBABLE SACRIFICEIN TRANSFORMER LIFE CAUSEDBY PROLONGED HOTTESTSPOT COPPERTEMPERATURE Period of High Temperature. The thermostatic element of this relay is immersed in the hot transformer oil. When these protective. (b) Shorttime loading for onehalf hour or more. 20.0 24. Fig. :lntf the final contacts can trip a circuit breaker to remove load from the transformer.0 111 105 99 1 119 112 104 126 119 109 134 126 115 are always within a permissible range and duration. THE COMPLETELY SELFPROTECTED TRANSFORMER (n) Based provision on for oil expansion.5 5.0 124 117 134 126 150 142 134 A power transformer design may include protective devices capable of preventing damage to the unit when it is subjected to electrical conditions that would probably damage conventional transformers. also carries a current proportional to load current: in this way the temperature of the element is geared to the total tempcrature that. Terminals or tapchangers might in some cases impose a limit lower than 200 percent.50 1 1. 21Fully assembled 3000 kva. \rnlch it is applied. . and so that peak loads can be carried without the use of oversize transformers. switching. so that emergency loads can be carried without interruption of power service. and metering features are all combined at the factory within a single unit.\lnximum shorttime loading =200 percent.16 kv CSP power .and lowvoltage circuits from lightning or other voltage surges.5 OA/FA 65 45 2. the next contacts can warn of temperatures approaching the maximum safe limit.. 150 142 8. Loading by copper temperature makes available the shorttime overload capacity of a power transformer.0 2.
13. without reference to the internal winding kva. The relation between primary and secondary winding voltages will depend upon the turns ratio z between these windings. 24Circuit for a twowinding autotransformer. = E. (33) (34) %N1.24.33 kv CSP power transformer. and a secondary winding S which is connected directly in series with the highvoltage circuit. Circuit BreakerLoad switching is accomplished by a circuit breaker in the lowvoltage circuit of the transformer. Here N is the overall voltage ratio between high. responsive to copper temperature (see section 26). the . AUTOTRANSFORMERS 27. CSP transformers are available in kva ratings up to 3000. “=+?a Es =l+Z=N. lowside circuit voltage E.5/4. The units may be used to supply distribution circuits from highvoltage lines in either industrial or electric utility applications. TwoWinding Autotransformer Theory Fig. 22Installation view of 1500 kva.and highvoltage circuits. just as a conventional twowinding transformer does. Autotransformers are normally rated in terms of circuit kva. will be the sum of the primary and secondary winding voltages.‘72 000/24 OOO2. will be equal to the primary winding voltage. if one unit is used individually on a radial circuit.33/5.and lowvoltage circuits. in the autotransformer the two windings are interconnected so that the kva to be transformed by actual magnetic coupling is only a portion of the total kvn transmitted through the circuit to which the transformer is connected.0/7. shown in operation at a substation site. operates to trip the secondary circuit breaker before damaging temperatures develop in the winding.Power Transformers and Reactors Chapter 5 vided in the highvoltage winding. n1 Fig. MeteringWatthour meters and ammeters are usually supplied for circuit metering. a bypassing switch can be supplied across the lowvoltage circuit breaker to permit withdrawal and maintenance of the breaker without a service interruption. RelayingOvercurrent relays normally are provided in the lowvoltage circuit to protect for secondary faults. so that the supply circuit can be cleared from internal transformer faults. highside circuit voltage E’. and secondary voltages up to 15 kv. E. When the transformer is carrying load current. primary voltages up to 69 kv. Overload ProtectionA thermal relay. 23Portable substation rated 2000 kva.5 kv. Tapchangingunderload equipment for the secondary circuit may be built into the transformer housing. X. 24 contains Fig. Voltage RegulationStandard noload taps are pro The singlephase twowinding autotransformer contains a primary winding and a secondary winding on a common core. a primary winding P which is common to both low. Internal Fault ProtectionFusible protective links of high interrupting capacity are connected between the highvoltage bushings and the winding. Under noload conditions. However. The autotransformer circuit shown in Fig.
like voltage regulators. an autotransformer rated 1000 kva. Two conversions may now be made. I.XIP or .’ (expressing voltages in kv). as well as current and voltage transformation. The series impedance may be evaluatcd by referring to Fig. referring to equation (37) up == E.5 has m ) 22 (b) EOUIVALENT OF TEST CIRCUIT I:N circuit voltage ratio of 22 kv to 33 kv equivalent twowinding kva of The reduced rating of transformer parts required in an autotransformer make it physically smaller. za=$2z. in the circuit where it is connected. and of higher efficiency than a conventional twowinding unit for the same circuit kva rating.l)%ps. %ps= (N. here the lowvoltage circuit terminals are shortcircuited. For this reason autotransformers. so that efficiency based on circuit transmitted power would be quite high. the first to move the series impedance to the lowvoltage side. though advantageous in its effect on transformer regulation. In the example just cited. and the second to express impedance in terms of Zps. An autotransformer will introduce series impedance. and this retluced kva would in practice furnish a fairly accurate basis for estimating the cost of the lOOOkva autotransformer. or E. This low series impedance. where z * = Nl 2xz ( N > PS. 25(c). . 25(a). (39) (40) N=I+$ (c) CONVENTIONAL Fig. and also correct through impedance. 25(a) is exactly the same as the circuit that would be used to measure the leakage impedance if Zsp were defined as the ohmic impedance measured :rcross the secondary winding with the primary winding shortcircuited.XI. less costly. Total losses in the autotransformer would be comparable to those in a 333kva conventional transformer. cannot always protect themselves against excessive fault current.Chapter 5 Power Transformers and Reactors 117 primary ampereturns should essentially balance the secondary ampereturns (noting that Ils=Is) : nJ.‘=N~I~.’ = n’Ip nc2 (35) (36) (37) I. with 51 ( N =g = 1.~Ip. Zsp= . Often the through impedance will be less than four percent based on the autotransformer nameplate kva rating. the conventional form of equivalent circuit is shown in Fig. so that the impedance measured at the highvoltage terminals will be equal to the series circuit impedance attributable to the autotransformer. The circuit impedance of an autotransformer is smaller than that of a conventional twowinding unit of the same rating. A circuit providing correct circuit voltage and current ratios.xIs. 0 E. which means that threephase short circuit current could exceed the maximum of twentyfive times normal rated current for two seconds as permitted by standards. Sate that the circuit in Fig.‘XI. 25(b). (41) 0 Sequence equivalent circuits for the threephase twowinding autotransformer are presented in the Appendix. but the winding kva is given by E. Ip Nl N I = iv’ ( NlP ) (33) For example. the autotransformer would theoretically be only as large as a 333kva conventional transformer. nl Q Nl The total circuit kva is given by E.=Is+Ip=I~+nzI. reactors or other connected circuit elements may have to be relied on for this protection.‘=NI. (41). is shown in Fig.XIp UC E&XI. 25Equivalent EQUIVALENT CIRCUIT autotrans circuits for a twowinding former. = nJs = nJ. may allow excessive shortcircuit currents during system fault conditions. The ratio between winding b) DETERMINATION OF IMPEDANCE TEST kva (Up or Us) and circuit kva (UC) is. as is evident from Eq. From this. =$zsp.
will be: AE. Furthermore. or in the secondary (series) winding. The percent change in 6)1+11 E.118 28. This winding could be very small in capacity if it were required to carry only harmonic currents. additional padding would be required throughout the tapped section.= 1 (42) nl = turns on primary winding.‘=E.+tz3 nl (43) remain (44) In this case the transformer volts per turn normal. This results in a threewinding autotransformer with terminals to accommodate three external circuits. The equivalent circuit for this type of transformer is given in section 59 of this chapter. taps placed adjacent to the line.’ is: AE& = trnl+n2 1 1 29.0 per unit based on normal rated lowside circuit voltage. if the transformer materials are to be used most effectively. The percent change in E. R= turns on secondary winding. E. tl = fractional part of n1 included in primary winding tap (+ tlnl indicates additional turns) t2= fractional part of nZ included in secondary winding tap. more cases are possible: Taps in secondary winding only: n2t2 &L?LA!(4% nl+n2 nl+lzzXnl+n41+t2) Fig. is assumed constant at 1. and E. The percent change in EJ is: A&‘= tl x n2 100.’ is: 0 NORMAL TAP E: 3 4 AE. and in most of these cases it is desirable to have a third winding on the core deltaconnected so as to carry the third harmonic component of exciting current.=tztlx~l()(). 26. The percent change in E. (50) (5) Taps in primary winding only: m nl E. It is frequently necessary to place taps in the windings of an autotransformer to regulate either or both circuit voltages. not considering taps. constitute a weakness that can be avoided by placing the taps in the middle of the winding as shown in Fig.‘=l 4 1 n2 1 t2t1~% m l+tl 121 As in case (2). under noload conditions. Autotransformer Taps n2 100. The ThreeWinding Power Transformers and Reactors Chapter 5 1 nl(l+td+n?(l+t2)E nl(l+td a* Autotransformer Threephase autotransformers for power service are usually starstar connected with the neutral grounded.‘=l+. some tap combinations are more desirable than others. (+tznz indicates additional turns).+ *.0 per unit based on normal highside circuit voltage. where the most severe voltage stresses occur. or in both windings: however. Taps (2) Taps in primary winding only: (45) The transformer volts per turn are times their normal value.=nl+~+nl+n2xnl(l+tl)+na n2tl (51) . If E. The transformer volts per turn are their normal value.=& n2 nl+n2U+t2) nltnz nl+n2(l+t2) > times may be placed in either the primary (common) winding. A widely used rule sets the deltawinding rating at 35 percent of the autotransformer equivalent twowinding kva rating (not circuit kva rating). l+tl nl+n2 If El rated three (4) (4% ElY 0 a is assumed constant at 1. is allowed to vary. by an equation which takes account of both primary and secondary taps: 100. nlfn2 l+tl (46) (3) Taps in both primary and secondary windings: E. It is not advisable to place taps adjacent to the line connections for voltages above 22 000 volts. The lowside and highside circuit voltages may be related. not considering taps. 26Autotransformer taps. it is often advantageous to design this winding so that load can be taken from it. three cases are possible: (1) Taps in secondary winding only: E. the transformer volts per turn are 1 times their normal value. because extra insulation is necessary on turns adjacent to the line terminals. Since it is necessary in most cases to have a deltaconnected tertiary winding. If taps were placed at the ends of the winding. but its size is increased by the requirement that it carry high currents during system ground faults.
building a grounded autotr:msformer to withstand a fullvoltage applied potential ((‘St mould not be economical because of the excess insulaticll near the neutral. These undesirable (4fcct. The percent change in E. . a significant incrcnse in voltsperturn at some tap setting would be rcllccted in a magnetic core of larger size than otherwise ncccssary. and the two circuits must operate with no angular Phase displacement unless a zigzag connection is introduced. this ia lnilike the conventional twowinding transformer which isolates the two circuits. the ari’:Lngement of taps is more complicated.b c c AE. then the percent change in E. because of the nullifying effect of the :tccompanying change in voltsperturn.=tl n2 n1(lSh)Sn2 100.and highvoltage circuits cannot be isobed .chapter 5 Transformer volts per turn are their normal Vahe.?. would be: a b (55) . it has disadvantages including low reactance which In:lY make it subject to excessive shortcircuit currents. regardless of tap position. and h(‘ttcr regulation as compared with the twowinding transformer.t2) E. ‘1’0 summarize the preceding discussion. Ihc lowvoltage circuit will be subject to overvoltages originating in the highvoltage circuit.and highvoltage circuits. higher efficiency. If the neutral of an :urtotransformer is always to be grounded in service. 30. Also. The percent change in Ed is: 100. 27Stardelta and zigzag grounding transformers. Unless the potential to ground ol’ each autotransformer circuit is fixed by some means.’ is: Power Transformers and Reactors %f% %(l+tl)+nz 119 times D b a .scan be minimized by connecting the neutral of the :ultotr:msformer solidly to ground. the autotransformer has advantages of lower cost. an intluccd potential shop test is more appropriate than an :Wied potential test. a tap change in c)rily one Gnding may be less effective than would normally 1)~ anticipated.lY have to carry fault currents exceeding its standard rating. the low. As inclicated in some of the cases above. = tlX 100. A. (6) Tups in both primary and secondary: n2(4 . because it represents more closely 11~ field operating conditions.nnt voltsperturn. The advantages of lower cost and improved effi (d) EOUIVALENT CIRCUIT OF AN INTERCONNECTEDSTAR GROUNDING TRANSFORMER Fig. (54 ‘~‘=nl(l+tl)+n2(l+tz) lf the transformer were designed for constant volts per n2(t12) (a ) STAR DELTA GROUNDINGTRANSFORMER (b) INTERCONNECTEDSTAR GROUNDING TRANSFORMER WINDINGS DRAWN PARALLEL ARE ON THE SAME CORE furn (t2= tlz).2!L+zx nl+n2 n1(l+t~)+n2(l+Q)’ nlfn2 Transformer volts per turn are nlfn2 (53) 777 Y ( nl(l+tl)+nz(l+tz) ) times their normal value. the delta tertiary *‘l.. lt is often advisable to specify a tap combination which .qll allow the autotransformer to operate at practically (:()rist. . Autotransformer Operating Characteristics (c) SCHEMATIC WINDING ARRANGEMENT OF AN INTERCONNECTEDSTAR GROUNDING TRANSFORMER OF THE THREEPHASE COREFORM CONSTRUCTION An autotransformer inherently provides a metallic conncction between its low.
be smaller than Zps by an amount depending on the forn lished in terms of an “equivalent twowinding 55 C kva” of core construction: a typical ratio of 20 to Zps is 0. 4 minutes 0. though the only exciting current circulates in the windings of a grounding transformer.065 0.153 0. The reduced size can be estab. El T=Zo=Zps.064 3 minutes 0.‘.060 for a specified time.170 0. for both threephase and singlephase units.225 0. assuming supply voltage to be fully StarDelta ImpedancesThe impedance to zeromaintained.092 for establishing a neutral ground connection on a three1 minute 0.102 Devices” (No.196 0. Zigzag ImpedancesThe impedance to zero sequence currents in each phase of a solidly groundec U.113 0.080 0.3f~~v~ ZigZag Connection 46 kv 1 69 kv 1 92 kv Time Rating A grounding transformer is a transformer intended solely 10 seconds 0. 0 Values for K are listed in Table 12 for various types and Percent zerosequence impedance for the zigzag connec .043 0. rarely exceeding 2 to 1. Conventional power transformers may be connected to to flow in the grounding circuit only when a fault involving serve solely as grounding transformers. = 3 Uo X Kl for a bank of singlephase grounding reference to Fig. When when a solid singlelinetoground fault is applied at the these modified ratings are desired.160 0.057 0.Oi. ards.170 0. they should be obtained from the transformer manufacturer.076 1 0. The values are applicable only for groundinp trsnsformers designed to have 100 percent impedance. by applying a reduction factor K to the shorttime though variation from this value for different design’ rated kva of the grounding transformer. May 1947) can be specified depending *These valuev are cnlculated on the basis that the initial uve~we winding temupon the probable duty to be imposed on the unit in perature is not over 7.073 0.080 0. = Uo X KS for a threephase grounding unit.046 2 minutes 0.064 0. 32.122 phase system.204 0. = equivalent twowinding .I. Percent zerosequence impedance is normally expressed ir Equivalent circuits for stardelta and zigzag grounding terms of shorttime kva and linetoline voltage: transformers with external neutral impedance are includZPS x UG (59’ ed in the Appendix. its size and cost are less than for a continuous duty transIn a threephase stardelta grounding transformer 20 ma: former of equal kva rating. (delta) winding: Sometimes a resistor or other impedance is connected in the transformer neutral. kva can be used for a price estimate. units (57) Eo=IoXZp~ep+ep.113 0. 030 where U. the table includes values When operated at rated threephase balanced voltage.85 U. (56) zigzag bank can be derived on a theoretical basis b: U. 27(d).:.104 0. Rated voltage of a grounding transformer is the linetoline voltage for which the unit is designed. 27. 5 minutes 0.127 O.216 0.064 0. is the product of normal linetoK1.068 0. threephase. and in these cases it may be ZO=ZPS desirable to specify that the grounding transformer shall have less than the conventional 100 percent impedance. Current of appreciable magnitude begins singlephase type is uncommon.078 0. Transformers so delta grounding bank made up of singlephase units is designed are said to have loopercent impedance based on equal to Zrs.088 0.167 0.55C kva. UC= (linetoneutral kv) X (rated neutral amperes).118 0.212 Fig.080 for grounding transformers.264 0. The transformer is usually of the stardelta 2 minutes 0.046 0. but the current and ground develops on the connected system.5’2. XI.074 0.295 0.245 The kva rating of a threephase grounding transformer.098 0.091 those suggested in AIEE Standard for “Xeutral Grounding 5 minutes 0.005 0.240 0.110 0.235 0.040 0. transformer terminals.Power Transformers and Reactors ciency become less apparent as the transformation ratio increases. normally are designed so that rated neutral current flows depending upon the form and details of construction. that the heat from load lossev is nil stored in the tram+ former.220 0. the ohmic leakage impedance between one primary (star) winding and the corresponding secondary rated kva and rated voltage.187 0.139 0.tkq.275 or of a grounding bank. time ratings for grounding service are open to question Grounding transformers. This is equivalent to loopercent zerosequence voltage impressed at the transformer terminals resulting in sequence currents in each phase of a solidlygrounded starthe circulation of rated neutral current.340 0. and that the final temperature will not ruceed VZLIUBYpermitted by standservice.242 0.7 0. A oneminute time rating is often used 0. For A Single Phase Unit (One of three in a bank) neuhal voltage (kv) and the neutral orgrozbnd amperes that the transformer is designed to carry under fault conditions 1 minute 0. so that autotransformers for power purposes are usually used for low transformation ratios. classes of grounding transformers.033 0.057 0. and this reduced is likely.051 0. GROUNDING TRANSFORMERS Ka.085 0.03i 0. Zo%= 10 x kv2 Because a grounding transformer is a shorttime device.055 0. particularly the zigzag type. .253 0.196 0.380 0.174 or interconnectedstar (zigzag) arrangement as shown in 3 minutes 0.084 0. though other ratings such as 4 minutes 0.082 0. For A Three Phase Unit Table 12“K” FACTORS FOR DETERMINING EQUIVALENT TWOWINDING 55 C KVA OF GROUNDING TRANSFORMERS* StarDelta ‘.
‘t’hc usual range of regulation is plus 10 percent and minus and sequence of operations of the type USR tap changer.chapter 5 Power Transformers and Reactors t. In this position. The connections for XII.r adjust the reactive flow between branches of loop cir. The 32 step. 16. IO percent of the rated line voltage. chart of Fig. Tapchangingunderload equipment is ap. the relacrrits. ~rpon whether voltage or phase angle control is required. AUTOTRANSFORMER REVERSING .1 to 9 merely select the transformer tap to which the load nlicd to power transformers to maintain a constant second. low impedance to the load current permits operation on Various types of tapchanging equipment and circuits :uc used depending upon the voltage and kva and also this position to obtain voltages midway between the transformer taps. The sequence of switching is so coordinated by the tap changing mechanism that the transfer Since that time the development of more complicated t. Taps from the transformer winding connect to selector switches 1 through 9. The UNR Mechanism Fig. plus and type UNR tap changer also has a reversing switch which minus 10 percent. to control tlrc flow of reactive kva between two generating systems. changer. S. and in other cases to control the phase is thusarestricted to switches R.. s ’ I . and T.olvcr transformers.% = 10xkv2 31. respectively. 2!3Seventeen single position. ITntlcrloadtapchangers are built for 8. changer. tapchangingunderload equipment has reverses the connections to the tapped section of the such wide acceptance as to be considered standard for winding so that the same range and number of positions many types of transformers. \vit. All arcing l. the preventive autotransformer .ion is normally expressed in terms of shorttime kvs and _~ linetoline voltage: ZPSX cc Z. and 32 steps. while switches ar. The UT Mechanism Figure 28 illustrates schematically the operation of the type UT mechanism for changing taps under load. On all evensecondary voltage with a fixed primary voltage. singlephase. But the third.xI^ SELECTOR SEQUENCE OF OPERATION SEQUENCE OF OPERATION swlTcHEs I O= SWITCH CLOSED Fi& 28Seventeen position. while its relatively (r:msformer output voltages.glc relation.bridges two transformer taps. The selector switches are connected to load transfer switches R.h the trend in recent years being toward the larger 32. in popularity.’ . numbered positions. 29 shows schematically the diagram of connections n~rmber of steps so as to give a finer degree of regulation.is to be transferred. to control the preventive autotransformer is shortcircuited. opening . phase. TAP CHANGING UNDER LOAD the tap changer positions are shown on the sequence The modern load tap changer had its beginning in 1925.ransmission networks has made tap changing under load switches perform all the switching operations. Type UNR tap . the LLryvoltage with a variable primary voltage.norc and more essential to control the inphase voltage of before and closing after the selector switches. and to control the division of power between branches tively high reactance of the preventive autotransformer of loop circuits by shifting the phaseangle position of to circulating currents between adjacent taps prevents damage to the transformer winding. Type UT tap Fig. with plus and minus The operation of the selector and transfer switches is es‘ii percent and plus and minus 5 percent being second and actly as described for the type UT tap changer. SWITCHES T n I. and T. When the tap changer is on oddnumbered positions. S. 28. PREVENTIVE AUTO TRANSFORMER TRANSFER SWITCHES1 R+ \ (.
Should A generate 10 000 kilowatts in excess of its own load. EXCITING TRANSFORMER transformer for voltage control. created by the flow of load current I.IO is the terminal voltage at 9. (c) Introduction of an in phase voltage. BO is the terminal voltage at B. at which time there is no current through the reversing switches and therefore no arcing on them. 31 connected by a single transmission circuit. The URS tap changer. one for each phase. or twice the range can be obtained with the same number of taps. The reversing switch operates when the selector switches are on position 17. is often objectionable. The flow of Dower from 4 to B. 32Regulating I I . B if there is to be no change in system frequency. The transmission of Dolyer from A to B results in a difference in marrnitude between terminal voltages and also a shift in phasl angle. Type URS tap kilowatts must go over the tie line to B. and each selector switch serves as a t. but the drop in voltage. and 0 is the phaseangle difference between terminal voltages. REVERSING SWITCHsm (a) A 8: 9 I YP B / 0 interchange between systems: Fig. Thl flow of reactive power over the interconnecting line is determined bv the terminal voltages held bv the machine excitations at A and B. introduca I SERIES TRANSFORMER I a’ Physically. AB’. can be equipped for hand operation. BC. or for full automatic operation under the control of relays. 31 (b). The URS Mechanism The type URS load tap changer is applied to small power transformers and large distribution transformers. Excessive voltage drop between the systems can be readily corrected by transformer taps of a fixed nature or by tapchanging equipment. . REGULATING TRANSFORMERS FOR VOLTAGE AND PHASE ANGLE CONTROL Consider two systems A and B in Fig. I SEPUENCE OF OPERATION I O=SWITCH CLOSED Fig. 30Thirtythree position. as they rotate about a center shaft. like the other load tap changers.~~ b b’ I Fig. In actual practice the phase angle is not always apparent. . . The schematic circuit diagram and operations sequence chart is shown in Fig. the 10 000 ~I ~~ ~. The moving s&lector switch contacts. both select the taps and make contact with them. singlephase. the stationary selector switch contacts are arranged in circles. An attempt to maintain satisfactory terminal voltages at A and B will often result in undesirable circulation of reactive kva between the svstems. or one of them may be a generating unit and the other a load. 33. 30. remote manual operation. to correct for excessive voltage drop. there can be but one result. changer. or vice versa. An increase in generator outDut bv A must be accomDanied bv a corresponding decreise in”output (increase in input) b. as illustrated in Fig. is ditermined by the governor settings. XIII.rnnsfer switch for the tap to which it is connected. A and B may both be generating units.4B is the vectorial voltage drop from A to B. The transfer switches are eliminated. The reversing switch is a closebeforeopen switch which operates at the time there is no voltage across its contacts.122 Power Transformers and Reactors Chapter 5 can be obtained with onehalf the number of tap sections. dlPower (a) Two systems with tic. (h) Vector diagram of voltages during interchange of power.
a \viII not entirely control the flow of power over any one OI t. a a’ B A 4 bB0 (4 c Fig. with no tract is in quadrature. or vice versa (see there are cases where both voltage and phase angle control Fig. For the case of phaseangle control. 33 (c)). The preceding statements follow from the fact that transmissioncircuit impedances are predominantly reactive. BC.hc interconnecting lines. and adjustment of governors 90 degrees. An attempt to adjust load on simple voltage control equipment can be used. and a part of the power going from rl to B over the lines AC and CB (seeFig. This simply I 110 between two systems results in a change of load on adds or subtracts a voltage in phase with the system volttic I. and from C to A form a closed loop. the equipment can be identical except the voltage selected to add or sub.v an exciting autotransformer xvith automatic tap changing equipment indicated by the arrons. As the earlier discussion showed. There now exists between B and C a difference SERIES in voltage and a difference in phase angle.tntl with power transmitted from A to B.\\oen the various lines is determined solely by the relative ilnPcclances of the interconnecting lines. The flow of reactive kva can be controlled by regulators for voltage control. With the tie line from B to C open. The regulator for phaseangle control introFig.ircly new element enters. from B to C. as in Fig. In general the distribution of real power flow over the various interconnections found in loop circuits can be controlled by regulators for phaseangle control. 33 (a). If the tie line 1)ctwecn B and C is closed under these conditions there is a a rctlistribution of power flow between A and B. This differs from the usual starand voltage control. duces a quadrature series voltage in the loop resulting in the flow of currents lagging the impressed voltage by nearly ctnt. 33 (b)). a part going 1 wcr the line from A to B. any place in the loop. to compensate for the voltage drop and bring the terminal voltage at B to a desired value.IIOother two tie lines. Figure 34 is a schematic diagram of a typical regulating transformer for phase angle control. The voltage regulator introduces a series inphase 0 voltage into the loop.chapter 5 Power Transformers and Reactors 123 ing an inphase voltage. An delta power transformation in that the angle of phase shift of current and voltage is not fixed but depends on the tap position. or a regulating transformer for vo1t:lge control were inserted in the loop. both kw and reactive kva beI. Conditions similar to that just described occur on inter“onnccted systems involving loop circuits. and quadrature current (reactive (b) (cl 4 kva) is circulated around the loop since the impedances are reactive. power being transmitted from A to C. ‘UK distribution of power. age. 33Power interchange with three interconnected systems. Figure 32 is a simplified sketch of a regulating transformer for voltage control. *TtB . 34Regulating transformer for phaseangle control. by the use of a regulating transformer Fig. Consider three systems interconnected with each other 8o that the interconnections from A to B. the terminal volttlges of A and C will be equal and in phase. or the circulation of inphase currents (kw). it would be Possible to make the voltage at C equal in magnitude to t1l:it at B but it would not have the same phase relationship. 1f at the time of closing BC an adjustment of transformer taps were made. For the case of correcting the voltage for line drop. usin. There would still be a flow of power from A to C ml from C to B. To control the circu1ation of kw and prevent overloading certain lines it is often necessary to introduce a quadrature voltage. 35Regulating transformer for independent phaseangle for phaseangle control.
where a = tan‘&n For negativesequence neglecting regulator impedance: E: = NejaE2 = 2/l f3n2ejaE2 (63) 1 I~=Ej~12= (06) d&3n2ti. It should be noted that the equivalent circuits for phaseangle control regulators involve an ideal transformer providing a phase shift of voltage and current. as in F7. Where the voltage and phase angle bear a close relation.47 kv.124 Power Transformers and Reactors Chapter 5 ample. E. This phenomenon. and its magnitude depends upon the noload losses of the transformer. This exciting current consists of two components. neglecting regulator impedance: . plus or minus 10 percent. but is in the oppcsite direction.d+ (69) dt where. Fig. 35. are tabulated in the Appendix under Equivalent Circuits of Power and Regulating Transformers. Zerosequence currents and voltages do not undergo an angular shift in being transformed. The loss component is in phase with the impressed voltage. The magnetizing component lags the impressed voltage by 90 electrical degrees. When such calculations become involved. 34. The angular shift for negativesequence voltage and current is always equal to the angular shift for positivesequence. This may dictate the location in a loop. rated are required. Positivesequence voltage and current are always shifted by the same angle in the same direction. Several common connections used for regulating transformers providing voltage control. neglecting regulator impedance: Ed = E. or are transformed with a different transformation ratio than for positive. where completely independent control is desired. and its magnitude depends upon the number of turns in the primary winding. a small exciting current flows.. considering the impedances of the tie line and the load conditions in the loop. IO flows through the regulator as though it were a reactor. Magnetizing Component of Exciting Current If the secondary of the transformer is open. For positivesequence. 36Regulating transformer for voltage control.lr N For zerosequence.J. the use of the network calculator provides a quick and accurate tool for obtaining the solution. the economical location for the control equipment is at the point of lowest load to be transferred.or negativesequence quantities as in Gl. one of them being shown in Fig. (67) r:=I” ((33) For this regulator zerosequence voltage and current are not transformed. two mechanisms with two regulating windings and one series winding. e = instantaneous value of supply voltage i= instantaneous value of current R=effective resistance of the winding 4 = instantaneous flux threading primary winding 721=primary turns j I I 1 i I . The differential equation for the circuit consisting of the supply and. The equivalent circuits of the regulating transformers to positive.8 XIV. or combined voltage and phase angle control. 12. which is the regulating transformer for phaseangle control shown in Fig. If it is desired to close the loop. EXCITING AND INRUSH CURRENTS If normal voltage is impressed across the primary terminals of a transformer with its secondary opencircuited. 20 000 kva. or with one regulating winding and two series windings are necessary. the transformer can be treated as an ironcore reactor. the shape of the transformer saturation curve and the maximum flux density for which the transformer was designed. the transformer can be written as follows: e=Ri+nl. The voltage to be added or the phaseangle shift that must be obtained can be determined by calculation. negative. Clem. and the use of the sequence equivalent circuits for regulating transformers has been discussed in papers by Hobson and Lewis2. There are a number of combinations of conto accomplish this. and the flow of both real and reactive power over the various lines forming the loop must be controlled. However. and zerosequence are given. and by . one mechanism may suffice. the loss component and the magnetizing component. A brief discussion of each of these components follows: 34. unless when in tying several companies together the boundary between systems determines the location. refer to F7 in Table 7 of the Appendix. phase angle control. For exnections It happens with several connections of regulating transformers that zerosequence voltages and currents are not transformed at all.
Under this condition the magnetizing current is not a sine wave. VOLTAGE COMPONENT OF EXCITING CURRENT (0) TRANSFORMER 8H CURVE (b) VOLTAGE. 38 shows that although the flux is a sine wave the current is a distorted wave. += GE cos(wt+X)+& (72) wnl d/ZE In this solution. the third harmonic component is included in Fig. The second tctrrn. &. A study of Fig. for the purpose of discussion. (71) . 15 percent fifth. be neglected. 38. 38(b) are shown the impressed voltage and the flux wave lagging the voltage by 90 degrees. Under these conditions Eq. Under steadystate conditions this component is equal to zero. R.\hcrc. represents a transient component of flux the magnit. 38. 38Graphical method of determining magnetizing current. 38. . 37Impressed voltage and steadystate flux. the magnitude of & is discussed in Sec. 39 are shown the variations in the harmonic content of the exciting current for a particular grade of silicon steel. These percentages of harmonic currents will not change much with changes in transformer terminal voltage over the usual ranges in terminal voltage. this component is quite small in comparison to the magnetizing component and has little effect on the maximum value of the total current. If there were no appreciable saturation in the magnetic circuit in a transformer. three percent seventh. (70) &‘E sin(ot+X) =n$$ Solving the above differential equation.Chapter 5 Power Transformers and Reactors 125 xormally the resistance. (72) it can be seen that the normal steadystate flux is a sine wave and lags the sine wave supply voltage by 90 degrees. An analysis of this current wave shows that it contains oddharmonic components of appreciable magnitude. resulting in appreciable saturation. the economic design of a power transformer requires that the transformer iron be worked at the curved part of the saturation curve. ~ cos (wt+X) is the normal wnl stcladystate flux in the transformer core. and the exciting current. In a typical case the harmonics may be as follows: 45 percent third. e=d/ZE sin(ol+X). and smaller percentages of higher frequency. resulting in a sinusoidal magnetizing current wave in phase with the flus. The above components are expressed in percent of the equivalent sine wave value of the total exciting current. (69) can be rewritten: (70) If the supply voltage is a sine wave voltage. the normal maximum flux and the residual flux in the core at the time the transformer is IMPRESSED 5 STEADY STATE Fig. i.ude of which depends upon the instant at which the tmnsformer is energized. In Fig. and its shape depends upon the saturation characteristics (the BH curve) of the transformer magnetic circuit. E = rms value of supply voltage w=2nf Sllbstituting in Eq. IMPRESSED  energized. However. Consequently the Ri term in the above equation has little effect on the flux in the transformer and can. From Eq. are small. the magnetizing current and the flux would vary in direct proportion. The supply voltage and the normal flux are plotted in Fig. The shape of the current wave can be determined graphically as shown in Fig. FLUX AND CURRENT WAVES Fig. However. In Fig. For any flux the corresponding value of current can be found from the BH curve. The current found in this manner does not consist of magnetizing current alone but includes a loss component required to furnish the hysteresis loss of the core. Following this procedure the entire current wave can be plotted. 37 as a function of time.
3 3.. In practice the iron losses are determined from laboratory tests on samples of transformer steel. Total Exciting Current As discussed above. In Table 13 are given typical exciting currents for power transformers. 38. 39Harmonic content of exciting current grade of silicon steel.97. watts per lb We= K. The iron loss in a GOcycle power transformer of modern design is approximately one watt per pound. and x are factors that depend upon the quality of the steel used in the core.3 3. = maximum flux density (73) inal voltage. 40Exciting curve applies for shape of the curve grade of current vs. Iron loss = Whf W. and the copper loss caused by the exciting current.2 2.4 3. hence any increase in terminal voltage above normal mill greatly increase the exciting current.5 Kv 15 Kv ~~3. IO&percent terminal voltage results in 200percent exciting current. the total exciting current of a transformer includes a magnetizing and a loss component.e. The exciting currents vary directly with the voltage rating and inversely with the kva rating. the residual flux in the core.1 4. Threephase KVA 500 1 000 Kh.6 3. Usually only the iron losses. maximum flux density. Wh = KJBx. voltage class. However. The ratio of hysteresis loss to eddy current loss will be on the order of 3.fzt2Bzm. Steinmetz. Inrush Current 4 When a transformer is first energized. Test values will as a rule come below these values but a plus or minus variation must be expected depending upon purchaser’s requirements. are important.1 3. and the characteristics of the magnetic circuit.7% 3. 161 Kv 230Kv 2500 5000 10 000 25000 50 000 2. 40 the exciting current of a typical transformer is given as a function of the voltage applied to its terminals. In Fig. The above one particular design of transformer: the may vary considerably depending upon the steel and the transformer design. TABLE 13 TYPICAL EXCITING CURRENTVALUES FOR SINGLEPHASE POWER TRANSFORMERS (In percent of full load current) The following values should be considered as very approximate for average standard designs and are predicated on prevailing performance characteristics.6 3. 35. the matter should be referred to the proper manufacturer’s design engineers. Typical Magnitudes of Exciting Current The actual magnitudes of exciting currents vary over fairly wide ranges depending upon transformer size. 25 Kv 69 Kv 1138Kv 3. terminal voltage. a transient exj citing current flows to bridge the gap between the ditions existing before the transformer is energized the conditions dictated by steadystate requiremen For any given transformer this transient current depe upon the magnitude of the supply voltage at the instant the transformer is energized. 2. for a particular Fig. Should closer estimating data be required.1 3.1 .Power Transformers and Reactors Chapter 5 92 go 70 PERCENT OF EQUIVALENT SINE WAVE 80 90 100 20 60 100 PERCENT 220 260 MO 160 OF NORMAL EXCITING CURRENT 300 400 Fig.9 3. 4.9 j 36.3 3. For example. Loss Component of Exciting Current The noload losses of a transformer are the iron losses. the formulas given below are useful in showing the qualitative effect of the various factors on loss. 3.1 2. In the original derivation of the hysteresis loss formula by Dr. The exciting current increases far more rapidly than the term *Reduced Insulation. The economic design of a transformer dictates working the iron at the curved part of the saturation curve at normal voltage...5 * 2.0....5’j’. hysteresis and eddy current losses.8 *. _. as they vary considerably with transformer design. Voltage Class (Full Insulation) 2. K. For modern steels 5 may have a value as high as 3. z was 1%.8 3.0$&o* 3. .4 .0 2. These figures should be used as a rough guide only.2 4.876 3.770 3. etc. i..0 with silicon steel and K with oriented steel. a small dielectric loss.1 $‘J 4. These losses depend upon frequency.8 3. watts per lb Wh = hysteresis loss We = eddy current loss j= frequency t = thickness of laminations Bm._. 37.
Assume that a transformer having zero residual flux is energized when the supply voltage is at a positive maximum. h is equal to 0. In this figure the transient flux has been assumed to have no decrement. Determination of Current Inrush After the flux variation has been determined by the method described. ln studying the phenomena that occur when a transformer is energized it is more satisfactory to determine the plus in the magnetic circuit first and then derive the curr(. (72) can be rewritten. X is equal to 90 degrees.nt from the flux. assumed for illustration only. Often the magnitude of this transient current exceeds fullload current and may reach 8 to 10 times fullload current. In the above discussion loss has been neglected in order to simplify the problem. if loss is considered the transient flux decreases with time and the %st value of the total flux is less than shown. (76) is plotted in pig. cos x= $* +o=o ho = dh Substituting in Eq. If the transformer is energized at zero voltage. the current reaches a value equal to many times the maximum value of the normal transformer exciting current. as shown in Eq. TRANSIENT FLUX mTOTAL FLux b 111the above equation the angle X depends upon the instantaneous value of the supply voltage at the instant the transformer is energized. 41a. This procedure is preferable because the flux does not depart much from a sine wave even though the current wave is usually distorted. Flux waves for any other initial conditions can be calculated in a similar manner using Eq. 41Transformer flux during transient conditions. . This high value of current is reached because of the high degree of saturation of the transformer magnetic circuit. Loss is important in an actual transformer because it decreases the maximum inrush current and reduces the exciting current to normal after a . 42. Fig. . For any flux the corresponding current can be obtained from the transformer BH curve. = transformer residual flux \v here 4m cos X=steadystate flux at 1=0 . The total flux in a transformer core is equal to the normal steadystate flux plus a transient component of flux. For these conditions 4. (75) VOLTAGE (Cl) PRIMARY CLOSED AT ZERO VOLTAGEZERO RESIDUAL FLUX. 39. . This relation can be used to determine the transient flux in the core of a transformer immediately 1/ZE represents the sftcr the transformer is energized. Although the maximum flux is only twice its normal value. 72.chapter 5 Power Transformers and Reactors 127 and the impedance of the supply circuit. (74). the current wave can be obtained graphically as shown in Fig. Eq. Sixty percent residual flux has been c$= dm CO6(ot)f$h / STEADYSTATE VOLTAGE (b) PRIMARY CLOSED AT ZERO VOLTAGE60% POSITIVE RESIDUAL FLUX. As ~ wnl (rest of the normal steadystate flux. These high inrush currents are important principally because of their effect on the operation of relays used for differential protection of transformers. zero supply voltage the following conditions exist: x=0 4. and cos X are both equal to zero so +to is also cc111sl zero. In Fig. whereas if the transformer is energized where the supply voltage is at a positivc maximum value. if a transformer having zero residual is energized :rt. 1lowever. The transformer flux therefore starts out to ~~ntlcrnormal conditions and there would be no transient. dto= rnitial transrent flux. In the case illustrated it was assumed that a transformer having zero residual flux was energized at zero supply voltage. (74) 4= An cos bt+X)+$bt . The total flux wave consists of a sinusoidal flux \vave plus a dc flux wave and reaches a crest equal to twice the normal maximum flux. 40= dh cos X+&o +. (74) (76) The flux wave represented by Eq.\t t=o. 41 (b) similar waves have been plotted for a transformer having 60 percent positive residual flux and energized at zero supply voltage. the flux therefore is equal to twice normal crest flux.
g The transient has a rapid decrement during the first few cycles and decays more slowly thereafter. The crest inrush currents are expressed in per unit inrush currents to singlephase. THIRDHARMONIC COMPONENT EXCITING CURRENT 41. 43Current inrush for a particular gized at zero voltage. The losses that are effective are the resistance loss of the supply circuit and the resistance and stray losses in the transformer.04 of crest fullload 2000 10 000 20 000 Note: current. During the first few current peaks. and hence the damping factor becomes smaller as the current decays. Estimating Inrush Currents The calculation of the inrush current to a power transformer requires considerable detailed transformer design information not readily available to the application engineer. the inrush current may be estimated from the relation ~i!iFig. Energizing a coreform transformer from the lowvoltage side may result in inrush currrents approaching twice the values in the table. Suppression of the ThirdHarmonic Component OF AS discussed in connection with Fig. R/L. the exciting current of a transformer contains appreciable harmonic . The per unit inrush current to a shellform transformer is approximately the same on the high. An order of magnitude of XV. The inrush currents in Table 14 are based on energizing a transformer from a zeroreactance source. The damping coefficient. X=Effective supply reactance in per unit on the transformer kva base. The inductance of the transformer increases as the saturation TABLE 14APPROXIMATE INRUSH CURRENTS TO 60CYCLE POWER TRANSFORMERS ENERGIZED FROM THE HIGHVOLTAGE SIDE Transformer Rating Kva Core Shell Form 58 47 Form 2. decreases. When it is desired to give some weight to source reactance. transformer ener 1fIoX (77) where 10= Inrush current neglecting supply reactance in per unit of rated transformer current. for this circuit is not constant because of the variation of the transformer inductance with saturation. For this reason reference should be made to the manufacturer in those few cases where a reasonably accurate estimate is required. making L low. 60cycle transformers can be obtained from the data in Table 14. 42Graphical method of determining inrush current. Figure 43 is an oscillogram of a typical excitingcurrent inrush for a singlephase transformer energized at the zero point on the supply voltage wave. 39. period of time.and lowvoltage sides.128 Power Transformers and Reactors Chapter 5 lNRUSH CURRENT I COA4 CURRENT la) TRANSFORMER SH CURVE tb) FLUX WAVE AND INRUSH CURRENT WAVE Fig. the degree of saturation of the iron is high. The values given are based on the transformer being energized from the highvoltage side at the instant the supply voltage passes through zero. 40.55 2.
. 5.TO GROUND. Effect of Transformer Connections The application of the above principles will be illustrated by consideration of a number of typical connections. The magnitude of this current will depend upon the impedance of the ground sources relative to the delta circuit impedance and is usually too small to cause trouble from telephone interlerence. The delta connection will furnish :L path for the flow of thirdharmonic currents and will minimize the thirdharmonic current in the external cir(Gts.5 1. This factor will help decrease the magnitude of the thirdharmonic voltages but may cause interference in telephone lines paralleling the secondary power circuits. 1 A 1 A COMMENTS II ) SEE NOTE I A UNGROUNDED (SMALL CAPACITANCE . The third harmonic is by far the largest harmonic component. usually not more than five percent. . 3. Some thirdharmonic current can flow in the secondary windings if other ground ~ouroes are present on the secondary side of the transformer bank. nlt. I. consequently the thirdharmonic flux in the transformer magnetic circuit is small even if the thirdharmonic component of the exciting current is suppressed. 1 SEC. interference due to possible coupling with parallel telephone circuits. If this component cannot flow. The paths permitting the flow of thirdharmonic currents are determined hy the system and transformer zerosequence circuits. under some conditions. Resonance with the secondary capacitance may produce high harmonic voltages. the esciting current must contain a thirdharmonic component. as affected by transformer given in Chapter 23. In a threephase coreform transformer the reluctance of the thirdharmonic flux path is high (see Sec. 4.hough a thirdharmonic voltage may be present in the linttoneutral voltages. there can be no thirdharmonic component in the linetoline voltage.s in the external circuits may. A discussion of telephone POWER SOURCE TRANSFORMER TRANSMISSION LINE 42. I &. Thus. In a threephase system. It has been shown that third harmonics must occur in either the exciting current or the voltage of a transformer. It is always preferable to have at least one deltaconnected winding in a threephase transformer bank. Very little thirdharmonic current will flow in the line and very little thirdharmonic will be present in the system voltage. 44 is shown a threephase transformer bank connected TABLETSINFLUENCEOFTRANSFORMER CONNECTIONSON THIRDHARMONICVOLTAGESANLJ CURRENTS I SOURCE 1 TRANSFORMER 1 CONNECTION PRIM. The thirdharmonic component of the exciting current is suppressed and so a thirdharmonic component will be present in the transformer linetoground voltages. because of transformer or system connections. The thirdharmonic flux will. The deltaconnected winding furnishes a path for the thirdharmonic exciting currents required to eliminate the thirdharmonic voltages. The deltaconnected winding furnishes B path for the thirdharmonic exciting currents required to eliminate the thirdharmonic voltages. 2. In Fig. 11. is 5 current. The same comments would apply if other ground sources are connected to the secondary oircuit. This is very desirable because thirdharmonic currcnt. The magnitude of the thirdharmonic voltage induced in a transformer winding. when the thirdharmonic current is suppressed. 4 n I II ” 01 (I ‘1 II I. will vary between 5 and 50 percent depending upon the type of transformers used. appreciable thirdharmonic current can fiow in the secondary windings. being as high as 40 to 50 percent of the equivalent sinewave exciting current. in turn. The thirdharmonic component of the exciting current flows over the line and may cause. If the capacitancetoground of the circuit connected to the transformer secondary is large. cause telephone interference.5 3 GROUNDED (GROUNDED GROUNDED BANKS OR TANCE TO GENERATORS OR TRANSFORMER LARGE CAPACIGROUND) . Note: 1. The exciting current will take the shape imposed by the particular connections used. If the flux in a transformer magnetic circuit is sinusoidal. No thirdharmonic current will flow in the line between the source and the transformer and very little thirdharmonic will be present in the system voltage. the thirdharmonic voltages will constitute a zerosequence set of voltages of triple frequency. with singlephase transformers or with threephase shellform transformers the thirdharmonic voltages may be as high as 50 percent of the fundamentalfrequency voltage. The thirdharmonic voltage induced is therefore small. 129 connections. (b) O* 7 Fig. the flux will contain a thirdllarmonic component. NO GROUNDED GENERATORS TRANSFORMER OR GROUNDED BANKS) I & /I. the thirdharmonic currents of each phase are in phase with each other and hence constitute a zerosequence set of currents of triple frequency. Sec. 4440 (d) 7 nnections which influence the flow of thirdharmonic exciting current. 56). 6.Chapter 5 Power Transformers and Reactors interference.ikebvise. induce a thirdharmonic voltage in the transformer windings.
(2) Magnetic forces tending to pull jointed core members together. or between a conductor and a magnetic member. 45. with a load current In assumed. a little thirdharmonic current can flow over the line. reflection. TRANSFORMER NOISE where P. For satisfactory operation the circulating . Noise arising from any of the sources listed above may be amplified by mechanical resonance in the tank or fittings. SinglePhase Transformers Transformers having different kva ratings may operate : in parallel. PARALLEL OPERATION OF TRANSFORMERS : 43. if the line is long and is closely coupled with telephone circuits. It is also necessary that the ratio of resistance to reactance in all transformers be equal. If the starstar connection in Fig. no thirdharmonic current will flow in the line. It is quite difficult to predetermine a sound level nhick will prove satisfactory in the surroundings where a new transformer is to be installed. it is necessary that the transformers be wound with the same turns ratio.002 dynes per square centimeter: 00 Ea 1 Fig. The same general comments apply when threewinding transformers are used. (3) Magnetic forces acting between two conductors. If the transformer bank is close to the power source no telephone interference should result from the use of this connection. If the primary is ungrounded or the generator is ungrounded. The most persistent of these sources of noise is magnetostriction. Either percent impedances or ohmic 2 I I:N c Transformer noise is a problem because of its disturbing effect upon people. is expressed in dynes per square centimeter. Transformers designed to have sound levels below standard levels are available. and careful design is necessary to avoid such reinforcement of the original sound. 44(b).130 Power Transformers and Reactors db = 20 logi. If the impedance of this path is low. 45Equivalent circuit for parallel connection phase twowinding transformers. which depends upon flux density and cannot be eliminated by tight core construction. With the primary neutral and the generator neutral grounded. the line in turn being connected to a power source. If the primarv is grounded and the generator is also grounded. XVI. as in Fig. in terms of decibels referred to 0. pumps. With this connection the magnitude of the thirdharmonic current in the line depends upon the relative impedances of the supply circuit and the delta circuit. the delta connection furnishes a path for the thirdharmonic currents required to eliminate the thirdharmonic voltages. this method is perhaps more appropriate when the circuit involves unequal turn ratios. This current is usually too small to cause any troublesome interference. Standard@ have been established for permissible sound pressure levels for various types of transformers. and resonance to a great degree. telephone interference may result. The division of current between transformers having unequal turns ratios and unequal percent impedances may be calculated from an equivalent circuit similar to the one shown in Fig. If one winding is delta connected. The only means of reducing magnetostrictive force now at hand is to reduce flux density in the core. because the use of percent values in this type of circuit involves extra complications. of single impedances may be used in an equivalent circuit for paralleled transformers. 45 contains ohmic impedances and actual turns ratios. The circuit in Fig. but at extra cost because the magnetic material is worked at an induction below normal. will indicate the division of current between transformers. Solution of this circuit. However. when each per. though most power transformers will likely be similar enough in this respect to permit calculations based on only the impedance magnitude.&& Chapter 5 ’ (78) to a transmission line. In Table 15 is given a summary of a number of typical transformer connections with a brief description of the effect of the connections on the thirdharmonic currents and voltages. Also. little thirdharmonic voltage will be present on the system. solution of this circuit with total load current set equal to zero will indicate the circulating current caused by unequal transformer ratios. To achieve : accurate load division. including (1) Magnetostriction. 44(c) and (d). Local conditions affect sound transmission. and these factors are hard to evaluate prior to transformer installation. or other transformer auxiliaries. centage is expressed on the kva base of its respective transformer. the small change in dimensions of ferromagnetic materials caused by induction. XVII. the sound pressure. 44(a) is used the thirdharmonic component of the exciting current is suppressed and a thirdharmonic component nill therefore be present in the linetoneutral voltages. with load division such that each transformer carries its proportionate share of the total load. (4) Fans.. little or no thirdharmonic current will flow in the supply circuit and little or no thirdharmonic voltage will be present on the system. When a deltaconnected winding is present in the transformer such as in Fig. Noise may arise from several sources of force induced vibrations. and that the percent impedance of all transformers be equal. a path is furnished for the thirdharmonic exciting currents.
46. aI:/N2 = r. The 10 oc= cl I 47. in this instance the transformers can be made to divide currents similarly at all loads. or higher.. twowinding transformers are given in Fig. so all these factors must be known before a solution is attempted. as well as winding ratios and impedances. ThreeWinding Transformers Currents flowing in the individual windings of parallel threewinding banks can be determined by solving an cbcluivalent circuit. selfcooled. Parallel operation of two such transformers is not usually satisfactory. If the insulation level of the lowvoltage winding is 15 kv. TRANSFORMER PRICES 47. 60cycle. etc. Transformers designed for star connection of the highvoltage winding may be built with a lower insulation level Fig. if the load currents I. particularly if this winding is of high voltage. Division of currents may be calculated from this circuit. 48. It is possible to design a threewinding transformer so that the load taken from the tertiary winding does not seriously affect load division between t’he paralleled windings of the two transformers. If the impedance 2~ is made equal to zero. The estimating prices per kva are based on net prices as of December 1. 48 should be corrected in accordance with Table 16.Chapter 5 Power Transformers and Reactors 131 current for any combination of ratios and impedances prolxLbly should not exceed ten percent of the fullload rated current of the smaller unit. Price additions are also required when the rating of either the high. of single terminal loads. it is necessary either to set up a comto I)lcte threephase equivalent circuit. In addition it is ncc:essary to make sure that polarity and phaseshift bet. Fig. A singlephase equivalent circuit Inay bc set up on a linetoneutral basis to represent one $ase of a balanced threephase bank. If the impedances are proportioned to divide the load properly for one load condition. the prices in Fig. XVIII. Such a design may result in a value of . and the division of load may be calculated from an estended equivalent circuit similar to the one in Fig. When t. and this impedance ratio nill remain independent of tertiary loading. the load division between tmnsfo’rmers at some other loading is likely to be unsatisfactory. comparing banks made of singlephase and threephase units.hreephase transformer banks having any consitlcrnble degree of dissymmetry among the three phases :LI‘C be analyzed.. 46. 45. affect the division of currents among the windings of a threewinding transformer. oilimmersed. 47Equivalent transformer circuit for a singlephase threewinding paralleled with a twowinding unit.\veen highvoltage and lowvoltage terminals are similar for the parallel units. using the theory of Symmetrical components. An exception is the case wherein the a” circuit of Fig. however._ oa” 44. ThreePhase Transformer Banks The same considerations apply for the parallel operation of symmetrical threephase transformer banks as have been orltlined for singlephase transformers. then current division at the a’ terminals Ivill be determined by 2s and Z only. or to interconnect c>cllGvalent sequence networks in a manner to represent the ~~nhalancedportion of the circuit according to the rules of symmetrical components.or lowvoltage winding is 1000 volts and below. It is difficult to obtain zero as the value for Zp. since a change in tertiary load will alter the distribution of load between the other two windings. 1949. with no load connected to the tertiary. 46Equivalent circuit for parallel connection phase threewinding transformers. As prices change frequently. such as that shown in Fig. TwoWinding Type OA Transformers Estimating prices for Type OA. the curves should be used principally for comparing the prices of different voltage classes. ThreeWinding Transformer TwoWinding Transformer in Parallel With The equivalent circuit for a threewinding transformer Paralleled with a twowinding transformer is given in Fig. . 45. 31ore than two transformers may of course be paralleled.’ and 1:’ are assumed. 4T represents a delta tertiary winding in a threephase bank.ZT which is undesirable for other reasons. values near zero can be obtained with special design at increased cost.
0 5.0 5. the price of the unit can be estimated from the curves used at the transformer neutral.0 2.0 1. type OA power transformers.5 7. Table 17 summarizes the possible savings in cost with these designs.5 2._ 14001 to 27000 27001 . *Reference should be mn< 1 ye&on 16 for a discmsion of Ininirnurn permissible de m neutral insulation levels.5 1. 48 INSULATION LEVEL IS 15 KV OR HIGHER Price Addition in Percent i501 to IS00 Insulation ChSS KV Basic Impulse Levelskv SinglePhase Equivalent .5 5.0 3.0 1. given for twowinding transformers by using an equivalent TABLE 17PRICE REDUCTION FOR GROUNDED NEUTRAL SERVICE Winding Insulation Class at Line End 69 92 92 115 115 115 138 138 THREEPHASE BANK (a) SINGLEPHASE RATINGWA UNITS Insulation* Class at Seutral End 15 15 2569 15 2569 92 15 2546 6992 115 15 2546 6992 115138 15 25G 69115 138161 15 2569 92138 161 15 2569 42138 161196 Price Reduction Percent 0 3.0 7.0 3. 48Curve for estimating prices of oilimmersed.5 2. 1 4 s 15 21 28 34 41 %/2’% 7 10 OY* 1 4 8 0% 2 5 9 16 23 30 36 15 21 28 34 41   44 at the neutral end than at the line end of the winding.75 I 2 4 6 IO 20 40 60 100 TRANSFORMER RATINGWA (b) THREEPHASE UNITS Fig. 287 287 287 287 .0 9.5 5.0 3.5 10.nd hove 0:.55C kva SelfCooled Rating  I .0 138 138 161 161 161 161 196 196 196 196 230 230 230 230 I 100 287\KV GLASS 2’ II II II III II II IIJ 0.0 6. 60cycle.5 3. 3Phase Equivalent 55 C kva SelfCooled Rating 501 to 3600 3%X 7 10 14 21 29 37 3601 to 7000 lW5 4 7 11 18 26 34 42  3501 to i000 0% 3 6 10 17 24 34 42 32 39 46 ioot to 1 13501mx I 13500 0% 2 5 9 16 23 30 36 44 above 7001 to 14000 0% 3 6 10 17 24 32 39 46 .0 9. Reference should be made to section 16 for a discussion of the minimum insulation level that should be 48.0 7. MultiWinding Units If a multi<Finding transformer is designed for simultaneous operation of all windings at their rated capacities.0 5.5 4.0 12.132 Power Transformers and Reactors TABLE 16ADDITIONS WHEN LOWVOLTAGE : Winding WINDING Chapter 5 TO BE MADE TO PRICES IN FIG. twowinding.
910.960.670.610.78~0. ant1 10 percent for fivewinding trttnsformers.820. orcedOil Cooled wit.930.920.hc unit can he estimated from the curves given for twowinding transformers. B=$(Sum of the maximum rated capacities of the various windings).900.690.710. the kva rating listed for OA/FA/FOA transformers is the FOX value. 49Oilimmersed aircore reactor.850. 50.790.75~0.h Water Coolers.870.850.750. (~1. A reactor whose inductance increased with current magnitude would be most effective for limiting fault current.890. and to perform this function it is essential that magnetic saturation at high current does not reduce the coil reactance.880.720. for example.750.970.810.810.91 0.82 FOA 20 50 100 2 0.750.850.Chapter 5 Power Transformers and Reactors 133 twowinding capacity equal to the sum of the rated capacities of the various windings divided by two.900.850.840.4Oi IImmersed SelfCooled/ForcedAirCooled. 5 percent must be added for threewinding transformers. O~~+illmmerscd Water Cooled.i9~0.880.75 0.690.81 0.02. Insulation ClassIiL IVA"' 1 15 34.69t0.680.970. Tn addition.640.670.05’1.920.590. O.810.4) is equal to 0.~/FA/FA OTripleRated. .620.660.051. SelfCooled/F orcedAirCoolcd/ForcedOilCooled.8i0.850.80 0. 7.670. Estimating Prices for Other Types of Cooling l’able 18 is a summary of the approsimate cost of threepl~:w power transformers employing auxiliary cooling systams.650.850. an iron core reactor designed to have essentially constant magnetic permeability proves overly expensive.900.600.660.870.i50. using an equivalent twowinding c:lpacity equal to I<quivalent=d+Q(BA) (79) Where rl = $(Sum of the simultaneous loadings).620.081. therefore air core coils having constant inductance are generally used for currentlimiting applications. If a multiwinding transformer is not designed for simultaneous operation of all windings at their rated capacities. based on the cost of an OA transformer having :I rating equal to the maximum of the special unit being considllreePhase BLtIllC 1ating Currentlimiting reactors are inductance coils used to limit current during fault conditions.83 0.91 0.810.780.00 times the cost oi B 15 kv OA threephase unit rated 10 000 ha.830.910.930.65~0.0.87 0.730.081.880.640. If fault current is more than about three times rated full load current.06~1.81’0.770. Application XIX. where the Oh rating used to determine the base cost is equal to the highest rating of the forcecooled or speciallycooled unit.82 0.920.910.700.840.07 OA/FA/FO 20 50 100 0.930. All cost figures are expressed in per unit of 01 TABLE ~&RELATIVE COSTOF THREEPHASE TRANSFORMERS WITH SPECIAL COOLING IVAN cost is in per unit. Fig.i410.061.610. the price of t.05~1.5 69 192 /1151138 161~196~230 1.780.710.071.670.031.5 percent for fourwinding transformers. cb) ~~~‘~~L~4 ratings tabulated for OA/FA and OA/FA/FOA units we the FA nnd the FOA ratings respectively.850.69 I ’ FOW 20 50 100 (:L)0.710.i30.4/F. REACTORS of CurrentLimiting Reactors 49.880.680.810.071.740. FOAForce~lOilCooled with ForcedAirCoolers.620.82 0.90 0. The kva ratings listed in the second column of Table 15 are the highest ratings of forcedcooled units.030.600.8b0. but this characteristic has not been practically attained.650.72 0.820.710.Example: The cost of a 15 kv OA/FA threephase unit rated 10 000 kvn (F.770.051.840.990.75 1.770.750.730.991.710.730.860.i60.7~0.800.820.90 ow 5 10 20 50 100 0.710. transformer cost.890.
Sometimes this spacing can be reduced by use of bracing insulators between units or using stronger supporting insulators and increasing the strength of the floor. these reactors are limited to 34. Free circulation of air must be maintained to provide satisfactory heat transfer. Reinforcing rods less than threefourths inch in diameter which do not form a complete electrical circuit are not included in these limitations. Ohms reactance. Structures such as Ibeams. oilimmersed (Fig. In order to avoid excessive floor loading due to magnetic forces between reactors the spacing recommended by the manufacturer should be observed. Dry. For the same size members. 49) and drytype (Fig. Drytype reactors depend upon the surrounding air for insulation and cooling. should also be kept at a distance from the reactor even though they do not form closed circuits. Oilimmersed reactors can be cooled by any of the means commonly applied to power transformers. Xo magnetic field outside the tank to cause heating or magnetic forces in adjacent reactors or metal structures during shortcircuits. JODrytype aircore reactor. Fig. 3. This should always be checked with the manufacturer since bracing increases the natural period of vibration and may greatly increase the forces to be resisted by the building floors or walls. Because of the required clearances and construction details necessary to minimize corona. and manganese steel about onethird the rise for ordinary magnetic steel. Reactor Standards The standard insulation tests for currentlimiting reactors are summarized in Table 19. aluminum abollt one and one half times. plates. either exposed or hidden.or oilimmersed type. High thermal capacity. information should be included on the following: 1. Drytype reactors are usually cooled by natural ventilation but can also be designed with forcedair and heatexchanger auxiliaries where space is at a premium. A high factor of snfcty against flashover. and an end clearance of onehalf the outside diameter of the coil will produce a temperature rise less than 40 C in ordinary magnetic steel. Determination of Reactor Characteristics When specifying a currentlimiting reactor. 2. 51. A side clearance equal to onethird the outside diameter of the coil.Power Transformers and Reactors Chapter 5 Oilimmersed reactors can be applied to a circuit of any voltage level. and other metallic members. Singlephase or threephase reactor. 50). for either indoor or outdoor installation. 4. 2. . brass will have about the same rise. 3.5 kv as a masimum insulation class. These coils should not be surrounded with closed circuits of conducting material because the mutual inductance may be sufficient to produce destructive forces when shortcircuit current flows in the coil. The advantages of oilimmersed reactors also include: 1. Indoor or outdoor service. channels. because the insulation clearances from the reactor should be sufficient to avoid undue heating in such small metal parts. 52. Drytype and oilimmersed currentlimiting reactors are designed mechanically and thermally for not more than 33% times (3 percent reactive drop) normal fullload current for five seconds under shortcircuit conditions. Air core reactors are of two general types. Drytype currentlimiting reactors are built with Class B insulation and have an observable temperature rise by resistance of 80 C with normal continuous fullload current.
6 I 2 4 6 IO 20 5.0 3.0 138. or fraction thereof.0 3. Voltage class. nt a suitable frequency. 52Curve for estimating prices of singlephase. Reactors for use in 1201 to 13 800 volt circuits may be estimated from the curve l:lMed “15 kv and below. Estimating prices for singlephase.s 5.0 115. TABLE 20PRICE ADDITIONS FOR DRYTYPE RATED BELOW 300 AMPERES Current Rating Amperes 250299 200249 150199 125149 100124 75.1 0.0 8. Circuit characteristics: (a) Singlephase or threephase. Continuous current rating.5 46.Chapter 5 FOR CURRENTLIMITING Low Frequency Testscb’ Dry TYW (cl kv rms 12 25 30 40 60 85 115 Power Transformers and Reactors 135 TABLE19STANDARDDIELECTRIC TESTS REACTORS Impulse Tests (Oil Type) (‘) Insulation Class kv (2) (Ii1 Type kv rms  Chopped Wave  1.2 0. above 600 amperes. price additions must be made in accordance with Table 20.4 0. testshave been established for drytype current ’ 0.0 3. across the reactor terminals: drytype reactors for olltdoor service require a turntoturn test voltage onethird greater than the tabulated values.0 345 0 12 17 21 29 36 GO 80 105 160 210 260 310 3G5 425 485 590 690 ?v Time Iin. For current ratings below 300 amperes. oilimmersed currentlimiting reactors.0 3.0 92. 48 (a) for twowinding transformers. 51 for kva ratings between 10 and 5000.0 3.0 Full WaW? kv crest 45 60 i5 20 40 6 ‘0 100 200 400 600 1000 2000 40006000 REACTOR RATINGKVA Fig. 53. above 800 amperes. but wmistandard values are available and should be used \vhcre feasible in the preparation of reactor specifications.5 69 88 110 130 175 230 290 ‘ml 520 630 750 865 1035 1210 1500 li85 1.0 3.0 3. oilimmersed.99 50.0 3. 7.0 3. 60cycle. 8. 6. 60cycle. amperes.6 1. Reactor Prices The estimating prices included in this section should be used for comparative purposes only because reactor prices arc subject to change from time to time. 52 for insulation classes between 15 and 138 kv.0 3. Reactor rating in kva. 250 350 450 550 650 750 900 1050 1300 1550 .0 3. oilimmersed currentlimiting reactors are given in Fig.5 1. (d) Type of circuit conductors.0 2S7. drytype currentlimiting reactors.0 69. When the current rating exceeds 600 amperes make a price addi tion of one percent for each 100 amperes.0 34. drytype currentlimiting reactors are given in Fig.O l!)G 0 230. (c) ‘L No standard impulse limiting reactors.0 3. Standardization of current ratings and ohmic reactances for currentlimiting reactors is not yet completed. (h) Turntoturn tests are made by applying these lowfrequency test voltages.” The prices given apply to singlephase reactors with current ratings between 300 and 600 amperes. selfcooled shunt reactors may be estimated by adding 10 percent to the prices given in Fig. (c) Linetoline voltage. (b) Frequency.74 REACTORS Price Addition Percent 5 10 15 2’2 29 36 43 . SlCurve for estimating prices of singlephase.8 2. Voltqe t o Flnshkv crest cwer in fi:jl 54 1. or fraction thereof.0 161 . Estimating prices for BOcycle. REACTOR RATING MVA Fig. For current ratings above 800 amperes make a price addition of two percent for each 100 amperes. 60cycle. 13kKV CLASS1 I  (a) Intermediate voltage ratings are placed in the next higher insulation class unless specified otherwise. 60cycle.2 2.66 I5 0 23. Estimating prices for singlephase.
and T windings are nl. ( = 3 ZPS+ZPTZST NT ) (80) $&T%+zP. S. EQUIVALENT CIRCUITS FOR SINGLE PHASE TRANSFORMERS Representation of a transformer by an equivalent circuit is a commonly used method for determining its performance as a circuit element in complex power and distribution networks. Equivalent former Circuits for ThreeWinding Trans I/+. respectively. The following sections discuss the equivalent circuits for threewinding and fourwinding transformers. in general.136 Power Transformers and Reactors Chapter 5 XX. (d). and n3. n2. The impedances which can be most readily determined by test or by calculation are those between transformer windings taken two at a time.% = f ( ZPT+ZSTZPS N: ZPT%+$zYT% Zps%) The quantities can be expressed in percent on any arbitrary kva base. S. The number of independent impedances required in an equivalent circuit to represent a multiwinding transformer shall be.zPT% > ) 1 zsT+zPsZPT N2. The number of turns in the P. 53 (c) with all impedances referred to the kva of the P winding. Thus.x 1 + Zpr% h%] z&=112[~Zs~%+Zpsr ZTY = 54. 2s zT= a ( 1 Z. three branch impedances for a threewinding transformer. 10. Without the simplifications offered by the use of such equivalent circuits the handling of transformers with their complex array of leakage and mutual impedances would be a formidable problem.. four. by multiplying each impedance by . therefore the impedances in an equivalent circuit can well be expressed in terms of these actual impedances between the transformer windings taken two at a time. one equivalent impedance is required to represent a twowinding transformer.g. The equivalent circuit for a transformer having three windings on the same core is shown in Fig. and in general multiwinding transformers. 2(c).% +$zsa Zps%] transformer. equal to the number of all possible different combinations of the windings taken two at a time. 53 (b) with all impedance in ohms on the P winding voltage base and with ideal transformers included to preserve actual voltage and current relationships between the P. and six independent branch impedances to represent a fourwinding transformer. as the choice may be. 53Threewinding (a) winding diagram. For three. (b) equivalent circuit in ohms. For the purposes of calculating short circuit currents. The equivalent circuit is shown in Fig. Uo. and stability of a power system. and T windings. an equivalent network can be always determined that will consist only of simple impedances (mutual impedances eliminated) and accurately represent the transformer as a circuit element. will adequately represent a twowinding transformer for calculation purposes. or (e). the normal magnetizing current required by transformers is neglected. where the magnetizing branches have been omitted. The equivalent circuit expressed in percent is given in Fig. (c) equivalent circuit in percent. with other windings considered idle. Thus Figs. Equivalent circuits for the twowinding transformer and autotransformer are presented in sections 1 and 27. On the P winding voltage base: 1 z. Note that Zp and 2s as defined and used here differ from Zp and 2s in Eq. 53. respectively. voltage regulation. 0 EP (a) a0 EP ] IP NI ‘y NZ’T “2 “3 zp=1/2 [ zps + ZP4ZST N: ] Zs=l/2 [+zsT+ZPSZPT] Zr=l/2 [ zpT+zsTzps NI: I \L/ Zp% = t/2 zp.=g Fig.
are derived from the relationships: %ps=ZpfZs ZPs%==P%+fZS% %pT=Zp+ZT (82) ZPT%=ZP%++ZT% Fig. Zrr=leakage impedance between the P and T windings as measured in ohms on the P winding with the T winding shortcircuited and the 5’ winding opencircuited. 033) Z. z.Zp70. and RPS is the total cffcctive resistance between the P and S windings.= $ ( ZPS+$$~TZPTK 1 $2Z~~++2Z~~$Z~~ \vhorc XPs is the leakage reactance between the P and S Cndings (with T opencircuited). etc.. (b) equivalent circuit. ~~~7~ leakage impedance between the S and T wind= ings. expressed in percent on the kva and voltage of the P winding.+K~ where. expressed in percent on the kva and voltage of the S winding. here again the magnetizing branches are omitted. as measured in ohms on the P winding with S shortcircuited and T opencircuited. 54Fourwinding transformer. 1 2 1 K ) Z. expressed in percent on the kva and voltage of the P winding.ities expressed in percent on the kva and voltage of the I’ winding.ZTV N. (a) winding diagram. K&Q+=ZefZf ( 1 ZTV+ZPVZPTK Ni 034) KI = ZPTf 1 zsv Nf 1 ZPS . 53 (b) and Fig. UC Power Transformers and Reactors The notation used is defined as follows: Up = kva of the P winding. Us = kva of the S winding.Chapter 5 the ratio up. SK Equivalent former Circuits for FourWinding Trans K~=ZPT+$~ZSVZPV$ST 1 The equivalent circuit representing four windings on the same core..ive. This form’is due to Starr11. and Zr will be positive.=dKlK2+Kl zf =~K~K.llc three legs of the equivalent circuit to be zero or neg:ut. with the X minding opencircuited. with the T winding opencircuited. It is possible for one of t. 53 (c) for %r. for example. and losses are concerned (except exciting currents and no load losses). The branches of the equivalent circuit are related to the leakage impedances between pairs of windings as follows: IdSO Zp = R&XP Zps = RpsfjXps = Rp+Rs+j(Xp+Xs) Zps% = Rps%+. shown in Fig.12and The windings will ordinarily be taken in the order that makes K1 and Kz positive so that Z. 54 (b) luring ohmic quantities. Zsr = leakage impedance between the S and T windings as measured in ohms on the S winding with the T winding shortcircuited and the P winding opencircuited.iX~~% etc. mutual effects between windings. The equations given in Fig. 137 3 ZFs =leakage impedance between the P and S windings as measured in ohms on the P winding with the S winding shortcircuited and the T winding opencircuited. The leakage impedances are defined as before. The equivalent circuits completely represent the actual transformer as far as leakage impedances. UT = kva of the T winding.& = leakage impedance between the P and T windings.. ZF9~O=leakage impedance between the P and S windings. 54 (a) is given in Fig.=+ Z. Zrs is the leakage impedance between the P . with the P winding opencircuited. Rps% and Xrs$$ are the respective (lunnt.
etc. tending to limit the area of the flux return path to that between tank and windings. Although the exciting im. The equivalent circuit in percent has the same form as Fig. and it is also affected by 1 tank construction..%=3 ( Zps%fZ~%+sv%K%). where the script refers to the winding on which the exciting impedance is measured in ohms. u&s. tertiary winding for neutral stabilization and third harI monic excitation. The zerosequence exciting impedance is measured by connecting the three windings in parallel and applying a singlephase voltage to the paralleled windings. Similar equations. the exciting impedance to zerosequence currents in a threephase coreform unit will lie in the range from 30 to 300 percent. except in the insulating medium. (85) ‘. Figure 55 illustrates m therellsGo ~~~r~~~~~~zeroseouence exciting flux in such a unit.___. . the . derived from Eq.and negativesequence voltages and currents but in opposite directions.. will always be of the same magnitude for both positive. negative. threephase core>&& ‘~m~&~n~~y’&ver . The equivalent circuits were developed by Hobson and Lewis2’13.: . if the phase shift is +cu degrees for positivesequence. Following tl same notation. or in the tank and metallic connections other than’the core. . .1 ing impedance of a 4000kva.“’ unit of the coreform ded autotransformer of the threep characteristic.‘and the sequence equivalent circuits. ZPs. The flux linkages with the zerosequence exciting currents are therefore low. For example._ K~%=Z~T%+~SV%ZPSYD~ZTV%. Sequence Equivalent Circuits The impedance of threephase transformer banks to positive. if any is involved. under Equivalent Circuits for Power and Regulating Transformers. The impedance to negativesequence currents is always equal to the impedance to positive sequence currents.. and zerosequence currents. The zerosequence exciting impedance of threepdq coreform units is generally much lower than the positi% d 6 Fig. Z.e_conditionsYZXlected ln@e du~&%~the through impedances to. it was measured to be 36 i percent at normal voltage after the core and coils were placed in the tank. the higher values applying to the largest power transformers.. In this case the tank saturated but acted as a shortcircuited secondary winding around the transformer. threephase.._. The number of branches require to define an equivalent circuit of threephase two. apply for the other quantities in the equivalent circuit. In small coreform units this characteristic 1 is particularly effective and can be utilized to replace a. The same notation as defined in the early part of this chapter is used to denote leakage impedances in ohms and in percent. _“_ _ posrtlvesequence impedance. Low exciting impedancg ! under zerosequen. as if it had a tertiary winding of relatively : high reactance. Thus. with the star point grounded.edance to the flow of zerosequence currents in _. 66 OOO2400volt unit was I measured to be 84 percent at normal voltage before the! core was placed in the tank. and the equivalent circuits are similar except that the phase shift. the zerosequence impedance viewed from the starconnected terminals for shellform units. the zerosequence impedance viewed from the deltaconnected terminals is infinite. XXI. pedance to positivesequence currents may be several thousand percent. and S windings as measured in ohms on the P winding with the S winding shortcircuited and with the T and V windings opencircuited..than . zeroy&uence current flow. s T etc.. The zerosequence exciting impedance is affected by the 1 magnitude of excitation voltage.. . are given in the Appendix. Zs&o is the exciting impedance of the winding to zerosequence currents expressed in percent c the kva of the S winding.or mult winding transformers is the same in general as has been d . A starstar grounded.Is_ . the zerosequence excit.imp.138 Power Transformers and Reactors Chapter 5 ’ r ’ . and the exciting impedance to zerosequence currents correspondingly low.The exciting impedance to zeresequence has been denoted by Zss.. omitting the ideal transformers. etc. The impedance of a threephase bank of twowinding transformers to the flow of zerosequence currents is equal to the positivesequence impedance for threephase shellform units (or for a bank made up of three singlephase units) if the bank is starstar with both star points groundb IO C ed. SEQUENCE IMPEDANCE CHARACTERISTICS OF THREEPHASE TRANSFORMER BANKS 56... If the bank is connected stardelta. 54 (b). The. two&%&@ I’. is equal to the positivesequence impedance. the phase shift for negativesequence quantities will be a! degrees. 55Zerosequence exciting currents and fluxes in a threephase coreform transformer. (84). or banks of three singlephase units.
ln this case an extra impedance branch must be provided in the equivalent circuit. 56(b) is approximated by Fig.o represent such a singlephase transformer is shown in Fig. 56(f). ..and threewinding tr:msformers. 56(c) where ZM is considered infinite. (a) 57. These disadvantages are evidenced particularly when analyzing transformer circuits wherein several windings or phases are interconnected. 56(b) with an ideal transformer. in which the two parts of the leakage impedance. or no load.. 56(d) which combines the circuit of l$.. respectively. are shown scnhcmatically in Fig. and scrvcs to transform voltage and current without imped‘uN’~ drop or power loss. 56Steps in the derivation twowinding of the equivalent transformer.. The customary equivalent c*ircrlit used t.’ v()lQe base (by multiplying ZB by the square of the voltage ratio). ZA and ZB. mutual . j6(f) becomes Fig.. sequence _. Zw is the socalled “magnetizing shunt branch.Chapter 5 Power Transformers and Reactors 139 scribed for single phase transformers.__. the circuit of Fig. and having a value dependent . To overcome these deficiencies it is expedient to use the equivalent circuit shown in Fig. 56(b) in which ZA and 2~ are components of t hc transformer leakage impedance. 56(e) becomes Il‘ig.)f an ideal transformer. and the circuit terminals a and a’ are not insulated from each other as in the actual transformer. Ifthe threephase bank connections are unsymmctrical~s.networks.pon the zerosequence excitation impedances of the windirIgs as well as the grounding impedance in the transformer neut. the voltage and currrnt transformation effected by transformer action is not rrpresented in the equivalent circuit. This process may be thought of as “sliding the ideal transformer through” the impedance ZB. and Zn.. for most calculations Fig. 56(e) is obtained from Fig. A notable exception to this will exist in the formulation of the zerosequence impedance of core form transformers with grounded neutral. TWO magneticallycoupled windings of a singlephase transformer having n1 and nz turns.~~~hec~~seofthe opendelta connection. The circuit of Fig. 56(g).” Since the numerical value of ZM is very large compared to %.ral..the. where ZPS=ZA+ZB (b) zA+z. Either of these circuits has serious deficiencies as a device representing the actual transformer. the ideal transformer thus restores actual voltage and current relationships at the tminnls a and a’. 56(d) by converting the impedance ZB to the E. Derivation of Equivalent Circuits b In the derivation of equivalent circuits for threephase tr:msformers and banks made up of three singlephase t I:insformers. with a more or less :Irt)itrary division of the leakage impedance between 2~ :rnd Zg.oupling ~111exist b+rcen. Finally. The ideal transformer is defined as having infinite exciting impedance (zero exciting current) and zero leakage impedance. combine into the complete leakage impedance Zps. this branch being always shortCircuited to the neutral bus. If the cs(~iting.. current may be neglected (ZM considered as infinite) the circuit of Fig. This method may be used in the .__. it is convenient to represent each winding of tllc transformer by a leakage impedance and one winding .l~~vclopment of circuits for two. circuit of a @f3) .* (cl =PS (d) Fig. if Z M I‘s considered infinite. 56(a).
E& 711 E. applied to the terminals abc. ZPS .I circuited. Eb. 2~s is understood to mean the leakage impedance. . and a threephase shortcircuit at the a’b’c’ terminals.= EL. The above relations show that the linetoground volt ages on the delta side lead the corresponding starsid voltages by 30 degrees.=e..SEOUENCE EOUIVALENT CIRCUIT I.. 57 each transformer winding is represented by an impedance and one winding of an ideal transformer. (c)NEGATIVE . = EL. ePand e. When the test is reversed.. respectively. by 30 degrees.@. 56(g) will be found most convenient. and E. nL = nJs I. the following relations can be written: EL= EL. Derivation Bank of Equivalent Circuit for StarDelta Designating the circuits connected to the abc and a’b’c’ terminals as circuits 4 and 5.(1a) =fi?!!E&~i~. . although in some cases it becomes desirable to have part of the leakage impedance associated with each winding. = IxZ. zIs=Ey=zp+ a ?! 0 722 2zs+s In Fig. which must be considered in complete positivesequence equivalent circuit for the tram former. terminals and the a’b’c’ terminals open circuited. e. (88) will show that th currentsI:. (b) POSITIVESEOUENGE EOUIVALENT CIRCUIT (91 where E: and El are the positivesequence voltages t ground at the transformer terminals. The voltages (89) .j30. (d) ZEROSEQUENCE EPUIVALENT CIRCUIT Fig. A consideration of Eqs. and voltage applied to the P winding. N . I = tEi30s N I1 I[=. The windings shown in parallel are assumed to be on the same magnetic core. Ebg= a2Eag EC. the transformer having nl turns in the P winding and nz turns in the S winding. Ib an I. = aE& 4 (90) i “p N’‘15n. With positivesequence voltages applied to the abc . As positivesequence quantities were used in this analy sis. the final equation can be expressed as follows: E/ = NE1ci30. and the P winding short circuited. as measured in ohms.=0 e. 57Equivalent circuits of a stardelta transformer bank.14u rower ~‘ransJ0rmer.. n2 DIAGRAM I LettingN=z &hi’ E&=NE. Id and I. the impedance is denoted by 2s~.. represent the voltages across the P and S windings of the ideal transformers. ZPS is the imped.= aE.~%2~~Lj~. n2 ZPS i(r” 7:. 56(f) may be used. (a) SCHEMATlC J I Ice 1.=z(ELgaEL) I ! E& = a2E& EL. when 2~ may be considered infinite.. with voltage applied to the S winding..s Keaclors and mapter 5 1 In most developments the circuit of Fig.’ also lead the currents I. with the S winding short circuited. It is obvious from the development given that. and the circuit of Fig. To be perfectly definite. Assuming positivesequence voltages E. = 3. Z45 is defined as the impedance between circuits 4 and 5 in ohms on the circuit 4 voltage base. = n’I* n”2 (88) 58. 1% =2&.: ante between the P and S windings as measured by applying voltage to the P winding with the 5’ winding short.. = 3e.
58Equivalent circuits of a threewinding former. c. Eb:f The zerosequence circuit is derived by applying a set of 7 zerosequence voltages to the abc terminals.Zs 0 B.j30.+Zp]=IaZps o (a) SCHEMATIC DIAGRAM Z.CL. Referring to Fig. windings of the transformer in Fig. 57(b) therefore includes the impedance Zps and an ideal transformer having a turns ratio N and a 30degree phase shift.Chapter 5 Power Transformers and Reactors 141 The complete positivesequence circuit in Fig.5=2 n1 R2 cw \ d (d) ZEROSEQUENCE EQUIVALENT CIRCUIT Similar relations can be derived for the impedances Fig.ZS = 0 because no zerosequence voltage can be present between line terminals. The zerosequence impedance of the transformer bank is therefore infinite as viewed from the delta side. (96) If zerosequence voltages are applied to the a’b’c’ terminals. or common and series. a’ (b) POSITIVESEQUENCE EQUIVALENT CIRCUIT 59. made with negativesequence voltages and currents. with the third winding open circuited. (95) IE I Ix=3B n2 * ep=?!eR= ?! 21.and negativesequence circuits are there*b fore identical excepting for the direction of the phase shifts introduced by the stardelta transformation. and the T winding open circuited.+IZp IZ ep=. The impedance between the primary and secondary.=::[ (.. = . In this case Eag=Ebg=Ecg=Eo I*=Ib=I. would show that I2 I~=Ei30a N E> = NE. no current can flow because no return circuit is present. autotrans .. = ‘y = Zps. Derivation of Equivalent Circuit for Autotransformer with Delta Tertiary The basic impedances of an autotransformer with a delta tertiary may be defined in term’s of the leakage impedances between pairs of windings. 58(a) may be obtained by applying a voltage across the P winding with the S winding short circuited. 59.=Io Eng= ep +ZJ. which is the same impedance as 9 was obatained with positivesequence voltages and currents. e. 0 a A (c) NEGATIVESEQUENCE EOUIVALENT CIRCUIT ea= 1 IZS 0nl E=e. (93) (94) @ The positive. A similar analysis. ..I.
zp=+ Z..59Representation of the primary.+e. z45=y=(1~n1)* (lfn1)’ 2 [z.+nJ. With positivesequence voltages applied to terminals abc.‘I. Eliminating . from the above equation: The above equation is the impedance between circuits 5 and 6 in ohms on the circuit 5 voltage base.&=(lhh)ZPT+ Expressing ZP. 5 and 6 can be derived in terms of the impedances between windings. With positivesequence voltages applied to terminals a’b’c’. terminals a’b’c’ short circuited and terminals a”b”c” open circuited.ZpII. 1 These equations can be solved for the individual winding impedances ZP. between the P and T. 2s and 2~.‘ZT The impedances among circuits 4. (l+nl)* l+nl+nl(l+nl) ZPS 121 (l+n1)2 (100) . the following relations can be written: E&=E&=E&=O e. it is convenient to express the circuit 5 to circuit 6 impedance on the same base.+e.zpTn.Z56 ZPT ZST e.=+> . 0.(I. z 46 = PS ratio (l+n~) by (102) Fig.fZp. terminals abc open and terminals u”b”c” short circuited. =.1 66 2ZT+Zp+z (lfnl S.=e. . with terminals a’b’c’ open and u”b”c” shorted circuited. (99) : I:I. .142 Power Transformers and Reactors Eag=ep(la)Ia)ZP Chapter 5 =ep+IJp( 1A) =I XZs+nl”ZP n IL.) 2. Ia n2 [z2+2. c 2zT+zp+zs 1 K ) ) n2 n2 E&=e. = nlep z. In this case: et = IxZT Ix = Ih2 (103) ZPS=?*+ZP ZPT=$+ZP 2 ZST= *zT+zs. and S and T windings. resulting in the set of equations The impedance between circuits 4 and 6 may be ohtained by applying positivesequence voltages to terminals abc.Zpy =1.to secondarywinding impedance of an autotransformer.(l +nd = +pp+zs) 1 IaZ.I:(Zp+Zd z=. As 245 and Z46 are ohmic impedances on the circuit 4 base. 1 cc 2ZsT+n. 1+nl e. 712 . and I’.+z” 1 &2 (101) Representing the circuit transformation N’.2 [ (98) E..‘Zs=O.+I. using the same procedure as employed in the derivation of the impedances of the starTdelta bank in section 58. et = IxZ~ 1 (99) eDSee= lfnl et= ( !I!? ?l. Dividing by (1 +n.+IJP =I* E ZT 24s=~=+Zp=ZpT.=O e. Zs and ZT in terms of impedances between windings as given in Eq.).
=Ia’= et= IxZ~ zx Defining3 E. 2... (112) I. Zerosequence circuitThe zerosequence characteristics of the transformer can be obtained as follows: 1. The general procedure in writing the necessary equations is similar to that followed in the positivesequence analysis given above. 3. the zerosequence equivalent circuit is altered considerably as shown in Fig. ZJI~.= (l+nr) EaK= N’E. tf GO. %rJsand Zpr are in ohms on the P winding voltage base and Z:Sr is in ohms on the S winding voltage base. et= E. ZHO = ZHl ZMO = Z&l1 2z45 ZPT ZST= = z46 (N’1) $+&z. are in ohms on the circuit 4 (abc terminals) voltage base. eB=nlet= .756 = &ST (iv’ 1)zPS The positive. 58(b). between terminals abc and a’b’c’ without transformation. 58(d) are the same as the positivese’quence quantities. The phase shifts between circuit voltages can be det crmined by applying positivesequence voltages to terminals abc with the other two circuits open circuited. The zerosequence impedance is therefore determined by the leakage impedance between the S and T windings. In this case zerosequence current flows Fig.3 21. bnder these conditions. Eilg=ep E&=ep+es EL.J’ . ZL~. and that the terminal voltages are related as follows: Ez’= N’E.ej30 N”.. (111) (114) 1 ZLO = ZLl In the above equations ZH~. Apply zerosequence voltages to terminals abc with the delta closed and terminals a’b’c’ open circuited. 245 and ZM. Current in the S winding is balanced by circulating currents in the tertiary. and the zerosequence analysis in section 57..‘z* R2 0 n2 . N’ is (Mined as lfnr.. la2 E.z. that is.a2) et= n2ep= GE.iGej30.\i’zPT + 2 ZPS ) . which shows that the one ideal transformer has an N’ ratio but no phase shift. 60Zerosequence equivalent circuit of an ungrounded threewinding autotransformer. (108) (113) z + ZST pqy2ZpT 1 ZPS+ZPT (109) (110) z = N’1 46 N’ ( z46 = ZPT . Apply zerosequence voltages to terminals a’b’c’ with the delta closed and terminals abc open circuited. This permits evaluation of the zerosequence impedance between circuit 4 and circuit 5. The relations between the impedances in the equivalent circuit and the impedances between circuits can be expressed as follows: Zhll+ZH1=245 z&I+z. which is the ratio of linetoline or linetoneutral voltages between circuit 5 (a’b’c’ terminals) ml circuit 4 (abc terminals).. with the a’b’c’ terminals connected to ground and the tertiary closed.. Applying zero sequence voltages to the abc terminals.pl’( 1 .. E w I?= n2 If the neutral of the autotransformer is ungrounded. NegativeSequence CircuitA similar analysis made with negativesequence voltages would show that the impedances in the equivalent circuit are the same as in the positivesequence circuit. with no current flow in the P winding. =F3E. It will be found that the zerosequence impedances in the equivalent circuit shown in Fig.and negativesequence circuits are therefore identical excepting for the direction of the phase shift introduced by the stardelta transformation. = 246 (107) The second ideal transformer therefore has an N” turns ratio and a 30 degree phase shift. Z56 is in ohms on the circuit 5 (a’b’c’ terminals) voltage base. Apply zerosequence voltages to terminals abc with terminals a’b’c’ connected to ground and the delta opened. Ei’ = N”E.Chapter 5 Power Transformers and Reactors 143 The transformer can be represented by the positivesequence equivalent circuit in Fig. 60.Ebs)’= E.:’ = N”E.
gitzeering. 871873. 10.E. M. March 1933. A. 328342. E. John Wiley and Sons (1938). . Prentice. (117) The equivalent circuits can be based directly on percent quantities as shown in Table 7 of the Anoendix. p.E. $. Fundamental Concepts of Synchronous Machine Rcnrtsnces.ion So. Electricnl En. Simplified Computation of Voltage Regulation with Four \Vinding Transformers. by H. Lewis. Con 13. E. An Equivalent Circuit for the FourWinding Transformer.E. Putman and W. by 0.snclion. C. These impedances can be converted from one circuit base to another as follows: 246 5. Lewis.lournnl Preprint. Eleclric . l22 of Supplement.Z. pp. Z’ransaclions.E. 1937. Dnnn A. Vol. E = linetoline circuit voltage in kv. by R. Theory of Abnormal Line to Neutral Transformer Voltages. = Z.4. Transformer Engineering. 4. LaPierre. 1918. 6th Edition. American Standards Association. where 4. 8. Similar relations can be written for the other impedances involved.Z. il. REFERENCES 1. by C. and Reactors. Vol. 11.. McGrawHill Book Company. Putman. E. pp. Clem. by B. of circuit 4. 3. 36. London. pp. 874. Vol.s. Tmn. 58. R. Eq=linetoline voltage. Hobson and W.144 Power Transformers and Reactors Chapter 5 sidering the autotransformer with delta tertiary (case Dl in Table 7).% = impedance between circuits 4 and 5 in per cent on kva rating of circuit 4. September 1932. 1939. Hobson and W. Blume. 50. D.E. by Stigant. 1Inrch 1331. 1939. pp. 9. CT4 threephase kva rating of circuit 4. F. . and P.Z. 58. Equivalent Circuit Impedance of Regulating Transformers. Vol. 1935. W. (11G) %= impedance in ohms. ZcO = impedance in percent.E. 1939. by J. AS4 CSi. Electric CircuitsTheory and Applications. A. 56. M. whereas Z64 would be in ohms on the circuit 5 voltage base. It should be noted that the impedances. Loading Transformers by Copper Temperature. St.E. 7. Transactions. by H. Transformer Book. September 1948. These percent vnlucs can be converted to ohms by the familiar relation The resulting impedances will all be in percent on the circuit 1 kva base. 58. Vol. 48132. Inc. pp. Surge Proof Transformers. ?‘runsaclions. (a book). Equivalent Circuits for Power and Regulating Transformers. American Standards for Transformers. 584600. pp. by J. Dnhl (a book) Vol.E. V. 5X584 and discussion. V. 14. A. 504509. by F.Z.E. Zd5=impedance between circuits 4 and 3 in ohms on the circuit 4 voltage base. Starr.January 1939. Regulating Transformers in PowerSystem Analysis. 6. in kv.. Johnson and Phillips. 420. General Electtic Review. the ecluivalent circuit impedances can be obtained from the impedances between circuits as follows: Percent QuantitiesThe manufacturer normally espresses transformer impedances in percent on 8 kva base corresponding to the rated kva of the circuits involved. kva = :&phase kva rating of circuit. Vol. Transxtions. et al. For example 245 is in ohms on the circuit 4 voltage base. pp. p. . G. I. 34. Evans. are expressed in terms of the voltage or kva rating of the circuit or winding denoted by the first subscript. E. by L.Z. J. Regulators. 150152. p. as used in this chapter and in the Appendix.andsrds for Transformers NE&IA Publicat.E. 7. by J. = lOZ4s% u4 Ed3 . October 1939.Z. 12. Z’ransackm. Sew York. A 15 2. Using the nomenclature employed in the derivations.E.
motor. conditions. Because of the necessity of matching the speed of waterwheelgenerators to the requirements of the waterwheels it is difficult to standardize units of this type.\vledge of machine theory. This chapter treats of the charnctcristics of synchronous and induction machines in the light of the tdevc~lopment of the past twentyfive years. such as transient reactance and subtransient reactance to describe machine performance under t. It {vi11 consitler steatlystate and transient conditions for both salient pole :tn(.pole synchronous generator. waterwheel The two general types of synchronous machines are the cylindrical rotor machine or turbine generator wliic*li has an essentially uniform air gap and the salientpol(> gc’nerator. and condenser \vil..\ro components.rnnsicnt. I.rrnkcrs. 6.LSthe need for more accurate determination of short(*irc:rlit currents for the application of relays and circuit t.jsc: to the analysis of the transient characteristics of r. other in quadrature thereto.y and Sickle. STEADYSTATE CHARACTERISTICS SYNCHRONOUS MACHINES OF Fig.ll its concentrated field windings requires that their analysis follow a different line from that for machines such :IS induction motors.l cylindrical rotor machines under both txd:~llccttl ancl r~nt):Jnncetl conditions. 1 to 5 illustrate the outward appc~a~~nces and crosssectional views of typical modern macliincx Typical saturation curves for a tiydrogc~ncoolc~~l turbine generator. which have a uniform air gap and tlistributed windings. one in line with the axis of the poles and I tic. 145 . who introduced into t. Since that time the method has been estcncled by subsequent invest’igators. Figs.r:rnsic:ntsassociated with system stability was undertaken B Fig. The variable character of the air gap of the conventional s:Jictnt. There follows a discussion of the characteristics of induction motors uncler such transient c*on(lit ions as might contribute to the shortcircuit currc:nt ot' :L sy. Blonde1 originally attacked this problem by resolving the armature mmf’s and fluxes into t. lcutaway view of umbrellatype generator.llc. machine perform:Ln(:cwas largely judged in terms of the steadystate char:~&xistics.l:tcahines and was largely responsible for our present I. an(1 8 respectively. When the study of the t.3g notably Dohert. F.he industry several new const:tntS.CHAPTER 6 MACHINE ()riyitllll dulhi: CHARACTERISTICS C. The cmcrgence of the stability problem gave .?tem and might influence the choice of a circuit breaker. A further contributing urge \Y. a waterwheelgenerator and a synchronous condenser are shown in Figs. Wagner EFORE the growth of the public utilities into their present enormous proportions with large generating stations and connecting tie lines. this conception was quickly rc%cognizedas an invnl~~r~ble too11*2. IIowever. 7. aCutaway view of conventional waterwheel generator.
familiarity with this diagram is assumed. 85percent SCR.146 Machine Characteristics Chapter 6 Fig. including saturation. 60 000 kva. which leads el by 90 degrees. is used. The vectors e. 600 rpm. is required. rent. Upon adding the armature resistance drop. aSteam turbine generator installed at the Acme Station of the Toledo Edison Company. Depending upon the application. either the field current for rated voltage in the air gap or the actual field current for rated voltage. However. and armature leakage reactance drop. Table 1 of Chap. 1. respectively. 1 gives some of the specificationsZo for these machines. It. This flus represents the net flux in the air gap. care must be exercised in the case of generators to use the same reference value for field cur The vector diagram of Fig. Consistent with the policy of this book. 9 is the wellknown diagram of a cylindricalrotor machine. These find their greatest application in the electric utility industry.. 90 000 kw. xli’to et. the vector er is obtained. jphase. 13 800 volt. which represents the voltage developed by the airgap flux a. and i represent the terminal voltage to neutral and armature current. 85percent power factor. Unsaturated SteadyState Fig. Let it suffice merely to indicate the significance of the quantities. The concept of perunit quantity is valuable in comparing the characteristics of machines of different capacities and voltages. CylindricalRotor Conditions Machine Under great strides have been made with large 3600rpm condensing steam turbinegenerators. 60cycle. 5Hydrogencooled the system of the City 50 cyclet1 frequency changer set installed on of Los Angeles. . ri. 500 volts. The current It can be taken from the noload Fig. 60 cycles13 200 volts. To produce this flux a field current.
Chapter 6 Machine Characteristics ov I 0 . gvector diagram of cylindricalrotor machine. can be viewed as a fictitious reactance drop. since it is proportional to Ai and consequently proportional to the armature current. To produce the net mmf represented by the current.0 2. Neglecting saturation. It is called the drop of armature reactance and is designated x&i. which shows that. formed by drawing A B perpendicular to i or Ji and OB perpendicular to OC. the foregoing statement would still be true. As ri is only about one or two percent in practical machines. of course) by merely calculating ei and taking Zr from the airgap line of Fig. But.5 3.i can be combined into a single term called the synchronous reactance drop and there results xd=x. is equal to the vector sum of et.5 2. hydrogencooled Fig. UNIT AMPERESPER Fig.0 1. saturation curve of Fig.. The reactance drops z& and x.0 I PER UNIT FIELD AMPERES Fig. Zr has now such position and magnitude that Zr and Ai added in vectorial sense equals I.5 FIELD 2. it is the voltage taken from the airgap line of the noload saturation curve for the abscissa corresponding to Zr. The field current. 10 as being the current required to produce el. leads the terminal voltage by 90 degrees.0 2. the line OC.5 turbine 4. 6Saturation curves for typical generator. The triangle OAB. hydrogencooled con 9 field current If. Zr. The side Z1B of the triangle. .+z. ei. the vector diagram reduces to that shown in Fig. 11. SSaturation curves for typical denser. OB. 10.0 FIELD AMPERESPER UNIT I I I I 3. the armature current produces an mmf by its socalled armature reaction. At no load the axis of the field winding. ri and j s. can be determined for any condition of loading (neglecting saturation.5 I /I 1. which is in time phase with it and in terms of the field can be expressed as Ai. except for the effect of the resistance drop. In other words. It. At zero powerfactor. is similar to the triangle OX’. the statement 1.. is thus the opencircuit voltage corresponding to the Fig. 7Saturation curves for typical waterwheel generator. the field current must be of such magnitude and the field structure must adjust itself to such position as to equal Zr. OB has the same proportionality to OC and AB to :li as ei has to It.ii. iI) It follows from the foregoing that the internal voltage. designated as ei.
generator at zero Fig.. This flux can be resolved into two components +pdand aQ. The angle __is a real 6 ~. Thus the unsaturated synchronous reactance per phase is equal to the phasetoneutral voltage ilB divided by the rated current.. In this case all of the internal voltage (the ri drop can be neglected justifiably) must be consumed as synchronous reactance drop (xdi) within the machine. the proportionality between the mmf’s and their resultant fluxes is not the same in the two axes. IONoload and fullload zero powerfactor istics of a generator. the internal voltage can be determined by simply reading the terminal voltage when the shortcircuit is removed. A vedor diagram for such machines is shown in Fig. 10 OJ is the field current required to circulate fullload clurent under shortcircuit conditions. As before e.i can likewise be thought of as arising from the two components of i in the form of z.a_cylir?dyical.or machine.Jand z. and er is the ‘Lvoltage behind the leakage reactance drop. same as for a uniform airgap machine. it can be seen that x:. but where expressed in generatorterminal voltage and field amperes. the directaxis component of Ai. The flux @d is produced by If and Aid. and i are the / 6 it Fig. If there were no saturation. T~isc~~~i‘i~~~~orne purposes to resolve the reactions within the machine into two components. fg FIELD CURRENT Fig. When this is terminal volt. conditions are different. When saturation effects are neglected +d canbe regarded as made up of a component produced by If acting . maintaining the field current constant ‘X&ti@h%e.. In Fig. respectively. can be obtained most conveniently from the noload curve and the fullload zero powerfactor curve. llVector diagram of cylindricalrotor powerfactor.rot.angle.equal to x. 10 be equal to nB. i.148 a Machine Characteristics NO LOAD SATURATION cuRvE Chapter 6 RATEDCURRENT SATURATION CURVE done. can be accepted as true for all practical purposes. In Fig.. but a case will soon be developed for whi& they are not equal. Because of the saliency effect.” The flux@ is required to produce ei.1. in percent or F m per unit. The synchronous react’ance. and a1 is produced by Ai. the armature current is divided into the two components.& and z. and. it can be measured without much difficulty. For other powerfactors. x. character 2. In the case of. However.. respectively. 12Vector diagram of salientpole machine. as the real load is applied to the machine the angle 6 increases from zero and the lead of OC ahead of et increases from 90 degrees to 90 degrees plus 6.age to neutral and the armature current. This voltage wolrld in Fig. X~. 12.i. the performance of an unsaturated salientpole machine at zero powerfactor% the _. id. in which the subscripts are significant of their respective components. in leading quadrature to id and i. the terminal voltage being zero. When the saturation curve is csprcssed in per unit or percent it is equal to AB. it is equal to p(100) fz Ifsi . UnsaturatedSalientPoleMachineUnderSteadyState Conditions Ai If It Given the proper constants. one along the axis of the field winding and the other in quadrature thereto. the quadratureaxis component of Ai. Here the similarity ceases. 9...are both.
i. It is the sinusoidal c:omponcnt of this flux that is. C. drop of Fig. 12. (b) quadrature sin ($+6) (4) * alone and a component produced by Aid.curves just as for the machine with uniform air gap. the distance BC is equal to CC~~ (++6). cylindrical rotor machine theory. mmf in 13) (a) direct axis.+x...i of Fig. and then j zq i. ~~1. 13b. Saturation in SteadyState Conditions Fig. 14. 13Flux resulting from a sinusoidal axis. 4. This determination of the angular displacement of the rotor as real load is applied to the machine and the use of Eq. The mmf produced by Aid has a general sinu. The soit1a. When this is done 1.> Tn the quadrature asis. which is proportional to If.=r.~_ relative strengths of the field and armature ampere turns. and excitation I)ol. : .roduced by If can be rkgardkd as producing the internal termined from the noload and fullload zero powerfactor v(jltage c.r .E. 10 the SCR is equal .d.Chapter 6 FUNDAMENTAL COMPONENT Machine Characteristics 149 ~r)$qx the projection of BF upon OG is equal to r. sin 6 :u’mature current. 6. The armature resistance is usually negligible in determining either the angle 6 or the excitation and for this case ft sin fi=r. It can be determined from a test involving the it. then since angle CBP is equal to 4+6.I.r. the which gives f. is much smaller than x. effective in producing the x.l distribution in the direct axis as shown by Fig. and excitation of *n unsaturated salientpole machine when loading is known. i. C2). In Fig. Another form of the vector diagram of the machine is FOR SIGNIFICANCE OF QUANTITIES IN PARENTHESES REFER TO FIG. llDet ermination of internal angle.1~internal voltage ed can be obtained by merely adding ri.ween those quantities that are most useful for calcula1ion purposes..to the fi&l current required to circulate rated armature current \mth the armature shortcircuited. quadratureaxis synchronous reactance is not obtained so ‘~11~: resultant flux because of the variable reluctance of the air gap has the general shape indicated.. The slip test is described in the A. and gives rise to the distorted flux form. quantity.i is drawn of a saturated salient pole machine when loading is known. can be de. 12. The effect of this componcnt is reflected in the z. ‘l’hc armature re&%ance and leakage reactance drops ~a11 also be resolved into its two components in the two ascs much as z. If from B the line BP of length z.i sin I$ From Fig.. It. or zgig.ii. 12. pc>rpendicular to i.general In L. Further. 3.i cosC++6) (2) Cpon expanding the last term and solving for 6 .E.i so that OG becomes equal to rrl. The component The unsaturated synchronous rcactnnce. Test Code for Symhronous 0 MachinesLo of 1945 for a determination of sd. the internal voltage fd=ft cos 6+x. the component of mmf is hkewis& smusoidal in nature as shown in Fig.18 presented in Fig... The notat.Shortcircuit ratiolis a term used to give a measure of the ~~. id drop shown in Fig. the fictitious internal voltage.ion ed is used to differentiate the internal voltage in this development from that used with the . rclnt’ion provides an easy construction for the determinaLion of the angle 6 having given the terminal voltage. 9 was resolved. In proportion to the mmf the sinusoidal component of flus is mllcshless than for the direct axis. and the powerfactor angle. andj xd id to the terminalvoltage c.l and ri.hat the minimum ratio of armature voltage to armature current is used..i cos $h tan 6 = _____ e. 14. 15Determination of internal angle. which shows much better the relation Fig.& (4 (b) Fig. .l. The test for the determination of Z~ is identical except t._.definedas the rgtis of tho field current required to is produce rated armature voltage at no load. 13(a). By easily but fortunately there is not as much need for this cos comparing this line with the corresponding line in Fig. (5) x”=i cos (q5f6) F or it can be determined by means of a slip test. (‘an be seen that the point P determines the angle 6..
are designed for SCR’s of 2. Consideration will be given first to the characteristics for zeropowerfactor loading. Fig. the trend in this country has been toward smaller values. It may be quite important for hydrogenerators. Therefore. Synchronous reactance. The Potier reactance is the reactance that.. The increased field leakage at the higher excitation produces greater saturation in the field poles and this in turn increases the mmf required to force the flux through the pole. xp. the two simplest are: 2. SR. Tests indicate that for practical purposes both the cylindrical rotor and the salientpole machine can be treated similarly. see Fig.8. 2d. The leakage from pole to pole varies approximately proportional to the field current and the point T was determined upon the basis that this leakage was proportional to the field current MS.=x*+0. The foregoing analysis is not strictly correct. can be resolved into the current OD necessary to o&&%i& leakage reactance drop and the current DA required to overcome demagnetizing effects. On the other hand. 11 shows that for zero powerfactor. Take any terminal voltage such as MN of Fig. will just fit between the two curves at rated voltage. however. For most economical design a high SCR machine usually has a lower xd’. For most machines it is sufficiently accurate to USCthe one value obtained at rated voltage and rated current. They may be calculated but their values are dependent upon the assumptions made for the calculations. 10. The net effect is to increase the field current over that determined by the method just discussed causing the two curves to separate more at the higher voltages. to produce the terminal voltage OM with rated current in the armature. both because of its lower xd’ and higher WR2 a high SCR has a higher transient stability. the Boulder Dam machines. This mmf is represented on the curve by the distance ST. It can be determined from test curves. because transient stability is not of great importance in the systems in which they are installed. it can be seen that the total field current required to circulate rated current at short circuit which is represented by the point 4. Potier reactance decreases with increased saturation. to OA and then EF parallel to OB. The determination of the ratedcurrent zeropowerfactor curve can be developed as follows..8 xd’. The distance PR then gives the field current necessary for magnetizing purposes. is introduced.150 ‘. then the other becomes determinable from Eq. Sterling Beckwithlg proposed several approximations of Potier reactance. When no saturation is present it is simply the la1 reciprocal of the synchronous impedance. The voltage behind leakage reactance is obtained by adding to this voltage the leakage reactance drop. is a definite quantity and is equal to the distance A0 expressed in either per unit or percent. = 0. 2 Machine Characteristics Chapter 6 to F. The effects of saturation arise primarily in the determination of regulation. This is not usually a significant factor particularly in condensing turbine applications.4 and 2. The distance FG is then the Potier reactance drop. Other points on the ratedcurrent zeropowerfactor curve can be obtained by merely moving the triangle RST along the noload saturation curve.&. Upon sliding’the triangle RST down to the base line. When either XI or x8 is assumed. which gives the line PQ. It is impossible to specify the best specific SCR for a given system. The Preferred Standards for Large 3600rpm Condensing Steam TurbineGenerntorszO specifies SCR of 0. In recent years.. used in a triangle of the general type described. Regulntion is also worse but both of these effects are alleviated in part by automatic voltage regulators. the ri drop of the machine is in quadrature to the terminal voltage and internal drop and can have little effect upon regulation. Neither leakage reactance nor the field equivalent of armature current are definite quantities in the sense that they can be measured separately.) and 2. It will therefore be neglected entirely. . the point for which it is determined. 16Zero powerfactor characteristics of generator. giving MT as the field current required. In addition. lG. as it neglects certain changes in saturation in the pole structure. zd. The desire for smaller SCR’s springs from the fact that the cost is smaller with smaller SCR. static stability is not as good with smaller SCR. the conventional For other loads at zeropowerfactor.63~xd’x. field current is required to overcome the demagnetizing effect of the armature current. The concept of the determination of the curve of rated current at zeropowerfactor by the method just described is valuable and in an attempt to retain the advantages of this method the concept of Potier reactance. however. (1) or from the triangle just discussed. by drawing DE equal Fig. In the past it has been the practice in Europe to use somewhat smaller SCR’s than was the practice in this country. for example. Potier reactance is thus a fictitious reactance that gives accurate results for only one point..74.
540 in Reference 10. the tlctermination of the terminal voltage when the load current. load powerfactor. is the Potier internal voltage or the voltage behind the Potier reactance drop. including the effects of the point F. respectively. The vertical distance OF is then the terminal voltage for the particular excitation. 17Determination of regulation other than zero by the “adjusted method. ISDetermination of excitation. saturation.hs rated current the regulation curve would be the line H. or the determination of the excitation when the load current. For powerfactors other than zero. P~WEFWACTO ANGLE Fig. which is laid off in proper relation with the terminal voltage as indicated hy the powerfactor of the load. rotor phase .’ / FIELD CURRENT Fig. Thus for threefourt. (b) General MethodFor lack of a better name this method has been called the ((General Method. 16 in proportion to the armature current. The noload voltage NO LOAD SATURATION CURVE 0 80% POWER FACTOR b’ . Fig. To obtain the excitation at any other powerfactor for rated current. several methods are available to determine the regulation. LOAD ATION CURVE \ FACTOR RATED (a) Adjusted Synchronous Reactance Method* This method utilizes the noload and the ratedcurrent zeropowerfactor curves. and excitation are given.Chapter 6 Machine Characteristics 151 method is to divide the lines BA and FD of Fig. The resistance drop is SO small that it is usually neglected.4 is then regarded as an internal voltage and the distance JB as an internal drop of pure reactance. Following this procedure another excitation is chosen and the construction repeated from which the dotted line is obtained. e.. and terminal voltage are given.1 in which RH and FJ are threefourths of BA and FD. If the machine is not operating at rated current. the zeropowerfactor curve corresponding to the particular current should be used. an arbitrary excitation is chosen such as OC of Fig. The problem may take either of two forms. particularly at lagging powerfactors.” curves for powerfactors synchronous reactance C. 17. The intersection of the line with the normal voltage gives the excitation for the desired powerfactor at rated load. A line equal to the distance AC is then scribed from the point A until it intercepts the Yaxis at _“Uescribed m Method (c) Para. 1.. 19 with all terms expressed in per unit. / ‘I. load powerfactor. 19Determination machine with saturation of field current for round included by adding s in with ep. They all give surprisingly close results. The voltage..” It is based upon the assumption that saturation is included by reading the excitation requirements from the noload saturation curve for a voltage equal to the voltage behind the Potier reactance drop. The method is described in Fig. The construction is as follows: The adjusted synchronous reactance drop AB is hid off to m ak e an angle with the Xaxis equal to the powerfactor angle.
the total excitation Oh is obtained.hr powerfactor angle. 0. On. KW . and (2) that due to the field.85 p. .. The leading kvar capacity (underexcited) of aircooled condensers is usually about 50 percent of the lagging kvar capacity but for hydrogencooled condensers about 42 percent.hroughout this is also equivalent to the field current. If kh. II. is added along a line parallel to Oc. It involves the construction of ep separately so that s can be obtained in terms of field current. The effect of saturation is introduced by the distance s. The line centering abont the origin represents the limit imposed by the condition of constant armature current whereas the other arc by constant field current.5 shows the entire construction.. 14 is used to determine the excitation before including the saturation factor s. the center is usually located at a point equal to (SCR) times (rated kva). Armature resistance will be neglected except as it influences decrement factors.hen the Z. I. there is little difference between those two methods. Oc then represents e. zd.. and t. (d) TwoReaction Potier Voltage MethodThis method is similar to that of (c).. and with the *Described as Method (a) Para. giving as a result. must be such that its radius. read from the abscissa. the terminal voltage of the generator. xd passes through the point of rated real power and rated reactive power. When using per unit quantities t. 20Reactive power capacity of steam turbine generator 20 000 kw. with the receiver voltage held at a constant value. the excitation corresponding to this voltage is obtained. 9 and 10 the power circle of a line of such character2 istics has its center in the negative reactive axis at %. 18. the construction is also shown in Fig. If there were no saturation the synchronous internal voltage would be Ok. in Fig. This method gives the best overall results.5 psig hydrogen. The particular name of this method was assigned to distinguish it from the next method.8 SCR. except that the tworeaction method of construction shown in Fig. I3y scribing this back to the ordinate and reading horizontally. Real power is plotted as abscissa and reactive power as ordinate. sending voltage held at a const. 23 529 kva.520 in Reference 10. 19 escept that they are in terms of field current.PER UNIT (c) Round Rotor Potier Voltage Method*This method is the same as (b) except that the effect of saturation s. ap. The presence of paths for flow of eddy currents as provided by the solid core in turbine generators and by the damper windings in some salientpole machines makes the treatment of these machines. however. This arises because X~is smaller than xd. s. 18. Oh. equal to the saturation factor. The construction is the same as (c) except that the line Oq is made to pass through the point ginstead of k. This is to take care of saturation effects. ot.ant value ed. For this reason the threephase shortcircuit of these types of machines will be discussed first.i. Two limitations from the heating standpoint are recognized: (1) that due to the armature.. .9 Fig. this condition both relations reduce to an equivalence. This quantity is then added to the diagram for no saturation in terms of the field current. in Fig. THREEPHASE SHORT CIRCUIT In addition to its steadystate performance.f. As shown in Chaps. the generator can be likened to a simple transmission line of pure reactance. Since. Fig. required to produce the terminal voltage at no load and jk the excitation. 0. For the sake of comparison with other methods. with no saturat. Reactive Power Capacity The capacity of a synchronous machine to deliver reactive power is dependent upon the real power that it delivers. ~.. In Fig. This method includes the effect of saturation by simply adding sI the increment in field current for this voltage in escess of that required for no saturation. These vectors correspond to et and jk. Actually. et. for the synchronous reactance drop.is equal to SCR. at 0. respectively. When per unit quantities are not used the construction is a little more complicated. the action of a machine under shortcircuit conditions is important. for the sake of simplicity laid off along 012. to Ok in phase with e. With regard to the latter.and xd edet its internal voltage. making om the desired excitation. first lay off from the terminal voltage. As can be seen. .. Figure 20 shows the reactive power capability of a standardized 3GOOrpmsteam turbinegenerator. The field current required if there were no saturation is obtained by the construction Oj andjk where Oj repwscnts the excitation. 18 is. 1. especially at leading power factors. less complicated than that for salientpole machines without damper windings.ion ?. ed. however.152 Machine Characteristics Chapter 6 The tlistancejk represents the synchronous reactance drop. from a practical viewpoint. All the curves are arcs of circles. it can be seen that for xd 4. drop ac at an angle with the horizontal equal to t.
It is a wellknown fact that for the flux linkages with a . the presence of the damperwinding currents of salientpole machines and rotor ‘%dy currents of turbine generators can. 23).* ? 1). It is approximately proportional to the instantaneous value of the field current 1r. Before shortcircuit occurs the flux associated with the field windings can be broken up into two components (see Fig. / Consideration will be given to a simultaneous shortcircuit on all phases while the machine is operating at noload normal voltage without a voltage regulator. the socalled transient component.s. is the current finally attained. id’. except for the first few cycles. effect on the field fluxes. 22. the leakage flux varies from the base of the pole to the pole tip.. its circuit can be regarded as having a powerfactor of zero. 22. The flux ~91 so weighted that it produces is the same linkage with all the field turns as the actual leakage flux produces with the actual turns. Actually. The manner in which this quantity is related to the esponcntial and steadystate terms is shown in Fig. In discussing this component.Chapter 6 Machine Characteristics 153 Voltoge from airgap to noload line corresponding excitation 5.I). Ijecause of the demagnetizing effect of the large shortcircuit current. a leakage flux that can be regarded as linking all of the field winding. Fig. I’~(w two components decay or decrease together with the xd = Steodystote value TIME Fig. In this case. They can be divided into two parts: a. This type of machine was chosen to show of more clearly the presence of field and damper currents. machine % same time constants. Values are rms. value of field current required to produce normal voltage divided by the synchronous reactance in ohms. They are: :L The steadystate component 1). The total flux linkages with the field jvinding are then those produced by the flux (@++. the r. as its name implies. Alternating Component of Armature Current This component can in turn be resolved into several components. Because the armature resistance is relatively small. Estending this straight line back to zero time and adding the steadystate component. As the field structure rotates. the flux density within the machine decreasesbelow a point where saturation is present. The general nature of the currents that appear is shomn in Fig. is an esponential function of time ithe points lie in a straight line). As wi’l be seen later such machines also contain a second harmonic ConWonent curlent.q . The steadystate value of shortcircuit current is thus equal to the linetoneutral voltage read from the airgap line for the . tion is important only as it affects the field current necessary to produce normal voltage at no load. be neglected. values of which are shown in Fig. The alternating armature component c:m be regarded as being produced by its associated unitlircctional component in the field. 5 I % id : . as contrasted with cross magnetizing. likewise. An alternating component in the armature and assoc. of course. Threephase ShortCircuit of Machines current Paths in Field Structures with . An unidirectional component in the armature‘&. called the transient reactance by means of t. for the moment. Satura*The machine used in this case was a salientpole machine.m.rd’an alternating component in the field or in the damper windings. the alternating current in the field winding can be regarded as produced by the uni(lircctionnl component in the armature. The transient component c. This component is defined through a new reactance. The symmetrical current thus produced develops an mmf that rotates synchronously and has a purely demagnetizing. SteadyState ComponentsThe steadystate component. a balanced alternating voltage and current of normal frequency are produced in the armature. The subtransient component loach of these components will be discussed separately.i:lted. Transient ComponentIf the cswss of the symmetrical component of armature currents over the steadystate component be plotted on semilog paper. a component Cpthat crosses the air gap and a component ai.. atThreephase with short circuit in salientpole damper windings.nith it an unidirectional component in the field. All phase components of the alternating current are essentially the same except I hat they are displaced 120 electrical degrees. 22Symmetrical current (threephase component of armature shortcircuit short circuit from noload rated voltage). it can be seen that this excess.he expression id’ = erated __ x’d 6. 21. or armature current is obtained.
Since inductance is defined as the flux linkages per unit current. but that the direct component of field current increases to s xd’ consequently el had remained constant during the trsnsition period. is the exciter voltage. The increments of both follow an exponential curve having the same time constant. the flux linkageswith the field winding can be regarded as constant. the increased armature current. an infinitely large voltage is necessary and the assumption is justified that. t is time in seconds.It follows then that the transient reactance must be greater than the armature leakage reactance. It is a reactance that includes the effect of the increased field leakage occasioned by the increase in field current. The time constant is equal to the inductance of the field winding divided by its resistance. can likewise be viewed as being produced by a fictitious internal voltage behind synchronous reactance. This. If it were erated xd times the open circuit value before shortcircuit. as just shown. 22. the airgap flux decreases and. then this component of shortcirc!lit current would remain sustained. then the transient component of shortcircuit current would be merely the noload voltage before the shortcircuit divided by the leakage reactance and the transient reactance would be equal to the armature leakage reactance ~1. If a constant direct voltage is ~%&l$ifiKapplied to the field of a machine with the armature opencircuited. in seconds xd The component of armature current that decays with / I this time constant can then be expressed by 1 When t is equal to Td’ the magnitude of the component/ has decreased to el or 0. If If increases then @I. It is important tk”@%$p the significance of this !#l%. as stated. is the opencircuit transient time constant of the machine or of the circuit in question in seconds. If + and possible to increase the exciter voltage instantaneously to an amount that would produce this steadystate field current. In order that this flux remain constant in the presence of the demagnetizing effect of the armature current. T&. must likewise increase. At the first instant of shortcircuit.368 times its initial value. for the transition period from the no load opencircuited condition to the shortcircuited condition. the armature current is less. The mathematical expression of this relation is: 03) in which e. The initial value of armature current. rf is the resistance of the field winding in ohms. no load. The armature current for shortcircuit conditions is equal to z. the armature current can be viewed as produced by a fictitious internal voltage equal to 2d id whose magnitude is picked from the airgap line of the noload saturation curve for the particular field current. id’. the airgap flux at the first instant of s circuit’ is prevented from changing to any great ext This results both from their close proximity to the air . Consideration of the steadystate conditions has shown that the airgap voltage. Attention will next be given to considerations affecting this time constant. the current’ builds up esponentially just as for any circuit having resistance and inductance in series.he armature and the unidirectional (often called directcurrent) component of current in the field winding. et. therefore. whose magnitude is x&d’ or if the shortcircuit be from rated voltage.154 Machine Characteristics Chapter 6 Fig.. which is proportional to it. whether the operating condition bc steadystate or transitory. Under steadystate conditions with no saturation. it follows then that the inductance of the field circuit under short’ I circuit must equal 5 times that for the opencircuit condition. . The shortc?cuit transient time constant that is the time constant that determines the rate of decjy of thl transient component of current must then equal Td’ = “T’d. 23Airgap and leakage fluxes at no load. gradually decreases to the steadystate and the induced current in the field winding likewise decreases to its steadystate magnitude. This is equivalent to saying that the flux (+++I) remains constant. Subtransient ComponentIn the presence of d er windings or other paths for eddy currents as in tu generators. xd’ ’ This voltage provides a means for determining the initial value of the unidirectional component of field current by picking off the value of If on the airgap line of the noload saturation curve corresponding to this voltage. l’here is always a constant proportionality between the alternating current in t. is proportional to the airgap flux +. it was shown that at the first instant the flux linkages with the field winding remain the same as for the opencircuit condition. instant is indicated in Fig. circuit to change instantly. It follows then that @ must decrease. In the case of the shortcircuited machine. it is necessary that the field current If increase to overcome the demagnetizing effect of the armature current. However.
mPoncnt gives the subtransient current. Since the copper section of this winding is so much smaller than t. Cb) Single vector projected on three axes. Therefore at times of threephase shortcircuits.05 second instead of the order of seconds as is (*tr:uacteristic of the transient component. 25Representation of instantaneous currents of a three phase system.hat of the field winding. The maximum magnitude which the unidirectional PROJECTION \’ AXIS FOR PHASE b 7. it follows that their currents cannot change instantly from zero to a finite value.= in x2 seconds (8) 2+. The application of this theorem thus gives rise to an unidirectional component of current in each phase equal and of negative value to the inst. The quantity 27rf merely converts the reactance to an inductance.o 0. This subl. Total Alternating rent Component of Armature Cur ‘l’lrc total armature current consists of the steadystate v:rllrc and the two components that decay with time conSl:IIlts Td’ and Td”. Tests on machines without damper windings show that becauseof saturation effects. The component of armature current that decays with this time constant is (id” id’) and can be expressed as a function of time as (i:’ id’) E Thus the time in seconds for this component to decrease t. (a) Fig. The magnitude of this time constant is dependent upon the ratio of the inductance to resistance in the armature circuit.‘i:)~ . As will be shown the negativesequence reactance.sicnl current is defined by the subtransient reactance in the csprcssion id” &ted = Xd)’ only one can equal zero at a time..antaneous values of the alternating component at the instant of short circuit. 8. ~2. 24The inclusion of a dc component of armature current whose existence is‘necessary to make the armature current continuous at the instant of short circuit.. The influence of current magnitudes as reflected by saturation upon the transient and subtransient reactance is discussed in more dcLai1under the general heading of Saturation. being :J)out 0.arl. called the armature shortcircuit time constant. Since the excess of the armature currents represented 1)~ t.. the straight line thus formed can be projected back to zero time.. SYMMETRICAL COMPONENT Fig. Since the armature circuits are inductive. 24. of the machine is a sort of average reactance of the armature with the field winding shortcircuited. There exists then the relation T. It can be expressed by the following ~‘~lllation $.Chapter 6 Machine Characteristics 155 and from the fact that their leakage is much smaller than that of the field winding.he subtransient components over the transient comlroncnts are sustained only by the damper winding currents. Consequently. 22. Each of the unidirectional components Tlrc subtransient reactance approaches the armature leakLLgc differing from that quantity only by the leakage of the &rmPcr windings. it would be expected that their decrement would be tlotcrmined by that of the damper winding.’ is very small. id”.368 times its initial value gives Td” as indicated in Fig. 9 in the three phases decays exponentially with a time constant T. This zerotime value when added to the transient . In this manner the armature currents are made continuous as shown in Fig. in which rB is the dc resistance of the armature. the initial shortcircuit currents of such machines are greater.1 Td”j((ii&)E . the alternating component of current in at least two and probably all three phases must change from zero to some finite value. the shortcircuit current even in this case can be resolved into a slow transient component and a much faster subtransient component. Unidirectional Component of Armature Current (b) To t)his point consideration has been given to flux linkages with the field winding only. If this excess of the symmetrical component of armature currents over tlrc transient component is plotted on semilog paper.ities are all expressed as rms values and are equal l)ut displaced 120 electrical degrees in the three phases. The “theorem of constant flux linkages” must apply to each phase separntely. (a) Three separate vectors projected on xaxis. . The requirement that tlrcse linkages remain constant at transition periods determined the alternating component of armature current..t Td’+id ‘Ilrc quant. so that it is the reactance to use in determining T. Td”.=(i. Since these components in the three phases have a phase displacement of I20 degrees with resnect to each other. it is found that the shortcircuit srrbtransient time constant..
The espression then becomes x2fxext in seconds Ta=27r. however.. in the vast.xd+&t A symmetrical threephase set of currents can be represented as the projection of three equalspaced and equal length vectors upon a stationary reference. for faults out in the system is so small as to justify neglecting the unidirectional component of current.~+j x. obtaining first the linkages with the field winding for zero 10. then current is then equal to the linetoneutral voltage read XA from the airgap line for the field current obtained for the loaded condition divided by xd. gives the transient component of : . They can also be represented as the projection. and r. component can attain is equal to the maximum alternating component. the flux linkages tc. m seconds . as it rotates. Short Circuit from Loaded Conditions The more usual case met in practice is that of a shortcircuit on machines operating under loaded conditions. For the armature time constant.156 Machine Characteristics of the The shortcircuit manner Chapter 6 time constant is affected in a similar Tdf. then the unidirectional components can be represented as the projection of a single vector onto the three equalspaced axes. short circuit. 11. In applying circuit breakers before short circuit whose power factor is zero.ions of circuits. Thus the components of shortcircuit Let $1 be the flux linkages with the field winding at current become noload at rated voltage. Effect of External Impedance armature current and any terminal voltage and then the If the shortcircuit occurs through an external impedflux linkages with armature current. idT. for machines in which zg” and Ed” are radically different. it is usual to use a factor 1. and (3) subtransient. The foregoing expression assumes that both the period from one circuit condition to another. the esternal reactance must be added to the negativesequence reactance of the machine and the external resistance to the armature resistance of the machine. a rms value as its name implies. is the flux alternating and the unidirectional components do not de.6 instead of 1/?. the shortcircuit current in the armature can be divided into two components. In reality the decrement is usually sufficient to* for analysis.. 9. For the shortcircuit from crease. by these amounts. For any other terminal voltage suchas et.. These axes can conveniently be taken as shown in Fig. Total RMS Armature Current Alternating ComponentThe alternating compoThe rms armature current at any instant is nent in turn can be resolved into three components: (1) steady state. divided into the linetoneutral rated voltage.. The field current can be determined by any that is. circuit will be determined by a superposition method. and zero terminal ance ~..~. The minimum current thus occurs in the phase in which The load on the machine affects the steadystate comthe unidirectional component is zero and the maximum ponent only as it influences the field current before the occurs when the unidirectional component is a maximum. majority of cases T. because of the natural decrement. This fact is used at times to determine the maximum magnitude which the unidirectional component can attain. will be equal to e rated ii’ = xd”+&t id’=.” Saturation will be more important than for the noload condition. Each of d/idc2 + cc2 these components will be discussed individually. of one vector upon three stationary axes. a symmetrical alternating component.linkages with the field winding. Consideration will be given first to a load make the effect noticeable.0r. Since the of the methods discussed under the heading of “Steadymaximum value that the unidirectional component can State Conditions.t) Because of the much lower ratio of reactance to resistance in external port. The flux linkages before short includes a small decrement. when maximum dissymmetry occurs.. The total flux linkages is the sum of the two be introduced by merely increasing the armature constants values so obtained. The steadystate shortcircuit attain is +‘2 %. as the horizontalaxis and two axes having a 120degree relation therewith. ide(max.+T. By its use it is unnecessary to await a test in which the maximum happens to occur. Therefore. lagging. As before. during the first loaded conditions this same quantity can be used as a basis cycle. is not too large. such as transformers or transmission lines. is an average component from the noload condition. their effect can voltage. say the real asis. spaced 120 degrees. 12) transient. Since the initial magnitude of the unidirectional components are the negatives of the instantaneous values of the alternating components at zero time. it was stated that quantity and is usually taken over a cycle or half cycle the quantity that remained constant during the transition of time.) = fi 5 (9) xd’fxext TdIE. This method is in error. and a unidirectional component. This factor and whose current is &L. In the discussion of the determination of the transient Of course. 25.&ted xd’+%xt id = eairgap *t no lOad Cl21 (13) erated xd+%xt By definition the transient reactance of a machine is ’ equal to the reactance which..
&ted xd’$l 9 By application of the superposition theorem. for nearly all practical purposes it is sufficiently accurate to replace ed’ by the amphtude of a quantity ei’.. which is equal to the terminal voltage plus a transient reactance drop produced by the load current.hc instant of transition. (17). ed’./. For loads of zeropowerfactor lagging the subtransient reactance drop. This gives i’ =  J/ = et+xd’idL xd’ . $1 &ted (18) Since the flux linkages with the field winding produced hY a unit of current id under shortcircuit conditions is qua1 to xA &ted then the transient component of short ner that edwas determined in Fig. namely A$1 ersted and th ose by the armature curxd”. it is permissible 1. The construction for this quantity is shown in Fig. But these conditions are always satisfied even under steadystate contlitions of short circuit. then the linkages are directly additive. in (19). 26 and to assist in the ready evaluation of the amplitude the curves in Fig. &. 27Machine internal voltage Fullload rated as a function voltage. 14. the vector sense of current and terminal voltage must be introduced. the total flux linkages with the field winding can then be regarded as the sum of the flux linkages produced by the terminal voltage.e t.0replace id’ in this expression by i&. assuming always that the armature is shortcircuited. of reactance. 27 are provided. and there results for the flux linkages with the field Fig. For any other armature current. The voltage ed’ should then replace et+xd’ &. This can be accomplished by computing cd’ for the operating condition in the same man While this c\xprrssion was derived from considerations applying only t. When the power factor of the loads considered is other than zero lagging. in general. i. can be eliminated giving +=id’..cnt is overcome by an increase in the field current. The demagnetizing effect of the armature cur. = (et+xdlidL). can be regarded as producing the flux linkages $1 with the field winding. namely idr. id’. However. of internal volt The subtransient component of shortcircuit current is obtained in a manner similar to the transient component except that the subtransient reactance is used in the calculation of the internal voltage ei”. The flux linkages with the field winding for the steadystate shortcircuit condition thus become idL. so. If the REACTANCE IN PERCENT armature current lags the voltage by 90 degrees. xd”& . Thus the armature current &‘I with its associated field current which is always proportional to it. except that xd should be replaced by xd’. synchronous or subtransient reactance is used. remain constant. xd’kh erated xd’$l er3ted (19) The numerator of this quantity can be regarded as an internal voltage. the flux linkages with the field winding are equal to 2. the flux linkages with the field winding. The transient component of shortcircuit current is then ‘k 26Construction for the determination ages ei’ and ei”. its application is more general. Combined with Eq.Chapter 6 Machine Characteristics If this (17) 157 shortcircurt current at noload normal voltage.it the instant of shortcircuit from noload at rated voltage.h &ted rent with zero terminal voltage.: $1. ‘rhc only necessary considerations that must bebsatisfied :~rcthat the armature be shortcircuited and that the field cllrrcnt contain a component of current to overcome the &magnetizing effect of the armature current. which is usually referred to as the voltage behind transient reactance to distinguish it from similar internal voltages for which leakage. shortcircuit current is designated as idI. then idll &ted = xd’ circuit current i’ can be determined by dividing these linkages into the total flux linkages just determined.
xd&. I xd” For terminaltoterminal short circuit.=e. into the steadystate. Id. III. however. Where et = Terminal voltage before shortcircuit. Before the shortcircuit the armature current is equal to Ida and has some position with reference to et such as shown in Fig. Other ConsiderationsTime constants are not influenced by the nature of loading preceding the shortcircuit. The magnitude of this quantity varies between this amplitude and zero depending upon the point in the cycle at which shortcircuit occurs. For other powerfactors ei” can be obtained from Fig.158 Machine Characteristics Chapter 6 cnuscd by the armature current is directly additive to the terminal voltage and for zeropowerfactor leading directly subtractive. transiel and subtransient components. that this is only an approximate solution. The unidirectional current performs a similar role for the shortcircuit from loaded condition. Where It = Initial 1 (2 value of field current. Doherty and NickleG have developed expressions for the armature currents for a threephase short circuit from no load. is Fig. the ac componeI of the phase currents are given by .% 2x62. Among other complications. as before.cos (27rft+a+ 120”) \ / (24) I (25. . xdxd’ z. r2+xd’x Td’ = qT&. (26)] 2. It must be recognized. The 42 times the vector difference between these two quantities (since they are rms magnitudes) gives the unidirectional component necessary to produce smooth transition. ThreePhase Short Circuit of SalientPole Machine without Damper Windings For most applications it is sufficiently accurate to treat the salientpole machine without damper windings just as other machines. T components of armature current and the time constar for the different kinds of shortcircuits are given below f stiortcircuits at the terminals of the machine. The instantaneous field current.& 2x.’ Id= If x*’ E c Td’6 C Gi cos 27gt + If ponent. Total rms currents can be determined by the relations already given. 28. The shortcircuit currer can be resolved. 28Showing that idofor a short circuit from load is equal to the negative of 1/2 times the difference between i” and i. These are given below. r2+xllxq 2xd’x. (23) A +.: cos tcr120’). lags ei” by ninety degrees so i” and Ida will be determined with respect to each other.% 2x:x.xd’ et. The difference betwe these components decreases exponentially as before. xq+zd’ cos (47rft+a+120°) et. i”.xd’ et. I For threephase shortcircuit: i”= fiI. the negativesequence impedance. . Phase Currents for Unbalanced Short Circui As explained in the chapter relating to Symmetric Components.= hd’+xq)’ c =Angle which indicates point on wave at which: shortcircuit occurs. T. i’=e’ xd’ 12. 27 by using zd”. in reality a strong second harmonic is present in the armature current. a the zerosequence impedance. 2x:x. UNBALANCED CONDITIONS 13. the unbalanced operating conditions of rotating machine can for most purposes be described terms of three characteristic constants: the positive! quence impedance. et. The subtransient come’f Unidirectional ComponentIn cos (27rjt+cr 2. . . the unidirectional component of current was introduced to prevent a noncontinuous transition of the instantaneous value of current from the noload to the shortcircuit condition.~h xdxd’ &+zd’ cos (47rft+cr120’) cos (crf 120”) cos (27rft+a+ 120”) I.x. The subtransient component of shortcircuit current is then zd ib = et+ x&d’ +z zrl cos (27rjt+a120°) 1207 the threephase shortcircuit from no load.
l’ + x:n .hen the current is large and if the rotor is moved through 90 electrical degrees then the clurent is much smaller. the sing]c value is used for all conditions.Chapter 6 . This is the position in which the subtransient reactance. d3Ci Xd’+XZ Xd+X? X.0t. The in%innry component of the impedance is called the nega(ivcsequence reactance and the real component the nega‘i\csequence resistance. should be somewhat the same as Q” and xq”. zd”.c those values corresponding to the particular load c:ondition.“. Thus.__Kcnerator s&i&Zls reqmrefliXBi6~ nibchine be. the terminaltoneutral shortcircuit currz %&ater th~~~~~~ph~~~sh~~~~~i~c~it current. UN] in the damper windin g if the machine has one. The’ . 29Relation between subtransient quence reactance. currents of normal frequency were induced in the field.6 0. the ac components ()f the phase currents are given by Z<.’ = l’ r Z. ‘i’crminaltoneutral short circuit 32d” Threephase short circuit =x~“+Q+x~ ‘I’hc negativesequence impedance. In order Ihat the terminaltoneutral current not exceed the three[chasecurrent a reactor should be placed in the neutral of the machine of such value as to bring the zerosequence impedance of the circuit equal to z~“. therefore. Machine Characteristics dei” +. The first position corresponds to the case of a transformer in which the secondary winding is shortcircuited. that is. These will be discussed separ:‘tely. then the field winding takes different positions successively as the armature field rotates with respect to it. The nature of the impedances in the two extreme positions. the neutral rwctor. and negativese 14.‘O the definition of negativesequence reactance is equal to “the ratio of the fundamental component of re .. 29. The ratio of the phase currents for terminaltoneutral t. is determined when the field is in this position and is equal to onehalf the quotient of the voltage divided by the current. is usually equal to s. NegativeSequence Reactance ‘I’hc negativesequence impedance of a machine is the irnpodance offered by that machine to the flow of negatives(‘(llwnce current. The unidirectional . is determined. ‘rllc subtransient time constant. for the phase currents i= 3ei z~i+z2+zo Td’ = up bv the armature current. Thus. does not change sigrlificantly with different conditions and. z~“.. Td”. should be 0.I3 (30) Q+Z2+xo in \vhich x0 is the zerosequence impedance of the machine.n:lc:hinc I?.hreephase short circuits can be obtained from Eq’s (30) :tt~tl (28).~(x~“x~) 3 OF ROTOR IN DEGREES Fig. T.. x2. A set of negativesequence currents in tllc armature creates in the air gap a magnetic field that rl)Mes at synchronous speed in a direction opposite to that of the normal motion of the field structure.&.Oomponents and the rms values are determined just as .and such is the case. For t lrcse cases.)r terminaltoneutral short circuit. ei’ and ei” will naturally l. The reactance fw intermediate positions varies between these two amounts in accordance with the curve shown in Fig. According to the AIEE test code. whereas. braced only for currents equal to the threephase values. but for many machines x0 is less than xd”. If the axis of the shortcircuited field winding lines up with the axis of pulsating field t.‘+xZ+x” T ..~“. It is equal to onehalf of the voltage from terminaltoterminal divided by the current.l~!sc:rit)ed under the general subject of “Short Circuit from l. One would expect therefore that the negativesequence reactance x2 is some sort of a mean between xd” and LX. the only signnificant difference being the fact that. in the order mentioned. The armature current is then determined by the magnetizing characteristics of the air gap in the quadrature axis. Currents of cl()ul)le frequency are thereby established in the field.2  ’ COPPER DAMPERS OO 30 ANGULAR 60 POSITION so 120 150 160 I x.If~” (29) in which ~41is the negativesequence impedance of the .” The above values of ei. The subtransient reactance. where the field winding lines up with the magnetic field and where it is in quadrature with it. x. _ dei' 159 Y _  z= Xd”+X:! . If a singlephase voltage is applied across two terminals of a salientpole machine without dampers while its rotor is stationary.. the resulting current is dependent upon the Position of the rotor with respect to the pulsating field set When a set of negativesequence currents is made to flow through the armature with the field shortcircuited and rotating in its normal direction. in the negativesequence case the currents are of twice normal frequency. the field winding in this case corresponding to the secondary winding of the transformer. For the second position the field winding is in quadrature to the pulsating field and consequently no current flows in the field winding.. in the determination of xd” and xq”.
In Table 1 are shown expressions” for x2 based upon different definitions.. the method of test to determine this quantity involves a sudden shortcircuit and from this standpoint proves rather inconvenient. (31) Fig.160 Machine Characteristics Chapter active armature voltage. l?he values for x9”. reading. XtJ+2d’ Zq+xd’ 2 JG d. but if the rootmeansquare figure of the resultant current is used then the expression in the second column should be used. For such machines the differen r between x4” and xd” is not great. In this table For each test condition it is possible to establish definitions based on whether fundamental or rootmeansquare currents are specified. This table is based on a machine without damper windings for which x9” is equal to x.rq Zq+5d’ 2 RootMeanSquare of sinusoidal negativese(2) Application quence current (3) Initial symmetrical component of sudden singlephase shortcircuit current (4) Sustained singlephase shortcircuit current (5) Same as (4) with 50% external reactance (6) A. 30. figures have been inserted in the expressions given in Table 1 for a typical machine having the constants xd’ = 35%. and xd = 100%. In order to orient one’s self as to the relative importance of the different expressions. the negativesequence reactance that would result in the proper rootmeansquare current for method (3) would appear to be the most important. and A. 30Determination of the negativesequence of symmetricallywound machines. The magnitudes are tabulated in the righthand columns of Table 1.ar xd” of a machine with copper dampers are given in Fig. due to the fundamental negativesequence component of armature current. several different defia * nitions can be given for x2. I. addition to the method impli Method of TestIn by the AIEE Code and the ASA whereby x2 is defined the arithmetic mean for zd” and x9/‘. and zd” is equal to xd’.I.S. the figure for x2 obtained from the use of the rootmeansquare values in a sustained singlephase shortcircuit current [method (4)]. Conversely if the currents are sinusoidal the voltages mill not be. in the first definition if the fundamental component of armature current is used in calculating x2 then the expression in the first column should be used.E. impedar OF NEGATIVESEQUENCE Analytical Expressions REACTANCE Kumerical Values xd’=%% 2. For these machines negativesequence reactan can be taken equal to xd”. That this is possible is dependent largely upon the fact that when a sinusoidal set of negativesequence voltages is applied to the armature the currents will not be sinusoidal. When the resistance is negligible this negativesequence reactance is equal to x2=zd I TABLE ~DEFINITIONS where I equals the rootmeansquare armature current j the shortcircuited phase. is nearly equal to this quantity. x2 can be determih directly from test either by applying negativesequen voltage or by the method shown in Fig. the same arguments can be applied to 0th types of machines such as turbine generators and salien pole machines with damper windings when the paramete xd” and x4” are used. In general. x. for turbine generators.rd’. zt=(sin 4. From the standpoint of practical application.1 However.A. = 70%.=70ojo xd= 100% Fundamental RootMeanSquare 44 74 48 47 50 Definition Fundamental (1) Application of quence voltage sinusoidal negativese2xCJ’xc.” A rigorous interpretation of this definition results in x2 equal to the ” + Xd” arithmetic mean X. saturation introduc variables of much greater magnitude than those just co sidered. For example.+jz. E then.31 where P= wattmeter ++j cos 4) =r.=I EC=E EA$= ?#O+dEaE)=q If ~=COSI&. On the other hand. to this component of armature current at rated frequency.. 2 For such machines the difference between xz based on tl different definitions of Table 1 will become inconsequentir In addition. However.E. .=0 Ea=O Ib=I Ee=E I. and E equals the rootmeal square opencircuit voltage between terminals before tl shortcircuit is applied or the noload voltage corresponl ing to the field current at which I is read.
Neglecting r. which can always be determined by solving the equivalent circuit involving the stator and rotor constants and the magnetizingcurrent constants. the 12. SO that. For a machine without damper. represents the S useful shaft power. Or E. A more convenient device. I rcrotor current. with the rotor turning at synchronous speed in the reverse direction. InductionMotor DiagramThe nature of the negativesequence resistance is best visualized by analyzing tlrc phenomena occurring in induction motors. of the rotor plus the shaft load. En= applied voltage. r. and the real part of z.I. A further increase in slip necessitates motion in the opposite direction.r. 31Equivalent circuit of induction motor.e.=:r. induces the voltage E. The justification for this diagram is shown briefly as follO\~~:The augap flux created by the currents I. represents the rotor cop ls per loss. In the rotor the impedance drop is T.between two of its terminals (neutral excluded) the sustamed armature current and the voltage between the terminal of the free phase and either of the shortcircuited phases are measured. For higher resistance tlnmper windings the negativesequence resistance increases to :I point beyond which the larger resistance diminishes the current in the rotor circuits sufficiently to decrease tt1c loss.= rotor resistance.eddy currents. The rotor current is therefore determined by the equation SE. The factors of fundamental importance are the power supplied to the stator and the power supplied to the shaft. The reading of a smglephase wattmeter with its current coil in the shortcircuited phases and with the above mentioned voltage across its potential coil is also recorded. i. With the rotor locked. l2 and the shaft load r.l. This resistance is designated the negativesc(lucnceresistance. %=shunt impedance to include the effect of magnetizing current and noload losses. = r. I$. (32) since the reactance varies with the frequency of the currents in the rotor. The total power absorbed by ?J must be the sum of the rotor losses and the useful shaft s power. the power absorbed by r. Half of the machine loss associated with the negativesequence current is supplied from the stator and half by shaft torque through the rotor. la=stator current.. i. the electrical input into the stator is 2. in the rotor.73. loss is I?r.. (33) 15. Physically that is just what would be expected.%t small slips the electrical input into the stator is equal to the copper loss. the copper Iqr.windings the only source of loss is in the armature and field rcsist.I*kjSX. the shaft load is zero.e. 31 It follows from this equation that the rotor circuit can be completely represented by placing a circuit of impedance :+jr. zp=rotorleakage reactance at rated frequency. . This is the condition obtaining with respect to the negativesequence in which the rotor is rotating at a slip of 200 percent relative to the synchronously rotating negativesequence field in the stator.ances. and ls ST.I. .. is given the usual equivalent circuit of an induction motor in which r8 = stator resistance. in the stator and SE. S=slip. and with a singlephase shortcu’cuit apphed . but the iron and eddy loss in the rotor may be considerable. resolving y into the resistances r. and iron loss. The relatively low resistance of this path results in a smaller negativesequence resistance than if the flux were permitted to penetrate into the rotor. At 200percent slip the electrical input into the stator is equal to the mechanical input through the shaft. for as the slip increases from zero the shaft power increases to a maximum and then decreases to zero for loopercent slip... Fig. and the total electrical input into the stator is equal to the rotor copper loss. divided by 1.chapter 6 Machine Characteristics 161 \Vith the machine driven at rated speed. and I. In Fig.. . ze=statorleakage reactance at rated frequency. which requires a driving torque. half of the copper loss is supplied from the stator and half through the shaft.I: or A negative shaft 2 2 . load signifies that the directcurrent machine instead of functioning as a generator is now a motor. The power absorbed by r.+ jsx. The copper loss in the armature and field is small as is also the iron and &ly loss in the armature.. At 200percent slip. hssume that the induction motor drives a directcurrent generator. NegativeSequence Resistance The power associated with the negativesequence current can be expressed as a resistance times the square of the current. the only real power is that concerned in the rotor circuit.+jx. Copper damper windings provide a lower impedance path for the eddy currents and lrintlcr the penetration of flux into the pole structure.I. The negativesequencereactance equals this impedance multiplied by the ratio of power to the product of voltage and current. across the voltage E. The negativeycqllence impedance equals the ratio of the voltage to the current so measured.
in turn is equal to the lc absorbed by the negativesequence resistance. The ratio of the negativesequence shaft power to the l( in the negativesequence resistance is then equal to t ratio of the power loss in 5 for unit negativesequer 2 current in the stator to r2. 12. their losses are supplied by power flowing into the machine from the system. the following method has 1 advantage that the machine supplies its own negati sequence voltage. = friction and windage loss. since there are no internal generated voltages of the negative. Cb) neglecting armature and no load losses. the negative. The total electrical effect of the negativesequence resistance in system analysis problems is obtained by inserting the negativesequence resistance in the negntivesequence network and solving the network in the usual manner. A !c u 2 supplied to the stator. 33. which produces a torque tending to decelerate the rotor. It was shown that this power tending to decelerate the rotor is numerically equal to the negativesequence power F in which P. Fig. which. and the power loss in r2 is equal to the loss supplied from the stator of the machine and the equal loss supplied through the shaft. The total electrical output of a generator. shown in Fig. = IL. Or viewed differently. not including the shaft torque developed by negativesequence current. 33Negativesequence resistance machine. and El = ES. (a) Negativesequence diagram for induction motor. the contribution of the negative. This ratio can be obtained eas by test by measuring the shaft torque and the negati7 sequence input when negativesequence voltages only4 applied to the stator. The components of this impedance will be called the negativesequence resistance r2.and zerosequence power outputs are merely the negative of their losses. In other words. While this analysis has premised inductionmotor CC struction. In addition to this electrical output. Howev the negativesepence power output per phase is equal the product of E2. Therefore. The current flowing through the negativesequence impedance is the current flowing through the stator of the machine. and the cosine of the angle betwc Et and IZ. the conclusions can also be applied to synch nous machines. The equ alent circuit and vector diagram for this connection . is equal to the total terminal power output plus the losses in the machine.162 Machine Characteristics Chapter 0 jxs 1 (a) L (b) B zm I jX.=shaft input. (c) simplified networknegativesequence resistance and reactance. there also exists the negativesequence shaft power supplied through the rotor. since s is constant and equal to 2 for the negativesequence. and the negativesequence reactance x2.or zerosequence. The positivesequence power per@ at the terminals is equal to the product of El and 11 a the cosine of the angle 4. is to reduce the equivalent network to a simple series impedance as shown in Fig. the stator resistance and the 10s: in the magnetizing branch can be taken into considerati by substituting them in the equivalent circuit and reduci that circuit to simple series resistance and reactant wherein the resistance becomes the negativesequence 1 sistance and the reactance the negativesequence reactan. The assumption was made that the stator resistan and the losses in the magnetizing branch were neglectc For greater refinements. Therefor the total decelerating power is equal to the positiv sequence power output plus the loss in the negativ sequence resistance. The total electrical output reduces then to that of the positivesequence and to include the positivesequence armatureresistance loss it is necessary only to use the positivesequence internal voltage in the calculations. and since I. the corresponding internal power must be zero. the negati Fig. This power is positive.+. P. Method of TestWhile r:! and x2 can be determined applying negativesequence voltage from another sou of supply to the armature. of a synchron . Two terminals of the machine uric test are shortcircuited and the machine driven at ral frequency by means of a directcurrent motor. All three of the sequence currents are thus affected to some extent by a change in the negativesequence resistance. 32Development of negativesequence resistance and reactance from equivalent circuit of induction motor. However. 32 (c).and zerosequences to the electrical output is zero.
the negativeyc. whereas.I. The per unit system4 of designation is advanced as possessing all the advantages of the percentage system but avoids this last mentioned disadvantage.chapter 6 seqllence Madhe Charucteristics 163 output is the negative of the positivewr seq. as a percentage drop would have much significance. A negative output is .L. its regulation a certain percentage of its voltage rating.and c~tlling P. can often be expressed by a single set of constants when those constants are expressed in percentages. the positivesequence resistance. are Therefore the total shaft irl’)\lt. By this is meant that the loss will be a certain percentage of its kilowatt rating. 34. PCF+Wj. It is the same as the percentage system except that unity is used as a base instead of 100. must follolv since (he output of the mnchine is zero. if the urmature windings were infinitely distributed so that each pliase produced a sinusoidal distribution of the mmf then t’~e mmfs produced by the equal instantaneous c&rents Of the three phases cancel each other and produce zero ‘e’d and consequently zero reactance except for slot and The performance of a whole line of apparatus.e. of course.t per phase is ~12. the figures are much smaller than those of positive and negative sequences. since I1=IZ. whereas. into the alternatingcurrent_ machine is equal J. The machine must. A disadvantage of the percentage system is the confusion that results from the multiplication of percentage quantities.I. from Rotor at synchronousspeed (or blocked) Zerosequence impedance. 34Connection for measuring ance. ZeroSequence Impedance ‘1’1~ zerosequence impedance is the impedance offered 10 fhc flow of unit zerosequence current. etc. This input is equal to r?li per @se.. since I.p(F+W)‘I 16. But since r. if z. Including the effect of friction and windage.iva’ent to a posltlve input. the voltage t1l. IV. The advantage of this method of representation extends to a better comparison of performance of machines of different rating.== :3 current. however.. the positivesequence terminal outpr. In this system the rating quantity is regarded as unity.. In general..c variable and depends largely upon pitch and breadth fllctors.totalshnft power due to the positiveL. be neglected. The foregoing . The departure from this ideal condiGon introduced by chording and the breadth of the phase belt determines the zerosequence reactance. Method of TestThe most convenient method for test of zerosequence impedance is to connect the three phases together. reduces to 6d:. the espression fl)r the negativesequence shaft power per phase can also Incidentally.p(F+W) 2 61. as shown in Fig. 7 heref_2re.)Ti. = rl.hcr hand.o J[r2~~+rl~f+ (~r$i] and.8 power of current. (3. regardless of size. or somenhat larger than. (34) ‘1s shown previously the negativesequence shaft po\\*r which on substituting .as in31 dicated in the illustration.ccl~l(~nce 3(rJ:+TJi). Any other amount of thequantity is expressed as a fraction of the rated amount. For this reason it is preferable to connecting the three phases in parallel.ncecurrents. In general. The zerosequence resistance is equal to. (35) where 1 is the actual measured phase circuited.~ence power output. be disconnected for otherwise no zerosequence current can flow. zerosequence imped T =p. A 100volt drop in a transmission line has no significance until the voltage base is given. from this the rotor ‘. Tvhich. E ZO'TJ Fig. Thus. of course. varies as the 1. The loss for turbine generators.T. and adding to this the copper loss due to I. The nature of the reactance is suggested by considering that. i..~~~~e input per phase is equal to endconnection fluxes. the correct answer is an gpercent voltage drop. 32(a). the total input into the alternatingcurrent nurclline from the driving tool. This connection insures equal distribution of current between the three phases. it is neglected in most calculations. Tests on nltlicnt pole machines with and without dampers verify Ihc fact that the loss varies as the square of the negativexoc’rlt. ‘l’he zerosequence reactance of synchronous machines is Wit. however.1) rctlucC3 t0 CT2. PER UNIT SYSTEM (36) 212 The foregoing neglects the effects of saturation..o’)across any one phase (starconnected) for unit current it1 ~ch of the phases. with the field short from which it follows that rz=~+rB.I: written (Tz~$$ ‘I)SSOS equal to 2(r~JIi. X3 sow from Fig. 2 . and. gives the. on the ol. a 20percent current flowing through a 40percent reactance would by simple multiplication give 800 which at times is erroneously considered as SOOpercent voltage drop.fSrphase is equal to .. T2= [p. The E zerosequence impedance is then equal to ZO=.
164
Machine Characteristics rBut and
multiplication example would in the per unit system be expressed as follows: A 0.20 per unit current flowing through a 0.40 per unit reactance produces an 0.08 per unit voltage drop, which is correct. A further advantage of the percentage and per unit systems lies in the elimination of troublesome coefficients. However, this is not an unmixed blessing as a definite disadvantage of the use of the per unit system lies in the loss of the dimensional check.
et++&
‘V’ E
& = ed xdid jxqiq
k=E cos 6jE sin 6 Upon substituting & and E in equation (37), there resul
j
[ed  Xdid jXqiq]
V. POWER EXPRESSIONS
It is frequently necessary to know the manner in which the power output of a machine varies with its excitation and internal angle. A particular application of this knowledge is the stability problem. Several simple cases will be considered. Equating reals
i,=
+ji[E
cos OjE
sin
O]

xq(x.+2c)
X&c
E
i,+
sin X0
8
17. Machine Connected through Reactance to Infinite Bus and also Shunt Reactance across its Terminals, Resistance of Machine Neglected
The schematic diagram for this case is shown in Fig. 35(a), which also shows the significance of the various symbols to be used in this discussion. The reactances x,, x0 and the one indicated by the dotted lines represent the branches of an equivalent g circuit, for which the resistance components are neglected. For the purposes of determining the power output of the generator the reactance shown dotted can be neglected. The vector diagram which applies is Fig. 35(b). The total machine current is equal to i,
=
x.E sin 0 x,xc+xqx.+xqx,
And equating imaginaries
id= x.+x0 [edxdid] x.xc (x.+xJedx.E =
X.Xc+ldX.+XdX,
kcE
cos
COS o
8
The power output, P, is equal to the sum of the products 01 the inphase components of armature current and termin& voltage, namely
P=iq(edXdid)+id(Xqiq) =ediq+ = [edi(XqXd)idiq (XqXd)id]iq
ed
t I_! ’
f
x.diA
The power is then obtained by calculating i, and id fro (39) and (40) and inserting into (41). If E and ed are e pressed in terms of rms volts to neutral and reactant ohms per phase, then the above expression gives the p in watts per phase; but if the emf’s are expressed in t of the phasetophase volts the expression gives total p On the other hand, if all quantities are expressed in then the power is also expressed in p.u. where equal to the kva rating of the machine. If ed’ rat ed is known then ed should be replaced by ed’ and wherever they appear in Eqs. (40) and (41). For the special case of a machine with cylindrical in which x, = xd, the expression reduces immediately P=edio = i,Eed sin e ~.x.2+xqx. + xqx.3
I
Id
(b)
connected to infinite actance. bus through a re
Fig.
35Machine
Another interesting special case is that for which t shunt reactance is not present or x.= co. Then
which from the internal obtain*
and external currents one can .
5 edE Sin 6+
edE
COS x +xd 0
e E

zqjid=$8+2L And inserting the equivalents of b. and ii
(Xd Xq)E2 %Xd+X.)
Sin 28 (X,+X,)
1xcfxq
sin 6
Xo+Xd
Andifx,=Oandx.=a,
*The symbol caret over a quantity indicates a phzisor quantity.
then
e+ (XdXq)E2 2XdXq Sin 28 Xd
P=
f?dE Sin
chapter 6
Machine Characteristics
165
ed
bdid id
“, I. Eo 0. 0 0 III OO 20 0 40 (II) 60 II (1 60 100 120 e IN DEGREES IIIllY 140 160 180
k$,
rid Fig. 37Vector diagram of salientpole machine effect of series resistance.
et cos e+riq+xdided=O
including
~18. 36Powerangle excitation determined power
diagram of a salientpole machineto develop rated kva at Wpercent factor. xd=1.15; x,=0.75.
/ From (45)
(46) (47)
In Fig. 36 is shown a powerangle diagram of a salientljol(: machine whose excitation is determined by loading at full kva at 80percent power factor. :\n expression frequently used to determine the maximum pullout of turbine generators is the following Pullout in kw =g(rsting of generator in kva)
iq=bq(et
sin
efrid)
Substituting
et
(47) into (46)
COS e+& 5, sin e+id+%&ed=O 2,
72
wllcre OC is the field current for the particular operating contlition and OD is the field current for the ratedcurrent zc’ropower factor curve for zero terminal voltage (see Fig. 17). ‘This expression is based upon the maintenance of mtotl terminal voltage up to the point of pullout. At pullout the angle 6 of Fig. 15 is equal to 90 degrees. Since the extent of saturation is measured by the voltage behind the Potier reactance drop, it can be seen from Fig. 15 that for 6 equal to 90 degrees this voltage is less than rated voltage, and that therefore little saturation is present. From Eq. (44) since zd =x, and 0 = 90 degrees, the pullout is @. But edis proportional to OC on the airgap line and %d Z,I is likewise proportional to OD on the airgap line. Examination of Eq. (44) shows that even if the excitation is zero (ed=O) the powerangle curve is not equal to zero, but equal to (zd$LT sin ze. This results from the d” (!llccts of saliency. Note that it disappears for uniform rrirgap machines for which zd=xQ. Advantage is somel.imcs taken of this relation in the case of synchronous condensers to obtain a somewhat greater capability in the k:ltling (underexcited) kva range. With some excitation systems (see Chap. 7, Excitation Systems) it is possible to obt.uin negative excitation. The excitation Voltage, edt in Eq. (4l) can besomewhat negative without producing an unstable powerangle diagram. By this device the leading kva range can be increased as much as 15 or 20 percent.
from which id=
r2+xdx,
1
++xd&
x’qedret sin exqet cos e
and substituting
in (47) 1 i q =red+x&
sin eret
cos 0
1 (4% 1 (49)
1 I.\
The power output, P, is equal to the sum of the products of the inphase components of i and et, or P=iqet cos 8+i& sin 8 (50) Upon substituting (48) and (49) this reduces to sin 2f3et
p=et
++xd&
c
ed(r ~0s e+x,sin 0) +‘ye,
18. Inclusion of Machine Resistance
Resistance or External
The power input into the machine is equal to P plus ri2. The expression for this quantity does not simplify and it is better to calculate it through the intermediate step of evaluating ri2, which is equal to r(ii+ii). The foregoing expressions apply to the steadystate conditions. In stability problems it is necessary to determine the average power from instant to instant. In general for this purpose it is permissible to neglect both the unidirectional component of currents and the subtransient component of the alternating current, leaving only the transient component. These latter are determined by the instantaneous value of ed’. It follows then that the power expressions are simply those derived for the steadystate condition with ed replaced by ed’ and xd by zd’.
If the machine is connected to an infinite bus through a resistance and reactance circuit, the external resistance and reactance can be lumped with the internal resistance and reactance and the following analysis used. The vector diagram for this case is shown in Fig. 37 for which et sin e+ridx&=0 (45)
VI. EFFECT OF CHANGE IN EXCITATION
Field forcing in certain industrial applications and considerations of system stability require that the voltage increase in response to a sudden need. This increase is brought about automatically either by means of the same
166
Machine Characterislics
Chapter 6
control that produced the increase in load or through the use of a voltage regulat,or. It is necessary, therefore, to be able to predetermine the effect of an increase in esciter voltage upon the output of the synchronous machine. In general, significant changes in exciter voltage never require less than about onetenth of a second to bring about the change. By the time this effect has been felt through the synchronous machine, which has a time constant of ahout a second, it will be found that the result is always slow when compared to the subtransient and unidirectional components of the transients associated with the change. In other words, variations in exciter voltage are reflected only in the transient components. As an example, suppose it is desired to calculate the armature current of a machine for a threephase shortcircuit while it is operating at no load with a voltage regulator set for rated voltage. Immediately after the inception of the short circuit there is a slight lag in the regulator until its contacts and relays close. The exciter voltage (and voltage across the field of the main machine) then rises as shown in the upper curve of Fig. 38. The bottom curve refers to the armature cur
the field circuit. The beauty of the per unit system is exemplified in the analysis of this problem. In p.u. the differential equation for the field circuit takes the following form ex= ed+ T&,dt
dr,,’
(52)
In this equation e, represents the esciter voltage or the voltage across the field if there is no external field resistor in the field circuit. The unit of e, is that voltage required to circulate such field current as to produce rated voltage at no load on the airgap line of the machine. The term edi is the synchronous internal voltage necessary to produce the instantaneous value of armature current for the given’ armature circuit regardless of what it may be. Its unit is rated voltage. It is synonymous with field current whefi unit field current is that field current necessary to produce, rated voltage at no load on the airgap line. It will be seen then that the use of ed is merely a convenient way of specifying the instantaneous field current during the tran, sient conditions; it is the field current, necessary to produce the armature current existent at that instant. As shown previously, ed’, is proportional to the flux linkages with the field winding. It is the quantity that, during the transition ! period from one circuit condition to another, remains constant. The foregoing equation has its counterpart in the more familiar forms (53) or (54) i To familiarize the reader with (52)) suppose that normal .{ exciter voltage is suddenly applied to the field winding at 4 no load. Since the armature is opencircuited ed’ and ed ; are equal and the equation can be written 4; (55) 4
LAG IN REGULATOR
+
TIME
Fig. 38Illustration showing relative importance of different components of armature shortcircuit current and response , of transient component to the exciter voltage.
When steadystate conditions are finally attained 2
4 is 1
rent, the dotted line showing the nature of transient component if there were no regulator, the exciter voltage remaining constant. The line immediately above shows how the transient component changes as a result of the change in exciter voltage. To approximately the same scale, the crosshatched area shows the increment in current caused by subtransient effects. The blackened area shows how the unidirectional component would contribute its effect. This component is quite variable and for a shortcircuit on the line might be entirely completed in a cycle or less. In any event regardless of its magnitude it can be merely added to the transient and subtransient component. It is independent of the exciter voltage.
equal to zero and ed=ex. This states that since eX= 1.0, 1 ed must also equal 1.0, that is, the excitation is equal toj the normal noload voltage. It will attain this value ex i ponentially with a time constant T&. Another esample. Suppose the synchronous machine i to be shortcircuited from noload and to be operating a without a regulator. At any instant the armature cur 3 rent, i, is equal to ed’/zd’. But since ed, which can be regarded as the instantaneous field current required to produce i, is equal to xd i, then eliminating i between these equations
xd ed = 7ed’
(56)
xd
19. Fundamental
Equation
Then equation (52) takes the form 1 =;$e,‘+T&,~
Being restricted to the transient component, the effect of exciter response can then be defined entirely by effects in
chapter 6
Machine Characteristics
167
or if it is to be expressed in terms of armature current l=xdi+T’
do ddt’
x 4e
(57)
The sustained magnitude of i, is then i= 1
xd
name, has been called by the author “the followup method.” It is a method that can be applied to any problem involving a simple time constant. To demonstrate the method, consider a simple resistanceinductance circuit to which the voltage, E, is applied. Let the differential equation for this circuit be
E=RI+L$
where the symbols have the customary significance. Dividing through by R, there results
6%
Tlrc initial magnitude of i, since ed’ remains constant during the transition and is initially equal to 1, is i= 1
xd’
(W
x @
T]rc homogeneous equation for (57) is
0=x d i+T’
The coefficient of $$ is called the time constant of the circuit and will be designated by T, giving
do d dt (58)
E %=I+T$f
E T
‘~11~s changes from 3 i constant equal to :dTio.
to i
exponentially
with a time
dI zl =
(61)
current that I
dt
T
In all problems involving a transition from one circuit c:onclition to another the one quantity (when subtransient c+fcc:t,s neglected and the time constant in the quadraare I,IIWaxis is zero) that remains constant within the machine is the flux linkages with the field winding, which in turn is reflected in the quantity ed’. It is necessary, therefore, to calculate ed’ for the circuit condition preceding the transition. All the discussion of the following cases assnmes that this point is understood and that ed’ is known for the beginning of the transient period. Several cases will now be discussed.
In this expression E is the steadystate
approaches for the instantaneous value of E. I is the instantaneous magnitude of current. If the current at any instant is plotted by the point a (Fig. 40) and the corresponding value of E for that instant is plotted as the point b (Fig. 40) displaced horizontally by a time T,
E
20. Threephase Short Circuit of Unsaturated chine
Ma
The problem is to determine the transient component of shortcircuit current in response to the exciter voltage gifen in Fig. 39. This is most quickly and conveniently found by a graphical method, which, for want of a better
Fig. IOConstruction derivation of “Followup Method.”
then the vertical distance between a and b gives the numerator of (61) and the horizontal distance the denominator. The slope of the line between a and b is equal to ‘i. If an interval of time At is chosen following the instant under consideration and E is assumed constant over the interval then the change in I during the interval, AI, is equal to At.
dI dt
The final value of current for the interval
is then given by the point c. If  at an instant Al later is then plotted and the line drawn from c then the value for Following such procedure it is possible to construct the complete curve for I. The construction is illustrated in Fig. 39, in which
E R
AI for the second interval is obtained.
Fig. 39Illustration
of “Followup
Method.”
168
Machine Characteristics i=1
Chapter 6
the curve marked lZ is the instantaneous magnitude of E from time t=O. Plot g displaced to the right a time T.
Starting from this value the actual magnitude of xd” i is obtained as a function of time.
Let li be the initial value of I at t = 0. Divide the time into intervals of length At. Draw the line ab, then cd, ef, etc. The accuracy will be greater the smaller the intervals and can be increased somewhat for a given element width by using I: instead of T for the distance by which the
21. Unsaturated
Machine Connected to Infinite Bus
steadystate curve which I tends to approach, is offset horizontally. Xow returning to the problem in hand. The differential equat,ion governing the case is given by (52). The esciter voltage e, is assumed given and expressed in p.u. For a threephase short circuit at the terminals of the machine ed is equal to zdi and ed’ =zd’i. Therefore Eq. 52 becomes
e,=sd+dT;o$
Dividing through by zd
(62)
2 =;+‘L
Xd xd
,
di TAO,
(63)
The construction
dictated
by this equation and the
xd
followup method is shown in Fig. 41. 2 is plotted against
As stated previously the subtransient and unidirectional components of current are not of importance in the stability problem. For this application it is desirable to determine how cd’ varies as this influences the power output of the machine and in turn dictates the degree of acceleration or deceleration of the rotor. The circuit shown in Fig. 35(a) is typical of a setup that might be used for an analytical study to determine the effect of exciter response in increasing stability limits. Another case of considerable importance is the action of a generator when a heavy load,* such as a large induction motor, is connected suddenly across its terminals or across the line to which is is connected. In starting the motor the line voltage may drop an excessive amount. The problem might be to determine the amount to which this condition could be ameliorate‘d by an appropriate excitation system. Since reactive kva is more important than the real pcmer in determining regulat’ion, the motor can be represented as a reactor and the circuit in Fig. 35(a) utilized. Having determined the manner in which ed’ varies, the power in the case of the stability problem and t,he terminal voltage (ed’rd’i) in the case of the voltage problem, can be calculated easily. Equation (52) must be used again to determine the manner in which ed’ varies in response to changes in exciter voltage and phase position of the rotor with respect to the infinite bus. The instantaneous armature current can be found in terms of the rotor angle 19 and ed’ by replacing ed and zd of Eq. (40) by ed’ and xd’, respectively, giving
id= (XsfXc)ed’XsE X.&+zd’x,+Xd’Xc COS 0
(64)
The synchronous internal instant to and upon substituting (64)
voltage, cd, is equal at any
ed=ed’+(XdXd’)id
(65)
=
(XsXc+XdXs+XdXo)ed’XJ;S(XdXd’)E XsXc+Xd’Xa+Xd’Xc
COS 8
Substituting
e,=
this expression in (52), there results.
X,Xc+XdXs+XdX, &X,+Xd’X.s+Xd’Xc X,(XdXd’) &X,+Xd’X.+Xd’Xo ed’
E cos 13 +T;ods
(66)
which can be converted to
x, (Xd Xd’)
Fig. 41Transient
component of shortcircuit influenced by excitation.
current,
i’, as
E cos d=edr+Tdrd$
(67)
XsXcfXdXafXdXc
in which from referTd’ = x&+x,,‘&,+xd’&
XsXofXdXsfXdXc
time, its zero being displaced an interval ‘$Ti,
Tdo’
(68) i
transient time
ence zero. The initial value of i is determined through ed’ which was 1.0 at t =O. This makes the initial amount of
The time constant Td’ is the shortcircuit constant.
Chapter 6
Machine Characteristics
169
If 0 were constant or if its motion as a function of time ,vere known then the whole lefthand side could be plotted (displaced by the time Td’) and treated by the followup method a~ the quantity that ed’ tends to approach. Unfortunately 0 is not in general known beforehand, and it is ncccssary t0 calculate 0 simultaneously in small increments in a simultaneous solution of cd’ and 0. The magnitude of 0 is determined by the electromechanical considerations discussed in the chapter dealing with System Stability. In solving for ed’ a progressive plot of the lefthand side can he made or (67) can be transformed to the following form
From this point the followup method can be used as before. After ed’ is determined as a function of time any ot,her quantity such as terminal voltage can be obtained readily,
23. Saturation
In analyzing transient phenomenon of machines in the unsaturated condition, the theory was built around the concept of the transient internal voltage, ed’, a quantity evaluated by using the transient reactance, xd’. In the presence of saturation it was found that for steadystate conditions by the introduction of the Potier reactance: x, (see Sec. 3) the proper regulation was obtained at full load zero’powerfactor. The use of x, and e,, also resulted in satisfactory regulation for other powerfactors. In extending the analysis into the realm of transient phenomenon. ep will continue to be used as a base from which to introduce additional mmf into the field circuit to take care of saturation effects. The treatment will follow quite closely the same assumptions as were used in determining the steadystate regulation according to the TwoReaction Potier Voltage method of Sec. 3(d). With this assumption the fundamental Eq. (52) for the field circuit becomes e,=ed+(s due to e,)+Td,,% (761
and the increment calculated from the equation
;\ shunt resistancereactance load such as an induction motor is not much more difficult to solve numerically but t,lr(:expressions become too inv&ed for analytical solution. lt is necessary only to calculate id in terms of ed’ and 8 just :~swas done before and then follow the same steps as used for the reactance load.
22. Unsaturated Machine Connected to ResistanceReactance Load
.\ case not too laborious to carry through analytically As before ed represents, neglecting saturation, the voltage is that for which a resistancereactance load is suddenly behind the synchronous reactance of the machine or what applied to a synchronous machine. Let reXtand xext be the is equivalent the field current required to produce the cstcmal resistance and reactance. The addition of a sub instantaneous ed’, including the demagnetizing effect of script t to machine constants indicates the addition of T,,~ the instantaneous armature current. The total field curor .crutto the respective quantity. The equations of Sec. 17 rent is obtained by adding s to ed. In some cases it is found I hrn apply to this case, if et in the equations is made equal simpler to convert all of the right hand side to the single variable e,, but in others it is simpler to retain the variable to zero and Ed replaced by Zdt, etc. Following the same procedure as previously, there re in the form of ed’. Two applications of this equation will Slllls from Eq. (48) when ed and zd are replaced by ed’ be discussed. Machine Connected to Infinite BusThe circuit :ultL z& and et is equal to zero. shown in Fig. 35(a) is the one under discussion and for X,t id= ed’ (71) which Eq. (66) applies for the unsaturated condition. This r:+xhqt equation can be expanded to include saturation, in accord‘l’lrc field current or its equivalent, the synchronous in ance with Eq. (76), to the following tcrnnl voltage, is then %&+xdx.+xd& x&dd’) e,= ed’ E cos 8
ed=ed’+bdtxi&d &,~c+~d’~.+~d’& xqt =ed’+(xdtx&j r’t+&xqt ed’ &&+~d’~,+~d’~o
(72) Substituting this expression in (52) there results that ex=
xdtxqt+< 2 ed’+ x;txqt+rt TL
ded’ dt
jvhich can be converted to
Td
~‘k=ed’+Td$ do
,ded
(74)
in Which Td=
xd’txqt+$ Td6. xdtxqtfr~
+(s due to eP)+Ttdd do dt ’ This can be converted to x&dxd’) E COS 6 ded’ e, (s due to e,) d (78) T’do iii= + (x.xc+ xd&+xdxc) Td’ Td in which Td’ is defined from Eq. (68). Before (78) can be used it mill be necessary to determine ep in terms of ed’. The components of current, i, and ‘id, can be determined from (39) and (40) by replacing ed, by ed’ and xd by zd’. Thus xsE sin 0 i,= (7% x.xo+2&+2q2,
id
(75)
(X.+X&d’GE &&+xd’x,+xd’&
COS e W) ’
170 The directaxis component of eP is equal to e*d= Q’  (Xd’ x,)i,,
Machine Characteristics
Chapter 6
assumption that is usually made in stability studies tha where quick response excitation is installed, ed’ may b regarded as constant.
Machine
and the quadratureaxis component of e, is epq= bk4G sin 0 = x,(x,z,)E wLfZ,~,f5,Z, The amplitude of e,, is then equal to e,=dK
Connected
to
ResistanceReactant
LoadThis case is the same as that considered in Sec. 2 except that saturation effects are to be included. Upo including the saturation term s into Eq. (74) there result that (82) in which (8t It is well to recall again that this analysis neglects sut transient effects and assumes that the time constant in t1 quadrature axis is zero. If in Eqs. (48) and (49) et is mad equal to zero, ed is replaced by ed’ and the correspondin changes in reactance associated with ed’ are made, and i addition the subscripts are changed to indicate total r~ actances, Then
(83)
While this quantity does not simplify greatly, it does not appear so formidable after numerical values are inserted. e,, can thus be calculated for any instantaneous value of e,,’ and the s corresponding thereto substituted in Eq. (78). Equation (78) provides a means for computing increments of change in ed’ for use in stepbystep solution. Thus
As s becomes small and saturation effects disappear, the solution relapses into the same type as used when saturation is negligible (Eq. 66), for which the followup method is frequently applicable. The relations just developed are useful in estimating the extent to which ed’ varies in system stability problems. Fig. 42 shows the results of calculations on a system in which a generator is connected to a large network, represented as an infinite bus, through a reactance equal to j0.6.
xiao.45 Xp~0.40 Xq’0.65 xd= 1.1, T&,=6, / 5 E0.91 / / / >! a
The total current is then (B! The voltage eP is ep= id
,+=3.0
% 6
Uljon substituting ed’ from (90) into (85) and using (81 also there results that
d(s;t+r,‘hl:+r:+S x due to e
Zdt&,t+T:
P
)
1.0
I.16 0 a I.14 0.5 1.0
0
0.05
0.05
0 I.121 0 ’ ’ 0.1 ’ ’ 0.2 TIME ’ ’ 0.3 ’ ’ 0.4 ’ J 0.5
IN SECONDS
Fig. 42Effect of rate of response upon ed’ as a linetoline fault represented by the threephase shunt load j0.2 is applied to generator which had been operating at 90 percent powerfactor. 20 percent of airgap mmf required for iron at rated voltage.
A linetoline fault is assumed applied to the connecting transmission line on the high tension bus at the generating end which is equivalent to a threephase short circuit through a reactance of j0.2 ohms. The curves justify the
As can be seen from Fig. 43 the solution of this equatic lends itself well to the followup method. On the righ hand side the assumed exciter response curve, e,, is plottl as a function of time. Multiplying this quantity by tl coefficient of e,, the term epmis obtained. This is the vail eP tends to attain if there were no saturation effects. 1 in the followup method, the zero of time from which tl instantaneous curve of e, is drawn, is displaced to the 1~ an amount Td’ minus half the interval of time chosen the stepbystep solution. Along the ordinate of ePa cur s1equal to the second term is plotted in which s is obtainl from the noload saturation curve shown in (b). For al instantaneous value of ep, s1 is plotted downward from e as the construction progresses. So starting from the initj value of e,, of which more will be said later, a constructit line is drawn to a point for which s1 was the value cork sponding to the initial value of eP. For the second inter\
Chapter 6
Machine Characteristics
xq i cos 0 et+xqi sin 4
tonb=
L
MACHINE CONSTANTS
OVERALL CONSTANTS
xPl=xP+xexl *it= XL +xext
XP
Xb Xd
xdt’ *qt=
Xd+Xext *q+xert
xq
1.0 FIELD CURRENT OR EXCITER VOLTS
(c)VECTOR DIAGRAM FOR MACHINE ALONE
r
rt. (a)DEFINITIONS
r +reXt
(b)SATURATION CURVE OF MACHINE
INITIAL VALUE OF ep
/
s z
s WHEN rt =o
CALCULATE $j FROM (Cl FOR INITIAL LOAD eb REMAINS CONSTANT DURING TRANSITION FROM ONE OPERATING CONDITION TO THE NEXT
5, FOR INTERVAL”2
INITIAL VALUE OF eP= 5, FOR INTERVAL”
1
l,At
= I
!I
!
I
’ = $T&u TIME
AAt’2 WHEN r,=o
!
0
0
TIME

Fig. 43Graphical
determination
of terminal
voltage
as polyphase applied.
series resistances,
rext and reactances
xext are suddenly
sudden disconnection of a loaded generator from the bus throwing its load upon the remaining units or the starting of an induction motor by direct connection to a generator. For the latter case, if the capacity of the induction motor is a significant fraction of the kva of the generator, a severe drop in voltage results. Thus a 500hp motor thrown on a 3300kva generator produces an instantaneous drop in et=idx2cxt+~2ext voltage of the order of 13 percent. The effective impedance of the induction motor varies with slip and to be rigorous this variation should be taken into consideration. It is W(l substituting ed’ from (90) usually sufficiently accurate to use the blocked rotor reactance for the motor impedance up to the speed correset= X2& +r*,*t (92) ponding to maximum torque in calculating the factor which ep 4 x&+ri determines ePa, terms of e,. Beyond the slip correspondin ‘lhis permits of the calculation of et from eP after the ing to maximum torque, the effective impedance varies ecristruction has been completed. rapidly to the running impedance. Simultaneously with 1)uring the transition from one operating condition to the increase in impedance the lagging kva likewise drops th nest, only ed’ remains constant; eP changes. It is es off which results in a considerable rise in voltage. This “crrli:~l therefore that ed’ be computed for the initial operat effect is clearly shown in Fig. 44 taken from some tests ing condition. The conventional construction shown in made by Anderson and Monteith.*O As running speed is l’i!& 43(c) can be used. This determines the initial value approached the generator voltage rises, the excitation becf cd’for the nelv operating condition from which the initial ing too high for the particular loading. To form a better “rrlrie of ePcan be computed by Eq. (90). idea of the magnitudes involved in such calculations, Fig. Common cases for which these calculations apply are the 45 shows curves of terminal voltage as an induction motor determination of regulation for loads suddenly applied to equal in horse power to 20 percent of the kva of a generator a generator. Instances in which this can occur are the is suddenly thrown upon an unloaded generator for differ
SIis taken for the value of F, at the end of the first interval or, to be slightly more accurate for an estimated average v:~lueof ePfor the second interval. And so the construction proceeds. 1%~ same reasoning whereby ePwas obtained in Eq. the (W)the terminal voltage et can likewise be obtained, giving
172
3500 3 100 2700 : 2 2300 P 1900 1500 1100 1200 1000 600 Iz = 600 5 O400 200 0
Machine Characteristics
240 200 I60 I20 % 2 2 ? : uz Fzj i
Chapter t
80 F P 40 (I E 03 E
Fig. 44Performance of 3333 kva, 0.6 powerfactor, 3600 rpm, 1.7 shortcircuit ratio generator as a single 500hp inductionmotor pump is started. Inductionmotor starting torque equal to fullload torque and pullout torque equal to 2.8 fullload torque. Full lines represent operation with fixed excitation and dotted lines under regulator control.
70
0
0.2
0.4
0.6
0.6
1.0
TIMESECONDS
Fig. 45Terminal voltage of a 500 kva, SOpercent powerfactor enginetype generator Cxd= 1.16, xq =0.59, xd’ = 0.30, 13 percent saturation) as a lOOhp induction motor is connected.
ent rates of response of the exciter. Ordinarily one is primarily interested in the minimum voltage attained during the accelerating period and so the calculations have been carried out to only 1.0 second. The curves show conditions for constant excitation and for exciters with 0.5 and 1.0 ratios, respectively.
24. Drop in Terminal Applied Loads
Voltage
with
Suddenly
problem generally and have plotted the results in curvl form. The analysis has been carried out for both selfescitec and separatelyexcited exciters. The results for the forme are plotted in Fig. 46, and for the latter in Fig. 47. Thes, curves are plotted in terms of the four parameters: (I magnitude of load change (2) XA,,, (3) Td,, and (4) rat, of exciter response, R. The response is defined in th, chapter on Excitation Systems. It is shown by Hnrde and Cheekz2 that variations in xq, saturation factor of th, generator and power factor between zero and 60 percen have little effect upon the maximum drop. The assumef value of zd for these calculations was 120 percent. AI accurate figure for maximum voltage drop can be obtainec for values of %dother than I20 percent by first expressin reactances and the applied load on a new kva base, sucl that xd on the new base is 120 percent, and then applyin the curves. For example, suppose a load of 1500 kva (ex pressed at full voltage) of low power factor is to be applie, to a 3000kva generator having 30percent transient re actance and 150percent synchronous reactance. Suppos that the generator time constant is 4.0 seconds and th exciter has a nominal response of 1.0. To determine th drop, express the transient reactance and the applied loal on the kilovoltampere base upon which xd is 120 percent The base in this case will be 3000~ 120/150 =2400 kva On this base the transient reactance 54’ is 30 X 2400/3000 = 24 percent, and the applied load is 1500/2400=62.5 per cent. If the exciter is selfexcited then from the curves a Fig. 46, the maximum voltage drop is 15 percent for 62.5 percent load applied to a generator having 24percent tran sient reactance, a time constant of 4.0 seconds, and al exciter of 1.0 nominal response. This same maximum drol would be obtained with the machine and load unde consideration. The initial load on a generator influences the voltag drop when additional load is suddenly applied. As show] in Fig. 48, a static or constantimpedance initial load re duces the voltage drop caused by suddenly applied loac However, a load that draws additional current as voltaf decreases may increase the voltage drop. Such loads ti be referred to as “dynamic” loads. For example, a runnir induction motor may drop slightly in speed during tl voltage dip so that it actually draws an increased currer and thereby increases the maximum voltage drop. Tl dynamic initial load curve of Fig. 48 is based on an initi; load that draws constant kilowatts and power factor as tf voltage varies.
When a relatively large motor is connected to a generator, the terminal voltage may decrease to such an extent as to cause undervoltage release devices to operate or to stall the motor. This situation arises particularly in connection with the starting of large motors on powerhouse auxiliary generators. The best single criterion to describe this effect when the generator is equipped with a regulator to control the excitation is the maximum drop. The previous section describes a method whereby this quantity can be calculated. However, the problem arises so frequently that Harder and CheekzzJ3 have analyzed the
VII. CONSTANTS FOR USE IN STABILITY PROBLEMS
The stability problem involves the study of the electrc mechanical oscillations inherent in power systems. 2 fundamental factor in this problem is the manner in whit the power output of the generator varies as the position ( its rotor changes with respect to some reference voltage The natural period of power systems is about one seconc Because of the series resistance external to the machine the time constant of the unidirectional component of am: ature current is usually so small as to be negligible i
Chapter 6
Machine Characteristics
I ! ! ! ! ! !tfvv!r
Fig. 46Maximum voltage drop of a synchronous machine WITH SELFEXCITED EXCITER tude of load change, (b) x’,, Bst, (c) T’do and (d) rate of exciter response. xd’ on curves refer to value. Assumptions used in calculations: xd’= 1.07 x d’ aat; xd=I.20; x,=0.75; noload saturation voltage = 1.2; time lag of regulator =0.05 second; added load is constant impedance of 0.35
as affected by (a) magnisaturated or ratedvoltage curve/air gap line normal pf.; initial load zero.
Chapter 6
Fig. 47Maximum
voltage drop of a synchronous (a) magnitude of load change,
machine WITH SEPARATELYEXCITED (b) x’d .,,,t. (c) Tdo’ and (d) rate of exciter
EXCITER response.
as affected
by
Chapter 6
Machine
Characteristics
175
Further consideration of this constant is given in Part XIII of this chapter.
26. Network Calculator
Studies
INITIAL
LOADPER CENT
u?
Fig. 48Effect of type (whether dynamic or static) and initial load, assumed at 0.80 power factor, upon the maximum voltage drop when loopercent lowpowerfactor load is suddenly applied to an ac generator.
For most problems the synchronous machine can bc represented by its transient reactance and a voltage equal to that behind transient reactance. For the rare case for which salientpole theory is required, the follo\ving procedure can be followed. It is impossible to set up the tno renctances in the two axes by a single reactor, but if the reactance, xQ, is used and a new voltage, cQ&introduced as representing the internal voltage, both position of the rotor and the variations in ed’ can be carried through quite simply. Fig. 49 shows a vector diagram similar to Fig. l1 in which epdis included. This voltage is laid off along cd and ed’ and terminates at the point a. The reading of power at eq,J the same as the actual output of the machine. 1s is the exciter voltage changes ed’ and esd likewise change.
comparison with this natural period. The subtransient component is likewise so small that its effects can be ncglccted. There remains then only the transient coml)oncnts, those components associated with the time constants of the field winding, that are important.
25. Representation
of Machine
The transient stability problem is primarily concerned with the powerangle relations during system swings folloiving a disturbance. Because of the dissymmetry of the two axes, it is necessary theoretically to take this dissymmetry into consideration. However, in most cases an impedance is in series externally to the machine so that the difference in reactances in the two axes becomes a smaller proportion of the total reactance. The results of c::rlculations presented in Chap. 13 show that for most practical purposes it is sufficiently accurate to represent
the unsymmetrical machine having the same xd’. with a symmetrical machine
Fig. 49construction
of eQdfor network
calculator
studies.
The incremental changes in epdcan be obtained as follons. From Fig. 49 it is evident that at any instant eqd= ed’+ (x9xd’)id (94) From Eq. (52)
ded’
$=z,(&ed)
In spite of the close agreement of salientpole with cylindricalrotor results, a few cases arise for which it is rrccessary to use salientpole theory. Relations for calcul:r( ing the power output have been given in Sets. 10 and 17 :md for computing the change in internal voltages in Sec. 22(:1). It is shown in the latter section that if the exciter is of the quickresponse type, the voltage ed’ can, for all lmrctical purposes, be regarded as constant. Methods for lhc inclusion of these factors into the stability calculations have also been treated in Chap. 13. 11knowledge of the inertia constant, H, is a requisite for tl~ determination of the acceleration and deceleration of Ihe rotor. It represents the stored energy per kva and can 1~ computed from the moment of inertia and speed by the follo\ving expression H=0.231WR2(rpm)2106 kva \Vtlere H = Inertia constant in kwsec. per kva. WR2= Moment of inertia in lbft2. (93)
1
and
where Aed’ is the increment of ed’ in the increment of time At. From Fig. 49 there results also that
ed=epd+ (xdx&d (96)
so that
1 r
exee,d((2dX&d
1 &.
(97)
In network calculator studies of system stability, e,, e,,d, and id are known at any instant. From Eq. (94) it is evident that the increment of egdis equal to the increment in ed’. Thus exee,d((Zdxp)id Ae,d=&L cl0L
1
At 1
(98)
176
Machine Characteristics
Chapter 6
This method can be applied regardless of the number of machines involved in the study. To obtain the initial value of eqd, calculate ed’ from the steadystate conditions before the disturbance. cd’ is the quantity which remains constant during the instant representing the change from one operating condition to another. The proper eqd is obtained by changing the magnitude of eQduntil Eq. (94) is satisfied. To include the effect of saturation, break the reactance xq, which represents the machine, into two components 5, and (z,zP), the latter being next to the voltage e,d. The voltage at the junction of these two reactances is eP, the voltage behind x,. The effect of saturation will be included by adding the saturation factor s taken from the noload saturation curve (see Fig. 17) for eP,to the excitation obtained by neglecting saturation. This corresponds to method (d) of Sec. 3 for steadystate conditions. Eq. (98) then becomes Aeqd=$
do
of unidirectional and negativesequence currents. The crest value of the negativesequence current, iz, is fiiz and the crest value of i” is G I,’ In the case of neztivesequence currents, part of the loss is supplied by the shaft and part is supplied through the armature. The loss associated with the circulating currents in the rotor as developed in Section 15 is approzimutely equal to 2(rzrl)i$ Assuming for the moment that the loss varies as the square of the current and neglecting the differences due to the frequencies in the two cases, the loss for the unidirectional components of current is
II
e,e,d(xdx,&s
1
At.
(99)
Actually, however, the loss varies more nearly as the 1.8 power of the current so that the expression becomes \
27. Armature
Resistance
For most stability studies the loss associated with the resistance of the armature is so small as to be negligible. The exception to this rule is the case for which a fault occurs near the terminals of a generator. The losses in an ac generator during a threephase short circuit can be large enough to affect significantly the rate at which the rotor changes angular position. This is of particular importance for stability studies. Two of the most important factors determining this effect are the location of the fault and the value of the negativesequence resistance. The latter is difficult of evaluation particularly for turbogeneratorsthe type of machine in which the effect is greatest. One must rely almost entirely upon calculations, which are extremely complicated. For ac board studies of system stability it is convenient to represent the machine losses by means of a resistance placed in series in the armature. The value of this resistance should be chosen so that its loss, with the reactance of the machine representcd by xdl, be equivalent to that of the machine under actual conditions. An approximate evaluation of this equivalent resistance will be developed for a turbogenerator. Let the initial value of the subtransient component of shortcircuit current be designated, i”. The components of the unidirectional current have a maximum value fii” and are related in the three phases in a manner as discussed in Sec. 8. The sum of the unidirectional components in all three phases produce anessentially sinusoidal wave of mmf that is stationary with respect to the armature. This stationary mmf develops a flux that in turn generates currents having a frequency of 60 cps in the rotor. This effect is similar to that produced by negativesequence currents in the armature except that the latter produce a sinusoidal wnf wave that rotates at a speed corresponding to 60 cps in a direction opposite to the rotation of the shaft and ultimately generates circulating currents in the rotor having a frequency of 120 cps. The magnitudes of the mmf waves in the two cases are equal for the same crest values
Now considering the effect of frequencies. Since the depth of current penetration varies inversely as the square root of the frequency, the resistance varies directly as the square root of the frequency. The loss for the unidirectional component is then (101) j
36 
3.77
I
j
i
1 I
2.6 : I 2.4
u)
I
.I4
.I2
$2.0 J
.I0
TIME
IN SECONDS
Fig. 50Development of req of a turbogenerator for the CC dition of a threephase short circuit across the terminals the machine for various duration of the short circuit.
(;hapter 6 The loss associated with this current is
177
To form an idea of the order of magnitudes of these losses, let x:‘=o.oo. T,=0.09. Ti=0.6. x:=0.15.
x3=1.25. T2=0.035. ?I 0.005. =
The results of the calculations are shown in Fig. 50. The upper dashed curve is the loss associated with the unidirerFtiona1 component and the lower dashed curve the load losses. The full line represents the total losses. The current flowing in the generator as represented on the board 1 is constant and equal to 7. The equivalent resistance,
xd
_Oo .05 JO
WITHOUT DAMPERS 1 I h .I5 (a)
.20
.25
.30
.35
TIME IN SECONDS
req, to be inserted in series with xd’ must be such that the integrated loss over any interval must be the same as that in Fig. 50. The dotdash curve in Fig. 50 gives the values of I,~ obtained by this method. Figure 51 gives similar values of req for other types of machines. The curves in Fig. 51(a) were calculated for short circuits at the terminals of the machines, those in Fig. 51(b) are for threephase short circuits across the terminals of a transformer connected in series with the machine. VIII. UNBALANCED SHORT CIRCUITS ON MACHINES WITHOUT DAMPER WINDINGS Because of the dissymmetry of salientpole machines without damper windings, the armature currents at times of threephase shortcircuits, as shown in Sec. 12, contain secondharmonic components. For unsymmetrical shortcircuits, such as from terminaltoterminal, the wave forms of currents and voltages become even more complex. Both odd and even harmonics are present.
types
TIME (b)
IN SECONDS
Fig. 5lTypical
equivalent resistance, of machines.
res, for different
(:L) for threephnse short circuit across the terminsls used (1)) for threephase short circuit across the terminals of a seriesconnected transformer of 10 percent impedance. Since the unidirectional with a time constant
~/z(rz
28. TerminaltoTerminal Short Circuit In particular consider a salientpole machine in which saturation is neglected and which is operating at no load to which a shortcircuit is suddenly applied across two terminals. The shortcircuit current5 in these phases is then i=(x,+xd’)+(xqxd’) cos2(!hft+$O)
current
1.81 7l) eE
decreases exponentially
T,, the loss as a function of time is
(102)
(105)
(%I”) l.B
*
ln addition to the losses associated with the unidirectional current, the load losses as reflected by TI can also be significant for a threephase fault across the terminals. xcglecting the subtransient component, the ac component of shortcircuit for a threephase short circuit from
noload is
in which &, indicates the phase position during the cycle at which the shortcircuit occurred. It will be observed that this can be resolved into two components V3 It sin (27rfL+& (106) First: (xq+xd’) + (xqxd’) cos 2 (2?rftf4o) 4 If sin & (107)
Second: (xg+xd')+(xqxd')
cos 2 (2rfl+$O)
[(+$&+$I.
(103)
The first component is shown in Fig. 52(a) for a typical machine and consists of odd harmonics only. The second
Machine Characteristics Thus &eadystate = &II xd’+d5gG x,1 d\/2skdl +
Chapter 6
sin (2~jL+&) (108) [(xq+xd’)+(xqE,,‘) 00s 2 ca?rfl+)] With the assistance of Fig. 52 it Gil be seen from Eq. (105) that the maximum amount of the odd harmonic component is equal to e. The masimum value of
the total current is dependent upon the instant during the cycle at which shortcircuit occurs and reaches a maximum of G. Xd’ Assuming no decrement for either the odd or evenA harmonics
TOTAL
++ARMATURE CURRElYT FIELD CURRENT
271 4
Fig. 52Armature current and field current in a synchronous machine when a terminaltoterminal short circuit is suddenly applied.
Xd’ = 0.30
3 2 I 0 I 2 3 4 2 Y 4 5 g0 I 5 4 3 2 I 0 I 2 ’
xd=l.l
x,=0.75
A,=90”
component is sholvn in Fig. 52(b) for &= +90” and consists of even harmonics only. The latter component is dependent upon the instant during the cycle at which the shortcircuit occurs and may vary anywhere between the values given and the negative of those values in accordance lvith the coefficient, sin 40. Figure 52(c) gives the total current, the sum of Figs. 52(a) and 52(b). The units chosen are the p.u. in which for the machine operating at noload at rated circuit voltage If would be equal to 1.0 and in this case the current i is given in terms of crest magnitude of rated phase current. The components of armature current shown in Figs. 52(a) and 52(b) have associated with them the field currents shown in Figs. 52(d) and 52(e), respectively, the former consisting only of even harmonics and the latter only of odd harmonics. In Fig. 52(f) is shown the total field current. The average magnitude of this current is equal to xd+lhzz If. Xd’t~ The oddharmonic component of field current and its associated even harmonic in the armature decay to zero with time. The even harmonics of the field and their associated odd harmonics of armature current decay to constant, steadystate amounts. Their initial values are in excess of their steadystate magnitudes by the amount the average of 1f is in escess of its steadystate amount, If. The steadystate value of i is then equal to the initial amount of the oddharmonic component multiplied by xd’+dgz Xdfd/Z,Zdl
(bl
\
I
\
(Cl
Fig. 53Wave form of voltage across terminals of a waterwheel generator without damper windings for a terminaltoterminal short circuit from noload. xq/xd’ =2.5.
(a) Initial
value of odd harmonic component (decays slowly); (b) initial value of even harmonic component for sin @,rl (decays rapidly). Its magnitude varies between that given and ita negative dependingupon the point during the cycle at which short circuit occurs. It may be zero. (c) Total initial value for sin GO= 1
Chapter 6
Machine Characteristics
179
Soa28+4b2cos40+6b3cos66+4
The rms total current is equal to t,hc square root of the Slun of the squares of those components. It must be remembered that the unit of current is the crest of rated terminal current. When expressed in t.erms of the rated ,.,I1,y current the above figures must be multiplied by ~‘2. ‘The voltage from the shortcircuited terminals to the froc terminal, neglecting decrements, is equal to ,~,,P,,=Q,= 3frK[sin(27rfl+&)+3bsin3(2?rfl+&) +5b2 sin 5(27rfl+&)+. .] +3If sin +[2b cos 2(27rfI+&)
+4bZ cos 4(2~rfl++) + .]
m[L+eb
v,
e31dK[sin0+3bsin36+5b2sin56+3 1 +3Id sin OC[Z!b cos 2 8 t4 b2 Cos 4 6 +I /T i
ClW
Fig. 55Equivalent&rcuit to which Fig. 54 may be reduced.
in which K=G7z
dGJxd’+ 1 (113)
b=%G7z1 1/x&d’+ 1
ant1 h has its previous significance. I,ike the shortcircuit current this voltage can likewise bc resolved into two components that together with the t,otnl voltage are plotted in Fig. 53. The maximum possible voltage, that which occurs when sin & is equal to unity, is 23l (114) ( ) When sin &=O, the even harmonic component is equal IO zero and for this case the maximum voltage is
eab(maxlmnm ror max. OuX Ilnkagea) 
iIf
where n represents the integers 1, 2, 3, etc., and also the order of the harmonic. The nature of this resonance phenomenon is illustrated more clearly by the curve of Fig. 57, in which is plotted the maximum voltage during shortcircuit in per unit. To orient one’s self with regard to the length of line 3If [cos (2njt + 4”)  cosr$o] involved in these considerations, the figure in miles which (116) appears below each oscillogram of Fig. 56 represents ap‘=(Zd’+S,,+Zg)+(Zd’ZR) COS2(27Tjt+I$o) proximately the length of singlecircuit 66 or 220kv 21. Unsymmetrical Short Circuits Under Capa transmission line that, with a generator having the charcitive Loading acteristics of the one used in the test, is required to satisfy the given value of xc/a. These figures were arrived When a salientpole machine without damper windings is loaded by a highly capacitive load,12v there is danger, at by assuming a generator capacity of 25 000, 75 000, and l3 200 000 kva for 66, 132, and 220kv lines, respectively. For smaller machines the length will decrease in proportion. The possibility of the existence of such resonant conditions can be determined for other types of loads and other types of faults by setting up the network for the system and replacing the machine by the reactance m. This circuit should be set up for the positive, negative, and zerosequence networks and the networks connected in accordance with the rules of symmetrical components. Any condition for which the impedance as viewed from the machine is zero or very small should be avoided. Since the danger of these high voltages arises from the ttissymmetry of the machines, it can be eliminated effecFig. 54sch ematic diagram of a threephase, salientpole of damper windings. Fig. 58 alternator to which a threephase bank of capacitors and a tively by the installation terminaltoterminal short circuit are applied simultanepresents oscillographic evidence of the voltages existing ously. for machines equipped with different types of dampers as
‘l’hc corresponding linetoneutral voltages for the termin:Jtoterminal shortcircuit are ?,$ of the above figures. In all of these expressions the crest value of rated linetonclltral voltage has been used as a base. When the rms ligure is used, the above quantity must be multiplied hy a. For a terminaltoneutral short circuit, neglecting decrrments, the shortcircuit current is
at times of unbalanced short circuit, that resonance occur between the reactance of the machine and the load with the possibility that dangerously high voltages might result. Considering a purely capacitive load such as an unloaded transmission line, the schematic diagram is shown in Fig. 54 and the equivalent circuit in Fig. 55 for the condition of a terminaltoterminal short circuit. The emf applied to the circuit is equal to the opencircuit voltage for the same shortcircuit condition. The oscillographic results of tests made on a particular machine as terminaltoterminal short circuits are applied for different amounts of connected capacitance are shown in Fig. 56. Resonance will occur near points for which the quantity *=n2,
.
Machine without damper winding. Nonconnected DampersThe dampers in each pole i:lcp:~r~ inclepcndent from those in adjacent poles. 3k Types of Damper Windings ILmper windings are of several general types. a nonc*onnccted damper minding having a ratio of s. (d) XonConnected copperdamper. ‘rllcy are continuous between poles as shown in Fig.terminal short circuits and capacitive react:m(‘(‘s arc applied simultaneously.ion of copper damper Tvindings to machines ~Ncctivcly simplifies the characteristics of the machines :IS vicw&l estcrnally in that harmonic effects are largely c4irninated. xq” and Q” have nearly thc same magnitudes. equal to all :II.2. Terminaltoneutral short circuit: (e) So damper. circuit: Terminaltoterminal short (8) X0 dampers. 59 in \Gh (a) sholvs t. (b) Connected copper damper. Connected DampersThese consist essentially of \viritlings similar to a squirrelcage or an induction motor. . %Effect of damper windings.35 will be found adequate for practically ~)llrposcs. in Fig.he connection between poles for a SLOW~cd machine and (b) shows the additional bracing rewirccl in the form of an end ring for higher speed mac*llincs. as shown (h) Fig. I. ‘1‘1~~ ir~fIllcnc:e of dampers can in most cases be evaluated in 1wrns of their effect’4 upon the subtransient reactances hi I IN: two ases.cyminalto. DAMPER WINDINGS ‘UN atltlit. (h) Kenconnected copper dumper. (f) Connected copper damper. 57Experimental values of crest voltages (twelfth cycle) from terminal a to b when switch in Fig. (g) Connected high resistnncc damper. However. the addition of other possible cirI’llits for crlrrcnt flow complicates the internal calculations. lcnst 1. They are somewhat cheaper than connected dampers but at the expense of no longer being able to make x~” and xd” equal.chapter 6 Mnchine Characteristics (d) [iid. sq/sd’=2. Unit of voltage is crest of terminaltoterminal voltage before short circuit. 60. 54 is closed. Tn this type of damper. While a continuous or’ connected damper winding is most effective. IX. (c) Connected high resistance damper.
.53 0.0.14 Both of these tables represent test results. one of high resistance and low reactance and the other of low resistance and high reactance. . The general characteristics of damper windings will be discussed under the following heads.. Because ‘of. Table 2 shows the effect of different types of windings upon a lOOkva generator12 and Table 3 upon a 5000kva synchronous condenser. 32.390 . which are in effect a double winding.171 0. t Xq _0. Machine Characteristics __.Q3 0.037 this classification fall such dampers as doubledeck windings.. Fig. the steady character of the load..Connected Copper.0.164 0. if this condition persists the currents that flow in the body of the pole pieces.146 0. 59Connected damper windings: (a) Slowspeed machine. phase balancers. Chapter 6 connected in series to slip rings. Another type of special winding is one that is insulated from the iron and Special Dampersh . later. The principal uses of this type are in motors where the combination provides better starting characteristics.028 i /2d1)19) . Damping Effect In the early days when prime movers consisted mostly of reciprocating engines the pulsating character of the .92 2.245 33. The former reflects its ability to prevent unbalancing of the voltage and the latter its ability to carry the negativesequence current without undue heating of the rotor. Voltage distortion similar to that discussed under unbalanced shortcircuit% occurs. These properties are particularly important for such fluctuating loads as electric furnaces. Xd’ Xq dXd’X0 0.2 AMPERES) (b) Fig.260 0. Xd” 0.” _.. At higher speeds the lowresistance winding becomes effective. The best criteria of a polyphase machine to carry unbalanced load are its negativesequence reactance and resistance. 31.151 0. Such machines if unequipped with damper windings have characteristics which resemble closely those of a threephase machine without damper windings when a singlephase load is drawn from it. (b) Highspeed machine.__0.. Connected Everdur. TABLE ~CONSTANTS OF A SYNCHRONOUS GENERATOR AS AFFECTED BY TYPE OF DAMPER WINDING (100 KVA. NegativeSequence Reactance and Resistance As discussed previously the negativesequence reactance and resistance of a machine are both affected by the damper windings.he starting characteristics can be varied at will.     Type No damper.. Not only do the dampers reduce voltage unbalance but also reduce wave form distortion. damper windings in singlephase machines and phase balancers must be heavier than those in threephase machines.._ __.157 0. 2300 VOLTS.157 0.036 0. In addition. which produces a high torque. Nonconnected Copper.154 0.___ . At low speeds the high reactance of the lowresistance winding forces the current to flow through the highresistance winding. produce excessive heating. 60Two types of nonconnected damper windings. Balancing Action and Elimination of Voltage Distortion One of the earliest needs for damper windings arose from the use of singlephase generators and.577 . The addition of damper windings provides t: lowresistance path for the flow of these currents and prevents both wave distortion and excessive heating.063 0. By connecting a variable resistor externally to the slip rings t. 25.388 Tad I.182 ~.
. although the latter phenomenon is usu:llly climinnted by altering the natural frequency of the s~slom by changing the fly wheel effect of the generator 1)~ mot.215 ratio is greater than 1. While a lowresistance damper winding \vill decrease the number of electromechanical oscillations f.. the benefits are greater for connected or continuous dampers than for nonconnected tl. ’ X.045 0. This was successf. 12 Calculated 0. Other Considerations ings Affecting Damper Wind Synchronous generators feeding loads through transmission lines having a high ratio of resistance to reactance t. Grits capacitors in decreasing the effective series reacta~‘(! increase the ratio of resistance to reactance and thus t(‘n(l to increase the likelihood of spontaneous hunting. the terminal voltage is regarded as the terminaltoneutral value.cllle made parallel operation difficult. 34. >lorc recently in consideration of the stability problem I. a curve of terminal voltage plotted against field current. upon the stability problem.. Mechanically there is no choice as both types ‘~‘1 be made equally reliable.026 0.195 0. 61.! ctcd brass.uVers. The conditions that must be satisfied for this phenomenon to occur are made manifest by determining the machine characteristics for a constant inductive reactive load. ln general. . e.35 and less than 1. Synchronous motors connected through lli!$i rcsistnnce lines or cables also develop spontaneous llllnt. Test 0.. ClConstruction of Ai ( regulation loading. When terminaltoterminal voltage is used the voltage drops considered will have to be multiplied by 4. the terminal voltage. ‘1’1~general influence of damper windings.ing but not so frequently as they are always provided \\.75 0. 4000 VOLTS.044 0. proceed as follows: Choose an armature current such that z. that if any damper winding is thought necessary.045 0.. SELFEXCITATION OF SYNCHRONOUS MACHINES When a synchronous machine is used to charge an unloaded transmission line whose charging kva is equal approximately to the kva of the machine. Damper windings arc very effective in suppressing such inherent huntin:: conditions and also alleviate hunting produced by I)Gotlic impulses. Suppose the machine is loaded with a threephase reactor equal to z.llowing a disturbance this effect in itself is not import:lrl~l4 in increasing the amount of power that can be t.ontl to set up spontaneous hunting. is approximately rated voltage. L’(. may very well aggravate the condition. where beneficial effects can accrue with the “se of damper windings. 721 AMPERES) T2 TYPC ?.029 0.r:lnsmitted over the system. to the price of the machine...1 ‘vitll connected dampers..215 0. By adding *In this discussion. This voltage is given by the distance BD in Fig. their negatives(:(lucnce resistance and reactance. . A ratio of zq” to Zd” as low as abollt 1. (.. the machine may become selfexcited and thevoltage rise beyond control. ohms per phase.15 This tendency is greater at light loads than at heavy loads.0‘40 0.chapter 6 TABLE ~CONSTANTS Machine Characteristics OF A SYNCHRONOUS CONDENSER As 183 AFFECTEDBY TYPE OF DAMPER WINDING (5000 KVA... respectively.20 Cdculnted 0.. . to initiate this phenomenon but iI. To determine the regulation curve for this impedance. In consideration of the many complicated problems involved in the selection of a damper winding it would appear.69 0. .. Damper windings for which this AA \ D x M T LKL xpi \ ’ I hi / I a 0 K N IFig. that is.195 0.. 13.215 0.i*.itli :L damper winding. 61 the line OAGrepresents the noload saturation curve. the connected type should be used. the criterion at which it tends to disappear being when the angle between tlw transient internal voltage and the load voltage equals 111~ impedance angle of the connecting impedance. There r~c~tlnot be any periodic impulse.vresistance damper windings have been advocated for tl.:ted copper. co . and also their purely tl:Lmping action. cc:td Evertlur.l:LIrrpcr.(! same reason.or or both... The nonconnected winding l(‘n(ls itself somewhat easier to the removal of a pole but “Ot to sufficient extent to constitute a consideration in the ‘lloice of type to install. curves for induction In Fig. such as the pulsating Iorcllle of a compressor...lly solved by damper windings in that the damper winding allsorbetl the energy of oscillation between machines :LI1d prevented the oscillations from becoming cumulative.125 x*= : (xd"+zq") Test 0. to. in view of the low increase in cost of the connected damper. is discussed \ ill more detail in Chap.35 can be obtained with nonconnected and 1..35 add 2 and 3 percent.
To produce this voltage the magnetizing current OB is required but since the armature current is magnetizing to the extent of Ai. the actual field current necessary is only OC. the impedance as viewed from the voltage behind Potier reactance is capacitive and the armature current is magnetizing instead of demagnetizing.hat for an inductivereactance load with some modifications as is shown in Fig. In this figure the distance CF represents the terminal voltage produced by the esternnl drop ~. When the load consists of balanced capacitors having a reactance xc in which xc is greater than xp.. the field current OC is necessary. the intersection with the vertical from H determines the terminal voltage for the excitation H. 62 is equal to tan l cx&. (X At The angle CY Fig. In conclusion. \I Fig. This critical condition occurs when the slope (xc . This signifies that when this point is reached selfexcitation doesnot occur. 10.i drop.pointed out that DA is the current it was necessary to ov&rMne the demagnetizing effect. in which the Potier reactance is replaced by a reactance equal to (x. The number assigned to each curve represents the percent kva delivered at rated voltage. Since the current leads the terminal voltage by ninety degrees the voltage behind Potier reactance for the assumed armature current is found by sub 0 FIELD CURRENT Fig.~use was made of Fig. The distance AB is the synchronous L . HG parallel to Oq. z zero excitation it can be seen that if this angle is sufficiently small. For normal current.a’. The magnetizing current to produce this voltage is given by the distance OB. Point q represents excitation and internal voltage. This determines F as a point in the regulation curve. ONN. The triangle BAC is a sort of Potier triangle.+ze).i giving the distance CD or B.a. to produce the terminal voltage F. curves for capacitive curve. i For other field currents such as the point H. the voltage behind Potier reactance denoted by the point A is obtained. Ai. of the armature current.184 Machine Characleristics Chapter 6 to this distance the x. This case can be treated in a manner similar to t. 63 depicts the regulation curves for different sizes of capacitors. 62Construction N FIELD C Ai  CURRENT I i E tracting the drop x. In addition to this. draw HG : $ parallel to CA until it intercepts the noload saturation curve at G..4. The intersection with the vertical from H determines the point J. but as Q: increases a point is finally reached at which intersection is impossible and the solution fails. D4. In discussing the significance of X. ‘Fig. in A..x.)i equals the slope of the noload saturation Ai of regulation loading. 62. intersection with the noload saturation curve is possible. Ai is the distance KN in the Potier triangle. Thus by drawing any line HG parallel to CA and GJ parallel to AF. where . neglecting saturation.. to produce rated terminal voltage with loopercent capacitive current. 63Regulation curves for constant capacitive load of such values as to give the loads at rated voltage indicated on the curves. however. the field current Ai is required to overcome the demagnetizing effect of the armature current. Then draw GJ parallel to A F.
among which are: (a) Grounding of Generators. (b) System Grounding. 0 0 5 IO 15 TEdhUNAL KV reactance drop xdi and DC the Potier reactance drop. . The contribution to the fault current for this condition is equal to 43 27rfGEX lows where f is the system frequency. The capacitance to ground of the windings must be known so that the associated resistance can be selected. For other voltages multiply by factor K in insert.ated otherwise. of the order of 120 percent. This is discussed in considerable detail in the chapter on Grounding. KVA Fig. The capacitance must be known so that the contribution of this element to the ground current can be determined for single linetoground faults.)i Ai < Ai Or (117) Xe<Xd St. the machine will become selfexcited if the kvs of the machine as defined by E is less than the !i 2 2 I charging kva of the line E.20 5 I z . danger may threaten when the charging kva requirements of the line exceed approximately 80 percent of the kvn of the machine.0. E the linetoline voltage and Co the capacitance per phase in microfarads. XI. Thus the slope of the noload saturation curve is equal to xdixpi ~ The condition for self excitation is then that ni * (xcx& (Xd. 64Capacitance to ground of TURBINEGENERATOR windings for 13 200volt machines in microfarads per phase. The capacitance of the rotating machines may be an important element in the determination of the system recovery voltage. (c) System Recovery Voltage. Since xd is. CAPACITANCE OF MACHINE WINDINGS A knowledge of the capacitance to ground of machine windings is necessary for several reasons.10 1.30 6 2 2 0.x. It is cus MVA Fig. except for special cases. 65Capacitance to ground of SALIENTPOLE GENERATORS AND MOTORS in microfarads per phase.chapter 6 Machine Characteristics 185 W40 2’ 2 5 a.
006 0 5 50. Fortunately. Substituting H from Eq.016 2 0.186 Machine Characteristics Chapter 6 !$ 0.231 (119) . Tn testing the insulation of machines. f is the system frequency.2. Now fY(A8) a== Ia2 KAB. (93) LIP= P.AP in deg/sec2 kva ZZ rf = AZ’ in rad/sec2 (kva) Ii where the kva refers to the rating of the machine and H the inertia constant. the capacitance will not vary more than f 15 percent from the values for 2300 volt. Thus. It is not within the scope of this work to discuss this subject in its entirety. The sign of P. for testing at time of installation or for testing after rewinding. 8 Fig. it is sometimes necessary to know the approximate charging kva of the windings so that a transformer of sufficiently high rating can be provided beforehand to do the job. the ASX C501943 Rotating Electrical ?\Iachinery Standards as fn= 35 200 P. but merely to derive the above expression. NATURAL FREQUENCY OF SYNCHRONOUS MACHINE CONNECTED TO INFINITE BUS J. is the synchronizing power in kw per degree displacement.AO in kw 0. (114) 7r106 fPr A0 “‘OX (WZ?)(rpm)? = . The generator data was obtained from reference 23 and the motor data from some unpublished material of Dr. E. however. L. the capacitance will not vary more than ?I 15 percent from the values for 2300 volts. I< is positive.004 capacitance per phase to ground in microfnrnds and E is the applied voltage from \\intling to groimd. From the Stability Chapter it can be seen that the acceleration of the rotor is 180 f a = . 0 0 100 200 300 HORSEPOWER 400 500 fP= (11~R2)(rpm)2’ 0. For details of this type of calculation refer to the chapter on PowerSystem Voltages and Currents During Abnormal Conditions.. If an incremental displacement A0 be assumed. it should be borne in mind that these characteristics vary greatly between machines of different designs. El a 0. not very great accuracy is required for the applications cited above. For voltages between 2300 and 6600.KA0. is actually negative as an increment in A0 { produces a torque which tends to return the machine to the 3 operating angle. . 00 0 100 200 HORSE 300 POWER 400 500 Fig.Xf cycles per minute (rpm) J. This information should be typical of other machines to within about +_ 50 percent. 67Capacitance to ground of 2300volt INDUCTION MOTORS in microfarads per phase.WR2 (113) tomary to represent the machine capacitance in this work by placing onehalf of the total capacitance to ground at the machine terminal.024 a=% (Wn’:(rpmj2AP in rad/sec2 (116) and substituting AP from Eq.012 2 2 i? p 0.. (d) Charging Kva. Figures 64 to 67 provide basic data calculated for Westinghouse turbine generators and salientpole generators and motors. A synchronous machine connected to an infinite bus possesses a natural period of oscillation which is given in. 66Capacitance to ground of 2300volt SYNCHRONOUS MOTORS in microfarads per phase to ground. (117) (118) K= %f. When given an angular displacement. the machine oscillates with this frequency and finally subsides unless subjetted to periodic impulse of proper magnitude.04 I E a 0. the corresponding synchronizing power is (114) and A0 is in degrees. Harder. The charging kva per phase is equal to 2rfCoE?X 10e6 where Co is the where P. particularly in the field. XII. For voltages between 2300 and 6600. In general. This is required either for normal routine testing.
Saturation within from which the machine is a minimum for the former and a masimum f2E n for the latter.(27rfn)2 A sin 27rf. 69 shows how from no load a transient component of the shortcircuit 2 J w 3 . Fig. All reactances from “Rated current” threephase short circuits without external reactance.: = 0. Phase short circuits from rated voltage no load.quantity. S. The ratcdthen substituting this relation into Eq. The ratedcurrent operates affect certain of the principal constants through figure is used as a base for all the curves. If threephase shortproblems.6 ii EINE GENERATOR I 2 3 4 5 6 7 6 9 IO $ 0.xd” is obtained from the same test as that for which the t111cc rated voltage at no load before the shortcircuit is ratedcurrent reactance of xd’ was obtained. TYPICAL CONSTANTS AND CQSTS current being altered by introducing different esternal 130th the voltage and the current at which a machine reactances in the armature circuits.these socalled constants are not iq reality constant. The ratedvoltage 2. limited by series reactance. In this case cnllcd the “ratedvoltage reactance” and the reactance rated current refers to the transient component and not obtained when the excitation is reduced so as to produce the subtransient component of current. Saturation factors for salientpole machine with damper winding is equal to unity. .~hstitut.1. because of lower excitation.9 ?I z 2 0. TRANSIENT COMPONENT ARMATURE CURRENT IN PER OF UNIT Fig.9 3 1 0. The rated voltage value is sometimes called 27r .circuit studies. (113). the “saturated value” and is the value usually given by the designer. lends itself more . ThreeCurrent 69Saturation factors for subtransient reactance. 0 0 TURBINE GENERATORS I.” A8 = A sin 27rfJ (121) X knowledge of these two values of xd’ is not sufficient for all applications for which zd’ is required.7 012345676 SALIENTPOLE AND MOTORS GENERATORS 1 ~~~ERATOR. the react.l= Kd sin 27rfnt readily for determination from test.chapter 6 Machine Characteristics 187 current equal to rated value is called the “ratedcurrent reactance. 68 obtained from data presented by Kilgore16 shows how the reactances whic*h converts to Eq. that the ratedcurrent reactance rcw:tance obtained when the excitation is such as to pro. (119) demand a quantity determined under conditions for which f. . z~‘. value used as base. Thus the antes. (120) current xd’.. = g&g cycles per set (122) the terminal voltage is near rated voltage and the armature current is likewise near its rated current.hc variability of the permeability of the iron. Certain applications. of typical machines of different classes vary if threephase shortcircuits were applied from rated voltage no load. such as stability studies.1 5 I. :Lnccsso obtained vary with the excitation. In this reactance on the curves for rated current is the one that sclnse. Fig.d’ is that required for short. with this difference. Two of these Similar considerations apply to the subtransient reactcllwntities have been given special designations. No specific name has been assigned to this circllits arc applied to a machine from no load.1 n SALIENTPOLE MACHINES 0 2 4 3 TRANSIENT COMPONENT OF ARMATURE CURRENT IN PER UNIT I 5 Fig. the XIII. 2 012345676 SALIENTPOLE SYNCHRONOUS CONDENSERS x F #$ I.ing K from Eq.0 E 2 0. 68&t uration factors for transient reactance. would have greatest utility for stability and regulation Consider the transient reactance. The particular t.6 .
74Opencircuit transient time generators. These values are plotted in curve form in Fig. of ac gen 6 ARMATURE CURRENT IN PER UNIT 2 0 I Fig. In general.60 0. all Chapter 6 points being obtained from threephase shortcircuits with no external reactance. 0 0.. as evidenced by Fig. it is unnecessary to make this distinction Fig. Td’ varies I7 in the same manner as zd’. The angular relations within the machine are determined to a large extent by x. 72 taken from Wright’s paper. by test. The variation. It must be kept in mind . B B p 0. . The normal value of I’d. . 72Variation of x0 for turbine generators. for several salientpole machines*2*17is shown in Fig. and the subtransient shortcircuit time constant Td” can be neglected. the cllrrent being altered by the for the negativesequence reactance. D = 100 kva generator with damper winding. 73 and 74 taken from a paper by Hahn and Wagner. In general. C = 33 1 kv:t motor with damper winding removed. Table 4 gives both the range of typical constants that are characteristic of normal designs and also an average that can be used for general purposes when the’specific value of a particular machine is not known. with the size of the unit the curves of Figs. of x..50 g 0. . The negativesequence resistance is that obtained at a negativesequence current equal to rated current. so that .1” varies with the transient component Machine Characteristics of current. 0 20 40 60 00 100 120 OF KVA 140 \ A = 7500 kva generator without damper winding. 71x. RATING IN THOUSANDS OF KVA Fig. RATING IN THOUSANDS i Fig. for salientpole machines..10 0. . designed into waterwheel generators varies with the kva capacity and speed.” is not affected to any great extent in the region for which it has greatest use. 73Opencircuit transient time constants erators and motors.188 X.40 0 d .01 ’ 20000 ’ 40000 ’ 60000 ’ 00000 1100000 ’ 120000 ’ ’ ’ ’ ’ I KVA Fig. the current during the sustained terminaltoterminal shortcircuit being limited to the rated current. is still maintained. . To obtain lower values than those indicated usually involves an increased cost. (xd”‘) for 7 6 under NegativeSequence Reactance. constants of turbine i 1 The zerosequence reactance.. The AIEE codelo suggests determination of x2 by means of the method discussed 0.‘* are also included. 71. Because of the wide variation of Td’. For practical purposes the effect of saturation upon the opencircuit transient time constant Td’. .601 g 0.50 2 2 0. the relation Td’=zTd’. 70. B= 750 kva generator without damper winding..20 PER UNIT ARMATURE CURRENT .30 5 3 0.10 p 0. .0. . I I escitntion before the shortcircuit. 70Normal unsaturated waterwheel transient reactance generators. 0.
The additional cost of the additional WR2 is about proportional as shown in Fig. The cost per kva of waterwheel generators depends “Pan its kva and speed. The extent of this variation is shown in Fig 78 Machines of higher shortcircuit ratio OrPowerfactor are more expensive in the proportion shown ”  4 I .. 75.. Column (0) in Table 4 refers to the ac re.ogcnerators and the corresponding figures for water. ‘f’hc inertia constant. The general variation of H of t.~crators. The WR2 represented ‘9 the curves of Figs..r..   >  .. The effect upon H IO I  8 c     THOUSANDS (a) OF KVA .  THOUSANDS Ibj OF KVA !A  Fig. T.. for snhentpole machines either with or without damper windings and as the 1.chapter 6 Machine Characteristics 189 that the loss assocrated therewith varies as the second Power of i.. II. turbine included.rmn (10) the dc resistance. including allow  (:L) I~rgr turbine generators. When m%nitudes of WR2 in excess of these are desired a more cxPcnsive machine results. or.ist:rncc. n   of incrcnsing the shortcircuit ratio and changing the P’)\vcrfactor is given in Fig. (11) fxge vertical type waterwheel generators. (which includes the effect of load losses) and .\. which is discussed in Chap. :LIICCof 15 percent for waterwheels. 75Inertia constants. 76. 75 and 76 are those obtained Ir()m a normally designed machine in which no particular cfT()rt has been made to obtain abnormally high H.l.8 power of it for turbine g’.vhccf gcnerntors are given in Fig. 77. 13 is likc\vise given in Table 4.
pilot exciter. and from 2. \ 1 111 16 I.0 79Effect shortcircuit 20 of shortcircuit ratio upon ratio and 0. 60 0 0 z 50 : 40 zi Fj 30 s g 20 2 z IO E E 0 1. 31. WCost of synchronous condensers and autotransformer. Directconnected pilot exciters cost approximately 30 percent of that of the exciter.6 3. including exciter I 0 ! IO I 20 THOUSANDS 30 OF KVA 40 50 J I in Fig. and 0.5 to 6 percent for high speeds. The loss in the resistor rr s shaft power and since the circuit is on a per phase basis.6 SHORT CIRCUIT 2. 50..8 powerfactor cost (Normal 1. SHORT CIRCUIT RATIO Chapter 6 1. 40 60 INCREASE IN H 60 100 values given PERCENT Fig.2 RATIO 2. directcon (0. 76Effect of shortcircuit ratio upon H.0 short circuit ratio) and 1. The larger figures apply for units of about 300C kva and the smaller figures for machines of about 50 OOC kva. 75.4 1. respectively. 46 THOUSANDS OF KVA Fig. XIV. The condenser cost per kva including the exciter. I OO 20.0 a.32 percent for a 1000. The exciter kw varies with the size of the unit.2. ranging from 1. Naturally these figures will vary from year to year with the cost of materials and labor. 5000.7. 79. and 50 OOOkva unit. 78Cost of waterwheel nected generators including exciters only. 77Effect of increasing H above the normal by Fig.9powerfactor (1. The cost of normal exciters for waterwheel generators varies from 7 to 13 percent of the cost of the generato] alone for slow speeds. the total shaft power is thus . and autotransformer is plotted in Fig. A Fig.8 powerfactor CO.Machine Characteristics .25 short circuit ratio) The equivalent circuit of the induction motor is shown ls represents the in Fig. Fig.1 short circuit ratio) and 1. 0.0 powerfactor and 1. INDUCTION MOTORS Fig.0 used as base).
those in which the machine is (lisconnected from the source of power and.. This approximation permits of T=7..ator resistance per phase in ohms) Fig. (123) (124) 191 Therefore.. 31 can be simplified consitlcrably by shifting the magnetizing branch to directly :tcross the terminals. the torque is (3Trir2) in watts lb ft. those in which the machine remains connected to the source of Power. second. As a first approsimation the shortcircuit current supplied by an induction motor can be resolved into an alternating and a unidirectional component much like that for a synchronous machine. SlApproximate equivalent circuit of induction SO motor. power = ?(3&2) =. however small. ‘[‘it :tl shaft horse power = ~o(Ti”Ib”)(‘pm). transients involving induction motors fall within one of two categories. 550volt squirrelcage motor. The time constants are namely.. (127) lb ft. that is. The total shaft power can also be expressed in terms of t.$. In the second case. 81. the efficiency decreases with increasing slip. 82 shows the shortcircuit current of a 25horsepower.04E2 T= (rpm) 8yn. I’~LtivcIy simple determination 1 Worncs 7. ‘l’hc equivalent circuit of Fig.or‘ques. for the alternating component. The dotted line in the upper total shaft power total shaft powerfrotor copper loss 1 s !9 c=l. for 90 percent slip the efficiency is 10 percent. 82Shortcircuit currents induction in armature motor. first.~tllcr losses. of squirrelcage . the rotor copper ll)ss is directly proportional to slip. (126) I~:cll~ating(124) and (126).?). (125) $+1 c ‘rll~~s. In the first case the transient is determined largely hY changes in magnetization and may be quite long. being 10 percent for 10 percent slip and 90 percent for 90 percent slip. Examples! of this case. The initial rms magnitude of the alternating component is equal to the terminal voltage to neutral divided by the blocked rotor impedance per phase. the maximum rms current the breaker can carry for any time. are the sudden energization of an induction motor or sudden short circuit across its terminals.. The resultant approximate circuit is shown in Fig.i. (blocked rotor reactance per phase in ohms) in cycles 2~ (st. Similarly. neglecting example of this case is the phenomena that. 2a (rotor resistance per phase in ohms) for the unidirectional component. Thus. the tmnsicnt is determined by reactions involving both the stator and rotor and the duration is quite short. occurs during the interval between the transfer of powerhouse auxiliaries from one source to another. 35 Contribution to System ShortCircuit Current In the calculation of system short circuits only synchronous machines are usually considered but in special cases where induction machines constitute a large proportion of the load. 3 of i..04 JXs 3 JXr Fib.chapter 6 Total shaft. For 10 percent slip the efficiency is 90 percent.(ls).7:6 ~(3rir2) ‘rllc rotor copper loss is (3r. (128) LIost. (127) l Sin per unit (wdsyn.. (blocked rotor reactance per phase in ohms) in cycles. the efficiency is: Efficiency = Machine Characteristics in watts in hp. An Fig. their contribution to the shortcircuit current even if its duration is only a few cycles may be large enough to influence the choice of the breaker from the standpoint of its shorttime rating. . s that Eq.
746T ibft 2drpm). the same acceler ation but a torque 2H times as great is required. The unit of shaft torque requires special comment. for which case the torque expression becomes The crosses close to the torque curve in Fig. the unit of torque is 33 000 2dO.5 if ln 0. Either the conventional method of the circle diagram or expression (132) can be used.0 4 5 1.6 8 TORQUE dt/dw AND TIME Fig. ElectraMechanical Starting Transient t 0.192 Machine Characteristics r. 83Illustrating calculation of speedtime curve of an induction motor upon application of full voltage.. Suppose that one per unit torque is applied to the motor’ which means that at synchronous speed the power input’ into acceleration .00 x.ROUE = Td 36.02 Chapter 6 ’ curve indicates the computed value of the envelope of the alternating component of shortcircuit current.u. and suppose further that the rotor is brought to synchronous. (132). . H. 31 shows the conventional diagram of an induction motor. I 0 0 0 I 04 2 3 a. Woundrotor motors.746) kvarated (rpm)synch. H may be comnuted bv means of Eq. To convert to accel: eration it is convenient to introduce a constant. In Fig.02 xg. include the WRiof the con.4 / . therefrom into Eq. The unit of both power and reactive voltamperes will be the rated kva of the motor and not the rated power either in kilowatts or horse power. 1.=.dls)~ (13o) in 33 000 Equating. 83. = I$. This can probably be attributed to using the ac resistance of the rotor rather than the dc resistance.<93). and equation (131) in per unit becomes T in p. At rated slip the voltamperes input into the stator must be 1S equal to unity but the power absorbed in the resistorr. This convention is consistent with the choice of units for the impedances. 83 is also shown the torque requirements of a particular load such as a blower. For most motors the magnetizing branch can be neglected. there results that T in lb ft= 33 000 W0.1 netted load. in which unit current is the current necessary to develop the rated power at the rated voltage.2 6 7 I . In the present discussion the per unit system of units will continue to be used. The shaft power can be expressed as ls Shaft Power in kw = kvarsbd Izrrs * In terms of torque the shaft power is equal to Shaft Power in km (129) / / / \Y /\ 1 \I \I TORQUE’T I I I ‘% CALCULATED TOROUE POINTS NEGLECTING MAGNETIZING INRUSH / TIME dt/dw 0. From this it can be seen that if the inertia is I such that at synchronous speed the stored energy is H then to develop this energy in one second. The amplitude shoivs a substantial check but the computed time constant was low.of the rotor is equal to rated kva. WR* must. The dotted line in the lower curve is the computed value of the unidirectional component which checks quite well. Upon applying voltage to the motor the difference between the torque developed by the motor and that required by the load is the torque available for acceleration of the rotor. the first step involves the calculation of the shaft torque as a function of the speed. Therefore there results that . speed in one second.oLTA:~gl$YFJ ?rL=3’o TO. which is equal to the stored energy in kmsec. operated with a substantial amount of external resistance. In using the latter method it is only necessary to solve the network of Fig. s 032) For the purpose of determining the nature of electromechanical transients upon starting a motor from rest.746) kVarated bpm) synch. T in per unit=ei =. unit of torque produces 1 per unit of acceleration if the inertia is such that s kva of stored energy is produced ix? one second. per kva of rating’ at synchronous speed.00 rr =. / Fig. mill have such small time constants that their contribution to the shortcircuit can be neglected. Thus 1 per . I$ (131) If unit torque be defined as that torque required to produce a shaft power equal to rated kva at synchronous speed. s will be less than unity and will be equal numerically to the ratio of the rated power of the motor to the rated kva. 83 were computed by this expression. of course. A solution of a typical motor is shown in Fig. During this interval the acceleration ’ is constant (1 per unit) and the power input increases’ linearly with time so that at the end of one second the: stored energy of rotation is (s kva) in kwsec. then from (130). (31) and substitute the solution of I.
Most economical used as base price. The following formula can be used to form an approxirnt1.t that $ is known as a function of w and the time to rc:lch any value of w can be determined integration. 83. If the voltage source is reapplied when the source voltage and residual voltage of the motor are out of phase. The curve of time so obtained is pl(jt.2500 e 250011 fI 92000 B 2000 I: I: 1500 2 > 1000 2 2 0 8 a 500 0 IO 20 30 40 50 60 70 193 ‘Ihis function is likewise plotted in Fig. %Decay c 210 TIMECYCLES of residual voltage% of a group auxiliary motors. . 20 1 40 I 60 I I I I I I 100 200 MOTOR SIZE 400 IN 600 1000 2000 4000 6000 lot HORSEPOWER Fig. variation of price with voltage and horsepower of squirrelcage These values apply approximately for 8 poles or less for 60cycle motors. c’ .lris form of the expression may be seen at once from the f:. currents exceeding starting values may be obtained.:tnt voltage. 37. SCApproximate induction motors. The utility of l. Residual Voltage If an induction motor is disconnected from its supply. the rate of deceleration being detcrmined by the inertia of its own rotor and the inertia of lhc load and also by the nature of the load.. it rotates for some time. 83. the time to reach any spcctl is obtained.Machine Characteristics Ablccelerntion can be expressed as ‘5 and its reciprocal “’ as &’ Thus from (134) dt dw 2H TTL (135) u. should include any external react:mce in the stator back to the point where the voltage may be regarded as constant and et should be that connl. of power house i( ) l<y summing up areas (such as indicated by the shaded portion) in a vertical direction. flux is entrapped and voltage Fig. 0 200 300 400 500 700 HORSEPOWER 1000 1500 85Typical time constants for 2300volt induction motors. to h:tlf speed Time to half speed= H(rsSnr)2in seconds (137) rre2t All of the above units must be expressed in per unit.tcd in Fig.. Thus t= g dw by a simple (136) Fig. squirrel cage appears at the open terminals of the machine. Remember also that 2.c idea of the time required to accelerate a motor: \vtrose load varies as the square or cube of the speed. Because of the inductance of the rotor.
E.ics. Jan. Vol. A.I.E. Vol. Transnclions.I. Synchronous Machines.E.I. Park. H. p. 13311350. A. Doherty and C. H. Park. TwoReaction Theory of Synchronous MachinesI. V. AIEE Standards Nos.s. A. 24. D. 310318. 1927. 2. Vol. Cheek. pp. 61. Synchronous MachinesIII. June 1933.I. Vol.E. 17. February 1925. Adams. Wagner. Bus Transfer Tests on 2300Volt Station Auxiliary System. t1. 54. Transactions. 22. 86 was prepared. Synchronous IMachines. Unsymmetrical ShortCircuits in Wat. Xickle.E. A. Determination by S. Preferred Standards for Large 360%RPM aPhase 60Cycle Condensing Steam TurbineGenerators. by C. Ovcrvoltages on Wat. July 1937. by R. 6.E. C. Wright. A. Practical Calculation of Circuit Transient Recovery Voltages. 351. 503. 2.Z. Transactions. Hahn and C. 47. 1931. 4. No.E. Tronsaclions. Test Code for Synchronous Machines. Skeats. Wagner.E. Vol. Park. July 1937.E.I. D.E. A. pp.E. Synchronous Machines1 Two ReactionTheorySteadyState Power Angle Chnracterist. July 1929.I. August 1938. No.E. May 1949. 11.E. Xckle. by C. 353361. A. pp. Doherty and C. 5180.I. E. Vol. pp. C. 49. Transactions. Discussion. 9. p. Johnson and H. 1950. Transactions. by R. This curve includes not only the effect of magnetic decay but the reduction in voltage due to decrease in speed. Cheek. by A.E. Standard Decrement Curves. C. A. A. A. 8097. E. 813. Wagner. C. 601 and 602. A.I. Transaclions. E. by L. R. pp. 15..E. Van Sickle and T. The opencircuit time constant for individual 2300volt machines is given in Fig. Evans and C.E. pp. A. H. As the rating increases the most economical voltage also increases.Z.erWheel Generators. p. p. Booth. W. 8. 14. 10. 1926. by W. 13. REFERENCES 1.E.E. March 1931.194 M’achine Characteristics Chapter 6 Figure 84 shows the decay of a group of powerhouse ausiliary motorsz5. 50.E. Sillers. pp. Doherty and C. Wagner. pp. A.E. W. pp.I. L.Z. There is a great variance in this constant between different designs but these curves give an idea of the magnitude for squirrelcage induction machines. Wagner. Vol. by R. by R. A. Anderson 20.E. A. Transactions. 716. A. .E. Vol. Damper Windings for WntcrWheel Generators. A. A. April 1930. To form a basis of judgment of the effect of voltage upon size the curve in Fig. 352. 321 and September 1938. Harder and R. April 1928. 18. 45.E.E. Transactions. 50. 48. L.E. 44. by R. by E. II. $00. 140151. F.Z. Transactions. 1927. 1942. Transactions.E. Further Studies of Transmission Stability. IV.E. Kilgore. Transactions.E. 46. Transaclions. Vol. p.E. A.E. A.E. June 1928. Effect of Armature Resistance Upon Hunting of Synchronous Rlachines.E. Nickle. Vol. by J. 457.Z. 545550. General Electric Reoiew. F. p. 114.E. Vol. G.E. by C.E.E. by J. Vol. by R. 2. 12. A. Tmnsaclions. presented before AIEE Winter Meeting. 1944. p. Potier Reactance. A. of Synchronous Machine Constants by Test. 49. The Electric Journnl. Doherty and C. 771778. 912942. Effects of Saturation on Machine Rcactances. F. A. pp.E. Transactions. F. 85. Wagner. 52. Auxiliary and A. F. 23. Wagne.I. by C. A.I. Vol.Z. 827. by R. pp. and IIAn extension of Blondel’s 3. A.E. 13851395. 1935. A.E. Nickle. C. 5. Harder and R. Approximating A. June 1945. Cost of Induction Motors The price of induction motors of a given rating varies with the voltage. H. Monteith. Power at Richmond Station. 1932. July 1930. Transactions. Sovember 1937. p. E. 1926. A. 10111024.I. Transaclion. by R. F. by V. Torque Angle Characteristics Tinder Transient Conditions. Regulation of AC Generators with Suddenly 4pplied LoadsII.E.I. TwoReaction Theory of Synchronous Machines. Publication So. 25. Regulation of AC Generators With Suddenly Applied Loads. by C. 63.E.E. Transactions.Z.Z. p. 332334. by E. b 16. i. ThreePhase Short Circuit Synchronous Machines. 21.I. Bush and R. 38. pp. 45. The group had a total rating of 2500 kw of which the largest was 1250 hp. 904.. Transactions. 19. A. F. F.E.Thompson. 1950. pp. Vol. A.E. Definition of an Ideal Synchronous Machine and Formula for the Armature Flux Linkages. Power System Transients. Tmnsnclions.crWheel Generators Under Capacitive Loading. by Sterling Beckwith.
CHAPTER 7 EXCITATION . it became necessary to operate synchronous machines in parallcl. and in case of trouble with the main exciter.jtability problem to determine the factors involved that most affected the ability of a system to transfer power from one point to another. E. transfer was accomplished manually. Early excitation systems were of many different forms (lcpending principally upon whether the main generators were small or large in rating and whether the installation ‘WS a steam or hydroelectric station. a group of engineers undertook solution of the . WKC accurate generator voltage regulators were soon introduced to the industry. and operating engineers had little concern about system stability. It usually had sufficient capacity to carry the excitation requirements of the entire station for at least an hour. voltage. Vol. which involved the transmission of power over long distances. Transac&ns.E. The standby excitation source \vas usually a spare exciter. Standby exciter capacity was provided in the commonbus system by a battery floated *dJ. 195 . Barkle. waterwheels. 1924. or combinations of these to provide a main and emergency drive.4uthor: J. appeared to be at least one method of increasing the stability limits of systems and preventing the separations occurring (luring transient conditions. The common exciter bus was generally energized l)Y several exciters driven by motors. E. steam engines. practically constant voltage was obtainecl on the bus and control of the individual ac generator field voltage was accomplished by using a variable rheostat in each field as shown in Fig. and it was pointed out that the synchronous machine excitation systems are an important factor in the problem of tlctermining the time variation of angle. In 1922. but it was not long before it was realized that directconnection of the exciter to the generator shnf t offered an excellent answer to the many problems encountered with separatelydriven exciters and this system grew rapidly in popularity.or turbinedriven. lCommonexciterbus compounded exciters excitation system using flatand ac generator field rheostats.E. :md esciters with higher speeds of response and faster. It soon became apparent that system stability was of vital importance in these cases and also in the operation of large interconnected systems. It was not recommended that this region of dynamic stability be considered for normal operation. Shand stressed the theoretical possibility of increasing the steadystate power that could be transmitted over transmission lines through the use of a generator voltage regulator and :tn excitation system with a high degree of response so that operation in the region of dynamic stability would be possible. In the commonbus system without a floating battery. The results of these studies were presented before the AIEE in a group of papers* in 1924. Jr. SYSTEMS RIOR to 1920 relatively little difficulty was encountered in the operation of electrical systems. but. I. pp. Improvement of the excitation systems. As the loads grew and systems expanded. Pilot exciters had not been used up to that time. the bus was operated at constant voltage supplied by compoundwound dc generators. and power quantities during transient disturbances. therefore. 16103. P on the bus. Greater interest in the design of excitation systems and their component parts developed. and difficulties encountered were not well understood. either motor. 43. that it be considered additional margin in determining permissible power transfer. Motor or turbine drive was also used in the individualexciter system. The exciters were invariably selfexcited. Thus. The two broad classifications were those using a common excitation bus and those using an individual exciter for each main generator. turbines. B. 111 certain areas it became necessary to locate generating slations some distance from the load centers. When a standby battery was EXCITER SHUNT FIELDS IT1 JS  ‘1 1 sr i AC GENERATOR FIELDS Fig.
time area as attained by the actual exciter. Methods of calculating and analyzing excitation system performance are also included. Second. In using the perunit system of designating esciter voltages. 39. First. the esciters were shuntwound to prevent polarity reversal by reversal of the seriesfield current. along with the methods of combining these parts bo form an escitation system having the most desirable features.e%itation voltage. and better controlling devices. there have been many developments in escitationsyst. The generator voltage regulators in use at that time were predominantly of the continuouslyvibrating type. rated speed and terminal voltage. 15511637. therefore. The achievement of all of these ideals simultaneously is a difficult problem. lower field voltages being obtained with a generatorfield rheostat so that the exciter could operate slightly saturated and be stable. and with the field minding at 25C. the same . which may be an ac or dc machine. in volts per second. DEFINITIONS In discussing excitation systems. the exciter was a shuntwound machine with field control enabling it to opernt. several choices are available from which to choose the unit. Nominal Exciter CeiZeiling VoltageNominal exciter ceiling voltage is the ceiling voltage of an exciter loaded with a resistor having an ohmic value equal to the resistance of the field winding *A symposium of papers on excitation systems was presented before the AIEE in 1920 and gives details of equipment and practices in use at that time. etc. The exciter usually operated at voltages between 30 and 100 percent. Excitation System StabilityExcitation system stability is the ability of the excitation system to control the field voltage of the principal machine so that transient changes in the regulated voltage are effectively suppressed and sustained oscillations in the regulated voltage are not produced by the excitation system during steadyload conditions or following a change to a new steadyload condition. 6. This resistanceshall be determined at a tempers ture of: (a) 75C for field windings designed to operate at rating with a tempernture rise of 60C or less (b) IOOC for field windings designed to operate at rating with a temperature rise greater than 6OC. pp. Exciter Ceiling VoltageExciter ceiling voltage is the maximum voltage that may be attained by an exciter with specified conditions of load. In the definitions of ratedload and noload field voltaie. sensitivity. “Definitions of Electrical Terms”.* In the past 25 years. but for system analysis it has very little utility. See AIEE Transactions. Publication C42. the exciter voltage necessary to circulate the field current required to produce rated voltage on the airgap i line of the main machine. ing meansfor its control. 1 review of the common excitation systems in use at the present time is presented in this chapter. reliability. is divided by the ratedload field voltage.196 Excitation Systems Chapter 7 floated on the common bus. Kzcitation SystemAn excitation system is the source of field current for the excitation of a principal electric machine. The following definitions are proposed for inclusion in the new edition of the American Standards Association.em design and practices. This definition does not apply to main exciters having one or more series fields or to electronic exciters. There is an unceasing search among designers and users alike to find ways of improving excitationsystem performance through use of various types of dc generators. Control of voltage was under handregulation. I. It should be noted that . Exciter ResponseExciter response is the rate of increase or decrease of the exciter voltage when a change in this voltage is demanded. Note: For a rotating exciter. Third. for specification purposes it has become‘standard through the adoption by the AIEE and ASA to use the ’ ratedload field voltage as unity. In the individualexciter system. The response is determined with no load on the exciter. includ to be excited. includes all of the equipment required to supply field current to excite a principal electric machine. ratedload field voltage is the voltage formerly referred to as “nominal slipring” or “nominal collectorring” voltage. The design and characteristics of each of the component parts are discussed. the rated voltage of the exciter would appear to be the fundamental basis. accuracy. Under steadystate conditions. if maintained constant. would develop. and then suddenly establishing circuit conditions which would be used to obtain nominal exciter ceiling voltage.age required across the terminals of the field winding of an electric machine under rated continuous load conditions with the field winding at: (a) 75C for field windings designed to operate at rating with a temperature rise of 60C or less (b) 1OOCfor field windings designed to operate at rating with a temperature rise greater than 60C. which response. The ultimate aim is to achieve an ideal in rate of response. RatedLoad Field VoltageRatedload field voltage is the volt. For analytical purposes this is j the one most generally used and is the one used in the 1 analytical work in Chap. The ac generatorfield rheostat required in the commonbus system was a large and bulky device. electronic converters. For rotating exciters the temperature of the exciter field winding should be considered to be 7X. simplicity. with the exciter voltage initially equal to the ratedload field voltage. Vol. response should be determined at rated speed. Part II. in onehalf second. MainExciter Response RatiThe mainexciter response ratio is the numerical value obtained when the response. and any equipment provided to regulate or control the amount of field current delivered. the meaning of which may not be entirely clear. NoLoud Field VoltageXoload field voltage is the voltage required across the terminals of the field winding of an electric machine under conditions of no load. 1920. which had considerable loss and required a great deal of maintenance. hn excitation system. the terminals of the field winding are considered to be such that the brush drop is included in the voltage in the case of an ac synchronous machine having slip rings. a number of terms are used. For rotating exciters ceiling should be determined at rated speed and specified field temperature. 1 . The fact that these regulators were not suitable for use with the new exciters with fast response and high ceiling voltages prompted the development of new types of regulators. however.e as a variablevoltage source.
f&l current and synchronous internal voltage become eqUd . With a main generator of any appreciable size. and the response ratio.5 second The ratedload field voltage is 200 volts. MAIN EXCITERS The main exciter is a source of field current for the principal electric machine. the slipring voltage necessary to produce rated voltage at no load or noload field voltage is sometimes. Exciters for turbine generators of less than 10 000 kilowtts capacity are rated at 125 volts. 2. but rather infrequently used. The curve aed is the actual voltagetime curve of the exciter as determined under the specified conditions.2 per unit. On a 250volt escitcr. The halfsecond interval is chosen because it corresponds approximately to onehalf period of the natural clcctromechanical oscillation of the average power system. obtained by dividing the response by the ratedload field voltage. Thus.er unit exciter voltage. Using this value as 1. or 3. the straight line UCis drawn so that the area under it. The response used in determining response ratio is the slope of the line UCin volts per second. the ratedload field voltage is approximately 2. relative values of important quantities for 250 . Fourth. 2. On this rating the ratedload field voltage is of the order of 200 volts or 80 percent of the exciter rating. The relative values of these ctlutntities are shown graphically in Fig.3 to 3.0 l. and t. II. abc. any dc machine that might be used to serve this purpose can be called a main exciter. and using this definition.7 per unit.Chapter 7 Excitation Systems 197 no saturation. 0.hose for larger (Inits arc generally rated 250 volts. The vast majority of exciters in use with all types of synchronous machines greater than 10 000 kilowatts in capacity are rated 250 volts. It is the time during which the exciter must become active if it is to be effective in assisting to maintain system stability. 2Construction for determining mainexciter response ratio showing volt main exciter. during the same interval. during the onehalf second interval from zero time is equal to the area under the actual voltagetime curve. The nominal exciter ceiling voltage is defined above and can be interpreted as being the maximum voltage the exciter attains with all of the fieldcircuit resistance under control of the voltage regulator short circuited. is 1. This definition includes the small amount of saturation present within the machine at no load. Seldom are storage batteries used as main esciters. Many other types of dc machines have been developed RATEDLOAD I I I I FIELD VOLTAGE I Ibl I I I I I I TIME IN SECONDS Fig. and in keeping the battery in good operating condition are such as to make it impractical. Some of the large units placed in service recently have exciters rated 375 volts. in maintaining the charge. point a. Reginning at the ratedload field voltage. which is 120 to 132 percent of the exciter rated voltage. exciter voltage. the difficulties encountered in finding room for the battery. abde. The work can also be done by expressing the voltages as perunit values.0. the ceiling voltage is usually about 300 to 330 volts. The construction of the response line in accordance with the definition for determining mainexciter response ratio is also included in Fig. The exciter voltage required to produce the field current in the main machine corresponding to rated voltage on the airgap line is usually about 90 volts or :3(i percent of the exciter rating. 100 volts = 200 volts per second.
The type of drive accepted as reliable depends upon the type of synchronous machine being excited. the use of exciter mg sets with synchronous condensers does not involve many complications. or it can be driven by some other form of prime mover such as a steam turbine or a hydraulic turbine. Each of these dc machines. that is. However. Most essential auxiliaries have a dual power supply comprising a normal and an emergency supply. A common requirement is that the exciter mg set be capable of delivering maximum forcing excitation to the generator field during a sy&em disturbance when the motor voltage is 70 percent of normal for a period of onehalf second. in the case of highspeed turbine generators. Main exciters. and it is necessary to construct the mg set so that it can withstand these disturbances without affecting the excitation of the main ac generator. can be grouped into two classifications. The exciter mg set might be classed as an essential auxiliary for operation of the generator. The major static or nonrotating form of main exciter is the electronic exciter. The Rototrol or rotating amplifier is very different in its operation from the conventional main exciter. Direct connection of the main exciter is widely accepted in the utility industry.198 Excitation Systems Chapter ‘7 to a high degree of specialization for use as main exciters that offer many operating and maintenance advantages over a battery. in general. Pullout torque of driving motor. blame for the resultant outage would be placed on the main exciter.4 ciably or the motor stall during momentary voltage dips. the main exciter. but this enables the two machines to operate at different speeds.. important. In this respect. The term “conventional” is used with reservation since a dc generator built for the purpose of supplying excitation for a synchronous machine has incorporated in it many features to improve reliability and reduce maintenance not found on dc generators used for other purposes.0. special features being included for operation at 3600 rpm. Conventional Main Exciters i Conventional main exciters. Problems of gear maintenance are introduced. The reliability of this form of drive is obvious and no elaboration is necessary. A separatelydriven main exciter is usually driven by a motor. characteristics of the exciter mg set have become fairly well standardized as follows: Inertia constant of the entire mg set. H = 5. The synchronous motor drive is undesirable. can be classified according to their method of excitation. and may receive its power from the auxiliary powersupply system.. A directconnected main exciter is one coupled directly to the shaft of the main generator and rotates at the same speed. the complete unit being called an exciter mg set. Hence a reliable source of excitation is essential. Prime Movers for Main Exciters Rotating main exciters are of either the directconnected type or the separatelydriven type. dual prime movers are used such as a motor and a steam turbine. therefore. Directcurrent motors have been used in some cases. Response ratio of main exciter &en operating at rated speed. A new form of rotating exciter that has made its appearance in recent years is the mainexciter Rototrol. DirectConnected ExciterThe most reliable prime mover for the main exciter is the same prime mover that drives the ac machine being excited. 2. A modification is the geared or shaftdriven exciter. of course. Considerable expense. It is then subject to voltage disturbances on the main system. it is neces. because of the possibility of transient disturbances on the motor supply system causing instability. Based on this criterion. rotating and nonrotating dc machines. driven through a gear by the shaft of the main generator. Induction motors are ordinarily applied where the exciter mg set is used. The inertia constant of the mg set and the pullout torque of the motor must be high enough to assure that the speed of the set does not change appre. Aside from these special features. Nominal exciter ceiling voltage when operating at rated speed. Loss of excitation of an ac generator generally means that it must be removed from service. the motor must be specially designed to drive the main exciter through any form of system disturbance. In some cases. sary that some time interval and voltage condition for the system disturbance be chosen. 1. i. whether it be a generator or a synchronous condenser. = 500 percent. The driving motor can be connected directly to the main generator terminals through an appropriate transformer. These difficulties have been completely overcome by adequate design of the exciter. This was realized many years ago when main exciters were first coupled to the shafts of the generators. Exciter MG SetThe exciter mg set can be driven by a synchronous or induction motor.0. can be justified to provide a reliable source of power to drive.. In any event. in regard to its application as a main exciter. is discussed in detail in the sections that follow. = 160 percent.. Power supply for the motor is.e. In arriving at values for these various factors. early installations experienced trouble in operation of the dc exciters at high speeds. In the former the field winding or windings are connected across the terminals cf the machine through variable resistors and in the latter the field windings with their resistors are connected to a source . the theory of operation is the same as the conventional dc generator. The motor is apt to be subjected to voltage disturbances regardless of the source of its power supply. If the main exciter should stop running while the main generator is still capable of operating. P.. When an exciter mg set is used with a synchronous condenser. the logical source of power for the motor is the system that energizes the condenser. E. The response ratio and ceiling voltage of the exciter must take into consideration any speed change that may occur. the turbine taking over the drive when the motor power supply fails. R = 2. The most common form of rotating main exciter is the more or less conventional dc generator. being either selfexcited or separatelyexcited. and automatic quicktransfer to the emergency supply is provided in case of failure of the normal supply. in general.
3. the operating point is likewise the intersection with OA. The field resistance line OA in Fig. namely a. any exciter voltage within limits can be obtained. the terminal voltage is varied 1)~simply changing the resistance of the field circuit. and the terminal voltages for noload and constantresistance load are at points g and h. This may differ in several essential points from the load condition which will be discussed later. 4. In this manner of changing the exciter field resistance. Calculation Exciters of Response of Conventional Main It will be observed that the definition of exciter response is based upon the noload voltage buildup curve. the output voltage theoretically can establish itself at any value between zero and the point where the noload saturation curve begins to bend away from the airgap line. then the operating point is c for the noload condition and d for the constantresistance load condition. On the other hand. 4 is drawn so that its slope is equal to the resistance of the field. of essentially constant voltage such as a small auxiliary fl:ltcompounded exciter. . Thus in Fig. if the machine were separatelyexcited by a pilot exciter. The basic connections of these two forms of main esciter are shown in Figs. Operation in this region is unstable unless some artificial means of stabilizing is provided. main exciters. 4 represents the noload saturation curve of a conventional dc generator that might be used as a main exciter. the field current of the exciter is determined by the intersection at f. Should the field resistance be increased so that the resistance line coincides with the airgap line. At. An examination of the curve reveals that for values of voltage less than approximately 7. For the particular constantresistance load for which the line odD represents the saturation characteristic.Chapter 7 Excitation Systems EXCITER TERMINALS EXCITER SHUNT FIELD RHEOSTAT (01 EXCITER SHUNT FIELD RHEOSTAT CONSTANT VOLTAGE SOURCE 0 I I EXCITER TERMINALS I tb) 0 CURRENT PER FIELD CIRCUIT=if Fig. the machine is saturated and a grcatcr proportion of the field ampereturns are used in forcing flux through the magnetic circuit. The field windings of the main exciter are frequently divided into two or more parallel circuits and in the present discussion the field current is always referred to as the current in one of the parallel circuits. respectively. (b) separatelyexcited. ~TWO common forms of shuntexcited (a) selfexcited. and a line drawn coinciding with the straight portion of the curve is called the airflap line. Fig. the voltage output is no longer proportional to the field current. and a given percentage increase in voltage output requires a greater percentage increase in the field current. the ordinate at any point divided by the field current is the total resistance of one circuit of the field winding. Above the straightline portion of the curve. namely b. For either the selfor separatelyexcited exciter. called a pilot exciter. The curve oca in Fig. the response will be calculated for the noload condition and mill be applied to a selfexcited machine. the intersection of the noload saturation curve with the line 0‘4 determines the operating point.. ISteadystate selfexcited operating points for unloaded and loaded and separatelyexcited machines.5percent of rated armature voltage substantially all of the field current is expended in forcing magnetic flux across the air gap of the machine. In this region the voltage output is directly proportional to the field current. for the resistance line OA and the constant pilotexciter voltage e. lYndcr this condition. 3(a) and (b). For the present. that is.noload. If some resistance is inserted in the field circuit so that its resistance line is changed to OB. the field current is determined by the intersection of the resistance line with the pilotexciter voltage line.
5. it will decrease. we may write total useful flux linkages number of poles per pole at rated voltage )C per circuit ) k.. which is equal to the number of turns per pole times the number of poles per circuit. The leakage flux may be said to contribute the flux linkages k&f to the total. N. These linkages may be designated as k. If. second. ff if ON SELFEXCITED MACHINE FORCING VOLTAGE FOR SEPARATELYEXCITED EXCITER = e. however. concerned in cal Equation (2) states that the time rate of rise of $ is proportional at any instant to a forcing voltage which is equal to the vertical distance between the terminalvoltage curve and the straightline curve of resistance drop at any given field current. the leakage fluxes are specified at some definite current such as that required to produce rated voltage at no load. rf if \ The flux linkages. those produced by the leakage fluses.+klif These quantities are illustrated in Fig.f.t) (~) ( rated voltage at rated voltage k. the leakage fluses would be proportional to the field current. = terminal voltage of the exciter and also the voltage across its field circuit ir=ficld current per circuit in amperes TY total resistance of each field circuit in ohms = $=flux linkages per circuit of the field winding in 1OW linesturns then there exists for the field circuits the following equation: CEILING ‘SATURATION CURVE) kuex VOLTAGE . is attained. that is. as shown in Fig. This expression can be rewritten in the following form drC/ . / where each term is expressed in volts.= if at rated voltage The total flux linkages per circuit are then $=k. linked by the flux.= ( (3) rated voltage The leakage component is more complex as not all of the leakage flux cuts all of the turns. Multip!ying this flux in 10e8 lines by the turns. Beyond this point (e. If. 6Graphical determination of response of flux linkages IJ with time. considerable error is permissible in the leakage component without affecting the result significantly. The coefficient kl can be defined by requesting from the designer both the flux linkages per pole at rated voltage due to the useful flux and the total flux linkages per pole at rated voltage. for any reason.r&f) is positive. e = e. The first component is proportional to the noload terminal voltage as this is the flux which produces that voltage. dt=exrfZf (2) FORCING VOLTAGE. the flux within the machine extends beyond this point. 5. The coefficient kl is then Total # per pole at . to be specific with respect to the particular voltage concerned. until the point of intersection of the two curves. gives the total linkages due to this component. those produced by the useful flux in the air gap and. JForcing voltages and flux linkages culating response. If there mere no saturation effects in the pole pieces and yoke. Fig. (5) if Fig.+ per pole flus to useful d”:)(~gT.200 If Excitation Systems Chapter 7 e.r&r) becomes negative. . where. It shows that the flux within the machine will increase so long as (e. then the leakage at higher currents will be less than proportional to the current and at lower currents will be more than proportional to that at the specified point.e. can be regarded as made up of two components. first.e. Inasmuch as the leakage flux is only about 10 percent of the useful flux. The designer can give the useful flux at any particular voltage or it can be obtained from the design constants of the machine. $.
rrtic .4 .0031 0. 125 volts.151 I 0.0997 0.0 I72 0. tabulate the terminal voltage and field currents from Fig. (7) it can be seen that t can be obtained as a function of 9 by simply obtaining the area of the vertical strata of increments.01 65 0. as will be illustrated by an esample.3 k.3 .1 093 0. obtained. After 1c. plot the value of g 130 120 from column (8) as ordinate against the value of # from column (5) as abscissa 0.OlO7 0. d4 ^ .corresponding to the starting value of e. the variable terminal voltage P.5 ohms per Total external resistance =2.0223 242 O. From this point a If the graphical choice may be made of two procedures. In Table 1.0 I66 0. 5. e.rrif).0177 0. Fig. When systematized.16 amperes per circuit Resistanceper circuit = 15.0682 0. can be plotted is 160 I50 140 as a function of time by taking corresponding points from Fig. 7.0200 . method is used.1 I62 0. 6 poles Separatelyexcitede.0 I 96 0. columns (I) and (2). Columns (3) and (4) are simply steps in the determination of the total # of column (5). 7.Chapter 7 Excitation Systems 201 In other words. it is posor g shown in Fig. Let it he desired to determine the exciter response for the separatelyexcited machine mhose characteristics are given in Fig.rrir /* uy choosing particular values of ir from Fig. = 125volts Three circuitstwo polesper circuit Ceiling voltage165 volts. 167 kw. The simplest method for obtaining the area is to divide the region into a large number of increments and then sum them progressively on a recording adding machine. 6. it is found that this calculation is quite simple. of the pilot exciter and the forcing voltage then becomes (e. in the expression for the forcing voltage should be replaced by the voltage e.1697 0.0 I60 0. siblC to plot # as a function of __ e.0244 0.0 I37 0. = (20.8 ohms External resistance circuit =8.0 I50 0. 5. Equation (2) can be transformed to dt=L from which ff e.0 I43 0. for exam . 7. starting from 1c.8 ohms # per pole nt 125volts due to useful flux = 18 Total + per pole at 125 volts =20. 1200 rpm.2566 /I 0 0 2 FIELD 4 CURRENT 6 PER CIRCUIT 0 Fig..0 I45 0.18)2 = 1 05 4. which is illustrated in Fig.0 I 55 0.2 I 0 I 0. 7Example for calculation of response of exciter.0367 0. If the machine is separately cscited. 5.0532 0.2324 0.1 334 0. The difference in these forcing voltages shows why separatelyexcited exciters are usually faster in response. if nt ceiling voltage=8. ~.3ohms Two field windings=0. Columns (6) and (7) are likelvise steps in the determination of 2 of column (8). the in§ion is a stable operating point. SAuxiliary curves for calculation of response ple given in Fig. From Eq.0107 0.
(1) decreasing the time constant of the field circuit.34 6.2 3.078 0. such as being more stable at low voltages. It is necessary to add more resistance to the external circuit so that the resistance per circuit remains the same.0728 0. is to annul the effect of the crossmagnetizing mmf of armature reaction. The former is usually accomplished by paralleling the field circuits placing at the same time resistors in series to limit the current.is onehalf and.9 4.7 05..6 3.1 ) 1.202 (4) klif Excitation Systems TABLE 1 (‘3  Chapter 7 (7) e. one side being positive and the other side negative.5 3.66 5.3 81. the terminal voltage rises twice as fast. 9Correction factor to be applied to calculated response to include effect of eddy currents. if the parallela are doubled. For higher speeds of response. A.& 0. 9 by W. for machines without compensating windings. of (7) 0. Thus.6 6.8 31. Associated with these changes one usually finds that the field i current of the alternator has increased a considerable .5 56. 10 (b) between CA represents the mmf distribution along the pole face.1581 0. then because of the linearity of QN of Fig. However.8 42. in the armature circuit of the synchronous machine.0409 0. Lewis..0 125C(6) 74.0208 0.30 3.0955 0.6 7. constitutes the mean of elementary areas. the eddy currents produced in the solid yokes can retard the buildup of the flux.152 giving the curve shown in Fig. which is wound into the pole face of the field winding. the mean of excited exciters or the ceiling voltage in the case of selfexcited exciters.2 21. faster in response than selfexcited exciters.0174 (9) x (10) .9 3:3 3.6 46. 8.’ in Fig. the same integration can be accomplished in tabular form.0186 0. 10 (a) in which MN represents the maximum magnetizing mmf at one pole edge and PQ represents the maximum demagnetizing mmf at the other pole edge.0462 0. Lewis’. amount.70 4. On the other hand. voltages at which selfexcited exciters may have a tendency to creep.014 0.5 3.8 51. They do. the summation of the At column. A.1 esriif (8) dt 4 100 110 120 130 140 1. Because of the high inductance of the field circuit of the synchronous machine.46 0. If A and C are so laid off that OA and OC equal MN and PQ. When current flows in the armature.8. 10 (a).0146 0. This curve supplies a correction to be applied to calculated responses. Fig.05X(2 .0403 0. have other advantages.3 4. Column (12) is merely a progressive summation of (11) and gives actual time.rfif =. Further. Likewise.288X(1 28.9 1. therefore.50 155 160 3.0231 0.80 7. By plotting column (1) against column (12). In Eq. such as short circuits. The extent to which this is effective is given by the curve (r 0 CALCULATED NOMINAL RESPONSEUNITS Fig.6 38.7 34. respectively.0162 0. except for the first element for which A# is only 0.343 0. however.0 5. but not necessarily.0603 0. This is done to obtain convenient divisions.4 45.2309 0. perhaps in excess of the armature current rating of the exciter.14 4. Time can then be determined by integrating this curve.5L 0. 10 (b) represents a section of the noload saturation curve in whic’h 0 represents the generated voltage on the vertical coordinate and the field mmf on the horizontal coordinate. This effect is shown in Fig.0476 0.1 7. the difference of successive values of 1c/from column (5).+ (3) 3.0 rec.4 61. One method of doing this is by means of the table constituting the insert of this figure.0287 0.6 63.112 0.2 44.0342 0. Improvement in speed of response can be obtained by two general methods. This is found by dividing $J into increments of unit width. 10 (a).x(11)0 0.__. (7) the only change is that 1c.3 43.7 43.0 49. Continuing in Table 1. the phenomenon of armature reaction must be taken into consideration except for those machines that have a compensating winding. 4. the number of poles and likewise $ per circuit ’ are halved.01. since the generated voltages are propor .8 104.6 53. the armature current of the exciter can usually be regarded as remaining substantially constant at this increased value during fhe period for which the response is desired. column (lo). the mmf of armature reaction produces an mmf that varies linearly from the center of the pole piece. Separatelyexcited exciters are usually. from Fig. and (2) increasing the pilotexciter voltage in the caseof separately Most of the cases for which the exciter response is desired are concerned with sudden changes. according to W. Calculation ditions of Response Under Loaded Con successive values of column (8). The product of these two values tabulated in column (11) is the increment of time. The second column represents time.2 0:.091 from (5) from (8) 0. Increments of time At are enumerated in the first column.5 68.7 53.8 (4) 15 3x(2) 50. the response curve is obtained.0 11. the abscissa of Fig.5 40.573 A  32. 8. column (9).26 6.4 71.421 0.0626 0.9 2.0134 0.3 29.0 114. constitutes the base of increments of area of curve $ in Fig. The function of the compensating winding.5 37.3 35.
For machines with compensating windings.” such as shown by the dotted curve of Fig. . Effect of Differential Fields on Response tional to the airgap fluxes. loEffect of armature reaction in reducing total flux across gap. 11. c=number of turns per pole of the main winding. such as shown in Fig. and Rd in series with the combined main and differential windings.. N = total number of poles of exciter. 5. With these increases the resistances of each of the main and differential circuits will be designated by the . [MN FROM (O)] machine is very much smaller under load than under no load. the total flux and consequently the generated voltage are decreased from the value indicated 1)~ CA to that indicated by FG. If the differential windings are not opened when the regulator contacts close to produce field forcing. = current per circuit of main winding.Chapter 7’ ‘1x. the forcing voltage (~r&r) for a selfexcited Differential windings are provided to reduce the exciter voltage to residual magnitude or below.. this effect is negligible. For separatelyexcited exciters. b = number of parallel paths in the differential winding. This effect is most pronounced in the region of the knee of the saturation curve as at both higher and lower field currents. respectively. 12 (a) shows schematically such an arrangement. There remains only the distortion effect to consider which amounts to only several percent. which is obtained by integrating the area under the curve DOB and drawing FG so that the two triangular areas are equal. d = number of turns per pole of the differential winding. resulting in a load saturation curve for constant current. (b) Section of noload saturation curve. the load response can be calculated in the same manner as the no load response. Fig. 11Load saturationcurves armature for exciter current. From this same curve it can be seen that for a given field resistance line. 11. so connected that the mmf produced thereby is opposite to that of the main windings. Fig. Excitation Systems NO LOAD SATURATION CURVE / 203 FOR CONSTANT ARMATURE CURRENT TERMINAL VOLTAGE FOR CURRENT (a) I RESISTANCE LINE ii SECTION TAKEN Fig. (a) Shows distribution of armature mmf. respectively. The extent to which this is effective may be calculated as follows: Let a=number of parallel paths in the main winding. In calculating the flux linkages in accordance with Eq. the differential circuit reduces the response of the exciter. the forcing voltage remains unaltered by the loading on the machine as it is independent of the terminal voltage. As a result. the section of noload saturation curve shows the effect of the superposed armature mmf upon the density of airgap flux across the pole. Except for these two changes. and bRd. They consist of a small number of turns wound on each pole. The higher mmf does not increase the flux on the righthand side as much as the lower mmf decreases the flux on the lefthand side. may be included in the calculation by increasing the actual resistances in each of the main and differential circuits by aR. The terminal voltage is reduced still further l)Y the armature resistance and brush drops.. id = current per circuit of differential winding. assuming constant REACTION AT EDGE OF POLE. i. there is a tendency to add on the one side of the pole just as much flux as is subtracted on the other. (5). the distortion curve should be used fore. The extent to which the average flux or voltage is decreased can be indicated by a “distortion curve. The resistors R. The armature resistance can be regarded as part of that of the main field winding.
+” d rd> I. 13 is of conventional construction so far as mechanical details and armature winding are concerned. and rr by [ I+R(. i. = Ra(i+$) ex ~~ multiplying (13) by (12) $ (13) by + (Rb+r&+$ 2. ‘&=a. then J/=$X (flux per pole in lo* lines) REVERSIBLE VARIABLE VOLTAGE $d = $JX (flux per pole in lo* lines) or (10) STABILIZING If it be assumed that the two windings be replaced by another winding having the same number of turns and circuit connections as the main windings. . and field 3 is a small capacity batteryexcited stabilizing field.+$)]/A. 13Schematic diagram of threefield main exciter.)(i+~id)+Rbid+d~ e. Field 1 is connected in series with a variable resistance across the MAIN EXCITER in which $ and $d are the flux linkages in each of the two respective circuits. 4=1g jrm+f$a)!?] (15) Iim id 1 lb) Fig. then MAIN WINDINGS ex=(Ra+r. $f can be eliminated 1 . In other words. i be the current read from the saturation curve.. the calculations should be carried out as though the differential winding were not present. and the voltage across each circuit be multiplied by the coefficient of e. respectively. except that instead of using the expression (e. should be multiplied by (l2)/ A. ThreeField Main Exciter i +g lnrn dt i +a dd dt 63) co The threefield main exciter shown schematically in Fig. If all the field flux cuts all turns.204 Excitation Systems Chapter 7 If (10) and (11) are inserted in (8) and (9). d. an d rd. 12 (b) the following equations can be written eI =R(ai+bid)+r G=R(ai mfbi d )+r 6. symbols r. e.ad c SELFEXCITED FIELD RHEO ‘I . The current id can then be solved in terms of i. and differential Equation 14 shows that the ordinary fluxlinkage curye for the exciter and conventional method of calculation can ’ beusedif the coefficient of i be used as the resistance of each circuit. then the instantaneous mmf of this winding is the same as that of the combination if its current.A.. field 2 is a1 separatelyexcited controlling field. . is .i+g m clt (14) . DIFFERENTIAL WINDINGS (a) subtracting from (12). but it is built with three electrically independent shunt fields. Referring to Fig. Upon substituting the expression for id into (12) there is finally obtained that lr4 !i e.r&r) todetermine the forcing voltage. 12Schematic diagram for main windings.i LOWENERGY DC SOURCE MECHANICAL INTERLOCK BETWEEN RHEO ARMS # :a 2. in which c rd = A l+R ( .: (” from which &Cifdid C *(ll) Fig. Field8 1 is selfexcited and provides base excitation.
Field 3 is used only when the esciter speed of response or range of voltage output makes it desirable. the separatclyexcited field 3 carries a negligible current. The trouble might involve the variablevoltage source for field 2 or the voltage regulator that controls it. If the resistance in the circuit of field 1 were increased to give a value of ampereturns less than Od in Fig. the machine would be unstable as pointed out in Sec. It is capable of supplying 5 to 10 percent of the normal total excitation requirements of the main exciter. but is generally high enough to supply sufficient field current to the ac generator field to mainta. 14 and can be esplained by assuming that the current in field 2 is zero. but even though the current in the field should become zero. and the generated voltage due to field 3 being energized is rrprescnted by hJ The ampereturns of the two fields :tntl the generated voltages add so that the distance Of is t. L . when the selfenergized shunt field is carrying a high excitation current. the total tciminnl voltage can be plotted as a function of the ampereturns in field 1 alone and is represented by the curve ebb in Fig. These are usually motoroperated under manual control. Operation in this region is the same as a selfexcited exciter. The combined effect of fields 1 and 3 is shown in Fig. Thus. the current in field 3 is increased. the resistance in the selfexcited field circuit would be increased to reduce the ampereturns produced by that field to Oj. 1. either increasing or decreasing the exciter terminal voltage. and if field 1 were the only field excited.in steadystate stability. such as Of. Since the effect of field 1 is that of a conventional selfexcited machine. 1 in that control of the exciter terminal voltage is not completely lost if any trouble should occur in the separatelyexcited field circuit. Operation at smaller values \vould not ordinarily be necessary except in the case of a synchronouscondenser csciter. 14Equivalent noload saturation curve of threefield main exciter showing effect of stabilizing field 3. The rheostat arms are mechanically connected together so that resistance is added in one field circuit as it. and by the voltage regulator controlling the input to field 2 under regulator control.chapter 7 Excitation Systems main terminals of the exciter and operates in the manner of the selfexcited field discussed in See. Field 2 is opencircuited. A]] cscitation is supplied by field 1. The voltage Wrcsented by Oe is usually less than 10 percent of the raxi voltage of the esciter.lic total terminal voltage. 14. The threefield main exciter has an advantage over the singlefield separatelyexcited main exciter described in Sec. is removed from the other. and its purpose is to provide exciter stability at low voltage output under hand control. The polarity and magnitude of the voltage applied to field 2 are then regulated so that the flux produced by field 2 either aids or opposes the flux produced by the base excitation in field 1. Excit. in the opposite direction. a small amount of energy input to field 2 can control the output voltage over a wide range. Since the current in field 3 is controlled by the amount of current in field 1 through the mechanical coupling of the fieldrheostat arms. the manuallyoperated field rheostat in field 1 circuit is set to provide some base amount of excitation. at the same time the current in field 1 is reduced. These ampereturns would cause a generated voltage eclual to Oh. Thus. When the voltage of the main esciter is under the control of a voltage regulator that varies the magnitude and polarity of voltage applied to the separatelyexcited field 2. the exciter will continue operating at a terminal voltage determined by the setting of the rheostat in the selfenergized field circuit. manual control of voltage is possible over the complete range necessary. The current in field 3 is usually negligible with such asetting of the rheostat when the generatoris carrying any load. and the relation between tlic exciter terminal voltage and the total field ampereturns is represented by the line ab.er polarity can be reversed by reversing both field circuits when the currents are zero and building up ” ~* 1 d CURRENT IN SELFEXCITED FIELD Fig. Fields 1 and 3 have rheostats in their energizing circuits. 14. If the fieldresistance characteristic of the selfexcited field is plotted on the same curve. there will always he a positive point of intersection between the resistance line and the saturation curve ekab and stable operation can he obtained for any voltage greater than Oe. To obtain a terminal voltage less than Oc. When the field rheostat is adjusted to give a voltage output greater than that represented by the distance Oc. Smaller terminal voltages are obtained by holding the current in the selfexcited field to zero and reducing the current in separatelyexcited field 3. This field also provides for stability of the exciter when the voltage regulation is under control of the voltage regulator. The only effect on the ac generator would be a change in its internal voltage which would cause a change in reactive loading of . Field 1 provides the base excitation for the machine. thus. Field 2 is a shunt field that is excited from a reversible variablevoltage dc source under control of a voltage regulator. This setting is determined by the operator. However. 1. The operation of the threefield main exciter is made stable by separate means for the two conditions of operation: by a separatelyexcited stabilizing field under manual control. Field 3 is a small separatelyexcited shunt field that obtains its energy from a station battery or any other source of substantially constant dc voltage.
RP ’ E3’ =Fixed voltage applied to field 3. substituting from Eqs. between 7. 2.. The voltage applied to each of the field circuits is &J absorbed in Ri drop in the circuit resistance and VTz drop P = Number of poles.’ is supplied from a constantpotential source. and added. Also. ohms. t _ LI _ PKN12X ’ RI 7’ Gw Equations similar to Eq. (25) = Ampere turns per pole (26) GO = R&o+ NMPAT leakage flux X = Flux leakage factor = 1 + useful flux Subtracting the two sets of voltage equations. i = Change in amperes in field winding = i’iO. (30).pI(N 1 2X. The following analysis presents a means of replacing the assembly of several fields with one equivalent field so that the response can be calculated. If Eqs.o+N~~~o+NE~Eo>. . (18) and (20) is p = Differential operator z d The initial or steadystate value of total useful flux per pole is ~o=K(N. and (29) are multiplied by RI ’ ?? Ez’ =Voltage applied to field 2. 13. 0 = R3i3 N&Ppb f (29 Et0 = Initial value of terminal voltage. = =Maxwells X lo8 per pole R = Resistance of the complete field circuit. the total useful flux at any later instant of time is 4’ = K(Nlil’ +:V&’ + N&‘).o = R&o+ N. 4 = qS. Ei . Eat= R2i2’+ N.=[(l&)+(t1+t2+tdP]dk (31) . K = Flux proportionality constant (24) CO= c+~= Rho+ NJPpdu Maxwells X lo8 per pole E. instant of time are $J’=Total useful flux per pole in Maxwells times 108.. The voltage equations at any N = Number of turns per pole in the field winding. Thus. the source of excitation for the ac gencrntor field would be lost and a shutdolvn of the unit would be necessary. The method uses stepbystep integration to take into account the saturated condition of the exciter. seconds. terminal voltage.EaO 0. when the total L = Inductance of field winding in Henrys. i and since the flux is expressed as Maswells per pole times 108. + N. EBI= R&s’+ NJPp$‘. . The basic formula for the selfinductance of any of the field circuits is L=Ndlo8 henrys. & = Initial useful flux per pole in Maxwells times 108. respectively.$0 = K(Nlil+ N&+N&). N1 is the turns per pole of field 1. the result obtained after and E’ EZ= Change in voltage applied to field 2. and the separatelyexcited field 2 as shown in Fig.XPpqS (22) i’ = Total amperes in field circuit. The three fields are wound to form a single element to be mounted on the field pole. (19) can be written for selfinductances Lz and L. The specific fields involved in the threefield main exciter are the selfexcited field 1. Nz the turns per pole of jield 2. Under similar circumstances of failure Fith the singlefield esciter. KN.’ =Terminal voltage applied to field 1. cc+ Rli. KNz Et = Change in terminal voltage = E. = R2i2+NJPp+ (2% E.XPp&.=Initial amperes in field circuit. useful flux is constant and p&=0. (17) The change in total flux per pole is the difference these two values. damping currents flow in those fields during voltage changes. in the circuit inductance. so that the mutual coupling is high and can be assumed to be 100 percent with small error. a set in c = Voltage proportionality constant terms of changes from steadystate conditions is obtained. (28). the batteryexcited field 3. If additional fields are present. [27). (16) When the field currents are changed to force an increase in 11 @iPNJ &&=~+(tl+t2+tdP~.E.. the same leakage coefficient can be applied to each of the fields. (20) for time constants t2 and t8. E.~. (19) * The time constant of the field circuit is the total selfinductance divided by the tot’al resistance. assumed to be connected in series. Their effect is to reduce the rate of change of flux in the exciter iron paths.206 Excitation Systems Chapter 7 the machine. Calculation Exciter of Response of ThreeField Main (18) X method of calculating the response of a singlefield esciter is given in Sec.’ = c$’ = Rlil’+N1XPp4’ (21) 4 = Change in flux per pole = 4’ do. terminal volts Since the voltage E.XPp+ = (27) t = Time constant of complete field circuit. ~. 2 (30) Rearranging the terms in Eq. In the following symbols the subscript indicates the particular field to which the symbol applies. During the initial steadystate conditions. etc. and similar to Eq. E. the selfinductance of the circuit of field 1 becomes L 1 = p4N1x il .?&=Initial value of voltage applied to field 2. (23) i.
L. 156Selfexcited singlefield equivalent exciter. the requirements to be met are given by Eqs.&o+N. Eeofc40 = R. The three fields on the exciter can be assumed to be replaced with a single equivalent selfescited field as shown in Fig. can be used provided the appropriate value of R. 2 . (31). refer to all coils which are self excited..Chapter ‘7 Excitation Systems Eliminating : by using Eq. the machine constants are appropriately adjusted to new values applicable to the next small interval of time. (40) 207 JVhen solved. The sum of the time constants should also include a value for the frame slab. the flux change can be calculated for the second interval and will he the same by either equation. Thus. (31) in the general form becomes ~~=[(lc~&J+Pq#J El+@‘= R. (44). and eddy currents in the slab cause a delay in the flux rise. (46). For dc machines of the size used as main exciters. = L e= PdNeX = pKN. which acts as a shortcircuited turn. The above equations can be generalized to the case of a machine having any number of the three types of fields considered.ZX 20 t =PKNA e R. it can be assumed that the machine constants do not change. The constants of the equivalent selfexcited field are determined from the following: t. If no regulatorcontrolled fields are present. e (41) The applied regulator voltage is Ee=$$. The quantities referring to the equivalent e E (42) Fig. is determined by Eq. 15. and N. and by comparing similar terms.L t. and t. it is derived that te=t1+12+t3 (46) (39) Ne=$te.ion may be used to determine the constants of the equivalent field. (36) c Re+f e and the regulator voltage to be applied (47) If no selfexcited fields are present in the machine. the only requirements to be satisfied are given by Eqs. E.‘+N. Equations (38) and (31) can be solved only if saturation is neglected. The various time constants for the machine in the unsaturated condit. and N.’ supplied by the regulator.XPpqS. refer to all coils to which regulator voltages are applied. for a small interval of time. Letting t. (32) from (33) E. However. and the change in flux calculated by either equation will be the same. (44) and (47). Using the relations q5 KN.+N. (36). the equivalent field is not selfexcited and has applied to it only the regulator voltage.+cc$= R. N.2 second. (45). (43) Subtracting Eq. (32) (33) (34) (35) (36) (37) i\t any instant of time. Eq. (31) expresses 4 and hence the terminal voltage as a function of time if saturation and the consequent change in constants are neg!ected.=Zlt (44) N. of threefield main ficltl arc designated by the subscript e. The selfinductance of the equivalent field is given by Eq.XPp$‘. Eq. (36) and (46).i. is calculated from Eqs. If at the end of the first time interval. the flux rise cnlculated from the equation for the single equivalent field by using the normal stepbystep methods that take into account saturation will be the same as the actual flus rise with the assembly of several fields.i.. The field has applied to it a voltage equal to the terminal voltage c+’ plrw an equivalent voltage E. and (45) IQ. When no selfexcited fields are present. the frameslab time constant may approach 0. Iking stcndystate conditions. Any value of N. (34) reduces to (38) Equation (38) is of the same form as Eq.’ where 8t= sum of time constants of coils of all types. and the resistance is Re=$.i.XPp4o.
. In the case of the Rototrol. The voltage E is determined by the source of voltage under regulator control. and the main exciter Rototrol is of the twostage type. However. are of conventional 3GOOrpm exciter construction. if the regulated field is a selfexcited field. the reduction in speed is compensated by an increase in the total flux._. and the equivalent field is a selfexcited field with no regulator voltage applied. In any event. For example. The slower the speed of a generator. the amplification factor can exceed 10” depending upon the design of the machine. In such machines. overcome the effect of the control field. but the electrical connections are quite different. A schematic diagram of the mainexciter Rototrol is shown in Fig. The mainexciter Rototrol has not been built in capacities large enough to supply the excitation requirements of large slowspeed ac generators. The mainexciter Rototrol is not adaptable at present to use with generators operating at less than 1200 rpm. and there are a number of speciallyconnected field windings to provide the high amplification factor. Fig. The field connected between terminals F3F4 is called the control field. but when used with 1800. 8. MainExciter Rototrol The most recent development in the field of rotating main exciters is the adaptation of the Rototrol rotating amplifier as a main exciter. its excitation requirements also increase and a larger controlling energy is required. and the equivalent schematic diagram is shown in Fig. I . 250volt. The Rototrol can be built with one or more stages of amplification. the volt. Any generator is in fact a “rotating amplifier” in that a small amount of energy input to the field is amplified to a large energy output at the generator terminals. therefore. Thus.or 1200rpm generators. the change in field energy required to produce loopercent change in output energy is usually within the range of 1 percent to 3 percent of the machine rating. A 210kw. commutator. As far as external circuits are concerned.. The mechanical details such as the enclosure. Reasoning identical to that in Sec. . 17 (a). the limits field is energized by a device for limiting the maximum or minimum output or both. Although this characteristic is undesirable in the selfexcited generator..208 Excitation Systems Chapter 7 (47). 1. The circuit between terminals F5F6 energizes a field similar to the control field. the amplification factor might be between 30 and 100. under certain conditions. and the line terminals supply voltage to the load in series with the series field. operation of a conventional selfexcited dc generator is unstable when the fieldresistance line coincides with the airgap line of the saturation curve as shown in Sec. The output terminals are LlL2. 17 (b). The armature winding is of the lap form but has no cross connections. 250volt. the mainexciter Rototrol can be represented as shown in Fig. *. A detailed discussion of the theory of operation of the Rototrol is beyond the scope of this chapter. This field operates in the same manner as the control field in controlling the Rototrol terminal voltage but it is called the limits field. and it also appears on only the two south poles. the Rototrol is directconnected to the generator shaft. The twostage mainexciter Rototrol can be built with sufficient capacity to supply the excitation requirements of the largest 3600rpm generator. the voltage E becomes equal to the exciter terminal voltage at each instant of time._. The control fikld is energized by the voltage regulator and normally has control of the voltage output. The windings energized by the circuit between terminals FlF2 are shuntfield windings used for tuning purposes as discussed later. etc. brush holders. 1 can be applied to a seriesexcited generator where the selfexcited winding is in series with the load and both the load and the field can be considered as a shunt across the armature. The principal field of application is with 3600rpm turbine generators. and can be found in the References. for The excitation requirements. 1 and 3. In the ordinary dc generator. and it should be noted that the circuit between the brushes of like polarity energizes additional field windings that are compensating and forcing fields and also serve as series fields. 17 (c): the control field is energized by some excitervoltage controlling device. the limits field is energized by devices that restrict the maximum or minimum voltage output. are greater for slowspeed generators. . 4pole mainexciter Rototrol directconnection to generator shaft at 3600 rpm.age response can be calculated by the stepbystep method of Sec.. For a given voltage output. as the Rototrol rated speed is decreased. 2. 16. However.I. the larger the physical size. The. requiring a larger volume of iron to maintain the same flux density. Using this equivalent singlefield representation of the multiplefield main exciter.. the name rotating amplifier has been specifically applied to a form of rotating machine possessing an unusually large amplification factor. 16A 2lOkw. 3600rpm mainexciter Rototrol is illustrated in Fig. The discussion here will be confined to a description of the operating principle as it applies to use of the Rototrol in excitation systems. The combination of these factors has largely restricted the use of the mainexciter Rototrol to directconnection with 3600rpm turbine generators. Furthermore. and to all outward appearances it is a conventional type of dc machine. it is an important part of the Rototrol principle. the change in input energy to the field is a small fraction of the resulting change in energy output of the armature. so that the limits field can. and windings appear on only the two south poles. the maximum rating of generator is restricted.
The Rototrol is operated on the straight portion of its saturation curve and the adjustments necessary to meet this condition are termed tuning of the Rototrol. the control field forces the change in ampereturns required to stabilize the machine or to change and establish the terminal voltage required for a new load condition. but can also be done by varying the air gap between the field poles and the rotor surface. which shifts the position of the airgap line. The ampereturns of the selfexcited field and those of the control and limits fields are superimposed. the selfexcited field of the Rototrol furnishes all of the ampereturns required to generate the terminal voltage. A . complete schematic diagram and equivalent representations. 17Twostage mainexciter Rototrol. The seriesfield current then is directly proportional to the armature voltage in the same way as the shuntfield current in the selfexcited shuntwound machine. two south poles. The resistance of the shuntfield circuit is adjusted to change the position of the terminal voltageseriesfield current relation to tune the machine perfectly. Exact coincidence of the resistance line with the airgap line cannot always be obtained by these two means so a smallcapacity shunt field is provided to serve as a vernier adjustment.Chapter 7 Excitation Sosterns COMPENSATING L FIRST STAGE LSECOND OF TWOROTOTROL STAGE (b)EQUIVALENT CIRCUIT STAGE MAINEXCITER F2 LI c5 F3 LEGEND + e @ 0 0 @ SHUNT FIELD COMPENSATING AND FORCING FIELDS SERIES FlELDS COMPENSATING FIELDS 1 LIMITS FIELD CONTROL FIELD INDICATES DIRECTION OF CURRENT INDICATES DIRECTION OF FLUX DUE TO CONTROL FIELD INDICATES DIRECTION OF FLUX DUE TO LOAD CURRENT BRUSH POLARITY FROM CONTROL FIELD EXCITATION BRUSH POLARITY INDUCED BY ARMATURE REACTION THIS REACTION IS THE RESULT OF THE CURRENT FLOWING IN THE ARMATURE BETWEEN THE POSITIVE BRUSHES TERMINALS kkCIRCUIT EXCITER REPRESENTATION ROTOTROL OF MAIN (alSCHEMATIC DIAGRAM OF ROTDTROL EXCITER SHOWING ALL FIELDS EXCEPT INTERPOLE FIELDS Fig. It is particularly significant that under steadystate conditions. I)Porating Principle of the MainExciter Roto The fundamental principle by which a small amount of energy in the control field forces a large change in Rototrol output is that of unbalancing the ampereturns on two poles of like polarity. 9. current in a given directron in the control field will weaken . Thus. However. in this case. the seriesfield circuit is tuned so that the resistance line of the circuit coincides with the airgap line. This is usually done by adjusting the resistance of the load or an adjustable resistance in series with the load. and the algebraic sum of the ampereturns on all of the Rototrol fields determines the terminal voltage.
210 Excitation Systems Chapter 7 (a) SIMPLIFIED WINDINGDEVELOPMENT DIAGRAM OF TWOSTAGE RoTOTROL SHOWING CURRENT FLOW WITH CONTROL FIELD ENERGIZED BRUSH POLARITY FROM ASSUMED CONTROL FIELD EXCITATION DIRECTION OF CIRCULATING CURRENT BETWEEN POSITIVE BRUSHES MAGNETIZING EFFECT OF LOAD CURRENT FORCING FIELD . . DIRECTION OF CIRCULATING CURRENT BETWEEN POSITIVE DIRECTION OF LOAD CURRENT DIRECTION OF CIRCULATING CURRENT BETWEEN THE NEGATIVE BRUSHES MAGNETIZING EFFECT OF LOAD CURRENT DIRECTION OF LOAD CURRENT BRUSH POLARITY ‘INDUCED BY ARMATURE OMPENSATING FIELDS COMPENSATING FIELD FLUX IZING EFFECT OF RRENT IN MMPENI (d).ROTOTROL EXCITER CONTROL FIELD WINDING (~1 ROTOTROL EXCITER FORCING FIELDS CONNECTED IN SERIES BETWEEN THE POSITIVE BRUSHES DIRECTION OF LOAD CURRENT. 18Principle of operation of twostage Rototrol.ROTOTROL EXCITER FORCING AND COMPENSATING FIELDS CONNECTED IN SERIES BETWE’EN THE POSITIVE BRUSHES b 0 b)RoToTRoL EXCITER C~MPEPJSAT~IUGFIELDS CONNECTED IN SERIES BETWEEN THE NEGATIVE BRUSHES FLUX DUE TO LOAD CURRENT FLUX DUE TO CONTROL FlELD BEING ENERGIZED Fig. DIRECTION OF CURRENT LOAD CURRENT FORCING FIELD FLUX FSI 1 F4 (b).
2.ive brushes and minimizing the armature reaction that would oppose the control field. Current flowing in the firststage machine sets up an armature reaction represented by a twopole armaturereaction generator. ?‘hc windingdevelopment diagram of Fig. . The forcingfield current also flows through the armature \\intling as shown in Fig. 18 (c).he conditions of Fig. however. 18 (d) oppose the armature reaction caused by current between t. For controlfield current in the direction shown. Reversing the polarity of the voltage appliccl to the control field would reverse the effect and (::L~~~c decrease in terminal voltage. by describing the sequence of events for a given oprr:lting condition. so the current directions in the armature conductors are as shown.)otl best.he positive brushes. Two of these windings appear as compensating windings on the armaturereaction generator since they further compensate for the armature reaction produced by the current between the positive brushes. 18 (a). two of them are twopole machines and the third is a fourpole machine. Tracing the circuit of the load current reveals that the load current must flow through the forcing and compensating fields. 18. l?urther analysis shows that the positive brush of higher potential is always under the south controlfield pole for t. the flux direction llnder pole 1 is out of the paper and under pole 3 is into 1he paper in Fig. which is in the direction to increase the terminal voltage of the machine.Chapter 7 Excitation Systems 211 one sollth pole and strengthen the other.). while the magnetizing effects of the unbalance currents add. The difference in potential between the two positive brushes caused by current in the control field is represented as a twopole generator excited by the control field and is the first stage of amplification in the Rototrol. an armature r(%t. The mmf’s produced in the armaturereaction and compensating fields are in opposition. 18 (c). The result is that the positive brush under pole 1 is raised to a higher potential than the positive brush under pole 3. and by virtue of the form of the armature winding. The two conductors in a wmmon slot u&r poles 2 and 4 carry currents in opposing (lirections. the forcingfield mmf’s decrease the flux densities. Thus. as shown in Fig. As stated in Sec. The compensating windings in series \\ith the forcing fields in Fig. 7. The resulting una t&nce of the southpole fluxes causes a phenomenon that is srlppressed in the usual dc generator. 17 (b). and the current flowing between these brushes energizes additional compensating windings on all four poles. The remaining two compensating windings compensate for the armature reaction caused by the current flowing between the negative brushes. For clockwise armature rotation. and the operation can be underst. thus. These windings produce the proper mmf in the commutating poles to assist commutation of the current in the armature. Currenttlircction arrows and corresponding fluxdirection arrows are shown in Fig. The Rototrol is represented as three separate generators. which shows all of the field windings and the current and flux arrows for the assumed condition. The current is in a direction to cause an increase in t 11c terminal voltage of the Rot. Thus. and that is the Ilnt&nce of voltage generated in the armature when the magnetic flux clensit. SeriesField Effect in MainExciter Rototrol The definition of mainexciter response ratio given in Part I does not apply to main exciters having series fields. causes a difference in pol:irity between two brushes of like polarity. holding to a minimum the voltage difference between the ncgat. The resulting current flow between the two negative brushes would cause an :lrmature reaction in opposition to the control field.ies in the field poles are unequal. 18 (a) is drawn for the controlfield flux in the direction shown in Fig. (d). The output of this machine is fed into the field of the fourpole generator which is the second stage of amplification. 17 (a). the series field of the Rototrol supplies all of the . The fourpole field windings are the forcing fields of the Rototrol. a set. li (a). So far as the controlfield flux is concerned. and (&). stage. The overall effect of current in the control field is sholvn in Fig. In addition to the field windings described above. 18 (aj. The armature reaction is represented by a field exciting this generator and the compensation for armature reaction between the positive brushes is another field on this same machine. 1s (c). carry currents in a common direction. The relative polarities of the two positive brushes are. pole 1 is a south pole and pole 3 is a north pole. A group of compensating fields are also connected in series in the circuit between the negative brushes. the fluxes produced by the forcingfield windings are in a direction to increase the flux densities in all four poles as shown by the openheaded flux Uows. the response ratio of the mainexciter Rototrol cannot be stated in the conventional manner. 18 (d). of commutatingpolewindings are included in the Rototrol. With the opposite controlfield polarity.i current is shown flowing in the control field in Fig. The potential difference between the two positive brushes is used to energize another special field called the forcing field. The (#cct of the unbalanced south poles on the armature winding can bc analyzed by assuming the unbalanced fluxes are the only ones present in the machine. 18 (b). as indicated by the encircled polarity marks. This is verified in the circuits of Figs.ion is developed which is in the direction to weaken north pole 2 and strengthen north pole 4.3. the equivalent circuit of the twostage mainexciter Rototrol. The conductors under poles 1 and 3. greatly rr(lucing its effectiveness if compensation were not provided in some way. ~8 (1. this armature reaction being in opposition to the control field exciting the first. I hc conductor moves under poles 1. and 4 in that order. The coils are wound on the field poles in such a direction that the load current cancels so far as any magnetizing effect is concerned. therefore. All of these currents and fluxes are summarized in Fig. except that the unbalance in generated voltage appears between the two negative brushes with polarities as shown ‘)Y the encircled marks in Fig. 10. These are shown in Fig. The effect is similar to that caused by current flow in the control field. The armature reaction establishes a potential difference between the negative brushes as shown. ant1 serve a purpose similar to that of the compensating fields between the positive brushes.otrol and produces fluses as sho\vn by the fl~~x arrows to strengthen south pole 1 and \vc:&rn south pole 3.
20Simplified from a sixphase circuit of electronic main exciter supplied alternator directconnected to the main generator shaft. The function of a quickrcsponsc excitation system is to incrcasc the exciter voltage as rapidly as possible under such conditions. a current of appropriate magnitude is induced in the field winding of an ac generator when there is any change in the terminal conditions. SWITCH IGNITRON TUBES IIS GROUNDING I JEGATIVE IERMINAL I POSITIVE TERMINAL 6 NEGATIVE TERMINAL Fig. AFI ALTERNATOR FIELD TERMINALS A a H2 I I I RECTIFIER 1 TRANSFORMER R2 RI R3 RI R4 R2 R5 t R3 l R6 l 1 R4 <. . because there would be no mmf produced by the series field.\s shown in Chap. 6. but this current cannot be sustained by conventional main excit’ers because their voltage cannot ordinarily be increased fast enough. Part IT. Thus. 19Simplified circuit of electronic main exciter supplied from the ac generator terminals through a rectifier transformer. The responseratio definition also states that t.. although to a smaller extent.. Fig.‘i” b POSITIVE TERMINAL ANODE BKRS GROUNDING . the internal voltage of the generator would he at a high value when the fault is removed. a short circuit at the terminals of an ac generator induces a large direct current AC L !jUPPL. in order to keep the field current at as high a value as possible.212 Excitatio’n Systems Chapter 7 ampereturns necessary to generate the terminal voltage rmtler steadystate conditions.. The same effect t. . 1 R3 RI R5 . This occurs when the generator voltage is low.akes place. The induced current is in the same direction as the current alreatly flowing in the field circuit and serves to maintain constant flux linkages with the field winding.Y AUXILIARY TRANSFORMER rHI H3 in the generator field winding. SIXPHASE SHAFTDRIVEN ALTERNATOR R4 t AF2 . 1 1 t 7 1 . and if the induced current were sustained at its initial value. Removal of a fault or sudden reduction of the load causes an induced current in the opposite direction due to removal of the armatlrre demagnetizing effect. which would seriously hamper the rate of voltage buildup in the Rototrol. I AUX TRANSF / \ &LA (L. when a load is suddenly applied to the generator terminals. The mainesciter Rototrol benefits directly from this induced current through its seriesfield winding and immediately increases the mmf produced by that winding.he test for voltage response should he macle under conditions of no load on the exciter.
excitation for the six (a) Selfexcitation using a t. Excitation Systems 213 SIXPHASE ALTERNATOR FIELD RIR3 R5 THYRATRON RECTIFIER TRANSFORMER 90 S6 s4 ~2 S3 54 s3 S5 S6 MOTOR OPER RHEO 1 4 QR The Rototrol terminal voltage is raised to a value that can sustain the induced current. 11. A simplified circuit diagram of an electronic exciter and (b) Separateexcitation using a threephase permanentmagnet generator and drytype rectifiers. No transformer is required when the sisphase shaftdriven generator is used as a power source. The result is that the Rototrol terminal voltage follows a magnitude dependent largely upon the induced current. Ac power taken from a separate generator which supplies power to the rectifier only. in the generator field winding. 3. Ac power taken from a separate ac supply that is essentially independent of the ac generator terminals. Thus. it is separatelyexcited. The transformer is connected delta on the highvoltage side and sisphase star on the secondary side. and so far as service as an excitation source is concerned. There issome timedelay before thecontrolfield current is effective in changing the terminal voltage.Chapter 7 FIELDFLASHING BREAKER._ . The battery is used to flash the alternator field to start operation. whereas the seriesfield effect is substantially instantaneous. The coordination of these component parts presents problems that must be solved in meeting the excitation requirements of a large ac generator.$LE. An electronic exciter consists essentially of a power rectifier fed from an ac source of power and provided \vith the necessary control. it gradually decays in magnitude. Ac power for the rectifier taken directly from the terminals of the ac generator being excited. A voltage regulator is used to hold the alternator field voltage approximately constant. The seriesfield effect in the Rototrol is a desirable phenomenon in improving the response of the escitntion system and in aiding to maintain system stability. and which has as its prime mover the 3 OL THYRATRON TUBES (0) FDRE = FIELD DISCHARGE RESISTOR SIXPHASE . If the induced current is caused by a short circuit. As the seriesfield mmf is following the induced current.NATOR CT] PERMANENTMAGNET AC GENERATOR SI W Fig. The output of a rectifier is only as reliable as the source of ac input power. the voltage regulator delivers energy to the control field to increase further the Rototrol terminal voltage. however. . and it cannot be duplicstcrl in a voltageresponse test with the exciter unloaded. When power for the rectifier is supplied by a highvoltage source such as the generator terminals. principally because they cost more than a conventional main exciter. since the generator can be designed for the proper voltage. 2. it must be reliable. 21Two methods of supplying phase alternator. In the first of these. It ennblcs the main exciter to anticipate the change in ac generator excitation voltage required. and in the second and third forms. Their use as main exciters for ac synchronous machines has been limited. protective. offers advantages over rotating types. a rectifier tmnsformer must be used to reduce the voltage to the proper magnitude for the rectifier. the electronic main exciter is selfexcited. The electronic main exciter.hyratron rectifier supplied from sixphase alternator terminals through a rectifier transformer. Three sources have been used in operating installations: 1. Electronic Main Exciters Power rectifiers of the ignitron type have been used for many years in industrial applications and have given reliable and efficient performance. since its power supply is taken from its own output. same turbine that drives the main ac generator. and the Rototrol voltage follows the decay in current. and regulating equipment. this ac source might be considered a part of the rectifier.
If an ignit. 22Installation photograph of electronic main exciter. voltage. The rectifier comprises three groups of two ignitron tubes each. Each pole of the anode breaker is equipped with a reversecurrent trip attachment and the breaker is automatically reclosed. the breaker is automatically opened at highspeed and reclosed when the arcback has been cleared. The delta primary of the transformer can be energized from the terminals of the main ac generator.214 Excitation Systems Chapter 7 Fig. The simplified circuit diagram of the electronic exciter supplied from a sixphase alternator is shown in Fig. The firing control . 10. R3 X AUX TR R4 R2 R5 FIL t lR  BIAS + RECTOXES RCSREGULATOR CONTROL SWITCH IN POSITION FOR MANUAL CONTROL Fig. the anode breaker again opens and locks in the open position to permit inspection of the unit.ron arcback should occur. highspeed anode circuit breaker. the two tubes of each group being connected to diametrically opposite phases of the sixphase transformer secondary through a twopole. Should aA second arcback occur within a short time. 23Method of controlling release of the thyratron firing circuit for ignitron tube to regulate the mainexciter tube 6 is shown. rectifier transformer is shown in Fig. 20. or from some other independent source. from the plant auxiliary power supply. Thus if a breaker is opened. both tubes of a group are deenergized.
the overload capacity of the ignitron tubes is such that the rectifier . that main exciters and excitation systems be capable of similar operation and that wearing parts be replaceable without requiring shutdown or even unloading. the electronic main exciter is designed so that it can supply full excitation requirements continuously with two of the six ignitron tubes out of service. E2 and ECAC add to Fig. cathodedisconnecting switch. The ignitron is then made conductive by current in the igniter and remains conductive for the remainder of the positive halfcycle of anode voltage. firing tubes and associated control circuit is located in one of three individual compartments of the main rectifier cubicle as shown in Fig. 20 is shown with six phases. since it is necessary to provide for cscitation of the sixphase alternator. therefore. The thyratron is made conductive when its anode voltage is positive with respect to its cathode and its grid is released. The voltage El appearing across rheostat RPB is a positive grid bias. Furthermore. The grid circuit of the thyratron firing tube can be traced from the cathode of the thyratron through the ignitron to rheostats RPB and RNB and through the grid transformer to the control grid of the thyratron. the details of the circuit are the same as Fig. the capacity is approximately 150 percent of the requirements. Ignitron Firing Circuit and DC Voltage Control EAFIRING TUBE ANODE VOLTAGE EGGCRITICAL GRID VOLTAGE OF FIRING TUBE EoncPHASE SHIFTED AC GRID BIAS VOLTAGE E. With all six tubes in service. 23. Each group of two ignitron tubes with its anode breaker. 21(b). give a total gridbias voltage represented by Ec and varying the negative bias determines the point at which the total grid voltage becomes more positive than the critical grid voltage Ecc of the firing tube releasing the tube for conduction. a Rectox supplying a fixed positive bias. however. Manual control of the exciter voltage is obtained by changing the setting of rheostat MVR which varies the negative bias. In its usual form. and consequently. 21. The circuits of these devices are shown in Fig. 10. for a wide range of control of the exciter output voltage. A permanentmagnet ac generator is used as the power supply in Fig. If the ignitron should fail to conduct for any reason. while the voltage EPappearing across RNB is a negative grid bias. Current then passes through the ignitron igniter which initiates a cathode spot and fires the ignitron. The angle cr in Fig. the thyratron attempts to carry the load current but is removed from the circuit by the thyratron anode breaker. While the shaftdriven generator in Fig. The use of a positive and negative grid bias in this manner provides for a wide range of control of the angle of grid delay. a Rectox supplying variable negative bias for manual control. A thyratron tube is connected in parallel with the ignitron through its igniter. Electronic Exciter Application Problems Modern ac generators have proven their capability of continuous operation over long periods without being shut down for maintenance. This point is determined by releasing the control grid of the firing thyratron. and it is essential that such a failure and consequent replacement be sustained without interfering with excitation of the ac generator. the manuallycontrolled negative bias Ez is replaced by a variable negative bias voltage from the regulator. 24 is defined as the angle of grid delay. which receives its power input from the same source used to supply the mainexciter rectifier. When the exciter voltage is under control of the automatic electronic regulator.Chapter 7 Excitation Systems 215 so far as the mainexciter rectifier is concerned.rolled by a sinewave grid transformer.FIXED POSITIVE GRID BIAS E2VARIABLE NEGATIVE GRID BIAS EGTOTAL GRID BIAS VOLTAGE CCANGLE OF GRID DELAY The firing circuit for each ignitron tube is of the anodefiring type as shown in Fig. The magnitude of the output voltage of the electronic exciter is varied by controlling the point on its anode voltage wave at which the ignitron tube is made conductive. and an electronic regulatorsupplyingvariablenegative bias for automatic control. 23. It consists of highquality permanent magnets mounted on the same shaft with the main ac generator to serve as the rotor and a conventional threephase armature winding on the stator. It is necessary. A complication is introducetl. 12. In the method of Fig. The ignitron and thyratron tubes in the electronic exciter are subject to deterioration and eventual failure and replacement. the excitation is provided through a sixphase tllyratron rectifier. The sinewave voltage ECACimpressed on the grid of the thyratron is delayed almost 90 degrees from the anode voltage and is connected in series with the positive and negative biases. Rheostats RPB and RNB are initially adjusted to give the desired values of positive and negative gridbias voltages. 24. 24Control grid voltages applied to thyratron firing tube. which is cont. 22. 21 (a). The output of the permanentmagnet generator is rectified by a threephase bridgetype selenium rectifier and fed directly into the field of the sixphase alternator. The bias voltages El. it can be a standard threephase unit in which case a rectifier transformer would t)c required to convert the ignitron rectifier input to sixphase. These voltages are shown in Fig. Two methods of accomplishing this are shown in Fig.
but maintenance problems soon prompted its replacement with rotating types of dc machines.0 per unit per second. This is overcome by making the rectifier power supply as reliable as possible.5second interval. which implies that the exciter voltage has not reached its ceiling value at the end of a 0. .or sixphase alternator. the input is no longer subject to variation during disturbances on the main system. fore. The actual time required for the voltage to increase from 0 to a is much less than 0. the input to the rectifier is subject to voltage changes during system disturbances. It is possible that a disturbance in the system supplying power to the rectifier may cause a disturbance in the excitation of the ac generator and a consequent disturbance on the main system.i / / / / / / / b CEILING I RECTIFIER VOLTAGE EQUIVALENT RESPONSE LINE ACTUAL / / 0 TIME IN SECONDS 0. it must be provided with a reliable source of ac power. Compared with the rate of voltage buildup of other types of dc machines. Two general classifications of pilot exciters are constantvoltage and variablevoltage types. PILOT EXCITERS 13. If the distance Oa is set equal to 1. and firingtube anode breaker are opened enabling replacement of the ignitron or firing thyrstron of any group without disturbing the continuous operation of the remaining two tube groups.5 second or 4. grounding switch.1 second. which has a selfexcited field and two separatelyexcited fields. The rate of voltage buildup is dc divided by 0. To compensate for the low voltage. this system can be used to supply the rectifier. Should a tube failure occur. When selfexcited from the terminals of the main ac generator. the dc machine which supplies the separate excitation is called a pilot exciter. Since the same philosophy applies to the system used to supply the powerhouse auxiliaries. the rectifier voltage output may be low due to the low ac voltage. it would convey a false impression. the rate of response is the slope of the line Oc.he rectifier. then the distance dc almost equals 2. 26Compoundwound conventional pilot excite. The circuit diagram SERIES FI EL0 SHUNT FIELO I TERMINAL VOLTAGE Fig. it might be considered instantaneous. When separateexcitation is used to supply power to t. It is also possible to use duplicate supply with automatic changeover during disturbances in the normal supply. A main exciter can be supplied with excitation from more than one source. but the sources of separate excitation are still considered as pilot exciters. but this is not justified normally. III. offers the most reliable solution. The line Oa in Fig. however. For the electronic exciter to be completely reliable.. If the response ratio of the electronic exciter were expressed in accord with the definition given in Part I.0 per unit. The shaftdriven three. 25Response of the electronic main exciter.bsecond interval. and the variablevoltage type is used where the pilotexciter voltage must vary to give variable voltage on the exciter field. The line Oc is drawn so that the area Ocd is equal to the area Oabd under the actual response curve during the O. the actual rate of voltage increase exceeds 10 per unit per second. Older excitation systems used a storage battery as a pilot exciter. Under normal load conditions the voltage is reduced to that required by control of the firing point. This method of compensation requires a larger rectifier transformer and means that the firing is delayed longer during normal operation. According to the definition. Response of the Electronic Main Exciter The ignitron rectifier has the ability to increase or decrease its voltage output with substantially no time delay. and there When the main exciter of an ac synchronous machine is separatelyexcited. The line ab represents the ceiling voltage. 14. that is.216 Excitation Systems Chapter 7 can supply full escitation for a short time with only two of the six tubes in service. as is the threefield main exciter.0 per unit. z 2 2 z a cz t” 3 5 / / / . The constantvoltage type is used where control of the main exciter voltage output is by a rheostat in the exciter’s separatelyexcited field circuit. the ignitron anode breaker. Thus during nearby faults on the system when it is desirable to increase the generator excitation as much as possible. CompoundWound Pilot Exciter i The most common form of constantvoltage pilot exciter is the compoundwound dc generator. the rectifier may be designed to produce normal ceiling voltage when the ac input voltage is $5 percent of normal.5 RESPONSE d Fig. the rectifier can be designed for a voltage output much higher than that required during normal operation. 25 represents the actual voltage response of the electronic exciter.
within :lcts as a stabilizing force to hold the voltage at any point which the regulator holds the voltage with steady or gradon the straightline portion of the saturation curve. is connected in series with the output circuit of the lji1ot exciter to regulate the voltage applied to the field of t1ic main exciter. either finder the control of a voltage regulator or under manual control. These were the indirectacting disturbances.suitable crosscurrent compensation provided between the state conditions.hc shaft of the main exciter. the regulator immediately applies corrective exciterrheostatic regulator and the directacting rheo. Indirectacting exciterrheostatic type 3. The singlestage Rototrol is a stabilized seriesexcited dc generator as shown in Fig. is :tlso used as a variablevoltage pilot exciter. 26. When the pilot exciter is operated :tt a speed lower than 3600 rpm.01 threephase ac generators excited by individual selfexcited exciters. The directacting rheostatic type of regulator controls the voltage by the regulator element varying directly the CONTROL regulating resistance in the main exciter field circuit. GENERATOR EXCITATION SYSTEMS tivity zone. Fig. The preceding sections have discussed the various types of main and pilot exciters in use at present. h typical SR. it. the ]<ototrol pilot esciter may be of either one or two stages of :Lmplification.action to restore the voltage to the sensitivity zone. 16. Each regulator is designed for and limited to the control of one exciter. The FIELD different sizes of SR1 regulators are suitable for the nutomatic voltage control of constantspeed. The exciter must be designed for shuntfield conLIMITS trol and selfexcited operation. SensitivityThe sensitivity of a generator voltage ti(. regardless of t.I. The principal difference tice is to provide each generator with an individual esciter. that is. Four types of voltage regulators are being used to control the excitation of synchronous machines: 1.Chapter 7 Excitation Systems 217 is shown in Fig. and l)ilot csciter all rotate at the same speed. l)ctween this and a conventional seriesexcited dc genera. Impedancenetwork or staticnetwork type 4. is of the twostage type. on the airgap line. the sustaining series field supplies prac. Where ac generators are operated in parallel and are within the range of application of this regulator. A rheostat. ually changing load conditions. Thus. Under steady. 28 (a). 27. Each of these are described in their application in varioiis types of excitation systems in the order named. 8 as a main exciter. Rototrol Pilot Exciter ‘The Rototrol. but does mean that when the regulated voltage In the tenyear period following 1935. Directacting rheostatic type 2. Electronic type. the ac generator. For generators rated above 100 kva. The DirectActing Rheostatic Regulator The Silverstat generator voltage regulator is a common and widely used form of the direct.md operating at 3600 rpm. It is specifically designed for the automatic voltage control of small and medium size generators. 27Equivalent circuit of singlestage Rototrol pilot exciter.regulators. The compoundwound pilot exciter is normally mounted (. adjusted to give substantially flntcompounding. 15. each filling a specific need of the industry. such as 1800 or 1200 rpm.and quickacting rheostatic type of regulator. Excitation systems are now in the midst Regulator sensitivity must not be confused with overall . main exciter. expressed as a percentage of the normal value of regulated voltage. the Silverstat or SRA regulator is available in five sizes. The control field is a of a period of changes by reason of progress in the develop ment of regulating and excitation systems.he load on the pilot excit. and quickeracting systems.ally all of the ampereturns required to maintain the 1iototrol terminal voltage. two. The pilot exciter is invariably a 125volt machine with a selfexcited shunt field and a seriescxciterl field. the pracs(>paratelyexcited shunt field. This does not mean that the regulated voltage does not vary outside of the sensiIV. one. more sensitive.er. Consequently. with its minimum operating FIELD voltage not less than 30 percent of its rated voltage. The input to the control field regulator is the band or zone of voltage. Depending ~ipon the excitation requirements of the main exciter. there are now many different escitation systems in use. described in Sec. static regulator. two basic types of varies more than the percentage sensitivity from the regugenerator voltage regulators filled substantially all needs lator setting due to sudden changes in load or other system Of the electrical industry. the magnitude of its terminal voltage is practically constant. Each generator 1or is the fact that the Rototrol is operated in the unsat. The remainder of the chapter will be a comprehensive discussion of the application of these dc machines in excitation systems in conjunction with various types of generator voltngc regulators. the largest being used with generators as large as 25 000 kva. Efforts havcl been directed part.l SERIES FIELD regulator of medium size is shown in Fig. and where the main exciter is directconnected. the pilot exciter has a single stage of amplification.and its exciter is provided with an individual regulator and timtcd region.with the exciters operated nonparallel. t. Generally.he development of mom reliable. when the main exciter and Roto1rol pilot exciter are directconnected to the generator shaft . more accurate.icularly toward t.
the two smaller sizes. This basic assembly consists of a group of springmounted silver buttons so arranged that the butt. provides the means for changing the resistance in the exciter shuntfield circuit under control of the regulator. 29Schematic internal diagram regulator.*$ I ! i Systems Chapter 7 regulation.ons are separated from each other SILVER MUTTONS hhAGNET REGULATOR COIL \ /ARM ii RMATURE I/ PUSHER / VOLTAGE 4DJUSTING 3HEOSTAT RRI (b) Fig.he regulator sensitivity zone. The rated sensitivity of the SRA voltage regulators depends on the size of the regulator.i ‘. The magnitude and rate of load change determine how far the voltage deviates out. Fig.side of t. 17. respectively. have rated sensitivities of plus or minus 2% and 1% percent. The A larger SRA3. of SRA3 Silverstat i . For these reasons only sensitivity can be specified so far as the voltage regulator is concerned and not overall regulation. The SRA1 and SRX2. 28 (b). which involves not only regulator sensitivity but also the time constants of the machines and the charactcr and magnitllde of the voltage changes. SRA4 and SRA5 regulators are rated at plus or minus x of 1 percent sensitivity. (b) Silverbutton assembly of Silverstat regulator. Fig. Operation of the DirectActing Rheostatic Regulator The silverbutton assembly. 28(a) SRA4 Silverstat generator voltage regulator. and the time constants of the machines chiefly determine the time required to restore the voltage to the sensitivity zone. which involves factors over which the regulator has no control.218 Excitation .
an infinite number of steps of regulat. For a given value of regulated voltage and load on the machine being regulated there is a corrtsponding value of regulating resistance required in the field circuit. the pressure of the silver contacts changes to provide the correct intermediate value of resistance. ~11 resistance can be inserted or removed from the field the (:il. and a corresponding position of the moving arm and silver buttons that gives this value of resistance. Varying amounts of the resistance are short circuited by closing of the silverI)ut. the time required for the regulator to remove all resistance from the exciter shuntfield circuit is “PDroximately 0. The springs or leaves that carry the silver buttons are insulated from each other and each leaf is connected to a tap on a resistance element as shown in Fig. 28 (a). The regulating EXC SHUNT I rP 3 TRANS COMPEN voltage is necessary. At the other end of 1. but can be closed or opened in sequence by a suitable driver having a travel of a fraction of an inch. At one end of the travel of the moving arm.mArm.Chapter 7 Excitation Systems 219 normally. and rectox . Excitation current in the generator field can be regulated by changing the esciter output voltage or by holding the exciter voltage constant and changing the generator field resistance. 30. 31Silverstat regulator damping transformer rectifier assembly.clglll:ltor is shown schematically in Fig. since the travel is short. With a sudden drop in ac voltage of 10 to 12 percent. The compensating resistance is used to provide crosscurrent compensation during parallel operation of ac generators or to provide linedrop compensation. When the voltage regulator is in . The fixed resistance in the esciter field circuit in Fig. and the voltage is restored to its correct value by the moving arm and silver buttons taking a new position._ REGULATING RESISTANCE REGULATOR COIL 2 I Fig. a corresponding change in voltage results... A typical excitation system under control of an SRA r. A springmounted armature is centered in the air gap of the electromagnet as shown in Fig.. placing maximum resistance in the field circuit.s&a. . the buttons are closed and the resistance is short circuited.ante of the contacts. In regulating ac voltage.1~~ travel. 30 is used when it is desired to limit the exciter shuntfield current when the maximum or ceiling current is such as to interfere with the best performance of the voltage regulating equipment. Since the pressure on silver contacts determines the resist..” .ton contacts.. The cont. 29.*u”sr.05 second or 3 cycles on a GOcycle basis. r&tance is connected directly in the exciter shuntfield (‘ircuit. The regulating action of the SRA regulator is that of a Sen&atic device that operates only when a correction in sea* . Fig. The speed of operation of the regulating element depends upon the magnitude and rate of change Of the operating force.ing resistance are obtained. .rol element of the regulator is a dc operated device.. If the required value of exciter field resistance should lie between two of t. a fullwave rectox rectifier is used to convert the ac to dc for energizing the control element. four being used in the SR=\4 regulator illustrated in Fig.cuit quickly. The exciter shuntfield rheostat and the generator field rheostat are provided primarily for control of the generator escitation when the regulator is not in service. The moving arm can hold the resistance :ktd any intermediate value and. 29. Undcl such conditions the magnetic pull on the moving arm is balanced against the spring pull at that position of its travel.*A‘. One or more of these basic elements are used in regulators of different sizes. When there is a change in lo&d on the machine being regulated. 30Selfexcited main exciter controlled by Silverstat regulator. all of the silver buttons are apart from each other..he tapped points of the regulating resistance...
However. On applications of this kind the type BJ regulator is particularly adapted to the control of ac machines employing quickresponse excitation. Parallel OperationAs is true with most generator voltage regulators. Since the damping transformer operates only when the excitation of the generator is changing. namely. and the interconnection of established systems. bus reactors. the main exciter must be separatelyexcited. where ac generators operate in parallel.ation change. The main control element of the regulator is energized from two singlephase potential t. The damping transformer is illustrated in Fig. and if the reactance is such as to cause from four to six percent reactive drop between the two generators. have reached a point where quickresponse excitation is valuable for improving stability under fault conditions and large load changes. and the wattless power is automatically divided in proportion among the paralleled ac generators. 18. energy is transferred by induction from one winding to the other of the damping transformer. and time relation to electrically damp excessive action of the moving arm. Two sets of contacts are on the moving lever arm of the regulator element shown in Fig. For threephase ac generators with. The B. DampingTo stabilize the regulated voltage and prevent excessive swinging under various conditions of excit. its minimum operating voltage must not be less than 30 percent of its rated voltage if stable operation is to be obtained. The damping transformer is of a special type having a small air gap in the laminatediron magnetic circuit. When there is a change in escitation voltage as a result of the regulating action of the regulator. it has no effect when the regulated voltage is steady and the regulator is balanced.ransformers connected . The adjustable resistance permits adjustment of the compensation to suit the application. and the other winding is connected in series with the voltage regulator coil. The current transformer is connected to an adjustable resistance in the ac supply circuit to the regulator operating element. In many applications. If each generator is excited by an individual exciter under control of an individual voltage regulator. preventing the moving arm from carrying too far the change in regulating resistance and consequent change in generator excitation. This is accomplished by means of crosscurrent compensation. Thus the phase relationship is such that for lagging reactive kva. 32. to raise or lower the ac machine voltage. Thus the division of kilowatt load is practically independent of the generator excitation. reactance in the form of power transformers. the compensation is obtained by a standard current transformer connected in one lead of each generator being regu lated as shown in Fig. and it is necessary to supply a means of assuring proper division of reactive kva between the generators. 31. If the exciter is selfexcited. Where several ac generators operate in parallel and all the generators are excited from one common exciter. the extension of transmission systems. IndirectActing ExciterRheostatic Regulator In recent years the increase in capacity of generating units. a damping effect is introduced into the regulator coil circuit. 30. ‘I’he use of this device eliminates the need for dashpots or similar mechanical antihunting devices. respectivelyThe quickresponse AR and AL contacts control the high: . 33 is used to describe the operation of the device. while the potential transformer that operates the regulator is connected to the other two leads. The indirectacting exciterrheostatic type of generator voltage regulator controls the voltage of an ac machine by varying the resistance in the field circuit of the exciter that excites the ac machine.to m the ac machine leads. One winding is connected across the field of the generator whose voltage is being regulated. the normalresponse contacts RL and the quickresponse contacts ARAL.J regulator is of the indirectacting exciterrheostatic type for the automatic control of medium and large size ac generators. exists between paralleled ac generators. The effect of the small droop required is usually negligible under operating conditions as found in actual practice. changes in the field excitation of paralleled ac generators do affect the reactive kva or wattless component of the output.. Thus. 30. The exciter is preferably separatelyexcited from a pilot exciter or other source. However. This energy introduced into the circuit of the regulator coil acts by reason of its direction. wattless current circulates between the paralleled ac generators unless some provision is made whereby the generators are caused to properly divide the reactive kva. In this manner each generator tends to shirk A reactive kva. This scheme of operation requires that the exciters be operated nonparallel. the voltage drop across the compensating resistance adds to the ac voltage energizing the regulator and subtracts in the case of leading reactive kva. the SRA regulator. This action tends to cause the regulator to lower excitation for lagging reactive kva and raise excitation for leading reactive kva. the exciter shuntfield and generator field rheostats are ordinarily turned to the “all ollt” position so that the regulator has full control of the excitation voltage. The current transformer is connected in one generator lead. This is because the reactance produces an effect similar to that obtained where crosscurrent compensation is used. A schematic wiring diagram of the BJ generator voltage regulator and its auxiliary contactors is shown in Fig. provided the exciter is of a size that is within the range of application of this type of regulator. 32. then stable operation and proper division of the wattless component can usually be obtained without using crosscurrent compensation between the regulators. which functions to cause each generator to shirk wattless current by means of a slight droop in the regulated voltage with increase in the wattless component of current. a single Silverstat regulator can be used. When lower voltages are necessary. the SRA regulator can control only one exciter at a time.220 Excitation Systems Chapter i’ operation and controlling the generator voltage. This diagram in conjunction with the simplified schematic of Fig. and the division of the reactive kva is directly affected by the operation of the voltage regulators. magnitude. the usual practice is to provide each one with an individual exciter controlled by an individual regulator. The division of the kilowatt load among paralleled ac generators is dependent upon the power input to each generator and is controlled by the governor of its prime mover. by means of a damping transformer as shown in Fig. etc. The normalresponse contacts control the rheostat motor contactors NR and NL.
The second is the circuit of the antihunting winding NH of the regulator main control clement and the third set of contacts complete a timingcondenser circuit. energizing the rheostat motor control contactor NR. tlepentling upon the sensitivity setting of the regulator. respectively.knc’<l inserted in t.he field circuit. The antihunt device operates to increase the gap distance between the contact faces of the regulator contacts R and L.or control contactor NR has three contacts that close in independent circuits simultaneously. The rheostatmot. which arc the “field forcing )’ UP and “field forcing down” contactors. Should the ac voltage fall below normal by a small amormt. . the normalresponse contact R will close. Where the deviation from normal voltage is small and within the recalibration effect of t.of BJ. 33 is closed.he antihunt device. thereby opening the circuit at the R contacts. This change in position of the R contact is equivalent to changing the regulator setting to a lower voltage so far as the raise contacts are concerned. t.h(! rcgulat.he rheostat in a direction to remove resistance from the exciter field circuit.~=FIELO FORCINGDOWN R R. and !&en (2L is opened by energizing its coil. 32.chapter 7 Excitation Systems 221 speed contactors QR and QL. REGULATOR OUICKLOWER FORCING CONTACTOR FIELO Fig. 32Schematic diagram of the BJ regulator controlling the voltage of a separatelyexcited main exciter.regulator in iFig.=FIELO FORCINGUP R I?.s RL nor the qrrickresponse contacts ARAL are closed.44OTOROPER SHUNT FIELD RHEO CONTROLLED BY REGULATOR NORMALRESP CONTACTS ROr’OIFF FIELD FIXED R OR = REGULATOR BUICKRAISE FORCING CONTACTOR OL. The contacts NR energize the rhcostnt motor which then turns t. a block of resist.c is shorkd out of the mainexciter field circuit. \l:hen contactor QR in Fig. the immediate Fig. and to a higher voltage so far as the lower contacts are concerned..=SHUNT FIELD FIXED R R. 33Mainexciter circuits:under control .or lever arm is balanced and in this position TO CT I nckhcr the normalresponse contact. thereby increasing the voltage applied to the exciter field. all external resist:rn(.. The one circuit is that just described which operates the rheostat motor. is Normal ResponseWhen the ac voltage is normal. R.
one compen. This causes the ac machine voltage to start to return to normal very rapidly by forcing action. it is desirable to provide some time delay to allow the ac machine voltage to reach its final value. L closes energizing the rheostat motor control contactor NL. Thus. The quickresponse contacts are set to a wider spacing than the normalresponse contacts so that larger deviations from normal voltage are required to close them.he required new position.act setting. The level of the regulated voltage is set by adjustment of the voltageadjusting rheostat. 20. As the ac voltage comes within the setting of the AR contacts and they no longer close. The compensator is designed to supply j a compensating voltage in two phases of the threephase / regulator potential circuit. CrossCurrent ulator Compensation with BJ Reg’ When crosscurrent compensation is required to give the’ voltage regulator a drooping characteristic.he AR ancl LL contacts in the same manner m described for the NH device and the R and L contacts. the fieldforcingup contactor closes and opens rapidly while the rheostat arm approaches t. for example. preventing overshooting of the rheostat position and bringing the ac voltage to normal in a minimum length of time. it takes only a minimum of additional movement after the normalresponse contacts take control to return the voltage to normal. This insures applying a bal1 anced threephase voltage to the regulator element. the damping effect of the differential field in slowing the exciter response is removed. a contact in the QR contactor opens the differentialfield circuit. 19. contactor NR opens to stop the rheost. Thus the rheostat arm is permitted to move a definite distance. from one button to the next on the rheostat faceplate. if the deviation from normal voltage is \vithin the recalibration effect of the QH antihunt device. This causes the rheostat motor to run continuously until the ac voltage is within the zone for which the antihunt device is set. and resistance being removed to decrease the level of the regulated voltage. Contacts AR close the circuit to the highspeed fieldforcingup contactor QR.222 Excitation Systems Chapter 7 result of the closing of the contacts on contactor NR is to cause the opening of the regulator R contact. This starts another cycle of operation such as just described and these cycles continue until the normal value of regulated voltage is established. However. In this case the regulator arm is caused to follow the change in contact position made by the antihunt device. Sensitivity of the BJ Regulator The rated sensitivity of the BJ generator voltage regulator is plus or minus x of one percent. and the R contact and the contactor NR remain closed. The action of the dashpot is also such that the time required for the contacts to remake is longer as the lever arm approaches the normal voltage position. If the deviation from normal voltage is greater than the recalibrated setting of QH antihunt device. Such delay is obtained by a dashpot on the antihunt device that prevents the regulator contacts from immediately returning to their normal position. contactor NR does not immediately open due to a timedelay circuit around its coil that maintains the coil voltage. The main coil of the control element in Fig. and at the end of its time delay. In this way. After the rheostat motor stops. The usual range of settings of the quickresponse contacts is from plus or minus 255 percent to plus or minus 10 percent. the normal response’ contacts R take control and by notching the rheostat: return the ac voltage to normal. 32. applying full exciter voltage to the field circuit. The vector diagram of the compensating circuit is shown. the normal response contact R again closes if the ac voltage has not returned to normal. By means of the continuous or notching action of the rheostat. 32 consists of a voltage winding energized by a dc voltage. device QH. would not be the case if only one leg was compensated. an action similar to that described for low voltage takes place.at motor and deenergize the antihunt device. dependent upon the magnitude of the voltage change. the normalresponse contacts R on the regulator close. The normal range of adjustment is approximately plus or minus 10 percent from the normal generator voltage. the fieldforcingup contactor closes and remains closed until the ac voltage is brought within the recalibrated setting. which operates to spread t. When the main exciter has a differential field as shown in Fig. Therefore. The sensitivity is adjusted by varying the spacing between the regulator contacts R and L. When the fieldforcingup contactor QR closes. When the ac voltage rises above the regulated value. except that the regulator contact. which short circuits all of the external resistance in the exciter field circuit. which operates the rheostat motor in a direction to increase the resistance in the exciter field circuit. Since the rheostat moves at maximum speed while the quickresponse contacts are closed. to stop the motor of the esciter field rheostat and thus stop the rheostat moving arm. resistance being added in series with the regulator voltage coil to increase the level of the regulated voltage.an auxiliary contact on this contactor closes at the same time in the circuit of the antihunt. the coil is energized by a voltage equal to the average of the phase voltages and the regulator holds this average voltage within the rated sensitivity zone. connected. which. 1 . as shown in Fig. rectified from the threephase ac source being regulated. followed by closing of the quickresponse contacts AR. time is allowed for the ac voltage to come to rest between each voltage correction as the voltage approaches its normal value. the setting depending somewhat on the setting of the normalresponse contacts and upon the operating conditions of the particular installation. This results in a decreased motor speed as the rheostat arm moves nearer to its new position. which in turn opens the circuit to the coil of contactor NR. Where the original voltage deviation is large enough the regulator contacts remain closed continuously even though the antihunt device changes the cont.1 sator and one current transformer are required. After this time delay has expired and the contacts have returned to their normal position. Quick ResponseWhen a large drop in voltage occurs. such as might be caused by a large block of load being thrown on the system or by a fault. 33. at which time the notching action takes place to bring the voltage to normal.
it is sometimes desirable to regulate for a droop compensation. The it is impractical to use pilot wires. 35Principle of linedrop compensation.respectively. the two components XI and RI are subtracted from it When the excitation of a synchronous condenser is inartificially by the compensation. For power factors greater than zero. Voltages ER and Ez are 120 degrees apart in time phase and. When . In order to compensate for reactive cross current between machines and for complete line drop’when machines are operating in parallel in the same station. can bc added to a threephase set of voltages without unbalancing it. these voltage vectors swing through an arc of 90 degrees and give zero compensation at 100 percent power factor.sary. 34Vector diagram of crosscurrent compensation with BJ regulator. and a simple compromise solution is available to inated to a large extent the need for linedrop compensa. If the components XI lagging (overexcited) current to the system thereby caus .. 32. LineDrop Compensation with BJ Regulator Complete linedrop compensation is not always necesThe wide use of interconnected power systems has elim. The resultant voltage Ez creased above a certain value. since the reactance component XI of the line predominates. each under the control of a voltage regulator. i. The current applied to the autotransformer of the compensator in Fig. 35. 34. only a proportionate component of these volt:~ges and ER add directly to voltages Ezl and E13. is no interference between the two compensators. E. and there vector diagrams of Fig. Circuit shown in Fig. Synchronous Condenser Excitation with BJ constant. vectors Ez and ER add directly to vectors Ezl and E13. The voltage regulator is to maintain the voltage Ez 22. the resistance component RI having a relatively small effect. the XI linedrop compensation must never exceed the XI crosscurrent compensation. 32 is taken from the secondary of a current transformer in phase 2 of the ac circuit. 21. the potential transformer secondary voltages being represented by I&.provide approximate linedrop compensation and reactivetion. The RIdrop compensation is set to constant voltage to be maintained at some point on the approach the XI drop of the line for some average power system external to or distant from the station where the factor.e. requires a droop in regulated voltage with an increase in wattless load. In general. Since in actual practice to control the excitation of a synchronous condenser.Chapter ‘i Excitation Systems 223 / Fig. under which condition maximum compensation is obtained. and E13. Is then supplied to the regulator. As the power factor approaches unity. Usually the coml)onsator should cause from four to six percent droop in voltage at zero power factor full load on the ac generator. the regulator potential circuit is essentially the same as that shown in Figs. In any event. one between terminals XlX2 designated as EZ on the vector diagram and the other between terminals YlY2 designated as ER on the vector diagram. TI” VECTOR Fig. used TO VOLTAGE REG El I2 EZ IX IR I OF SYSTEM DIAGRAM in Fig. three current transformers and two compensators with suitable auxiliary equipment must be used for each machine. At zero power factor. and and 33.. the regulator controls the voltage as if it were connected by pilot wires to the load center. The Ez addition of these compensating voltages to the line voltages :LS the’load increases or the power factor changes gives the regulator element a high voltage indication resulting in a reduction or droop in regulated voltage. If it were possible to supply the regulator with Regulator Pilot wires so that it could measure the voltage at the load center. 32 “inding is energized from the generator bus voltage El. therefore.the RIdrop compensation is so set. the condenser furnishes a . there must be a net droop in regulated voltage with increase in wattless load. the sc machine and its regulator are located. it is necessary to compensate mainly for this component of the line drop. and RI are proportioned to and in phase with the corresponding values of line reactance and resistance voltage drops. On the other hand. However. reactance linedrop compensation requires a rising characteristic for the regulated voltage with an increasing mattless load. the regulator could adjust the excitation of the The type BJ generator voltage regulator can also be used generator to maintain Ez constant. The vector diagram shows Ez and ER for zero power factor. Parallel operation of ac generators. The principle XIdrop compensation can be adjusted independently to 1~ which this is accomplished is shown by the circuit and provide the required crosscurrent compensation. Two compensating voltages are produced.
a timedelay currentlimiting device is used.224 Excitation Systems Chapter 7 ing the voltage to rise. and the other set deenergizes an auxiliary relay. the case of a slowly rising load current to a predetermined limiting or unsafe value. One set of contacts initiates a timing cycle. Protection against overcurrent is provided by a currentoperated device having its operating coil energized by the line current and having its main contacts connected in series with the main control contacts of the voltage reguI I EP later. ImpedanceType Voltage Regulator The excitation system shown in Fig. This protection against a gradual increase in load operates in the normalresponse RL circuits of the voltage regulator. a second contact of the currentlimiting device energizes the “lower” control circuit of the regulator. voltage adjusting unit. and automatic control unit. decreasing the excitation lowers the voltage. and second. Thus. When the BJ regulator is used to control the excitation of a synchronous condenser. These voltageregulator devices consist entirely of imped AUTO. The equipment is designed to recognize two conditions. causing the excitation and load current to be reduced to the safe limiting value. the regulator in trying to hold the line voltage overexcites the condenser. Tf the synchronous condenser load is gradually increased. 36Block diagram of the impedancetype voltage regulator as used in a mainexciter Rototrol excitation system. it regulates the line voltage to a constant value by varying the excitation of the condenser. a situation may occur where the condenser does not have sufficient corrective rkva capacity to handle all. it is necessary to use a separatelyexcited main exciter.‘! A . 36 employs a mainexciter Rototrol to supply excitation to the ac generator. This is accomplished by the differential field in the conventional main exciter. Deenergizini the auxiliary relay allows the contacts of the voltageregulating element to remain in control for the time setting of the timing relay. In the case of a sudden increase in load current. a currentlimiting device is used to limit the masimum excitation voltage to a level that does not cause damage due to continuous overloading of the condenser. thus permitting the use of both normaland quickresponse escitation for stability purposes under fault conditions. system requirements. It is often necessary that the condenser furnish leading (underescited) rkva as well as lagging (ovcrcxcitcd) rkva. Tn a similar manner. and by reversing the pilot exciter voltage in the case of the Rototrol pilot exciter. the latter maintains control to prevent increase in excitation. the energy requirements of the control field are sufficiently small that they can be supplied by instrument transformers. CONTROL J :II VOLTAGE ADJUSTING UNIT EN EP MINIMUM 1. In many cases it is necessary to reverse the excitation voltage to obtain full leading rkva capacity from the condenser. With the high degree of amplification obtainable with a Rototrol. To protect against this condition. a sudden increase in load current such as might be caused by a system fault. The intelligence transmitted to the control field of the Rototrol as a function of the generator terminal voltage is determined by the voltageregulator potential unit. when a generntor voltage regulator is applied to a synchronous condenser. At the same time. the currentlimiting contact in series with the R contact of the regulator opens the “raise” control circuit and prevents any further increase in excitation. 23. Should the decreasing overload remain for a time below the setting of the instantaneous overcurrent relay but within the setting of the currentlimiting element. causing it to carry excessive current and become overheated. i Fig. provided the condenser has sufficient corrective rkva capacity. Control of the excitation is automatically returned to the voltageregulator control element when the overload disappears. and it is necessary to reduce the excitation to an extremely low value. Where the minimum value is less than 30 percent of the main exciter rated voltage. At such a time. In the operation of a synchronous condenser under abnormal conditions. first. or the most severe. an instantaneous overcurrent relay set to pickup at a higher value of current than the currentlimiting device closes its contacts.
in the absence of a zerosequence component.uc determined solely by the magnitude of the impressed ac voltage. (b) Vector diagram.. consists of a potential transformer. a filter reactor and a set of resistors. The automatic control unit is :L voltagesensitive device.iun to decrease the excitation voltage. the output of which is a dc voltage. When the voltage regulator is not in service. To guarantee synchronous machine steadystate stability. If the generator voltage increases above t. the voltage regulator is not affected by generator voltage unbalance and regulates to constant positivesequence voltage.gulsted value. 3SVector diagrams showing how pensation is obtained with the potential ancetype regulator. ER ER / 62 ‘52 I \:A R En LOCUS EC \ ER= E. The output of the automatic control unit is the control signal that energizes the control field of the mainexciter Rototrol. When the generator voltage falls below the selected value. Ex3 (b) Fig. The polarity and magnitude of this dc voltage .2 + E. the dc output voltage is in the direct. applied to the series connection of the voltage adjusting unit and the automatic control unit. 3’i. (a) Schematic diagram. working through the Rototrol exciter. insure adequate excitation for all kilowatt loads. crosscurrent comunit of the imped . shown schematically in Fig. The circuit is a negativesequence voltagesegregating filter so connected that the negativesequence voltage is subtracted from the line voltage which.+E. yields positivesequence voltage. 37Impedancetype regulator potential unit. (b) (a) @ ER h+Ex. the dc output voltage of the automatic control unit is in the direction to increase the ac generator excitation. The primary of the filter or mutual reactor is energized by the phase 1 and 3 current transformers. that is. 36 is referred to as an impedancetype or statictype voltage regulator. The output voltage of the potential unit is directly proportional to the positivesequence component of the generator terminal voltage. manual control of the ac generator excitation is by means of the man~ral control unit. ‘I’he minimum excitation unit used with the Rototrol excimt’ion systems is of a form that provides a variable minimum limit depending on the kilowatt load. a minimum excitation unit is used.Chapter 7 Excitation Systems 22s sncc elements and from this consideration the combination cf devices in Fig. (cl Fig. = = + EXI 62 Ex2 Exa ERS *. When the generator output voltage is exactly at the rc. Potential UnitThe voltageregulator potential unit. and therefore.hc regulated value.. The voltage regulator potential unit is energized by the gcner:btor linetoline voltage and the currents of two nhases. Its output is a singlephase ac voltage. The flus produced thereby induces a voltage in the secondary wincling which is added vectorially to the phase3 drop in the Es1 \ II / E23 13 (a) ER= E. the output voltage of the automatic control unit is zero.
It is subtracted vcctorinlly from the generator voltage to give the desired positivesequence voltage across the terminals Vll and V12.ivesequence voltage at the gcnerntor terminals. the vector diagram becomes that shown in Fig. This point of intersection is called the balance point of the two impedances.. must vary for the automatic controlunit input voltage ER to remain constant. If the ohmic value of RC is reduced to 50 percent. and the currents through the two branches of the circuit are equal at only one value of voltage where the characteristics intersect. one containing a capacitor and the other a saturating reactor. the sum being proportional to the negat._ A. The curve of the reactor indicates that its current increases more rapidly than voltage. By means of the voltage adjusting unit. Assuming a given lagging power factor generator load. which are connected with additive re. the resistance between the generator terminals and terminals Vll and V12 of the automatic control unit can be changed. EC also increases and El2 must decrease. 39 (b). It measures the voltage to be regulated and delivers energy to the mainexciter control field only when necessary. th current of the capacitor circuit is large compared of the reactor circuit raising the potential of term1 above that of ACT. I PI lrh ! lrl Fig. Automatic Control UnitThe automatic control unit is the voltagesensitive element of the impedancetype voltage regulator. causing a directly proportional change in voltage drop in the circuit. circuit. the current in the reactor is greater than the current in the capacitor. 38 (b). reactor. The difference vector EC can appropriately be called the reactivedroop compensator voltage. of saturating drytype rect. The voltagesensitive circuit in Fig. The voltagecurrent characteristic curves of the capacitor and saturating reactor are shown in Fig. 38 (c) shows how the generator terminal voltage E. ‘The operation of the voltage regulator depends upon the fact that when the voltage increases above this point. the vector diagram of Fig. the output currents of the rectifiers are of the Rototrol control field and no current flowsin th Should the ac voltage become low. Thus the generator voltage is given a drooping characteristic with increase in lagging power factor load. since ER remains constant in magnitude. As the generator load increases. The potential unit can also provide compensation for parallel operation of ac generators when each machine is equipped with a voltage regulator. The drop requires a change in ac generator voltage to produce the regulator balancepoint voltage across the terminals Vll and V12. When the voltage decreases below the balance point. 38 (a). 37. (a) Circuit diagram. The control field of the Rototrol is connected between amidtap on the resistor and the opposite side of the rectifier. 1 1 kh ’ AMPERES regulator automatic control unit. Reactivedroop compensation is obtained by adjustment of the resistance RC in the potential unit in Fig. 36 is a rheostat that enables the operator to set the ac generator regulated voltage at any value within a band of plus or minus 10 percent of the rated generator voltage. and causing current to flow in trol field in a direction to increase the excitation voltage . 39 consists essentially of two parallelcircuit branches. If the ohmic value of the resistor RC is 100 percent. The output of the reactor circuit and the output of the capacitor circuit are rectified by singlephase fullwave INSULATING TRANSFORMER VII0 VI20 (0) E+L E Ei . When the applied voltage is and the capacitor and reactor currents are equal in tude. 39Impedancetype ‘cl L.226 Excitation Systems Chapter 7 resistor RC.ifiers.. lation in series through a resistor and smoothing reactors. The vector relations of the generator line currents and terminal voltages are shown in Fig. This negativesequence voltage is the component of the threephase voltage that represents the unbalance in voltage resulting from load unbalance. Voltage Adjusting UnitThe voltage adjusting unit in Fig. 37. (b) Intersecting impedance characteristics and capacitor. the capacitor current is the greater. En and El2 no longer are identical. the voltage equation of the circuit and the vector diagram are those shown in Fi g. although for unity power factor their difference in magnitude is of negligible proportion. however.
36. and the reactance of the ac coils is high. there is a voltage at which the seriescircuit current begins to increase rapidly with small increases in voltage. t.. For an increase in ac voltage. current in the series circuit is substantially zero. both windings produce an ac flux in the same direction through the center leg of the core. 41. The small current output of the :uitomatic control unit is sufficient to control the Rotot ml output over the entire range of the Rototrol capability. the direction of current flow in the control field would be reversed causing a reduction in excitation voltage.chapter 7 Excitation Systems 227 r&c the aC voltage. The controlfield c:llrrcnt as a function of the ac voltage applied to the :ultomatic control unit is shown in Fig. A substantial amount of ac current is allowed to flow through the reactor windings under this condition. The two outside legs are connected in parallel. Since the mainexciter Rototrol is limited to use with 3600rpm turbine generators. 40. the minimum excitation unit is of the variablelimit type. the core has a higher degree of saturation and the reactance of the ac windings is low. 41Schematic diagram of the impedancetype minimum excitation unit and vector diagram showing how variable minimum limit is obtained. thereby controlling the inductance and reactance of the two outer ac windings. However. The minimum point can be a fixed limit or a variable limit. 300 200 100 200 0 100 AUTOMATIC CONTROL UNIT OUTPUT MILLIAMPERES 300 Fig.ith normal UC voltage applied to the automatic control urrit. The dc current in this winding controls the saturation of the iron core. thereby maintaining a margin of excitation current above that at which the machine would pull out of synchronism. 40Typical output curve of automatic function of ac voltage. ‘The current in the control field of the Rototrol is directly . the current in the reactor control winding is relatively high. saturation of the core is slight. the minimumexcitation unit normally used is comprised of impedance elements.rolfield current is approximately proportional to the iTHREELEGGED SATURATING REACTOR OLEAD n OLAG Fig. The minimumexcitation unit establishes a minimum Point or limit below which the excitation of the ac gencmtor cannot be lowered. 30 (b).series circuit composed of the saturating reactor. such that at any given instant. capacitor and rectifier. Minimum Excitation UnitLike other units of the irnncdnncetype voltage regulator. The centerleg winding is energized by the mainexciter Rototrol output voltage as shown in Fig.roportional to the horizontal difference between the caI. When the voltage is low across the. and when the dc current is high. When the ’ ac generator is operating at normal voltage and the excitation voltage is normal. The winding on the center leg is the dc control coil. If for some reason system conditions should cause the voltage regulator to introduce current into the control field . Maximum current in the direction to raise the Rototrol terminal voltage occurs when the ac voltage is approximately 85 percent of the balancepoint voltage. A saturable reactor with coils on the three legs of a Bshaped core is used.:Lcitor and saturating reactor voltampere characteristics in Fig. control unit as change in ac voltage for small changes. On machines that carry considerable real or kilowatt load it usually is desirable to make the minimum limit vary approximately directly proportional to the kilowatt load. Examination of the curves shows that the com. because of the impedance characteristic of this series circuit. The schematic diagram and vector diagram of the minimumexcitation unit is shown in Fig. Thus . and consequently the reactance of the ac windings is low.hc controlfield current is nearly zero and any dcviation in aC voltage causes a corrective current to flow in the control field. When the dc control current is low.j. The relatively high ac current through resistor Rl causes a large voltage drop such that the ac voltage appearing across XY is relatively small.
If the magnitudes of the line currents are held constant and the power factor changed to 70 percent lagging.. The selenium rectifiers form the. the magnitude of the voltage EC is dependent on the magnitude of the inphase component of the line current. Manual Control UnitThe manual control unit used with the mainexciter Rototrol excitation system of Fig. The main exciter terminal voltage is applied across: two terminals of the bridge and the control field of the Rototrol is connected across the other two terminals. A compensating voltage that is a function of line currents 1211is added vectorially to El2 such that the ac voltage applied to the saturating reactor is equal to EC. rectifier and load) are shown in Fig.he reactor control winding is also reduced. maintaining essentially constant mainexciter voltage and constant ac generator voltage for a given load. The sudden large increase in current shown when voltage E1 is reached in Fig. The voltage El? across terminals VlV2 is held constant by the automatic control unit under balanced load conditions. As the voltage across XY is increased. The minimum excitation control field is the limits field in Fig. and the current through resistor Rl is reduced. the voltage ERzl is shifted such that the magnitude of EC becomes that represented by the dotted vector. The currents 1. (b) Effect of Rl included. Thus. since the voltage input to the saturating reactor is a function of the kilowatt load. causing less voltage drop in the circuit and increasing the voltage across XY. 42 (a). and this current rectified is supplied to the minimum excitation control field of the Rototrol exciter. The current then undergoes a large increase to the value 11.228 Excitation Systems Chapter 7 of the Rototrol to reduce the excitation voltage. Thd exciter terminal voltage is adjusted by changing the posi. In addition. and the other giving an operating range of O750 milliamperes. Such a circuit is required to reverse the direction of current in the control field as required to raise or lower the Rototrol voltage. The direct current supplied to the minimum excitation control field is in the direction to raise the excitation voltage. controlling element of the bridge circuit since the voltagil drop in this leg of the bridge is practically independent oft 9 i . The unit having the larger operating range is used with the mainexciter Rototrol. the regulator again takes control and holds the voltage for which it is adjusted. 42 (a) is eliminated. 36 is a bridgetype circuit as shown in Fig.l tion of the potentiometer. The bridge circuit consists of two fixed resistors. The individual and combined voltampere characteristics of the saturating reactor. and the minimum excitation unit thus begins to regulate for a preset minimum excitation voltage to keep the circuit of the unit balanced. The reactnncc of the ac windings increases. 36. 41. Two ratings of minimumexcitation units are available. and Iz in the vector diagram of Fig. the combined characteristic shows that the circuit conducts practically no current until the voltage E1 is reached. When system conditions cause the automatic control unit to increase the excitation above that provided by the minimum excitation unit. The minimum excitation limit becomes a variable quantity dependent upon the kilowatt load of the generator. The locus of the magnitude of EC for a particular magnitude of current at various power factors is represented by the semicircle as shown. If the volt>nge across XY rises to the conducting point of the series circuit. a potentiometer and two selenium rectifiers connected as shown. ac current increases sharply in this circuit. 41 are drawn for the unity power factor condition and the resulting magnitude of EC is represented by the vector drawn with a solid line. The practical operating range of the unit is determined by the intersection of the capacitive reactance line Xo with the saturating reactor line Xs. the unit is a dc voltage regulator in itself. 42Voltampere characteristics of individual components of minimumexcitation unit and combined voltampere c characteristic. (n) Effect of RZ omitted. When the voltampere characteristic of the resistor Rl is included. Therefore. 42 (b). capacitor and resistance (equivalent resistance of the reactor. the current in t. but the current increases rapidly and linearly with increase in voltage in the range above E. The variable minimum excitation limit is obtained by the compensating circuit shown in the lefthand portion of Fig. the voltage across XY applied to the series circuit also varies with kilowatt load. the combined characteristic ismodified to that shown in Fig. and hence varies with the kilowatt load on the generator. 43. one giving an operating range CURRENT (a) I CURRENT Fig. of O300 milliamperes.
since the Rototrol is a highspeed machine with airgap dimensions the same as any other form of dc machine.Chapter 7 EN MAIN EXCITER TERMINAL VOLTAGE Excitation Systems EP INDIRECT ACTING EXCITER TYPE “EJ30” REGULATOR RHEOSTAT1 Fig. In each case. it can be directconnected 10 the shaft of a turbine generator. Evidence of the importance of this effect and illustration of the comparative performance of the mainexciter Rototrol excitation system is given in Fig. This arrangement might be used where the generator speed is less than 1200 rpm. the mg set having sufficient inertia to carry through system disturbances without appreciable speed change. Also. without contactors.O + 6. and the manual control unit holds the excitation voltage constant. 44. For a drop in. Fig. In the conventional excitation system.O I 9. the control field current would be in the raise direction. 43Schematic diagram manual of the impedancetype control unit. and will remain substantially constant. For a given setting of the potentiometer. and the dashedline curve shows the variation under control of an indirectacting exciterrheostatic type of regulator and a conventional main exciter with 0. The circuit of the mainexciter Rototrol excitation system is that shown in Fig.O 1. the current through the bridge increases. The effect on the mainexciter Rototrol of induced field current caused by changes in generator load was discussed in Sec. exciter voltage. The rapid recovery of the voltage under control of the impedancetype regulator and mninexciter Rototrol is an important factor in maintaining system stability. regulator CYCLESSECONDSO y 3. When the speed of rotation of the main ac generator is 1200. 43 can be considered constant. operation and maintenance 1)~completely eliminating the pilot exciter. The latter arrangement is applicable with a generator of any speed. 36. 120 h 150 & 180 i 2iO & 240 d. The directconnected mainexciter Rototrol is a step toward simplification of turbine generator construction. The mainexciter Rototrol excitation system has the advantage of a voltage regulator without moving parts. a threephase reactance load was suddenly applied to the generator to cause approximately 20 percent of rated generator amperes to flow in the circuit. the ac voltage may be adjusted for any value from zero to maximum. Rototrol Exciter Pilot Exciter with SingleField Main The simplest form of an excitation system using a Rototrol pilot exciter is shown in Fig. Approximately 20 percent of generator rated amperes at 0 percent lagging power factor added at zero time. 45. 10. Current then flows in the control field in a direction to reduce the exciter voltage until the bridge circuit is again balanced. 24. If the main exciter voltage should increase for any reason. The overall performance of the system shows marked improvement in voltage dip and recovery time when compared with a conventional mainexciter excitation system. The solid line shows the time variation of the ac generator voltage under control of an impedancetype regulator and a mainexciter Rototrol. 1800 or 3600 rpm. the bridge circuit is balanced when the voltage El is equal to Ez and nndcr this condition there is no current in the Rototrol control field. 270 d. A second possibility is to have the main exciter mounted on the shaft of the ac generator and the Rototrol separatelydriven by a small motor.5 response ratio. 44Voltagerecovery performance of mainexciter Rototrol excitation system compared with performance of conventional mainexciter system under control of BJ regulator. particularly during the period of overshoot when the generator voltage is greater than 100 percent. MainExciter System Rototrol Generator Excitation The Rototrol with its two stages of amplification can be built with large power output capabilities while the control field energy requirements are sufficiently small to be supplied by instrument transformers. The Rototrol pilot exciter is a variable voltage pilot exciter and the method of operating the excitation system of Fig. and requiring no large motoroperated mainexciter field rheostat. Thus. The system also eliminates the use of any pilot exciter. A third arrangement is to have the main exciter and the Rototrol pilot exciter driven by a motor and operating at 1200 or 1800 rpm. 300 b the current through the rectifiers. Thus the voltage E. 45 is essentially no different than the operation of conventional exciterrheostatic . in Fig. the main exciter and Rototrol pilot exciter can be directconnected to the generator shaft. the pilot exciter is a constantvoltage generator. which increases the voltage drop El so that MCR is positive \vith respect to MCL. 25.
automatic control unit and the manual control unit are those described in Sec.he mainesciter field by the Rototrol pilot exciter. provides performance characteristics at least equal to those obtained with conventional excitation systems. 1 \ 26. 23.“30 Excitation Systems SELF ‘ENERG. ) . The Rototrolexcited field of the main exciter also acts as a stabilizing field under regulator control. The manual control unit is not required. The Rototrol buckboost pilot exciter supplies the proper voltage to field 2 of the main exciter to control the output voltage. 45. complete excitation is not lost or is the continuity of the load disturbed upon the occurrence of any trouble in the Rototrol buckboost pilot exciter fircuits or in any part of the impedance. The voltage regulator potential unit. and the excitation provided by field 2 adds to or subtracts from this base excitation to vary the output voltage. 46 are those described in Sec._C. the Rototrol pilot exciter operates in a different manner from that in Fig. depending on the energy requirements of the mainexciter shunt field.hi 71 _. MAIN EXCITER Chapter 7 systems. Even in the event of a short circuit or open circuit in the pilot exciter output circuit. ROTOTROL <SERlm ES FLD. Rototrol BuckBoost Pilot Exciter The buckboost Rototrol excitation system using a twoor threefield main esciter. 6. All of the voltage regulator component parts in Fig. In this respect this system is identical with exciterrheostatic systems using pilot exciters. SELFENERG. In the system of Fig. The essential advantage is the elimination of the comparatively complicated exciterrheostatic regulator with its moving parts and elimination of the motoroperated mainexciter field rheostat. The Rototrol can easily be constructed to provide rates of response and ceiling voltage equal to or in excess of those obtained with conventional dc machines. 25.J”$ UISIK. pilot Fig. as shown in Fig. therefore. The excitation system shown in Fig. which is connected directly to the field and is under the control of the voltage regulator automatic control unit or the manual control unit. and voltage control is by means of the shuntfield rheostat. If a circuit failure occurs when the ac generator is carrying a load other than that used to determine the rheostat MAIN EXCITER SHUNIT FIELDS I r4 ‘: 5ch . The Rototrol pilot exciter in Fig. 46Excitation exciter system with and threefield Rototrol buckboost main exciter. CONTROL &Vl. 45. the preset base excitation remains rheostat controlled and undisturbed. loss of the pilot exciter through a short circuit or open circuit causes loss of excitation on the ac generator. TRANSE . voltage adjusting unit. Thus the Rototrol must be capable of bucking or boosting the main exciter base excitation to give the necessary range of main exciter voltage.T . A Fig.errheostatic excitation system. except that no regulatorcontrolled. As is the case with the excit. since manual control is obtained by operating the main exciter as a selfexcited exciter with a stabilizing field. 45 supplies all the excitation requirements of the main exciter. 45Excitation singlefield main system with Rototrol pilotexciter and exciter controlled by impedancetype regulator. the excitation provided by field 1 is set by the operator to give a base ex I J Fig. Briefly. The Rototrol pilot exciter used in this escitation system can provide either one or two stages of amplification. offers a number of advantages over the singlefield main exciter system described in Sec.. citation in the main exciter. The operation of the threefield main esciter was described in Sec. motoroperated csciterfield rheostat is used.type voltage regulator elements. Variable voltage is supplied to t. 47Excitation system for hydroelectric generator with motordriven Rototrol buckboost pilot exciter and threefield main exciter. 46. 23. Since the main exciter base excitation is supplied by the selfexcited field. 46.
the Rototrol can he stnrtcd and the voltage regulator placed in service. Where some form of current limit is desired as discussed in Sec. 47 for waterwheel generators. The threefield main exciter and Rototrol huckt. Rototrol Excitation Condensers for Synchronous The Rototrol escitation system for synchronous condensers is similar to that shown in Fig. 41.Chapter ‘7 Excitation Systems 231 setting.oost pilot exciter excitation system of Fig. 25 + I 1 REFERENCE 1 VOLTAGE  1 1 II 1 SL Y3GT AC GENERATOR VOLTS FROM POTENTIAL TRANSF. Ii t L M TI i T2 The variable output appearing across terminals exciter or can be adapted to control a Rototrol Fig. but at a different power factor. the main exciter for the condenser is also motordriven so t. in the usual case.rol to the &aft of a waterwheel generator. must he driven by a small motor. this is accomplished by building sufficient inertia into the Rototrol mg set to carry it through such disturbances with very little change in speed even under severe forcing conditions. the ac generator continues to carry its kilowatt load. and to control the magnitude I I I I O24 1 VARIABLE NEGAT IVE BIAS i I ot . Rototrol ators Excitation for Hydroelectric Gener it is impractical to directconnect the Rotot. 27. However. Electrically. Under shortcircuit conditions on the ac system. the TO DAMPING TRANSFORMER y excitation system must he capable of supplying full escitation to the generator field. During startup of the generator when no outside source of supply is availahle for driving the Rototrol motorgenerator set.hat the motor supply circuit has to be modified to supply sufficient power for the motorgenerator set. The Rototrol pilot exciter. . introducing the prohlem of a reliable power supply for the driving motor. The rectified ac load current of the condenser is used to energize the center or control winding of the threelegged reactor. because of the multiplicity (If speeds and sizes involved. the circuit is the same as Fig. the main exciter is operated as a selfexcited machine and provides excitation for the main generator. a static currentlimit device can he used with the Rototrol excitation system. hut the main and pilot exciters arc normally on the same shaft and driven by a large motor. As soon as ac voltage is available. 47. 24 and 25 can be used to control the 5ring point of an electronic main excitation system. or operation can he continued under hand control with the operator controlling excitation with the selfexcited shunt field rheostat. 28. The circuit of the currentlimit unit is similar to that of the minimum escitation unit shown in Fig. 47.48&h ematic diagram of the electronic generator voltage regulator. 46 is readily &ptnble to use with slowspeed generators and is shown in Fig. therefore. With the system shown in Fig. 47. 22.
Valentine. which is applied to the limits field of the Rototrol. 2. April 1935. 601606. Recent Developments in Generator Voltage Regulators. R2.E. At a certain magnitude of ac current. The grid circuit of the GVGGT tube can he traced from the grid through resistor R7 to the cathode. September 1934. the output of the bridge circuit. Transactions. Transactions. by A. Vol. c i. The current in the limits field is in the direction to lower the excitation voltage.age applied to the series reactorcapacitorrectifier circuit. E. This grid bias establishes the current in the 5693 tube and the drop across IU. G1. The rectified output of transformer 1’1 is fed into a twosection condenser input filter giving a smooth dc voltage with polarities as indicated. The Electric Journal. 361364. by A. E. A. Ttansaclions. Tramaclions. The dc reference voltage is obtained from the voltage drop across a type VR105 voltage regulator tube connected in series with resistor R6 across the dc power supply. The output of the voltage amplifier is fed into a power amplifier using a GV6GT tube. Vol. M. 143144. which is an industrialtype tube with characteristics the same as a type GSJ7 tube.E. the output of which is applied to a voltageadjusting rheostat and a modified Wein bridgetype filter. The amplified voltage from the 5693 tube appears across the load resistor tz?’ with polarities as shown and this voltage drop isapplied to the grid of the BVGGT tube. pp. vnn Siekerk. Hanna. 23.i 232 Excitation Systems 5 i Chapter 7 : of the volt. Static Voltage Regulator for Rototrol Exciter. Time delay can be provided in the limiting circuit to enable full forcing of the condenser escitation during transient overloads. by J. Vol. Transadions. 29. The bridgetype filter provides a high degree of filtering without adding unduly long time constants to the regulator input circuit. 8.E. 23. 1947. the differential connection of the rectified generator voltage and the reference voltage makes the grid bias of the 5693 tube more negative than previously. Reference to Fig. 4. The rectified generator voltage is connected differentially nith the reference voltage and applied to the grid circuit of the 5693 tube. from the positive side of i he rectified generator voltage to the positive side of the wference voltage. The bridge. 66. Lynn and C. G. A fullwave rectifier (5Y3GT tube) is used to supply the plate voltage of the 5693 tube. The lower voltage drop across R7 reduces the negative bias voltage on the grid of the GVGGT tube and causes an increase in current through the tube and load resistor R9. a typical one being shown in Fig. Rs and R4. Vol. Thus.E. 31. 529534. Vol. A dc voltage proportional to the average threephase ac generator voltage is obtained from a threephase bridgetype rectifier. Under balanced conditions when the ac generator voltage is equal to the regulated value. the grid of the 5693 tube is established at a particular bias voltage depending on the magnitudes of the reference voltage and the rectified ac voltage. by W. August 1934. 31. low ac voltage causes the grid bias of the 5G93 tube to be less negative than previously. Valentine. 7375.E. Harder and C. 67. pp. A. The generator voltage regulator consists of two dc amplifiers and a reference voltage. Vol. 67. Jr. comprising resistors RI. bY C.E. Valentine. The reference voltage is also a smooth dc voltage that remains constant for wide variations of supply voltage. This particular regulator is used with the electronic main exciter in Fig. 838844. is a smooth dc voltage. A. 3. IMain Exciter Rototrol Excitation for Turbine Generators. Should the control field be conducting current in the raise direction. February 1935. Determining the Rstio of Exciter Response. Vol. 3. by A. 308312. The Generator Rheostatic Regulator. which is the input to the regulator. 9.I. G.E. 32. Vol. pp. A. 32. Vol.Z. ‘C !J . G.E.. Bnrkle and C. 564568.I.E.E. Gower.E.I. pp. QuickResponseExcitation. 7.I. W. the negative voltage output across terminals 24 and 25 is increased. Lewis. pp. A. pp. which causes a reduction in the voltage across terminals 24 and 25 and a consequent reduction in the thyratron firing tube angle of grid delay. The ExciterRheostatic Regulator. Regulation is obtained by comparing the rectified generator terminal voltage with the reference voltage. filters the 3GOcycle ripple voltage in the dc output of the rectifier. TwoStage Rototrol for LowEnergy Regulating Systems.yi r’ :g. 58. The Electric Journal. E. REFERENCES 1. Current in the 6V6GT tube is thus fixed. 1948. by C. which reduces the mainexciter voltage. The first dc amplifier is a highgain voltage amplifier using a 5693 tube. This circuit can be traced from the grid of the tube through the grid resistor R5 to the negative side of the rectified generator voltage. A. the combined effect of the two fields is such that the excitation voltage is held constant at the limiting value. Gower. 1939. 24 shows that the increase in negativebias voltage on the thyratron firing tubes causes an increase in the angle of grid delay. 48. by /G . by E. Tranmtions. The voltage output is constant as long as the ac generator a voltage is equal to the regulated value. Valentine. 1948. 15071511. The highgain voltage stage gives the regulator its high degree of sensitivity and the power amplifier supplies the variable negative bias voltage for controlling the thyratron firing tubes in Fig.. Rototrol Excitation Systems. 10. The Electric Journal. Jr. pp. Should the ac generator voltage increase above the normal value. GG. In a similar manner. ElectronicType Voltage Regulator I%zctronictype voltage regulators are available in many dif’frrent forms. A. 1947. which reduces the current in the tube and in resistor R7. pp. R. The Multistage Rototrol. A. but it can be modified for use with Rototrol escitntion systems. 535539. Liwschitz.E.I. E. pp. A. Thus. 1945. The variable negative dc voltage output of the regulator is obtained across the load resistor R9 of the 6V6GT tube and applied to the grid circuits of the thyratron firing tubes in Fig. which in turn establishes the grid bias of the GVGGT tube. pp. the series circlIit begins conducting a rapidly increasing current. The Eleclnc Journal. Oplinger and C. as is the drop across load resistor R9. Kimball. 23. K. and from the negative side of the reference voltage to the cathode of the 5693 tube. Vol. E. L. by M. capacitors Cl and C%’ and potentiometers P2 and PS.
If the capacitor kvar is small. capacitor use has increased phenomenally year by year. As shown by Fig.rical equipment and circuits on the source side of where they are installed. 10 000 and 15 000 kvar are in service in a number of larger substations. Increases voltage level at the load. It supplies the kind of kilovnrs or current to counteract the outofphase component of current required by an induction motor as illustrated in Fig. 7. 2. Their use. 233 .. Usual voltage ratings of capacitor banks star> at 2400 volts and range upward for groups of capacitors connected in series for 36 kv. f~. it may be desirable. Shunt capacitors are applied in groups ranging from one capacitor unit of 15 kvar to large banks of these standard units totaling as much as 20 000 kvar. Reduces PR power loss in the system because of reduction in current. say ten percent of the circuit rating. History Shunt capacitors were first applied for powerfactor correction about 1914. to increase the field currents to these motors. Extensive utility use started after the appearance of outdoor units. where the capacitor kvar is large. SHUNT CAPACITOR FUNDAMENTALS The shunt capacitor affects all elect. Johnson I. Thus. A. Consideration is being given to voltages up to and including 138 kv. Shunt capacitors applied on the load end of a circuit slIpplying a load of lagging power factor have several (‘l’fects. 9. 2. Larger banks of 5000.CHAPTER 8 APPLICATION Aulhor: OF CAPACITORS TO POWER SYSTEMS A. 3. it is usually sufficient to make an analysis on the circuit involved for the application. T ES Z=R+jX m =F  EM I. 1. if possible. Decreases kva loading on the source generators and cir cuits to relieve an overloaded condition or releasecapacity for additional load growth. its effect on each part of the system back to and including the source should be considered. Many small banks of 4. In determining the amount of shunt capacitor kvar required. was limited during the next twenty years because of high cost per kvar. Banks of 520 kvar to about 3000 kvar are common on distribution substations of moderate size. lShunt 1. as shown in Fig. If the load includes synchronous motors. : 0 FEEDERS GEN xc’ 1 = IND MOT SHUNT CAPACITOR EM QFig. which eliminated steel housings and other accessories.. by an in Capacitors supplying kvar required duction motor. one or more of which may be the reason for the :Ipplication: I. At about this time the introduction of chlorinated aromatic hydrocarbon impregnating compounds (askarels) and other advances in the capacitor construction brought about sl’@$?eductions in size and weight. Improves voltage regulation if the capacitor units are properly switched. 5. Correction to 100 percent power factor may be economical in somecases. Before 1937 practically all capacitors were installed indoors in industrial plants. ReducesZ*Xkilovnr loss in the system because of reduction in current.I shunt capacitor has the same effect as an overexcited synchronous condenser. Reduces lagging component of circuit current.5 kvar to 360 kva are installed on distribution circuits. large size and weight. 2 the present weight per kvar is less than 5 pounds compared with over 20 pounds in 1925. However. 3. 4. 1. however. Increasespower factor of the source generators. This is feasible provided the bank is sufficiently large in kvar. in system facilities per kilowatt of 10. TO reduce demand kva where power is purchased. P$?$‘to 1932 all capacitors employed oil as the dielectric. . By 1939 capacitor costs had been reduced almost proportionately with weight and they had been proved in service. to get the desired correction some additional capacitor kvar may be required above that based on initial conditions without capacitors. The acceptance of capacitors has been due to the following: HE function of a shunt capacitor applied as a single unit or in groups of units is to supply lagging kilovars to the system at the point where they are connected. heduces investment load supplied. generator or motor. By reducing kva load on the source generators additional kilowatt loading may be placed on the generators if turbine capacity is available. it must be recognized that a voltage rise increases the lagging kvar in the esciting currents of transformer and motors. Starting in 1939 and continuing to the present.
Capacitor Failure Rates To evaluate the operation and ecaonomics of shunt capacitors. 0 0 I 2 3 4 5 6 7 8 9 IO YEARS IN SERVICE Fig. Better understanding of system benefits that xcrue from their use. Reduction in failures. such as given in Curve A. By adding shunt capacitors at the load. 5. 5. 4Failure rate of shunt capacitors. Reduction in selling price. the economics of using cnpscitors is fnvornble. 1. 2. For such performance each unit should be inspected and tested at the installation to weed out units damaged in transportation. Individual capacitor fuses are also essential for best performance as discussed later under Capacitor fusing. 3. the more effective the capacitors are. Due to the large volume of production (luring the war and since. In view of the benefits a performance as given by Curve I3 has been considered economical and satisfactory. Sot only cl0 unit failures mean the loss of the units but also. 4. the kva from the source is reduced materially. during the war emergency of 1939 to 1945. unc!cr certain conditions a unit failure may damage other good units. poletype units and standardized mounting brackets. This situation is illustrated in Fig. other hazards. Also less critical material was required for capacitors thnn for other kvnr generators. exposed instnll. Development of outdoor. 6. By force of circumstances. This clwve gives cumulative unit failures per 1000 units in service regardless of how or where they are installed or how they are protected. Improved design and manufacturing 3.ztions. 2Evaluation of the size and use of Shunt Capacitors. lightning prot c>c*tion and subject to ‘is10 1918 1926 YEAR 1934 1942 1950 Fig. 4. 2. . Fundamental Effects Fig. of unprotected. Fig. of well protected larger installations. 3Cutaway view of 25 kvar 2400 volt outdoor unit. Curve C represents performance of large banks of capacitors where careful attention has been given to operating conditions and protective devices. The lower the load power factor. capacitor To illustrate the effects of shunt capacitors. Prediction of failures can he based on past experience. c AAverage BAverage CAverage of all types of installntions. manufacturing facilities for capacitors lvere more available than other means of supplying kilovnrs. it is helpful to predict the number of unit failures that may occur.234 Application of Capacitors to Power Qstems Chapter 8 7. methods resulting in small size ant1 weight. Curve 13 represents unit failures of small groups of capacitors distributed over a system lvithout. assume that a lOOkva circuit or piece of apparatas has to carry 100 kva at various power factors.
y. 5(C) sug . 5Fundamental effects of shunt circuits. 5(F).ity by the cost per capacitor kvnr. then the increase in ability to supply load is obtained at a cost of 1. 5(D) ).. KVAR IN % OF CIRCUIT KVA 90% F! F.00 per kvar. then adding 4000 lcvar of wpacitors permits the km to be increased from fO’)O t. the capacitor kvar per increase in kva of the load is 1. Thus if the load power fwtor is 70 percent and a capacitor kvar of 40 percent is added. &Adding capacitors releases circuit capacity for more load. the ordinate in this 0 3 I CAP. (If the load should be 10 000 kva at 70 percent power factor.65 timrs J7. The incremental capacitor kvar required for an increment in kva of the load is Fig.65. as well as transmission and distribution lines. Expressed mathematically. The load kva can thus be incrrnsetl to 12 400 kva at 70 percent power factor. up to the Illtimate point at which capacitors supply al1 of the kiloVW required by the load and the circuit supplies only the IciIonatt component. If capacitor cost is $7. is of particular interest. By adding 40 kva of capacitors to a ‘00k~~ load of 70 pcrccnt power factor the load can be i~~~~~~:wd from 100 kva to about 124 kvn.Chapter 8 Application of Capacitors to Power Systems 235 (4) INCREASE IN LOAD KVA gests.) Shunt capacitors can be vicwcd in two lights. The incremental cost of adding transformer capacity may be much greater per kva 07 increased capncit.‘as Pig. The same data apply equally well to any equipment other than transformers in which current might constitute a limiting factor such as generators. I I . 5) as the load through the transformer approaches unity power factor. can be compared with the cost per kvs of increasing the transformer or supply circuit rating. &Reduction in losses in the capacitors. capacitors on power Increasing the capacitors lessens the current carried by tI)c supply circuit from the source (Fig. 5(E). regulators.00 or $11. smaller and smaller incremental gains in load are obtained for incremental increases in capacitor kvar. I OF LOAD . Icor a constant load in the circuit adding Garious amounts of capacitors allows the useful IwI to be increased. neglecting other advantages of the capacitor.o S700 nithollt incrcnsing the circuit loatling above IO 000 kva. This cost. 60 eo too CAPACITOR KVA IN PERCENT OF CIRCUIT KVA Fig. I. source circuit to shunt . and adding capacitors relieves overloaded circuits. The capacitor kvnr per kva of load increase. cables. Fig.55 per kva. because multiplying this quant. the product is the average cost of supplying each additional kva of load. Fig. In the example taken (Fig.
Curves are also shown for half load or 7500 kva at 70 percent power factor. the reactive component ES ER R I X LOAD=I. 4. particularly if the circuit impedance is high. The new voltage drop becomes approximately : Voltage drop = RI.F. If a capacitor is placed in shunt across the end of the line. This expression also shows that if the Fig.lL=HL I JLAG (a) OA. the I*R and 1% loss will be 59 percent of its former value. on voltage drop in source 0 2 SHUNT 4 CAPACITOR 6 0 KVAR X 1000 IO as shown in Fig. I. and 1. whereas with the use of 6000 kvar of capacitors the generator power factor is raised to 82 percent. By providing 3 steps of capacitors and removing 4000 kvar from the system at i/z ’ load.000 7. 7 are shown where it is assumed that the load bus voltage is maintained constant at 4160 volts and the generator voltage varies with load.236 curve is equal to Application of Capacitors to Power Systems Chapter 8 d (Cap.9 kv at the generator. which is almost equal to the capacitor kvar. The 12R loss in the circuit is reduced by about 800 kw (19001100) and the PX losses are reduced by about 1600 kvar (40002400).1 kv at half load with 6000 kvar of capacitors. To illustrate the effect of shunt capacitors applied to a large load. both the RI. 13.500 SHUNT T CAP.8 kv at full load with 6000 kvar of capacitors to 12. X the reactance. removing all capacitors from service.6 KV wt 4160 V 15 MILE33 KV 4/O ACSR KVA 15.VOLTAGE AT LOAD .6  6000 I 1 I c I I I/ I .300 > Y 3 2 J ? . the drop immediately decreases or the voltage rises. 6 For a 70 percent power factor load with 40 kvar of capacitors added for each 100 kva of circuit capacity.ER O&VOLTAGE AT SENDING END WITHOUT CAPACITOREs OC. the curves in Fig. it is desirable to switch part or all of the capacitors off during lightload periods. The voltage and power factor at the generating end determine whether switching in steps should be applied. drops can be neutralized. E 0 l51413l2 LOAD KVA 15. 70 % 70 % LOAD 15.VOLTAGE AT SENDING END WITH CAPACITOREs (b) Fig. Thus if 10 be made sufficiently large.4 kv is required for 4160 volts at the load.+XI. the power component of the current. In the case cited.900 9. and the XI. P CAI? KVAR 0 0 A I5 OOOkva. is the current drawn by the capacitor. the remaining 2000 kvar gives a voltage of 12. As Fig. Voltage Drop The voltage drop in feeders or short lines can be expressed approximately by the relation Voltage drop = RI. assuming a constant voltage of 4160 at the load. at full . Capacitors applied to a given load reduce the I*R and 1*X loss in the supply circuit in accordance with Fig. a generator voltage of 13.0 $ 0. kvar) These curves d (Increase in load kva)’ show that the final increment is attained at much greater expense than the initial increment.XI. 7 indicates A the voltage at the generator would have to vary from 13.I.0 a’ 5 0.000 KVA 1% 5% LOAD P. The resistance and reactance losses are also reduced in all circuits and transformers back to and including the source generators.000 5% GEN. The 6000 capacitor kvar reduces the loading on the generator by 5850 kva.+XIx (1) where R is the resistance. 8. SEffect of shunt capacitors circuit.O SHUNT d1 Y PER UNIT CAPACITOR ~ II 7000 FULL LOAD 1 1 /I 1 SOURCE’ L 1. (2) where I. Without shunt capacitors the generators must supply 19 900 kva at a power factor of 62 percent.70 percent power factor load is supplied over I5 miles of 33kv circuit. 7Effect of various amounts of shunt capacitors and half load on a practical system problem. This loss in the particular circuit supplying the load can be calculated directly and may be a big factor. KVA 19.
At light loads the voltage rise might be so much in excess of normal as to represent an ‘indesirable or even intolerable condition. total impedance 0.Chapter 8 Application 90. 5 or 10 percent voltage cuit. ratio R/X=0. ER 237 60 50 60 v/ A/ /I/ Y / A’ / 60 PERCENT CIRCUIT REACTANCE X PERCENT CIRCUIT REACTANCE X PE RCENT CIRCUIT REACTANCE X Fig. These data and the capacitor kvar required . 5 and 10 percent voltage drop over the supply circuit. Es=l.36 ohms or 0.45 and a circuit reactance of 0. Assume the full load is 10 000 kva at 80 percent power factor.54 per unit. In this case 1.62+j15. All percent values are referred to full load kva as 100 percent base. 9 show the amount of shunt capacitor kvar required for loads of three power factors and for 0. is (lcpendent only upon voltage and not upon load. drop in the source cir voltage drop is compensated at full load with permanently connected capacitors. 9 for R/X ratio of 0. Referring to Fig.8 kv. TO illustrate their use. transformer impedance 0. a solution is to Provide manual or automatic switching to add or remove 8rOUPs of capacitors as desired. the shunt capacitor kvar required for a 10 percent voltage drop on the line is 0. of Capacitors to Power Systems I . 9Shunt Capacitors required for various power factor loads to give 0.05 .45. 0.212 per unit. 33kv line of 2/O copper conductors which steps down through a 10 OOOkva. and I.096 Therefore.212 per unit.0 per unit is 10 000 so 5400 kvar of shunt capacitors are necessary.0%36+jO. become smaller and the line is overcompensated because I.07 per unit. Regulation of the line is practically unchanged by the capacitor I)ecausethe capacitor effects an increase in voltage both at light load and at full load.142 per unit on 10 OOOkva base.212=0. The curves of Fig.008+ jO. Also assume: line impedance 9.Tpercent reactance transformer to 13.096+jO. . then at light loads I. assume a 20mile.
~cl I Circuit I Percent Potvcr Factor At 230 V 460575 30 GO 00 120 v 2400~lIti 30 45 GO 90 v tioo* 900 * 1500* 2100* 2.!4 15 15 25 25 25 II.9 60 00 135 1so 270 360 . TABLE IDATA FOR 20 MILE 33 Kv LINE WITH TRANSFORMATION TO 13.Ol Es =E. To give a more complete view of the llse of Fig. as wtll as the pwvcr factors nt the scntling (Es) and rccei\.izing current required by the load. 15 & 10. 5.800 Capncitors At.02 5.x..0 ii Ic:td 5 lag 0. Each capacitor has a builtin high resist. CAPACITOR ON INDUSTRIAL CIRCUITS PLANT l&3 l&3 1 i 1 13800 1 Note: 25 KV. It can be located at the service entrance. is onethird of one percent of the kvar rating.540 720 1080 1260 13. kvnr can he determined.5. for outdoor open mounted units it. The average operating loss for capacitors.vpe KYAR 1. Overvoltage on Capacitors Capacitors arc designed for operation on circuits whose average voltage over a 24hour period does not exceed the rntetl voltage by more than 5 percent. which means that the voltage applied to the capacitor is 150 percent of its rating.ol kvar in use.4pplication period: 3600 ryrles or 60 cydes. or 110 percent in the case of higher voltage unit)s. !) c~wves the shunt capacitor kvar required for 3OOOkva %Ipercent power factor 10x1 is included in Table 1.ing (I:‘n) cntls of t.acitom. during the starting of large induction motors the voltage rating of capacitors applied in shunt with the motor may be as low as 67 percent of the voltage applied to the motor. The load may be a single motor. for any part of full load by reactance based on the partial load.5 percent of the rated voltage. thus removing magnetizing current .. The maximum momentary voltage.4R Units TABLE ~FACTORY Voltage Rating of Cnpnciton 3 ~ 230 460 I 5.8 Kv LOAD BvS 6. TABLE 2 STANDARD ~ TEST VOLTAGES ON CAPACITORS i’ TerminaltoTerminal Test Voltage 500 Indoor 3000 1000 1200 so00 5000 575 2 400 4 160 4 800 7 200 7 9GO 12 4i0 13800 so00 9000 10 000 15 000 IG 600 25000 28SOO 10 000 19 000 26000 26 000 26000 34000 34000 I . in kw. should not exceed Ifi.500 194 156 234 94.050 2.os EEL 0. is 0. The variations above the average may go to 115 percent in the case of 230. The ratio Thus the capacitor voltage drop in the using the per unit 540 630 lSO0 21G0 2520 I *Using 25 KV.lO E 0.00* 3300* 4200* V A 15 30 45 !I0 180 360 600' 900” 1500* 2100' 1.c Es = 1. 9 1:1g lead on.%‘C and for housed lmits between 40°C and 50°C depending on rack type. For example. 10 & 10. Phlt% 1& 3 l&3 l&3 l&3 4GO 575 2 400 4 160 4 800 7 200 7 060 12 470 13 800 10 & 15 10 85 15 15 & 25 15 & 25 15 & 25 15 & 25 15 & 25 15 7.’ 180 100 81 0 13R 540 7’0 1080 12GO 1440 2iOO' 3300* 4200* 5100" 180 270 360 270 360 540 $20 1080 1260 For 5000 kva the circuit reactance R/X remains const. 5.106.IR Units are only single phase ( 15 15 15 1 1 1 1 A capacitor can be installed in shunt with any load of lowpowerfuctor to supply the magnet. The ambient temperature TABLE ~STANDARD RATINGS FOR INDOOR AND OUTDOOR HOUSED CAPACITORS BANKS VOLTAGE AND KVAR RATINGS 4800i2007060 12. or it may be a large industrial plant.ance device which automatically discharges the capacitor for safety. In adtlition.o. is . and 575 volt capacitors. The capacitor can be chosen to supply the magnetizing current under peak load conditions.ant for all loads. 10 & . #X3. 2 lag 58. Standard Ratings and Tests on Capacitors Tal)le 2 gives the standard ratings of capacitor units for indoor and outtloor typc. For short periods of time. Table 3 gives t hc Standard ratings of indoor and outdoor hor~sctl capacitors. for a given circllit. calculated losses in the circuit are given.he circrlit with t!he wlccted capwit. 15 $ 15 & 15&25 15&25 limit covering a11 capacitors is 40°C. Table 4 gives the factory test voltage \\hicti are npplicd to c. I 34000 34000 CAPACITOR RATINGS Volts 230 KVAR 57x T Phase l&3 l&3 lh3 l&3 l&3 l&3 I__ Volts 230 460 575 2400 4160 4800 7200 7960 12470 Outdoor T. 255.50 E=l. or it can be chosen only large enough to supply the reactive kva hours accumulated over the month.4iO13. tlw T.238 Application of Capacitors to Power Systems Chapter 8 for 0 and 5 percent voltage drop are given in the following table. such as in welding applications. shunt capacitors can safely withstand higher voltages.
ocnlizedcorrection on smnll feeders. Esceptions lvill arise nhere feeders are long and where the gain from intlivitlual load application warrants the greater initial investment in capacitors. In this case. the \‘:ll%ttion in load. and considerable savings can be made by supplying this mugnetizing current with capacitors. The size of the capacitor or the merits of their use can only be determined by systematic analysis. magnetizing current is drawn from the line. motor voltages 3re low such as 230 volts. if) Localized correction tlirect on motors. A centrallylocated group capacitor in this case would be an advantage since it would tend to be the same distance from the loads at all times. or units can be applied to the individual loads. 10. 3.Chapter 8 Application of Capacitors to Power Systems 239 fr()~nthe utility system only. and better maintaining voltage at the l0UlS. therefore. such as an industrial plant with a day load averaging 960 kw and 67 percent powe: factor. It is obvious that a lnrgt. 8s for example for one buil~ling. reducing their loss. Recause of the higher cost of lowvoltage capacitors their application to 230volt motor circuits may more than double their cost.he service entrance to various widelyseparated parts of the plant and if the loads shift about a great deal from one feeder to another.s in the plant. IOEnclosed indoor bank of 2400/4160 volt shunt capacitor units with protective screen removed. This voltage boost. Voltage is raised near the loads. no transformer kva is released. constancy of load distribution. (1)) &oup correctionat secondary of transformer. Losses are reduced in the circuits between the loads and the metering point. (C) Group correctionout in a plant. ‘Uw capacitors can be locatcci% many ways as follows: 01) Group correctionnt primary of trnnsformer. Location of Capacitors Many factors infltlence the location of the capacitor such :LJthe circuit. Fig. Where 1. distribul. Capacitors placed aheatl of the main bank of transformers do not benefit the transformers. with peak loads running up to 1200 kw and 75 percent power factor. and the transformations. and local conditions. use of the 230volt capacitors on the feeders or near the motors is frcquently warranted. superimposed on the normal voltage. The selection of the capacitor size. Group Correction The two principal conditions under which group correction is better are: . would be better for the operator to switch off portions of the central capacitor t. and increasing their load capacity. 2400 to 7200 volts. the power rates. This is one step voltage control with a RCOC oil contactor. or groups of motors and switched with the motor. the corrcction may be needed first in one part of the plant and late in another.[cpcntlcnt8 on what is to be accomplished. Localized Correction Capacitors should be placed as near the load as possible or near the ends of feeders for three main reasons: 1. capacitor units is shown in Fig. the length of the circuits. If a group capacitor remains connected during light loads the voltage rise is less if this capacitor is installctl at or near the transformer hank since the reactance of the plant circuits does not contribute to voltage rise. application of capacitors to individual motor w01~lcl represent a larger investment because of the diversity factor. This gives considerable aclvantage to group installation if this can be on the primary side. 8. This varies with : 2.o meet the varying load conditions. ((1) T. 10. (e) Lwrlizetl correction on branch motor circuits. R%ere loads shift radically as to feeders. if any. If the power flows from t. is practically constant from no load to full load on the feeder. Whatever gains are found in released transformer capacity and reduction in losses in transformers and circuits are added gains. thus removing magnetizing current from the plant circuit also. An outdoor bank of 7. giving better motor performance. It. and its location is . 2. The first point can be evaluated easily by investigating the length of the circuits. the load factor. 9. Rates and Capacitor For the purpose of analyzing the different types of rates a typical application can be considered. Capacitor kvar can be reduced automatically as the load drops off by installing some of the capacitors direct on loads so they are switched off with the loads. types of motors. The effect of the capacitor is to raise the voltage permanently at any given point where it is connected. Thus.iw of loatls.
the size of the blocks depending on the kva demand. (c) Sometimes a check is made on the average power factor under day load conditions. The size of the capacitor required to 0.‘.t Reactive kva at 9. This ratio amounts to a power factor of 68 percent.240 Application of Capacitors to Power Systems 960 kvv at 67 percent power factor. (b) If the rates include a kvs demand charge.95 accomplish this is determined from the reactive kva at the two values of power factor as follows. Assuming that rates indicate that it will be worthwhile to reduce this rkva hours to a point corresponding to 95 percent power factor. and the billing thereafter based on this check until some future check is made. Capacitor required is 1035 minus 313 which equals 729 kvar. the day load averaged Using 730 hours per month the capacitor kvar required equals 339 000 . (a) Power factor is not consiclerctl in the rates. Reactive kva at 75 percent power factor =d1600’1200”= 1060 Reactive kva at 9. in order to install capacitors to raise the power factor as high as warranted by the expected savings. the current in the circuit is reduced to the desired figure.e applies to a larger proportion of the energy consumed. (d) A method commonly encountered in industrial plants takes into account monthly power factor obtained by integrating kw hours and rkvs hours. or the net billing is adjusted up or down according to this power factor. 67 percent power factor = t/iZjiCF6B = 1(X3.106 000 or 319 where the kvar meter has 730 200 I no ratchet so that full credit results even if the power PER CENT RATED CURRENT Fig. such as to reduce current in circuits. In the case above. . reduce lo:& on transformers. (d) Power factor is determined by the ratio of kw hours and rkva hours and is used in different ways to calculate the ~lemnnd charge or energy charge or both. 1st block% per kw hour 2nd blocklxc per kw hour 3rd blocklc per kw hour Additional Zdc per kw hour In this case the energy cost is reduced by a decrease in kva demand. voltage. the lower rat. With a demand of 1200 kw at 73 percent power factor the kva demand is = 1200 0. &suming the plant mentioned above is billed for 322 250 kn. ACapacitor voltage. The capacitor should usually he located near the loads of low power factor. In such cases it is necessary to determine how this check is to be made.75 lGO0 kva. current less than motor current ut no load rated : BCapacitor current equal to motor current at no load and rated ‘1 CCapacitor current equal to 100 percent. (I)) Power factor is taken into account in demand charge. because if the blocks are smaller. Some rates involve several energy charges for successive blocks of power. brought up to 95 percent power factor. the kva can he reduced by raising the power factor tlrlring the demand kva at 95 percent power factor =@? = 1010 kva. and under what conditions. or it may reduce the energy charge depending on the rate structure. Such a capacitor usually is made proportional to day load requirements. (c) Power factor is checked by test and used to determine energy charge thereafter.5 percent power factor =d126021200*=387 Kvar rating of capacitor is 1060 minus 387 which equals 673 kva.5 percent power factor =2/1010”960?=315. the capacitor can be used only to secure savings in the plant. . If the power factor is raised to 95 percent the demand kva 1200 is = 1260 kva. 720 kva of capacitors are required as follows: 960 kw = lL30 kva 67 percent Reactive kva at. and that the reactive kva hours equals 346 000. and to reduce loads on customeroperated generators.50’ = 106 000 peak. 11Self excitation of induction motor with various amounts of shunt capacitors when supply breaker is opened.hours. For example: Size of block = (70) X (kva demand). kvar (a) If power factor is not taken into account in the rate titrrlcture. l<y using a capacitor large enough to supply all or part of this reactive kva. The energy charge. 000 kva hours at 95 percent power factor = E+g Reactive kva hours at 95 percent power factor = d339 0002 322 2. The size can be determined by calculating the reactive kva. The reduction in the kva demand from lGO0 to 1260 may result in either a reduced kva demand charge. Assuming Chapter 8 this is to he One of the following conditions may exist.
chapter 8 Application of Capacitors to Power Systems 241 factor is leading at times.he load points. not cause excessive voltage at the motor due to selfdo TABLE ~MAXIMUM CAPACITOR KVAR FOR USE WITH OPEN TYPE THREE PHASE 60 CYCLE INDUCTION MOTOR 1800’ RP5I 7 I *c  1200* RPM 900* RPN Kvar 5 7.tgc on plant circuits. Where the individual loads are not known.5 15 17. the effect of capacitors in reducing the current is dependent upon load power factor.5 15 15 17.zI KVX livnr ** 2 5 9 2. 11 shows.j.5 21 10 20 19 10 12.. Usually the minimum reactive kva determines the am=ofGed%$iiitors to apply without automatic control. 5.5 22.** Kvar .5 .5 14 27. Detail analysis of the load and its variations at each nlsnt. Capacitors on Induction Motor Terminals Qpacitors frequently arc installed across the t.ation when the breaker is opened.5 15 17. such as network ‘iruts. KVU 4 5 5 i . certain system data are required. CAPACITORS ON DISTRIBUTION CIRCUITS Shimt capacitors offer a convenient and practical means of rclicving lines and source equipment of wattless current. as Fig.5 10 IO 12. shunt capacitors cannot materially change feeder loading. Fixed shunt capacitors raise the voltage level at the point where they are applied on a given circuit by practically a constant value as given by XI. In certain cases more fixed capacitor kvar can be applied where voltage conditions at light load permit and where leading power factor is not objectionable.5 45 ** 17 16 15 14 13 12 12 11 11 10 9 9 9 i20* RP. 2.5 low. ‘1’:tble 5 gives the maximum recommended capacitor kvar for direct connection to the terminals of induction motors t&n from the 1047 National Electrical Code.. in Eq.5 10 12. 12. lVlwc the transformers are expensive. This gives.ll . groupfused. installation. 600* PM ** 28 26 24 22 21 19 17 16 15 14 13 13 12 Fig.5 27.5 20 25 30 37. 1. Determine variation. If the power factor is In applying shunt capacitors to distribution circuits. They usually arc arranged in three1~1~~~ banks of 45 kvar or more and are distributed over tlw system at distribution voltage. the capacitor kvar to decrease voltage drop is dependent not only on load power factor but also on the ratio of resistance to reactance of the distribution feeder. Where voltage is the limiting factor. 17light load periods to prevent excessive v.). The :unc)rint of kvar SOconnected should be limited to values t.5 . A 180 kvar inst:tllation is shown in Fig.5 22.5 11 10 60  L 7. of kw and kva on each feeder for a typical 2lhour period at both minimum and maximum daily loads.i 10 10 12. should 1)1) matlc to obtain the greatest benefit from using capacil. taking into consideration the type of rates.‘s. about unity polier factor at minimum load.5 9 Kva.5 35 42.5 15 17.5 !I I !5 9 I7 5 9 22. preferably by graphic instruments.5 13 12 35 11 40 47. polemounted capacitor 35 42.5 25 30 37.erminals . usually 2400 volts and 1% in accordance with local requirements.5 9 25 9 32.r 4 5 5 7. a large reduction in feeder current or kva can be obtained as indicated by the curves in Fig.5 4i. ‘h capacity of a distribution feeder can be limited by (!irrrcnt or by voltage drop. III. secondary capacitors may be justified.3 27.hc Capacitor must he large enough to build up nccumlilatcd kvarhours while the power factor is not leading. it is reasonable to assume they are proportional to the installed transformer capacity for minimum and maximum feeder load.5 1s 16 15 15 20 22.5 10 10 12. 2.5 22. Where current is the limiting factor. 11.. 12. ‘l’lwy can be installed in relatively small banks and placed i~:tr t. To simplify calculations singlephase loads can be grouped together to form balanced threephase loads and adjacent threephase loads can be grouped to simplify the calculations. .5 55 67.\I:Lt.5 7. When the meter has a ratchet t.vit~:tiecloff durin. To calculate the voltage at various points on the feeder the circuit characteristics and the load distribution must be known. Application Factors w:it.f induction motors and switched with the motor.5  I1 11 10 I) 9 !I s 8 8 8 8 8 8 5 23 7. Obtain actual voltage measurements on the feeder during full load and light load at a sufficient number of points to determine the ‘optimum location for capacitors. 12180 kvar.5 9 12. If the load power factor is high..5 9 5 !I 5 9 7 5 ‘9 IO 11) 12. ** _ 12 11 11 10 10 10 9 9 8 s 8 8 8 . In some cases part of the capacitors may have to be . At present it is not economical to apply capacitors on the secondary side of distribution transformers because of the much greater cost.
Two nonautomatic breakers are provided for controlling two steps. It is dcsirablc to supply the kvar required by the load as close to the load as possible to reduce feeder losses.i[C TTT (d) Fig. thus. A few years a. 13. The effect on all equipment back to and including the source generator should also be evaluated when the total capacitor kvar become appreciable relative to the total source circuit or system reactive kva. Several typical layouts for switching large capacitor banks are shown in Fig. capacitor units should be located at load centers or near the ends of feeders. Calculate the reduction in kw losses and the reduction in kvar losses in the feeder. 6. From the above brief summary on applying shunt capacitors to distribution systems. the released feeder capacity. then seven steps in capacitor kvar can be obtained with three circuit breakers. IV. Each case is different and requires a complete study in more detail than has been given in this general discussion. The disadvantage to this scheme is that during the switching process. Figure 13(e) is another scheme where one automatic circuit breaker supplies a number of nonautomatic breakers which control equal amounts of capacitor kvar. 13(a) is for one group of capacitors switched by one automatic circuit breaker. it is evident that no fixed rules can be stated regarding the location of capacitors nor can the tlcgree of importance of each of their elrects be stated. large changes of capacitor kvar are made to get from one kvar to another. but general practice is to provide switching so that a large bank is connected to the system as needed in several equal steps. the current increases. Summarize the tangible effects namely. This may involve capacitors installed at several locations on a given feeder. High Voltage Banks Supplying kilovars direct to highvoltage circuits is often desirable to meet certain system requirements even though : a greater portion of the system is benefited by placing the . installation of voltage regulators. the reduction in losses. the released capacity back to and including the source generator. 5. raising the distribution voltage. Figure 13(d) is a scheme in which three groups of capacitors properly proportioned provide seven equal steps. such as construction of a new feeder. 4. Also compare the cost of capacitors with other ways of doing an acceptable job. 15 or 25 kvar. and so on far all three switches giving the full capacity of the bank. switching large ca Shunt capacitors have been applied at substations and at the ends of primary feeders in banks ranging in size up to about 20 000 kvar. LARGE CAPACITOR BANKS TTTT (b) TTT (c) BREAKER . There are many combinations of the use of automatic breakers. depending upon the operating requirements and economics. disconnecting all capacitors before closing switch 3. Therefore. transformers were used to step down the voltage to the range of the capacitor unit ratings. 13. Each nonautomatic breaker has a highcapacity fuse that will clear a faulted capacitor group ‘ihead.he effect on voltage. switch 2 gives ‘/i. capacitor nearer the load and on lower voltages. Calculate the released feeder capacity in kw and kva for the capacitor kvsr installed. and generator capacity is also immediately available.of tripping the main supply klreaker. Fig. depending on the voltage change per step and the variation in load. If the voltage change during these changes can be tolerated. nonautomatic breakers and highcapacity fuses for capacitor banks that can be applied. A capacitor bank can he switched all in one step. This. or switch 3 must be closed putting all of the capacitors in service before switches 1 and 2 are opened. For many years. the third step being controlled by the main breaker. Switch 1 gives ‘/7 of the total. Ideally each load point. namely.242 Application of Capacitors to Power Systems Chapter 8 3.o the practice of connecting lowvoltage capacitors in series parallel groups and directly to the highvoltage line Was . The worst condition is changing from “/7 to J/T of the total kvar where switches 1 and 2 must be opened. 13Schematic X 2X 4XKVAR (e) arrangements for pacitor banks. etc. switches 1 and 2 give 3/~. Fig. Figure 13(c) shows three equal steps where one automatic breaker supplies the entire bank and trips for short circuits in any one of the three groups of capacitors. would have the exact amount of capacitor kvar to supply the necessary load kvar. t. however. The circuit breakers must be capable of handling short circuit currents. is not possible because standard size units rnllst be used. 13(b) shows four automatic breakers controlling four equal steps in a large capacitor bank. The usual large sizes are between 5 000 and 10 000 kvar. Overcompensation of feeder branch circuits with capncitors to obtain a higher voltage results in increased copper losses because at lower and lower leading power factors. Also it is more economical to use the large size units. and evaluate the economics to determine whether or not capacitors are justified. Three equal steps are quite common although more or less steps are used. etc. transmission. Released substation.
Second. when proper thought is given 10 sl~h factors as fusing.Chapter 8 Application of Capacitors to Power Systems 243 ostatAishcd because they are more economical than the use of highvoltage cnpacit. units in parallel. operation of capacitor units in series WLLS looked l~l)on :LSrisky due to the everpresent possibility of subiwhg capacitors to overvoltage as a result of changes in v(jlt:kge distribution either due to a change in impedance of wtions of the phase leg or due to grounds at some point ~1 the assembly. 14Connection for fused capacitor of a three phase bank. e.he remaining units in a group following the operation of one fuse.inue for a long time and such a rupture may endanger other units in the bank. the current through the fuse is limited.ic~~ the phase t. The number of units in parallel in a single group is important. number of units in parallel. These fuses need not be of high interrupting capacity because there are always two or more capacitor groups in series. (‘()llnection of one bushing of capacitor to the insulated I)l:ltform on \vhich it rests and means of detecting un‘l~~la~~~e conditions before the unbalance becomes escesSi\c. Q.r~)~~ps 2 Mvolt of’ outdoor capacitor units operating in . The fuses can be omitted but at a sacrifice in the protection to the capacitor bank.he voltage on the remaining units in a group should not.llrc 1I shows ho\v capacitor units are assembled for one piusc: of a hank.Rated voltllgc a~!ross group 1. co. I eX GROUP X Mx T T I i NEUTRAL units for one phase to Eqs. nunher of cupttcitor units per group NKumt)er of units out of one grunt>. t. or making inconvenient tests. when a unit becomes short circuited for any reason. Actud voltage across group 1. It is also desirable to avoid voltages in excess of 110 percent on t. This assumes that in the case of the minimum size bank not more than one fuse operation is To accomplish this. One of the first such installation consisted of six .o neutral voltage of a 24kv circuit. on l<& group of 2400volt units consisted of lo15kvar  T/ e2 e.ors or transformers ant1 lowroltage . Symbols apply XSunher MXormal of capacitor groups in series. Korrnl~l system voltage to neutral. periodic checks are permitted. and.l. the current1 through t. necessary. Fig. e. become excessive with t. and these 150 kvar groups were supported 1111 ins(IIators to take care of the line to ground voltage. 156000 kvar 34. Tf the number of parallel units is too small. I Ilit ially. however. Most of these risks are minimized or clrltircly eliminated.5 kv outdoor capacitor mounted fuses. It is also easy to make R check and determine if all units in the bank are operating properly. (3) to (11).he operation of one fuse in a gro~~p. and taking into account the arc energy required to rupture the capacitor case. bank with bus never be less than 10 times the normal capacitor current through the fuse. The amount of current that flows through a fuse when a unit is shorted is also affected by the number of series . Several things affect this.llits. ICtachcapncitor unit in a highvoltage bank d10dd be Provided with a fuse of the indicating type. it has been established that the current through the fuse when a unit becomes shorted should I%!.he fuse on a single unit blows when the unit becomes short circuited and the fuse is called upon to carry the total phase current. First the number should be sufficiently large to insure that t..hc fuse may be so low that it lvill not blow. An arc of 50 amperes inside a capacitor unit may rupture its case if allowed to cont.ScSr. After considering the size of fuses that must be used to avoid operation on snitching transients. or take too long in doing so. With individual fuses a faulty unit can be located without resorting to the risky procedure of searching for the source of noise or arcing.
The voltage across the remaining capacitors can be determined from Tables 6 and 7.4 11.6 11.5 110 110 110 110 110 than 110 than 110 than 110 than 110 than 110 than 110 than 110 than 110 1 IN SERIFS Number Groups Series 1 Minimum Units per Group 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 4 8 9 9 10 10 10 10 11 11 11 11 11 11 11 11 10.5 Less than Less than Less than Less than Less than Less than Less than Less than periodic checks of fuses are necessary to avoid abuse of good capacitors as result of a faulty one.9 11. all units have the same volt.8 Voltage on Remaining Units in Group Percent 109 109 109 109. Very often large banks contain many more than the minimum number of units in parallel.5 11.age rating.5 4 PERCENT UNITS REMOVED FROM GROUP Less Less Less Less Less Less Less Less Fig. Tables G and 7 show the recommended minimum number of fused capacitor units that should be used in parallel for a given number of groups in series in each phase leg. For all equations the system impedance up to the capacitor bank was neglected. Refer to Fig.8 110.8 11.7 11. 16 and 17 or calculated from the equations given below.8 11.0 109.0 12. In such cases PER CENT UNITS REMOVED FROM GROUP .0 10.8 11. The equations simplify quickly.6 11.8 11. 16Ungrounded wye connected shunt capacitor bank. All capacitor units are assumed to be the same voltage and kvar rating.4 119. Fig.2 10.5 11.244 Application of Capacitors to Power Systems Chapter 8 groups and whether or not the neutral of the capacitor bank is grounded.6 Voltage on Remaining Units in Group Percent 109 109 109.8 11. more than one fuse can operate and still not seriously raise the voltage across remaining units. the curves of Figs.2 11. 17Grounded wyeconnected shunt capacitor bank Curves give the percent overvoltage across the remammnl units in a group. 14 for identification of symbols in the following equations.2 11.9 10.5 11.7 11.0 10.0 11. .7 110 110 110 110 110 110 110 110 based on meeting the previously discussed requirements.1 11. TABLE ~GROUNDED WYE CURRENT AND VOLTAGE RELATIONSHIPS WITH SHORTING AND REMOVAL OF ONE UNIT IN ONE PHASE LEG Minimum Units per Group 1 Current During Fault Through Fuse Times Normal Line Fault 12 12 12 11. When this is the case.9 11. Number Groups Series 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 6 8 9 9 9 10 10 10 10 10 10 10 11 11 11 11. Curves give the percent overvoltage across the remaining units in a group. for ungrounded or groundedwye connections respectively TABLE ~UNGROUNDED WYE CAPACITOR CURRENT AND VOLTAGE RELATIONSHIPS WITH SHORTING AND REMOVAL OF ONE UNIT IN ONE PHASE LEG Current During Fault Through Fuse Times Normal 12.
with N units out of group 1 in one bank. . L . 5.and 25kvar capacitor units are protected usually by a fuse whether inst. . Potential transformers connected acrosseach phaseor each seriesgroup per phaseof ungrounded wye banks to trip the bank circuit breaker on phase or group voltage unbalance.) . GroundedNeutral Capacitor Bank Normal voltage el across group 1 is same as for ungrounded neutral bank as given in Eq. Individual capacitor fuses. 15. Capacitor group (or bank) fuses.’ %2+. Two Identical Capacitor Banks with Neutrals Solidly Tied Together and Ungrounded The normal voltage across any group of capacitors in an installation consisting of two similar groups with the neutrals tied solidly together and ungrounded is el as given by Eq. Protection of Large Banks of Shunt Capacitors The usual types of protection for large capacitor banks are : 1. the voltage e1across the remaining units is The current in the fuse of a completely shortcircuited capacitor unit in group 1 of one bank of two similar banks wit. 3. In addition the ground path for harmonic currents is not present for the ungrounded bank. Potential or current transformers connected between the neutrals of two or more wye ungrounded banks to detect unbalance in one bank and operate a relay to trip a single breaker through which all banks.+$ . With iV1 units out of group 1 in one bank the voltage across the remaining unitsingroup l. Large capacitor banks can be connected in wye ungrounded. Potential transformer placedbetween the neutral and ground of a wye ungrounded bank connected to a grounded system to operate a relay and trip the bank breaker on a shift in the neutral voltage. However.Chapter 3 Application of Capacitors to Power Systems 245 14. wye grounded or delta. 6. however. Thus. (3).is . units removed from group 1 the voltage el across the remaining units is The current through the fuse of a completely shortcircuited capacitor unit in group 1 in times normal operating current for a groundedneutral capacitor is $+&+. in the protective scheme. 4. Ungrounded Neutral Capacitor Bank elN = Normal voltage across group 1 is (9) \Vith IVYunits removed from group I. are supplied. For wye grounded or deltaconnected banks. with the wye ungrounded (8) I 2 x 16.alled in an outdoor or indoor bank for any type of capacitor connection. the wye ungrounded connection is preferable from a protection standpoint. in terms of the normal current through one capacitor is eNo= (5) The current through the fuse for a completely shortcircuited capacitor unit in group 1 in times normal operating current is 17. (3) for any bank. For the wye ungrounded system of connecting single capacitor units in parallel across phasetoneutral voltage the fault current through any fuse is limited by the capacitors in the two sound phases. Ir=(M.+g x 2. Overcurrent relays or trip coils to trip a bank circuit breaker. With N.h the neutrals solidly connected and ungrounded in terms of normal current in one capacitor unit is (10) lVith Ni units removed from group 1 the voltage shift of the neutral of the capacitor bank eNO is The current in the neutral connection between two similar banks of capacitors. Individual singlephase 15. the fault current can reach the full shortcircuit value from the system because the sound phases cannot limit the current. This schemecan be used for delta or wye groundedneutral banks that have two or more groups in series..
In addition. One of the potential transformers can also be used for an indicating lamp to show when the group is energized. Individual capacitor fuses are provided for each unit). 4. 18(h). 19Deltaconnected. A current or potential transformer connected betneen the neutral points of two equal parts of a grorlp of capacitors provides protection for unbalanced kvar per phase as shown in Fig. two potential transformers connected in open delta should be Ilsed on automatically controlled banks across the supply leads to the group to provide a fast discharge path when the capacitors are deenergized. however. of un providing a discharge path to dissipate quickly charges left on capacitor units when the supply is disconnected. usually applies to voltage classes of 2400 volts or less. POWERCIRCUIT TO BREAKER TRIP COILS P3 BREAKER SHUNT TRIP CAPACITOR USES / Fig. The capacitor hank should have a protective tlevicc to disconnect the bank from the system if individual units bccome defective thereby callsing a bad unbalance of capacitor kvar among the three phnscs. A deltaconnected bank of capacitors.he group is switched on again before the charge is dissipated high transient switching currents result. Two protective schemes for wye connected ungroundctl hanks for all voltage classes are sho\vn in Fig. fused capacitor at 2400 volts or less. 18. units usually used OWER CIRCUIT TRIP COILS SHUNT TRIP INDIVIDUAL CAPACITOR (b) Fig. it can be compulsory that potential transformers be applied for rapid dissipation of charges remaining on capacitor units. In special cases such as for indoor capacitor banks. When the number of units in a single installation . as shown in Fig. potential transformers should be applied across each phase leg to provide fast discharge when the group is deenergized. The scheme shown in Fig. 18. 19. Fig. If the bank is controlled automatically. Current flow between the two groups can also be used for protection. With two or more groups of capacitors in series per phase. (b) Residual voltage trip in event of unbalnrxe between the two 3phase groups of capacitors. Capacitor Fusing GeneralEach capacitor unit contains a large area of insulation and the probability of unit failures must be recognized even though the record is good. the shortcircuit current is limited by the capacitors in the unfaulted group.446 Applicatioti of Capacitors to Power Systems Chapter 8 cmnection smaller fuses and less material are needed for protecting the capacitors. This time of five minutes is considered to be too long for banks that are controlled automatically because when t. The individual capacitor units have a very high resistance provided across the terminals inside the case to discharge the capacitors in five minutes after being disconnected from the source. 18Two Q V protective schemes for large banks grounded wyeconnected capacitors. 18(a) is preferred because the potcntial transformers serve the dual purpose of protecting against unbalance(1 cxpacitor kvur per phase leg as iyell as I. (a) Residual voltage trip in event of unbal:tnce among the three phases due to failure of capacitor units.
The presence of other capacitors in parallel with and discharging into the shorted capacitor increases the tendency to melt the fault clear. and not as protection against high fault current within an individual capacitor unit. If the arc in a capacitor unit is allowed to persist until the case is ruptured. (c) To protect the other units in the bank from mechanical damage due to a unit case rupture. This may damage the currentcarrying connections of some adjacent units and cause simultaneous or later failure. it (d) To minimize the hazard to the operators and maintenance personnel. These are usually on circuits where the fault currents are moderate and group fusing has been Satisfactory. If the fault current in a capacitor is limited to a few hundred amperes. the pressure builds up slowly and many cycles of current flow may be endured before case rupture takes place. desirable to provide adequate protection against shortcircuited capacitor units.y of the thin aluminum foil that forms the electrode surfaces. The removal of a unit failure of insulation is of faulted units is important for Ihe protection of the remaining good units. A breaker should be considered primarily as a switching device and circuit protective device. This WC~may be used for fuse selectivity with remonnble safety. This has considerable bearing on the fusing problem because a fault within a capacitor can melt the foil rather easily and the fault tends to clear and sometimes restrike. Group FusingA shortcircuited capacitor is in reality a conducting path having timemelting character .hc C:LWis not likely to rupture. and its relation to case (‘urve AWhere fault currents arc cleared in a time to the left of this c:urve t. other units and parts in the bank may be damaged either mechanically or by consequent arcs. If these foils are allowed to carry heavy fault current. :\bout sixtyfive percent of existing capacitor kvar on utility systems are “pole type” and usually total about 180 kvar per installation. (b) To protect other capacitors in the bank against electrical damage due to current transients. does not remove the hazard associated with a bank where unit fault currents are high. Large capacit. however.chapter 8 Application of Capacitors to Power Systems 247 \ 8 2 CURRENT IN AMPERES RMS 60 CYCLES usually does not result in case rupture or damage to other units. unless supplemented by individual capacitor fuses. It may. 20Capacitor fault current rupture. and adequate interrupting rating protect the circuit. therefore. 20. bitt usually do not protect the capacitors against damage in case of a shortcircuited unit. and additional problems are crcntcd due to the close association of large numbers of capacitor units. When the current exceeds about 3000 amperes a rupture results in mechanical damage to adjacent units and often in shortcircuited bus connections. Protection Inherent in R&kersBreakers with overload protection. The function of this protection is: (a) To protect the circuit and capacitor bank so as to minimize the chance of an outage. When a capacitor unit becomes shorted. Use of breakers alone. . (~urvc I3Where f:hult currents are on for LI time to the right of this wrve the c:tse is likely to rupture with sufficient force to dnmitge other units. Cnder certain conditions the arc restrikes each half cycle.\rw ABFault currents in this areit may open case seams. is large the probability greater. IQ. be considered as backup protection in case the individual unit protection or other protection fails. The ability of a shortcircuited capacitor to pass currrnt is limited by the currentcarrying capacit. It is. the foil may act as its own fuse. or relay means to trip the breaker as a result of current or voltage unbalance.or banks are generally on circuits capable of producing high fault current. however. the greater the shortcircuit current the more violent the case rupture. thus allowing the atljaccnt capacitors to be repetitively charged and discharged. The current a capacitor unit can pass before case rupture is likely to occur is shown in Fig.
Individual capacitor fuses give indication of a blown fuse and give electrical as well as mechanical protection to parallel units. so that a faulted unit is disconnected promptly from the circuit for a number of reasons: (a) Their current rating is small and coordinated with the timecurrent characteristics of the capacitor. or failures in potheads or accessories. Likewise there is a masimum safe number of individually fused capacitors that can be placed in parallel amperes in rating (on a 100 per cent rating basis. Fuse A4160 volt 4160 volt 7200 and Fuse B2400 volt 2400 volt 4160 volt 2775 volt 4160 volt 7200 and Fuse C2775 volt Fuse D2400 volt 2400 volt delta connected 15 kvnr units. group fuse to the fault. Other units may also fail at a later date when the reasons are not immediately apparent. harmonic currents and the number of times in rapid succession a fuse must carry discharge current from a good capacitor unit to a defective unit. or are made up of series groups.05 0. If too many units are in parallel per fuses should be supplemented with individual fuses when group. which has a bearing on the maximum size of the group fuse. roof bushing flashovers. through one have sufficient interrupting capacity. rupture of the fuse with the possibility of damage to other even though the group fuse interrupting rating is ade100 quate for the expected fault current. (d) They remove a shortcircuited unit before the inside foil material is fused to the point where repetitive clearing creates high transient current in adjacent units. Unbalance in these cases is detected by voltage transformer and relay schemes so as to trip the breaker under abnormal conditions such as might occur if a unit becomes shortcircuited.025 0. Relaying should be provided where possible to detect ground faults even though their occurrence is very rare. . 7960 volt ungrounded wye 15 kvar units. Table 7 shows there is a minimum number of capacitor units required in parallel per group to give sufficient current for positive operation of an individual fuse on a failed (a) It is preferable not to apply group fuses greater than 55 unit. ungrounded wye 15 kvar units. so that the problem of high fault currents does not exist. at high current.) (b) The circuit is protected adequately by group fusesif they per group because if a unit fails all other parallel units discharge their stored energy. this fuse is not expected to clear for ground faults within the unit. ture and subsequent mechanical damage. delta connected 15 kvar units. 5C Large banks of capacitors have been installed with dcpendence placed solely on group fuses or breakers. ungrounded wye 15 kvar units. The objection to this arrangement is that it is difficult to identify a defective unit and there is the possibility of electrical damage to parallel units before the breaker deenergizes the bank. (c) To minimize the danger of mechanical damage. (e) They protect units against transient currents set up by parallel arcs in the bank such as bus flashovers. Individual capacitor fuses should be used.5 t= 0. Since only one fuse is used with each unit.1 0. delta connected 25 kvar units. 7960 v. particularly in large hanks. The size of the group fuse is also determined by the normal current of the bank and harmonic currents. thereby multiplying the damage considerably.248 Application of Capacitors to Power Systems Chapter 8 istics.5 z z Y m ~ z w z 0. Individual FuseThe individual fuse rating is dependent upon the normal current rating of the capacitor unit. Some of these large capacitor banks without individual fuses are wye connected with the neutral ungrounded. grounded wye 15 kvax units. In general. the following rules are recommended for group fusing: (f) They permit uninterrupted use of the capacitor bank since a faulty unit need not take the bank out of service. the current rating of the fuse is usually at least twice the current rating of the capacitor. or arcs in shortcircuited units in the bank.25 0. grounded wyeconnected 25 kvar units. ungrounded wye 25 kvar units. (b) They indicate the defective unit. the current is high enough to cause mechanical the unit fault current is expected to exceed3000 amperes. ungrounded wye connected 25 kvar units. ungrounded wye 25 kvar units. Individual fuses are used primarily to remove units following failure of the dielectric. the failure of one unit is likely to damage other units in the bank. Where fault currents are high. (c) They reduce to a minimum the chance of unit case rup 25 IC F 2.01 AVAILABLE CURRENT IN AMPERES Fig. 21Typical type BAG capacitor fuse characteristics for use with housed units where the fault current is less than 15 000 amperes from the system. delta connected 25 kvar units. To provide for the later requirement.
Where the fault current is less than 15 000 amperes type BAC fuses are used.02 second or less. the characteristics of which are shown in Fig. units. Lower voltage units with a fewer number in parallel per group with more groups in series may be a solution also. Where the fault current exceeds 15 000 amperes from the system. . Actually the discharge current from the good capacitor units operating in parallel Ivith the faulted unit supplies a considerable portion of the Fig. Individual Fuse Characteristics AVAILABLE CURRENT IN AMPERES (a) Housed Banks2400and 4160volt deltaconnected and 2400volt wyeconnected groundedneutral. (b) Housed Banks (Ungrounded Wye) IIoused banks for circuit voltages of 4800 volts and above are usually wye connected with the capacitor neutral ungrounded. 22Typical type CLC current limiting fuse characteristics for use where the fault current is high or in excess of 15 000 amperes from the system. Fuse E2400 2400 Fuse F2400 2400 volt volt volt volt delta connected 15 kvar units. grounded wye 15 kvnr units. AVAILABLE CURRENT IN AMPERES Fig. on large banks of capacitors. delta connected 25 kvar units. 22. individual capacitor current limiting fuses (CLC) are used. when the number of units in parallel per group exceeds two or three t. special consideration should be given to the application particularly with regard to arrangement.imes the minimum required number. 28 and reaches half value in &I$ 0. If it mere not for the current from the parallel units the system shortcircuit current would have to be limited to about 3000 amperes to prevent rupture of the capacitor case. units. units. The discharge current from the parallel capacitors is high in magnitude as shown in Fig. 23Typical UT fuse characteristics used on ungrounded wyeconnected outdoor capacitor banks. Where such limitations are involved. the characteristics of which are shown on Fig. units. 21. units. grounded wye 25 kvar units.Chapter S Application of Capacitors to Power Systems 249 energy to blow the fuse on the faulted unit. Therefore. Fuse G4160 7200 Fuse H2775 4160 7200 Fuse I2775 volt volt volt volt volt volt 15 15 15 25 25 25 kvar kvar kvar kvar kvar kvar units units. whether or not the source neutral is grounded. the bank can be divided into two or more parts \vhcre there are two or more groups in series. Housed banks usually contain indoortype individual unit fuses.
then the operating circuit is energized and the capacitor breaker closes. then the load kvar or total current must be used as the means for control. 1 breaker always closes first and trips last. 24Inrush current 8ooO KVAR (3 PHASE BANK) CAPACITOR from system tor bank. (c) Outdoor Struc:tur:tl Type Banks (Delta or Grounded WJT) Where the fault current is likely to be high as for a delta conncctctl or grounded wye. when energizing capaci voltages the maximum rms inrush current for different system shortcircuit currents available at the capacitor terminals. ((1) Outdoor Structural Type Banks (Ungrounded Wye) Outdoor structural type hanks for voltages of 4800 volts and above are usually wye connected with t. A similar process in reverse trips the capacitor breaker. For more than twostep control. or kvar. zz n 7000 1 4000 Fig. and if the masterrelay contacts stay closed for the time required for the timedelay relay contacts to make. Curves in Fig. then the capacitors can be switched on when the voltage is low or off when the voltage is high. This applies to delta conncctcd 2400 volt banks. For a twostep control the sequence is the same as for onestep control except that auxiliary contacts on the No.hcir use. ous and involved. outdoor bank current limiting individual fuses (CLC) are desirable. the control systems are similar. wye connected 2400 volt and delta conncctcd 4160 volt banks of capacitors. current. Where the need for capacitor kvar follows a fixed schedule. it is possible for a large instantaneous current from the system to flow. No. and thoroughly proved by many applications. etc. control switches. 22. This current can be calculated using the following formula: I=&[l+J$g Where ELC is linetoground capacitor bank.ime relay that initiates on or 08 at predetermined times. but their operation is accurate. Automatic Control for Capacitor Banks The intelligence required to switch banks of shunt capacitors automatically depends upon the reason for t. each additional breaker. the characteristi:s of which arc shown on Fig. that is. 1 breaker set up the circuits for the control of the second step. 24 show for several linetoline CURRENT FROM 19. whether or not the source neutral is grounded. other devices are required in the control scheme such as timedelay relays. 2. (12) . The characteristics of the fuses are the same as for similar indoor banks as shown on Fig. I ! operating voltage on the! . sets up the control circuits for the next operation whether it be to add or remove capacitor kvar. If the system voltage is regulated by other means and t. the circuit is set up to either trip No. The sequence of operation is the same in all cases. Inrush Current When the first step of a capacitor bank is energized.he neutral of the capacitjor ungrounded. 23. If they are used primarily to control voltage.age. 20. Kvar control is used where t. and a voltage relay supplies the control. If the No. 1 or to close No. This arrangement limits fault current and permits fuses of lower interrupting rating. 21. Whether the control is accomplished by volt. by means of auxiliary contacts. For onestep automatic control the master relay energizes the “closing” element of a timedelay relay. reliable.250 Application of Capacitors to Power Systems Chapter 8 This arrangement limits fault current and the type MC fuses are us~rl. the capacitors can be switched by a t. Current control is commonly used where the voltage is regulated by other means and the power factor is practically constant through wide variations in load. In addition to the master control relay. The control circuits become numer 5. The charncteristics for these fuses are given on Fig. It is always desirable to use t. 1 breaker is closed.he simplest type of control that will accomplish the desired result.he capacitors arc used for powerfactor correction.he load po\ver factor varies over a wide range as the load changes.
2 fat:t. it can be limited by the insertion XL(IOO/Xe) % NOTE 1. the masiFor deltaconnected banks the equivalent singlephasetomum inrush current is greater than 7250 amperes. grollnd capncit. To illustrate its use assume a 13%kv. I.he source. capacitor group to be C. 25Magnitude and Frequency of transient inrush current .hough the bank Where the inrush current when switching banks of was wyc connected.CAPACITORS. the peak tor step has a series inductive reactance of O. If one or more steps in the C8aLpacitor hank are already energized. then the masimum I1e& current that flows into the next. THESE CURYES APPLY FOR 60 CYCLES SYSTEMS.0225 which for slyitching in the third 2520 kvar step of capacitors allo\vs a masimllm peak :~rltl with the third step being energized then inrush current of about 69 times normal rms rated current c= l of each step or 69 X 105. XL is the inductive reactance in ohms per phase of t.elyby using the following equation: being energized and that portion of the bank already energized. The two capacitor steps already cncrgized in parallel are 50 ptrcent on 2520 kva.kvar. or 7250 amperes. ELM is rms all of its leads between the capacitor units and a common linetoneutral voltage applied to the capacitors. therefore.5 percent XL for use with Fig. IF THE SWITCH CLOSED CbPICITOR IS FULLY CHARGED AN0 TtiE WHEN SYSTEM VOLTAGE IS OPPOSITE TO CHART. Now assume that each capnciohm in If the st. The 1. threestep capacitor installation consisting of three 2520kvar banks.0076 ohm.nc:rg~zed tletcrmined largely by the momentary disis c. when energizing a bank in parallel with one existing .Chapter 8 Application of Capacitors to Power Systems 251 XC is the capacitive reactance in ohms of one phase to neutral of the capacitor bank. The frequency (14) of this current is about 6000 cycles.or is applied to account for some feed in from the system and also possible current unbalance due to unequal pole operation of the breaker. If the inductive &+$ 1 2 ‘3 reactance of the leads is less than 0.or kvnr must be used as t. ‘l’hc breaker controlling the last step in a bank of capaci101‘s the one that is subjected to maximum peak current is \v)lcLn step is energized.01 percent expressed on 2520 total cspncitancc per phase of the capacitors already ener. thus giving a current that is too high and. The above formula applies to deltaconnected capacitor I&. Ising this data the cwqizetl.S if X. For :L three step bank with two steps energized XL (X.r~ctoncutr:tl capacitor kvar. This value of L is difficult to determine accurately. 25. on the safe side. can be determined approsim:~t. NOTE 2. for use with Fig. the estimated L is usually a low value rather than a&&one.2) (4) (13) step to be energized are 150 percent. So the X.cp being energized is fully charged. The current values are for the first step of a bank. two being energized and the third step to be energized. 25. C is the point on the bus which is 0. 25 is 150 percent. bank. These two steps in series with the one 1penk= (1./100) equals 0. The peak currents if no charge this is on the step being energized. Two such units in parallel plus one in series gives giz(L(l combined with the capacitance of the step being 0.]~:~rgc from those capacitor units already in service. SWITCHED APPLIES TWICE THAT OWEN BY THIS FOR ENERGlZlNG DISCHARGED XL AND Xc ARE IN t ON THE CAPACITOR Fig.01.: is determined as the reactance of the equivalent l. due to inductance in the capacitor leads and bus structure. but. THE CHART . is the inductance between the step capacitors is excessive. The inrush current and frquency when a bank of capacitors is energized in parallel with one or more esisting banks is given in Fig. The percent capacitive reactance for each step on 2520 kvar is 100.OOSG inrllsh current can be about twice this value.
For a normal breaker opening. \\hen A opened. Voltages When Switching Off Capacitors The voltages of Table 8 are brought about by the fact that 100 percent voltage is left on L4phase. the first phase to open. Since the current goes out at normal current zero when deenergizing a bank of shunt capacitors... it can have a value of E = N 1. Similar analyses can be made for deltaconnected capacitors...35E: ~n J (15) B 2nd 87 187 C 2nd 87 187 100 37 137 where n is the order of the harmonic.UF k! a 7 6000 s kf Q 2 4000 Y ZTi a 2000 ADDED L OHMS AT 60 CYCLES o”. . a charge of 50percent voltage is left on phases B and C because the instantaneous voltage across these two phases is 50 percent. the peak current was measured. The voltage across the contacts of the circuit breaker is important because if the recovery rate or the magnitude is too great.. The kvar loading of a capacitor expressed as a fraction of its rating with harmonic voltages applied can be obtained as follows: KVA= E12+3E3?+5ES2+. System Harmonic Voltages Since the reactance of a capacitor varies inversely as the applied frequency relatively small harmonic voltages cause relatively large currentwave distortion.o RESISTANCEOHMS Fig. Reactance is much more effective than resistance... one phase is interrupted first even for a well adjusted breaker. A ___1st 150 250 22. If the transformers are operated near their rated voltages. . Maximum Voltage across Capacitor Leg Following Interruption. The sltbsequent voltage applied across B and C when R or C clears is 173 percent.. TABLE 8 Percent of Peak Voltage Phase of Wye Connected Bank Sequence of Opening.Ol 0. Harmonic frequencies usually encountered are the third and fifth... Capacitors do not generate harmonic voltages.. is still present and adds or subtracts from half of 173 percent giving a net of 37 percent or 137 percent.4mlF T 360 KVAR 192 .. But the 50percent charge left on these two phases. The neutral point of the capacitor bank remains at a potential of 50 percent above ground.... recovery voltage across circuit breakers.1 0’ 0 0. and the linetoline and linetoneutral voltages across the capacitors are important.000 Dcs2r: Application of Capacitors to Power Systems Chapter 8 v) 0000 z i 360 KVAR 192. The capacitor has lower reactances to higher .... The curves in Fig. The standard margins in capacitors are usually more than sufficient for the amounts of harmonic voltages present in most systems and.. where all voltages are expressed as a fraction of the rated voltage.. very little trouble is experienced. the voltage was then removed and Ivhen the switch was closed between the two groups.6 0. and 90 degrees later the other two phase currents are interrupted simultaneously at current zero by the clearing of either B or C breaker contact... the rms voltages resulting can be calculated. 21. at current zero. If only one harmonic is present.. which appears across the capacitance to ground. 26Test results indicating the effect of reactance resistance on limiting energizing inrush current. and of reactance or resistance into the circuit... Special consideration should be given to this problem in each case. 26 give the results of tests showing the effect of adding resistance or reactance in reducing the peak inrush current. The voltages of Table 8 expressed in percent of normal peak line to neutral voltage are obtained when the supply system is grounded solidly and does not suffer neutral displacement while switching a wyeconnected ungrounded capacitor bank.. The problem is more acute at voltages above 15 kv. In extreme cases it may be necessary to limit restriking on deenergizing by inserting in series or parallel with each phase of the capacitor circuit a suitable resistor just prior to the operation of the circuit breaker to deenergize the bank. Such restriking cause switching surges that may produce peak voltages of several times the normal peak voltage to ground. laboratory and field tests may be necessary. restriking occurs across the contacts. Capacitors are therefore built to permit combined harmonic and GOcycle kvar to equal not more than 135 percent (AIEE Standard) of the capacitor nameplate rating. The voltages to ground. Special treatment with respect to the oil flow in the breaker grid during interruption usually solves the problem.. The principal cause of harmonic currents in capacitors is the magnetizing requirements of system transformers. therefore... The very instant A phase opens. half of which is across B capacitor and half across C capacitor. hlasimum E to Ground after Opening. the harmonic voltages are limited to minimum values. A careful analysis of the problem should be made for each application.0 I . Dc voltage was used to charge one group of capacitors..4 0.2 0.252 FIXED L 15X IO6 HENRY 10.. Maximum E across Corresponding Breaker Pole... Careful adjustment of the breaker can make an otherwise unsatisfactory condition one which is acceptable.
By resonance or partial resonance with RMS MEASURED VOLTS IN ?&OF RATED VOLTAGE capacitors these voltages can be magnified.000 expected harmonic currents along with the ratedfrequency 5 current.Chapter 8 Application of Capacitors to Power Systems 253 frequencies and therefore allows proportionately larger clrrrrnts. fuses all over the system. the frequency of the current and the time constant of the circuit can be calculated for a given situation. the harmonic currents may become so high as to endanger the life of the capacitor.he total current the capacitor can carry is 161 percent. The peak current is high in magnitude but since the frequency is high and the time constant of the circuit low. m 5 I I An unbalanced fault on a system supplied by watera . Discharge Current When a capacitor is short circuited.000 0 f 3 i 60.001 ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ 100 110 105 115 wheel generators without damper windings may produce harmonic voltages. Because of the lower impedance of capacitors at higher frequencies. 27Permissible harmonic currents. Q Breakers applied with shunt capacitors must have ‘. either at its termi~1s or through a length of feeder.000 are 1OZpercent fundamental and 25percent fifth harui monic. Then 220. 1 sizes of capacitor banks. The peak current. Harmonics Circu$. sufficient continuous currentcarrying capacity to handle . For 135 percent kvar for different fundamental voltages without exceeding therinjure the capacitor. For the transformer magnetizing current the third hnrmanic components and their multiples are supplied usually 140 by circulation around the delta connected windings. . the current decreases rapidly. Considerable study has been given the effects of shunt t. . depending on how much total rms voltage. which standard capacitors can WL’I’V. .DlJCTOR :: $ 2 40*ooo 2 If the short circuit occurs at the instant the voltage on the capacitor is a maximum. exists at the same time. are usually sufficient so that for the amount of harmonic voltage present in 160 most systems no undue amount of trouble is esperienced. For example. The stored energy is dissipated in the resistance of the circuit which includes the capacitor and the feeder up to the short circuit. . it may result in blowing of capacitor mal limits. . the discharge current reaches half value in about 0. . 28Peak current supplied to a three phase fault through various lengths of bus from shunt capacitor banks. Figure 27 shows the amount of total rms current. ( . and the total rms voltage is 105 percent.000 IO00   500 MCM II BUS II CON. fun~lnmcntal and one harmonic. or cause excessive fuse blowing. . or less. This hazard is reduced by properly designed damper windings and system arrangement. suppose the fifth harmonic and the fundamental arc present. 24. flrndament~al and harmonic. While the duration of the fault might not be sufficiently long to Fig.02 second. it discharges its stored “ncrgy determined by Stored energy= l ClP (16) s 80. . This is made up of about 102percent rated amperes at capacitors on the inductive coordination of power systems fundamental frequency and about 125percent rated :lmperes at fifth harmonic. ) . ( . The corresponding voltages I20. For all practical 0’ 0 I 50 LENGTH I 100 OF BUSFEET I 150 200 I Fig. The higher harmonics are usually so small that they give no 120 appreciable trouble as long as the transformers are operated near their rated voltage. then the stored energy is a maximum. 16 which were mentioned previously. 100. Figure 28 shows the peak value ot current calculated for various lengths of bus consisting of singleconductor cables with an equivalent delta spacing of four feet. Breakers normally applied with capacitor banks are capable of handling these currents. . The standard margins built into capacitors. is THREEPHASE FAULT 23. or overheating of breakers and switches. and Coordination with Telephone The principal cause of harmonic voltages and currents in capacitors is the magnetizing requirements of transformers.
In the opendelt. The arresters are therefore available and in service at all times. are effective in relieving overloaded facilities until more permanent changes in the system can be made.a application the most effective use of the capacitors is to plan twice as much capacitive kvar across the phase lagging the open side of the delta as is placed across the open side. As the exciting current is decreased. 26. This is particularly important where large amounts of shunt capacitors are planned for systems where generators are already operating at high power factors. Surge Protection of Shunt Capacitors Fig. some of which are difficult to evaluate. Where the capacitor bank is switched. In addition. This problem also has a direct bearing on how much capacitor kvar can be permanently connected through minimumload periods with few generators in service and how much capacitor kvar can be installed with switching to provide needed kvar during maximum load periods and maximum generation.utc Bulletin. Therefore. 29. The static stability limit of a generator for a given set On circuits exposed to lightning it is recommended that lightning arresters be provided on all deltaconnected capacitors either housed or hanger type large or small banks. it has generally been practicable to improve conditions by rclntivcly simple measures applied to either the power or communication systems or both. may be affected by systemshunt capacitors and therefore the problem should always be taken into consideration. Likewise arresters are recommended for all wyeconnected capacitor banks where the neutral is ungrounded. A summary of the available measures is included in the article mcntionetl above and in Chap.al to the voltage on the airgap line of the generator corresponding to the excitation current.tct: on Development and Rcsenrch of the Edison Electric Tnstitut. Capacitors and System Stability Shunt capacitors reduce the static stability limits of generators (and systems) because they reduce the field currents used for a given km load and terminal voltage. if any. arresters are required to protect this other equipment whenever the capacitor bank is disconnected. Where capacitors have resulted in increased noise. 23 of this book. depending on the particular conditions in each case. These studies have bcon carried on by the *Joint Subcommit. as more shunt capacitors are added to a system. Actually many factors are involved in determining the static stability limits of generators.e and I{ell Telephone System. phase mobile capacitor units can be used to reduce the overload on opendelta banks of transformers occasioned by the failure of one transformer of a threephase deltaconnected bank. 29Portable capacitor bank. generators are operated consistently at power factors in the lead unless the generators are designed specifically for such service. the voltage on the generator airgap line rlecreascs. 1938 issue of the Edison Electric Instit. It.C 1 254 Application of Capacitors to Power Systems Chapter 8 and exposed t. if the operating power factor at full load is no greater than 95 percent 1~ Lgging. so that it affords added lightning protection to the capacitor bank itself and to transformers and other adjacent equipment. 27. Hydragenerators may also be affected by shunt capacitors. Any generator. However. Two single of conditions is directly proportion. In some cases generators are operated between 95 percent lagging and 100 percent power factor with satisfactory performance. The static stability limit is therefore proport’ional to generator exciting current. it is best practice to provide arresters on the capacitor side of the circllit breaker. Advance planning by the power and communication industries has reduced the number of trouhlcsomc situations to a small percentage of the capacitor installations. Generally on turbogenerators. for those surges where arresters are required there is also some hazard to the arrester because the capacitor discharges through the arrester when the arrester operates. Portable Capacitors Portable capacitor units such as shown in Fig. has been found that t)hc use of capacitors may bc either detrimental or beneficial from the inductive coordination standpoint.elephone circuits at noise freqllencies. A capacitor bank connected in mye with the neutral grounded has the ability of sloping off the front and reducing the crest of traveling waves. 25. the power factor of the generators increase and consequently the exciting current decreases. regardless of its primemover. Where t. Few. Thus there is some question as to whether or not arresters are needed. When the capacitors are connected to a bus with transformers and other circuits. X few power systems have this problem now and more will probably have the problem as future plans are made to get better overall system economy by taking advantage of the characteristics of shunt capacitors. the effect of shunt capacitors can be determined rather directly.he capacitor bank is the only load on a transformer winding the arresters can be omitted if the transformer is rcmovcd from service when . but usually thesc generators are so far removed electrically from capacitors that the generators are affected more by other factors such as the characteristics of transmission lines and the sending of power over relatively long distances. experience has shomn that the operation is safe. The results of their preliminary study of the problem were inch&xl in an article published m the August. The effect is noticed by an increase in generator power factor as more and more shunt capacitors are added.
wye and switched in five steps. The noload losses of aircooled synchronous condensers are about 60 percent of the fullload losses and fol hydrogencooled synchronous condensers about 10 pcrcent. the comparison should be on a basis of the synchronous condenser against the capacitor at full kvar plus a slllmt reactor of 50 percent kvar. has a further stabilizing cf’fect upon the other synchronous machines comprising the system.o 100 000 kva. For best surge protection of the . however.lcar before the capacitor unit is ruptured.tcrmine the proper adjustment of the breaker or to tlctcrminc what changes are necessary. 28. By moving . Figure 30 gives an idea of the relative cost of aircooled outdoor VITHOUT LOSS AAT EAT EVALUATED $150 $250 / PER KW PER KW IO 20 MVA 30 RATING 40 50 Fig.itor of shunt capacitors. wyeconncctcd capacitor banks should be ol. ‘J’llc gro~~ntledncutr:ll bank provides a path for the third or rcsitlunl harmonics. h standard synchronous condenser is capable of supplying kvars equal to its rating to the system as well as :It)sorbing them to an extent equal to 50 percent of its r:~t. coolers on synchronous condensers. at fractional loads the losses of the synchronous condenser are not in proportion to the output in kva. the arresters may be damaged. In this regard the synchronous condenser IUS greater stabilizing eRect upon system voltages and likewise tends to maintain synchronism between machines. For capacitors the losses arc about onethird of one percent of the kvn rating. The following points should IIC conaiclcretl in comparing these two types of equipment.Chapter 8 Application of Capacitors to Power Systems 255 all capacitors are disconnected. An instantaneous drop in terminal voltage. 4. For those applications requiring these chnracterislics.pplietl from a third winding of the transformer. thereby increasing the probability of communication interference. which is impractical with synchronous condensers. erection. Capacitors Versus Synchronous Condensers In large units synchronous condensers constitute a compct.which may occur. the neutral should he grounded and arresters ~. For short periods the synchronous condenser can steeply kvar in excess of its rating at normal voltage. The s(Jlllt. Capacitors can be installed easily.ightning arresters protecting highvoltage capacitor i. increases the kvar supplied to the system in the case of a synchronous condenser whereas a similar c411angc the case of capacitors decreases the kvar supplied in to the system. and $250 per kw. if a capacitor unit becomes sllortocl. With restriking across breaker c~ontacts. fa