LIBRARY ANNEX 2 .ftiS^rfV /r. CORNELL UNIVERSITY LIBRARY Given to the COLLEGE OF ENGINEERING Henry A. Ward. Cornell University Library TK2182.R981 Notes on the app»catlons of elert^^^^^^ 3 1924 003 959 412 Cornell University Library The original of tiiis book is in tine Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924003959412 NOTES ON THE APPLICATIONS ELECTRICAL MACHINERY HARRIS J. RYAN AND HENRY H. NORRIS Sibley College, Cornell University, Spring Term, 1897. TABLE OF CONTENTS. Direct Current Motors. The electric motor i The transformation of electrical into mechanical energy . 2 The practical principles of motor construction ... 3 Closed coil, constant current series motors 4 Constant potential series motors . 5 The constant potential shunt motor . 5 Theory of Motor Control. Control of open and closed coil constant current motors 6 Notation for the discussion of constant current motor control . 8 Constant potential series motor . . 8 Constant potential shunt motor . . . . . . 11 The series motor in practice . 12 The shunt constant potential motor in practice . . 12 The Storage Battery. The electric accumulator . ... 14 The lead accumulator . 14 Characteristic behavior of storage batteries . 14 Accumulator troubles . . 15 Applications of the storage battery 15 Line Efficiency and Copper Economy. Line eiBciency ... 19 Transmission of given power over given distance at fixed loss . 19 Transmission of given current a given distance ... 20 Kelvin's Law . 20 The distance a current can be transmitted . . 21 Direct Current Systems of Distribution. Constant current apparatus . . . . . . 22 Constant potential apparatus 22 Two-wire system . . .23 Expedients for saving copper . ... 23 Three-wire system . . . 23 Leonard three-wire system 24 Five-wire, Siemens-Halske system 24 Combined systems . 25 Calculation of Incandescent Wiring. Wiring calculations . . . .26 Case I, lamps on mains . . . 26 The tapered conductor . 27 Wire table . ... .28 Case II, lamps on branches .... 29 The Static Transformer as a Commercial Apparatus. The alternate current transformer . 31 Details of practical transformer . . 32 Table of Contents. iv E.M.F. developed in the secondary coil .... . . 34 Impedance of the transformer with the secondary on closed circuit 34 Effect of primary and secondary inductance . 36 Function of the primary exciting current . . 37 The Induction Motor. Description . 40 Induction motor behavior . 40 The Bradley motor . 43 Induction motor characteristics , . 44 The Synchronous Motor. Armature without self-induction . . 46 T-toothed armature . . . 46 The rotary transformer 50 Phase Transformation in Polyphase Practice. Phase changing transformers . 52 Vector diagrams . . . • ■ 53 Two phase — three phase transformation . • ■ • 55 The Incandescent Lamp. Manufacture . 57 Caudle-power 58 Life . 58 EflBciency .... ' 59 Commercial features . . 60 The "smashing point " . . 60 The Arc Lamp. The arc . . . ... 61 Consumption of carbon . . ,62 The feeding mechanism . 63 Systems of Distribution. Transmission lines . 65 Two and three phase transmission . . . 65 General Electric Co. 's plan . . . 65 The two phase sy.stem . . . 66 The three phase system . . .68 The monocyclic system . 68 Commercial Electrolysis. Copper electrolysis . . 69 Aluminum electrolysis ... 70 Electrolysis of pipes . 71 Transmission of Intelligence. The telegraph ... 72 The Morse alphabet 73 The duplex system . . ... 74 The telephone ... 75 Underwriters'' Rules. Abstracts from National Rules . ... -78 ELECTRICAL MACHINERY. DIRECT CURRENT MOTORS. I. The Electric Motor is a practical machine constructed and used for the transformation of electrical energy into mechanical energy. It is the necessary terminal apparatus where power is transmitted electrically. The principles of operation of the elec- tric motor are in all ca.ses those of the reversed dynamo. From this it does not follow that one will always obtain a .satisfactory electric generator by driving backwards an excellent motor. In fact, excellent street railway, elevator and some shop driving motors when driven as generators, will give very unsatisfactory results. The reason for this is that motors are generally required for totally different classes of service from that required of dyna- mos. A generator is required without sparking at the commu- tator to furnish a variable current at constant E.M.F. or at an E. M.F. that .slightly increases with the current. It is the duty of the electric motor, with reasonable freedom from sparking at the commutator, to furnish from a constant pressure circuit a variable mechanical torque at variable speed in some types and at constant .speed in other types. The student will remember that many of the characteristic features of design of the dynamo had to be adopted because being driven at constant speed it must maintain at variable current output a definite terminal pressure. In prac- tice the motor that is required to furnish a constant speed under varying mechanical load by taking current from a constant pres- sure circuit, is practically a reversed constant pressure generator. The far greater proportion of motors in practice are required to produce both variable torque and variable .speed. These are u.sed for driving street cars, and for operating hoists, elevators, etc. They are designed for a normal effort to rotation, speed and eiE- ciency, and with conductors of such size that the motor shall not heat unduly during prolonged normal operation and that the com- mutator performance .shall be satisfactory without shifting the brushes regardless of the direction of rotation of the armature. As a rule in this class of motors less attention is paid to the re- Electrical Machinery. 2 ductiou of armature reaction. Sparking at tlie brushes is limited as much as possible by the use of symmetrical windings, a maxi- mum number of commutator bars producing a commutator of ample proportions, and by the use of carbon brushes. The brushes may thus be set on the normal diameter of commutation and do not have to be changed with the load or direction of arma- ture rotation. These motors carry more copper in proportion to iron than do the generators, and as a rule they are lighter and more compact than generators. 2. The Transformation of ElecTtrical into Mechanical Energy. Any portion of an electric current through which cur- rent is established by an electric generator when placed in a mag- netic field, is acted upon by a mechanical force that tends to move the circuit or conductor across or at right angles to the field. The relation of the directions of field induction, current and motion of the conductor is given in fig. i. The direction of motion of the conductor will be reversed when either the field or the current are reversed ; it will remain the same if both field and current are reversed. The actual value of the mechanical force exerted by the field on the conductor is y= Bli sin 6, where f is the force in dynes, B the induction in lines per sq. cm., /the length of the circuit located within the field, and i the current .strength in c.g.s. units, while 6 is the angle between the directions of the motion and of the field. The corresponding force in grammes per ain- pere is/' = Etc sin Q -h (980 X 10) ; for grammes per ampere per centimeter length, /" = 5c sin Q -h- (980 X 10) ; for pounds per ampere,/'" = Bel sin 6^ (980 X 454 X 10) ; and for pounds per ampere per foot, / = 30.5 Be sin -j- (980 X 454 X 10). In all ordinary cases met with in practice, 6 = 90", sin 6 = i, so that this term is generally unity. Whenever any portion of an electric circuit carrying current is located in a magnetic field and is made to move, an E. M.F. is gen- erated, E= {Blv sin 6) -^ lol When the direction of motion is the same that the field would naturally give and as shown in fig. I, the E.M.F. generated is opposed to the current and is opposite to the E. M. F. of the generator. The field under the.se circum- stances assists the conductor to move by an amount of mechanical power equal to the amount of electrical energy removed from the electrical circuit. This amount of energy is equal to the product of the E.M.F. generated into the current in the circuit. Should the motion of the conductor be opposite to that which the field Electrical Machinery. ' 3 would naturally produce, the E.M.F. generated by the motion of the conductor through the field will have the same direction as the current and the generator E.M.F. The field when the conductor moves thus assists the generator to establish the current. The amount of mechanical power absorbed by the field and conductor will be equal to the increase of electrical power in the circuit. The mechanical power, /*, produced or absorbed by the circuit in the magnetic field equals in dyne-centimeters per .second, P ^ i Blv sin e ; Watts = W =- P ^ id = C.E. ; <: == 10 z ; E = Blv sin 6 -=- 10* ; W= i Blv sin 6 -=- 10' If the original of the prime mover is considered positive, the dynamo power will be negative and the motor power will be positive. It follows, therefore, that in any case where the field and current are each positive sin 6 will be positive for the motor and negative for the dynamo generator of E.M.F. 3. The Practical Principles of Motor Construction. The mechanical and electrical principles are applied for motors in much the same manner as for dynamo construction. The same variety of armature windings, mechanical details of armature core construction, methods of mounting and insulating armature and field conductors, commutator construction and of field forms that are applied for generators are applied also for motors. The pro- portions vary from those of dynamos according to the particular practice as explained in section i. There are, therefore, motors with open and closed coil armatures, each with bi-polar or mul- tipolar series or shunt fields. For operating these there are two general methods of electric power .supply. By the one method the current in the circuit is kept constant as in arc lighting, the pressure varying with the power taken, and the motors with series fields are, therefore, put into the circuit in series, the cur- rent passing first through one and then another, etc. , as in arc lighting. This is known as the constant current system. By the other method the parallel system of current distribution is adopted. A practically uniform pre.ssure on the supply leads is maintained by the generator. From these leads the motors take current in parallel as in incandescent lighting, and the generator supplies current in proportion to the electric power demanded by the motors. Series and shunt motors alike are operated most exten- sively in practice by this system. This is known as the constant pressure system. Open coil series motors have been tried in practice to a small Electrical Machinery. 4 extent. They are inherently less adaptable to the demands of practice than the closed coil machines. The.se motors are adapted to work at constant current with the electrical pressure at the brushes varying as the lead varies. The requisite regulation is effected by varying the angle of lead of the brushes by hand or automatically by means of a centrifugal governor. There are so few of these motors in actual use at the present time that space cannot be given for their further discussion. Closed coil constant current series motors are used occasionally to obtain power from arc lighting circuits. They give better practi- cal results than the open coil type. The torque is varied to .suit the demands of varying loads by either of two methods. One is by shifting the brushes, and this is the method more generally applied. The other is by varying the field turns. When the former method is used the motor is so designed that the magnetic difference of potential produced by the field between the pole faces is in excess of the armature m.m.f. by the amount which will establish an induction sufficient to commutate the armature cur- rent. Since the current is con.stant at all loads this relation is easily obtained in practice. As it holds for all diameters of com- mutation, it follows that sparklcss commutation will be obtained with the brushes in any position. The torque is a maximum when the brushes are in the normal diameter of commutation and it is zero when at right angles to this diameter. When variation of speed at various torques is desired the brushes are shifted by hand. If the speed is to be maintained constant at all loads the brushes are shifted automatically by means of a centrifugal gov- ernor. When the method of varying the field turns is u,sed for regulation, the field will vary from practically zero when the mo- tor is running light to a maximum when fully loaded. The brushes are kept on the normal diameter at all times. To insure that commutation will occur under all variations of load without serious sparking, carbon brushes and a commutator ample in num- ber of bars and dimensions must be employed. Th-; regulation is done by hand or automatically as before, according as the speed is to be varied or maintained constant. These motors are em- ployed practically where an arc light circuit furnishes the only available source of electrical energy, and then in sizes generally smaller than twenty-five horse-power. At the present day the con- .stant current system is nowhere adopted as a standard method of power transmis.sion. One serious objection to the system is due Electrical Machinery. 5 to the high pressures that must be employed when the current is maintained constant at a small value. The high pressures that must be employed by the motor produce danger to life, increase the cost of attendance and repairs, and introduce added fire risk owing to the serious arcing that must occur when the electric circuit about the motor becomes impaired at any point or becomes partially opened. Constant potential series motors are applied at the present time more extensively than any other type. This type is used for traction purposes, operating hoists, elevators, and generally for the supply of all sorts of services where the power demanded is intermittant and at all times irregular or unsteady. A series motor when operated from a constant potential circuit will natur- ally produce a different speed for each value of torque that it is required to produce. The speed is high when the torque is small, and low when the torque is great. Independently of this the speed at a given torque is increased when field turns are cut out, while on the insertion of resistance into the circuit of the motor the speed for a given torque is lessened. Thus any desired rela- tion of torque and .speed may be obtained within the limits of safe operation of the motor. A point of greatest practical importance lies in the fact that the armature and field coil always receive the .same current. The armature reaction always tends to weaken the field. The evil effects of this are overcome in the series motor where the field excitation increases with the armature current thus fitting the motor for sudden overloads and irregular work. The armature of a motor connected to a supply line acts like a short circuit the moment that the field es!?itation is interrupted. The series motor construction avoids the possibility of the appli- cation of the line pressure without due field excitation. The constant potential shunt motor next to the constant poten- tial series motor is used most extensively in practice. It con- struction is the same as the shunt or compound dynamo. It is used quite largely for "stationary" power supply ; that is, for driving shops or factories and for similar classes of service where constant speed is desired. Its behavior is quite exactly that of the shunt or compound constant potential dynamo. Electrical Machinery. 6 4. Theory of Motor Control. Owing to the simplicity of control that is adopted for constant current motors and the fact that thej' are used to a very small extent in practice, the theory of their control will be discussed merely qualitatively, while the methods of control that are used in practice for the operation of the series and shunt constant potential motors will be analyzed more accurately. Those methods present a greater variety of con- trol and the motors are used almost universally. Control of Open and Closed Coil Constant Cwrejit Motors. In these motors all regulation is effected by changes in the armature torque. Thus a given amount of power is obtained at a given speed in this type of motor by adjusting either the position of the brushes or the magnitude of the field induction so as to obtain the corresponding torque. The torque of a given motor depends on the product of the field induction and the armature current. In any given motor the induction through the armature established by the field is limited to that amount which causes saturation, while the current is at all times invariable and fixed at a definite value. In this type then, the torque due to the maximum field induction, and constant line current is the highest possible me- chanical effort to rotation obtainable. The demands of much practice are .such that motors, for very short intervals of time, are required to produce torques that are many times in exce.ss of the normal full load running torque. A duty of this sort is impos- .sible with the constant current motor. To obtain such a duty would nece.ssitate the construction of a motor field many times larger than that which is needed for average or normal operation. The constant potential motor has the same field induction limit ; however, it may draw from the line, for short intervals of time, any amount of current and thus produce the necessary torque to overcome a momentarily large increase in the load. This is an- other important rea.son why the practical application of the con- stant current motor is so very limited. The con.stant current motor must be constructed so that the maximum torque that it can produce will exceed that which will be demanded of it in practice. As explained in section 3, the torque is varied in prac- tice in each instance by one of two methods. The first method is by shifting the bruishes. This is the more commoaly u.sed method. As .stated, all motors of the constant current type are of the series class. The field coils in .series with the armature carry the main current. The number of turns in Electrical Machinery. j these coils is such that the maximum obtainable induction through the armature is produced at all times. Then when the brushes are on the normal diameter of commutation, the motor produces its maximum torque measured in pounds at one foot radius. The product of this torque into i-k times the safe speed of the motor, divided by 33,000 will be the normal output of the motor in hor.se power. The corresponding electrical horse power input is CE-^ 746, where C is the line current in amperes and E is the sum of the motor E.M.F., E^, and the fall of potential through the motor due to its resistance, R, and the line current, C, and which, therefore, equals CR. Hence .£= ^,^ + C^. The effi- ciency is the ratio of the output to the input horse-powers. The interval losses or wastes are of three classes just as in generators : CR, hysteresis and eddies, and friction. The input minus the output always equals the sum of these losses. For lower power outputs the torque must be lessened. This is done by shifting the brushes. The effect is to change the sign of the torque pro- duced by a portion of the armature conductors, thus opposing the balance of the conductors and lessening the total torque. The number of opposing conductors will increase with the angle of displacement of the brushes until points are reached midway be- tween those at which normal commutation occurs. At this point the number of opposing conductors equals the balance and the torque is, therefore, zero. Further .shifting of the brushes will reverse the direction of rotation of the motor, attaining a maxi- mum negative torque when the brushes again occupy the normal points of commutation in their reversed position. In lessening the torque by shifting the bru.shes the motor E.M.F., generally called by text writers the counter E. M.F. , is lessened proportion- ally. The same conductors that have their torque reversed have the E. M.F.s that are developed by their motions, also reversed. This occurs because these conductors are made to move under opposite poles. For this reason as the torque is diminished, the field and .speed remaining constant, the E.M.F. with which the motor opposes the current furnished by the generator diminishes a corresponding amount. It is seen that the.se conditions of operation involve no change in the adjustment of the motor when the load demands the same torque at a higher or lower speed. The motor will operate from zero to the maximum speed at which it may be safely operated without any change of adjustment at any torque for which it is set, should the load demand this uniform torque over such a range of speed. Electrical Machinery. , 8 The second method of control is by means of the change of field excitation. This method is used but very little in practice. The change in field excitation is produced generally by cutting sections of the field winding out or in. Sometimes it is done by shunting the current past the field coil by increasing or diminish- ing the amount of resistance put in parallel with the field winding. Thus by changing the field excitation from zero to a maxinuim the torque is made to change through a corresponding amount. Aside from this difference in the method of torque variation the behavior of this motor is the same as that desCi'ibed in connection with the method of torque variation by shifting the brushes. Notation for the Discussion of Constant Potential Motor Con- trol. E is the E.M.F. impressed at the termiuals of the motor. E^ is the motor E.M.F. or in volts, that pressure which is de- veloped in opposition to the generator pressure by the motion through the field induction of the motor armature conductors. C is the current in amperes through the motor armature. c is the current in amperes through the motor shunt field. R^ is the resistance in ohms of the motor armature. R^ is the resistance in ohms of the motor series field coil. R^ is the resistance in ohms of the motor shunt field coil. R^ is the resistance in ohms external to the shunt field. R is the resistance in ohms external to a .series motor. R is also the resistance in ohms external to the armature cir- cuit of a shunt motor. Q is the torque or effort to mechanical rotation measured in lbs. at one foot radius. n is the motor speed in r.p.m. A constant potential series motor receiving electrical power from service mains in practice is illustrated by the diagram in Fig. 2. From the point c at the .service mains the current is taken through the safety devices and switch and thence to the rheostat controller at H. From /f the current passes through the controller and field coil to the switch R.S. and from there through the motor armature and finally it passes back to the other side of the service main at c^ by way of the main line switch D.P. and the safety devices. As the connections of the controller indicate, its function is first on starting to limit to the amount of current nece-ssary for the de- sired motor torque by inserting all or a portion of the resistance between the contact points from O to F. By putting the lever H O in contact with points on the controller beyond F, turns in Electrical Machinery . 9 the field coil F. C. may be cut out. The following relations be- tween current, torque and speed exist for this type of motor when a constant supply of pressure is maintained on the service connections C C\ The fundamental characteristics of the motor are due to design and construction. These are the ampere-torque curve given as the unbroken line in Fig. 3 and the ampere-speed curve which is tlie unbroken line drawn in Fig. 4. The ampere- torque characteristic of Fig. 3 is predetermined for a given mo- tor only through the processes applied in designing ; and for a fin- ished motor is determined by trial with the ammeter and prony brake. Whether the amature is or is not permitted to rotate the result will be practically the .same. This ampere-torque character- istic is always determined by the u.se of the full number of field coil turns. The corresponding characteristics with portions of the field coil cut out are readily determined when the fundamen- tal has once been obtained. The broken characteristic in Fig. 3 was determined in the following manner for a case in which one third of the field coils were cut out at the controller by manipu- lating the contact lever // O : With all of the field coils in cir- cuit 30 amperes produced a torque of 122 lbs. Two thirds of this field coil when passing 30 amperes will produce the same arma- ture field induction as two thirds of 30, or 20 amperes, through all of the field coil turns. Now 20 amperes through the whole coil produced a torque of 57 lbs. ; 30 amperes in the presence of the same field induction will produce a torque = 57 X 30 -^ 20 = 86 lbs. thus locating, as indicated in the diagram, a point on the new ampere-torque characteristic. Other points were thus determined, enabling one to draw in the broken curve as the ampere-torque characteristic for one third of the field coil cut out. Other char- acteristics for more or less field coil turns cut out would be deter- mined in the same manner. This process of determining secon- dary characteristics of this class from the primary or fundamental is given by the following reaction which always exists : 8" = ^ where n is the fraction of field coil turns included in the motor circuit, Q' is the torque due to the current n C on the fundamental characteri.stic, and Q" is the desired torque produced by the cur- rent C and the changed number of field coil turns. The ampere speed characteristic is deduced from the ampere torque characteristic, as follows : Electrical Machinery. lo Equating the mechanical and electrical powers jj p ^^x^^ C^ 33,000 746 ■ E^'^E— C(^R -\-R^-k- J?f) . For the fundamental ampere-speed characteristic, R^ 0. Substituting and transposing, ^_ 33.ooo[C^-C'(^.+ ig;)] 746X27r(2 For example, the resistance of the motor whose characteristic is given in Fig. 3, is R^+ R,= 1 ohm. What speed will the motor make on a 500 volt circuit at a current of 30 amperes and a torque of 123 lbs.? 33,000 [30x500 — 30'' X il o n = "^^ !=^^ -i i =1 = 809 r.m.p. 746X27rXl23 In the same manner speeds corresponding to other currents are determined by obtaining a sufficient number of points on the curve necessary for its complete location. Thus the unbroken curve in in the diagram of Fig. 4 was determined. The lower broken curve in the same diagram was determined in the same manner for the case where R='] ohms, or a total re,sistance oi R + R^+ Rf= 8. ohms in the motor circuit. This shows the control effect by inserting additional resistance to be the lowering of the motor speed. Thus by variation of R any desired change of speed for values below those of the fundamental ampere speed characteristic- may be obtained. As R is increased, the loss of the electrical energy, CR, is increased and the efficiency correspondingly lowered. In practice the use of resistance for regulating purposes is only momentary, chiefly in starting the motors. Where greater speed at given current with its corresponding torque is required , than that which the fundamental characteristic will furnish, it is obtained by cutting out a portion of the field coil turns thus lowering the armature field induction, motor E.M.F. and torque at a given current and speed. The upper broken curve in Fig. 4 is drawn through points determined from the broken ampere torque curve of Fig. 3. In this case one-third of the field coil turns were cut out. Thus it is seen that any value of speed may be obtained within the maximum speed limit that the motor will endure at any value of the current or torque by adjustments of the resistance and field coil turns in the motor circuit. Under all conditions of operation the efficiency will be the ratio 2TT Qn _^ C E the output is to the input. Efficiency = 33,000 746 Electrical Machinery. 1 1 The Constant Potential Shunt Motor for which a complete dia- gram is given in Fig. 5 has a behavior that is so nearly like that of the reversed shunt constant potential dynamo that a study of that dynamo leads to an understanding of this motor. The m.ain field exciting m.m.f. is derived from a field coil of many turns of fine wire shunted across the terminals of the motor and taking a small current, the value of which may be adjusted at the rheostat i?'. Current is admitted through the safety devices and switch, then first to the shunt field coil. After the field is e,stablished the armature current is established first through resistance by means of the contact lever of the controller at V. Armature current is thus in sufficient amount admitted to start the motor light or under load as the case may be. As the motor conies to speed the motor E.M.F., .fi'^ limits the current and the resistance R may gradually be all cut out. The small series coil S.F. C. is not always applied. Many shunt motors perform their duty without the aid of this series coil. Its purpose is to weaken or strengthen the field set up by the shunt coil as the load increases, for the purpose of maintaining either constant or lesisening speed with such in- crease of load. The range over which the field induction of a shunt motor is applied in practice is .such as not to exceed the point of .saturation. The induction through the armature may be taken as being proportional to the field exciting ampere turns. The relation between the speed and load currents of a shunt motor operated from a constant potential circxiit is as follows : ~^^^=^„C^746. E^ = E-C{R^ + R^ 33000 K = E -=r- normal motor speed. E-= Kn. Ill From this it is seen that the value of E^ is fixed for each value of armature current, and that if n is to be maintained constant K must vary with E^. As E^ diminishes with the load current by an amount, C {R ^-\- R,), it follows that for constant .speed A" must be diminished with the load in the same proportion as E^. Kn = E^= Bvl-T- ID*, A' can only be diminished with the load by causing the load current in passing through an opposing series field coil to diminish the armature induction, B, in the ratio ^,„-^ E. Itwas found in connection with the shunt constant potential dynamo that the armature reaction caused the induction B to diminish with the load current. This same result occurs in a motor and Electrical Machinery. 1 2 the amount of diminution of the armature induction from this cause is frequently sufficient to cause the speed {n) to remain constant at all loads without the use of series turns whose m.m.f. opposes the shunt ampere turns. Occasionally in practice series field turns are applied so as to assist the shunt turns. This is done so as to increase 5 as C in- creases, thus causing the speed n to drop off proportionally. This behavior is often desirable where the motor must handle very sudden changes of load occurring through wide ranges. 5. The Series Motor in Practice. As above stated, this type of motor is used for a variety of irregular kinds of work. It is used to drive street cars, hoists, elevators, and for many special duties. When properly installed the lead wires from the service connections before entering the building or structure where the motor is located, pass through impedance coils LL' , Fig. 3. On both sides of these impedance coils, lightning arre.sters are connected and furnished with a good ground connection G. Only the inside set of arresters is shown in Fig. 3. The connecting wires are then taken into the building and through safety fuse cut-outs for the purpo.se of cutting off the service connection .should the cur- rent for any reason become excessive. The conductors continue through a double pole main line switch to be used for breaking the motor connections completely from the supply lines ; then one side of the circuit passes through a double pole armature re- versing switch and through the armature, then through the field and finally out through the controller to the other side of the main switch. All safety or manipulating appliances are mounted on .slate, porcelain or marble insulating and fire proof bases. Great care is taken in manipulating the circuits of a motor so as not to form short circuits or destructive arcs that might cause serious fire risks. Fig. 7 gives a diagram of a series motor in which a magnet in series with the motor operates a rheostat for automatic control of the current in elevator work and, therefore, to control the me- chanical effort with which the elevator is driven when coming to speed from a full stop. 6. The Shunt Constant Potential Motor in Practice. In instal- ling and operating a shunt motor it is necessary to apply the .same character of .safety devices and manipulating apparatus as were described above for the series motor. The controller omits the .series field commutating points and in connection with it, the motor Electrical Machinery. 13 is provided with an additional safety device. The purpose of this device is to open the armature circuit should the electric service be momentarily discontinued, as is often the case in practice. The .stoppage of the supply pressure causes the magnet .^in Fig. 5, which is connected in series with the field circuit, to relea.se the contact lever which, actuated by a spring, is promptly returned to the position occupied prior to starting. Without this device the armature, standing still, would be connected acro.ss the supply mains when the pressure would again be applied. The supply line would at once be short circuited through the armature as it would be developing no motor E. M. F. , and would be in circuit with no resistance, and there would be nothing but the safety fu3es, therefore, to limit the current from attaining dangerous proportions. Fig. 6 gives a diagram of a shunt motor circuit where the mo- tor is applied to elevator driving. Here the magnet M, operating the controlling rheostat is wound with fine wire and connected across the brushes of the motor. This magnet receives current, therefore, in proportion to the motor E. M. F. , and cuts out re- sistance in the armature in the same proportion, so that when the motor is operating at full speed the motor E. M. F. alone acts in limiting the armature current to that amount which is neces- sary to carry the applied load. EIvECTRICAIv MACHINERY. Plate i. A ^ s s F^g^ L X CURMHT *2 CONDUCTOR^ UI MOVES TOWWBD VOU. 5 c Fii. I. v/» • s Electrical Machinery. 14 THE STORAGE BATTERY. 7. The electric acaivtulator is composed of two sets of plates immersed in a liquid which acts upon one of them more than on the other. In this respect it is exactly like the primary cell, from which it differs only in the fact that the plates are " formed " by the electrolytic action of the current used in charging it. An electro-motive force is always produced when one plate of a vol- taic cell is acted upon more than the other, and in the primary cell, the plates are of different metals. In the secondary cell or accumulator, the plates are usually of the same metal, but in different chemical states, so that the action of the electrolyte is the same as in the primary cell. For example, in the lead accumulator, the positive plate is lead per-oxid and the negative plate spongy lead when the cell is fully " charged." 8. The lead accumulator is the one which is mainly used at the present time. It consists of two sets of "grids" (see fig. i) made of antimonious lead, the interstices being originally filled with paste of litharge (red lead), which is afterward changed by electrolysis to per-oxid in the positive plate and spongy lead in the negative. The average electromotive force of the cell is two volts, and its capacity depends on the amount of ' ' active material ' ' present. The accumulator does not store energy in the form of an electrical charge, but the energy is used in changing the chemical composition of the plates and is re.stored to the circuit by the chemical afl&nity between the elements of the cell. When a number of accumulators are connected together, either in series or parallel, or both, they are called a " storage battery." 9. Characteristic Behavior of Storage Batteries. When in normal working order the average discharge voltage of an ac- cumulator is two volts. When full it is higher, and when nearly empty it is as low as 1.8 volts. After this point is reached the falling off is very rapid and the cell is apt to be injured by the sulfating of the plates. For this reason the electromotive force of an accumulator is never allowed to fall below 1.8 volts. A much higher electromotive force is required to charge a cell than can be taken from it, for the reason that the cell has some resist- ance. This causes a "drop" in the electromotive force both in Electrical Machinery. 15 charging and discharging. At least 2.3 vohs are required to charge a lead cell and often ranch more, the exact amount depend- ing on the value of the charging current. Fig. 3 shows curves of charge and discharge electromotive forces for a lead cell. From the fact that when the cell is discharged, both plates are sulfated, it follows that in this .state the electrolyte (sulfuric acid, sp. gr. 1.2) is weaker than when the cell is charged. When a cell is fully charged the energy of the current is used in decomposing the electrolyte, which is given off in the form of gases. Accumulator troubles. A cell is often short-circuited by paste plugs falling between the plates. This is frequently accompanied b3' warping or "buckling" of the plates, due to the unequal electrolytic action on the two sides. To remedy tlie.se difficulties the cell should be kept clean, and if the plates buckle they should be hammered flat between two boards. An important disease of lead plates is known as "sulfating." It is the formation of a hard layer of white sulfate on the surface of the plates, which in- sulates them from the electrolyte, making a high resistance and reducing the capacity. Sulfating is usually due to allowing the electromotive force of the cell to fall below i.S volts, which may occur if the cell be left uncharged for any length of time, for the leakage current soon empties the cell. 10. Applications of the Storage Battery. The commercial uses to which the accumulator may be put, are: (ij Traction purposes ; (2) Isolated work of small extent ; (3) In connection with central stations. Storage Battery Traction. Up to the present time the accumu- lator has not proved itself well adapted to traction purposes for the following reasons: (i) Its lack of durability under the shocks of ordinary traffic ; (2) Its great weight compared with the work to be done; (3) Its cost ; (4) Inconvenience in hand- ling ; (5) The care and expen.se of its maiutainance. Of these troubles, all but the first one have been to some exttnt overcome, by proper engineering, but up to the pre'sent time a cell has not appeared which will endure the shocks of ordinary traific for more than a few months. All the cells on the market have been re- peatedly experimented with for traction purpo.ses, but all have been abandoned after a fair trial. These failures do not, how- ever, prove the undesirability of the storage battery for traction purpo,ses, for the independent unit is the ideal system of traction. The custom, in the cases in which the experiment has been tried Electrical Machinery. i6 is to place from 96 to 150 cells under the seats of the cars, or in a special receptacle in the truck, and to change these few every few trips, the extra cells in the meantime being charged and ready for the car on its return. By means of suitable machinery, the time required for changing the cells has been reduced to less than one minute, and the operation is, to a large extent, automatic. Isolated work. There are many cases in which a small amount of power is desired at a distance from an electric circuit, and in cases of this description the accumulator has a large field. For launches, train lighting, small motors and .such work the accumu- lator, in many instances, proves to be a convenient and econom- ical secondary source of power. Application to central stations. The great field for the accumu- lator at the present time is in connection with central stations. The importance of their use has long been known in Europe and at present 80 per cent, of the stations in Germany and Au.stria are supplied with one or more batteries of accumulators. The im- portant .stations in this country are also adopting the accumulator as an auxiliary device. The uses to which the battery may be put in a central station may be summarized as follows : ( i ) To straighten out the load curve; (2) to keep up the voltage on heavy load ; (3) to carry the entire load at times when it is mini- mum ; (4) to unify the engine load ; (5) to act as a regulator of voltage; (6) for use in sub-stations; (7) for the transformation and subdivision of voltage. These topics will be considered in detail. To straighten out the load curve. A curve of the load of a central station, plotted for twenty-four hburs, (see Fig. 2) will show that at certain hours a greater expenditure of power is re- quired than at others, and much greater than the average. This maximum may be double the average load, but it only lasts for a couple of hours. It nece.ssitates, however, an engine power .sufficient to supply the maximum load, and with the engine must be furnished dynamos, boilers and corre.sponding accessories. This surplus engine power remains idle for the greater part of the day and thus represents a waste of capital invested. . The function of the accumulator in this connection is to receive a charge during some part of the day, and when the extra load, or "peak" comes on, to assist the engines by discharging into the line. In this way the extra equipment is rendered unnecessary and economy is secured. Electrical Machinery. 17 To keep up the voltage on heavy load. At some distance from the generators, a heavy load will often cause a considerable drop in the voltage, on account of the resistance in the line. The electromotive force of the generator cannot be raised to any ex- tent, on account of injury to incandescent lamps and apparatus near it. Suppose a storage battery to be placed at the end of the line furthest from the generator, which at light loads will receive a charge, but which when the voltage drops owing to the resist- ance of the hue, will furnish current to the line and thus obviate the necessity of raising the voltage of the generator. A common example of this kind of practice is as follows : It may be desir- able to extend a line beyond its original limits, in which case a new line, or its equivalent, will be required from the extension back to the generator. If, however, a battery be placed at the beginning of the extension, it may be charged in the hours of light load on the existing wires, and discharged as required, and there will be no need for additional copper back to the generator. To carry the entire load when it is mi^iimum. At certain hours during the day or night, the load on the engines is very light and they are run at considerable disadvantage, the same amount of attendance being required. If, during the time of heavier load a battery be charged, it may be discharged into the line at the time of light load and the machinery may be shut down, with the con- sequent saving for attendance. To unify the engine load. The great variations which occur in the load of an engine lower very much its average efficiency of operation. A steam engine works most efficiently at, or near, its normal load ; but in a station, e.specially a power plant, the load fluctuates rapidly and between wide limits. The accumulator, if so arranged as to be charged when the load is light and di.scharged when the demand for current is heavy, will render the engine load uniform, and this can be arranged so that the engine will always be running at maximum efficiency. Similar reasoning ap- plies to the construction of the line. Suppose that the power to be delivered at a certain point, di.stant from the station, fluctuates between wide limits. Evidently the line must be of capacity suf- ficient to supply the maximum power demanded. An accumu- lator placed at, or near, the point of demand of the power, may be charged when the load is light and discharged when it is heavy and the current carried by the line from the station may thus be kept constant, which will give the condition for the maximum economy of line as well as engine. Electrical Machinery. i8 To act as a regulator of voltage. In certain prime movers, such as the gas engine, the supply of power is apt to be irregular, and for this reason cannot be applied to uses where a steady electro- motive force is required ; e.g., electric incandescent lighting. In this case an accumulator, placed across the terminals of the dyna- mo, acts to steady the electro-motive force by taking a charge when the electro- motive force is high and restoring it when the electro-motive force drops below the normal value. For use in sub-stations. A common practice in some places is to charge a number of batteries of accumulators in series, these being located at certain centers from which the current can be economically distributed. In this way the charging current is small and a small wire may be used for carrying it. For the transformation and sub-division of voltage. Finally, ac- cumulators may be used for the transformation and sub-division of voltage. . The first is accomplished by charging the batteries in series and discharging in smaller batteries either in parallel or separate. Or the opposite process may be used to raise the avail- able voltage by charging in parallel and discharging in series. For sub-dividing the voltage, the cells are charged in series and the series may be tapped at such points as to give the voltage re- qu;red, each cell furnishing two volts. In this way the three or the five wire system of distribution may be used from one generator. By placing a battery across the terminals of a 220 volt dynamo and connecting the neutral wire to the central part of the battery, the Edison system may be used. ElyECTRICAL MACHINERY. Plate II. ^ ^ ® ® ® R9.I. Xe^^, ly HP \ Ll M! CL »i / \ J \\ / 1 ) 1 ''9 z \ \ , y \ \ r//if£ 2.S zz if 19 IB '7. ' y t ICCC 1/10 '-^r ■>/< COA res ,-' / i--'s s. ^r^ ^ ^ — ■H»_ S>^ ^ '~~" '-^ ' — 1 :SWj .H}, ffp. CO /T-a ■ires ■~~ -^ -^ ^ Electrical Machinery. 19 UNE EFFICIENCY AND COPPER ECONOMY. 1 1 . The problem of the transmission of electrical energy with commercial economy divides itself into several phases, namely : (i) The transmission of a given amount of power a certain dis- tance with a fixed loss ; (2) the transmission of a given current a certain distance ; (3) the distance to which a given power can be transmitted. All of these items are to be calculated for maxi- mum commercial efficiency. Line efficiency is the ratio of the power delivered by a line to that which it receives. Evidently, the more copper a line con- tains the more electrically efficient will it be. Also any improve- ment in the quality of the copper or other material used, will have its effect upon the efficiency. A line may, however, be electrically efficient and yet not be commercially so, for the cost of the extra copper may be greater than the saving in power would warrant. 12. Problem i. To transmit a given amount of power at a fixed loss over a given distance with maximum economy will necessitate the employment of a high voltage. In this case the electrical efficiency of the line is given. A high voltage is more economical than a low one, because the current decreases at the same rate at which the voltage rises, with the same amount of power delivered; With a smaller current a smaller wire may be used with the same line less. As the power lost in the line is equal to the product of the square of the current by the resistance, the size of wire will vary directly as the square of the change in current or inversely as the square of the change in the voltage. For example, if a given line would transmit 100 H.P. at 10,000 volts with a weight of 1,000 pounds, an increase in the voltage to 20,000 volts would reduce the weight of wire needed to 1000 H- 2^ = 250 pounds. The limit to which the voltage may be raised without danger to apparatus is the determining factor in this ca.se. It is expen.sive to insulate a dynamo and line for excessively high pressures, and finally a point will be reached at which the saving in copper from an increased voltage will be more than counterbalanced by the ex- pense and difficulty of insulation. In alternating current systems, pressures of as high as 30,000 volts have been used successfully, but expensive insulators are needed to prevent leakage of the cur- rent from line to line or from the line to earth. Electrical Machinery. 20 13. Problem 2. To determine the most economical size of wire to carry a given current a certain distance. This is an entirely different problem from the preceding, for, as the current is fixed in value, the loss will vary directly as the resistance. As the size of the wire is increased its cost is also increased at the same rate, while the energy lost is less with a larger wire. There must be some point at which the increa.sed cost for copper is no longer counterbalanced by the money value of the power saved. Kelvin's Law. Lord Kelvin stated this principle about as follows : The proper size for a zt'ire which is to transmit a given airrent will be such that the iiiterest on the investment in the line, and accessories which vary with the size of the line, will be equal to the annual jnoney value of the loss of electrical power. This law is almost self-evident, for if the saving does not compensate for the expenditure it is evidently a poor investment, and on the other hand, if an increase in the line would more than pay interest on the investment, it would have to be increased in order to have an economical line. This law is graphically shown in Fig. i, which is the method for solving this kind of problem. The curves there given illustrate the following problem : Example. With copper at 14 cents per pound and electrical energy worth 7 cents per kilo-watt-hour, and money at five per cent, per annum, what size of wire should be used to transmit 10 amperes a distance of 1000 feet ? In solving this problem the in- terest on the cost of the wire and the annual value of the loss in power are plotted as ordinates while the sizes of wire are the ab- scissae. In calculating the kilo-watt-hours for a year the assump- tion is here made that the current flows for 3000 hours per 3'ear. It would, of course, vary with circumstances. The cost of insu- lators, etc. , has been neglected, although that would often be taken into account when they vary in size with the size of the wire. In the problem we find that -the curves cross each other at No. o wire. Brown and Sharp gauge. The total cost of delivering the power at one end of the line after it has been received at the other will be given by the sum of the ordinates of the two curves at any point, and this forms a third curve which shows the cost of using the line with the different sizes of wire. This curve has its mini- mum value where the other two curves cross. Houston and Kennelly have .shown that, after the size of wire has been determined for one current, the proper size for another current may be found by .simple proportion. That is, if a wire of Electrical Machinery. 2 1 one-tenth of a square inch section would carry a certain current at maximum commercial efficiency, one of two-tenths of a square inch section would carry twice the current with the same efficiency. This is evident because the cost of copper due to the added cur- rent and the loss in the line due to the same current have been increased in the same proportion. 14. Problem 3. To determine to what distance a current can be transmitted economically. The question to be deter- mined in a case of this kind is whether the power can be delivered from the end of the transmission line cheaper than it could be generated on the spot, taking into account the cost of copper in the line. Here the matter of allowable voltage becomes promi- nent, for, as has already been seen, the cost for copper diminishes very rapidly as the voltage rises. The first point to be settled is the highest practically obtainable voltage. Having determined this and the amount of power which is to be delivered, Kelvin's law will give the proper size of wire to use for any particular dis- tance. A number of such distances can be calculated in this way. The point at which the interest on the cost of the line is equal to the annual- money difference between the values of power at the two ends of the line is the ^naximum. distance to which the power can be trans- mitted commercially . A transmission line would onl}' be put in where there was a considerable difference between these two costs. As an example, take the Niagara Falls and Buffalo transmission. Power can be generated cheaply at the Falls, while to produce electrical power by steam engines is more expensive at Buffalo. It is found that even with the great cost of the transmission line, the power can be sold at a profit in Buffalo, cheaper than would be possible by steam generation on the spot. It is an open ques- tion, however, with the voltages at present obtainable, whether the same power could be transmitted as far as New York at a price which would compete with steam power. All of the foregoing conclusions apply mainly to a power traus- mi.ssion line. Incandescent lighting circuits present a different problem, for in this case a certain drop in the voltage is always specified, and with a certain drop and a given current the size of wire is determined without much calculation. Electrical Machinery. 2 2 DIRECT CURRENT SYSTEMS OF DISTRIBUTION. The earliest and simplest method of connecting electrical appar- atus with the generator consisted in running a pair of wires from the terminals of one to those of the other. - It is usual" to have more than one piece of apparatus connected to the same generator, and in this case certain modifications- of the simple transmission line are introduced in order to obtain the greatest economy of distribution. There are two main cla,sses of apparatus to which direct currents are furnished, namely ; machines arranged for a constant current and those adapted to the use of a con.stant pres- sure or difference of potential. Constant current apparatus is always connected in .series, for as the function of the generator is to keep the current constant, each piece of apparatus must be in such connection with the generator that the latter may have control of the current. The employment of a high potential is thus rendered necessary and a given amount of energy may be transmitted with a smaller current than would be required with a lower electro-motive force. This form of trans- mission, that is, connection of apparatus in series, is economical of copper and energy loss, for the wire is much smaller than for constant potential work, and the loss in power is proportional to the square of the current if the resistance is uniform. Arc lamps are usually connected together in series in this way, (see fig. 2) about fifty volts being allowed for each lamp in the series, with a cur-rent of ten amperes. Sometimes incandescent lamps are also connected in series, but the system is objectionable from the fact that the burning out of one lamp in the series, extinguishes all the others at the same time. Arc lamps are so arranged that when the arc is broken the lamp automatically short-circuits itself. Constant potential apparatus comprises the large part of elec- trical machinery and a different sy,stem of transmission is required. Each piece of apparatus must be in separate connection with the generator in order that it may receive the full electro-motive force. Two wires are run from the generator, usually known as ' ' mains ' ' , between which the difference of potential is practically constant at all points. The apparatus is connected to these mains in "parallel", that is, the terminals of each piece of apparatus are connected to the wires independently of the other pieces. (See fig. 3. ) Electrical Machinery. 2 3 The plain two-wire system has the disadvantage that, owing to the resistance of the mains, the difference of potential at the end of the line furthest from the generator is less than that at the near end. Apparatus which is sensitive to small changes in the voltage, such as incandescent lamps, will not operate satisfactorily unless the mains are made of such size that the "drop" is in- appreciable. To do this requires a large amount of copper, and various expedients have been tried in order to save copper in the line, while keeping the difference of potential between the mains fairh' constant. It should be said, however, that the simple two- wire system is in very general use, for circuits of reasonable length, and with satisfactory results. Expedients for saving copper in two-wire transmission. The simplest method to obtain a uniform pressure at all points of the line is what is known as a " loop " distribution, (see Fig. 4). One main is connected in the ordinary way to the dynamo, while the second dynamo terminal is attached to the far end of the sec- ond main by a separate wire, or loop. In this way each lamp on a lighting circuit is subjected to the same pressure, but it depends on other circumstances, whether the amount of copper in the loop, if used in the simple system, would not produce nearly the same results. If both the beginning and the end of the pair of mains are near the dynamo, as in figure 5, the method is an eco- nomical one. Although theoretically nearly perfect this system will only apply in a simple circuit. In a large lighting sy.stem, the lamps are very seldom connected directly to the mains, but to branches, as shown in figure 5. There are also a number of mains carried from the dynamos in different directions. Now if the mains be cross-connected at the points where they approach each other, the difference of potential between the wires will be equalizad, to some extent. This forms what is known as a " net-work " of mains, and it is the universal custom in locations where the cross-connection can be convenient- ly made. Before leaving the two-wire system it must be noted that this system can only be economically used where the load is not very far from the generator, except a high potential be used. In street railway work the pressure is high, and a perfect uniform- ity of pressure is not necessary, so that the two-wire system is used in this case. The trolley wire and feeders form one main and the track and ground form the other. The three-wire system. It is evident that the higher the pressure used in distributing current, the smaller will be the re- Electrical Machinery. 24 quired size of the wires. Edison has taken advantage of this fact by using two dynamos in series, furnishing a total of 220 volts. The lamps are connected in series of two by connecting one terminal of each lamp to an outside wire and the other to a smaller wire which runs to the connection between the two dyna- mos (fig. 6). This inner wire is known as the " neutral " and its office is to carry back to the dynamos the excess of current in one branch over that in the other. When both branches are balanced, that is having the same number of lights on each, the neutral carries no current. The neutral wire is made of one-half the cross sectional area of the other wires. The use of the sys- tem re.sults in a large saving of copper, in the proportion of two and three-fourths to four, as compared with the use of two dis- tinct circuits. In spite of this great saving in copper the eco- nomical commercial u.se of the Edison system is limited to short distances, say one-half mile, and it is thus only adapted to light- ing in the densely populated sections of large cities. The three wires which are used for the mains are encased in the same tube of iron, from which the}' are insulated by rope and pitch. In the houses the wires are run either in tubes in the wall, or on porce- lain knobs, which rai.se the wire clear of the wall. Leonard Three-wire System. This is a modification of Edison's plan. (See fig. 7.) A single dynamo supplies the outside wires with a pressure of 220 volts. The neutral is used as before, but it is connected at one end to the terminals of a small auxiliary dynamo which is belted to the large generator. As before, one end of the neutral is open. When the system is balanced the neutral carries no current and the auxiliary dynamo gives none, for it is adju.sted so that this will be the case. If the pressure on the branch, to which the small dynamo is connected, falls, due to the pres,sure of more lamps on that side,-the auxiliary dynamo will assist the main generator in supplying current to that branch. If, on the other hand, the pressure on this branch rises, owing to the other branch carrying more than its share of the load, the auxiliary machine runs as a motor, thus assisting the generator, and at the same time balancing the two branches by taking cur- rent when its branch is underloaded and by supplying current when it is overloaded. Five-wire Siemens-Halske System. Siemens improved the system of Edison by using four wires instead of three with a pro- portionate saving in copper. This was afterward extended to the use of five wires, for various reasons. Five hundred volts are Electrical Machinery. 25 used between the extreme wires and this voltage is convenient for the following reasons: (i) The same dynamo maybe used for the running of eight arc lamps in series ; {2) the same dyna- mo may be used for running street railway cars ; (3) the saving in copper due to the high voltage ; (4) the .use of motors on the lines, which run efficiently at this voltage. As the use of a num- ber of dynamos in series would be inconvenient, a different de- vice has been adopted. As shown in fig. 8, a single dynamo, which furni.she^ 500 volts to the mains, is allowed to drive a motor which has four armatures connected in series and revolv- ing in the same magnetic field. These armatures are known as the "equalizers"; and they are located in sub-stations near the point at which the load is located, .so that but two wires are re- quired to be run from the generator to the sub-stations. The neutral wires, of which there are three in this system, are con- nected to the junctions of the equalizer armatures and they carry current back to the armatures when an}^ of the branches become unbalanced. The function of the equalizers is to keep the pres- sure on the various branches at the same value, 125 volts, and it is accomplished as follows : If the electro-motive force in a branch falls below 125 volts, as would be the ca,se when one branch was taking more current than the others, the corresponding equalizer armature would act as a dynamo and make up tiie deficiency in the current. On the other hand, if a branch took less current than the others, its electro-motive force would be high and the equalizer armature would take current from this branch and would run as a motor. In this way the electro-motive force of the various branches is kept uniform. The system has some dis- advantages owing to the presence of the equalizers, which require .some attention and which must be brought up to speed if the main circuit is broken for any reason. Both the three and five- wire sy.stems can be connected to form a net-work as in the simple two-wire sy.stem and thus produce an equalization of pressure in the mains. Combined systems. Combined series and parallel systems are sometimes employed, e.g., when two arc lamps are run in series across no volt incandescent light mains. Or a number of incande.scent lamps are run in series across a power circuit of 500 volts. If it is desired to run a large number of lamps under the latter .system, the current must be switched into a resistance if one lamp is to be extinguished, or the current will be cut off from the whole series. This latter method is not in general use. ELECTRICAL MACHINERY. Plate III. T M CutiyeS ILCUSTHATIWfr , KELVIN'IS law: I X X >C X T/e-.Hr. £'A ' FQC/fl/Zf/i ^^^- S Electrical Machinery. 26 CALCULATION OF INCANDESCENT LIGHT WIRING. Incandescent lamps are not usually connected directly to the mains, except in small installations, but to branches connected to the mains at different points. In cases of this kind, the mains and branches will be of different sizes and must be calculated sep- arately. Wiring Calculations. The specifications for an incandescent lighting system always call for the dehvery of a given current at the lamps with a certain "drop" in the pressure in the line. This drop must not be very great, for the voltage at the lamps varies by the amount of the drop, when lamps are turned on or off. If but few lamps are in circuit, the drop is .slight and the lamps have nearly the voltage of the generator. When all the lamps are connected the resistance of the line causes more volts to be lost in it. As a difference of a few volts is noticeable in an incandescent lamp, it is evident that those near the generator would burn brigliter than the distant ones, if some precaution were not taken to prevent it. By the use of lamps of a higher resistance near the generator this trouble may be overcome, but the method is not used much as the lamps are apt to become mixed. As previously stated, a certain drop in the line is always allowed from generator to lamp, and this may be as much as two volts on a no volt circuit and correspondingly smaller on a 50 volt cir- cuit. To calculate the proper .size of wire to cause this drop is the problem now in hand. There are two .separate cases which are likely to be met, namely : ( i ) A number of lamps distributed along the mains ; (2) A number of groups of lamps on different branches. Another case which might occur would be that of a net-work of mains and branches, but this is only a difficult appli- cation of the principles which will be here laid down for the sim- pler cases. Case I. (See fig. i). A certain drop in voltage is allowed between the generator and the extreme lamp, at which point the voltage will be lowest. The current gradually drops off as lamp after lamp is passed, and if the same size of wire is used through- out, as is usual with a small circuit, the drop also becomes less per unit length as the end of the line is approached. This can be Electrical Machinery. 27 seen by reference to the figure, for from the generator to b the current as well as the drop is maximum, and the drop is less with a smaller current, the resistance per unit length being uniform. The line must be divided into sections in which the current is uni- form and the drop calculated separately for each section. The total drop will be the .sum of the ' ' drops ' ' in the .separate sections. Example : A circuit of 200 feet length contains eight lamps twenty-five feet apart, as shown in figure i. The dynamo supplies 112 volts and a drop of two volts at the extreme lamp is specified when all the lamps are on. If each lamp takes one-half ampere, what size of wire should be used ? Refer to the wire table on page 28. The drop in each section is the product of the current by the resistance of the section. Let 075 ft) 2 = .9 0) 050 0) 025 ft) ft) = 2.22 ohms per 1000 ft. 900 ft) volts. The nearest size to this is number 13 wire, B. & S. (See table. ) The Tapered Conductor. In, the case just taken, the wire was of the same size throughout the length of the circuit. This necessitated the use of a conductor between h and i large enough to carry four amperes, although but one-half ampere actually flowed between these two points. Figure 4 shows a similar case in which the first hundred feet of conductor carries sixty amperes ; the second, fifty ; and so on. The lo.sses in volts and watts are marked on the diagram and their values show that at the right hand side the wire is much larger than is necessary. The ideal arrangement for wiring a system of lamps like figure i is shown in figure 2. Having determined the allowable drop from dynamo to lamp, a .separate wire is run to each lamp and the same drop Electrical Machinery. 28 allowed in each. In this way each lamp has the same voltage and one lamp does not interfere with another. This plan would not be wasteful of copper, for exactly the amount of copper would be used as would cause the required drop, and no more. Moreover, each wire would be carrying a current in proportion to its size, which was not the case in figure i . WIRE TABI^E. FOR use; in house wiring calculations. B. &S. Diam- Current Capacity. Gauge eter. Ohms per 1000 feet. Amperes. Pounds per 1000 feet. Number. Inches Concealed. Open work. 0000 .460 .0490 218 312 640.5 000 .409 .0618 181 262 507-9 00 .365 .0780 150 220 402.8 •325 .0983 125 185 319-4 I .289 .1240 105 156 253-2 2 •257 ■1564 88 171 200.9 3 .229 .1972 75 no 159-3 4 .204 .2487 63 92 126.3 5 .182 •3136 53 77 100.2 6 .162 •3954 45 65 79-4 7 .144 .4987 38 55 63.0 8 .128 .6529 33 46 49-9 9 .114 .7892 29 37 39-6 10 .102 .8441 25 32 5I-4 II .091 1-254 20 27 24-9 12 .081 1.580 17 23 19.7 13 .072 1-995 14 19 15-6 14 .064 2.504 12 16 12.4 Note. — Number 14 is the smallest size of wire allowed by the fire insur- ance underwriters, except in lighting fixtures and flexible cords. The car- rying capacities here given are those specified by the underwriters. The large number of separate wires that would be required for the ideal arrangement makes it necessary to adopt a scheme which will be easier to install, and this is accomplished by combining the Electrical Machinery. 29 conductors in figure 2 into one conductor, the section of which will diminish as the current to be carried is less. This produces the form shown in figure 4, which is known as the "tapered " conductor. This is not as efficient as the same amount of wire used separately, for, as the calculations show, the drop at the re- mote end of the line is greater than it would be with separate wires by .26 of a volt. By the use of the tapered conductor, therefore, the drop has been increased a slight amount. Now, if the cross- section of the whole line be increased by an amount equal to twenty-six one-hundredths of its first value, the total drop will amount to the same as in the ideal case. The tapered conductor either causes a slightly greater drop than the ideal case, or a larger wire must be used. In the ideal case all of the lamps have the same presisure, while with the tapered conductor, those nearest the dynamo have a higher pressure than tho.se more remote. Although the tapered conductor is not equal to the ideal case, it is a great improvement over the single wire of uniform section. This is shown by a comparison of the calculations in figures 3 and 4. These two conductors contain exactly the same weight of cop- per and both carry the same current, yet the tapered conductor causes ten one-hundredths of a volt less drop and 12.2 watts less loss in power. Case 2. (See fig. 5). As before, the specifications call for a given drop between the dynamo and the most remote lamp. The figure shows a ca.se in which the lamps are connected to branches which are them.selves branches of the main line. It will evident- ly be impossible to use the same size of wire throughout, and an adaptation of the tapered conductor must be employed. As it would be u.seless to carry the calculations to the case of each indi- vidual lamp, these are considered in groups, and the distance from the dynamo is measured to the center of gravity of each group. The lamps on each of the .small branches are considered as mak- ing up a group. The first approximation will be made on the supposition that each group of lamps is connected with the dynamo by a separate pair of wires, and that these are made of such size as to cause the specified drop at each group of lamps. The size of the.se wires is readily determined on knowing the amount of current to be car- ried, the length of wire, and the allowable drop. This operation is shown in figure 6, where each pair of wires is represented by a .single line. Then all the wires which run parallel are combined Electrical Machinery. into a single conductor with an area equal to the sum of the s rate areas. As before, this causes a total drop greater than ideal case and the area of the wh,ole tapered conductor must b' creased so as to bring the extreme drop to the specified amoui Greater refinement of calculation than is here obtained is necessary because all the lamps are never burning at the s time, and any disarrangement of the number of lamps in cii changes the drop at each. It is then useless to make elabc calculations of the drop, the only requirement being that wher lamps are all on, the maximum drop, shall not exceed a spec amount. In practice it is never allowable to connect together directly different sizes of wire. The insurance companies insist that w a branch joins a main the wires shall be connected through " blocks," the fuse being a piece of fusible wire which will mel fore the current has attained a value at which it would unduly the conductor. In figure 6 these fu.se-blocks would be inserti a', b', d, d', e' f, and^' EI.ECTRICAI. MACHINERY. Plate IV ^A. ° XS/1. f 3/1. 'IJ.SA. '.^A f:/.S/l. i JCA. \.SA j tfOl/ F7e}.;. A^e/\ M £■//?- I ,1 , ' , , ' ,1 , '■ \/aA. /i/.\ I 66 000 sirooo ^■f-ooo J3000 HZooo rr :>oo A/ff/l 4VCAI- \JSSf0oo \ZAoooo\/faooo jVArr'S zoar | ^» I f*-* ■ <«r 0/f()/' //f yoCTS ] I I 1 a I /O/l. fK\ \/a/l. /K I I \/o/i./i/:\ I I F/q-.JS'. /OA. ft^. I /srooo /Z/ooo I £6ooo I /6SS-00 c. /v I I ^-ToTAi. ^-^6 t-r- o a. O o y /"t! -F :S: * o ?;^..r- 7?> / /'' 100. X ) -Loi p-lv- Hts s-Pf B.- /•■' z. = 11 100. aOO. 300. 400. 500. 600. Watts in Secondary Fig. 3. delivered pres.sure has dropped from 54 to 49 volts, or 5 volts, the primary pressure being 1080 volts throughout. This is a serious variation for incandescent lighting. Had L' and L" in this tran,s- former been practically zero this drop would have been, E' t" E'rt' t' [j^' + {'j,,)\j^"+r)]r ^' = 21.8 ohms ; R" = .04 ohms ; and r Substituting, 1080 X 5 X 700 54- [-•8+(^°}(-°4 + 5)]x 5 ohms. — = .94 volts. 35 Electrical Machinery . 37 It appears, therefore, that of the 5 volts observed drop in secondary pressure practically but i volt is due to ohmic resistance and the remainder to L' and L" The existence of L' and L" in annoy- ing quantities is due to the fact that these primary and secondary coils of this transformer establish magnetic induction about them- selves within the space they individually occupy and independently of one another. In all modern transformers the coils are so formed as to make the length of the path of the local inductance as great as possible. This is accomplished by making the coils thin, as shown in Fig. 4. By this means the values of L' and L" are diminished to such an extent that the total secondary drop in \ Fig. 4. pressure at full incandescent lamp load is not more than double the drop due only to the ohmic resistance of the coils. The function of the primary exciting current which was found when the secondary circuit was open, is to establish the induction in the magnetic circuit, which in turn forms the resistance E.M.F. As this resistance E.M.F. plus the fall of potential through the primary coil, which is always small, must at all times equal the line pressure, it follows that the induction and the current which establishes it diminish with the load but slightlj'. In the above ten light transformer at full load the resistance E.M.F. equals 1000 — I 10 X -5^ X 21.8 I = 989. This is a diminution of about V 700 J one per cent. , a proportion in which the induction and the reactance current are also diminished. Since this variation is .small, these quantities are generally thought of and spoken of as remaining practically con.stant at all loads. This reactance current mu.st be superimposed or added to the primary transformed current, as no Electrical Machinery. 3.8 part of the transformed current can set up induction through the magnetic 'circuit that is common to both coils. Since the purpose of this current is to form reactance E.M.F. and, therefore, to . establish induction, it is sometimes spoken of as the transformer magnetizing current, though more frequently than by any other expression it is called the transformer exciting current. This ex- citing current has a rather complex duty to perform owing to the fact that it must in addition to providing the alternating ampere turns of m. m. f. that are necessary to alternate the induction in the iron by the necessary amount, it must convey the electrical energy that is taken up in the iron through the phenomenon of hysteresis and given out as heat. The maximum value to which the induction in either direction must be conveyed so as to pro- duce a reactance E.M.F. of looo volts is, j^ lOOO X ID max y — V 2 TT 700 X I 33 257,000 lines. Figure 5 gives two curves .showing the relation between the s 50;0 00. S5,AoO. ^__ =- 7 260,000. ^ f 175,000. Z- / y / ^ // f 150,000. 6 / / /' / / 125,000. g 1 / 7 400,000. £ 1 / / re. m. / 1 ,60, 300. / 25, 300. 1 Prir lar 'Ci rre -ttJ Vmr ere .20 .1 -1 1 J .( 5 .06 / 1 .1 > .'. n \ f /- 26, m , \ [_ 60,000 75,1)00. to 1 1 100,000 ~1 / !l25,()00. / / 1 1 1 160,000 ■M- /, / y 175,000. ^ / / y /I / 200,000 ■B // -«-i 1 1 2S5.000 -^ ^ — 260,000 ^^ Fig. 5. current through the primary coil, with the .secondary on open cir- Electrical Machinery. 39 cuit, and the corresponding induction established through a com- plete cycle. The inner curve or card was obtained by taking a a long time in which to carry the induction through a complete cycle of variation ; the outer card gives this relation by taking the induction through the complete cycle in y^-j second. The difference is due to the eddy currents in the iron that set up m. m. f. 's in the iron circuit opposite to those formed by the pri- mary coil, thus requiring more current for the quick than the slow cycle. In this process of cyclic variation of induction the iron absorbs electrical energy from the electric circuit that sets up the induction, by requiring more current to establish a given in- duction when the induction is increasing than when it is dimin- ishing. On this account the transformer exciting current, when producing smooth curves of induction and reactance E.M.F., is irregular in form, (see Fig. 6) and occupies a time position with respect to the line E. M.F. so as to extract line energy that mu.st be supplied to the iron to be transformed into heat through the hysteresis of alternating induction. Since 1887, when this trans- former was manufactured, great improvements have been made in "■ — n v)» ^ ~ ^ ■ ■ ■" — . r- / i:-i)u 's 138 ~per Second Primary E.M F. 1030 Secondary E.'M.F. 54.5 Secondary Current.~v ^ J 12 H) s - / II )U. ,,-/ MX) ""V / 9(U >. , KIK) « / - —i 7 7(HI '- \ SIM / r mo. c. \ ^ so s J ~ 501). i / >0 t \ 41)0 ^ ^ M ^_ ^ / II) \ " J ^•1 It r ■^ f 30 V i t ~ t o >^ / \J ', iO W \ _ '■9 <^t r 0. / f- V V 10 3 ' f- r / V / \ "" I V ■■ f\ \ fe ^ ~y "■ ' / '\ \ 1 10 t Y '-~ "* r '~ / ' He c. A — ^ " / ■M •> / ' ^s. -/ -^ s ~ ■ ( 1 ■s as -1 - V- ~ ^ / \ / L - 2.i y (^ 1 ^ V' -- -- J p - - - - ~ - - - /-- -- - - - - - - - - -- -- - ^ ^ - - - - - - - 1 — -- - - - — — ~ ~' ~~ \ _, l_ _ _ "~ \ 1 _ ~ ~| 1 1 \ / _ _ _ _ ~" ~ ~ ' 1 L s / ~ ~ ~ _ _ J L 1 1 1 L _ L L 1— x L □ Pig. 6. the manufacture of sheet iron or steel for these purposes. A brand of soft sheet steel is now marketed by the Apollo Iron and Electrical Machinery'. 40 Steel Co. , of Pittsburg-, Pa. , which if used to make the core of this ten light transformer, would have caused the transformer ex- citing current energy to be 25 instead of 96 watts. THE INDUCTION MOTOR The Induction Motor is a special type of transformer in which the secondary coil moves with respect to the primary. Mechanical power is given out by the secondary, the measure of which is the product of the secondary current into the E.M.F. mechanically formed by the motion of the secondary conductors through the alternating induction established by the exciting cur- rent of the primary and into the cos Q if there is secondary local induction. When the induction motor is to be self starting, it is necessary that there be more than one set of primary and sec- ondary coils. Each primary is supplied independently with a line pressure that is equal to the line pressure on the other primary coils, but differing in phase from each of them. It is through these phase differences that the proper relation of magnetic induc- tion and .secondary currents to maintain continual mechanical motion is obtained. There are two types of self starting induction motors, distinguished by the number of differing phases .supplied to corresponding primary and secondary circuits. The two phase motor is marketed chiefly by the Westinghouse and Stanley elec- tric companies. This motor uses two sets of primary and sec- ondary coils. The primaries are supplied with equal and separate line pressures differing in phase by one-quarter period. The three-phase induction motor is principally marketed by the Gen- eral Electric, and the Siemens and Halske companies. In this type three sets of primary and secondary coils are used. The pri- maries are supplied with independent line pressures equal in , amount and each differing in phase from the remaining two by 120°. The principles of behavior of the two types of motor are identical. The two-phase motor appears the simpler to the begin- ner. For this reason it is selected for the following demonstration of the characteristic behavior of the induction motor. Induction motor behavior. In Fig. i is given a diagram show- ing the arrangement of a .single portion of the two sets of primary Electrical Machinery. 41 and secondary coils with respect to themselves and their magnetic circuit for a two-phase induction motor. This diagram is drawn ^777 ^" i^'^^'>//^)Ji^>^^r/,//}/,^/,^,,^f!^^j^^^y^y^^^^JJ^^^^^^; r^ Fig. I. so as to be characteristic of the particular form of induction motor marketed by the Stanley Electric Co. , of Pittsfield, Mass. Though the action in this type is identical with that of all others, it is so modeled as to make this action more readily apparent to the stu- dent who is new to this subject. The coils i and 3 form one pri- mary, and 2 and 4 the other ; 1 and 3 have a common magnetic circuit independent of i and 3. The magnetic circuits are closed across an air gap through the laminated iron that supports the secondary coils, a a, and bb. The air gap is neces.sary for me- chanical clearance, permitting the secondary coils, together with that portion of the magnetic circuit upon which they are mounted, to move relatively to the primary coils and their magnetic circuits' As,sume now that the two-phase line pressures are applied at the terminals i and 3, and 2 and 4. Assume also that the i and 3 pressure is one-quarter period or 90° in advance of the 2 and 4 pressure. Consider the .secondary circuits, a a and b b, on open circuit for the time being. An exciting current will now be e.stablished through each set of primary coils equal to R" is very small when compared with «''/,'' and may be neglected, when the expression for the exciting current becomes C^j= E'-i- wL. The reactance E.M.F under these circumstances is prac- tically equal and opposite to, or 180° back of, the impressed E.M.F., E' The phase position of the magnetic induction is always 90° in advance of the reactance E. M.F. that it produces and, therefore, 90° back of the impre.ssed E. M. F. When an air gap exi.sts in a mag- Electrical Machinery. 42 iietic circuit, as is the case here, the position of the total exciting current, including hysteresis and Foucault current effects is almost identical with the phase position of the magnetic induction, being slightlj' in advance of it and, therefore, not quite go° behind the impressed E.M.F. Consider now that the secondary a a is closed through a non-inductive resistance sufficient to limit properly the current induced in it by the magnetic induction from 2 and 4. The current thus established through the secondary conductors aa is due to and in unison with the reactance E.M.F. that is 180° the pressure on 2 and 4. These conductors are located in a iield of induction that has exactly the same time position with reversed sign. This is easily seen from the fact that the pressure on 1 and 3 is 90° in advance of 2 and 4 ; that the i and 3 iiiduction is 90° behind the i and 3 pressure and, therefore, in the same phase position as the 2 and 4 pressures ; hence the current in a a is 180° behind both the line pressure 2 and 4 and the induction i and 3 in which a and a are located. As a change of phase of an alter- nating quantity of 180" is equivalent exactly of a change of sign of that quantity, the fields at a a will act on tho,se currents just the same as though they were in unison, with the exception that the effort to mechanical motion will be opposite in direction. The student should remember that a change in the sign of both field and current at the- same instant will not change the direction in which the current will move through the field. Similarly d and l> will have a current induced in them that will be acted on mechan- ically by the induction through the circuit 2 and 4. A study of the signs, directions of currents, inductions and mechanical motions in each of these secondary coils, will reveal the fact that the coils a a and bd will be moved in the same direction, in this instance toward the left. Permit these conductors to move. In an instant the coil bb has moved into the position just occupied by a a and is acting, therefore, exactly as a a acted, while a a has moved for- ward to a position corresponding to bb, behaving as b b did, and both tending at all times to move to the left. Reversing one of the primary pressures would make this motion take place from left to right. The conductors, by continuing their motion, soon attain a velocity such that the E.M.F.s produced by the mechan- ical motion across the magnetic induction through which they are passing, become appreciable. As the E.M.F.s are opposite the secondary induced E.M.F., the currents in aa and iJ i5 are cut down and there is no tendency for the speed to increase. The Electrical Machinery. 43 resistance in circuit with a a and b b may now be cut out gradually, the secondary velocity will again increase, thus developing more mechanical E. M.F. until finally a point is attained where all of the secondary resistance has been cut out and the secondary is driving at full speed. At this point the counter E.M.F. mechan- ically formed, is practically equal to the the secondary induced E.M.F. Should the secondary structure now be called on to do mechanical work, it will lessen its speed a trifle, thus lessening the counter E.M.F. and permitting the induced E.M.F. to .set up more current, producing more torque and maintaining the load without further lowering the speed. What drop in the speed does occur is small, because the resistance in the secondary conductors is low and the currents producing the load torque effect but .small falls of potential when compared with the impressed and counter E.M.F.s. As the difference between the secondary in- duced and counter E.M.F.s will be proportional to those falls of potential, it follows that such differences will be small, the varia- tion of counter E.M.F. be small, as will be the corresponding variation of speed. Fig. 2 gives a side and vertical cross-.section of a Bradley two- pha.se induction motor. There are two independent motor sec- ondary structures mounted on the same shaft. In the right Fig. 2. structure, which is enclosed by the primary or outer structure, a starting resi-stance is permanently inserted in the secondary cir- cuits. In the left structure not enclosed by the primary, the sec- ondary circuits contain no additional resistance and are to be used when the secondary structures are brought to full speed. When this motor is started the primary is in the position shown in the figure. As the secondary comes to speed by means of the hand lever, ^a, the primary is moved to the left so as to use the sec- ondary that contains no starting resistance. Many other forms Electrical Machi, inery. 44 are adopted in practice to control properly the starting currents. Induction motor characteristics. The torque speed characteristic curves of an induction motor operating from a multiphase con- 8 § i 8 8 8 8 8 8 8 in vD r- 8 To, que, Ib-s. t.fJ ot n dius -I a g -1 ---- — — . L . - — _j__ ■ — . ^..^ - — -~~~^ ^ ___^ i ^- --. .^ ^ -^ "~" X o ^ -^ / \ 1 o / ^ ^ ^ / / y K ■o / / y \ a CO / A \ / / / * / - / 1 I lOOjS 90 80 70 60 EO 40 30 20 10 Fig 3 stant potential circuit are given in Fig. 3. Curve / shows the relation of torque and speed when all of the secondary starting resistances are cut out. Note that the starting torque is small, while at full speed and load the motor drops but little below the no load speed. The excessive dropping off of the torque at the low speeds is due to secondary inductance, which shifts the phase position of the established currents so that they produce but little torque. The insertion in the secondary circuit of fall of potential due to non-inductive resistance or of mechanically formed counter E.M.F. , limits the secondary currents and maintains them more nearly in unison with the secondary induced E.M.F. , the position they must occupy in order to produce torque. Curve /// is like curve /, except .some secondary starting resistance has been used. Curve // was obtained, when the secondary starting resistance was so adjusted that the .starting torque should be a maximum. In practice some starting resistance is used that will give a sufficient starting torque. The motor will start and come to speed along a characteristic curve like // or ///. As this is done the resistance is gradually cut out, being all out when the motor is running at full speed and when it will give a performance like that of curve /. Electrical Machinery. a 5 The Synchronous Motor. The first alternators that were used in our country about ten years ago had smooth bodied arma- ture cores, upon which was operated a minimum amount of cop- per. The armature self-induction and reactive effects amounted to so little that we may neglect them, when making an examina- tion of that which should be their behavior when operated as synchronous motors. In Fig. \, E, E and E" are the generator, motor, and resultant pressures,—^" is the difference between E and E'. When E and E are equal, as in this figure, at the moment of synchronized connection of the motor with the genera- tor, E" will ha\'e no value. .£" is the E.M.F. upon which de- pends the establishment of current through the motor. With Fig. I. Fig. 2. E" at zero the motor current will be zero, no power will be de- veloped, and the motor will immediately fall behind the generator to some point like d. Here E" has a large value. The circuit of the two armatures and line pos.ses.ses some resistance and a negli- gible amount of self-induction. Current will be established in uni- .son with E" . We see, however, that no current which occupies the time position of E" can produce power in the motor, for all such current is as much in unison with the motor as it is with the generator E. M.F's. Under such circumstances the motor will quickly come to rest. We now diminish the field of the motor and thus lower its E. M.F. , E , as in Fig. 2. At the moment of synchronized connection iT— ^' =^" will be in unison with i? and opposite to E The current that E" establishes occupies the same time relation, and mechanical power equal to CE is pro- duced at a power fraction of unity. In general this will be more Electrical Machinery. 46 than sufficient to keep the motor at this speed position, and it will be accelerated to some position like d. At d we find E" and, therefore, C is greatly increased while the developed power which is proportional to a/ is much diminished. Thus the motor takes up some position where the developed power is equal to the de- mand made upon it. Increase in the demands on the motor for power will cause it to fall back in speed position until the point of exact synchronism is again attained, after which it will come to rest, for it is evident from the diagram that at this point af attains a maximum value. Note that the power factor is high only at the point where the motor is about to stop on an attempt being made to further in- crease the load. It is evident that such a motor would be very unsatisfactory in practice. On adjusting E so that the motor will be reasonably free from the danger of lo.sing synchronism, the power factor is much diminished. Under the most favorable circumstances a motor of this class should stand a half over-load Fig. 3. suddenly applied without going out of step. Then frequently in practice a motor is loaded on the average to no more than 50 per cent, of its rated output. All this points to an average working power factor that is less than 50 per cent, to be realized by this class of .synchronous motor. Five years ago the T-toothed alternator armature was intro- duced for electric lighting. (See Fig. 3.) This design is now largely superseding the smooth-bodied type because of its struct- ural superiority. Makers had not gone far in their experience Electrical Machinery. 47 with machines of this type before thej- realized that they possessed some features of special excellence for operation as synchronous Fig. 4 motors. They differ electro-magnetically! from the smooth core armatures through their large self-induction and armature reactive properties. The former property changes the time position of the current with respect Xa E" , while the latter acts to equalize E and E' , however these may differ, though such action amounts to very little when the power factor is; high, as it must be in most cases. Consider then that in Fig. 4 we have represented the conditions for a synchronous motor with a T-toothed, or iron clad armature. The fall of potential will be small as compared to the reactance due to inductance, and E" will establish current at a time position that is practically 90° behind itself. By starting with E and E' equal, the motor will promptly take up such a lagging position as d. Now note that C, as established by E" , is right in the posi- tion required for the maximum obtainable power factor. See that this current has come up in proportion to the power demanded, and that it will continue practically to do so, even to the point where E" is 50% of the value of E. Ultimately, as in all other cases, a~.point is found where the power factor diminishes so rapidly that ag, proportional to the developed power, no longer increases, and beyond that the motor quickly comes to re,st. The design is easily arranged so that this poiat will be at almost any desired over-load. On coming to rest, too, this motor will suffer nothing from the .sudden rush of current that takes place, because the inductance of the armature places a convenient limit upon the Electrical Machinery. Pig. 6. Fig. 8. Electrical Machinery. 40 greatest possible current that can be established. The circuit breaker will cut off such a current before damage can be done. Fig. 6 illustrates the actual behavior of a motor in which the field was adjusted so as to bring the current, 6.46 into the position, where the greatest power factor is obtained. The relation of the inductance and resistance of the motor circuit was such as to make C82.5° behind E" . ^and E' are nearly alike, -they differ 1.3 volts, the drop in the line. Fig. 8 shows the behavior of this same motor when its field ex- citation is greatly reduced, and Fig. 4 is the corresponding be- havior with field excitation greatly increased. You see that with the weak field the motor currents take up a position behind E and with a .strong field a corresponding position in advance of E, the general pressure. The power factor is greatly reduced, and the Fig. 8. conditions, therefore, are undesirable for practice. Note that the reactive effects of these currents on the motor field in all cases are to make the motor E. M.F. more nearly like that impressed by the generator. In this connection it is necessary to remember that a lagging current lowers the field of the generator, and an advance current raises it ; the opposite occurs in the motor. In order that the output of a motor be as high as practicable Electrical Machinery. 50 for a given size of machine at a given first cost, it is necessary to operate at the highest practicable power factor. This is easily accomplished when we do not have to deal with a line on which there is considerable drop, say from 10% to 30 or 50%. The effect of the resistance of the line is to lessen the angle that the current occupies behind E". In Fig. 7 we have illustrated a case where the inductance and resistance of the lines and motor circuit are such that the current will be 60% behind E" Here, for the highest power factor at average loads, we must have the current remain slightly at the rear of E. To do this it is ntce,ssary to compound the generator producing E, as indicated in the diagram, so that E" will remain approximately in the same position, as Fig. 6 shows at all loads. Single pha.se synchronous motors are generally started by means of a special, single phase auxiliary induction motor, that is self- starting under load, though inefficient, and is used for starting purposes only. (See Fig. 8.) A small generator answers very well as an exciter, — though frequently alternating current from an auxiliary winding on the synchronous motor armature is used for excitation when rectified by a two part commutator. The Rotary Transformer. Any ordinary direct current bi- polar generator that emploj'S a closed coil armature may readily be converted into an alternator by mounting over the commutator two collector rings, connected to diametrically opposite commuta- tor bars. An alternating pre.ssure will be developed between brushes placed on these rings, the maximum value of which will be equal to the direct current pressure that the armature will de- velop between the normal points of commutation on the commu- tator. The working value of an alternating E.M.F. obtained in this way would therefore be E^ -==- v/g where E^^ is the direct cur- rent pressure. The periodicity of a two pole machine would, in general, be too low for most practical purposes. When, there- fore, a direct current armature type of alternator is to be designed the multipolar two path armature and field are selected. The col- lector rings would be mounted on that end of the armature shaft which is remote from the commutator, and connected to con- ductors that pass directly to commutator bars that are separated by an interval of bars that correspond to the pole interval, or the distance on the commutator between neutral points. Two addi- tional rings may be used with commutator connections, midway between tho.se for the fir.st set of rings. The result is that between Electrical Machinery. 51 the second set of rings an alternating pressure is produced that is maximum when the pressure between the first set is zero, and zero when the first pressure is a maximum. Thus a two phase source of alternating pressure is produced. Similarly a three phase pres- sure may be obtained by connecting three sets of rings at inter- vals on the commutator of two-thirds of the direct current com- mutation interval or the interval assigned to one pole. Views of various types and forms of machines of this class will be exhibited to the class with the aid of the lantern. Such machines may be used to furni.sh either multiphase alternating or direct currents or both. Being alternators they will work as synchronous motors. When working as synchronous motors they form alternating counter E.M.P.s, due to the rotation of their armatures in the magnetic fields. The same conductor E. M.F. that produce alter- nating E.M.F.s from collector ring to ring will also form a direct current E.M.F. at the neutral points on the commutator, from which direct current energy may be drawn, just as alternate cur- rent energy is drawn from the secondary of a static transformer. Thus this class of electrical machinery has come to be called a rotary transformer. Note then that by supplying one of these rotary transformers with a two phase current at 350. volts, that the direct current energy would be discharged from the commu- tator at 350 X \/2 = 500. volts. The practical utility of the rotary transformer will be discussed in the next lesson. Electrical Machinery. 53 Phase Transformation in Polyphase Practice. In Fig. i two sources of alternating pressure are available as o' y' and o' x' . o'y is one-quarter period ahead of 0' x' Transformers with numerous secondary taps, as shown in the figure, enable one to lessen these pressures through efficient transformation by any desired amount, or from the maximum effective secondary pres- sure to zero pressure. x and oy are the maximum effective secondary pressures, and o c and b are corresponding smaller values obtained by adjustment as indicated. In Fig. 2, oa\s an E.M.F. that is obtained by combining in the proper proportions E.M.F.s from oy and ox, having the values b and c. Taken with respect to phase or time position, a new E.M.F. is thus obtained that is B° ahead oi x and 90°—^ behind oy. Taking the phase position of the new E.M.F. with respect to o x, and remembering that angles taken anti-clock wise are positive and those taken clockwise are negative, the relations between 6, o a, oh and c, are a a cos S = oc n o a .sin d = — . ob As it is entirely practicable to vary b and c independently from zero to oy and ox respectively, by the device of tapping the sec- ondary coils of the transformers at points corresponding to the desired E.M.F., so 6 may be varied without changing the value of the pressure o a, from zero to 90°. We are not limited to this amount of variation of Q. In Fig. i it is seen that in joining the secondaries in series, the circuit might have been arranged so as to go from o through b to instead of to c, oi x and from o through o c to a, then through the working circuit back to o of oy. The effect in the circuit of this reversal of one of the E.M.F.s is better seen in connection with Fig. 3. There it is seen that by reversing the sign of x the new time position that it will occupy is the broken curve x. The phase position of this broken curve is 90° ahead oi ox or a total forward change oi o x oi 180°. The result in the diagram of Fig. 2 of this reversal in the circuit of the secondary ox is that o cis changed to — o c. The result is that cos 6 = and — c n a b sin a = — . ob Electrical Machinery. 53 It is evident now that B may be made to vary from 90° to 180° by selecting values dob and c ^ as indicated. By reversing h and c, both sin 6 and cos 6 each change sign and values of or the positions of the new E.M.F. ahead of ox may be obtained from 180° to 270° Then by rectifying the direction of <7csoasto make cos again positive, and by permitting o b to remain re- versed, causing sin 6 to remain negative, 6 of the new E.M.F. may be made to vary from 270° to 360''. Thus by combining sin 6, cos 6 portions of two sources of alternating pressure in quadrature, E.M.F.s may be obtained in any desired new phase position. We will now turn our attention to the system of vector diagrams that is used for determining the new E. M. F.s that result from combinations of polj'phase sources of E. M. F. The diagram of fig. 4 is used in lieu of the one in fig. 2. When the vectors are drawn as in fig. 4 the broken line b which completes the triangle corresponds to the diagonal of the parallelogram ob (^broken line) and is, therefore, the pressure put upon the line by the combined E.M.F.s ob and c. But the interpretation to be put upon ob\s, that it is the external or working circuit through which the new E.M.F, establishes current. That, however, makes a diagram of mixed E. M. F.s. It is always desired to know the value and pha.se posi- tion that the new pressure will have, not in the direction that it is exerted through the line circuit but in the direction that it is ex- erted through the circuit of the equivalent source. The equivalent source in this instance would be an alternator coil or another secondary coil whose terminals are connected from o, ob, o c to b oi ob and in which the E. M. F. oa would be generated. Such an E.M.F. would impress through the line the pressure ob {broken line') just as is accomplished by the combined E. M. F.s i? i5i and oc. The equivalent source E. M. F. is more conveniently found by a continuation of the process by which ob {broken line) was found. The positive direction of the circuits in which the source E. M. F.s oc and b are produced is taken from b to ob\)y way of the route oc, o b. Now to pass from b to o in the same conductor direction it is necessary to go back through b and o c\yy the route bo, co as there is no actual circuit from b to direct. In continuing the circuit from btoo'xt is seen that the direction of the electric circuit is opposite to the E. M. F.s ocand c^and that these E. M. F.s are, therefore, reversed with regard to the direct circuit (5 and the directions of the vectors that represent them should be reversed for determining the equivalent .source E. M. F. On reversing the Electrical Machinery. 54 arrow heads oi o b and o c so as to make these E. M. F.s read co and bo\V\s found that their resiihant equals oa. Thus by com- pleting^ the cycle of circuit direction taken throughout with the same sign where two or more sources of E. M. F. are joined in series, that is, by starting through the circuit in a positive direction and continuing through all of the connected sources of E. M. F. their equivalent sources of E. M. F may be determined. It is neces- sary that the sum of the actual and equivalent E. M. F.s shall be zero at all instants of time. In fig. 4, omitting the broken vector we may designate any instant of time in an alternating cycle by drawing an}' line of reference through the center of the triangle. The projections of the E. M. F. vectors forming the sides of this triangle will be the instantaneous values of the E. M. F.s. It is seen that the algebraic sum of such projections taken with their proper .signs must alwaj's be zero. Fig. 5 gives a vector triangle which determines the equivalent source of E.M.F. when the circuit through the ijx secondary of fig. I is reversed. In fig. \ oi o c \s now joined \.o o oi ob, b and fare now connected to the line to deliver the new E.M.F. instead of b and o as before. Note in fig. 5 that the positive or anti-clockwise direction of the circuit now begins with b and passes through bo and a c. The actual E-M.F.s that occur in the circuit are taken from b Xa c through 00 and are b and c, while the reverse of this is the positive direction for the circuit of the equivalent E.M.F. , Fig. 4-. Fig. 5. Fig. 6. Fig. 8. Electrical Machinery. 57 THE INCANDESCENT LAMP. The incandescent lamp is a device for transforming the heat generated by the passage of an electric current through a sub- stance of high resistance, into light. At low temperatures all the power absorbed by the substance passes off as heat, but when a temperature of 350° C. has been attained, a certain percentage of the energy becomes light. The light consumes from five to ten per cent, of all the energy absorbed, the remainder passing off by invisible radiation. The incandescence of the filament in such a lamp is caused by the rapid movement of the particles of carbon, and combustion is prevented by extracting the air from around the hot filament. A complete lamp consists of the bulb ; the filament ; the base, which includes the support for the filament ; and the connecting wires between the filament and the base. These wires are usually of platinum, for this metal does not shrink away from the glass in which it is sealed, on cooling. Figure i shows the various parts of the lamp. A is the bulb ready for the filament, B the filament and holder, and C is the finished lamp. The bulb is of glass and of sufficient size that it will not be unduly heated by the hot filament. The filament is now made only of carbon, though platinum wire was formerly used. Its size is such that it will consume a certain number of watts with the production of the required candle power. The base consists of a core of por- celain or glass on which are mounted two brass pieces which come in contact with the terminals of the circuit when the lamp is placed in its socket. The manufacture of an incandescent lamp consists of the fol- lowing processes: (i) Forming the filament; (2) carbonizing the filament; (3) "flashing" the filament; (4) blowing the bulb ; (5) mounting the filament in the bulb ; (6 ) exhausting the air from the bulb ; (7) testing and rating the lamp. Filaments are usually made from vegetable fiber which has been changed into a form called amyloid, by treatment in .sulfuric acid. The fiber is made uniform in section, by passing it through a series of die plates. This prepared fiber is wound on -carbon blocks of a proper form to produce the required shape of loop and the blocks so prepared are carbonized in a carbon box, which is closed, but supplied with vents so that the volatile matter may pass off. This box is gradually raised to a white heat, and after cooling down, the filaments are removed in the form of pure car- Electrical Machinery. 58 bon, though more or less granular in character. The filaments are now temporarily mounted for electrical connectiqji and are heated by an electric current for some seconds in a hydro-carbon gas, vapor or liquid, which is decomposed by the heat and which deposits carbon on the filament, especially in the hottest parts where it is needed most. This process is known as "flashing" and when the filament is removed from the vessel in which the operation is carried on, it is found to have lost its granular char- acter and to have a smooth, hard and apparently metallic surface, more suitable for the production of incandescence and more durable. The filament is mounted on its platinum wire supports by means of a carbon paste, which is deposited on the joints. Klec- troplating and soldering was the method formerly employed. The platinum wires are next sealed into the base of the bulb, which has been made in the meantime as follows : The bulb may be blown either from a piece of glass tubing or directly from the pot. In the latter case the proper form is obtained by blowing in a mold, while in the former case the shape of the globe depends upon the skill of the operator. The air is removed from the bulb finally by means of the Sprengel's or mercury vacuum pump, which is attached to a small piece of tubing left on the tip of the bulb by the blower. After the removal of the air this tip is sealed over. After the lamp has been mounted in plaster in the base, which consists of a porcelain or glass core, carrying the bra,ss contact pieces, it is tested and rated at the voltage and candle-power at which it has the desired efficiency. The candle-power of a lamp depends on the direction in which it is viewed, for the light given out in any direction will depend upon the luminous area exposed in that direction. Figure 2 shows the variation of candle-power in a flat filament lamp, as different angular positions are occupied by the filament. Twisted filaments are often used to cause a more utiiform distribution of light. The real illuminating power of the filament would be the average value in all directions, and this value is called the "mean spherical" candle-power. It is not usual, however, to give this value, for the base cuts off all the hght in that direction. The ordinary method is either to rotate the lamp about its own axis while making the measurements, or to take the mean of the meas- urements when the edge and when the side of the filament is turned toward the photometer. The life of an incandescent lamp means tlje time that it will Electrical Machinery. 59 operate before being "burned out." This life will depend upon the quality of the filament and the voltage at which the lamp is operated. The lower the voltage the longer will be the life, and vice versa. Complete tests of a lamp includes measurements of the following quantities: (i) Life; (2) voltage; (3) power consumed ; (4) candle power. When a lamp is new it is much more efficient than when it has been burning for some time, so that, after a time the candle-power will be less if the voltage remains constant, or if the voltage be increased to raise the candle-power to its original value, the power consumed will be greater. This is shown in Figs. 3 and 4. This decrease in efficiency is due to the decomposition of the filament, as a perfect vacuum cannot be obtained. The life is subject to changes in the voltage and will be much less if the voltage be variable. The useful life of a lamp is the time taken for its candle-power to fall off 25%. Roughly speaking, the candle- power of a lamp is proportional to the square of the voltage, which shows that a slight increase in the voltage will produce a great change in the candle-power and consequently in the rate of decomposition of the filament. The so-called efficiency of an incandescent lamp is the number of watts used to produce a candle-power. This efficiency will vary in the same lamp, depending on the voltage which is used, and the efficiency then is not really a property of the lamp. For ex- ample, if a sixteen candle-power lamp consumed 56 watts at 100 volts it would be, at this voltage, a 3.5 watt lamp. Now if the voltage be increased to 104, 60 watts would be consumed and 20 candle-power produced, making an efficiency of 3 watts per can- dle-power. A further increase of voltage to 1 10 would produce a candle-power of 28 and a consumption of power of 78 watts and and the efficiency will rise to 3.5 watts per candle. The efficiency at which it is best to run a lamp is purely a commercial matter, for as the life of the lamp is shorter at a high efficiency, a point will soon be reached at which the cost of renewing the lamps is equal to the saving of power, and this point will be the limit to which the efficiency of a lamp should be increased. High effi- ciency lamps are also very susceptible to changes in the voltage so that they can never be used on a circuit where this is the case. By a high efficiency lamp is meant one which consumes less than 3 watts per candle-power, while a low efficiency lamp may use from 3 to \yi watts. Two and one-half watts per candle is the hio-hest efficiency which is found economical at present. Electrical Machinery. 60 Commercial Features of the Incandescent Lamp. The commercial problem to be settled is the proper efficiency at which lamps should be run and the proper time at which they should be discarded. It is very seldom economical to run a lamp until it burns out, for the efficiency after a time becomes very low. The determination of the proper initial efficiency depends upon the useful life of the lamp at that efficiency. If the .saving in power is more than overbalanced by the cost of renewing lamps, evi- dently the efficiency of the lamp used is too high. The second important problem is that of the proper time to allow the lamps to burn. Mr. Carl Hering has shown that there is a time at which the lamps should be discarded, and this has been called the "smashing point." The calculations of this point should be made on the basis of the cost of a candle-power of light for a certain time, say 3000 hours. Lieut. M. K. Eyre {^Electri- cal World, Vol. XXIII, p. 54) published calculations, from the data of which the following example is worked out : Problem — To determine the ' ' smashing point " of a lamp under the following conditions. Efficiency and candle-power curves as shown in figures 3 and 4. Co,st of electrical energy, 7 cents per kilo-watt-hour. Initial efficiency of lamp, 3.2 watts per candle-power. Cost of lamp, 20 cents each. There will be a point beyond which the candle-power has dropped to such an extent that the wasted energy more than makes up for the cost of new lamps. ^Assuming that the lamps run a total period of 3000 hours and assuming different smashing points, the following table has been calculated : Lamps Smashed at No. Lamps. Mean C.P. Mean Eff. Total Cost I C. P. 3000 Hours. 200 hours 15- 19.7 3.22 83.05 cents 300 " 10. 18,7 3-38 81.7 400 " 7-5 17.9 3-54 82.7 500 6. 17.2 364 83-7 600 " 5. 16,5 383 86.7 700 4-3 16.0 3-96 88.6 800 " 3-75 15-5 408 90.7 goo " 33 15- 1 4.18 92.3 1000 3- 14.7 4.26 93-9 Figure 5 is the curve plotted from these values, and this shows that with the conditions assumed, the most economical smashing point is after the lamp has burned 320 hours. ELECTRIC AI, MACHINERY. Plate VI. 27o" CANote PoWEf? 2o 16 /\^ ^^-.^.^ I FE Af 4D CANDLE POWE 5 - 14- ~- — -Zrt:>; "■■ --IT :rrr^ ■^_ lo ■~' • ^ ™^^ ng.3 EFF. WATTS 4. z Qnfi>SHiNeC^iftT^^— FigS Zoo fsoo Aoo \foo 600 900 Sno / 3 times the value of current in ab, be or ea. It should also be noted that in the three phase system the pres.sures between wire and wire in all instances are the same, so that the gain in copper of 25% is genu- ine for all instances. It is also a fact that a three phase line will develop, when trans- mitting a given amount of power at a given pressure, less react- ance E.M.F. due to inductance, than a two phase or .single phase line. For these rea.sons the three pha.se line is much preferred for long distance transmission of power. The Monocyclic System is used where local distribution of light and power current is to be effected within a territory covered by a radius of a few miles. Where much incandescent lighting is to be done, the single phase 1000 or 2000 volt alternate current transformer system is much preferred in practice because of its simplicity. There is, however, a demand for facilities to operate numerous .small stationary motors from incandescent lighting circuits. The.se motors vary in size from yi to 50 or 100 horse-power. A service of this character is best made by means Electrical Machinery. 68fl of the monocyclic system of generating and distributing elec- trical energy. The circuits of the generator for this system are given in the diagram of Fig. 4. There is a main alternator coil developing a .single phase pressure of 2080 volts. Should the entire output of the alternator be used for incandescent lighting, this coil would furnish such output in .single pha.se current as in the single phase system of distribution. For furnishing power current to induction motors an additional generating coil is provided. This is called the "teaser" coil. It generates a pressure at right angles to the main coil, at the mid- dle of which one end of the teaser coil is connected, while the other end is taken to a collector ring. For power purposes then the main circuit and the tea.ser terminal by line conductors are run to the motor locality. There two transformers are used to change the special two phase or monocyclic source into three phases of properly reduced pressure. The diagrams of connections and of values applied respectively, are given in Figs. 6 and 7. As the tea.ser coil needs to have but one-quarter the capacity of the main coil, it follows that the cost of the generator is but httle in excess of one in which it is entirely omitted. A monocyclic generator may profitably, therefore, be installed, even where the amount of power current bu.siness will be small or uncertain. ELECTRICAL MACHINERY. Plate VIII. LONG DISTANCE POWER TRANSMISSION STEP-UP AND STEP-OOWN TRANSFOREMERS NalSMI General Electric Ca June K'^^. ^' fTTTrrrr^.l ^ Lj rnxo: Electrical Machinery. 69 COMMERCIAL ELECTROLYSIS. Electrolysis is the name given to chemical decomposition pro- duced by the electric current. When a current of electricity is passed through a liquid, decomposition of the latter usually occurs and this may be apparent in the following ways : (i) by a resolution of the liquid into its gaseous components, (2) by a deposition of the base of the salt in soluion, (3) by chemical changes in the electrodes. Of the first class of electrolytic phenomena, practical applications are made in the manufacture of gases, such as hydrogen, oxygen, chlorine, etc. The second class furnishes a method for the commercial pro- duction and refinement of various metals, among which the im- portant ones are silver, copper and aluminum. In the ca,ses of silver and copper the metals are deposited from a solution of the metals in an acid, while aluminum is depo.sited electrolytically from a fused salt, the action being the same in both cases. Both of these classes furnish accurate methods for determining the strength of an electric current, for the amount of chemical action is always directly proportional to the current, other conditions being the same. A very important application of electrolysis is that included under the third class, namely, the electric accumulator, which has been partly discu,s.sed. No electricity is stored in this device, but the energy of the electric current is changed into potential energy in the electroly.sis of the chemicals which go to form the cell, and this energy may be restored to the circuit by allowing the chemicals to act upon each other as in the primary cell, with a production of an electro-motive force. Copper Electrolysis. In the mining of copper a number of impurities are found in the ore which cannot be removed by roast- ing and smelting. Electrolysis offers an easy means for separating the copper, silver and impurities from each other, and the metals are obtained in a pure state. The copper after smelting is cast into large ingots, which after being surrounded with heavy bag- ging are placed in a solution of copper .sulfate, nearly saturated, and a large electric current is .sent through the liquid from the in- Electrical Machinery. 70 got to a plate of copper which forms the other terminal of the bath. The copper is dissolved from the impure ingot and is de- posited on the plate in a pure state. The electrode or terminal which is dissolved by the current is known as the anode, the other being the kathode, while the intervening liquid is the electrolyte. As only copper is dissolved from the anode, any other metals pre- sent in the ingot will be left behind in the bagging and may be redissolved and electrolyzed for the silver or other metal in a sep- arate bath. After the pure copper has been deposited in this man- ner it is remelted and is cast into ingots of the proper size for rolling into sheets and bars, or drawing into wire. The copper obtained in this manner is finer than the best copper obtained in any other manner and ordinary copper is now about as fine as the standard samples used by the physicist Matthieson when he made his famous determinations of the specific resistances of metals. For this reason copper is sometimes spoken of as 102% fine. A current of one ampere will liberate .0003307 grams of copper per second, or in other words this quantity is deposited per coulomb of electricity passing, and this constant is called the electro- chemical equivalent of this metal. Aluminum Electrolysis. The commercial production of al- uminum is an important electrolytic process. This metal cannot be smelted from its ores, which are very common, as these are very refractory. The oxide of aluminum, known as alumina, (AljO,), is the .source of the metal. This alumina is formed from Bauxite which is found in some of the southern states and it is prepared as follows : The ore is crushed in a stamp mill and treated with so- dium hydroxide and the aluminum is thus dissolved. The in- soluble residue is discarded and the carbon dioxid is forced through the sodium aluminate, forming sodium carbonate and precipitating aluminum oxide, which after washing and drying is shipped to the smelting works in the form of a white powder. The alumina is placed in a carbon lined iron box, in which a bath of fused cryolite is kept liquid by the heat due to the resist- ance offered to the flow of a current which passes through the box from large carbon rods in the center to the carbon lining. When the alumina is thrown into the bath it is reduced at the negative terminal, or lining, and the high temperature of the bath keeps the reduced metal in a molten state. It is removed from this box in a practically pure state. Aluminum has a number of uses in the arts, principally in con- Electrical Machinery. 71 nection with other metals in the form of alloys, which possess peculiar properties. For example aluminum bronze is one of the hardest metallic substances with the exception of some of the rare metals. Electrolysis of Pipes. An unfortunate application of electro- lysis is found in the effect of the stray current from the return rails of electric street railway lines. Naturally the current takes the easiest path to return to the generator and this is often found in metal pipes which run in the same general direction as the rails. To produce electrolytic action it is necessary to have an anode, a kathode and an electrolyte and these conditions are met in the above case by the pipes, the rails and the chemicals present in the moist soil. The loss of metal occurs where the current leaves the pipes so that if they are po.sitive with respect to the rails they will be eaten away. Five grains per ampere-hour is an average value of the consumption of the anode, the exact amount depending on the minerals present in the soil, chlorides, nitrates and sulfates being the most active in the order named. Two methods for the prevention of this trouble are possible : either by keeping the current away from the pipes, or by making them a part of the return circuit. The first could only be done by coating the pipes with an insulating substance, which would be a troublesome arrangement. The second method is the one in use and the pipes are connected in such a manner that the current neither enters or leaves them except through a joint of low elec- trical resistance. If the pipes are kept at the same or at a lower potential than the rails there will be no injurious effect on them, for no decomposition takes place where the current enters a metal. Electrical Machinery. 72 THE TRANSMISSION OF INTELLIGENCE. Two classes of electrical apparatus will be considered under this head, the telegraph and the telephone. The original telegraph instrument, as its name indicates, was intended to write at a distance, and this was accomplished by- means of the attraction of an electro-magnet for its keeper or arm- ature when excited by an electric current. The movement of the armature was used to operate a stylus, which either wrote upon a ribbon of paper or indented it, the paper meanwhile being fed through the apparatus by suitable clock-work. These machines are still used to a slight extent, but they have been replaced for land use by the ' ' sounder, ' ' which simply makes a series of clicks instead of writing. . The name telegraph is not strictly accurate, but it still clings to the apparatus. The following parts comprise a telegraph system: (i) the sending apparatus ; (2) 'the transmitting devices ; and (3) the re- ceiving apparatus. The sending end of the line requires only a key or switch which will make and break the circuit as desired. The ordinary kej' or lever key, as shown in figure r, makes contact at the point c on being depressed and thus completes the circuit through the line. A short-circuiting switch sw, is closed when the instrument is not in u.se, so that the circuit is complete for the other operators. In automatic and rapid telegraphy, which is not used much in this country, the contact is made through perfora- tions in a strip of paper. Holes are punched in the strip by the operator and when this strip is fed through the sending machine, a metallic contact, which rests on the strip, passes through the perforations and completes the circuit and thus sends an impulse through the line. This machine is capable of very high speeds. Transmission. A battery of primary cells is needed to supply the current for these in.struments and as the circuit is to be closed nearly all the time, a cell is needed which will not " polarize " when the current flows. The blue-stone or gravity cell answers these conditions, for it works best when kept on a clo,sed circuit, hence its name, closed-circuit cell. A single wire is used to trans- mit the current for a telegraph line, and all the instruments are Electrical Machinery. 73 connected in series. The ground is used for the return circuit and but one operator can use the line at a time in the usual system. On long lines an excessive battery power would be required if but a single battery were used, for the resistance of the line is high and each instrument requires a considerable current. As the resist- ance of each instrument is considerable, and as they are connected in series, a high voltage would be required to send a current through them. In order to economize wire in the line and to make the operation more satisfactory, the separate instruments are worked through relays. The relay (.see Fig. 3), consists of an electro-magnet with an armature adjusted to move when a very slight current is sent through it. This armature then acts as a key and completes the local circuit when it is attracted to its core. The local circuit contains a small auxiliary batterj of one or more cells and the sounder, or other receiving instrument. By this relay system the large battery on the line is done away with, and on account of the smaller current required a much smaller wire may be used. The receiving apparatus consi.sts usually of a sounder, as in figure 2, but it may be a writing or printing device known as a recorder. The sounder is made to click with one sound on being attracted, and with another when it is released, so that the time which elapses between these two sounds is a measure of the time during which the sending key has been held down. This interval makes the distinction been the dots and dashes. The armature sometimes carries a ,st3'lus which marks the dots and dashes, while in the apparatus used on cables, the current deflects the moving coil of a galvanometer to which is attached a capillary tube, which traces a sinuous line on a paper ribbon. The sending key in this case is double, one key sending the current in one direction, while the other sends it in the opposite direction. The galvanometer coil is thus deflected either to one side of the zero position or the other, according to the direction of the current. The fact that the ca- pillary tube is usually made in the form of a syphon, the upper end dipping into a vessel of ink, gives the name syphon recorder to this instrument. The Morse Alphabet. As the only available motion in the Morse apparatus is such as to produce clicks on the sounder or indentations in a strip of paper in the recorder, the alphabet must be arranged according to the time which the armature carrying the sounder arm or recorder stylus remains in contact with the Electrical Machinery. 74 core. This leads to the use of two signals, a dot and a dash which is three times as long. Another and still longer dash is used for one letter (1). The alphabet is changed slightly to suit the syphon recorder, and a twitch of the capillary tube in one di- rection (above the center line) represents a dot and one in the opposite direction a dash. In case of such long cables that the current is too weak to operate the syphon recorder, a reflecting galvanometer is used and a spot of light is allowed to be thrown upon a screen, and by this means the twitchings of the needle are observed. The Duplex System. As the erection and maintenance of a long line is a considerable expense, various devices are used in order to make the line as efficient as possible. One plan already mentioned was to send the messages very rapidly by a special ma- chine. Another plan is to send more than one message at a time on the same wire. Of the various plans for doing this the sim- plest one will be described. This system can be applied to the quadruplex system al.so, but this becomes quite complicated. Figure 4 shows the connections for one duplex system. The sounders, 5", and S ^ are each wound with two coils in opposite directions, or differentially. One of these coils on each instru- ment is connected in series with the line, the keys. A', and K^, the batteries, B^ and B^, and the earth. So far these are the usual connections. The back contact of each key is connected to the second coil of the sounder through the resistances, R^ and R^, each of which is equal to that of the line plus one-half of the sounder coil. If both operators wish to use the line at the same time they will depress their keys and block the line, for the bat- teries will oppose each other. The local circuit of each .station will, however, be complete through the resistances and the .sound- ers which will both click just as though the currents had passed each other on the wire. Suppose that but one operator wislies to use the line, on depressing his key the current will divide in his sounder, the two parts going around the magnet in opposite di- rections so that there will be no effect. On the other end the cur- rent will go to earth through one-half the sounder coil and this will be operated. In this way the one wire will give the same re- sults as if two had been used. Electrical Machinery. 75 The Telephone. The classification of apparatus is the same in this case as before. The sending apparatus consists of (i) the microphone; (2) the induction coil; (3) the battery; (4) the call bell. The transmission takes place through a pair of wires, for a ground return is never satisfactory. At the receiving end the receiver or telephone proper changes the electrical oscillations into sound again. The first apparatus u.sed by Bell for transmitting sound con- sisted of a diaphragm of thin iron mounted near the end of the bar magnet, which was surrounded by a coil of fine wire as in figure 5. When the diapliragm was .set in motion by the vibra- tions of the air which go to make up the sound, it changed the reluctance of the magnetic circuit of the bar. The induction in the magnet was thus altered and an electro-motive force was set up in the coil surrounding the bar. By connecting a similar ap- paratus at the other end of the line, the coils of wire being in series, the alternate weakening and strengthening of the magnetism of the bar by the current surrounding it caused vibrations in the second diaphragm similar to those in the first. This device is still used for the receiver. In this case the amplitude of the vibrations in the receiver were very much smaller than those in the transmitter, on account of the lo.s.ses in transformation and transmission, ,so that, except for short distances the sounds were barely audible. An amplifying device was necessarj' and this led to the invention of the micro- phone by Hughes and others. The microphone depends for its working upon the variable re- sistance of carbon contacts when subject to varying pressures. (See figure 6. ) An electric current meets le.ss resistance when the contacts are under greater pressure. By so placing one or more carbon contacts in an electric circuit so that the resistance of the joint would be disturbed by the vibrations of the air, a device was obtained in which the electric current in the circuit bore some relation to the amplitude of the .soimd wave. Two forms of mi- crophone are common : one in which the jar produced by the sound varies the contact, and one in which the carbon is in con- tact with a vibrating dissc by which the actual pressure on the carbon contact is varied. This latter is the most successful form for transmitting speech. A telephone when placed in .series with a battery cell and such a microphone, will be effected by the varying current which flows Electrical Machinery. 76 through the circuit. The effect on the telephone will, however, be greatly amplified if the microphone be connected in series with the primary circuit of an induction coil the secondary of which is connected to the telephone. (See I, figure 8.) As the secondary circuit contains a much greater number of turns than the primary, the electromotive force is raised and the telephone coil can be wound with many turns of fine wire, and a large magneto-motive force will be thus obtained. The microphone is made in various ways, but the underlying principle is that the vibrations of the diaphragm shall vary the pressure between one or more carbon contacts. This carbon is sometimes in the form of rough buttons, and in other cases it is powdered. The resistance of a telephone circuit is ordinarily high so that the ordinary call bell cannot be used to send signals over the line. This work is done by means of the magneto generator and bell. A pair of bobbins is rotated by means of suitable gearing in front of the poles of a powerful permanent steel magnet. (See G, fig- ure 8.) There is thus generated in the bobbins an alternating current. The current is received in an electro-magnetic bell as is shown in figure 7. The magnets and the armature are kept po- larized by the permanent magnet N S. The windings on the magnet are so arranged that when the current flows in one direc- tion it weakens one magnet and .strengthens the other and the armature is attracted to the strongest core. This operation is re- versed for every period of the current, and the hammer h vibrates in uni.son with the current. These magneto bells are made to ring through as great a resistance as 30,000 ohms, if required, but they are usually wound for 10,000 ohms, and they form a convenient apparatus for many kinds of electrical testing. An important feature in a telephone set is the switch for con- necting the various parts in circuit at the proper time. The call bell must be in circuit when the telephone is not in use, and it must be cut out when the telephone is being operated. The dia- gram shows the arrangement of the various parts. The current enters by the line which is connected to the light- ning arrester and short circuiting plug A. Wire No. i then passes to the hinge of the switch 5, by which the current is trans- ferred to the bell or the receiver. The weight of the receiver holds the switch down when hung upon it, but when the receiver is removed the spring draws the blade of the switch into contact Electrical Machinery. 77 with the upper jaws. When the switch is down it is in contact with jaw No. j which is connected to the magneto machine, but as this is short-circuited by the switch sp, the current passes through it to the bell and thence to the line. When the main switch is up, contact No. 2 connects the micro- phone M in series with the battery B, and the primary of the in- duction coil /. It al.so connects the receiver in series with the secondary of the induction coil, and with the line so that the ar- rangement is complete both for sending a mes.sage and receiving one. The only remaining connection to be noted is that in the magneto ma- chine As the self-induction of the armature coils is high they must be thrown out of circuit except when the machine is being used to ring the bell on the other end of the line. It is cut out of circuit by the short-circuiting switch. The handle and arma- ture are movable in and out, being normally kept short-circuited by the spring sp. When the machine is to be used the handle is pres,sed in,, the armature disconnecting from the switch sp and it is then in the main circuit. Releasing the handle allows the ar- mature to come back to its normal short-circuited position. ELECTRICAL MACHINERY. Plate IX. Tya. Electrical Machinery. 78 ABSTRACTS FROM THE UNDERWRITERS' RULES. The fire insurance companies, through their board of under- writters, lay down certain rules for the guidance of parties install- ing electrical apparatus in insured buildings. All electrical work done is subject to the inspection of the insurance companies and their rules cover mainly methods of wiring and of mounting ap- paratus. Installations are divided into the following classes : A. Central Stations for light and power. B. High Potential Systems. Over 300 volts. C. I