LIBRARY OF CONGRESS. Shelf..'JyL5- UNITED STATES OF AMERICA. r- THE PRACTICAL APPLICATION OF y^^SWMllMMs^ Dynamo Electric machinery. ^ y C. K. MacFadden and Wm. D. Ray, 2 ^^"2 V > "^2^ PUBLISHED BT DATE & RUGGLES, 189 Washington Street ^ Chicago^ iLLr Index. CHAPTER I.— KI^EME^NTARY DATA - - 7 Units of Klectrical Measurement. Ohms Law. CHAPTER II. — MAGNETISM AND INDUCTION - I5 Methods of Current Generation. CHAPTER III. — THEORY OF DYNAMOS - - 20 Action of Commutators. Methods of Dynamo Control. CHAPTER IV.— CURRENT DISTRIBUTION - - 41 Losses in Copper Conductors. Fuses and Safety Cutouts. CHAFrER v.— TRANSFORMERS - - - - 62 Construction and use. Alternating Current Distribution. CHAPTER VI.— TYPES OF DYNAMOS - - - 76 Direct and Alternating Current. Their Application in Practice. CHAPTER VII,— CAUSES of troubi^e in dynamos 88 Their Remedy and Prevention. Methods of Testing for Faults, etc index: CHAPTER VIII.— ARC I.AMPS - - - ~ lor Direct and Alternating Current. Various Types and Makes. Incandescent Lamps of Various Types, CHAPTER IX. — BI.KCTRIC MOTORS - - - 121 Transmission of Power Dynamos. Direct and Alternating Current, Various Types of Motors. Street Car Systems. Methods of Motor Control. CHAPTER X.— STORAGE BATTERIES - - - I39 Various Types. Their Care. Directions for Charging. CHAPTER XI.— EI.ECTRIC HEATING - - - 154 Electric Cooking. Electric Welding. Electric Metal Workings. Station Instruments, Switchboards. HORSE POWER EQUIVALENTS - - - 165 COPPER WIRE DATA ----- 166 Copper Wire Table. Safe Carrying Capacity. Preface. Our aim in bringing out this little book has been to reach a class of readers who, realizing the need of a gen- eral fundamental understanding of the application of Electricity, will read with some benefit, we trust, a few descriptions of the modus operandi of the most generally used class of Electrical Machinery, It has not been our intention to take up the subjects treated on, in any but the most simple and as we believe, the most easily understood way. It is becoming more necessary each year for the well qualified steam engineer to be somewhat familiar with Dynamo Electric Machinery in order to advance in his calling. A partial understanding at least is now or soon will be almost a necessity for those engaged in nearly all branches of engineering work. There is hardly a pro- fession which electricity in some way has not entered. The vast majority of the men in charge of our practical work have never had the advantage of a technical educa- tion and are therefore unable to follow the advances that are so rapidly being made. The volt, ampere, and ohm and their relations to each other are the first stumbling blocks and the cause is easily seen by inspecting a few books for definitions of these words, We have endeavored to impress as much as possible, the formula expressing Ohms Law, E R on which all calculations necessarily take their ^tart. A 6 PRKF'ACK. thorough understanding of the relations of the volt, ampere and ohm to each other, is without doubt the foundation of all electrical knowledge. We have also endeavored to keep up to the times and believe we give some interesting descriptions of modem electrical apparatus, which will be of value to those whose main source of light on electrical matters come from cat. alogues and newspapers. The dynamo tender, unless partially conversant with the principles on which his machinery operates, will often be perplexed at even the most trivial troubles to which a dynamo is subject. A motorman on our usual electric street cars, could often lessen motor repairs to a great extent by obtaining even an elementary understanding of the motors' action. An understanding of the proper dis- tribution and installation of electric wires would be the means of averting many thousands of dollars loss each year by fire from electrical causes. There are many good books on the various branches of electrical work but they are too often of such a technical nature as to bar the uneducated reader from obtaining much benefit from them. We hope that a close study of the following pages will place the average beginner on such a foundation as to make the other more complete • electrical books more easily understood. Cari, K. MacFadden. June I, 1894. Wm. D. Ray, :^I,KMENTARY DATA. CHAPTER I. ELEMENTARY DATA. What is electricity ? A question often asked and prob- ably never as yet clearly answered. Those interested in the practical field, find it almost impossible to keep up to the times, in regard to the laws that govern its generation and control. We know that by means of certain combinations of coils of wire and magnets or by means of chemical action, we can produce, we may say, electricity. We must be content if we master a few leading laws ^governing the generation and application of the Electric current. Let the Scientists and Philosophers discuss the question as to what electricity is. In dealing with the simple electric current that is generated by dynamos, batteries, etc., we will find that there are several broad and easily understood laws that govern its practical applications. These laws hold good in all cases and un- der all conditions, and should be thoroughly understood by anyone desiring to learn even the first and most sim- ple effects of a flow of electric current. Probably the easiest way to understand this law will be to take a simple case of a pump connected to a loop of water pipe. The pipe is filled with water and it is evi- dent that if the pump is started there will be a circulation of water from the pump through the pipe and back into 8 KI.EMKNTARY DATA, the pump again. The pump furnishes the power to move the water in the pipe — and it is evident that the water moves through the pipe owing to the pressure exerted by the pump on the water. In (figure i) an open tank, (d) is shown into which the water flows from the pipe. The pump takes the water from the tank to keep the pipe filled, and the speed of the water through the FIGURE I.— PUMP FORCING WATER THROUGH PIPES. pipe and therefore the quantity of water passing through the pump and pipe in a minute of time, will depend on the pressure given the water by the pump. If we double the pressure of the water and the friction or resistance to the flow of water in the pipe remains constant, the quantity of water handled by the pump will be doubled. In other words by increasing the pres- sure, we increase the quantity of water passing through the pipe at exactly the same ratio. Now if we found, that a guage placed at (a) would have to register 50 lbs. pressure to make 100 gallons of water pass through the pipe in a minute, it is evident that if the friction in the pipe remained the same, that 100 lbs. pressure ought to put 200 gallons through the pipe. It is also evident that KIvKMKNTARY DATA. 9 if loo feet of pipe a certain size offers a definite amount of friction, that twice the length of pipe would have two times the friction or resistance to the flow of water that the loo feet has. Thus to put loo gallons of water a min- ute through a pipe will take %, the pressure required to force 200 gallons through the pipe. This example may serve the purpose of illustrating the principle of the flow of current from a source of electricity. We will let the dynamo take the place of the pump, which will generate the pressure to send the electricity through the circuit which may consist of lamps, etc., connected by means of conducting wires. The friction in the pipe is represen- ted by the * 'resistance' ' of the wire and circuit, and the amount of water used, represent the quantity of the cur- rent of electricity. Instead of using the pound as the unit of water pres- sure we will use the term ' Volt' ' which is the unit of electrical pressure. We also have a term which denotes the unit of * 'resistance, which is the equivalent of ' 'fric- tion" used in the illustration ot the pump and pipe. The unit of resistance is called the ' 'ohm. ' ' Then lastly the quantity of current in electricity is measured by the unit of current quantity, the "am- pere." This is the quantity of current that a pressure of one volt will force through a resistance of i ohm. The resistance of a conductor of electricity varies not only with its size or cross section but also with the ma- terial of which it is made. Silver, when pure, is the best conductor of electricity known, but copper, when pure, nearly approaches silver and is so much cheaper that it is used in nearly all cases to distribute current for practical purposes. lO EI.EMKNTARY DATA. The metals in their order for conductivity are as follows: Silver, Copper, Gold, Aluminum, Zinc, Platinum, Iron, Ivead, German Silver, Platinum Silver alloy and Mercury. In practice the wires or conductors to carry current are either designated in size by their diameter in thousandths of an inch (or mils) or by the sectional area or cross sec- tion of the wire expressed in circular mils or by the size .in number, as measured by the Standard American or Brown & Sharp Wire Guage. A circular mil is a circle iq^qo i^ch in diameter. As in the case of pump, the higher the pressure in volts at the dynamo, the larger quantity of electricity (expressed in amperes) will be put through a circuit which has a resistance (expressed in ohms) to the flow of current. Thus if I volt pressure will force i ampere of current through a circuit having i ohm of resistance, it will take 5 volts to force 5 amperes thiough this same i ohm of re- sistance and if this resistance is increased to 2 ohms, the pressure would have to be 10 volts to force 5 amperes of current through it. It will be seen that these terms are dependent on each- other and their relation to each other is expressed by what is known as Ohms I^aw which is expressed: Current Pressure in Volts B in = or C = — Amperes Resistance in Ohms R **C" standing for current, *'E)" for electro-motive force or volts, and '^R" for resistance expressed in **Ohms." This relation C=E/ R must be remembered for it is the fundamental law of the governing of electric currents, and is used as a foundation to obtain all of the more com- E:i,EMEN'rARV DATA. tl plex formulas known to the Electrical Engineers. Take the simple case of a certain make of incandescent lamp, the resistance of the lamp in question is found to be 200 ohms The pressure of the circuit on which this lamp is designed to run is 100 volts, and according to Ohms law the current in amperes which 100 volts pres- sure will force through 200 ohms of resistance is ^gg or Yz which is the number of amperes such a lamp would allow to pass through it if current at 100 volts pressure was applied. It will thus be seen that it is a very easy matter to ob- tain any one of these quantities, provided we have the other two given, by a simple multiplying or dividing of the two known quantities. The relations to each other are expressed E E C= — , E=R X C, and R= — . R C There is still another term with which the practical man is brought in contact and that is the unit of power, the * 'watt. ' ' This watt is the power represented by the passing of i ampere of current through i ohm of resis- tance and can always be obtained of any current by mul- tiplying the number of volts by the number of amperes. Thus in the incandescent lamp before spoken of, the watts used by the lamp would be 100 (volts) x>^ (amp) = 50 (the number of watts), this being the amount of elec- trical energy necessary to be applied continually to keep the lamp burning. Such an incandescent lamp would be termed **a 50 watt incandescent lamp.*' An arc lamp which needed a current of 10 amperes at a pressure of 50 volts to keep it in operation would be termed a * '500 watt arc lamp". 12 ELEMENTARY DATA. There are two entirely different methods of distribu- ting current to lamps etc., connected to dynamos. The plan illustrated by means of the pump in (figure I) may be seen as applied to a dynamo and lamps in (figure 2) (a), in which the dynamo D supplies current for lamps L and is known as the Series System. It will be seen in plan (a) that the lamps are connected in "series" that is, the current which passes out from the positive (+) D D D 0) (!) (!) 0) (!) Q O Q Q Q Q ^ revolution, the current flows in such a direction as to make the brush (+) a position brush, that is, the current flows from + to — , i^igure; 6.— awkrnating current wave. the current generated depending on the rate at which lines of magnetism are being cut by the loop of wire. The current with the loop at the line (A to B) w^ill be zero, for this is the point at which the current is revers- ing its direction in the loop, owing to the fact that THEORY OF DYNAMOS. 25 the direction in which the loop cuts the lines of mag- netism is being changed. From this position on dotted line A — B, the current will gradually rise to a maximum when the loop is on the line C — D, which is the point at which a certain given movement of the loop will cut the greatest number of lines of force. The rise and fall of current in an alternating current circuit may be shown readily by the cut in figure 6. The line A — B repre- senting the zero line or line of no flow of current in the coil. The distance from i to 2 represents one complete revolution and the curved line C represents the current produced by the revolving loop. The portion of the curved line above the zero line represents current flow- ing in a positive direction and the portion below the line will represent the negative flow of current. The total curve from i to 2, represents one complete alternation, which in the combination shown in figure 4, means one revolution. If we had the coil making 2000 revolu- tions per minute, there would be 2000 of such waves as shown in curve i — 2. With a commutator such as shown in figure 5, and with the brushes placed as shown, it will be seen that just as the current in the coil is at zero the commutator has moved in such a position that the brushes are just changing from one segment of the com- mutator to the other, thus keeping the rising side of the loop connected to the negative brush and the down- ward moving side of the loop connected to the positive brush. In this way we can send all the impulses or waves of current from the revolving coil on the circuit in one direction, thus producing a pulsating current, but at the same time one whose flow is always in one direc- tion. The current curve of such a dynamo will now be 56 I'HEORY OF DYNAMOS. such as shown in figure 7, in which it will be noticed all of the waves of current are above the zero line A — B. This type of current is known as a direct current of pul- sating character. As before stated the generator, or dy- namo just spoken of, is of the simplest possible form and to make large dynamos for supplying continuous direct current in an economical manner such a primitive dyna- IflGURE 7. — PUI^ATING DIRECT CURRENT WAVE- SINGLE coil. ARMATURE. mo as shown, must be greatly improved. In chapter 1, we have spoken of electro-magnets as being the only practical form for large and powerful magnets, and we will find that all field magnets for large dynamos are of this type The armatures of large dynamos, instead of having but a single coil, will often have 100 or more coils, each con- sisting of one or more turns, for if a single turn coil will generate, for instance, i volt, when cutting the magnetic lines at a given rate, a coil of 10 turns of wire in it will generate 10 times the pressure that the i turn will, or 10 volts. And as has been explained, if a piece of iron be placed in a magnetic field, a large number of the lines of THEORY OF DYNAMOS. 21 --B FIGURE 8. — F*I,OW OF MAGNETISM THROUGH A RING ARMATURE CORE. I I FIGURE 9.— FtOW OF MAGNETISM THROUGH A DRUM ARMATURE CORE. 28 THEORY OI^ DYNAMOS. magnetism will take to the iron in completing their in- dividual circuits, and so it has been found advisable to wind armature coils on an iron core, so that the largest possible number of lines of magnetism flowing from the poles of the field magnets, will flow through the iron ar- mature cores, and in this way, the coils of the armature will cut a larger portion of the lines given out by the field magnets. The reader will easily understand, from the previous description, that the current given out from a single coil armature may be a direct current, but still of a pulsating type, there being in the cases shown, two impulses in each revolution. There are many cases where such a pulsating current would be nearly as objectionable as an alternating cur- rent. To overcome this trouble and to also make a dyna- mo whose efficiency is high enough for practical work, has taken an immense amount of study. To make the principal used to obtain a continuous current, very clear to the reader, it will be well to take up the case of a **ring" armature on which a single coil is wound. From figure 8 and 9, it will be seen that nearly all the lines of magnetism shown between the pole pieces take the iron path in preference to the air. In the case of the iron ring shown between the pole pieces, N and S, the lines practically divide on the line A — B half taking their path by way of the upper half of the ring, and the remaining half through the lower portion of the ring. Thus with a coil placed as in figure 10, it is evident that practically only the wire on the outer face of the ring will be cutting the lines of magnetism as they pass from the pole pieces to the ring. The coil will generate a THKORV Olf DYNAMOS. ^9 current while revolving from C to D, the line C— D being the neutral line, or line of commutation, which is the point at which the current will reverse its direction in the moving coil. This is the simplest form of ring armature, a step in advance is the adding of a coil BtGURE lO.— RING ARMATURE OE ONE COIL. on the opposite side of the ring, and connecting them ill multiple, that is, the current generated by one coil has added to it the current of the other coil, which add? whatever current it may be generating, to that of the original coil. In this case, the amounts generated in 30 THEORY OF DYNAMOS. two coils will be equal, for when the coil on one side of the ring is generating current, the coil diametrically op- posite must also be generating a like amount, and when connected in multiple as shown the total result at the brushes will be the sum of the two eflfects. See figure II. It will be evident also, that while the coils are FIGURK II. — RING ARMATURE. — TWO COII,S IN MUIvTlPLE. moving past the line C-D, fig. 8, they generate no current, since they cut practically no lines of magnetism, and that if two coils were placed on the ring so as to be moving past the line, A — B, they would be generating a maximum amount of current. By connecting these four THEORY OF DYNAMOS. 31 coils as shown in figure 12, we will generate a current having twice as many impulses as in the case of an arma- ture having but one pair of coils in series. The current wave will be represented by figure 13. The line i — 2 representing one revolution. FIGURE 12.— GRAMME RING ARMATURE. FOUR COIL TYPE. This multiplying of coils can be carried on with econ- omy, until we have some dynamos of this type, having hundreds of coils, and giving practically a perfectly con- tinuous current. The proper placing of the armature coils in the mag- netic field between the pole pieces of the field magnets, 3^ O^HKORY OI^ D\:NAMoS. has taken an immense amount of study and experiment^ and we to-day have two general types of armature, the drum and the ring type. The drum armatures are so called from their shape. A cylindrical piece of iron with the armature shaft running through its length from end to end, is covered with coils of wire, which in dynamos FIGURE 13.— CURRENI^ WAVES OE EOUR COIL RING ARMATURE* having but two field magnet poles, are so wound as to form loops similar to that shown in figure 5. Thi^ type of armature, with many coils each of several turns, is the type usually used for incandescent lighting and power Work by direct currents. Representative American made dynamos of this type are the Kdison, Mather, Bddy^ C & C, ancj direct current dynamos made by the West- inghouse Co. The drum armature is sometimes called Siemens armature, from its inventor. Dynamos having ring armatures have been used to a great extent for arc light work and are certainly a very much easier armature to repair than drum armatures, whose windings usually cross and overlap at the ends of the iron armature cores, and thus increase the liability to make trouble from THEORY OF DYNAMOS. 33 short circuits etc., at these points, which often make it necessary to remove nearly all the armature windings to repair the damaged coil. Ring armatures or Gramme armatures as they are often called, may be repaired quickly by removing the defective coil from the ring, and rewinding with a new coil, or in many cases of I^IGURE 14. — OPKN COII< RING ARMATURE. trouble, the damaged coil may be disconnected from the commutator bars, and the two bars to which the coil was connected are then connected by a short piece of wire, and the dynamo will then be able to generate current until repairs are made. Arc light dynamos of the Wood, Excel- 24 THEORY OF DYNAMOS. sior, Schuyler, Ball, Standard and otlier smaller compan- ies, use the ring armature of the Gramme type and gener- ate voltages up to 5000 or 6000, depending on the number of lamps for which they are designed to furnish current. The arc light dynamos of the Brush and Thomson- Houston makes, have armatures of the ring form, but owing to the peculiar windings on them, cannot be called Gramme armatures. They are known as **open coiP* armatures, while all Gramme ring and drum armatures are known as * 'closed coil". This distinction is brought about from the fact that drum and ring amiatures are, as has been shown, connected betv/een windings or seg- ments, of commutator, in such a way as to leave the armature windings alwa^^s connected in a permanent way from coil to coil, whereas in open coil armatures as shown in figure 12, it will be seen that the coils are sep- arate and distinct from each other. In both the Brush and Thomson-Houston dynamos, the armature coils are provided with terminals which alter the connections in such a manner as to let the coils in the most active positions give the bulk of the current, and either cut out the less active coils entirely, as in Brush dynamos, or reduce the resistance of such coils to the flow of current by placing them in parallel or multi- ple arc, and then in series with the active coil or coils. The current from both of the dynamos mentioned, is extremely pulsating, compared to current from the usual ring armature, but owing to the well worked out details of construction, insulation, regulation, and to the relia- bility derived therefrom, both Brush and Thomson- Houston arc dynamos are known world wide, and at present probably furnish more current for arc lamps, THEORY OF DYNAMOS, 35 than all other makes of arc lighting d3mamos combined. Chas. F. Brush, of Cleveland. Ohio, was the pioneer in arc lighting work as the w^orld now knows it, and Profs. Elihu Thomson and Edw. J. Houston were not far be- hind him in pioneer work. FIGUR:^ 15. — SEPARATE^LY EXCITKD DYNAMO. We have taken up the fundamental study of armatures and have spoken of electro-magnets for field magnets, and the various methods in vogue for energizing them will now be taken up. It takes a current of electricity, passing around an iron core or center to make an electro magnet, the power of which will vary with the ampere l^ THEORY OF DYNAMOS. turns, or the product of the number of amperes passing and the number of turns of wire around the iron core. The first dynamos built with electro magnets for field magnets were * ^separately excited," that is, had a separ- ate battery or generator of electricity to furnish current W -:s=. FIGURE 1 6. — SERIES WOUND DYNAMO. to energize them, the plan of connections being shown in figure 15. Then it was seen that the current from the dynamo itself might be used to excite its own field mag- nets and owing to a slight amount of * 'residual magnet- ism", which always remains in a piece of iron after hav- ing once been magnetized, being present in the field THEOIEIV OF D\^AMOS. 37 magnets, this was easily accomplished, as shown in Hgtire i6, and is known as a ^'series winding" and such a dynamo w^ould be known as a * 'series wound" dynamo. The field winding carries the whole current of the arma- ture and is connected in series with it. This winding is I^IGURE 17, — SHUNT WOUND DYNAMO. generally used in arc light dynamos, and others gener- ating a constant current of high voltage. The plan of a shunt winding is shown in figure 17, and is so called from the fact that the winding forms a ''shunt" path around the armature. To make an effici- ent dynamo, the resistance of the shunt winding is made SS I'HKORY OF DYNAMOS, quite high, several hundred times the resistance of* the armature and as a result, the current through the shunt is small in quantity, but the immense number of turns of small size wire in the coils on the field magnets make the necessary number of ampere turns, and thus the resultant magnetism is the same as that produced in the series dynamo with its large current and small number of turns. Shunt wound dynamos are usually used to generate currents of "constant potential'* or ^'constant voltage'* such as is used in operating incandescent lamps, electric motors, etc. Owing to the high cost of wire necessary for shunt windings for dynamos of high volt- age, we will find that practically all shunt dynamos oper- ate at a voltage under 600, in fact, by far the largest number of shunt dynamos operate at a voltage of not over 125 volts. For regulating purposes a rheostat (R) containing resistance vnre, is placed in the shunt circuit, and in this way a practically uniform voltage is main- tained at all loads by varying the total resistance of the shunt circuit, and thus the current through it, which must in turn vary the magnetism of the field and in this way raise or lower the voltage, as the case may require. Still another winding is shown in figure 18, knoviTi as the compound winding, and is often used to make a dy- namo self-regulating It is evident, that on a constant potential circuit when additional lamps are turned on, that the dynamo must respond at once and send out a largei number of amperes to take care of the load. . Under these co::ditions, to maintain the voltage constant, we must increase the magnetism of the field magnets to compensate for the increased output. This may be ac- complished in an automatic manner by winding a large THEORY OF DYNAMOS, 39 portion of the field with a shunt winding, which should be of such strength as to generate the rated voltage of the dynamo when there is no load placed on it. Then the series windings must be sufficient to add enough ampere turns as the load rises, to keep the voltage up to standard or in some "over compounded" dynamos to increase the FIGURE 1 8. — COMPOUND WOUND DYNAMO. voltage slightly as the load increases, so as to compensate for loss in line or feeders supplying lamps, etc. This type of dynamo is largely used in lighting plants having a fluctuating load, and is invariably the type used to gen- erate current for street railway work, where the load is an ever varying quantity, a condition under which the 40 TH^ORV or DYNAMOS. compound wound dynamo is practically the only one which gives satisfaction. There are a number of modifications of these principal windings which are seldom run across and for this reason are not explained in detail; suffice to say, that dynamos can be^ and have been made, which, by means of the prin-^ ciples described, will give a constant current and varying voltage, a constant voltage and a varying current, in both cases the speed being maintained uniform, or as in some cases of dynamos designed to be connected to the axle of a car for train lighting, the voltage remains practically uniform, with a varying current, while the speed alters several hundred per cent. It will be seen that we have various types of armatures and fi!eld magnets with their various windings, and it will be be easy to see that it is possible to build dynamos of almost any size, and for any kind or character of current. CURREN'T DISI'RIBUTION. 4^ CHAPTER IV. curre;nt distribution. In the previous chapters, we have treated of direct and alternating current dynamos, and to a certain extentj their application. In this chapter we take up the meth- ods of distributing current to lamps, motors, etc., all of which are deserving of much study. The dynamo for lighting or power purposes, usuailj^ sends current a considerable distance before it is used in the arc or incandescent lamps, motors, etc. It is evident that it is desirable to have as little loss as |)ossible in power, between the dynamo or generator and the point at which the current is used. For this reason, conductors of copper are used, owing to its * 'conductivity, ' ' that is, its small resistance to the flow of current. But even in the purest copper, there is some resistance, the amount varying with the lengthy and also with the diameter or * 'cross-section'* of the cop- per. If we attempt to reduce the loss to a very small amount, the cost of copper will be high and if there is not enough copper, the loss in pressure will be excessive. To prevent a loss of current from the conductors, froni them accidentally coming in contact with the ground or other conductors of electricity, the wires are insulated from each other and from all connections to the ground. In high potential work, this insulating of conductors 42 CtJRRKN'r DISTRIBUTION. would have to be done for safety to human life, for pres- sure of 600 volts and over are exceedingly dangerous. The distribution of current for series arc lighting is a simple matter, since the current in amperes is constant and uniform in all parts of the circuit and the loss in one portion of the wire circuit will be the same as in any similar length of the same size wire. Thus in calculat- ing losses in the wiring leading to arc lamps in series circuits, the main thing to determine is the total resis- tance of the wire, and, having the resistance in ohms, we easily calculate the number of volts lost in passing the 6, 8 or 10 amperes, as the case may be, through the wire. The loss in volts will be the number of amperes, multi- plied by the number of ohms or, expressed in symbols of ohms law, K=CXR. Thus, on an arc light circuit 10 miles long, consisting of No. 6 B & S Guage Wire, which we may see from con- sulting the wire table in the back of book, has a resis- tance of practically, 2 ohms per mile (2.088) that with 10 amperes of current flowing, that the loss in volts, per mile, will be 10X2 or 20 volts, 10 miles would thus be 200 volts, which is the pressure required to force the cur- rent through thd wire circuit, this being independent of the number of arc lamps in series, each one of which adds from 45 to 50 volts to the 200. Thus a circuit, 10 miles long, of No. 6 wire and 50—50 volt lamps connected in series on it, will take a total electro-motive force in volts of 200 (line resistance) + 2500, which is the total voltage required for the lamps themselves (50X50) which makes a total of 2700 volts required to force 10 amperes through the circuit with its lamps. The loss in volts being 200, and the total voltage necessary to operate the CURRENT DISTRIBUTION'. 43 lamps on such a circuit, being 2700, it is evident that the per cent, of loss on such a circuit will be Vo^o^> ^^ nearly 7/4%, which in practice would not be considered exces- sive. If No. 4 B and S wire were used in place of No* 6, the loss would have been only 130 volts or 5 % loss, but the extra cost of copper wire provided with a good rub- ber insulation, would have been nearly |8oo over a No. 6 wire and the extra loss in current is not enough to pay for putting up a No. 4 wire. Smaller wire than No. 6 can hardly be recommended, however, on account of the increased trouble in keeping up a long line of small wire which is likely to be broken easily by sleet, wind, etc. ^ Incandescent lamps are sometimes connected in series in the same manner as arc lamps; but the current will usually be found to be less than 5 amperes, although there are some series incandescent lamps made to run on 10 ampere circuits in series with arc lamps. Owing to the danger connected with the handling of such scries incandescent lamps, due to the high voltage on which they usually operate, they are not in very general use and are being discarded more and mure each year for indoor illumination. It will be seen that in any series circuit that if the cir- cuit be broken at any point that it will stop the flow of current through all the lamps connected and for this reason all arc and incandescent lamps designed for series work are provided with ' 'cut out' ' which preserves the circuit in case of trouble with an individual lamp, so as to allow the remaining lamps to operate. In arc lamps on series circuits, the cut out ''short circuits," the lamp in case of the carbon being consumed or broken or in ease of a carbon rod in the lamp, sticking or * 'hanging 44 C^XJkRK^J^I' DiSI'kliBWioN. up/^ In the case of incandescent circuits, there is lisu^ ally provided a socket for the lamp in which is a cut out, designed to preserve the continuity of the circuit in case from a lamp being broken or removed from its socket. Arc lamps for series circuits are nearly always operated on direct current dynamos. The series arc lighting sys^ tern (alternating current) of the Westinghouse Company, probably being the oaly exception. The second and without doubt the most generally used plan of current distribution for either power or illumin- ating purposes, is that by means of the constant poten- tial dynamo and a multiple or multiple series system of distribution. Practically, all incandescent lighting, all distribution of current for power purposes and quite a portion of recent arc lighting plants are furnished with current from constant potential dynamos of either alter- nating and direct types. To distribute current at a constant potential or voltage^ great care must be exercised in designing the plan of wiring to be used, for it is a very necessary thing to have the pressure in volts as near a constant quantity as possi^ ble. This will be found especially the case in incandes- cent lamp installation. A slight rise of voltage above that for which the lamps are designed, will decrease the life of the lamp to an alarming extent. A slight reduction in the voltage, will increase the life to a great extent but the light given out by the lamp will decrease so much as to be unsatisfactory, The method of calculating the size of wire for constant potential distribution may be easily understood after a study of the relation of size of wire to its resistance. A copper wire 98% pure, which is CURRENT DISTRIBUTION. 45 circular mil. in cross section, will be found to measure 10.355 ohms per foot of length, at a temperature of 20° Centigrade or 68° Fahrenheit. Knowing the resistance of a wire one mil in diameter and one foot long to be 10.355 ohms, we may then calculate the resistance of any wire, provided we know its length in feet and area in circular mils. On a foot length, a wire 2 circular mils in cross section will have but half the resistance of the wire having one circular mil area or 5.1775 ohms per foot length. The smallest wire usually carried in stock by dealers in wire for magnets, etc., is 25 C. M. in area and is known as a No. 36 wire and has a resistance of .4142 ohms per foot of length or ^ as much as a wire i CM. in area. The smallest wire used in wiring for incandescent lamps and other electric light distribution (No. 16 B. & S.) has an area of 2583 C. M. and a resistance at 68° Fah- renheit of .004009 ohms per foot of length. A No. 6 B. & S. copper wire which has been spoken of as a very largely used size for distribution of arc light current on series circuits, has an area of 26,250 C. M. and a resistance of .0003944 ohms per foot. The temperature of the wire has an effect on the resistance of the metal of which it is made. Copper wire increases its resistance as the tem- perature rises, but for ordinary conditions the rise is so slight that it need not be considered. Knowing the resistance of a certain size wire in ohms per unit length and the distance to lamps or motors from the source of current, we may easily calculate the loss or drop in volts with a given current in amperes passing, by means of the equasion, V=CXR, or, the volts lost will equal the num- ber of amperes multiplied by the total resistance, ex- pressed in ohms, of the copper wire. It must be remem- 46 CURRENT DISTRIBUTION. bered that we must look at the loss in the wiring as a distinct and separate expenditure of power, which is entirely independent of that taken by lamps, motors, etc to which the wiring conveys current. We have shown how the loss in volts may be calcu- lated, provided we know the total resistance of the con- ducting: wires and the current passing through them. The condition most usually encountered is that where the maximum number of volts available to overcome con- ductor resistance is known, and also the distance to the lamps from the source of supply. The unknown quan- tity is the area or size of the wire necessary to carry the amount of current needed at lamps, etc., with the loss in volts decided on. On electric light plants, for example, where no volt incandescent lamps are used, we will find that tbe volt- age, at the dynamo, will probably be from 115 to 125 volts, depending on the distance from the dynamo to the lamps, the difference between no volts and the voltage found at the dynamos, being the number of volts used to overcome the resistance of the conducting wires. Dyna- mos for supplying direct current for constant potential work are usually shunt or compound wound and by means of a rheostat in series with the field magnet cir- cuit, can have their voltage raised as the load increases, so as to maintain a uniform voltage at the lamps. This energy or power lost, shows itself in heating the copper conductors and of course is a loss which must be made as low as possible without an excessive outlay for copper wire. The loss in watts in a given conductor varies with the square of the number of amperes passing through it Thus, in a conductor having i ohm resistance, 10 volts CURE.ENX DISTRIBUTION. 47 will pass 10 amperes. The loss in watts being the pro- duct of the number of volts and amperes, 10X10 or 100, which is the number of watts used in such a conductor when 10 amperes are passing. If 20 amperes were then passed through the same conductor we would find that it took 20 volts pressure to put it through. The watts used being now 20X20 or 400, or 4 times the power that 10 amperes required. It has been mentioned that this loss shows itself by heating the conductors and in this connection it must be stated that but a slight rise in temperature can be allowed, on account of the danger from fire at points in buildings where the conductors pass near wood etc. If a conductor is enclosed in an insulating covering, its radiating capacity is reduced, for as a general thing, in- sulators are poor conductors of heat. Thus, after a great deal of experimenting under various conditions, a table was made, showing the * 'safe carrying capacity' ' of copper wires of various sizes, (see table in back of book). The carrying capacity given in this table is that allowed by the Board of UnderwTiters for wires used for interior work. We can see that there are two limits between which we must work. The wires must not be allowed to carry more than their safe carrying capacity, in which case we will probably find that the per cent, of loss would be higher than it should be, nor can we increase the size of our conducting wires to any great extent over that abso- lutely necessary, without making the cost of copper excessive. The fundamental elements of the case now having been explained, let a practical case be taken. Let the 48 CURRENT DISTRIBUTION, dynamo D, figure 19, designed to furnisli a constant potential current, be connected by wires to 100 incandes cent lamps in each of two buildings, 1000 feet distant- The incandescent lamps are to be made to give 16 c. p at 1 10 volts pressure. The wires from the dynamo to the * 'centre of distribution" will be called ''feeders". The centre of distribution being the point at which the feed- ers are connected to the ' 'mains. ' * T" Fee^'t'5 <^yf PreJ^urc Iodic At*" FIGURE 19. — PI.AN OE CURRENT DISTRIBUTION SHOWING FEEDERS, MAINS AND PRESSURE WIRES. When the buildings are reached, the wiring then con- sists of * 'mains' ' and the service wires, or tap circuits from the mains on which the incandescent lamps are placed. Thus the distributing system of such an incan- CtJURENT DISTRIBUTION. 49 descent ligM plant will consist of feeders, mains and the branch circuits to the lamps to the mains. The aim will be to keep the voltage at the mains, constant and uni- form, the pressure in this case being about 112 volts, and thus allowing for about 2 volts loss at full load, between the mains and the lamps themselves. The dynamo will generate a maximum of 125 volts, thus giving a maxi- mum voltage to be used in overcoming the resistance of the ''feeders" at full load, the difference between 112 and 125 or about 13 volts, which is about 10% of 125 volts. To show the pressure of the mains which we have shown have a voltage but slightly higher than that used by the lamps, "pressure wires" are usually run back from the mains to ''pressure indicators" or voltmeters situated in the dynamo room. Thus at a glance, the dynamo tender can see the exact voltage at the lamps and regulates his dynamo accordingly. Assuming that this is a practical case, we first desire to know what the size of the wire must be for the feeders to carry current for the 200 lamps with a loss of 13 volts or 10% in the wire. The 16 candle power lamps of no volt type, will take practically ^ ampere each. Thus, the dynamos will have to supply 100 amperes for the 200 lamps. The formula for calculating the size of the feeder, is 21.21 C D C. M. == E C M.=Area in Circular Mils. 2i.2i=Resistance of 2 feet of copper i mil in diameter. C=Current in amperes. .D=Distance in feet, to the lamps. $a ClTRKENO* DISTHIBTJTI'ON. E=I/OSS in volts in the wire. In the case given^ C=ioo, D=:iooo and E=i3. Thus the size of the wire in circular mils or C. M. is' 21. 21X100X1000 — ^ -=163,153 13 Thus the wire must have a sectional area of 163^153 cir- cular mils to carry the 100 amperes^ 1000 feet, with a loss- pressure of 13 volts. By consulting the table of wire in sizes in the back of the book, it will be seen that the nearest size, is No. 000 B. & S., which has an area of 167,800 C. M. The diameter in mils or thousandths of an inch of such a wire is 409.6. Thus it is seen that to maintain a pressure of 112 volts at the buildings 1000 feet from the dynamo, when the 200 lamps are burning, will require a No. 000 B. & S. wire. We have shown the buildings i and 2 to be 200 feet apart, and as before stated, it is desirable to njake the loss in the mains as low as possible*^ in this case for ex- ample, I volt. What size wire must be used ? We will use the formula used in the first case, 21.21 C D C. M. = E there will be a current of 50 amperes in either branch from the centre of distribution to the buildings. Thus in the formula, C=5o amperes, D=ioo feet and B=i footj thus: n/^' 21.21X50XT00 =::C. M. or 106,050 I which is slightly larger than a No. o B. S. wire^ which has an area of 105,592 C. M. We now have our CURRKNT DISTRIBUTION. 5I "wire sizes to deliver current inside the buildings at iii volts, which leaves one volt to be expended in overcom- ing the resistance of the service wires, from the mains to the lamps themselves. This loss is calculated in the same manner as the other two cases. We may in this calculate the size wire necessary to deliver any number amperes any distance at any loss and the formula should be remembered by anyone having any distributing work to do, as it makes him entirely independent of wiring tables, since he has the key to the whole plan himself. In all constant potential distributions, safety devices must be so placed in the conducting wires, as to make it impossible to overload the dynamo or wiring by an extra flow of current, due to metallic contact between the wires, accidentally or otherwise and thus produce what is known as a * 'short circuit". Short circuits may be caused by two wires having a difference of pressure com- ing in contact with each other or with any other con- ductor, so as to cause an excessive flow of current which may overload the dynamos or perhaps melt the conduct- ing wires unless a safety device is so placed as to cut off the current until the trouble is remedied. On the usual constant potential circuits used for lighting *^fuses", made of an alloy having a low melting point, are placed in the circuit, so as to melt when an excessive amount of current pass through them and thus open the circuit. In figure 19, fuses at A. B. and C. are so placed as to pro- tect the wiring, as shown. The fuses at A. would be of such size, as would carry 100 amperes safely, but any excessive amount above this, would speedily heat the fuses to their melting point and the circuit would then be "open". At B and C the fuses would be $2 CURRKNT DISTRIBUTION. of 50 amperes carrying capacity and any rise above gd amperes would melt the fuse and protect the wiring be- yond it. It should be understood that fuses are used simply to prevent more curreat to pass through a wire than its safe carrying capacity allows, and whenever placed should be so arranged as to size as to open the circuit before the wire is carrying more than its safe carrying capacit}^ They must melt before the smallest wire which they protect, shall have passing through it, more current than the law allows. Fuses even at their best, are often sluggish in operating, especially when of large size, and great care must always be exercised in putting the proper fuse in the proper place. For places where an unusually heavy fuse would have to be placed, such as the dynamo room of a street rail- way plant, instead of fuses, * 'circuit breakers*' are often placed, which open the circuit b}^ mechanical means and they are undoubtedly far more reliable and satisfactory than any s^^stem fuses could be. The usual plan of oper- ating mechanical circuit breakers, is to have a magnet carrying the whole current of the wire it protects, so arranged as to trip a catch when the maximum current is reached, and this catch releases the contacts, which separate and thus open the circuit. The loss of electrical energy in a conductor of a given resistance varies with the square of the current in am- peres passing through it. If a certain vdre has for in- stance, 5 ohms resistance, it will take 10 volts pressure to put 2 amperes through it, the loss in watts being 2 X 10 or 20. If we pass 4 amperes through this same wire it will take 20 volts, the watts being now 4X20 or 80, thus in doubling the current in a given wire, the loss in watts CURRENT DISTRIBUTION. 53 will be increased 4 times or as before stated, it varies with the square of the current. For the reason that the loss in a conductor varies with the square of the current in amperes passing through it, it has always been the aim of electrical engineers and inventors to make lamps, motors, etc. , operate at as high a voltage as is permissable with due regard to safety and reliability. Up to the tDresent time incandescent lamps um [ FIGURE 20. — SIMPLE MUWIPI^E SYSTEM, DYNAMO SUPPI.YING 10 LAMPS. have not been made so as to give good results at any higher voltage than about 1 10 volts and for this reason we are practically limited to no volts pressure as the highest to be used for the operating of incandescent lamps when placed in multiple. The Edison 3 wire system was designed to make it possible to carry current a much greater distance from the dynamo than was possible by the simple multiple sys- tem without great loss, and by its use the loss can be greatly reduced in a system of distributing current for lighting. The amounts of copper necessary to distribute current for a certain number of lamps, by the 3 wire sys- tem is about ^ of that used in the simple multiple system. The explanation of the plan of wiring will be shown in 54 CURRENT DISTRIBUTION'. the figures 20, 21^ 22 and 23. In Fig. 20, the no volt dynamo D, is shown connected to its load of 10 incandes- cent lamps. The current flows out from the positive brush and then through the lamps back to the negative FIGURE^ 21 — TWO DYNAMOS SUPPI.YING 5 I.AMPS KACH^ ON MUWIPI^EJ SYSTEM, UU6 6 n FIGURE 22. — ^MUI.TIPI,E SERIES SYSTEM FOR lO I.AMPS, brush; assuming that each lamp is 32 candle power at no volts pressure, it will take one ampere of current, thus the 10 lamps take 10 amperes of current. This plan CURRENT DISTRIBUTION. 55 as shown in Fig. 20 is a simple multiple plan of wiring Fig. 21 shows the same number of lamps but they are supplied by two dynamos Di, D2, each of half the capacity in amperes but the same voltage as dynamoD in Fig. 20. Thus with the 5 lamps connected to each dynamo, there will be 5 amperes flowing out from the positive brush and through the lamps back to the negative brush of each dynamo. Now, if the two dynamos Di and D2, were connected in series and each of them was designed to generate no volts it is evident IflGURE 23.— EDISON 3 WIRE SYSTEM SUPPI.YING 10 LAMPS- that the no volt lamps may be connected in series of two and give their proper candle power as shown in Fig. 22. The current flowing out from the positive brush of Di will be but 5 amperes, for there are now 5 series of 2 lamps each, each series taking i ampere at 220 volts pressure. This plan would not meet practical conditions, for if one lamp of a series of two were turned out, its mate 56 CURRIONT DISTRIBUTION. would also be extinguished. A single lamp could not be turned on and off at will and to make it possible to do so, a third wire must be added, Fig. 23, which will make it an Kdison 3 wire system. In this case D-|- and D — repre- sent the dynamos Di and D2 in figure 21 & 22. The 3 wires are marked -}-> i and — , and are called respec- tively, positive, neutral and negative wires. The two dynamos are connected in series as in figure 22, and the pressure between the two outside wires -|- and — , is therefore 220 volts. The pressure of each dynamo being no volts, there must be but no volts pressure between the -)- and the +, or between the d: and — wires. The current flowing out from the -\- brush of the dynamo D-|- must be but 5 amperes and as long as the loads in am- peres are equal on both * 'sides' ' of the system there will be no current flowing in or out on the middle or ± wire. If, however, a lamp is turned off on the ''positive side'* of the system, (the lamps supplied by D-|-) the current flowing out on the -\- wire will be but 4 amperes, which will destroy the balance and we will then find a current of one ampere flowing out on the neutral wire to make up the 5 amperes needed for the ' 'negative side' ' of the system. The current flowing on the neutral wire will be the difference between the loads in amperes on the two sides of the 3 wire system. If the loads in amperes on the two sides balance, the station switch on the neutral wire can be opened and the lamps will not be affected. With the neutral switch closed, it is evident that by opening station switch on the positive wire, that all the lamps on the -{- side will be put out, but that the lamps on the — side will burn as usual and if the station switch on the negative wire is opened and the positive and neu- CURRKNT DISTRIBUTION. 57 tral switches closed, the + side will burn and the — side be put out. As to the saving in copper and relative losses in this system as compared to the simple multiple system, it will be noticed that the outside wires carry but half the number of amperes that w^ould be necessary on the mul- tiple system, figure 20, and that the middle or neutral wire usually carries but a small amount as compared to the outside wires. Thus, if for a moment we ignore the necessity for a neutral wire, the size necessary for the 2 outside wires will be found to be but X ^^^ size necessary to supply current for the same number of lamps, the same distance from the dynamos, in a simple multiple system of distribution, for the current at 220 volts is but 5 amperes, and the loss is twice as many volts, for if we assume a loss of 2 volts from the dynamo to lamps on the simple multiple system, the dynamo voltage must be 112 volts and the voltage at the lamp is no. Thus in the 3 wire system with each dynamo generating 112 volts, we will have 224 volts between the outside wires at the dyna- mo and since the two 1 10 volt lamps in series need but 220 volts, it will be seen that we can allow a 4 volt loss in our wiring and still have the lamps up to candle power. Thus it will be seen that our loss in volts is 4 instead of 2 and our current in amperes is reduced from 10 to 5, thus, each of our outside wires need but be X "^^^ ^^^^ that would be necessary for the same number of lamps on a simple multiple distribution. As has been stated the middle or neutral wire should carry but a very small amount of current in a well de- signed 3 wire system, but for the reason that the fuses on either of the outside wires might be melted in case of a 58 CURRENT DISTRIBUTION. short circuit and thus make the middle wire carry as much as the outside wire, the rule has been followed, of making the neutral or middle wire as large as either of the inside wires, thus we have 3 wires, each X ^^^ size that would have been necessary for each wire of a simple multiple sj^stem and the relative amounts of copper will thus be 2fX)4=H f^- 3 wire or 2X1=2, the size for sim- ple multiple wiring, the relations are thus : ^ to 2 or ^ to I. In calculating wiring for 3 wire distributions we may get the size necessary for a simple multiple system of wiring for the number of lamps we desire to run on the 3 wire system and then divide the area of the wire in cir- cular mils needed for each of the wires of the multiple system by 4, which will give the area of the size wires needed to distribute current to the same number of lamps by means of the 3 wire system. The Edison 3 wire system is used by nearly all the larger size stations supplying direct current for incan- descent lighting. There are 4 and 5 wire systems sometimes used, which are operated on the same plan as 3 wire systems with the exception that an extra dynamo is used for each addi- tional wire, thus a 5 wire system will have 4 dynamos, or their equivalent, all working on the same lighting sys- tem. It is doubtful if the extra complication necessary with such a system is in the line of economy or not, es- pecially with medium sized plants. . For supplying current to street car motors, a plan of current distribution is used, which in the usual single trolley systems is very different from that used to supply current for illumination. CURRENT DISTRIBUTION. 59 The usual method employed will be readily understood from figure 24. The dynamo D of the compound wound constant potential type is designed to generate a varying amount of current at about 500 volts pressure. The pos- itive brush is connected to the trolley line L and the negative brush should be connected to the rail or to the copper wires laid in the ground near the track, which serve as the return circuit for the current. Thus the neg- ative side is always ''grounded" and in fact, the earth itself is used to a limited extent for the return circuit of the current used in operating the motors on the cars. In most cases it has been found that the earth cannot be FIGURK 24 — ^STREET RAII^WAY TROI,I,EY SYSTEM. depended on to furnish a path of low enough resistance to insure satisfactory results and a system of * 'bonding" is always resorted to. The "bonding" consists of uniting the ends of the rails together by means of heavy copper ' 'bond wire' ' thus making use of the metal section of the rail to form a continuous metallic circuit from the cars back to the power house. The question of a good return circuit that is durable, is one that has worried the de- 6o CURRKNT DISTRIBUTION. signers of electric street railways a great deal. This is owing to electrolytic action on the rails, bond wires or on water and gas pipes or other metal conductors near the line of electric road. If a current of electricity be made to flow from or to a metal plate immersed in a con- ducting fluid, an electro-chemical action is set up which disintegrates or destroys the metal, the rate depending on the amount of current flowing. This same action is taking place on the rails and other metal conductors when they are placed in moist earth, and its destructive action will depend on the amount of current flowing from the metal to the earth. In an electric street railway track, if the rail circuit contains considerable resistance, part of the current will flow to the earth from the rail, and if water pipes or other good electrical conductors in the earth are in such a position as to make a part of the return circuit, current will be conducted through them, and the chemical action set upon the metal surfaces will rapidly destroy them and make great trouble. The only way to obviate this trouble is to make the return circuit through the track so good that there will be but little current flowing from the rail to the ground, for the cur- rent will always divide depending on the relative resist- ance of the paths offered it. There are some cases of distribution of street railway current by means of two trolley wires, one of which is connected to the positive brush and the other to the neg- ative. In this case the earth and rails do not form a return circuit, the entire distribution being effected by means of the two trolley wires, between which 500 volts pressure is maintained. The motors are of course con- nected between the two wires, by means of two trolleys, CURRKNT DISTRIBUTION. 6l which bear against the trolley wires and thus make con- tact with them. Underground conduit distribution for street railways is gradually being developed, and in some cases we will find the rail being used as a return circuit and in others two underground trolley wires are used, insulated from each other and from the earth, in which case the con- ducting wire of course have no more connection with the rail as a return circuit than the double trolley system of overhead distribution, dot-s v^ith the track circuit. 6i aWErnating current. CHAPTER V. ^ransEormkrs and amernating current DISTRIBUTION. In the preceding chapter,* we have not spoken of alter- nating current distribution and have purposely avoided doing so for the reason that there are several peculiar characteristics of pulsating and alternating currents which should be thoroughly studied by themselves. By suddenly completing and then breaking an electric circuit, it will be found that there seems to be an action take place, similar to that of inertia. The current does not rise instantly rise to its full value and when the cir- cuit is broken, there will be evidence of the current tend* ing to resist the breaking of the circuit* This effect Varies with the "inductance" of the circuit. The induc- tance being the magnetism producing effect of the circuit. If the circuit is through a coil wound on an iron core, the effect will be much greater than if the wire is not wound in coil form. This action is caused by the lines of magnetism or the magnetic field being generated around the wire when the current is started through it. Each wire has its magnetic influence which is created the instant that the current starts flowing through it. If this wire is part of a coil, its magnetic influence must affect other neighbor- ing wires of the same coil. The effect will always be AI^tERNATlNG CURRENT. 63 gucli as to retard the flow of current through such a coil, until the full current strength is reached. The "self- induction" as it is called, will vary with the square of the number of turns in the coil. Thus, a coil of 10 turns has 100 times the self induction which a coil of one turn would have. If, then, we connect the terminals of a large coil of wire which surrounds an iron core, to a source of electricity, we will find that it takes an appre- ciable time for the current to reach its full strength, and that when the terminals are disconnected, that a spark will show itself in breaking the contact which is many times larger than it would have been, in case the same amount of current was interrupted, which had not passed through a coil such as described. The self-induction of a circuit, always resists the sudden starting and stopping of a flow of current. This eflect may be very easily dem^ onstrated by placing an ammeter in series with a field magnet circuit of a shunt wound dynamo and watching the gradual rise in current through the circuit, on con- necting it to a source of electricity. On a **dead beat" ammeter, that is, an ammeter whose needle comes instantly to the correct reading, without going past it, the gradual rise can be readily seen, and on breaking the field magnet circuit, a flash will take place which clearly shows that the current resists being broken. In fact, the voltage at the terminals of a no volt dynamo field coil circuit, may be several times no volts the in- stant the circuit is broken and a shock obtained in this manner is often exceedingly painful, if not dangerous. It will now be evident that when a current of electricity is started through a coil of wire, that each wire is send- ing out its magnetic influence, which may effect other H AI.'r^RNA'riNG CURRKN^. conductors in its vicinity and in fact, if two coils of wire are placed near each other so that the magnetic influence of one coil may affect the other, it will be found that the instant current is started in one coil that a current will at once be generated in the second coil, the amount gen- erated depending on the resistance, etc. , of the second coil. It will be found that the current in the second coil jRon coftfi*. J) tOOO VOLTS* PRIMARY- COIL OP 100 TURPIS. 100 VOLTS. SECONDARY, t^lGURK 25.— IRON CORK WITH PRIMARY AND SECONDARY COII.S. will only be generated while the current in the first coil is being increased or diminished and that the instant that the current in the first coil becomes a constant quan- tity that the current in the second coil falls to zero. Also that the current in the second coil flows in one direction during the rise of current in the first coil and that the current flows in an opposite direction when the current in the first coil diminishes in AIvTERNATING CURRENT. 65 strength. Thus by sending a pulsating or alternating current through the first or primary coil, a pulsating or alternating current may be generated in a second or sec- ondary coil^ although there is absolutely no metallic connection between them. See figure 25. Alternating current dynamos are easier to build than pulsating direct current dynamos, and are certainly eas- ier to handle. They have no commutator, bat instead have collecting rings which are the terminals of the coils on the armature. Brushes bear on these rings and in this manner connect the armature coils to the wiring connected to the lamps. The usual alternating current dynamos used for lighting purposes, give out from 15,000 to 16,000 alternations per minute, although dyna- mos lately constructed are being made from 6,000 to 9,000 alternations per minute, which is in the line of for- eign practice. The usual alternating current dynamo is designed to generate either 1,000 or 2,000 volts and a varying number of amperes, and this pressure is reduced at a transformer^ to 50 or 100 volts for use in operating the usual incandescent lamp. The electro- motive force or voltage generated in the secondary coil, as compared to the voltage on the pri- mary, will depend on the relative number of turns in the two coils. If the primary coil is connected to 1000 volts pressure of alternating current and has 100 turns in it, we will have generated in a secondary coil of 10 turns, a pressure of 100 volts, or if there are but 5 turns of wire in the secondary coil, we will have but 50 volts between its ends. We can in this manner generate alternating cur- rents of high voltage and distribute them long distances from the dynamos, with a small loss and when the build- 66 AI^TBRNATING CURRENT. ing is reached which is to be lighted, the his:h pressure current is connected to the primary coil of a transformer ^ on the secondary coil of which the lamps are connected. If there is i ampere at looo volts or looo watts ]:)assed through the primary coil, we will find practically the same number of watts given out by the secondary coils, the only change being that at loo volts we will have a current of lo amperes or at 50 volts — 20 amperes. Thus the current in amperes is increased in proportion to the reduction of pressure and it is possible in this way to generate any voltage desired on the secondary winding, by proportioning the number of turns in the primary and secondary coils. The efficiency of transformers, that is, the proportion of the energy supplied the primary coil given out by the secondary varies in different makes, but at full load, large transformers can be made to give to the secondary from 95% to 97% of the energy supplied the primary. Figure 26, represents an alternating current dynamo, generating a constant pressure of 1000 volts, connected to 2 transformers, i of which (No. i) reduces to 100 volts, the proportion of turns on its primary and secondary coil being 10:1, and the second transformer (No. 2) reduc- ing to 50 volts the proportion of turns in the two coils in this case being 20:1. The same dynamo may supply a third form of trans- former which is called a "step up" transformer, the pri- mary coil being supplied with 1000 volts current and the secondary coil supplying a higher voltage, 5000 volts. This is accomplished by winding a proportionately larger number of turns on the secondary coil than is wound on the primary, the proportion being 1:5. ALTERNATING CURRENT. 67 It must be understood that the secondary coil has no metallic connection with the primary coil, and that what" ever current is generated in it, must be due to the induc- tive effect of the current in the primary circuit. To get the maximum effect of the current in the primary cir- cuit on the secondary winding is the first requisite in a good transformer. The coils are therefore both placed on an iron core in such a position that as man}'' as possible of the magnetic DVMAAVO tooo vocrs. _ I^IGURE^ 26.— PI,AN OF ALTERNATING CURRENT DISTRIBUTION. lines from the primary coil will embrace the secondary winding. The coils are usually placed close to each other and in fact, would be interlaced with one another, were it not for the fact that aside from difficulty of man- ufacture, the danger of contact between the high voltage current carried on the primary circuit and the lamp cir- cuits from the secondary windings would prohibit it. In practice we will find that the primary coil is extremely 68 AI.1^BRNATING CURR:^NT. well insulated from its neighboring secondary coil, so as to make it quite unlikely for contact between them. The theory of the transformer is quite complex when all of its internal actions are taken into consideration, and many books have been written, filled with algebraic formulas in regard to the transformer and its actions. We may take up a few simple actions however, without the aid of mathematics. We have already spoken of self-- $9 V01.T.S. RATIO 10:1 vouXS RATIO! :5 FIGURE 27. — FORMS OF TRANSFORMERS. induction of a coil of wire surrounding an iron core, and have shown that it takes an appreciable time for current to reach its maximum in a coil having self-induction. Also the apparent resistance offered to the breaking of the circuit. If an alternating current bc^ supplied to a coil surrounding an iron core, we will find that owing to the self-induction of the coil, it will be impossible to pass AI.TERNATING CURRENT. 69 ^s much current through it with a given voltage as would be possible with direct current of the same voltage. In fact, in a coil having for instance, 10 ohms resistance, we might be able to put but ^^ ampere through it with a voltage of 1000, provided the coil was wound on an iron core after the manner of a primary coil of a transformer. 1000 volts of direct current would put 100 amperes through the same coil. This difference is due to the self- induction or ''impedance" of the coil when supplied with an alternating current. The reason is, that with the current alternating 15,000 or 16,000 times per minute that the electro-motive force or voltage, although high, does not have time to force the current to its maximum, before having fallen to zero and exerted an electro- motive force tending to reverse the current in the coil. The amount of energy expended will be in this case, only a small amount and in a transformer will be termed the Leakage or magnetizing current. The leakage will depend on several factors. In the first place there must be enough iron iu the trans- former core to form a path of low magnetic resistance for all the lines of magnetism given out by the primary coil, so that with the secondary coil placed on this core, practically all of the magnetism of the primary current will effect the secondary winding- Great care must be taken in selecting the quality of iron, for the magnetism of the iron transformer core is being reversed each alter- nation of the current and some kinds of iron will mag- netize more quickly than others. It takes power to reverse the magnetism in the iron and this loss is called hysteresis loss, hysteresis being the loss occasioned by altering or reversing the magnetism of iron. This hys- 70 ALTERNATING CURRENT. teresis loss is smallest when a soft iron core is used and to provide against eddy currents in the core, it is lamin- ated in the direction parallel to the lines of magnetism in the core. The eddy currents in a solid iron core would be caused by the magnetic effect from the primary and secondary coils generating currents of electricity in the iron itself, which although of extremely low voltage, would have sufficient current strength to heat the iron and core, thus not only endanger the insulation of the transformer but also counteract part of the magnetic effect in the iron. Thus the cores of transformers and in fact, all coils which carry alternating currents should be di- vided in small sections either by building them up out of thin sheet iron or out of soft iron wire. These divisions should not be made so as to break the continuity of the magnetic circuit, but must be made parallel to the lines of magnetism in the iron core. The leakage in the primary coil is of course partly de- pendent on its resistance, but this resistance has practic- ally no effect at all in the case, we have mentioned of a single coil wound on an iron core for the impedance due to self-induction in the coil^ exerts by far the most powerful tendency to keep the current from passing through the coil. If now, on the laminated core of iron, we place a sec- ond coil and for instance, have the ratio of turns of wire in the high pressure as compared to the low pressure or second coil, lo to i, we will find that when the primary coil is connected to a looo volt constant potential alter- nating current circuit, that the low pressure coil forming the secondary winding will have a voltage of loo volts at its terminals and if we have a correctly designed trans- AI.TKRNATING CURRENT. 71 former of say, 5000 watts capacity (100 — 16 c. p. lamps) will find that with practically no current in the second- ary circuit, that there will be a leakage on the primary of about ^Q ampere at 1000 volts pressure. If now, lo lamps each taking )4 ampere at 100 volts are connected to the secondary winding, we will find that the current in the secondary circuit will now be 5 amperes and that the primary current has increased from ^q ampere to /g. If 10 more lamps are now connected to the second- ary, the primary currer-t will be found to be ijo ampere. In fact, as the current in the secondary winding is in- creased, the primary current is also increased and this increase in the primary should be as many watts as is added to the secondary load. The self-induction and impedance of the primary circuit is being decreased by the mutual induction taking place between the second- ary and primary windings, for as the load increases in the secondary winding, just so much is the counteracting effect of the secondary winding on the primary. Thus owing to the decreased impedance of the primary coil, owing to the effect of the current in the secondary wind- ing more and more current flows through the primary <:oil, until at full load the primary winding is carrying its maximum current and the secondary is exerting its full contracting effect on the impedance of the primary winding. The efficiency of such a transformer should be about 97% at full load. That is, the secondary wind- ing should be delivering 97 % of the energy supplied to the primary coil. Thus it will be seen that alternating current can be distributed at high pressures and then reduced at trans- formers to a voltage suitable for operating incandescent 7^ AI^TERNA^TING CURREN'I^. lamps with but a small loss in transformation. The loss in watts in transmitting a certain amount of electricity through a wire of given resistance may be stated as vary- ing with the square of the pressure or voltage at which it is transmitted. Thus to deliver 5000 watts, i mile from the dynamo at 1000 volts pressure, will take xig ^s much copper as that necessary to send the same 5000 watts at 100 volts pres- sure with the same loss. For several reasons, 1000 or at most, 2000 volts pressure is as high as is safe to go in generating current for ordinary lighting plants and most alternating current dynamos are made for either one or the other voltage. A transformer may have more than one secondary coil or a single coil may be divided into two or more sections and by varying these connections, it will be possible to get for instance, from the usual form of tratsformer used in America for incandescent lighting, 50 or 100 volts as desired by connecting the two sections of the secondary in multiple or series w4th each other. Fuses are usually placed on the primary wires only, in the modern tj^pe of transform r. In case of a short circuit in the secondary winding or the wires leading from it. the primary fuse would immediately be melted, and thus open the circuit. When the secondary wires enter buildings the usual method of fuvsing all circuits must be carried out, not as a protection to the transformer, but as a protection to the smaller tap wires leading to lamps, etc., which unless provided with fuse wires might in case of a short circuity melt before the primary transformer fuse would open the circuit. The wiring from the transformers to the lamp should ALTERNATING CURRKNT. 73 be carefully calculated, owing to the fact that the trans- formers on a constant potential primary circuit cannot provide extra pressure as the load increases so as to com- pensate for loss in the wiring. The wiring on all second- ary circuits should be done so as to provide for a very small drop, for in fact, the secondary voltage gradually falls as the load increases and although in the well de- signed transformers, this drop amounts to but from i % to 2 % , it is oftentimes sufficient, when combined with loss in the secondary circuit, to cause a marked diminuation in candle power of the lamps. Special transformers are often wound so as to give from 30 to 35 volts on the secondary winding, which is the voltage needed to operate the usual type of alternat- ing current arc lamps now on the market. Step up transformers are usually used in places where it is desired to send electricity for lighting, etc. , a con- siderable distance from the dynamo. The voltage of the alternating current dynamo or ''alternator" is usually looo or 2000 and this is raised in the step up transformers to pressure sometimes as high as 20,000 volts. The cur- rent at this pressure is then transmitted in some cases 20 or even 100 miles and then the pressure is again reduced to that desired for lighting, etc* During the past years, great progress has been made in the perfection of alternating current motors of various types, and the next few years will undoubtedly see many plants in mining districts supplying light and power for miles around water falls supplying cheap povv^er for run- ning the alternating current dynamos. The usual alternating current dynamos as is used for lighting, supplies a single phase current as distinguished ^4 AI.TKRNATING CURRKNT. from alternating current dynamos supplying multiphase currents, which may have two, three or more phases. 5rpfGLfr Phasb-. AtTERl^ATlNO Wav6. FiGURK 28 — PI.AN A. cofts 90 TVO PMAS6 CtRHtnt Figure 28.— Pi^an B. 3 PH/^E. ^tjBRWATlfW CtWJV«:«JVV^v^ Figure 28.— Pi,an C. ALTERNATING CURRENT 75 A dynamo which supplies an alternating current which consists of a succession of single alternating impulses will be called a single phase dynamo and such a current is a single phase current. If, however, there are two windings or their equivalent on the armature, each of which is sending out a single phase alternating current, the dynamo is now a two-phase dynamo and by designing the armature coils so that the rise of current in one armature winding is not coincident with the rise in the other winding, peculiar magnetic effects may be produced in a suitable form of magnet by providing it with two windings, which are supplied with current from the two armature windings of the two phase dynamos. Likewise^ three windings may be placed on a single armature and thus make a three-phase dynamo generating a three-phase current. The figures 28 show the current wave of a single phase alternating current dynamo (plan a). Plan b, shows a two-phase current, in which the relative amounts of current during a revolu- tion are shown in the two windings. The waves of cur- rent in this case are in quadranture, that is, one coil's current wave is 90° ahead of the current in the other coil. Plan c, shows the relations of the currents in the coils of a three-phase generator. The currents are in this case 120° ahead of each other. Alternating current motors of recent construction are almost invariably designed to operate on multiphase cur- rents. The large Westinghouse alternating current dy- namos supplying the incandescent lighting of the World's Fair at Chicago, were of the two phase type. The cur- rent was distributed to the transformers on the single phase plan, the voltage being over 2000 volts. y6 'TYPE.^ O^ DYNAMOS, CHAPTER VI. i'VPES OE DIREC1^ AND AIvTERNATlNC CXJRRENl? DYNAMOS. Direct current dynamos are manufactured principally in three types, shunt, series and compound wound. The shunt dynamo is found largely in isolated and central station electric lighting plants, operated on the two and three wire systems, and supplying incandescent lamps, small electric motors and constant potential arc lamps. These three types of machines are made in bipolar (two poles) and multipolar (more than two poles) design as respects field magnets. In the previous chapters, the reader has learned that a current is generated in the armature by the movement of coils of wire in a magnetic field which is usually pro- duced by electric currents flowing through coils encir- cling bodies of iron, called field magnets, and that the electric currents generated in the armature are taken from it by means of the commutator and brushes bearing on it. Consider a U shaped piece of iron forming the field magnet, wound with wire so as to form an electro-mag- net when supplied with the electric current. Now in the shunt dynamo, the two terminal wires of a magnet similar to this are attached to the armature terminals TYPES OF DYNAMOS. 77 and form a shunt around the armature from whence the name "shunt dynamo" is derived. By previous application of an electric current to the coils surrounding the iron cores, they have been magnet- ized and iron once magnetized always retains a little of its magnetism, called "residual magnetism". If then, there is the feeble magnetism remaining in the magnet, it follows there must always be a slight magnetic field between the poles of the field magiets and as there is an armature revolving in that field, a current is generated which passes out to the commutator and by brushes is led ofi" to the circuit. One path which this current can take is that through these field coils which at once causes them to be more powerfully magnetized and a greater magnetic field is thus produced, hence a greater amount of current is generated in the armature. By this step by step process, the machine slowly "builds up" to the proper potential until the normal magnetization is reached. A rheostat or a variable regulating resistance is in series with and connected in the field circuit which regulates the potential or voltage by varying the current through the fields. When a dynamo is running with no load, all the resistance in the rheostat is generally in circuit. Now as the load increases, whether it be that electric lights are switched on or motors are run, either of these will require dynamo current and the potential of the ar- mature falls slightly. To get more current in the fields so as to raise the potential, we must increase the current through them by manipulating the rheostat. A shunt or compound wound dynamo generally speaking, has its pressure remain constant and the current quantity varies 78 type:s of dynamos. as more or less lamps are turned on. The shunt or com- pound wound dynamo for supplying constant potential current usually depends on the varying strength of the field magnets for regulation, but the series wound dyna- mo supplying a constant current is often regulated by * 'armature reaction" alone, the armature reaction being the internal electrical action of the armature windings, which may be used for regulating. The series dynamo is used almost exclusively for series arc lamps, but series motors can be placed on the series dynamo, and operated. The field magnets are in series with the armature and full current of armature must pass through them. Again we find that the series dynamo usually generates a variable pressure and constant cur- rent. Arc lamps are run in series with the dynamo, and if this current supplying them fluctua.ted, as does that on a shunt machine it would produce great variations in the candle power of the lamps which would make a very un- satisfactory light. The arc lamp when burning on a ten ampere circuit has a resistance between the carbons of A% to 5 ohms and to force the requisite lo amperes of current through the arc, a pressure from 45 to 50 volts (CXR=B) is required. If 10 lamps are to be run, the dynamo must supply the 10 amperes at a pressure of 500 volts and the dynamo for this purpose usually has its brushes shifted so as to cut in additional active coils in the armature, which means an increase in pressure or voltage on the line for the extra lamps cut in the circuit. No rheostat is necessary in this case, as the regulation of the armature by the shifting of the brushes keeps the current constant through the fields, the magnetic field of the dynamo always containing the same number of mag- TYPES OF DYNAMOS. 79 netic lines. In the series wound machine the brushes are shifted against the direction of rotation for an increase of load. The compound wound dynamo supplying constant potential current, though embodying the shunt and the series principles, is more a shunt machine than series. Its regulation depends upon its fields. Its voltage is con- stant and ampereage variable. In a well designed dyna- mo after the voltage is regulated by means of the rheo- stat, the machine takes care of itself. The armature rotates in a powerful magnetic field. In either the shunt or compound wound dynamo, it is possible to so propor- tion the field magnets and armature that the ''non-spark- ing" point on the commutator is not shifting as the load varies, although in many makes of dynamos, as the load increases the brushes must be shifted forward in the di- rection of rotation, or sparking will result. This is caused by the magnetic efiect of the armature distorting the flow of magnetic lines given out by the field magnets so as to alter the position of the neutral line in the arma- ture. In compound wound dynamos all the current gen- erated in the armature passes through the series windings on the fieldmagnets. The machines are so built that the series winding does the regulating and the magnetic field does not reach its full strength until the dynamo is deliver- ing its full current. The dynamo * 'builds up" by virtue of its shunt winding and as current is required from the dy- namo for the outside circuits, this same current passes around the series coil of the field magnets, increasing the magnetic field and consequently maintaining the pres- sure uniform. Compound wound dynamos may be com- pounded for any percentage increase in pressure from no 8o I^YPKS OF DYNAMuS. load to full load, thus compensating for the **drop" in voltage that occurs on the line when large currents are being passed through them. Alternating current dynamos consist principally of three classes, self excited, separately excited and com- posite or compound. They are almost invariably high voltage machines, from looo volts up and consist in an armature of as many coils in series as there are field magnets and these are much in excess of those on direct current dynamos. A great number of alterations of the current are required for practical work and consequently to keep the armature speed down to a reasonable point, a greater number of mignet poles are used. Currents generated by the armature are led out in two wires to collector rings, where by brushes connected to the circuit the current is taken to the transformers, etc. The mag- netic field extends from one field magnet to its neighbor on either side of it. The field magnets which are usually wound in opposite directions on each successive pole, are excited in the self excited machine b}^ taking the current of one or more of the armature coils and passing it through a current rectifier. It would be useless to excite the fields by the alternating current as the rapid reversals in the cur- rent unfit it for such Service. The alternating current must be commuted to a direct current and the rectifier or two part commutator performs this function, by sending all the impulses through the field magnets in one direction and they are thus excited by a pulsating direct current. The residual magnetism in this case plays a part in the alternating dynamo as it does in the direct. As the ar- mature rotates in the magnetic field, weak alternating currents are generated passing through the rectifier, TYPES OF DYNAMOS. 8l thence around the field magnets, again developing great- er currents in the armature until normal magnetism is reacher^ . The separately excited alternator is devoid of a rectifier and has its fields excited by an independent direct current dynamo. Regulation is obtained by vary- ing the current in the field magnets as the load varies by means of a rheostat in the circuit of the direct current dynamo. The composite field or compound alternating dynamo is analogous to the compound direct current dynamo, inasmuch as an additional winding passes around the field magnets in addition to the usual winding found on the separately excited alternating current dynamos. This winding is supplied with a pulsating direct current from a rectifier which commutes a portion of the output of armature current, the amount depending on the load in amperes. As the current is increased on the circuit, just so is the current increased in the fields and the potential is gradually increased to overcome the ''feeder loss." A resistance is bridged or shunted across the rectifier which can be varied so as to produce different increases in the pressure at full load to allow for "drop" in line. Other classifications of alternators are made, namely: dynamos in which the field magnets are stationary and armature rotates, dynamos in which the armature is sta- tionary and field magnets rotate and those in which field magnets and armature are stationary and an irregularly shaped iron inductor rotates between the two. The principal of the ' 'inductor' ' type of alternating current dynamo is that if the number of lines of magnetism is varied through the stationary armature coil that the same effect will be produced as v^hen the armature coil 82 TYPES OF DYNAMOS. moves so as to cut these lines of magnetism. The revolving iron ''inductor" is so shaped that as it revolves it completes and then breaks the magnetic circuit through the armature coil and this of course must generate cur- rent in it. Alternating current dynamos are built for any phase and frequency but it is not timely to lead the reader further than has been gone into in the preceding chapter, as other books cover this advanced work. It is frequently the case that one dynamo is insuffic. lent to supply the circuits that it is feeding at certain hours during the run and it becomes necessary to place an additional dynamo on the circuit. In incandescent light- ing, the machines would be placed in multiple, but in arc lighting they would be connected in series. In the shunt wound dynamo an increase in load means an increase in amperes output with voltage constant, while in the series dynamo an increase in voltage results with the current constant. Therefore, if it is required to run an extra machine on a circuit, if it be a shunt or com- pound of the constant potential type, it is connected in multiple, but if the usual series dynamo, it is put in series. Instances where series dynamos are run in multiple or compound wound dynamos in series are few but explanation will be given their operation. SHUNT WOUND DYNAMOS IN MUI.TIPLK. The directions to be followed in placing shunt dynamos in multiple is as follows: One dynamo already running — Start second dynamo up to full speed — Set brushes on commutator — Move rhe- ostat handle until voltage of dynamo is the same or slightly greater than that of dynamo already running. TYPES OF DYNAMOS. 83 This can be indicated by a voltmeter. A pilot lamp is usually placed on shunt or compound dynamos and they will roughly indicate when dynamos can be placed in circuit. As soon as switch is thrown connecting both dynamos to the circuit, the load should be equalized on each by cutting in resistance in field circuit on the dyna- mo first running and cutting out resistance on dynamo just switched in. If the voltage of the second dynamo be less than that on the circuit, the dynamo will receive current from the first and operate as a motor turning in the direction of its previous rotation. In taking a ma- chine from the circuit proceed in reverse step. The Kdison three wire systems use shunt and compound dynamos. The two dynamos in this case, the positive dynamo supplying the -f- side and the negative dynamos supplying the — side, are started up as previously explained. Being independent of each other and working on separ- ate circuits, no especial precautions are necessary in starting dynamos. The potential should be kept alike on both machines and when possible, the current in amperes should be the same. If one dynamo carries more current than another, that difference existing, is is carried by the + wire, if, as the load increases, addi- tional dynamos are to be placed in parallel on either side of the system they are placed in, the circuit in the same manner as has been described, one dynamo being in mul- tiple with the -|- dynamos and the other with the — dynamo. One dynamo may be made to supply a 3 wire system by using a switch that will connect + wire with — wire making the neutral a common return for the two outside wires. This method is not to be recommended 84 TY^nS OF DYNAMOS. unless the original wiring was designed with this object in view. Shunt dynamos may be connected in series when long distance transmission is to be accomplished. The field circuits should be connected so as to form one shunt across the dynamos so run in series and they will thus all be excited equally. All machines in this case should be of the same current capacity and each must be able to carry the maximum current on the circuit, or in the case dynamos of various sizes in series, the current must never rise above the carrying capacity of the smallest armature in curcuit. SERIKS DYNAMOS IN SKRIKS. Dynamos to be thus connected must have same current capacities and the + terminal of one must be connected to the — terminal of the other. In a lighting plant, this is readily performed at the switchboard by plug con- nectors. This not so satisfactory as making other com- binations but is often done. If series dynamos are to be connected in multiple, let the armature of one dynamo excite the fields of the other and vice versa, so that if one generates not enough current, it weakens the field of the others and both are equalized. COMPOUND DYNAMOS IN MUI,TIPI,K. The compound dynamo embracing the characteristics of the shunt and series machine, the coupling together becomes an operation including both. In figure 29 two generators are connected for multiple working. One machine is running and the switches Ax for shunt circuit and Bi for series circuit are closed. The armature Di is then generating its normal electro-motive force and cur- TYPES OF DYNAMOS. 85 rents are flowing in the shunt field and the series field circuits. Armature D2 is then run at its normal speed, the switch A2 is thrown, allowing the shunt winding to excite the fields of dynamo D2. The switch C on the equalizer wire is closed and when switch B2 is closed, the machine takes its part ot the load. Before the sec- CONNKCTIONS O-^ TWO COMPOUND DYNAMOS IN MUI.TIPI,]^. ond dynamo is coupled in circuit, that is, before switch B2 is closed, the voltage should be about the same as that of the dynamo D i first running. After the two are coupled in circuit, the load on- each machine should be balanced by the rheostat. By examining the diagram of circuits, it will be seen that the equalizer wire practically places the two series windings in multiple, and this is necessary, owins^ to the fact, that in case two compound dynamos in miultiple were feeding a circuit and were not provided with an equalizing wire, and one dynamo had its voltage slightly decreased from any cause, for instance a slipping belt, that the current in the series coil of the dynamo 86 'TYPES OE DYNAMOS. whose voltage was lowered, would necessarily be weak- ened and this of course would still further reduce the voltage of the dynamo in trouble. But in the case of the dynamos provided with an equalizing wire, the two coils being in multiple and of equal resistance, will have the total current output of the two dynamos divide equally between them and thus tend to keep the two dy- namos balanced. The equalizer should have a very low resistance compared* to the series windings so as to per- form its office satisfactory. In cutting out a machine the same steps are taken only in reverse order. Compound dynamos of different current capacities can be run in multiple, if the voltage is the same and the resistance of the series windings are inversely proportional to the cur- rent capacities of the several machines, in other words, if a dynamo produces half as much current as another, its windings should have twice the resistance of the other. The machines also govern each other, as when one ma- chine runs too fast, it does more work and consequently lowers its speed, and momentarily it robs the other machines of part of their load, which makes them run faster and thus producing equality. Compound dynamos may be connected with good results in the manner described under shunt dynamos in series. AlyTERNATORS IN MUI.TIPI,E. To couple direct current dynamos in multiple we said that their potentials should be alike, but in alternating current dynamos not only this is usually required, but the machines must correspond in phase and frequency. To couple an alternating current dynamo in circuit with another, the impulses in both machines must rise and fall together or be 'in step." The frequency, period TYPES OI^ DYNAMOS. 87 and alternations are directly affected by the speed, for tlie faster the speed the ^eater the alternations, that is, frequency, and vice-versa. Now when one generator is coupled with another generator or motor and running in step with it, we say they are in synchronism. The instru- ment provided to indicate synchronism is called a synchronizer and is explained in chapter x. To discon- nect alternators when running in parallel, is not as diffi- cult as when coupling in. The main switch of dynamo is opened and then the switch on the exciter circuit to dynamo should be opened. It is a better plan while machines are running on single circuits to reverse this operation, throwing exciter switch first and then machine switch, as there are less chances of injury to alternator. The same effects which cause alternators to work well in parallel causes them to be opposed and get out of step in l^ROUBI^E IN DYNAMOS. CHAPTER VII. CAUSES OF TROUBI^E IN DYNAMOS. — THEIR REMEDY AND PREVENTION. The rotating portion of any dynamo electric machine or motor is its vital part. In some machines, this ele- ment is the armature, in others the field magnets. In case the rotating part is the armature, it will be evident that means must be provided to take the current gener- ated from the moving conductors of the armature to the lamps, motors, etc. , for which the dynamo is to supply current. This is done usually by means of brushes bear- ing on the commutator or collector rings, as explained in previous chapters. The first fault developed in a machine should be speed- ily removed, and the second fault never allowed to appear, as the machine will rapidly destroy itself. The following directions apply particularly to direct current dynamos of the * 'closed coil" type and do not apply to some ''open coil" dynamos used for arc light- ing, etc. We have to-day, several different styles of brushes in general use. They come under two general heads, metal brushes and carbon brushes. Metal brushes are made usually of copper, either of several leaves of thin copper ribbon or a number of cop- per wires soldered together at one end, or a combination of wire and leaf copper. A patented brush of consider- I'ROUBI.E IN DYNAMOS. 89 able merit is made with leaves of copper arranged be- tween leaves of high resistance metal with a number of sheets of oiled or parafined paper interlaid for lubricating purposes. The idea of high resistance metal either side of the copper, is to stop the sparking by reducing the short circuiting action of the brush on coils that are in the weak field near the neutral point on the commutator. Carbon brushes are used to a great extent on street railway machinery, both on dynamos and motors. The brushes are cheap, self-lubricating to a great extent, and do not wear the commutators near as much as copper brushes. The sparking which results from rapidly fluc- tuating loads is not only lessened but is made practically harmless to the commutators, since the burning action of tho sparks seems to concentrate itself on the carbon and does not injure the commutator. For the best results, a good grade of carbon made especially for this purpose, should be used. The vapor of the carbon, generated at a spark, is undoubtedly of higher resistance than the vapor of copper and this must reduce the sparking to a great extent. The carbon brush has been applied to arc dyna- mos of recent construction with considerable success. On motors which are designed so as to be able to reverse their direction of rotation, such as street car motors, etc., the carbon brush is a necessity and is usually set at right angle to the face of the commutator. Brushes should have more than sufficient cross section to enable full current of armature to be delivered through them continuously without undue heating, and the cross section of armature segments should be governed by the same rule. Every part of the whole width of the brushes should bear on commutator. When brushes are set on 90 TROUBIv]^ IN DYNAMOS. commutators of direct current dynamos, they should usually be diametrically opposite each other and then placed evenly so that every part makes true contact. There should be no dirt on contact surfaces, for if there is, severe sparking and heating will result. Brushes should rest upon commutator with slight pressure, but not enough to cause undue cutting or heating. They should be removed at regular intervals for inspection and cleaning and if necessary, placed in a brush jig or form, and filed to a proper bevel, as brushes will, even with the best care, wear uneven, burr and collect dirt. To remove diit and grease from brushes, soak them in gaso- line or benzine. Sparking resulting from the causes above named is usually distinguished from the nature of the spark, which is present at the brush points during the full revolution of armature. A position known as the neutral point, exists on all commutators of direct current dynamos. This is the posi- tion of non-sparking in most cases, and generally varies with the load. If the brushes are ahead or behind this point, the sparking is considerable and can be remedied by shifting the brushes till the non-sparking point is reached. The following is the cause of this shifting of the non-sparking point. The field magnets tend to set up a magnetic field in one direction through the arma- ture and the armature owing to its conductors carrying currents of electricity, magnetize the iron armature core in such a manner as to oppose the magnetism of the field magnets and the resultant field, will depend on the rela- tive strength of the two opposing influences. Such an action called ''distortion of the magnetic field" occurs in all direct current dynamos, and tends to shift the field of TROUBI^K IN DYNAMOS. 9I force out of its natural position and it is evident that since it is caused by the current carried in the armature conductors, that it varies with the load, thus making a change in position of brushes necessary to keep them from sparking. If there existed no field distortion in dynamos, no movement of brushes would be required. Though the care and inspection of the brushes has been spoken of, as much must be said of the commutator. No matter now nicely filed the brushes are, or how evenly they are set, or in what cleanliness kept, little v/ill avail, if the commutator is uncared for. A well cared for com- mutator should have a glaze and polish, and with a good dynamo tender, this is always attainable. Sparking is sometimes caused by what is termed, a **high bar" or a flat bar on the commutator, and can, by close scrutiny usually be detected. At first you are warned by a spark appearing on the commutator, quite difierent from that caused by dirty brushes. By applying the fingers to commutator, once in every revolution a depression or an elevation will be felt. Commutators sometimes become **out of true'' owing to improperly * 'smoothing down". This will be noticed by a rise and fall of the brushes when armature is revolved slowly. Again, if brushes are not properly trimmed or if not properly lubricated, the commutator will often present a bright coppery appear- ance and a disagreeable **sing" will be noticed, and when felt, it will be found rough, and plenty of copper dust will be found on brushes, etc. If in very bad condition commutator should be turned down in a lathe or better than this and without removing armature, a device simi- lar to a slide rest and usually furnished by makers of machines can be used with a narrow cutting tool. If ^2 I^ROUBI^K IN DYNAMOS. only in rough condition with no deep groves, sand paper of different sizes from coarse down should be fastened on a suitable block for bearing pressure and applied until smooth. The use of a file, unless in experienced hands, is not recommended as it will often cause fine bits of cop- per or burrs to lodge in the insulation between the seg- ments and short circuit sections. -'Flats'* are sometimes caused after arcing on a particular bar has occurred for some time, and if one of the segments is of softer metal than the others, a flat will gradually develop. From time to time commutators should be calipered and the cross section of copper determined by subtracting the interior diameter of commutator from the exterior and multiply by the mean width of each bar. Excessive heating in commutator other than that produced by local causes, is many times observed and can be accounted for when it is found that commutator bars are not sufficient- ly large to carry the armature current without heating. Where a light load is constant on a machine, this may not be noticed until the commutator is worn through, but if on the other hand, a heavy load is always deman- ded, heating will take place, a new commutator should replace the old. The commutator should be supplied with a suitable lubricant and probably the best for the purpose, is vaso- lene, which is not only cheap but does the work well. A small quantity should be rubbed on a piece of cloth or canvas or even better, a piece of leather and on the slightest sign of cutting of brushes on commutator, it should be applied. Great care must be taken in keeping copper dust and dirt aw^ay from commutator and arma- ture, for if allowed to gather, it will surely make trouble. TROUBIvE IN DYNAMOS. 93 In case of trouble with armature, first take a "mag- neto'' described in previous chapter, which should be part of the equipment of every electric light plant, and find whether the' windings of the armature are connected in any way to the core of the armature. This is easily done by connecting one terminal to the commutator segments and the other one to the shaft of the armature If it is possible to get a ring, you may be sure that something is wrong in the insulation of either the commutator or armature, if the trouble has made itself known by a violent flashing at the brushes, and on ex- amination it is found that the fault is not in the brushes themselves and that one or more commutator segments are badly burnt, it may be inferred that the armature coils connected to the segments are out of order. In event of a short circuited armature coil, the particular coil will usually be found to be hotter than its neighbor or even burnt, or in case of an open circuit, the armature will refuse to generate at all. If there is simply a bad contact at the commutator or m the coil itself, there will be considerable local heatmg at the point of bad contact, which will Us>ually be at the point the armature coil is joined to the commutator lugs. In case of an armature that has been heavily overloaded, it may be found that the solder in the lug connections at the commutator, has melted and thus short circuited the coils, in this case clean out all the loose solder and removes all solder that make connections between commutator segments, so as to short circuit them. Ring armatures are much easier to locate trouble on than drum armatures, but it is hardly advisable for anyone but on experienced man to take any armature and try to repair burnt out coils. 94 TROUBIvE IN DYNAMOS. In case it is found that a series arc liglit dynamo will not generate current after having been started up to speed, the first thing to determine is whether your cir- cuit to your lamps is "closed" or not. If it is "open", that is, if the circuit is not complete, the break should at once be located. In case a circuit seems to be partially open, which may be the case where several bad contacts or a defective lamp are in the circuit, it will often be possible to "ring" through it by means of a testing mag- neto, and from the fact that the ring of the magneto is much fainter than it should be, we know that the circuit contains much more resistance than usual. The usual testing magnetos as has been previously described, is a simple alternating current s^enerator of small size oper- ated by hand, which when orenerating current will ring a small bell. The winding of its armature is of fine wire and will generate sufiicient current to ring the bell when connected to a circuit containing as high as 20,000 or 30000 ohms resistance. Since the usual arc light circuit sel- dom has a resistance of over 500 ohms, and owing to the fact that a series wound dynamo will not "build up" on a circuit having any unusually high resistance, it follows that, although we may be able to ring through a circuit, it may be impossible to start the dynamo on it. In many cases of this kind current may be started through the cir- cuit by connecting the dynamo to the circuit and then short circuiting the dynamo by means of a piece of wire until it commences to generate current and then sudden- ly opening the short circuit where the momentary rise in pressure at the dynamo terminals due to its self-induction will start the dynamo current through the circuit. This plan should only be used as a last resort. Series dyna- TROUBI.K IN DYNAMOS. 95 mos for arc lighting are usually provided with a switch which short circuits the series coils forming the field magnets and thus shuts down the dynamo by destroying the magnetism of the fields. This is the proper way to shut down a series arc dynamo, for if the circuit was broken while the dynamo was in operation, the rise in voltage due to the self induction of the dynamo and circuit is likely to injure the armature. For this same reason the field circuit of the usual shunt wound dynamo should never be broken while in operation, for the self induction "discharge" from the field magnet circuit is likely to injure the insulation of the machine. In stopping a shunt dynamo after a run, the brushes are often lifted from the commutators by careless dyna- mo tenders, while the field magnet windings are still receiving current from the armature before the armature has ceased revolving and the flash seen at the commuta- tor in such cases shows conclusively that the field coil insulator m.ust be undergoing a much greater strain than when the dynamo is generating its maximum voltage. A separately excited alternating current arc light dynamo is usually provided with an arrangement which acts as a safety valve for the excessive extra voltage caused by an open circuit in the line. It consists of two pointed carbons connected to opposite terminals of the dynamo. The points of the carbons, between which the maximum voltage of the dynamo will be found, are sep- arated a slight distance, the distance being such that when the dynamo generates an excessive pressure, the current will jump across the points and thus short-circuit the dynamo and save it from possible damage. The two well known types of open coil armature dyna- 96 TROUBI^K IN DYNAMOS. mos for arc lighting are the Brush and Thomson-Houston makes. The regulation of the Brush arc light dynamo is accurate and positive, and is performed in a simple manner. The field magnets which are of the series type and are provided with a shunt circuit of variable resistance between their terminals. In other words, the current in passing from the armature to the lamps has two paths to divide between. The amount of current in either path v/ill depend on the relative resistance of the two circuits and the amount of current through the field circuits may thus be varied to suit the load which the dynamo is to have. The circuit in multiple with the field magnets has its resistance varied by means of the ''Dial regulktor" which consists in its simplest form, of a magnet in the form of a selonoid through which the main circuit passes. The core of the selonoid is attached to a lever which varies the mechanical pressure on several piles of thin carbon plates, the usual form of dial regulator ha\dng four piles or columns of carbon plates in series, whose resistance will vary with the increase or decrease of pressure applied to them by the lever attached to the iron core of the selon- oid. These piles of carbon plates are thus capable of hav. ing their resistance varied and being in multiple with the series coils whose resistance is constant, we have a means of regulation. This style of regulator, when given proper attention, keeps the current practically constant at all loads, and should never be allowed to get out of repair. From time to time carbon plates may have to be added, and the contacts and moving parts of the regulator should be inspected often, to insure proper working when needed. troubi^e: in dynamos. 97 Thomson-Houston arc dynamos are very largely used in the United States, and should always give satisfaction if kept clean and dry. The armature consists of but three coils, which in the later type, are wound on a large iron ring and are thus known as ring armatures, although electrically, they do do not resemble the Gramme wind- ing in the least. The three coils are placed equi-distant from each other on the ring and each of the three main coils are divided into ten smaller coils in series, five of which are placed on either side of the ring diametrically opposite each other but connected in series. The three terminals of the three main coils on the pulley end of the armature, are all connected to a metal ring which serves as a common junction for the three main coils The re- maining three terminals at the commutator ends are con- nected to the three commutator segments. The brushes, four in number, are in pairs, the leading brushes of each pair being called the secondary brushes and the trailing brushes are called the primaiy brushes. The brushes should be set with great care and also the nozzles to the air blast whose office is to direct a blast of air at the point of the brush just as the brush is passing from one seg- ment of the commutator to another and thus reduce the sparking. The regulation is effected by moving the brushes slightly in direction of rotation and also moving the primary and secondary brushes of each pair with relation to each other, which tends to allow the armature to generate more or less current. The usual Thomson- Houston dynamo will not run safely for any length of time on less than about ^ its maximum load unless pro- vided with a ''light load switch" which simply cuts out a portion of the field winding of the dynamo and thus weakens the magnetism. gS TROUBIvB) IN DYNAMOS. The spark at the commutator should vary from >^ to X^ inches long, depending on the load. At full load the spark should be about ^^ inches long. Great care should be observed in keeping all dust and dirt from the com- mutator supports and from all moving parts of the regu- lators. The automatic movement of the brushes in the Thomson-Houston dynamos is done by moves of a regu- lator magnet, which is in turn supplied with current intermittently by means of a wall controller located in any convenient position near the dynamo. The Bxcelsior Dynamo regulates by means of an auto- matic cutting in and out of convolutions or windings of the field magnets. The Wood dynamo and the Standard arc light dyna- mos both provide automatic regulation by means of shifting the brushes on the commutator of closed coil Gramme ring armature. The repairs of arc light dynamos are confined gener- ally to the armature and commutator and as a general rule, it will be found that such trouble as may occur, other than that due to over-loads, can be traced directly to carelessness on the part of the dynamo tender, in cleaning dirt and oil from his dynamo. Too much care cannot be bestowed in cleaning and keeping clean all dynamo electric machinery. A careful dynamo tender is a necessary adjunct to any well regulated electric light plant, and a greater mistake cannot be made than to place inexperienced men in charge of high voltage machines. An arc light dynamo may * 'flash" if carrying too heavy a load or if oil or dirt is present on the commuta- tor and brushes. TROUBI^K IN DYNAMOa 99 Arc light dynamos of ail types whicii generate pres- sures of 500 volts and over, should be handled with care, and it is always advisable to use rubber mats to stand on, when adjusting brushes, etc., while the dynamos are running. It is also a good plan to make it a rule never to use both hands if it can be avoided, in handling arc lamps, dynamos, transformers, etc., on live circuits. By placing one hand in the pocket and keeping it there when working on circuits of high pressure, there will be little chance of injury from touching two points at the same time in a circuit having a high difference of poten- tial. It must of course always be understood that it is extremely foolhardy to touch any part of a lamp or wire on a live circuit, unless standing on a thoroughly insu- lated floor, or in case of outside lamps on live circuit, a wooden box should always be used to stand on when adjusting them. Many fatal accidend have occurred through carelessness or neglect on the part of men not following these directions. Series wound dynamos may have their field magnets partially short circuited by over-heating and this will usually show itself by the machine refusing to generate its usual current. Rewinding the field magnets is the usual remedy. Portions of coils may be cut out by grounding on the core or frame of dynamo in two places. This can be located by means if a magneto or a simple battery and bell. If a shunt wound machine when running separately from other machines, refuses to * 'build up*' open the main switch leading from dynamo, and usually the trouble will be righted, which in this case was probably caused by a short circuit of some nature on the external circuit. lOO TROUBXK IN DYNAMOS. If the machine refuses to generate after opening switch^ look for trouble in socket of the pilot lamp. The reader will understand that when a large motor is not revolving, the resistance is low, and if a shunt dynamo should be started up with this motor connected, it would likely re~ fuse to generate, consequently many motors now made are provided with a magnetic retaining switch which automatically disconnects motor from circuit when cur- rent for some reason, is thrown off. Again there may be an open circuit in the field magnets, or the magnetism (residual) may have become reversed by the close prox- imity of another machine. It will be noticed, if tried, that under these conditions, there will be little magnet- ism exhibited, even less than when machine is not run- ning, being due to the neutralizing effect of the residual and current magnetism. Seek first to understand the principals upon which your machine depends, as it is then more possible to rem- edy the troubles that your machine is subject to. Field magnets, armature, commutator and current collecting devices make the dynamo. Understanding each of these you understand the whole. If the reader will grasp the facts in the preceding chapter, he can much easier cope with the troubles that will likely come. TYPES OF ARC LAMPS. lOI CHAPTER Vin. ARC LAMPS FOR DIRECT AND ALTERNATING CURRENT, INCANDESCENT LAMPS. -• One of the most common uses of the electric current, is for illumination by means of the arc lamp and a rather detailed account of what should be expected of an arc lamp, will be of interest to every man in charge of dyna- mos used for either arc or incandescent lighting. By far, the greater number of arc lamps in use to-day are supplied from constant current dynamos. That is, a dynamo generating practically a constant number of amperes, and a voltage that varies with the number of lamps in the circuit. These dynamos may be of either the direct or alternating current type, although alternat- ing dynamos supplying a constant current for arc light- ing are rather rare. Arc light dynamos are built in sizes that supply current for from i to 125 lamps. A 125 light arc dynamo was exhibited at the World's Fair at Chi- cago, in 1893, built by the Brush Electric Co., and proba- bly is the largest machine yet constructed for operating arc lamps in series. The dynamo was built to be direct- connected to a high speed Willans engine and attracted considerable attention. The voltage of the dynamo mentioned, would be about 6250 volts when running the 125 lamps. The current was 9.6 amperes. Assuming that the reader is fam- tOi 'TYPES OF ARC IvAMPS. iliar from study of previous chapters of the difference between constant current and constant potential dyna- mos, it will be evident that an arc lamp operating on a constant current dynamo, must have a mechanism capa- ble of performing several different functions. The ^'Vol- taic arc^', which is so called from the noted philosopher, Volta, is formed in the usual arc lamp by separating the points of two carbons from ^ to j% inches, through which FIGURE 30. — VOMAIC ARC. curtent is passing and when supplied from a suitable source of electricity, the current instead of being broken, jumps the space between the carbon points and generates in- tense heat, which makes them emit the usual dazzling light, known as the arc light. Figure 30 shows the appearance of the arc as viewed through a pair of dark glasses. The light is not given TYPES OF ARC IvAMPS. I03 bound together so as to pre- vent this vibration. If a coil which is to carry an alternating current is wound on a metal spool, it should be slit so as to prevent FIGURE 32. — DIAGRAM OF CIRCUITS OF HKAT MOVEMENT I,AMP. current from being generated in the spool, which might otherwise act as a secondary coil of a single turn. The feeding mechanism of alternating current arc lamps should be as free from magnets as possible for any mag- net especially with an iron core will have sufficient self induction which, if it be a series coil, will act in a great- TYPKS OF ARC IvAMPS. II3 er or less degree so as to choke bi^ck the current in the circuit. Then again as the same series coil will not act the same on 16,000 alternations per minute as it will on 15,000, the lamps will have to be adjusted different for each variation in the number of alternations found in various plants. The heating effect of an alternating current of a given strength however, is the same as the same amount of direct current, and various methods of feeding arc lamps by means of heat generated either at the arc itself or by passing the main or shunt current through a resistance have been experimented with. The heat generated by passing a current through a resistance will vary with the square of the current, or in proportion to the watts used. Thus if we pass two amperes through five ohms resistance, we will be expending 20 watts (C2R=(2X2)X5=2o, if now, the current is increased to four amperes, we will find that although the current is doubled, that the watts ex- pended and thus the heating effect will be increased four times the watts, being 80. One arc lamp in which the plan of a heat feeding mechanism is thoroughly worked out, is the lamp manufactured on the Nutting *'heat movement'* patents whose feeding mechanism is purely a heat movement. The lamp is also probably the only practical lamp whose feeding mechanism feeds the car- bons continuously while burning, although the rate of feed may vary considerable. The plan of the lamp may be understood from figure 32 which shows the plan of circuits. The lamp uses a shunt circuit for its feeding and separates the carbons in start- ing the arc by pulling down the lower carbon by means of ai series arc drawing magnet in the bottom of the lamp. 114 TYPKS on ARC lyAMPS. The shunt circuit has a total resistance of about 1200 ohms, a portion of which is wound around a heating pin, one end of which is embedded in the surface of a wax disc which is mounted on a shaft and geared to the upper rod. The pin being stationary, the wax disc is not allowed to turn until the heat of the current passing through the heating pin melts the wax around its end. The pin being in the shunt circuit, will carry more or less current as the arc is longer or shorter and as the heat in the pin varies with the square of the current passing through it, the field is very sensitive and con- stant. The end of the pin does not plow a furrow in the wax disc, for the melted wax surrounding the pin fills in behind it at:d thus a wax disc lasts an indefinite length of time. These lamps may be made for either direct or alternating currents of constant current or constant potential type. A very important and at the same time a very little thought of subject in connection with arc light, is that of carbons. A good carbon is an absolute necessity in obtaining good results in arc lamps, and a carbon which will give good results under one condition, may not in another. A soft carbon usually burns faster and gives a steadier and more perfect light than one that is hard. A ''coppered" carbon burns much longer than the same size and make that does not have a copper coating. The higher the quantity of current to be carried in a carbon, the larger its diameter or the harder its texture should be. If the burning carbons are exposed to a wind, they will burn much faster than in a still atmosphere, and for this reason, if for no other, the globe should be added. For constant potential lamps either for direct or alter- TYPKS OF ARC I,AMPS. II5 nating currents, a special carbon should be used. For direct current work, a cored carbon should be used for the upper or positive carbon and a solid carbon for the lower. A cored or treated carbon is usually a moderately hard grade of carbon having a core or hole extending through its length. The core of the carbon is a much softer made carbon than the main carbons. This hole in the positive carbon forms a permanent body and crater, and thus steadies the arc. The core being softer, makes an arc that is longer and less noisy than a hard solid carbon. In alternating current lamps, cored carbons should be used for both upper and lower carbons, to ob- tain the best results. At present, foreign carbons made in Germany and Austria, are being sold in large quantit- ies in America for constant potential lamps. Their ad- vantage over American makes, are longer life, better light, less dust and dirt and altogether a much better made carbon than the American makes. This is largely due to the fact that American carbon manufacturers have not experimented sufficiently as yet, to make as finished a product as the foreign makes. INCAND^CEJNT I-» — t>-^ <2)rWr — ^ t X POJHTS 5 • 6 7 • AHE NOT ' RtjHNma poir^T^ ' 'i:>-^ •J>^ FIGURE 36 — SERIES MUWiPtE STREEl* CAR MOTOR CONNECTIONS. On the second point, ^ of the resistance Rhi and Rh2 is cut out and on the third point, all the resistance in series with the motors is removed, and on the fourth point the field coils are shunted by the resistance Ri and R2. The points 5, 6 and 7^ are not running points and are not marked on top of the controller box or stand. During these points, the motors are being changed from series to TRANSMISSION OF POWER. I35 parallel relation, as shown in point 8, and from this to the maximum speed point at number 10. Motors are known in types, as *'gearless", **single reduction" and ''double reduction", owing to the various methods of connecting the revolving armature to the car axles. A gearless motor is mounted on the axle usually, except in one case where a single large motor is con- nected to the two car wheels by means of connecting rods, such as are used on locomotives. The gearless and single reduction type of motor has been brought out within the last year or so and to-day the majority of roads use the double reduction motors, which although noisy and necessitating large repairs on the double set of gear- ing, are usually of light weight for the power developed and are, if anything, more efficient than the heavier slower speed motors. The single reduction motors of various makes do away to a large extent with the gearing repairs and seem to be the most desirable for the usual street car service. These motors are not excessively heavy and are as a gen- eral rule more efficient than the gearless types. For high speed service, the problem is somewhat dif- ferent and the gearless motors should be well adapted to this service, for the armature speed would be high and the motor quite efficient. Owing to a lack of room under a car for motors, gearless motors are difficult to build of sufficient power and efficiency that will go in the limited space allowed. It should be remembered in this connection, that the high speed motor is a much lighter motor for a given power than a slower speed motor. Thus as the armature speed is reduced, the magnetic field must be increased in 136 TRANSMISSION OF POWER. strength or the armature coils must be increased in size and number, which means a heavier motor. Street car motor repairs are often "heavy and expensive and in many cases are due to careless handling. The usual street car motor of any make or type is always working under disadvantages compared to stationary motors protected from the weather. The street car motor is exposed to dust and dirt in dry weather, and water and mud in wet weather, and owing to the high pressure used, 500 volts, it is often most difhcult to keep armature and fields from grounding and thus disabling them. The frames of the motors are of course connected to the metal car trucks, which are grounded, and in wet weather water or mud may ground or short circuit the fields or armature of the motors. **Bucking" is the usual name given to a violent jerk which often takes place when a motor is grounded. Its act- ion is very much like that of a * 'bucking bronco" and may strip the cogs from the gears or shake up the passengers badly, depending on its severity. If a motor is running at full speed and a ground occurs on the wire between the fields and the armature of the motor, the fields are often made to carry a much greater amount of current than they should. The armature at once begins to act as a generator and ''bucks". A flash from brush to brush across the commutator will often cause the same effect. In case of trouble of this kind, look for a grounded field or brush terminal. If the motor is permanently ground- ed, cut it out and proceed to the barns with one motor. The brushes and commutator of a motor should receive excellent care. In case of sparking, see that your brushes are being pressed against the commutator firmly by the TRANSMISSION OF POWKR. 137 Springs and that they are properly fitted to the commuta- tor surface. Copper coated carbon brushes are superior to those not coppered, for they heat less. Never reverse a car when running if it can be possibly avoided, and then it should be done in a moderate man- ner or the fuse is likely to * 'blow' ' or the cogs to break, and thus effectually stop all chances of stopping suddenly in this way. In going down grades it is bad policy to run at an excessive speed, experience having shown that several of the worst accidents which have ever occurred on electric railways, were caused by run-away cars on grades. The series multiple controller or start- ing switches are provided in some cases with a locking switch, which prevents a motor being reversed while cur- rent is on the motors. In climbing grades, always run on one of the *'best running points", and thus avoid any possible damage from overheating field and armature coils. On a slippery rail, care must be used in starting, to avoid slipping. A moderate use of sand is recommen- ded and in case wheels slip as speed increases, move starting handle back and throw on current again gradu- ally. Go slowly around curves, for your trolley is not only likely to jump off the wire, but broken and dam- aged trucks are often the result of such reckless running. The trolley pole is also likely to break span wires, etc. , of the trolley line. A motorman should never leave the car without first removing the starting handle from starting box. Motor cars, if properly designed, should be able to mount grades of 15% to 18% and several roads in Amer- ica are operating daily on grades of 13% and over. Incandescent lamps in street cars are usually connected 138 TRANSMISSION OF POWER. SO as to place five 100 volt lamps in series. Some cars have one set and others two, and as a general rule electric cars are the best lighted of any cars used for passenger trans- portation in the country. Incandescent lamps, when placed in series, should always be of the same make and candle power, to give the best results. Lamps should not be allowed to become loose in their sockets, for socket repairs are sure to follow. STORAGE BATTERIES. 139 CHAPTER X. STORAGE BATTERIES. Electricity passing through a liquid solution from on^ metal plate to another will produce chemical action, The action will depend on the quantity of current used, the kind of solution and the material of which the plates are made. The history of the storage battery or *• electrical accu^ mulatof dates back to 1 80 1, when one of the scientists of the day noticed that if two plates of the same metal Were immersed in an acid solution and current be passed from one plate to the other, that after they had been dis^ connected, electric currents could be obtained from the plates by connecting them together by a conductor, the current flowing in an opposite direction to that of the current with which the **cell" had been charged. No material progress seems to have been made from this date until 1859, when Planted, while experimenting in this line, made a storage battery consisting of sheets of lead immersed in a solution of dilute sulphuric acid* He found that when currents of electricity were passed though the solution from one plate to the other, that a chemical action was at once set up, tending to change the chemical composition of the lead plates, one of which being connected to the position pole of a primary battery Would gradually assume a reddish color and the other I40 STORAGE BATTKRIKS. remaining practically unchanged. He found that after current had been sent through the battery, that it would exert a counter Klectro Motive Force, (counter B. M. F.) of from 2 to 2.5 volts and that in discharging the cell, that it would show an B. M. F. of about two volts until there had been given back from the cell nearly the amount of current used iu charging it. He also found that by charging a cell and then discharging it, and then revers- ing the cell and charging it in the opposite direction that its storage capacity would be increased to a large extent, and this process of **forming" the plates was always gone through until the plates became porous and would hold a charge of many times what they w^ould at first. This forming was necessarily a long, tedious and expensive operation, and some time later it was discovered by Faure, that if a paste made of oxide of lead .be supplied to a lead supporting plate, called a ' 'grid' ' that the pro- cess of forming was so shortened as to be practically done away with. It also made it possible to reduce the weight of the plates to a great extent and the majority of electrical accumulators or storage batteries used m Amer- ica are now made on this plan. It will be evident that it will be necessary in any bat- tery to provide means for keeping the positive and neg- ative plates from touching each other, and thus short circuiting. Various methods have been tried, a few of which will be mentioned Hard rubber * 'combs" or **hair pins" may be placed on the plates and thus keep them separated from each other. In one type of battery using pasted plates and ' 'grids' ' for supporting the active material, plugs of active material in the negative plates are removed in certain places in the plate and rubber STORAGE BATTERIES. 141 FIGURE 37. — STORAGE BATTERY PIRATES. PASTED PIvATE TYPE. plugs are placed in these openings so as to hold the posi- tive plates away. In other cases a perforated hard rub- ber plate or a sheet of asbestos paper is placed between the plates. Owing to the * 'buckling" of the lead plates, it is often a very difi&cult matter to keep the plates apart, 142 STORAGE BATTERIES. and in case of contact, the cell will at once become dis- charged and very likely injured. A deposit of active material often forms at the bottom of the retaining cell and unless the plates are raised some distance from the bottom of the cell, trouble may arise from a short circuit at this point. It should be understood that no matter how large a single storage cell, either of the Plante^ or Faure type may be, or how many plates it may contain, that its volt- age will never be higher than from 2 to 2.5 volts. The **ampere hour'' capacity will vary however, with the size and number of plates exposed to the solution, and to get a voltage of say, 100 volts, it will always be necessary to connect at least 50 coils of battery in series, each cell having about two volts K. M. F. There is always a max- imum charging rate for a given size plate, which should never be exceeded. A plate when being supplied with more than this rate will be likely to be injured by warp- ing or by being ''buckled" as this bending or warping of the plates in called. The cells are rated on their ' 'ampere hour capacity" and each size of cell with its * 'elements" will have a rate at which it may be charged and dis- charged giving its maximum efficiency. A well known make of storage battery of 150 ampere hour capacity, may be described as as follows: — voltage about two volts — number of plates 23, 11 of which are positive and 12 negative, thus giving each of the positive plates a neg- ative plate either side of it. The size of both positive and negative plates are the same, I2X6X/^ inches thick. The normal charging rate is about 25 amperes and the discharge rate from 25 to 30 amperes. The weight of cell and liquid complete is about 45 pounds. STORAGE BATTERIES. I43 The discharge rate may be slightly increased, but the capacity of the cell will be diminished to a considerable extent. We have stated that the large number of American made storage batteries are manufactured on the ' 'pasted plate'* or Faure principle, but of late many cells of the Plante'' type are being used in America. The mechanical construction however, of the plates is very different from the original Plante^ battery. In one leading make of the Plante'' type of battery, the plates are formed of lead ribbon whose surface has been previously roughened, the ribbons being about ^ to j4 inches wide and placed in a horizontal position between heavy lead end supports. The plate really consists of a large number of thin lead strips, piled one over another until the plate when complete measures in the medium sizes about 6x8X>^ inches thick. A number of such plates are then connected by means of lead lugs on the heavy frame of the plates, and in this way a completed cell is put together, and the plates are now ready for '^forming". This is done in the usual manner, the per- oxide of lead formed from the ribbons fills up the spaces between them and at last forms a practically solid plate of * 'active material" as the peroxide of lead is called. There is alwa3^s supposed to be enough of the lead ribbon left to form a support for the active material and when such batteries are properly cared for, they should give good results. They will stand a heavy discharge with- out buckling and will withstand considerable hard usage such as is experienced in train lighting, etc. The positive and negative plates of lead batteries, may be easily distinguished by their color, the positive plates 144 STORAGE BATTERIES. being of a reddish color and the "negatives" of a metallic lead color. When in good condition and fully charged, the * ^positives' ' should be of a dark plum color. The solution of sulphuric acid and water in the plates are immersed will be found to vary in its specific gravity with the charge in the cell. When the cells are com- pletely discharged, its specific gravity should be from 1. 15 to 1. 16 and when fully charged, its specific gra\dty will be somewhat greater. The mixture is about ^q acid and f^ water and should be tested after mixing with a hydrometer, which gives the specific gravity. The solu- tion evaporates rapidly when in a cell which is in actual service and water should be added to keep the solution right. If the solution does not contain enough acid, pour in solution already mixed and never pour clear acid in on the plates to increase the specific gravity of the solution. In charging storage batteries, the positive terminal of the dynamo should be connected to the positive terminal of the series of cells. In charging the usual lead storage battery, it will be found that until the batteries are almost charged, the electro motive force of each cell will be from 2 to 2.1 volts, but as the charging progresses, the voltage may rise as high as 2.3 volts, but when the charging current is stopped, the voltage falls to about two volts. A low reading voltmeter should be used in testing storage bat- teries and the terminal reading of a single cell is usually a correct showing of the condition of the cell. If it is found that a single cell tests lower than the others in series with it, the cell should be removed and exam- ined. It may be found that the plates are buckled and in this case they must be straightened again by mechani- cal means. STORAGE BATTERIES. 145 The negative plates as a general rule, do not need renewals, but the positive plates are often subject to repairs which cost at least lo fo per annum of the original cost of the cell. In many cases of train lighting, the positive plates last but a year on an average, but this service is very severe. The connections between the batteries and in fact, all corrodible parts of the battery plant should be liberally treated with asphalt paint. All connections should be made in a strong and servicable manner and unusual care taken in insulating all parts of a battery which is to be charged from a high voltage constant cuirent circuit, When charging from a constant potential circuit, a resis- tance is usually put in series with a set of cells, so as to keep the charging current uniform. An automatic safety cutout should also be provided, so as to cut out the batteries in case of a dynamo being stopped while connected to the storage batteries, for if this was not done, the dynamo would be supplied with current and run as a motor. A shunt wound dynamo is better adapted to storage battery charging, than the series wound dynamo for several reasons one reason being, that it is not easily reversed by failure of cutouts working, etc. Although we have spoken of the lead type of stor- age battery only, there are several other types which are worthy of considerable study. One of these is called the * 'alkaline' * accumulator, and has for its positive plates, a mass of finely divided copper surrounded by a copper wire gauze which holds the copper in position. The plates are then placed in an iron containing cell, which is so constructed that iron partitions come between the 146 STORAGE BATTERIES. positive plates, but are held away from them. The solu- tion used is one in which potash is dissolved and before the cell is ready for use, a quantity of zinc is dissolved in the solution and held in suspension in it. When the battery is charged, the zinc is deposited on the iron case and partitions between the positive plates, and the cop- per in the positive plates is oxydized and the electro motive force of the cell will be found to be about -f^ volts, much lower than the lead types of battery. When the cell is discharged, the copper oxide is reduced and the zinc on the iron partitions is again dissolved in the pot- ash solution. Although the voltage of the cell is low, its current capacity is high and a cell capable of furnishing about 300 watt hours, will weigh but ^ as much as the usual lead cell. The K. M. F. of this type of cell is quite constant and owing to its small weight and size as com- pared to the same capacity of lead cell, its application to street car propulsion is to be watched with interest. Storage batteries cannot be charged by means of the alternating current, a fact which is at once evident when it is remembered that with the current reversing its direction many times a second, a chemical action such as is necessary in any storage battery, would be out of the question. It will thus be seen that the storage battery of any type will alw^ays be used in connection with direct current stations, and their value is now becoming gener- ally known in America and Europe. There are many electric light stations supplying low pressure direct cur- rent for incandescent lamps, etc., in large cities where current must be supplied at all hours of the day or night. In such a station it will usually be found that during the brightest hours of the day and between mid- STORAGE BATTKRIKS. 1 47 night and morning, that the load on the station is very light, so light in fact that the smallest dynamos and engines in the station may be under-loaded. We will find in many such cases as this, that a set of storage batteries may be installed and effect quite a sav- ing in the operating expenses of such a plant. During the hours of the day when the load is smallest, the stor- age batteries may be charged. Then as the heavy load comes during the early hours of the evening, the batter ies may be connected so as to help furnish current to the circuit and later after the load has lessened to the capac- ity of the battery, the engines and dynamos may be stopped, and the batteries will furnish the necessary current until morning, when the dynamos are started to carry the daily load. Thus the running hours of the station are not only shortened, but as the dydamos, dur- ing the day, are carrying a load nearer their maximum, both the engines and dynamos should operate at a higher efi&ciency. By this means, a considerable saving can often be effected, and many central stations and office buildings having their own plants, are now using storage batteries to obtain these results. By means of the storage battery, a smaller dynamo may be made to furnish current for a much larger num- ber of lamps, for a limited time, than it would be able to carry alone. This is well illustrated in the electric lighting plants installed on some of the steam railroad trains in America. A dynamo, direct connected to a high speed steam engine has a maximum capacity of about 80 amperes at 70 to 80 volts pressure. The incan- . descent lamps used are 16 candle power and are about 200 in number on a six car train and require 148 STORAGE BATTERIKS. about 150 amperes at 64 volts pressure. It will tbus be seen that the lamps on the train require 150 amperes of current of which the dynamo can supply but -80, when working at its full capacity, and since the engine driving the dynamo gets its supply of steam from the locomotive, it is often impossible to get any steam at all in certain sections of the road when all the steam in the boiler must be used to operate the locomotive in pulling the train. Thus it will be seen that to maintain a reliable light, a storage battery is the only means which can be used to obtain good results, under such conditions. In practice 32 cells of 150 ampere hour batteries of the lead type, are generally placed under each car and their voltage will thus be found to be about 64 volts for the set of 32 cells. The dynamo is run continuously, except at such times as the locomotive is disconnected or steam cannot be obtained for other reasons and in this way the batter- ies are always in condition to supply the current needed in additional to that from the dynamos. Train lighting service is very severe and much time and money has been spent in developing practical and successful systems of train lighting. The storage battery, operating a motor, was applied to a larger extent to the propulsion of electric launches on the lagoons of the World's Fair at Chicago in 1893. There is little doubt but that they provided the best means of propulsion that has yet been devised for the particular conditions under which they operated. Sev- eral private pleasure boats both in America and abroad are equipped with storage batteries and motors for power. A French torpedo boat has also attracted considerable attention, whose propelling power comes from storage batteries and motors. STORAGE BATTERIES. 149 FIGURE 38.— TYPICAL STORAGE BATTERY. An ''electric carriage" was also shown at the World's Pair, which was propelled by a motor supplied with cur- rent from a set of storage cells. Considerable work of this kind has been done in Burope. By means of a set of storage batteries from the usual arc lighting circuits of constant current type, incandes- cent lamps may easily be operated on the usual multiple plan. This would hardly be an advisable thing to do 150 STORAGE BAMKRIKS. however, in residence lighting, unless an automatic device prevented the incandescent lamp circuit from being thrown on while the high voltage arc light circuit was charging the cells, for otherwise the handling of the sockets and incandescent lamps might be a dangerous thing to do. The storage battery has also been used to a large ex- tent for operating small motors in phonographs and other automatic machines. Their use in medical and surgical work is also quite general. The application of the storage battery to street railway work is at present far from general, but as suggested in previous chapter, the storage battery system is an ideal one and a fortune awaits the successful investigator in this line. A few points in regard to the care of the usual lead type of accumulator, in addition to those already men- tioned, may be of value to those charging or handling them. Storage battery plates are usually received from the makers after having been formed, and, after having placed the plates in position in their rubber or glass cells and having taken due care in seeing that the positive and negative plates do not come in contact with each other, they should be covered with acid solution of about 1. 17 specific gravity and allowed to stand until the solu- tion has thoroughly entered the pores in the plates. The lead connecting lugs leading up from the positive and negative plates, should be painted with an asphalt paint to keep the acid from attacking the bolts and nuts usually used to connect one cell to another. Care should be taken to scrape the contact surfaces of the lugs and con- STORAGE BATTERIES. 15I sections between the cells, and thus reduce all needless resistance between the batteries. After a connection has been made, it should be painted with an acid and water proof paint or varnish, to prevent corrosion. The cells are connected in series. The positive plate of one cell connected to the negative plates of the next cell and so on. If the battery is to be charged from a constant potential circuit, care must be taken to allow but a safe amount of current to pass through the battery. The amount of current will depend on the counter E. M. F. of the cells and their resistance. Thus if we are to charge from alio volt circuit, we may safely charge from 50 to 55 cells in series, each of the two volts E. M. F. A resistance should always be placed in series with the set of batteries in such cases and in this way be varied to keep the charging current uniform. The amount of current that no volts will put through, say, 50 cells will depend of course on Ohms law, E R But E or the number of volts, will be the difference between the counter E. M. F. of the set of batteries and the 1 10 volt circuit. Thus the voltage of the 50 cells may be 100 volts, in which case E=io or the difference between 100 and 1 10. R or the resistance is likely very low, probably not over ^ ohm, and thus 10 C= — or 50 i which in a cell whose maximum charging rate is but 25 or 30, will be too much. The only way to prevent this excessive flow of current, is to either add extra cells to 152 S'TORAGE BATTERIES. the set or place a variable resistance in series with the set of batteries. The larger the size of the plates and the more their number in a e^iven cell, the lower its resis- tance will be. Knowing the charging and discharging rate of a given battery, and also its voltage and capacity in ampere hours, any problem in the application of the storage battery, may be worked out. New cells should receive several long and steady charges before being put on regular heavy work. The number of amperes multi- plied by the number of hours charged, will give the ampere hour charge and new plates after having been dried out in shipment, should be carefullv charged the first few times. A cell should never be allowed to discharge lower than 1.80 or 1.85 volts, and under no condition should a dis- charged cell be allowed to stand any length of time with- out recharging, for the plates are likely to become coated with a white coating of "sulphate'* which not only injures the plates, but can only be removed by a most careful and tedious process of charging at a low rate. Buckled cells caused by short circuits or heavy charg- ing or discharging, should be taken apart and straight- ened by mechanical means. A heavy deposit of active material or "mud*' may be found in the bottom of a retaining cell and may be enough to short circuit the bottom of the plates. Rubber gloves should be used in handling the plates and solution, and woolen clothing should be worn, for cotton goods are soon destroyed by splashes of battery solution. Ammonia may be applied to discolored cloth and will often counteract the effect of the acid. Great care should be exercised in mixing the sulphuric STORAGE BATTERIES. I53 acid and water. The acid should be poured in the water in small quantities and should be stirred well as it is being mixed. Considerable heat is always generated under such conditions, and the acid should be allowed to cool before passing over the battery plates. 154 HEATING AND MHTAI, WORKING. CHAPtKR XI. E1.ECTRIC HEATING AND METAI, WORKING. STATION INSTRUMENTS. Blectric heating is a subject that at present is interest- ing many able workers in the electrical field. Its advan* tages over coal or gas for heating are many, and its only drawback is its cost when current at the usual lighting rates is used. The principle of all heaters both for direct and alter- nating current is that of passing current through resist- ing conductors which of course consumes energy and ex- hibits itself in heat. The conductors used in the usual heaters for electric street cars, are generally made of Ger- man silver or iron wires, and these wires are in most cases surrounded by some insulating material which is a good conductor of heat, such as fire clay, sand, or enamel. This material really furnishes the wire with a larger heat radiating capacity. It will be evident, for example, that if a heated wire is placed against a plate of cold glass^ that it will at once lower its temperature and gradually raise the temperature of the glass. The wire cannot in this case be raised to a dangerous temperature without passing through it several times the amount of current that would melt it in open air. The wires in the usual electric heater, thus carry a much greater amount of cur- HEATINC AND METAI. WORKING. I55 rent without being over-heated, than would be possible without the radiating material surrounding the wire. In one form of heater, the wires are fastened to an iron plate by means of enamel, the enamel not only complete- ly covering the wires and uniting them to the iron backing but also insulating them from the iron. The wire is in intimate contact with the iron plate by means of the enamel, and of course cannot become much hotter than the iron plates, which having considerable radiating surface, make efficient electric heaters. One of the earliest forms of electric heaters, patented in the United States, used iron or German silver wires imbedded in fire clay, the whole being incased in an iron box. This form of heater was used in the earliest elec- tric street railway put before the public. There are to-day about 200 patents on various forms of electric heating and cooking devices. The usual form of electric street car heater, takes from two to five amperes at 500 volts pressure, and after a street car is once heated, from 1200 to 1500 watts of current will provide enough heat for the coldest weather. In a large electric street railway plant, the current will cost about three cents an hour per car to keep the heaters in operation, and this figure will be found to be little if any more than stoves using anthracite coal for fuel. The heaters are usually placed under the seats and are of course out of the way of passengers. A large number of street railways are now using them. Cooking by means of electricity is being advocated by several companies and without doubt there are many cases where electric cooking devices can be used at a cost of operating about on a par with coal stoves. 156 HKAI^INC AND MEI'AI. WORlClNG. Quite a number of patents have been taken out on heaters designed for alternating current work whose operation depends on the setting up of eddy or secondary currents in cores of coils of wire carrying alternating current. Such a heater would not of course operate on direct current circuits* One of the most interesting applications of heat from electricity is that of metal working and welding. Electric welding machines are at present doing work that would have been practically impossible with forge and hammer. The Thomson electric welding machines use alternating current for welding purposes by sending an alternating current of moderately high pressure through the primary coil of a large converter, the sec- ondary of which furnishes a current of immense volume at a voltage of but a few volts. The pieces of metal to be united in the weld are placed in the secondary circuit by means of clamps, with their ends in contact with each other. The point of contact being the only appreciable resistance in the secondary circuit, the ends are at once raised to a high temperature. The cur- rent is then increased in the primary, and the junction of the two pieces of metal to be welded, is raised to a welding heat. While this heating is in progress, pres- sure is being applied so as to press the pieces to be welded into even more intimate contact. The whole operation of welding a large bar of iron occupies but a few seconds and the joint made, in many cases is found to be the strongest part of the bar. Many metals may be welded in this way which are very difficult or practically impossible to weld in any other way. Wrought iron pipe bent in various awkward shapes, may be united in this STATION INSTRUMENTS X57 way in a perfect manner, an operation which is often- times very expensive when done in the usual manner. Large crossing frogs and steel rails are often welded on the electrical welder. The intense heat of the voltaic arc is used to some extent in metal working. The usual plan is to make the metal on which the work is to be done, one pole, and a carbon provided with a flexible conducting cord and handle as the other pole, and form the arc between the metal body and the carbon. If a piece of metal be connected to one pole of a suitable source of current supply, and a pail of salt water be con nected to the other, it will be found that by dipping the end of the metal in the water that it may be raised to a white heat in a few minutes, the water still remaining cool. This may seem impossible at first thought, but neverthe- less a fact. A large number of small arcs probably form between the metal and the water, and with metal pieces of proper size, they may be quickly raised to a high heat SWITCH BOARD AND STATION INSTRUMENTS. All electrical machinery should, when performing its usual duty, be capable of being controlled, started and stopped in an exact and simple manner, and to know whether a given dynamo or motor is performing its duty, there must necessarily be connected suitable measuring instruments. The proper fitting of a station switchboard is an extremely important consideration, for in many cases, without the use of simple and reliable means of dynamo regulation and control, a station could never perform its proper work. What we cannot see being devel- oped in machinery, we must have indicated by some means. 158 STATION INSTRUMENTS. Every electric light or power station should be provid- ed with all instruments that are necessary for the gov- erning and regulating of its machinery. There should be instruments which indicate the amount of loid on the dynamos in amperes, also their pressure in volts. The dynamo regulating apparatus may be either auto- matic or performed by means of rheostats, etc. Ground detectors should be used to detect or locate contacts between the wiring or dynamos and the ground. To prevent damage from lighting in the station, lighting arresters are placed on the lines exposed. Switches should be provided to connect the dynamos to the cir- cuits or to make various combinations of the dynamos and thus get various currents. Fuse or magnetic cutouts are used to prevent a load being applied to the dynamos beyond their maximum capacities. The switch boards themselves should be made of a non-combustible insulating material, such as marble or slate free from metallic veins, marble being the best possible material usually, for it has high insulating qualities and does not crack or chip as easily as the usual grade of slate generally used. **Marbleized'* slate how- ever, is much superior to the usual slate and is largely used. In the score of economy, wooden switch boards are often placed in otherwise first class plants. An oak or pine switch board in a plant using low voltage cur- rent, may undoubtedly be made reasonably safe, but it is an exceptional case when one is found, and as a rule a wooden switch board for high potential circuits, when made safe, will cost nearly as much as a slate or marble board. STATION INSTRUMENTS. I59 There should always be from two to three feet space behind a switch board and it should be kept free from waste material from the plant. Station electricians often pile or throw everything imaginable behind them and when trouble comes behind the board, it is a hard job to do anything in a quick manner. Many of the larger electric light companies are building *and selling very superior switch boards at reasonable figures. Rheostats or Field regulators used with shunt or com- pound wound dynamos pro^dde means of regulating their output by varying the current through the field windings. They usually consist of a series of resistances in the form of German silver or iron wire coils that are connected at several points to contacts on the face of the rheostat, and by means of a contact brush rubbing on their surfaces^ more or less resistance is put in series with the dynamo field circuit. Rheostats of this description are very clumsy, and a better type now produced, is the enamel or cement rheo- stat in which the resistance wires are imbedded in cement or enamel, only a small amount of wire being required, and that very small in size, as it is well known that a wire imbedded in such a manner will carry a current several times greater than in open air. They occupy but little room and are compact and fire proof. Quite often in central stations where circuits are une- qually loaded, it becomes necessary to raise the potential on individual feeders. To increase the potential of the dynamo would not suffice because circuits having a light load would have too high a pressure. The '^booster" for direct current circuits consist of a small series dynamo placed in series with the circuit whose pressure is to be l6o STATION INSTRtJMEJNXS. raised. The conductors on its fields and armature are suflBciently large enough to carry full current. An increase of current in the series field would mean an increase in potential at the armature and this added to the potential of the generator gives the desirable pres- sure. This machine increases the pressure automatically as the current increases. For alternating circuits, this scheme is not possible, but the flexibility of the transformers is admirably utilized by Mr. ly. B. Stillwell in the Stillwell regulator and described by him as follows: **If each supply circuit receives current from an independent generator, that is, a generator which is called upon to furnish current to other supply circuits, the necessary adjustment of pressure is obtained by regulating the field charge of the generator by means of the rheostat provided for that purpose. If however, several supply circuits are receiv- ing current from the same generator, it becomes neces- sary to provide means for adjusting the pressure of each without disturbing the others. The regulator consists of a transformer having a secondary coil adjustable in length. Connections are brought out from different points on the secondary coil, to a multi point switch, by means of which the secondary coil, or any portion of it^ may at will, be thrown in series with the supply circuit. When this is done, the electro-motive force due to the whole or a part of the secondary coil of the regulator is added to the initial potential of the circuit. The potential of the supply circuit may therefore be acurately adjusted, independent of. whatever may be the potential at the terminals of the generator' ' . An instrument, the Compensator, is always used with STATION INSTRUMKNTS. l6l the regulator and in the same circuit. It consists of a small transformer which supplies current to the volt- meter. The primary circuit has two windings, one of which is on the usual high pressure constant potential circuit and the other is a winding in series with the circuit whose voltage is to be measured. The secondary circuit supplies current to the voltmeter and when current is flowing, a current is induced on the secondary coil from the primary, which causes voltage to be shown at the voltmeter, corresponding correctly to the voltage at the end of the line vdth that current. To sum it up, the compensator acts upon the voltmeter to give the potential at the end of the line. Voltmeters and ammeters are of two general types, those whose reading is due to magnetic effects and those whose reading depends on the expansion and contraction of a wire due to current passing through, and heating it. The measuring instruments using magnetism, are of various types, some of them using a simple selonoid of wire acting on a movable iron core, to which is attached the indicating pointer. In instruments of this type for use on alternating current, the selonoid spool, if made of metal, is always slit to prevent the spool acting as a sec- ondary coil of low resistance, in which currents would be generated by the passage of current through the coil windings. The iron core of such a coil would have to be laminated, or built up of small iron wires to prevent cur- rents being generated in it. Other magnetic instruments use the effect obtained by mounting a small armature between the pole pieces of a permanent horse-shoe magnet, and sending the current to be measured through the armature, which tends to 1 62 STATION INSTRUMENTS. revolve on its shaft and thus produce a movement which gives the reading. Such instruments are usually pro- vided with jeweled bearings and are quite expensive, but the leading measuring instruments of this type, the Weston ammeters and voltmeters, are the standard instru- ments to-day in America for direct current measuring. The **hot wire'^ instruments are mainly used for alter- nating current work, for since the heating effect of a given current is the same for either direct or alternating current, it follows that such an instrument should be well adapted to the measurement of alternating current. All high grade instruments should be very carefully handled and in case repairs are needed, it should.be done only by one thoroughly acquainted with the work. The rougher classes of cheap instruments are ofted found to be incorrect and it is always policy to calibrate them by means of a standard instrument as often as possible. In selecting a switch board ammeter or voltmeter, an illuminated scale with large figures is preferable. Dead- beat instruments should be used as much as possible as an instrument whose pointer comes at once to the correct reading and stays there without needless swinging, saves time, and is by far preferable to those whose needle swings to and fro before coming to the exact reading. All instruments should be placed in such a position as to be easily seen by the dynamo tender, but should not be placed in such close proximity to a dyna- mo^ as to have its magnetism effect the reading. In case of it being impossible to place them away from the vicin- ity of a dynamo, they should be provided with magnetic shields, which may be made in various forms. Voltmeters should be chosen having as high a resis- STATION INSTRUMENTS. tance as possible, and ammeters should have the least possible resistance, for it will be found that station instruments often take many times more current to operate them than should be used on proper instnmients. In placing instruments on a switch board, care should be taken to so place them that their needles or pointers will be at zero when no current is flowing. Direct read- ing instruments are always to be preferred to those read- ing in ^*degrees'\ etc. A voltmeter should read directly in volts, and an ammeter in amperes, or a resistance measuring instrument in ohms. All electric light or power stations having conductors in the open air, must have devices to protect the dynamos from injury from lightning. It should not be understood that lightning must actually strike a line to injure the apparatus connected to it. The majority of cases of trouble from lightning occur from currents of high volt- age induced in the line by the passage of lightning through the air near or parallel to the line. The voltage is generally very high, and ruined armatures and field coils result unless means for protection are employed. In many cases the actual damage to the dynamo is caused by the dynamo current following the high voltage lightning discharge, and the damage is done before the dynamo can be stopped. A great deal of time and ingen- uity has been spent in devising various lightning arrest- ers. To be reliable, a lightning arrester should always be ready to operate. It should allow the lightning to pass to the ground, but at the same time prevent the dy- namo current following. The lightning takes the path of least resistance to ground and will of course break through the system at its weakest point. STATluN INSTRUMENTS. The term * 'resistance' ' as here used does not necessar- ily mean the ohmic resistance but the sum of the ohmic resistance and the impedance due to the self-induction of the circuit. A lightning discharge rather than pass through a coil of even very low resistance will often jump a large air gap and pass to the ground. The lightning arrester usually places a small air gap between the systems and the ground, and this is designed to be the path of least resistance to ground. After a dis- charge takes place across the air gap to ground the next operation is to interrupt the dynamo current which we have said, usually follows. This is accomplivshed in various ways, one of the most common being to place the air gap near the poles of a small electro magnet, which * 'blows" out the arc by means of the magnetism, it being a well known fact that if a magnet is placed near an arc so that the arc is in the magnetic field that the arc will be apparently blown aside as if it were in a strong current of air. By using a strong magnetic field in this way, an arc may be instantly inter- rupted and blown out. Another method is to have the air gap over which the arc would start, inclosed in an air tight box and as soon as an arc is started, the confined air immediately expands due to the heat of the arc, and operates suitable mechanism for breaking the circuit. The air gap may be made between two terminals made of non arcing metal and thus make it impossible to main- tain an arc. The non-arcing metal is an alloy lately dis- covered which apparently on being melted by the arc, forms a gas having a high resistance, for a few small air gaps in series between pieces of this alloy, will rupture an arc on the highest pressure used for commercial elec- tric lighting. 5 z UJ o u. u. UJ 1- z Ul ce UJ Q. O 05 So ^> ^^^O lOCO-r-KMGO §1 s| 5^^ GfO d CC CJ CQ O »C t- O 05 >^ O z UJ o u. u. UJ »- z U! o oe UJ 0. 05 i| S ^ S S r^ t- ^ l; t^ »-- ^ 03 ^CO^t-^^g.^0 ^§fc2§??OGCiC:0-*05 -^^s^a^^fsss §1 05G0 50CO OOir-iOT-(0^rO«OOCO 'r^COt-'^THiOt^i-icCO^GO n ci§0^t-»C0500iO 4 THCOt^»Oi-it^CO»pt-iC^05 >^ o z UJ o u. u. UJ 1- z Ul o oe UJ Q. 1 so G^'O ZU o > i| g| 1 31 o Ml 1-1 05 CC lO 00 O iO o ^ >- « z UJ O U. U. UJ H- z UJ o QC UJ a. ifl CO S^ 03iC^5 ?D CO CO 05 r*^ CO CO '•r-i05^diCCC'-^0500'+*O T-l T-l 03 '^ o t- o ro C0CO05'«*iC!D05 05 1-11-105 05 §1' •^ OS t- XO 05 -«* r-i05iOi-ii-i»O0Di-l^05O5CO i-l05C0»OO5i-lt-0500 s| c; t- -* Ci cc oc 05iC^05iCOCOCOC5iaoO^ r-l05^t-T^0CO5^iCt- tH T-i 05 CO -* O >- o z UJ u. u. Ul 1- z iJ QC UJ Q. lO i| ^ ^^ §_ 05 C5 05 C5 CO 05 CO C5 'i-l?0xoC5-^o:^o:O5 S^Si;:COiOxOi-l05tr-COC5 S| ^^»boOi-li-l05CO ^0ci^0.-0C|:^05O||| 8?' 05-^O5Oi CXJt-CO l-105-*O5O;C:C5'*O5COC5t- i-l05-*t-Oi-*C5'-t< 1-11-105 S| '■Jti 05 Ci 00 t- iC 05 05^cJaiaiciC5coo5ccoit- r-iroioos-rt-oscsoso: 1-1 1-1 05 CO Ml > z UJ C£ U. U. UJ H- z UJ oe UJ a. s s^ 05 ^ Ci CO' t- CO CO 05 CO OS CO XO i-ICOOOS-rtlCOt-t-CO r-^i-H05CO^ i| xC0^05^CO OSCOCOt-CO •r^0i^'cc505^CO^«^g 1^ Oi-i05^0S'.*C0J>»t- 1-1 T-l 05 1^ iHCOCOCOO lO ^oi^dcc=od^«|g| 4 CO CO CO CO CO o Hi o ^ S £ 1-H05C0U5Q0O10O10 COPPER WIRE TABLE. Giving weights, lengths and resistances for A. W. G. TBrown and Sharpe) Guage. GUAGE), To the nearest fourth significant diofit. WEIGHT. I.KNGTJI. RESISTANCE. O '/ Diam- eter. Inches. Area Lrbs. per Ft. per Ohms Ohms ^^ Circu- lybs. per Ohm. Feet Ohm. per lb. per foot. lar foot. @ 50° C. per lb. @50°C. @ 50° C. (^ 50° C. <^ mils. 123° Fah 132°Fah 123° Fah. 123° Fah. 0000 0.460 211.600 0.6405 11.720 1.561 18,290 0.00008535 0.00005467 000 0.4096 167,800 0.5080 7,369 1.969 14,510 0.0001357 0.00006893 00 0.3648 133.100 0.4038 4,634 2.482 11,500 0.0002158 0.00008693 0.3249 105,500 0.3195 2,914 3.130 9,123 0.0003431 0.0001096 1 0.2893 83.690 0.2533 1,833 3.947 7,235 0.0005456 0.0001383 2 0.2576 66,370 0.2009 1,153 4.977 5,738 0.0008675 0.0001743 3 0.2294 52,630 0.1593 725.0 6.376 4,550 0.001379 0.0003198 4 0.2043 41,740 0.1264 455.9 7.914 3,608 0.002193 0.0003771 6 0.1819 33,100 0.1002 286.7 9.980 2,862 0.003487 0.0003495 6 0.1620 26.250 0.07946 180.3 12.58 3,369 0.005545 0.0004406 7 0.1443 20,820 0.06302 113.4 15.87 1,800 0.008817 0.0005656 8 0.1285 16 510 0.04998 71.33 30.01 1,437 0.01403 0.0007007 9 0.1144 13,090 0.03963 44.86 25.33 1,133 0.03339 0.0008835 10 0.1019 10,380 0.03143 28.21 31.83 897.6 0.03545 0.001114 11 0.09074 8,234 0.02493 17.74 40.13 711.8 0.06636 0.001405 12 0.08081 6,530 0.01977 11.16 60.59 564.6 0.08963 0.001771 13 0.07196 6,178 0.01568 7.017 63.79 447.7 0.1435 0.003334 14 0.06408 4,107 0.01243 4.413 80.44 355.0 0.3366 0.003817 15 0.05707 3,257 0.009858 2.776 101.4 381.5 0.3603 0.003553 16 0.05082 2.583 0.007818 1.746 137.9 333.3 0.5739 0.004479 17 0.04526 2,048 0.006200 1.098 161.3 177.1 0.9109 0.005648 18 0.04030 1,624 0.004917 0.6904 203.4 140.4 1.448 0.007133 19 0.03589 1,288 0.003899 0.4342 256.5 111.4 2.303 0.008980 20 0.03196 1,022 0.003093 0.2731 323.4 88.31 3.663 0.01133 21 0.02846 810.1 0.002452 0.1717 407.8 70.03 5.823 0-01428 22 0.02535 6424 0.001945 0.1080 614.3 65.54 9.259 0.01801 23 0.02257 609.5 0.001543 0.06793 648.4 44.04 14.73 0.02271 24 0.02010 404.0 0.001333 0.04273 817.6 34.93 33.41 0.03863 25 0.01790 320.4 0.0009699 0.03687 1,031 37.70 37.33 0.03610 36 0.01594 254.1 0.0007692 0.01690 1,300 31.97 59.18 0.04553 27 0.0142 201.5 0.0006100 0.01063 1,639 17.42 94.11 0.05740 28 0.01264 159.8 0.0004837 0-006683 2 067 13.83 149.6 0.07239 29 0.01126 126.7 0.0003836 0.004303 3,607 10.96 237.9 0.09128 30 0.01003 100.5 0.0003043 0.003643 3,287 8.688 378.3 0.1151 31 0.008928 79.70 0.0002413 0.001663 4,145 6.890 601.6 0.1451 32 0.007950 63.21 0.0001913 0.001045 5,227 54.64 956.5 0.1830 33 0.007080 50.13 0.0001517 0.0006575 6,591 4.333 1,621 0.2308 34 0.006305 39.75 0.0001203 0.0004135 8,311 3.436 3,418 0.2910 35 0.005615 31.53 0.00009543 0.0003601 10,480 3.735 3.845 0.3669 36 0.0050 25.0 0.00007568 0.0001636 13,210 2.161 6,114 0.4637 37 0.004453 19.83 0.00006001 0.0001039 .16,660 1.714 9,722 0.5835 38 0.003965 15.73 0.00004759 0.00006454 21,010 1.359 15,490 0.7357 39 0.003531 13.47 0.00003774 0.00004068 26.500 1.078 24,580 0.9377 40 0.003145 9.888 0.00002993 0.00002')59 33.410 0.8-548 39.080 i.rc Specific gravity of copper==8.89. Resistance in terms of the international ohm, from American Institute of Electrical Engineers Transaction Oct. 1893- COPPER WIRK DATA. SAFE CARRYING CAPACITY TABLE. 167 Below is a table showing- the safe carrying capacity of different sizes in Birmingham, Brown & Sharpe, and Edison guages, which must be followed in the placing of interior conductors, Taken from recommendations of the Underwriters Klectrical Association. Brown & Sharpe Birmingham. Edison StandaxId Guage Guage Guag- - No. Amperes No. Amperes No. Ampere 0000 175 0000 ....175 200 175 000 145 000 . ... 150 180 100 00 .120 00 . . . ....130 140 135 100 . . . . .. no no no I - 95 I . . . . •... 95 90 95 2 70 2. . . . .... 85 80 85 3 60 3 • • • .-.. 75 65 75 4 50 4.... ... 65 55 65 5 45 5.... .... 60 50 60 6 35 6.... .... 50 40 ....... 50 7 30 7.... .... 45 30 40 8 25 8... .... 35 25 35 10 20 10 .... 30 20 30 12 15 12 ... 20 12 20 14 10 14 ... .... 15 8 15 16 5 16.... 10 5 10 18 3 18... . •• 5 3 5 20 .... 3 2 3 Outside overhead conductors may carry from two to three times the amount g^iven above, without being dan- gerously warm, provided they do not have an insulating covering which always acts as a wire conductor of heat, and prevents radiation.