®1|E ^. p. ^ill ^ibrar^ 2fartii Caralina State CnUtae *TKI46 H3 V. 3 7157 This book must not be taken from the Library building. J Al^»^ THE THOUGHT IS IN THE QUESTION THE INFORMATION 18 IN THE ANSWER riAWKINs ELKJRia mt ^NSWER^ ILLUSTRATIONS A PROGRESSIVE COURSE OF STUDY FOR ENGINEERS. ELECTRICIANS, STUDENTS AND THOSE DESIRING TO ACQUIRE A WORKING KNOWLEDGE OF aEaRiciTv m it5 melons A PRACTICAL TREATISE ® HAWKINS^^ND STAFF THEO- AyDEL\& C0^^72 FIFTH AVE. NEW YORK. COPYRIGHTED, 1914. BY THEO. AUDEL & CO., New York. I'finted in the United States. TABLE OF CONTENTS; GUIDE NO. 3 ,ry~'" TABLE OF CONTENTS GUIDE NO. 3. GALVANOMETERS 431 to 464 Action of compass needle — simple galvanometer — difference between galvanoscope and galvanometer — sen- sibility — action of short and long coil galvanometers — classes of galvanometer — astatic galvanometer — tan- gent galvanometer — graduation of tangent galvanometer scale — table of galvanometer constants — mechanical ex- planation of tangent law — sine galvanometer — table of natural sines and tangents — • comparison of sine and tangent galvanometers — differential galvanometer — - ballistic gal- vanometer — kick — damping effect — use of mirrors in galvanometers — lamp and scale — damping — D'Arson- val galvanometer: construction, operation; uses — gal- vanometer constant or figure of merit — shunts. TESTING AND TESTING APPARATUS- 465 to 536 Pressure measurement — Clark cell — Weston cadmium cell — pressure measurement error with ordinary voltmeter — International volt — -hydraulic analogy of amperes — coul- ombs — current measurement — International ampere — voltameters — Ohm's law and the ohm — International ohm — ohm table — • practical standards of resistance — various methods of resistance measurement^ — direct deflection method — method of substitution — resistance box — fall of potential method — differential galvanometer method — ■ drop method • — voltmeter method ■ — Wheat- stone bridge — usual arrangement of resistances of Wheat- stone bridge — ratio coils of Wheatstone bridge — the de- cade plan — two plug arrangement — "plug out" and "plug in" type of resistance box — testing sets — direct deflection method with Queen Acme set — ohmeter — fall 7/^7 TABLE OF CONTENTS; GUIDE NO. 3 TESTING AND TESTING AVVXKXTY?,— Continued. of potential method with Queen Acme set — apparatus for measuring low resistances — how to check a volt- meter — Kelvin bridge — Queen slide wire bridge — in- ternal resistance measurement — Evershed portable ohm- meter set — L and N fault finder — ammeter test — dia- gram of Queen standard potentiometer — diagrams illustrating loop testing — the Murray loop — the Varley loop — special loop — the potentiometer — location of opens — to pick out faulty wires in a cable — voltage of cell measurement with potentiometer — care of potentio- meter — location of faults where the loop is composed of cables of different cross sections. AMMETERS, VOLTMETERS, AND WATTMETERS 537 to 572 Definition of ammeter — classification of ammeter and voltmeters — moving iron type instrument — Keystone voltmeter — winding in ammeters and volts — connections for series and shunt ammeters — voltmeter connections — Westinghouse ammeter shunts — various types of in- strument — plunger type instnmient — magnetic vane in- stnmient — inclined coil instniment — Whitney hot wire instruments — principle of electrostatic instruments — multipliers — portable shunts — Siemens electro-dyna- mometer — station instruments — Thompson watt hour meter — how to read a meter — installation of watt- meters — Westinghouse watt hour meter — Thompson prepa^-ment watt hour meter — how to test a meter — Sangamo watt hour meter — Coliunbia watt hour meter — Duncan watt hour meter. OPERATION OF DYNAMOS - - - - 573 to 596 Before starting a dynamo — adjusting the brushes — brush position — how to set the brushes — method of soldering cable to carbon brush — brush contact pressure — direction of rotation — method of winding cables with marlin — method of assembling core discs — starting a dynamo — tinning block for electric soldering tool — shunt dynamos in parallel — shunt dynamos on three wire system — how to start a series machine — the term "build up" — how to start a shunt or compound machine — "pick- ing up" — indication of reversed connections — how to TABLE OF CONTENTS; GUIDE NO. S OPERATION OF DYNAMOS— Continued. correct reversed polarity — finding the reversed coil — loss of residual magnetism — remedy for reversed dynamo — attention while running — lead of brushes — method of taking temperature — lubrication — oils — allowable de- gree of heating — attention to brushes and brush gear. COUPLING OF DYNAMOS - - - - 597 to 610 Series and parallel connections — coupling series dynamos in series; in parallel — equalizer — shunt dynamos in series; in parallel — switching dynamo into and out of parallel — to cut out a machine — dividing the lead — compound dynamos in series; in parallel — equalizer connection — switching a compound djmamo into and out of parallel — equalizing the load — shunt and com- pound dynamos in parallel. DYNAMO FAILS TO EXCITE - - -611 to 622 Various causes — brushes not properly adjusted — de- fective contacts — incorrect adjustment of regulators — speed too low — testing for break — insufficient residual magnetism; remedy — open circuits — test for field circuit breakers — probable location of breaks — Watson armature discs — Fort Wayne commutator truing device — short circuits — Watson armature — wrong connec- tions — reversed field magnetism. ARMATURE TROUBLES 623 to 634 Causes — how avoided — various faults — short circuit in individual coils — location of faulty coil — test for break in armature lead — bar to bar test for open or short circuit in coil or between segments — short circuits between ad- jacent coils — alternate bar test for short circuits between sections — short circuits between sections through frame or core of armature; between sections through binding wires — partial short circuits in armatures — method of testing for breaks — burning of armature coils — Watson field coils — grounds in armatures — method of locating grounded armature coil — magneto test for grounded arma- tures — method of binding armature winding — breaks in armature circuit. TABLE OF CONTENTS; GUIDE NO. S CARE OF THE COMMUTATOR AND BRUSHES - - 635 to 652 Conditions for satisfactory operation — oil for com- mutator — attention to brushes — Bissell brush gear — two kinds of sparking — commutator clamp — causes of sparking — bad adjustment of brushes — rocking — bad condition of brushes — brushes making bad contact — bad condition of commutator — detection of untrue commu- tator — high segments — "flats " — causes of flats; remedy — method of repairing broken joint between commutator segment and lug — segments loose or knocked in — how to re-turn a commutator — Bissell commutators — over- load of dynamo — method of repairing large hole burned in two adjacent bars of a commutator — operating dynamos with metal brushes — indication of excessive voltage — method of smoothing commutator with a stone — causes of excessive voltage — loose connections, terminals, etc., — breaks in armature circuit — sandpaper holder for com- mutator—short circuits, in armature circuits; infield — breaks in field — sandpaper block — short circuits in commutator. HEATING 653 to 662 Various causes — how detected — procedure — heating, of connections; of brushes, commutator and armature — excessive heating — ventilated commutator — self -oiling bearing — some causes of hot bearing — eff'ect of hot bear- ings — points relating to hot bearings — operation above rated voltage and below normal speed — forced system of lubrication — heating of field magnets — causes of eddy currents in pole pieces — detection of mois- ture in field coils — indication of short circuits in field coils. OPERATION OF MOTORS 663 to 696 Before starting a motor — starting a motor — various starting resistances — starting boxes — speed regulators — Cutler Hammer starter — tinie required to start motor — how to start — sliding contact starters — series motors on battery circuits — starting a shunt motor — multiple switch starters — eff'ect of reverse voltage — rheostat with no voltage and overload release — failure to start — starting panel — Cutler Hammer starting rheostats — Allen Bradley automatic starter — Monitor starter with TABLE OF CONTENTS; GUIDE NO. ."? OPERATION OF MOTORS— Continued. relay for push button control — a remote control of shunt motors — regulation of motor speed ; various methods — Monitor printing press controller — speed regulation of series motor, by short circuiting sections of the field winding — varying the speed of shunt and compound motors — Cutler Hammer multiple switch starter — regulation by armature resistance — Compound starter — regulation by shunt field resistance — Holzer Cabot instructions for shunt wound motor — Reliance adjustable speed motor — Cutler Ham- mer reversible starter — combined armature and shunt field control — selection of starters and regulators — Watson commutators — organ blower speed regulator — General Electric controller — speed regulation of traction motors — controller of the Ranch and Lang electric vehicles • — two motor regulation — controller connection dia- grams — stopping a motor. t\y.^. '% W. GA L VA NOME TERS 431 CHAPTER XXVI GALVANOMETERS If a compass needle be allowed to come to rest in its natural position, and a current of electricity be passed through a wire just over it from north to south, the north seeking end of the needle will be deflected toward the east. If the wire be placed under the needle and the current continued from north to south TTdTT^ Fig. 503. — Effect of neighboring current upon a magnetic needle. Above the needle and parallel to it is a conductor carrying an electric current, the current flowing in the direction indicated by the arrow. This causes the north pole of the needle to turn toward the east. If the conductor be held below the needle, its north pole will turn in the opposite direction or toward the west. These movements are easily detemiined by Ampere's rule as follows: // a man could swim in the conductor with the current, and turn to face the needle, then the north pole of the needle will be deflected toward his left hand. the needle will be deflected toward the west. Again, if the ctu*- rent be passed from north to south over the needle, and back from south to north under the needle, as shown in fig. 504, the magnetic effect will be doubled, and the needle deflected pro- portionately. Upon these phenomena depend the working of galvanometers. UtOKItTT UMAST 432 HAWKINS ELECTRICITY Oues. Describe a simple galvanometer. Ans. It consists essentially of a magnetic needle suspended within a coil of vdve, and free to swing over the face of a graduated dial. Oues. What is a galvanoscope and how does it diflfei from a galvanometer.' Ans. A galvanoscope, as shown in fig. 504, serves mereh' lo indicate the presence of an electric current without measuring -^ Fig. "MM. — Effect upon a magnetic needle of a neighboring current in a loop. In this arrange- ment the same conductor is simply carried back beneath the needle and hence both the upper and lower portions tend to turn it in the same direction, while the side branch or vertical section is ineffective. In accordance wnth Ampere's swimming rule, the upper wire causes the N pole of the needle to turn to the left, while if a man imagine himsel' swimming in the lower wire in the direction of the current, and facing the needle (that is, swimming on his back), the N pole of the needle will turn to his left — that is to the east. The effect of the loop then has double the effect of the single wire in fig. 503. its strength. It is an indicator of currents where the movement of the needle shows the direction of the current, and indicates whether it is a strong or a weak one. When the value of the readings has been determined by experiment or calculation any galvanoscope becomes a galvanometer. GALVANOMETERS 4.i:i Oues. For what use are galvanometers employed? Ans. They are used for detecting the presence of an electric current, and for determining its direction and strength. Oues. How is the direction and strength of the current indicated? Ans. When a galvanometer is connected in a circuit, the direction of the current is indicated by the side towards which the north pole of the needle moves, and the current strength by the extent of the needle's deflection. Fig. .")0o. — Effect upon a magnetic needle of a neighboring current in a coil. The coil as shown, is equivalent to several loops, that is, the force tending to deflect the needle is equal to that of a single loop multiplied by the number of turns. Hence, by using a coil with a large number of turns, a galvanometer may be made very sensitive so that the needle will be perceptibly deflected by very feeble currents. An instrument, as shown in the figure is called a galvanoscope. When it is accurately constructed, and supplied with a scale showing how many degrees the needle is deflected it is then called a galvanometer. Oues. How should a galvanometer be set up before using? Ans. When no current is flowing, the coil should be parallel to the magnetic needle when at rest. 434 HAWKINS ELECTRICITY Ques. What is a " sensitive " galvanometer? Ans. One which reqviires a very small current or pressure to produce a stated deflection. It does not follow that a galvanometer which is sensitive for current measurement will also be sensitive for pressure measurement. Ques. Define the term " sensibility." Ans. With reference to mirror reflecting galvanometers it may be defined in three ways. First, in megohms, the sensibility Fig. 506. — Bunnell simple detector galvanometer. It has middle clamps and scale divided into degrees. being the number of megohms through which one volt will pro- duce a deflection of one millimeter with the scale at one meter distance. Second, in micro-volts, the sensibiHty being the number of micro-volts which applied directly to the terminals of the galvanometer will produce a deflection of one millimeter on a scale one meter from mirror. The sensibiHty is best stated in megohms for high resistance galvanometers and in micro-volts for low resistance galvanometers, and is frequently given both for galvanometers for intermediate resistance. Third, in GALVANOMETERS 435 micro-amperes, the sensibility being the number of micro- amperes that will give one millimeter deflection with scale at a distance of one meter. Ques. Upon what does the sensibility depend? Ans. 1, Upon the number of times the current circulates around the coil, 2, the distance of the needle from the coil, 3, the weight of the needle, 4, the current strength, and 5, the amount of friction produced by its movement. Fig. 507. — Breguet upright galva- nometer with glass shade. Fig. 508. — Bunnell horizontal galvanometer. It has two coils, one of which is of zero resistance and one of fifty ohms resistance adapting it to a variety of test. The needle is usually quite small, and often a compound one. In very sensitive galvanometers, the coils are wound with thousands of turns of very fine wire, and shunts are generally used in connection with them. Ques. What two kinds of coil are used ? Ans. The short coil and the long coil. NOTE. — Strong currents must not be passed through very sensitive galvanometers, for even if they be not ruined, the deflections of the needle will be too large to give accurate meas- urements. In such cases the galvanometer is used with a shunt, or coil of wire arranged so that the greater part of the current will flow through it, and only a small portion through the galvanometer. 436 HAWKINS ELECTRICITY Oues. What is the difference between a short coil and a long coil galvanometer? Ans. A short coil galvanometer has a coil consisting of a few turns of heavy wire; a long coil galvanometer is wound with a large number of turns of fine wire. Fig. !)09. — Bunnell galvanometer for measurements of instruments, lines, batteries, wires and any object from lij to 10,000 ohms or more. Oues. What is the action of short and long coil gal- vanometers? Ans. With a given current, the total magnetizing force which deflects the needle is the same, but with a short coil, it is produced by a large current circulating around a few turns, instead of a small current circulating around thousands of turns as in the long coil. The short coil being of low resistance is used to measure the current, and the long coil with high resistance, is suitable for measuring the pressure. Hence, a short coil instru- ment with its scale directly graduated in ami^eres is an ammeter, and the long coil type with graduation in volts is a voltmeter. GA L VA NO METERS 437 Classes of Galvanometer. — ^There are numerous kinds of galvanometer designed to meet the varied requirements. According to construction, galvanometers may be divided into two classes, as those having : 1. Movable magnet and stationary coil; 2. Stationary magnet and movable coil. Pig. 510. — Astatic needles. Two magnetic needles of equal moment are mounted in opposition on a light support. The whole system is suspended by a delicate fibre, and when placed in a uniform magnetic field such as that of the earth, there will be no tendency to assume any fi.xed direction, the only restraining influence oil the needles being that due to torsion in the suspension fibre. Either type may be constructed with short or long coil, and there are several ways in which the deflections are indicated. The principal forms of galvanometer are as follows: 1. Astatic; 2. Tangent; 3. Sine; 4. Differential; 5. Ballistic; 0. D' Arson val. 438 HAWKINS ELECTRTCTTV Astatic Galvanometer. — It has been pointed out how a compass needle is affected when a wire carrying a current is held over or under it, the needle being turned in one direction in the first instance, and in the opposite direction for the second posi- tion of the vnre. The earth's magnetism naturally holds the compass needle north and south. The magnetic field encircling the wire, being at right angles to the needle (when the wire itself is parallel there- with), operates to turn it from its normal position, north and south, so as to set it partially east and west. However, on account Fig. 511. — Connections of single coil astatic needles. The coil surrounds the lower needle and the direction of the current between the two needles tends to turn them the same way. of the fact that the earth's magnetism does exert some force tending to hold the needle north and south, it is evident that no matter how strong the current, the latter can never succeed in turning the needle entirely east and west. The accomphshment of this is further prevented by the reason of the points of the needle, where the magnetic effect is greatest, quickl}'- passing out of the reach of the magnetic field, where it is now practically operated on only in a slight degree. Thus it would take quite a powerful current to hold the needle deflected any appreciable distance. The use of a shorter needle is. therefore, more desirable. GALVANOMETERS 439 It is evident in this style of instrument that the effect of the current cannot be accurately measured, because it acts in opposi- tion to the earth's magnetism, and as this is constantly varying, some method must be employed which will either destroy the earth's magnetism or else neutralize it. In the astatic galvanometer, the earth's magnetism is neutral- ized by means of astatic needles. These consist of a combination of two magnetic needles of equal size and strength, connected S «gs^p^^»|y] ) HD Fig. 51*. — Connections of double coil astatic needles. With this arrangement, the direction of current in both coils will tend to turn the system in the same direction, making the needles more sensitive than with a single coil as in fig. 511. rigidly together with their poles pointing in opposite and parallel direction?, as shown in fig, 510. As the north pole of the earth attracts the south pole of one of the needles, it repels with equal strength the north pole of the other needle, hence, the combination is independent of the earth's magnetism and will remain at rest in any position. If one of the needles be surrounded by a coil, as shown in fig. 511, the magnetic effect of the current will be correctly indicated by the deflection of the needle. 440 HAWKINS ELECTRICITY Sometimes each needle is surrounded by a coil, as in fig. 512, the coils being so connected that the direction of current in each mil tend to deflect the needles in the same direction, Oues. For what use is the astatic galvanometer adapted? Ans. For the detection of small currents. It is used in the "nil" or zero methods, in which the current between the points to which the galvanometer is connected is reduced to zero. Fig. 513. — Queen reflecting astatic galvanometer. It is mounted on a mahogany base with levelling screws. A plain mirror is attached above the upper needle. The entire combina- tion of mirror and needles is suspended by unspun silk from the interior of a brass tube, which also carries a weak controlling magnet. A dial 4 inches in diameter and graduated in degrees, enables the deflections of the needle to be accurately read. The mirror can be used with a reading telescope and scale, or by means cf a lantern, the image of a slit may be reflected from the mirror to a screen. Resistance, .5 to 1 ,000 ohms. Oues. Upon what does the movement of the needles depend? Ans. Upon the combined effect of the magnetic attraction of the current which tends to deflect the needles, and the torsion GALVANOMETERS 441 in the suspension fibre which tends to keep the needle at the zero position. Oues. Does the astatic galvanometer give correct readings for different values of the current? Ans. When the deflections are small (that is, less than 10° or 15°), they are very nearly proportional to the strength of the currents that produce them. Fig. 514. — Central Scientific Co. tangent galvanometer. A 9 inch brass ring is mounted on a mahogany base which rotates on a tripod provided with levelling screws. The needle has an aluminum pointer and jewelled bearing. The winding consists of 300 turns of mag- net wire so connected to the plugs in front that 20, 40, SO, or 160 turns ox any combination of these numbers may be used. For heavy currents a band of copper is used by connectiiig to the extra pair of binding posts in the rear of the instrument. Thus, if a current produce a deflection of 6° it is known to be approxi- mately three times as strong as a current which only turns the needle through 2°. But this approximate proportion ceases to be true if the deflection be more than 15° or 20°. Oues. "Why does the instrument not give accurate readings for large deflections? Ans. The needles arc not so advantageously acted upon by the current, since the poles are no longer within the coils, but 442 HAWKINS ELECTRICITY protrude at the side. Moreover, the needles being oblique to the force acting on them, part only of the force is turning them against the directive force of the fibre ; the other part is uselessly piilling or pushing them along their length. Ques. How may correct readings be obtained? Ans. The instrument may be calibrated, that is, it may be ascertained by special measurements, or by comparison with a Fig. 515. — Bvinnell tangent galvanometer. This instrument is mounted on a circular hard rubber base, 7?^ inches diameter, provided with levelling screws and anchoring points. The galvanometer consists of a magnetized needle "s inch in length, suspended at the center of a rubber ring six inches in diameter, containing the coils. There are five coils of 0, 1, 10, 50 and 150 ohms resistance. The first is a stout copper band of inappreciable resistance; the others are of different sized copper wires, carefully insulated. Five ter- minals are provided, marked, respectively, 0, 1, 10, 50 and 150. The ends of the coils are so arranged that the plug inserted at the terminal marked 50 puts in circuit all the coils; marked at the terminal 50 — all e.xcept the 150 ohm coil; and so on, till at the zero terminal only the copper band is in circuit. Fixed to the needle, which is balanced on jewel and point, is an aluminum pointer at right angles, extending across a five inch dial immediately beneath. One side of the dial is divided into degrees; on the other side, the graduatieas correspond to the tangent of the angles of deflectioa. GALVANOMETERS 443 standard instrument, the amounts of deflection corresponding to particular current strengths. Thus, if it be once known that a deflection of 32° on a particular galviinometer is produced by a current of jj^ of an ampere, then a current of that strength will always produce on that instrument the same deflection, unless from any accident the torsion force or the in- tensity of the magnetic field be altered. Pig. 516. — Tangent galvanometer. It consists of a short magnetic needle suspended at the center of a coil of large diameter and small cross section. In practice, the diameter of the coil is about 17 times the length of the needle. If the instrument be so placed that, when there is no current in the coil, the suspended magnet lies in the plane of the coil, that is, if the plane of the coil be set in the magnetic meridian, then the current passing through the coil is proportional to the tangent of the angle by which the magnet is deflected from the plane of the coil, or zero position — hence the name: "tangent galvanometer." The Tangent Galvanometer. — It is not possible to construct a galvanometer in which the angle (as measured in degrees of arc) through which the needle is deflected is proportional throughout its whole range to the strength of the current. But it is possible to construct a very simple galvanometer in which the tangent oj the angle oj deflection shall be accurately proportional to the strength of the current. 444 HAWKINS ELECTRICITY A simple form of tangent galvanometer is shown in fig. 516 The coil of this instrument consists of a simple circle of stout copper wire from ten to fifteen inches in diameter. At the center is delicately suspended a magnetized steel needle not exceeding Fig. 517. — Horizontal section through middle of tangent galvanometer, showing magnetic whirls around the coil and corresponding deflection of needle. one inch in length, and usually furnished with a light index of aluminum. When the galvanometer is in use, the plane of the ring must be vertical and in the magnetic meridian. A hori- zontal section through the middle of the instrument is shown in V^' Fig. 518. — Diagram of forces acting on the needle of a tangent galvanometer. fig. 517. For simplicity, the coil is supposed to have but a single turn of wire, the circles surrounding the wire representing the magnetic lines of force. By extending the lines of force until they reach the needle, it will be seen that with a short needle, the GALVANOMETERS 445 deflecting force acts in an east and west direction when the gal- vanometer is placed with its coil in the magnetic meridian. If, in fig. 518, ab represent the deflecting force acting en the N end of the needle, the component of this force that acts at a right angle to the needle will be ab cos X in which, x is the angle of the deflection. The controlling force is aJ = H i I 1 b / i i H ( 1 R 1 I ^ o^ . — M :^ < >■ >■ Fig. 519. — Diagram illustrating the tangent law. This is the law of the combined action of two magnetic fields upon a magnetic needle. If two magnetic fields be at right angles in direction as indicated in the figure, the resultant field is obtained by the parallelogram of forces and it makes an angle with one of the component fields such that tan = M + H where M and H are the strengths of the component fields. In the tangent galvanometer this principle is employed in the measurement of currents. A magnetic needle is pivoted in a field of known strength. The current to be measured is passed round a coil (or coils) which generates a field at risht angles to the original field. The needle then lies along the direction of the resultant field, and by finding the tangent of its angle of deflection, and knowing the field strength produced by unit current in the coil, the current strength can be found. and when the needle is in equilibrium, the component ae = H sin x is equal and opposite to ac, hence from which ab cos « = H sin x ab = H = H tan x cos X Since ab is proportional to the current, ab = k C = K tan x 446 HAWKINS ELECTRICITY in which it is a constant depending upon the instrument. For any other current C, kC = K tan x' hence C: C = tanx : tan a;' This means that the currents passing through the coil of a tangent galvanometer are proportional, not to the angle of deflection, but to the tangent of that angle. 30 T Fig. 520. — Graduation of tangent galvanometer scale with divisions representing tangent values. In the figure let a tangent OT be drawn to the circle, and along this line let any number of equal di\-isions be set off, beginning at O. From these points draw lines back to the center. The circle will thus be divided into a number of spaces, of which those near O are nearly equal, but which get smaller and snialler as they recede from O. These unequal spaces correspond to equal increments of the tangent. If the scale were divided thus, the readings would be proportional to the tangents. Oues. Upon what does the sensibility of a tangent galvanometer depend? Ans. It is directly proportional to the number of turns of the coil and inversely proportional to the diameter of the coil. Oues. How may the tangent galvanometer be used as an ammeter? Ans. The strength of the current may be calculated in am- peres by the formula given below when the dimensions of the instrument are known. The needle is supposed to be subject to only the earth's magnetism and to move in a horizontal plane. The current is calculated as follows: amperes = H_x_ N tan X. (1) GA L VA NOME TERS 447 in which H = constant from table below; r = radius of coil ; N = number of turns of coil ; X = angle of deflection of needle. The constant H, given in the following table represents the horizontal force of the earth's magnetism for the place where the galvanometer is used. Each value has been multiplied by -— so that the formula (i) for amperes is correct as given. Fig. 521. — Mechanical explanation of the tangent law. Construct an apparatus as shown in the figure. The short wooden block, NS, represents the magnetic needle. This fiece of wood turns around its center, C, which may be an ordinary nail. It will now be seen that two different forces act upon N; namely, the weight. G (oneor two ounces), and the changeable weights which are placed in the scoop, W (made of cardboard). Tue height of the roll, or wheel, R, is such that the cord, RN, runs horizontally, when NSstands vertically, i.e., when there is no weight in the little scoop. If the wheel, R, be placed '-ufficienlly far from NS, the string RN, will always remain almo.=t horizontal, even if NS be deviated. The thin hand on NS moves over a vertical scale, which is divided into equal parts, as shown. This scale may be made of cardboard. If the hand point to division 1 when one ounce is placed in the scoop, it will point to 2 for two ounces, to 3 for three ounces, etc. At 45° the needle is debated at its greatest angle, and this is, therefore, the sensiti\^ty angle of the tangent galvanometer. The deviating values are, therefore, proportionate to the scale parts 01, 02, and 03, and so on; and, inasmuch as these themselves are tangents, the tangent law will hold good. Table of Galvanometer Constants. — Values of H. Boston Chicago . Denver . Jacksonville London . Miimeapolis New York .699 .759 .919 1.094 .745 .681 .744 New Haven 731 Philadelphia 783 Portland, Me 674 San Francisco .... 1.021 St. Louis 871 Washington 810 448 HAWKINS ELECTRICITY Oues. How is the tangent galvanometer constructed to give direct readings? Ans. To obviate reference to a table, the circular scale of the instrument is sometimes graduated into tangent values, as in fig. 520, instead of being divided into equal degrees. Pig. 522, — Queen tangent and <«"«' galvanometer. This instrument properly adjusted can be used as a standard instrument for laboratory work. The brass ring is 12 inches in diam- eter, and the grooves in which the wire is wound are carefully turned so as to be of true rectangular cross section, thus allowing the constant of the instrument to be accurately cal- culated and compared with the constant as obtained by other methods. The compass box is 5 inches in diameter and is so held in position that it may be raised or lowered, rotated on its vertical axL shL'ted out of the plane of the coil, etc., thus enabling the opverator to acquire proficiency with the instrument and to meet all cases of derangement possible The dial is graduated to single degrees, and the needle is suspended by a very light cocoon fibre. The whole instrument can be turned about itr. vertical a.'us, and a quadrant graduated in degrees upon the base allows the amount of rotation to be accurately measured, and the laws of the sine galvanometer investigated. The instrument is wound to measure .25 ampere to 8 amperes, Oues. What is the objection to the scale with tangent values? Ans. It is more difficult to divide an arc into tangent lines with accuracy than into equal degrees. Oues. What disadvantage has the tangent galva- nometer? Ans. The coil being much larger than the needle, and hence far away from it, reduces the sensitiveness of the instrument. GALVANOMETERS 449 The Sine Galvanometer. — This type of instrument has a vertical coil which may be rotated around a vertical axis, so that it can be made to follow the magnetic needle in its deflections. In the sine galvanometer, the ceil ts mov^ed so as to follow the needle until it is parallel with the coil. Under these circumstances, the strength of the deflecting current is proportional to sine of angle of deflection. Fig. 523. — Central Scientific Co. universal tangent galvanometer. This instrument may be used as a tangent, Gaugain, Helmholtz-Gaugain, sine, cosine, Wiedemann or detector galvanometer. The coils, which slide on a beam parallel to the one carrying the needle box, are wound on brass rings 12 inches in diameter. On each ring are wound two coils of 48 turns each, c nnected to separa.e binding posts, and double wound so as to be of equal resistance. The coils and needle bo.x are each pro\'ided with an indicator for reading their position on the scale. The needle box is swivelled and removable and one coil may be rotated about its vertical axis and its position read on a disc graduated in degrees. Cur- rents may be measured ranging from .000002 ampere to ICO amperes. Oues. Describe the construction of a sine galvanometer. Ans. A form of sine galvanometer is shown in fig. 524. The vertical wire coil is seen at M. A needle of any length less than the diameter of the coil M, moves over the graduated circle N. The coil M, and graduated circle N may be rotated on a vertical 450 HAWKINS ELECTRICITY axis, and the amount of angular movement necessary to bring the needle to zero, measured on the graduated circle H. Oues. How is the current strength measured? Ans. It is proportional to the sine of the angle measured on the horizontal circle H, through which it is necessary to turn the coil M, from the plane of the earth's magnetic meridian to the plane of the needle when it is not further deflected by the current. Fig. 524. — Sine galvanometer. It differs from the tangent galvanometer in that the vertical coil and magnetic needle are mounted upon a standard free to revolve around a vertical axis, with provision for determining the angular position of the coil. The needle may be of any length shorter than the diameter of the coil. In the figure the parts are: M, coil; N, graduated dial of magnetic needle; H, graduated dial by which the amount of rotation necessary to bring the needle to zero is measured; E, terminals of the coil; O, upright standard carrying coil and graduated dial of magnetic needle; C, base with levelling screws. Oues. How is the sine galvanometer operated? Ans. In using the instrument, after the needle has been set to zero, the current is sent through the coil, producing a deflec- tion of the needle. The coil is then rotated to follow the motion GALVANOMETERS 451 of the needle, the current being kept constant, the rotation being continued until the zero on the upper dial again registers with the needle. The current then is proportional to the sine of the angle through which the coil has been turned, as determined by the lower dial. Oues. Has the sine galvanometer a large range? Ans. For a given controlling field, it does not admit of a very large range of current measurement, since, for large deflection, on rotating the coil the position of instability is soon reached. TABLE OF NATURAL SINES AND TANGENTS ^ Sin. Tan. ^ Sin. Tan. ^ Sia. Tan. I Sin. Tan. ^ Sin. Tan. 0° .0000 .0000 18° .3090 .3249 36° .5878 7265 54° .8090 1.3764 72° .9511 3.0777 19 .3256 .3443 55 .8192 1.4281 73 9563 3.2709 1 .0175 .0175 37 .6018 .7536 56 .8290 1.4826 2 .0349 .0:^49 20 .3420 ,3640 38 .6157 .7813 74 9613 3.4874 8 .0523 .0524 39 .6293 8098 57 .8387 1.5399 75 .9659 3.7321 21 .3584 .3839 58 .8480 1.6003 76 .9703 4.U1U8 4 .0698 .0699 22 .3746 .4040 40 .6428 .8391 59 .8572 1.6643 6 .0871 .0875 23 .3907 .4245 77 .9744 4.3315 6 .1045 .1051 41 .6.561 .8693 60 .8660 1.7321 78 .9781 4.7040 24 .4067 .4452 42 .6691 .9004 79 9816 5.1446 7 .1219 .1228 25 .4226 .40C.:f 43 .6820 9325 61 .8746 1.8040 i .1392 .1405 26 .4384 .4877 62 .S829 1.8807 80 .9848 5.0713 9 .1564 .1564 44 .6947 .9657 63 .8910 1.9626 27 .4540 .509i 45 .7071 1.0000 81 9877 6.3138 10 .1736 .1763 28 .4695 .5317 46 .7193 1.0355 64 .8988 2.0503 82 .9913 7.11.14 29 .4848 .5543 65 9063 2.1445 83 9925 8.1443 11 .1908 .1944 47 .7314 1.0724 66 .9135 2.24o0 12 .2079 .2126 30 .5000 .5774 48 .7431 1.1106 84 .9945 9.5144 13 .2250 .2309 49 .7547 1.1504 57 .9205 2.35.59 85 .9962 11.43 31 .5150 .6009 68 .9272 2.4751 86 .9976 14.30 14 .2419 .2493 32 .5299 .6249 50 7660 1.1918 69 .9339 26051 15 .2588 .2679 33 .5446 .6494 87 .9986 19.03 16 .2756 .2867 51 .7771 1.2349 70 .9397 2.7175 .S8 .9994 2-<.64 34 .5592 .6745'52 .7880 1.2799 S9 .9998 57.29 17 .2924 .3057 35 .6736 .7002^3 .7986 1.3270 71 .9455 2.9042 Oues. What is the position of instability? Ans. The position of the needle beyond which the rotation of the coil will cause it to turn all the way round. Oues. How may the range be increased? Ans. By an adjustable controlling field or a shunt. 452 HAWKINS ELECTRICITY' Oues. What advantage has the sine galvanometer over the tangent instrument? Ans. Its advantage is in the case where the relative values of two or more currents arc required to be measured, or where the constant of the instrument is obtained by comparison with a standard measuring instrument and not calculated from the dimensions of the coil, because all galvanometers thus used follow the sine law independently of the shape of the coil, while only circular coils will follow the sine law. Pig. 525. — Differential galvanometer. It consists of two coils of wire, so wound .is to have opposite magnetic effects on a magnetic needle suspended centrally between them. The needle of a differential galvanometer shows no deflection when two equal currents are sent through the coils in opposite directions, since, under these conditions, each coil neutralizes the effect of the other. Sometimes the current is so sent through the two coils, that each coil deflects the needle in the same direction. In this case the instrument is no longer differential in action. If, when this condition obtains, the magnetic needle be sus- pended at the exact center of the line which joins the centers of the coils, the advantage is gained by obtaining a field of more nearly uniform intensity around the needle. When the needle is suspended by a silk fibre, a final and most delicate adjustment can be obtained by raising or lowering one of the levelling screws slightly, so as to tilt the needle nearer to or farther from one of the coils. The Differential Galvanoirieter. — This is a form of galva- nometer in which a magnetic needle is suspended between two coils of equal resistance so wound as to tend to deflect the needle in opposite directions. The needle of a differential galvanometer GA L VA NOME TERS 45S shows no deflection when two equal currents are sent through the coils in opposite directions, since under these conditions, each coil neutralizes the other's effects. Such instruments may be used in comparing resistances, although the Wheatstone bridge, in most cases, affords a preferable method. Ones. What is the special use of the differential gal- vanometer ? Ans. It is used for comparing two currents. Ques. What is the method of comparing currents? Ans. If two equal currents be sent in opposite directions through the coils of the galvanometer, the needle will not move ; if the currents be unequal, the needle will be deflected by the stronger of them with an intensity corresponding to the difference of the strengths of the two currents. Ques. How are the coils adjusted? Ans. This is done by coupling them in series in such a way that they tend to turn the needle in opposite directions, and when a current is passing through them, they are moved nearer to the needle or farther from it until the needle stands at zero with an^- current. If the coils be not movable, a turn or more can be unwound from the coil giving the greatest magnetic effect until a balance is obtained, the wire so unwound can then be coiled in the base of the instrument. Ballistic Galvanometer.— This type of galvanometer is designed to measure the strength of momentary currents, such for instance, as the discharge of a condenser. In construction the magnetic system is given considerable weight, and arranged to give the least possible damping efect. 454 IIA IVKINS ELECTRICITY The term "damping effect" means the offering of a retarding force to control swinging vibrations, such as the movements of a galvanometer needle, and to bring them quickly to rest. If a momentary current be passed through a balHstic galva- nometer, the impulse given to the needle does not cause appre- ciable movement to the magnetic system until the cturent ceases, o'^'ing to the inertia of the heavy moving parts, the result being a slow swing of the needle. rRonnrr unAMt S. C. State CoUt^ Pig. 526. — Queen dead beat and ballistic reflecting galvanometer. As illustrated, the coils are easily removable and enclose a heavy block of copper fixed in a central fork. In a cylin- drical hole bored in this block hangs the bell magnet which n-ith its mirror is suspended by a long cocoon fibre, and the eddy currents induced in the copper bring the system quickly to rest after a deflection. By lifting the copper block out of the frame the instrume *• is made ballistic. The instrument is made with coils of any desired resistance up to 1 ,000 ohms. Oues. What name is given to the swing of a ballistic galvanometer needle? ^ Ans. It is called the kick. ^^ Cr_.-rfl^ Oues. How is the current measured ? Ans. As the needle swings slowly around it adds up, as it GALVANOMETERS 455 were, the varying impulses received during the passage of the momentary current, and the quantity of electricity that has passed is proportional to the sine of half the angle of the first swing or kick. If a reflecting method be used with a straight scale, the observed deflection depends upon the tangent of twice the angle of movement of the needle. For small deflections, however, the change of flux can be taken as directly proportional to the observed deflection. Fig. 527. — Thompson galvanometer with mirror reflecting system for reading the deflections of a galvanometer needle by the movements of a spot of light reflected from a mirror attached to the needle or movable magnetic system. Use of Mirrors in Galvanometers. — In order that small currents may be measured accurately, some means must be pro- vided to easily read a small deflection of the needle. Accordingly, it is desirable that the pointer be very long so that a large number of scale divisions may correspond to small deflections. In construction, since sensitive galvanometers must be made with the moving parts of little weight, it would not do to use a long needle, hence a ray of light is used instead, which is reflected on a distant scale by a small mirror attached to the moving part. 456 HA WKIXS ELECTRICITY In the Thompson mirror reflecting galvanometer, as shown in fig. 528, a small vertical slit is cut in the lamp screen below the scale, and the ray of light from the lamp, passing through the slit, strikes the mirror which is about three feet distant, and which reflects tne beam back to the scale. It should be noted that the angle between the original ray of light and the reflected ray is twice the angle of the deflection of the mirror; the deflections of the ray of light on the scale, however, are practically proportional to the strength of currents through the instru- ment. The mirror arrangement as shown in fig. 528, requires a darkened room for its operation, but such is not necessary when a telescope is used as in fig. 529. Here the scale readings are reflected in the mirror and their value observ'ed by the telescope wiQiout artificial light. TELESCOPE \ (fr-SCALE MIRROR ^^^ MIRROR ON GALVANOMETER COIL Pig. 528. — Telescope method of reading galv.inometer deflections by reflection of scale reading in mirror. Here two mirrors are used, but in most cases the telescope is pointed directh toward the_ mirror on galvanometer shown in fig. 527, because the two mirror s>-stem, as illustrated in the figure, is used on ix>rtable galvanometers since it is the more compact. Damping. — This relates to the checking or reduction of oscil- lations. Thus, a galvanometer is said to be damped when so constructed that any oscillations of the pointer which may be started, rapidly die away. Galvanometers are frequently pro- vided with damping devices for the purpose of annulling thesq GALVANOMETERS 457 oscillations, thus causing the moving part to assume its final position as quickly as possible. Sometimes the instrument is fitted with a damping coil, or closed coil so arranged with respect to the moving system that the oscillations of the latter give rise to electric currents in the closed coil, whereby energy is dissipated. Again, air vanes are employed, but anything in the nature of solid friction cannot be used. Figs. 529 and 530. — Galvanometer lamp and scale for individual use. The scale is etched on a ground glass strip 6 centimeters vnde by 60 centimeters long with long centimeter divisions and short millimeter divisions the entire length, reading both waysfrom zero in the center. It is mounted in an adjustable wooden frame. A straight filament lamp (110 volts) is enclosed in a metal hood japanned black to cut out all reflected light. This form of filament makes a single brilliant line on the scale, enabling closer readings than the "spot of light" arrangement. The lamp hood is adjustable to any desired height on the support rod. D'Arsonval Galvanometer.— This instrument has a movable coil in place of a needle, and its operation depends upon the principle that if a flat coil of wire be suspended with its axis perpendicular to a strong magnetic field, it will be deflected when- ever a current of electricity passes through it. 458 HA WKINS ELECTRICITY Oues. Describe the construction of a D'Arsonval gal- vanometer. Ans. The essential featttres are shown in figs. 532 and 533. The coil, which is rectangular in section is wound upon a copper form, and suspended between a permanent magnet by fine wires to the points A and B. The magnet has its poles at N and S. It is a soft iron cylinder fixed between the poles in order to intensifj' the magnetic field across the air gaps in which the coil moves. Fig. Ml. — Queen reading "--:-:i-jpe. This arrar.genezt is utilized tomeasurethe deflections of a galvanometer ha\-ing suspended mirror moving system. It consists of a reading telescope mounted as illustrated with a millimeter scale, ha^•ing a length of 50 centimeters. In use. the image of the scale is seen in the galvanometer mirror through the telescoi>e. The eye piece of the telescope has a cross hair which acts as a reference line so that by noting the particular di\'ision on the scale when the galvanometer is at rest, the amount of deflec tion can be readily obser%-ed when the galvanometer is deflected. The instrument has all the necessary adjustments to set it up quickly and for bringing the cross hair and scale in focus. It is generally placed at a distance of one meter from the galvanometer mirror. Oues. Explain its operation. Ans. An enlarged horizontal cross section of the galvanome- ter on line XY is shown in fig. 533. The current is flowing in the coil as in fig. 532, up on the left side and down on the right. GALVANOMETERS 459 The position of the coil when no current is flowing is indicated by n' s'. By applying the law of mutual action between mag- netic poles, it is seen that when the current is applied, the poles developed at n' s' will move into the position n" s" . See fig. 119. Oues. How is the coil affected by a change in the direc- tion of the current? Ans. The polarity of the coil is reversed and consequently the direction of the deflection. -♦A Figs. 532 and 533. — Diagrams showing essential features of construction and principle of operation of D' Arson val galvanometer. Oues. Upon what does the sensitiveness of the instru- ment depend ? Ans. Upon the strength of the field of the permanent magnet, the number of turns in the suspended coil, and the torsion of the wires by which it is suspended. Oues. When is this galvanometer called " dead beat " ? Ans. When the construction is such that the moving part comes quickly to rest without a series of diminishing vibrations. 4G0 HAWKINS ELECTRICITY Fios. 534 to 536. — Queen hori- zontal magnet D'Arsonval galvanometer with telescope and scale. Itis very sensitive and is used in many electrical measurements, including com- mercial testing, such as meas- uring insulation of cables, fault location, etc. It is not affected by surrounding mag- netic disturbances, and may, therefore, be used in proximity to dynamos and switchboards. The instrument has a pair of binding post terminals, one of which connects to a bottom spiral of the system and the other forms a junction with the top of the tube holding the system, forming a complete circuit through the coil. The '^ tube containing the system may be readily removed from the magnet and another tube hav- ing a different system inserted as is required for various kinds of electrical measurement. The entire system with its suspension may be inspected by the removal of a thumb screw. To inspect interior of tube first be sure that the screw B is turned so that the coil is clamped. Entirely remove screw C, and, holding the outside tube near the window, press firmly with the finger on the extreme top of the suspension support. The inside rib, with complete suspension, will draw from the tube, and the working parts can be fully inspected. Care- fully return sarne to its original position in tube, setting tight the screw C. The galva- nometer is designed so that the coil is clamped in position when the galvanometer must be transported. The insulation of the galvanometer terminals and binding posts is such as to guard against any possible leakage. As a further protection, each levelling screw is provided with a hard rubber insulator. This feature is essential since, in making insulation measure- ments, the operator wishes to be assured that the deflection being obtained is the result of leakage upon the cable or wire being measured and not leakage between the galvanometer terminals. _ The galvanometer is pro\'ided with an attached telescope and scale for noting the deflections. The deflections produced by this galvanometer are proportional to the cur- rent. To facilitate quickly setting up the instrument, two way levels are provided. GA L T VI NOME TERS 40 1 Oues. What causes this? Ans. The instrument is made dead beat by winding the coil on a copper or aluminum frame, so that when in operation, cur- rents are induced in the frame by the motion of the coil in the magnetic field; these currents oppose the motion of the coil. Oues. For what service is the D'Arsonval galvanometer adapted? Ans. It is desirable for general use as it is not much affected by changes in the magnetic field. It may be made with high enough period and sensibilit}' to be satisfactory as a ballistic instrument, but for extreme sensibility an instrument of the astatic type is more generally used. Galvanometer " Constant " or " Figure of Merit." — In order that a galvanometer shall be of value as a measiuing in- strument, the relation between the current and the deflection produced by it must be known. This may be obtained experi- mentally by determining the value of the current required to produce one scale division. The galvanometer constant then may be defined as the resistance through which the galvanometer will give a deflection of one scale division when the current applied is at a pressure of one volt. Accordingl}^ the deflection as indicated on the scale must be multiplied by its constant or figure of merit, in order to obtain the correct reading. If the scale readings be not directly pro- portional to the quantity to be measured, the law of the instru- ment must also be considered. Thus in a tangent galvanometer as previously explained I = K tan ^ where I = current,

-, and for its negative electrode, amalgamated zinc. The electrolyte consists of a saturated solution of zinc sulphate and mercurous sulphate. The pressure is 1.434 volts at 15°C.. and between 10°C. and 25°C. the pressure decreases .001 15 of a volt for each increase of 1°C. The containing glass vessel consists of two limbs, closed at bottom and joined above to a common neck fitted with a ground glass stopper. The diameter of the hmbs should be at least 2 cms., and their length at least 3 cms. The neck should be not less than 1.5 cms. in diameter. At the bottom of each limb a platinum wire of about .4 mm. in diameter is sealed through the glass. To set up the cell, place mercury in one limb, and in the other hot liquid amalgam, containing 90 parts mercury and 10 parts zinc. The platinum wires at the bottom must be completely covered by the mercury and the amalgam, respectively. On the mercury, place a layer 1 cm. thick of the zinc and mercurous sulphate paste. Both this paste and the zinc amalgam must be covered with a layer of the neutral zinc sulphate crystals 1 cm. thick. The whole vessel must then be filled with the saturated zinc sulphate solution, and the stopper inserted so that it shall just touch it, lea\'ing, however, a small bubble to guard against breakage when the temi)erature rises. Before finally inserting the glass stopper a strong alcoholic solution of shellac is applied to the upper edge, after which the stopper is pressed firmly in place. TESTING AND TESTING APPARATUS 467 Oues. What error is introduced in measuring the pressure of a battery with an ordinary voltmeter? Ans. Since the measurement is made on closed circuit the reading does not give the total pressure of the battery. The error is very slight because the resistance of the voltmeter is very high and the current so small that the loss of pressure in the battery can be neglected. Fig. 541. — Weston Cadmium Cell. It is made in two forms; one known as the Weston normal cell, in which the solution of cadmium sulphate is saturated at all temperatures at which the cell may be used. The other, known as the Weston standard cell, in which the cadmium sulphate solution is unsaturated at all temperatures above 4° C. The Weston normal cell, or saturated form is slightly affected by changes in temperature, but, on account of the fact that it can be accurately reproduced, it was adopted by the London Conference in 1908, as a convenient voltage standard. The value of its voltage suggested by the committee of the London Conference on Electrical Units and Standards, and adopted by the Bureau of Standards at Washington, Jan. 1st, 1911, is 1.0183 International volts at 20° C. At any other temperature its voltage is: Ej = Ejo- .0000406 (t- 20) -.00000095 (t - 20)' + .0000000 (t- 20"')» The Weston standard cell, or unsaturated form is practically unaffected by changes in temperature and is the form most commonly used for laboratory work and general testing. The average pressure of this form i$ 1.01S7 Int. volts. 468 HAWKINS ELECTRICITY Ques. Define the International volt. Ans. It is the electromotive jorce that, steadily applied to a con- ductor whose resistance is one International ohm, will produce a current of one I titer national ampere, and which is respresented ^5?^^^55^^J^%5!^^;:^^^^^55??^ Pigs. 542 and 543. — Diagrams showing hydraulic aaa]og>' illustrating the difference between amperes and coulombs. If the current strength in fig. 543 be one ampere, the quantity of electricity passing any point in the circmt per hour is 1 X 60 X 60 = 3.600 coulombs. The rale of current flow of one ampere in fig. 543 may be compared to the rate of discharge of a pump as in fig. 512. Assuming the pump to be of such sire that it discharges a gallon per stroke and is making 60 strokes per minute, the quantity of water dis.-harged per hour (coulombs in fig. 543) is 1 X 60 X 60 = 3.600 gallons. Following the analogy- further (Ln fig. 543). the pressure of one volt is required to force the electricity through the resistance of one ohm between the terminals A and B. In fig. 542. the boiler must furnish steam pressure on the pump piston to overcome the friction (resistance) offered by the pip* and raise the water from the lower level A' to the higher level B'. The difference of pres'sure between A and B in the electric circuit corresponds to the difference of pressure between A' and B'. The cell furnishes the energ>- to move the current by maintaining a difference of pressure at its terminals C and D; similarly, the boiler furnishes energ>- to raise the water by maintaining a difference of pressure between the steam pipe C and exhaust pipe D'. sufficiently well for practical use by of the voltage between l,4o-i the poles of the Clark cell at a temperature of 15° C, when prepared as in fig. 540. The relation between the units volt, ampere and ohm, are shown graphically in figs. 542 and 543. TESTING AND TESTING APPARATUS 469 Current Measurement. — It is necessary to adopt some arbitrary standard in order to compare currents of different strengths. The term strength of a current, or current strength means the rate of flow past any point in the circuit in a given unit of time. The unit of current, called the ampere, is defined as the unvarying current which, when passed through a solution of nitrate of silver in ivater (15 per cent, by weight of the nitrate) deposits silver at the rate ^/ .001118 gramme per second. Fig. 544. — Queen weight voltameter for determining the strength of current by the weight of metal deposited in a given time. The two outside plates form the anode and are joined together and to one binding post, while the cathode is placed between them and connected to the other binding post. The cathode thus recives a deposit on both sides. An adjustable arm serves to lower the plates into the electrolyte. To calculate the strength of an un- known current which has passed through a weight voltameter, divide the gain in weight by the number of seconds the current flows through the instrument and by the weight deposited by one ampere in one second. That is, current strength in amperes = gain in weight -j- (time in seconds X .0003286). Oues. How much copper or zinc will one ampere de- posit in one second ? Ans. .0003286 gramme of copper in a copper voltameter, or .0003386 gramme of zinc in a zinc voltameter. 470 HAWKINS ELECTRICITY Oues. What is the difference between an ampere and a coulomb? Ans. An ampere is the unit rate oj flow of the current, and a coulomb is the unit quantity of electricity, that is, the ampere is the rate of current flow that will deposit .0003286 grammes of copper in one second and a coulomb is the quantity of electricity that passes a given point in one second when the current strength is one ampere. In other words a coulomb is one ampere second. Fig. .545. — Gas voltameter for determining the strength of current by the volume of gas evolved. To use, connect up as shown in the illustration. Adjust so that the zero position of the burette is about one-half inch below the level of the top of the U tube. Pour acidulated water into the mouth of the burette till the water in the U tube is about one-half inch from the top. With the. electrodes inserted through the corks, carefully place each one in position .by giving a slight twist to the right as the cork enters. The water level in the U tube and burette should now be the same, or further adjustment must be made to attain this result. The level in the burette does not necessarily have to correspond with the zero graduation, but must not be below it. Unclamp the burette and hold it nearly horizontal. The liquid will not run out if the corks be tight, so that this is the air leakage test. Attach the connectors and wires from the current source (which should have a pressure of 2 or more volts) placing a switch in the circuit. When the switch is closed, bubbles of gas will rise in the U tube from both electrodes, displacing the water and forcing it up the burette. Hydrogen will be liberated over the negative electrode, and oxygen over the pos- itive electrode in the proportion of twice as much hydro- gen as oxygen. To calculate the current strength, divide the volume of gas liberated by the time in seconds, and by the volume of gas liberated (in cubic centimeters) by one ampere in one second and by .1733; that is: amperes = volume of gas liberated -H (time in seconds X . 1733). EXAMPLE. — If an arc lamp require a current of 8 amperes, how much electricity does it consume per hour? Since one coulomb = one ampere second, the quantity of electricity consumed per hour is amperes I ^ {^^J^^ndsj = 28,800 coulombs. TESTING AND TESTING APPARATUS 471 Voltameter. — A voltameter is an electrolytic cell employed to measure an electric current by the amount of chemical decom- position the current causes in passing through the cell. There are two classes of voltameter: 1. Weight voltameters; 2. Gas voltameters. Oues. What is the difference between these two classes of voltameter? Ans. In one, the current strength is determined by the weight of metal deposited or weight of water decomposed, and in the other by the volume of gas liberated. Fig. 544 shows a weight voltameter and fig. 545 a gas voltameter. Oues. How should the plates of a weight voltameter be treated before vise? Ans. They must be thoroughly cleaned and polished with sandpaper, the sand being afterwards removed by placing them in running water. The fingers must not be placed on any part of the plate which is to receive the deposit. Oues. What form of voltameter has been selected to measure the International ampere? Ans. The silver voltameter arranged as here specified: The cathode on which the silver is to be deposited shall take the form of a platinum bowl, not less than 10 cms. in diameter, and from 4 to 5 cms. in depth. The anode shall be a disc or plate of pure silver some 30 sq. cms. in area, and 2 or 3 cms. in thickness. This shall be supported horizon- tally in the liquid near the top of the solution by a silver rod riveted through its center. To prevent the disintegrated silver which is formed on the anode falling upon the cathode, the anode shall be wrapped around with pure filter paper, secured at the back by suitable folding. The liquid shall consist of a neutral solution of pure silver nitrate containing about 15 parts by weight of the nitrate to 85 parts of water. 472 HAWKINS ELECTRICITY Oues. What is the value of the International ampere as lAeasured with the silver voltameter? Ans. The International ampere is represented sufficiently well for practical use by the unvarying current which, when passed through a silver voltameter (as described above) deposits silver at the rate of .001118 gramme per second. Pig. 546. — Single contact and short circuitinc key. This key is intended e.<^peciaUy for use with D'Arsonval galvanometers in zero deflection methods. The key is connected in circuit vnXh. the galvanometer so that whenever the key is not depressed, the galvanometer is short circuited, and its oscillations quickly damped out by the currents induced in its coil. The back end of the spring is held in a slot in a rubber block attached to the base. Ohm's Law and the Ohm. — The various tests here de- scribed depend for their truth upon the definite relation existing between the electric current, its pressure, and the resistance which the circuit offers to its flow. This relation was fully in- vestigated by Ohm in 1827. Using the same conductor, he proved not only that the current varies with the pressure, but that it varies in direct proportion. Ohm's law has already been discussed in a pre\dous chapter and the several ways of expressing it are repeated here for con . venience : volts 1. Amperes = ; ohms TESTING AND TESTING APPARATUS 473 2. Volts 3. Ohms amperes X ohms; volts amperes Various values have been assigned, from time to time, to the ohm or unit of resistance, the unit in use at the present time being known as the International ohm. This was recommended Fig. 547. — Double contact key. It is of especial value in connection with a Wheatstone bridge. When used with the latter it forms a combination battery and galvanometer key. The battery is wired to the top leaves of the key and the galvanometer to the lower leaves. Hence, when operated, the battery circuit will be closed before the galvanometer circuit, as it is desirable to avoid undue disturbance of the needle. at the meeting of the British Association in 1892, was adopted by the International Electrical Congress held in Chicago in 1893, and was legalized for use in the United States by act of Congress in 1894. The International ohm in graphically defined in fig. 548. The previous values given to the ohm which were more or less generally accepted are as f oUows : 471 HAWKINS ELECTRICITY The Siemens' Ohm. — A resistance due to a column of mercury 100 cm. long and 1 sq. mm. in cross section at 0° C. B. A. (British Association) Ohm. — A resistance due to a column of mercury approximately 104.9 cm. long and 1 sq. mm. in cross section at 0" C. -ISQ.MM Fig. 548. — The international ohm. It is defined as the resistance of 14.452 grammes of mercury in the form of a column of uniform cross section 106.3 centimeters in length, at a temper- ature of 0°C. This is approximately equivalent to a column 106.3 cm. long, having a uniform cross section of 1 sq. mm. In the figure the international ohm is defined graphically. The resistance of the external circuit and the standard one volt cell is assumed to be zero. Legal Ohm. — A resistance due to a column of meicury 106 cm. long and 1 sq. mm. in cross section at 0° C. This unit was adopted by the Paris conference of 1884. TESTING AND TESTING APPARATUS 475 OHM TABLE * Date Inter- national Ohm Legal Ohm B. A. Ohm Siemens' Ohm International Ohm. Legal ()!im 1893-4 1884 1864 1. .9972 .9866 .9407 1.0028 1. .9894 .9i34 1.0136 1.0107 1. .9535 1.0630 1.0600 1.0488 1. B. A. Ohm Siemens' Ohm Fig. 549. — Leeds and Northrup one ohm standard resistance (Reichsanstalt form); ad- justed at 20° C. Low resistance standards may be properly divided into two classes: 1. those which are designed primarily as resistance standards, and 2, those designed as current carrying standards. Those of the first mentioned class are often used to measure currents up to their capacity. The above standa.d has both pressure and current termi- nals. The binding posts for the former are mounted on high posts so as to be easily accessible when the standard is immersed in oil. When used as a resistance standard of precision, it should not be subjected to a current of more than one ampere, and when used as a current carrying standard of lesser accuracy, a current of 2 or 3 amperes may be used. • NOTE. — In the above table to reduce, for instance, British Association ohms to Inter- national ohms, multiply by .9866, or divide by 1.0136; to reduce legal ohms to International ohms, multiply by .9972. or divide by 1.0028. etc. 470 HA WKINS ELECTRICITY Practical Standards of Resistance. — The column of mercury as shown in fig. 548, is the recognized standard for resistance, however, in practice, it is not convenient to compare resistances with such a piece of apparatus, and therefore secondary standards are made up and standardized with a great degree of precision. These secondary standards are made of wire. The material generally used being manganin or platinoid. 6MVAN0METEI? TWO WAV KFY P^?===^ f] 1 >- yf— M 1 ^^K 1— ^-^^^- _ 'V ^^ ^ / > m / ( KNOWN RESlST^NCE ^ / 1 • UNKNOWN RESISTANCE Fig. 550. — Direct deflection method of testing resistances; a useful and siinple method which may be used in numerous tests. Galvanometer readings are taken through the knowTi. and tinknown resistances, and the current being proportiozial to the deflections, the \'alue of the unknown resistance is easily calculated. Resistance Measurement. — Resistance is that ii'hich offers opposition to the flow of electricity. Ohm's law shows that the strength of the current falls off in proportion as the resistance in the circuit increases. This gives a basis for measuring resistance. There are various methods by which an unknown resistance may be measured, as by the: 1. Direct deflection method; 2. Method of substitution; 3. Fall of potential method; 4. Differential galvanometer method; TESTING AND TESTING APPARATUS 477 5. Drop method; 6. Voltmeter method ; 7. Wheatstone bridge method. Direct Deflection Method. — This method is based on the fact that the greater the current through a galvanometer the greater the deflection of the needle ; it is a simple method and is capable of extended application. The apparatus required consists of battery, galvanometer, known resistance, and double contact key. The connections are Fig. 551. — Charge and discharge key, adapted to condenser testing where the condenser is to be alternately charged and discharged. The insulated handle enables the key to be used without insulating the body. Fig. 552. — Pohl commutator. This is equiv- alent to a two pole double throw switch. The depressions in the base are filled with mercury into which the contacts dip in closing the circuit. made as in fig. 550. The known resistance is put in circuit with the gah^anometer and after noting the deflection, the key is moved so as to cut out the known resistance and throw into circuit the unknown resistance. The deflection of the galva- nometer is again noted and compared with the first deflection. If the deflections be proportional to the current, the unknown resistance will be as many times the known resistance as the deflection with the known resistance is greater than the deflection with the unknown resistance. 478 HAWKINS ELECTRICITY Method of Substitution. — This is the simplest method of measuring resistance. The resistance to be measured is inserted in series with a galvanometer and some constant source of current, and the galvanometer deflection noted. A known ad- justable resistance is then substituted for the unknown and adjusted till the same deflection is again obtained. The value of the adjustable resistance thus obtained is equal to that of the resistance being tested. o -o- o rO o O o o CO -o o UJ O o o -o o o 1/5 ■o -o- o ■o ^ 1 ^\ ^ 6ALVAN0METER .:-,vKYKrr BATTERY Hill Fig. 553. — Substitution method of testing resistances. The connections and apparatus are the same as in fig. 550, except that a resistance box is used in place of the known resistance. In maJdng the test, first note deflection with unknown resistance in circuit, then press key so that the current will pass through the resistance box, and adjust the resistance in the box so that the deflection of the galvanometer is about the same as with the unknown. Now switch from one circuit to the other, changing the resistance in the box until equal deflections are obtained. When this obtains, the resistance in the box is the same as the resistance being tested. Oues. What kind of adjustable resistance is used in making the above test? Ans. A resistance box. TESTING AND TESTING APPARATUS 479 Oues. Describe a resistance box. Ans. It consists of a box containing numerous resistance coils with their ends connected to terminals and provided with plugs so that they may be thrown into or out of circuit at will, thus varying the resistance in the circuit. Figs. 554. — Ordinary resistance box. It contains sets of standard resistances consisting of coils ot insulated wire having low conductivity and small temperature coefficient. The ends of the coils are joined to the section of the bar between the plugs. The insertion of a plug cuts out a coil. In using, care should be taken to put the plugs in with a slight twist so that there shall be no resistance introduced by poor contact. Fall of Potential Method. — This is a very simple method of measuring resistances, and one that is convenient for practical work in electrical stations because it requires only an ammeter, voltmeter, battery and switch — apparatus to be found in every station. The connections are made as shown in fig. 555. In making the test the ammeter and voltmeter readings are taken at the same time, and the unknown resistance calculated from Ohm's law. Accordingly, since: 480 HAWKINS ELECTRICITY amperes = volts ohms solving for the resistance, volts ohms = amperes (1) (2) AMMETER VOLTMETER SWITCH Fig. 555. — Fall of potential method of testing resistances; a convenient method for testing at stations, requiring only the usual instruments to be found at a station. The resistance of the voltmeter must be verj- high, otherwise the test must be made as in fig. 556. EXAMPLE. — If in fig. 555 the readings show 6 volts and 2 amperes how many ohms is the resistance being tested? Substituting in formula (2) ohms = - TESTING AND TESTING APPARATUS 481 Ques. Can this test be made with any kind of voltmeter ? Ans. Its resistance must be very high to avoid error. When a voltmeter having small resistance is used, it should be con- nected so as to measure the fall of pressure across both ammeter and unknown resistance as shown in fig. 556 SWITCH BATTERY UNKNOWN RESISTANCE VOLTMETER Fio. 556, — Fall of potential method of testing resistances; diagram showing connections for testing with low resistance voltmeter. The resistance measured with this connection will be the sum of the resistances of the coil and the ammeter. The resistance of the ammeter is usually known and can be subtracted from the sum to obtain the required resistance. Differential Galvanometer Method. — This is what is known as a nil or zero method, that is, a method of making electrical measurements in which comparison is made between two quantities by reducing one to equality with the other, the absence of deflection from zero of the instrument scale showing that the equality has been obtained. 482 HAWKINS ELECTRICITY The test is made with a differential galvanometer, and re- sistance box connected as in fig. 557. The current then will di\'ide so that part of it flows through the resistance being tested and around one set of coils of the galvanometer while the other part will flow through the resistance box and the other set of coils as indicated. When the resistance box has been so adjusted that its resistance is the same as the unknown resistance the current in the two branches will be equal, and the needle of the galvanometer will show no deflection. DIFFERENTIAL GALVANOMETER RESISTANCE BOX ^'6660 I — K) 6 6 > \ L^^ UNKNOWN RESISTANCE •^WVWVVVW BATTERY Fig. 557. — Differential galvanometer method of testing resistances. In making the test, the resistance bo.x is adjusted till the galvanometer needle shows no deflection. When this condition obtains, the resistance in circuit in the resistance box is equal to the unknown resistance, hence, a reading of the bo.^c gives the value of the unknown resistance. Oues. What name is given to this method of testing? Ans. It is called a zero method, distinguishing it from de- flection methods. Oues. For what kind of resistance is the method adapted? Ans. Since it is a nil or zero method, it is better adapted to the measurement of non-inductive than of inductive resistances. TESTING AND TESTING APPARATUS 483 Ques. What precaution should be taken with inductive resistances? Ans. The current must be allowed to flow until it becomes steady to overcome the influence of self-induction. Oues. What may be said with respect to the differential galvanometer method? Ans. With an accurate instrument it is very reliable. TWO WAY SWITCH KNOWN RESISTANCE UNKNOWN RESISTANCE ■*>A\A/VvVvVvW — — " BATTERY I I I Fig. 558. — Drop method of testing resistances. The apparatus is connected as shown and readings taken with voltmeter across known and unknown resistance. The unknown resistance is then easily calculated. Drop Method. — This is a convenient method, and one which may be used for measuring either high or low resistances with precision. It is used for many practical measurements, and requires only a voltmeter, battery, known resistance and a two way switch. The instruments arc connected as in fig. 558, and in making the test, the voltmeter is switched into circuit across the known 484 HAWKINS ELECTRICITY resistance and then across the unknown resistance, readings being taken in each case. The value of the unknown resistance, is then easily calculated from the following proportion: drop across known resistance known resistance drop across unknown resistance unknown resistance Fig. 559. — Leeds and Northrup portable galvanometer (pointer type A). The sensitiveness of this instrument is such that it may be substituted in numerous cases for the non-portable reflecting type of galvanometer; as for instance, in the checking of ammeters and volt- meters to an accuracy of .2% by the potentiometer method, and on almost all Wheatstone bridge measurements to commercial accuracies. A current of 2 micro-amperes will cause the pointer to move 1 mm. over the scale, that is, it has a sensibility of 500.000 ohms. The method of suspending the moving system is such as to practically eliminate initial friction which is of value in all zero deflecting methods. The suspensions and moving system are guarded by springs, which together with the solid construction of the case, render the instrument capable of withstanding rough usage. Overall dimensions are 5^" X 2fi" X 3H; weight about 3 pounds. from which unknown \ known resistance X drop across unknown resistance resistance/ drop across known resistance TESTING AND TESTING APPARATUS 485 Oues. What may be substituted for the voltmeter? Ans. A high resistance galvanometer, whose deflections are proportional to the current, the value of the deflections being substituted in the formula. Oues. What precaution should be taken in making the test? Ans. The current used should not be strong enough to appreciably heat the resistance, and if the current be not very steady, several readings should be taken of each measurement and the average values used in the formula. Fig. 500. — Voltmeter method of testing resistances. Knowing the resistance of the voltmeter, turn switch to the left and from reading calculate resistance corresponding to one division of the scale. Turn switch to right and multiply reading by resistance required for deflec- tion of one division This gives resistance of voltmeter and unknown resistance; sub- tracting from this the resistance of voltmeter gives value of the unknown resistance. Oues. How are the most accurate results obtained? Ans. By selecting the known resistance as near as possible to the supposed value of the unknown resistance. Voltmeter Method. — This is a direct deflection method and consists in determining first the resistance that will deflect the needle through one division of the scale on a given battery 486 HAWKINS ELECTRICITY current, then with this as a basis for comparison the voltmeter is connected across the unknown resistance whose value is easily calculated from the reading. In making the test, the instruments are connected as in fig. 560. The current from battery is first passed through the galva- nometer by turning switch as shown. Pig. 561. — Megohm box or set of standard high resistances. The box contains five resist- ances of 200,000 ohms each. The six pillars are petticoat insiilated, the resistances being placed between each pair of pillars. There is a double contact post on top of each piUai so that these can be connected together with copper links. Assuming that the resistance of the instrument is 8,000 ohms and that the current deflects the needle through 10 divisions of the scale, then for a deflection of one division the resistance is 8,000 X 10 = 80,000 ohms. Accordingly, if, when the switch is moved to the right, connecting the voltmeter across the unknown resistance, the needle be moved through 6 divisions of the scale, the combined resistance of the volt- meter and unknown resistance is 80,000 -^ 6 = 13,333i ohms, and subtracting the resistance of the voltmeter, the value of the un- known resistance is 13,333i - 8,000 = 5,333J ohms. TESTING AND TESTING APPARATUS 487 Ques. For what kinds of test is the voltmeter method best adapted ? Ans. For measuring high resistances, as the insulation of wires, etc. Ques. What may be said with respect to the current used? Ans. Its voltage should be as high as possible within the limits of the voltmeter scale. Fig. 563. — Standard resistance box: lOO.OOOohms.infour units of 10,000, 20,000, 30,000, and 40,000 ohms. An "infinity" plug separates each coil from the ones adjacent. Seg- ments are elevated from the hard rubber top by special washers in order to increase insulation. Bind- ing posts are so arranged as not to be in the way when plugs are used. Pig. 562. — Standard high resistance box: 100,000 ohms. It is mounted in a brass box with a hard rubber top. Connections should be made to terminals marked 3 and 4. When the flexible cord is on plug 1, the box is short circuited, but when on plug 2, the resist- ance of 100,000 ohms is in series. The box is especially suited to rapid cable testing. Ques. In testing cable insulation what is desirable with respect to voltmeter and current? Ans. A low reading voltmeter should be used in connection with a large battery. 488 HAWKINS ELECTRICITY Wheatstone Bridge Method. — For accurate measurements of resistance this method is almost universally used. The so- called "Wheatstone" bridge was invented by Christie, and improperly credited to "UTieatstone, who simply applied Christie's invention to the measurement of resistances. The bridge consists of a system of conductors as shown in fig. 564. The circuit of a constant battery is made to branch at P into two parts, which re-unite at 0, so that part of the cturent ^'^ J^A % ^"-^ ^^ ^■<^^> 'cs> feW-VANOMETOf ^>^' .to ,^■^5. BMTEf?Y II Pig. 564. — Diagram showing principle of Wheatstone's bridge. A, B. C. and D. are the fout members which constitute the bridge. The current from the batten,- dixndes at P. part traversing DC, and part traversing B.\. The galvanometer connected to M and X will indicate when the currents are equal in the two branches by gi\-ing no deflection. This is then a zero or nil method of testing. The resistances and keys required in testing are shown in fig. 565. In the actual instrument, the members A, B, C, and D are known by the names given in the figure. flows through the point M, the other part through the point N. The four conductors A, B, C, D, are spoken of as the arws of the balance or bridge. It is by the proportion existing between TESTING AND TESTING APPARATUS 489 the resistances of these arms that the resistance of one of them can be calculated when the resistances of the other three are known. When the current which starts from the battery arrives at P, the pressure will have fallen to a certain value. The pressure in the upper branch falls again to M, and continues to fall to Q. The pressure of the lower branch falls to N, and again falls till it reaches the value at Q. Now if N be the same proportionate BATTERY KE.Y Fig. 56.5. — Diagram showing arms of Wheatstone bridge, resistances and method of connecting galvanometer, battery and unknown resistance. distance along the resistances between P and Q, as M is along the resistances of the upper line between P and Q, the pressure will have fallen at N to the same value as it has fallen to at M ; or, in other words, if the ratio of the resistance C to the resistance D be equal to the ratio between the resistance A and the re- sistance B, then M and N will be at equal pressures. To find out if this condition obtain, a sensitive galvanometer is placed in a 490 HAWKINS ELECTRICITY branch wire between M and N which will show no deflection when M and N are at equal pressure or when the iour resistances of the arms " balance " one another by being in proportion, thus: A :C = B :D (1) KEY 6MVAN0METEf{ • O N 1,000 100 10 Q 10 100 1000 M » » » INF I B t 2 (!> g- m » » » <> i) P 50 10 ZO 10 10 UNKNOWM RtSlSTWK.E ■ r> .• 1 1 1 10 , 1.000 1.000 1.000 100 plug from each arm A and B. Ratios pyjj. —. j^. etc., or "y • -^q" '"loo ' Too'^"^"^'^ obtainable in this manner. This simple arrangement is oi)en not only to the objection that the contact resistance of the plugs which remain in is always included with the re- sistance unplugged, but also to all other objections to be urged against the use of many plugs where a few will do. The method has the limitation that it is not possible to reverse the arms of the bridge, that is, to transpose the arms A and B. sistance then in the arm B X (C of the unknown resistance. A) will give the value Ques. Why should the battery key be depressed before the galvanometer key. Ans. To avoid the sudden swing of the galvanometer needle, which occurs on closing circuit in consequence of self-induction. TESTING AND TESTING APPARATUS 493 Ques. How is it known whether too much or too little resistance be unplugged ? Ans. The galvanometer needle will be deflected to one side for too much resistance, and to the opposite side for too little i-esistance. Fig. 569. — Method of reversing arms of Wheatstone bridge with reversing blocks. The arrange- ment shown in the figure is •classical, being that used in the English post office type of Wheatstone bridge. It is open to the objections which apply to the use of several plugs, one of which is withdrawn to obtain the desired resistance. Ques. What is the meaning of " Inf.," marked on the bridge? Ans. It stands for "infinity," because the resistance coil at the point marked infinity is omitted so that adjacent sections of the arm are disconnected when the plug is taken out. In fact, the air gap interposed by the removal of tlie plug by no means provides an infinitely great resistance, but is usually called such because it is vastly greater than any of the other resistances of the bridge. 494 HAWKINS ELECTRICITY * Figs. 570 and 571. — Dia- grams illustrating the decade plan of combin- ing resistance coils. In this method the coils are connected in series and the arrangement avoids the disadvantage of the ordinary Wheatstone bridge in that the lat- ter requires a large number of plugs to short circuit the resistances not in use, which introduces an element of uncertainty as to resistance of the plug contacts and the necessity of adding up the values of all the unplugged resistances in order to deter- mine the total resistance in circuit. The necessary regular succession of values in a rheostat built on the decade plan can be obtained with either nine or ten coils per decade. The chief reason for using the latter number is found in the facility with which all the coils of one decade can be compared with one coil of the next higher decade, thus permitting the coils of a rheostat to be checked among them- selves. Thus, the ten 1 ohm coils can be checked with a 10 ohm, the ten lO's with a 100, etc. In some sets the ten coils of a a decade can be connected in series or in parallel, and it then becomes an advantage to have ten coils to a decade, since the coils in one decade in parallel equal one of the coils of the next lower decade. When these latter advantages are not required, and especially when dials or sliding switches are used, there is Uttle or no advantage in using more than nine coUs per decade, as shown in fig. 570. Here all the coils of the set are cormected in series so that the circuit is never open. Thus it imnrlnnriUrij /O /O M lO /O /O /O lO /O is a slight advantage to have permanent connections a, 6, and c, because all the coils of a decade can be thrown in circuit by simply pulling out a plug, it not being necessary to insert it again, as would be the case if the a, ft, and c connections were not used. More- over, if any plug make bad contact, its eflfect is somewhat lessened by having this bad contact shunted by the remaining coils of the decade. Again, there are occasions where \-iolent deflections of a galvanometer are prevented by not having the circuit entirely oi>en when a plug is taken out. TESTING AND TESTING APPARATUS 495 The Decade Plan. — In this method of combining resistance coils, there are 9 or 10 one ohm coils for the units place, 9 orlO ten ohm coils for the tens place, 9 or 10 one hundred ohm coils for the hundreds place and so on. Each series of coils of the same value is designated a decade. The connections are usually made as shown in figs. 570 and 571. It is apparent from the figure that any value in any one decade can be obtained by inserting between a bar and a block, only one plug; moreover if several decades be in series, any value up to the limit of the set can be read off directly from the position of the plugs without having to add up the unplugged resistance as in the ordinary arrangement. Fig. 572 — Two plug arrangement of ratio coils. Each of the ratio coils has one of its terminals connected to a common center which corresponds to the block marked C in the figure. The other terminal of each coil is connected to an individual block, there being one block for each coil. The bar B on one side of these blocks is joined to the rheostat and the bar A on the other side to an X post. In the ordinary use of this set of ratio coils two plugs only are used. One plug is inserted between the bar A and one of the blocks, 1, 1', 10, 10', etc., of the central row of blocks. The other plug is inserted between the bar B and any one of the other blocks of the central row. There are two ratio coils of each value. To obtain an even ratio as 1,000 to 1,000', one plug is inserted between the block 1,000 and the bar A, and the other plug between the 1,000 block and bar B, the ratio arms are reversed; that is, the 1,000 ohm coil is connected to the X post and the 1,000 ohm coil to the end of the rheostat. When uneven ratios are used, the same ratio can be obtained by four different combinations. To obtain the ratio one to ten, insert a plug between A and 1, and another between B and 10, or between A and 1', and B and 10, and get 1:10, or between A and 1, and B and 10', and get 1 :10', or again, between A and 1' and B and 10', and get 1' to 10'. Other ratios are obtained in a similar manner. By using more than two plugs and connecting certain of the coils in parallel combinations, a large number of other ratios may be obtained. This arrangement offers a convenient method of measuring the sensi- bility of a bridge and galvanometer combination that is frequently applicable. If for instance the one ohm coil is used on either side after a balance has been obtained the one ohm may be shunted with the 1,000 ohm on the same side. This will make a variation of — of 1% and the galvanometer deflection may be noted for this variation. Similarly, the 1 ohm may be shunted with the 100 for a variation of 1%, or with the 10,000 for a variation of rrr^ of 1%. The ten ohm coil may be shunted with the 1,000 for a variation of 1% and with the 10,000 for a variation of tt; of 1%. In the arrangement of ratio coils, errors due to plug contacts are negligible because only two plug contacts enter the circuit, and with an even ratio, it is only the difference in the resistances of the two plug contacts that can affect the result. In measuring any of the ratio coils while in the bo.x it is only necessary to connect to the bar C and to either the bar A or B and plug in the coils to be measured. 496 HAWKINS ELECTRICITY /VWWWWVNf — -f- {^ 3' A^AAA/vA/vVv/J Of)- <(^ lAAAA/VWVNAA^ (-5) Figs. 573 and 574. — The Leeds and Northrup decade. The object of this arrangement is to reduce the number of coils required. In fig. 573, the 1, 3', 3 and 2 are con- nected in series. Let the ter- minals of the 1 ohm and 2 ohm coils be numbered (1), (2), (3), (4) and (5) (fig. 573). The cur- rent enters at point (1) and leaves the coils at the point (5), traversing 1, 3', 3, 2 =9 ohms in all. If this series be multiplied by any factor n, then n ( 1 +3' -H 3+2) = n 9 ohms. It will be seen that if the points (I) and (5) be connected, all the coils are short circuited, and the cur- rent will traverse zero resistance. If the points (2) and (5) be con- nected, the 3', 3 and 2 ohm coils will be short circuited and the current will traverse 1 ohm. By extending the process so as to connect two and only two points at a time it is possible to obtain the regular succession of values n (0, 1,2. 3, 4.5. 6, 7, 8, 9). the last being obtained when no points are connected. Fig. 574 shows Leeds and Northrup's method of connecting these points two at a time \vith the use of a single plug. The circles in the diagram repre- sent two rows of ten brass blocks each. To the first two blocks at the top of the rows, the points (5) and (1), fig. 573, are connected; to the second two, the points (2) and (5) are connected, etc., no points being connected to the last pair of blocks. Hence, if a plug be inserted between blocks 1 and 5, fig. 575, the points (l)and (5) of diagram fig. 573 are connected, giving the value of 0, if between the blocks 2 and 5 the points (2) and (5) are connected, giving the value 1, and so on. The value 9 is obtainable when the plug is in the last pair of blocks, which have no connections. Fig. 572 shows a top view of the blocks of a simple decade constructed upon this plan. o o TESTING AND TESTING APPARATUS 497 Oues. What other advantages are gained with the decade arrangement? Ans. The single plug used with each decade is never out of use, being either in the zero position or set on some value, and is therefore not easily lost by being laid aside. The use of only one plug in a decade makes it easy to ascertain that the plug is Fig. 575. — Leeds and Northrup dial Wheatstone bridge. Rotating switches are used instead of plugs, which permits quicker adjustment of the resistances, adapting it to rapid working. The ratio coils are arranged as in fig. 568. There are four dials which form the rheostat. The units dial contains 9 one ohm coils; the tens dial, 9 ten ohm coils; the hundreds dial, 9 one hundred ohm coils, and the thousands dial 9 one thousand ohm coils. The values of the ratio coils are 1, 1. 10, 10, 100, 100, 1,000, 1,000, 10,000, 10,000. making good contact as only one block in a row is plugged at a time, the other blocks are not kept under a strain by having plugs forced tightly between them. This strain on the blocks, which always exists in those sets in which a resistance is thrown in by removing a plug, tends to separate or loosen them and often to warp the hard rubber upon which they are mounted. Another advantage of the decade plan is that it permits obtaining a succession of values by means of sliding contacts or dial switches, a method which is becoming deservedly more appreciated. 498 HAWKINS ELECTRICITY Oues. What is the difiference between " plug out " and *' plug in " types of resistance box? Ans. In the plug out type, resistance is put in the circuit by remo\'ing plugs, as in fig. 565; in the plug in type, resistance is put in the circuit by inserting plugs as in figs. 570 and 571. Fig. 576. — Queen Acme portable testing set. It consists of a Wheatstone bridge, with reversi- ble arms, battery of four dry cells, D'Arsonval galvanometer, batter>' and galvanometer keys. There ^re sixteen resistance coils, having a combined resistance of 11,110 ohms. Each bridge arm is provided w-ith three coils of 1, 10, 100 ohms, and 10, 100, 1.000 ohms respyectively. The commutator admits of a ratiool 1 to l.OOOon either bridge arm, gi\-ing the set a theoretical range from .001 of an ohm to 11.110.000 ohms. For resistances above 1,000.000 ohms, the normal battery power must be increased. The contact keys are located as shown. The battery key has single contact, but the galvanometer key has double contact ; depressing it closes the galvanometer circuit, and releasing it short circuits the galvanometer, bringing the latter quickly to rest. TESTING AND TESTING APPARATUS 499 Testing Sets. — For convenience in testing, a combination of the instruments used is put up in a neat and substantial case, and known as a testing set. There are innumerable forms of testing set, a few of which are shown in the accompanying illus- trations. The usual combination is a Wheatstone bridge, galva- nometer, battery and necessary keys and connections. Fig. 577. — Connections and circuits of Queen acme portable testing set. There are three rows of blocks, LL', MM', NN. LL is connected to NN' by means of a heavy copper bar, joining L' and N'. LL' and NN' constitute the rheostat, from which any resistance from 1 ohm to 11,110 ohms may be obtained by removing the proper plugs. The block N of the rheostat is connected to the lower line post D. The upper line post C is connected to the block X of the commutator. The block C has no other permanent connection, except key G. The block R of the commutator is connected to the block L of the rheostat, and has no other connection, excepting by plugs. Each half of MM' constitutes a bridge arm, designated A and B respectively. Beginning at the lower line post D, the cornections form a continuous circuit through the rheostat, thence through the bridge arm B, thence through the bridge arm A, thence to the upper line post C. The commutator serves merely to reverse the bridge arms A and B. The battery terminals are connected as shown: the positive terminal directly to the common junction of the two bridge arms, and the negative terminal through the battery key to the rheostat. The positive terminal of the galvanometer is connected through the galvanometer key with the block X, and the negative terminal with the block R of the commutator, or. what is equivalent, with the block L of the rheostat. The commutator blocks A B, R and X, are connected by plugs as shown. When the commutator plugs are in the position PQ, the bridge arm B is connected to the rheostat and the bridge arm A is connected to the line, the ratio between the bridge arms ratio being A -^ B = X -^ R but when the plugs are in the position ST, the bridge arms are reversed in position A. being connected with the rheostat and B, with the line, and the bridge arm ratio becomes A -^ B = R -^ X. The connections of the testing set may be more readily understood from the simplified diagram fig. 578. 500 HAWKINS ELECTRICITY Ques. Describe the operation of the Queen Acme test- ing set figs. 576 and 577, in measuring resistance. Ans. Connect the terminals of the resistance to be measured to the line posts C and D. Place the battery connections on the two upper tips and 1 , thus thro\Ying one end of the battery into circuit, which is sufficient until an approximate balance is obtained. Employ the 100 ohm coil in each bridge arm, and Fig. 578. — Simplified diagram showing connections of Queen Acme portable testing set. place the commutator plugs in the position PQ, or in the position ST. Then remove plugs from the rheostat until the value of total resistance employed, or nearly as may be guessed is equal to that of the unknown resistance. Now press the battery key Ba, and holding it down momentarily, press the galvanometer key Ga. If the galvanometer needle swing to the right toward the symbol + the resistance employed in the rheostat is too high and must be reduced. If the needle swdng to the left toward—, the resistance employed is too low and must be in- creased. By altering the resistance of the rheostat accordingly, TESTING AND TESTING APPARATUS 501 a value will soon be found, which when varied slightly either way, will reverse the deflection of the galvanometer needle. Now remove the battery connection from tip 1, and place it on the tip 4, thus throwing the whole battery into circuit. Then press the keys again as before, first the battery key, then the galvanometer key. This will increase the deflection of the gal- vanometer needle for the same variation in the rheostat, thus Fig. 579. — Diagram of the Queen dial decade portable testing set. Its dimensions are 9H" long, 7" wide, and 7" deep, and weighs 11 J^o pounds. The resistances are arranged upon the dial decade plan, being placed in circuit by means of a rotating switch contact. The switches are so constructed that they maybe turned in either direction, thereby permit- ting them to be turned quickly from the highest resistance in any dial to the lowest resistance in the same dial. This arrangement avoids the necessity of turning back through all the remaining resistances in any particular group of coils and is of value in locating swinging crosses or conditions of momentary balances. The connections for the various tests are made by the manipulation of one small knife switch (W.B. — M.L.) and the switch 'BA..; these are plainly lettered, thus avoiding the necessity of referring to a diagram of connections. In construction , the dial switches are made up of eight laminations of No. 28 B. & S. phosphor bronze, and the form is such as to prevent wearing grooves on the top of the contact studs. In this instrument the electrical circuits are soldered throughout excepting the switch contact whose resistance is negligible. The resistances are wound with manganin. The battery comprises six cells sub-divided which are easily replaceable. The galvanometer is the same as in the Queen acme set, but has the addition of an Ayrton shunt, which is useful in making insulation measurements. The necessary keys, binding posts, and switches are provided so as to facilitate the use of the instrument for the various measurements that can be made with it. 502 HAWKINS ELECTRICITY enabling the making of a more accurate adjustment. The measurement thus made will be the best result that can be obtained with bridge arms of equal value, but by selecting more suitable values of the two arms from the following table of bridge ratios a much higher degree of accuracy may be obtained. Table Showing the Best Values of Bridge .\nns for Measuring any Desired Resistance Value of Resistance being measured Best values of Position of Commutator Plugs as shown in fig. 582 A = B = Below 1 5 ohms 1 1.000 PQ PQ PQ PQ PQ or ST ST ST ST Between 1 5 and 1 1 ohms 1 10 100 100 1,000 1,000 1,000 100 100 1,000 100 100 10 1 " 11 and 78 ohms 78 and 1,100 ohms 1,100 and 6,100 ohms 6,100 and 110,000 ohms 110,000 and 1,110,000 ohms. " 1,110,000 and 11,110,000 ohms Oues. In testing with the Queen Acme set how should the plugs be placed in the commutator? Ans. iVlways make the arm A the smaller except when the two arms are of equal value. Oues. If the resistance being measured is higher than 6,100 ohms, or lower than 1,100 ohms, how should the com- mutator plugs be placed? Ans. If higher than 6,100 ohms, they should be placed in the position ST; if lower than 1,100 ohms, in position PQ. TESTING AND TESTING APPARATUS 503 When the plugs are placed in the ST position, the unknown resistance is found by dividing the value of the larger bridge arm by that of the smaller, and multiplying the total employed resistance in the rheostat by the quotient. When the plugs are placed in the PQ position, the employed resistance in the rheostat is divided by the quotient. Fig. 580. — Queen portable silver chloride testing battery, advantage of long life, light weight, and compactness, new is .8 volt. The silver chloride cell has the The pressure of each cell when Direct Deflection Method with Queen Acme Set. — To measure for instance, insulation resistance by direct deflection connect a known high resistance, say 100,000 ohms between the line post C (fig. 577), and the positive battery post. Remove all plugs from the commutator, and place all plugs in the rheo- stat, as any employed resistance in the rheostat will be in circuit 504 HAWKINS ELECTRICITY with the galvanometer and the battery. Place the battery connection so as to throw only one cell into circuit. Now press the keys and obtain a deflection of the galvanometer needle. For example: asstmie that the needle to be deflected about 8 di\'isions of the scale. Since this deflection is due to the current from one cell passing through a resistance of 100,000 ohms, then Fig. 581. — Ohmmeter. It consists essentially of a slide wire Wheatstone bridge, with the scale di%4ded to read either directly in ohms, or in per cent, of a fixed resistance value. A gal- vanometer is mounted on the containing case of each, and battery and galvanometer keys are pro\nded. In the direct reading type, the scale is so cut that when the galvanometer is balanced, the pointer of the instrument indicates the value of the resistance between the X posts. The scale is calibrated for any desired range. These ohmmeters being slide wire bridges, the greatest accuracy is at the center of the scale, hence one should be selected that will bring the part of the scale likely to be the most used at or near the center. A con- venient tvi)e is that in which the scale is cut in per cent., 100 per cent, being at the center of the scale. Fixt-d coils of 1, 10, 100, 1,000 and 10,000 ohms are contained in the instru- ment with apluggingarrangement allowing any one to be used. When a balance is obtained, the actual resistance is determined by mu!tipl>-ing the dial reading by the value of the fixed coil in use. This amounts simply to shifting the decimal point. For instance, if the 100 ohm coil were being used, and the pointer were at .875, the resistance would be 87.5 ohms. 100,000 X 8 = .8 megohms represents the resistance through which one cell will produce a deflection of one division on the scale. Hence, .8 megohms is the constant of the galvanometer. TESTING AND TESTING APPARATUS 505 Now, replace the known high resistance (100,000 ohms) by the unknown resistance (for instance such as a cable) the value of which is to be determined. Add enough cells to produce as large a deflection of the needle as possible. Assume that 75 cells give a deflection of 1.5 scale division. Then, the galvanometer con- stant multiplied by the number of cells and the product divided by the deflection will give the insulation resistance of the cable ; or Fig. 582. — Commutator plug setting for comparing electromotive forces by the fall of potential method with Queen acme set. .8 megohm X 75 cells = 60.0; and 60.0 -T- 1.5 = 40 megohms as the resistance of the cable. Fall of Potential Method with Queen Acme Set. — To compare electromotive forces by this method, place the battery connection (fig. 577), so as to throw into circuit all the cells, taking care not to reverse them by crossing the battery cords. Plug the commutator as shown in fig. 582, and remove 1,000 ohms from bridge arm B. Place all plugs in arm A. From the rheostat unplug 5,000 ohms. Then connect one of the cells being tested, with its positive terminal to the + battery post and its negative terminal to the line post C. 506 HAWKINS ELECTRICITY When the keys are pressed, the galvanometer needle will swing either to the right or to the left. If it swing toward +, reduce the resistance in the rheostat ; if it swing toward — , add resistance to the rheostat. When a value is found wherein a variation of an ohm either way reverses the deflection, add to this value the resistance unplugged in arm B, and divide the sum by the resistance in arm B. The result gives the ratio Pig. 583 — Diagram of apparatus for measuring low resistances based on the principle of the Kelvin double bridge. In the diagram AB represents a heavy piece of resistance metal of uniform cross section and uniform resistance per unit of length; CD is another piece of resistance metal of smaller cross section, and the two are joined together by a heavy copper bar, AC, into which both are silver soldered; LL are the current terminals and PP are the pressure terminals. The resistance of AB between the marks and 100 on the scale S is .001 ohm. From the point 1 on the resistance CD to on AB is also .001 ohm, from 2 to is .002 and so on, and from 9 to 100 is .01 ohm. The slider M moves along the resistance AB and its position is read on the scale S which is divided into 100 equal parts and can be read by a vernier to thousandths. Subdivided in this way the resistance between the tap off points PP may have any value from .001 to .01 ohms by steps of .000001 ohm. between the voltages of the testing set battery and cell being tested respectively. The division is decimal and may be readily accomplished by merely pointing off as many places as there are ciphers in the resistance employed from arm B. This operation repeated with any number of different cells, will give their voltages in terms of the voltage of the testing set battery, and from these ratios their relative values may be readily obtained. TESTING AND TESTING APPARATUS 507 If the testing set battery be replaced by a standard cell, the first measurement gives at once the voltage of the cell tested. If the voltage of the cell or battery being tested exceed that of the testing set battery, reverse the position of the two batteries, and the subsequent operations, as outHned above, will give the desired results. How to check a Voltmeter with the Queen Acme Set.— In using a set as in fig. 576, first remove about 10,000 ohms from the rheostat, plug the commutator as shown in fig. 582, remove 100 ohms from the arm B, of the bridge, and connect a standard Fig. 5g4.— Kelvin bridge. This includes a low resistance standard of .1 ohm variable by steps of .00001 ohm, a set of ratio coils, and a holder for rods or wires to be measured, with a scale to measure their length. It is also provided with heavy flexibles to be used in measuring the resistances of irregularly shaped pieces. The connections are clearly shown in the diagram. The range of measurements of this bridge is : 1 ohm to .1 ohm hv steps of .001 ohm readily estimated to .0001; .1 to .01 ohm by steps of .0001 ohm readily es- timated to .00001; .01 ohm to .001 ohm by steps of .00001 ohm, readily estimated to .000001; .001 ohm down by steps of .00001 ohm, readily estimated to .000001 ohm. cell with the positive terminal to the + battery post and the negative terminal to the line post C. Then, connect the circuit to the battery posts of the testing set the positive lead to the + post and the negative lead to the — post. Now, press both keys and note the direction of the deflection of the galvanometer needle. If it move toward + , the rheostat resistance is too high; if toward — , too low. 508 HAWKINS ELECTRICITY Change the rheostat resistance accordingly until the balance attained is su :h that a very slight variation of the rheostat re- sistance one way or the other will reverse the galvanometer de- flection. To find the pressure on the circuit, add 100 to rheostat resistance and point off two places. Multiply this value by the voltage and the product will be the desired voltage. If the voltage of the standard cell be exactly one volt, the total employed resistance represents the ^•ni^aje on the circuit. P^G. 585. — Oueen slide wire bridge. It consists of a portable slide wire, Wheatstone bridge arranged to read directly in ohms in addition to its use for locating crosses and grounds. It is complete with battery, galvanometer and telephone receiver. The bridge is balanced by moving the hand stylus until the galvanometer shows no deflection oruntil there is no sound in the telephone receiver. In order to provide a wide range of measurement and maximum accuracy, ratio coils or multipliers ha\'ing values of 1. 10, 100, 1,000 and 10.000 are pro- vided. The scale of the instrument is arranged in two parts, one of which indicates ohms and the other is divided into uniform di\'isions for use when locating crosses and grounds by the Murray and Varley loop methods. A small induction coil is included so as to furnish an alternating current when using the telephone receiver. For instance, in making a measurement on a 110 volt circtiit, assume that the emplo3ang of 7,840 ohms rheostat resistance produces balance, and that increasing or decreasing this resistance by two ohms, reverses the galvanometer deflection. This indication that the setting 7,840 is uncertain, about ^ of 1 per cent. Since the rheostat coils are ad- justed to an accuracy of only | of 1 per cent., that will be about the accuracy of the measurement. TESTING AND TESTING APPARATUS 509 If the pressure of the standard cell be 1.018 volts, then 7,840 + 100 = 7,940. Pointing off two places, gives 79.40, which multiplied by 1.018 gives 80.82 for the voltage on the circuit. To Measure Internal Resistance of Cell with Queen Acme Set.— First compare its voltage on open circuit with the pressure of the testing set battery.. Then, shunt the cell with a known resistance, about 100 ohms, and again measure its ter- minal voltage. The difference between the two values thus Fig. 586. — Evershed portable ohmmeter set. This testing set consists of a direct reading ohm- meter which indicates by direct reading the value of the resistance being tested, also a portable hand dynamo which provides at any required pressure the current necessary to make the test. It is adapted to the needs of supply stations, wiring contractors and dynamo builders. It is also useful in testing the insulation of underground and aerial cables, and is designed so that it can be used by ordinary workmen who are not experienced in handling delicate instruments and who, by its use, are able to obtain accurate results. The dynamo is wound for 100, 200, 500, or 1,000 volts, and is fitted with spring drum inside the case on which is coiled a twin flexible cord provided with a connector adapted for clamping under the ohmmeter terminals. obtained, divided by the value of the shunt resistance, will give the value of the current. To find the internal resistance, mul- tiply the value of the shunt resistance by the ratio between the first and second measured values. 510 HAWKINS ELECTRICITY For instance, assiune that the open circuit voltage of the cell being tested as compared with the voltage of the testing set batten,' is .212 of the latter, and that when it is shunted with a resistance of 1,000 ohms, its terminal voltage is .179. Then, the total resistance is to the 1,000 212 ohms shunt resistance as .212 is to .179 or ^^^^ X 1,000 = 1,184, from .1< 9 which deducting the 1,000 ohms shimt resistance, gives 184 ohms as the internal resistance of the cell. Pic i. lineman's instniment for the location of faults, -r.d telegraph circuits, and for the measurement Ammeter Test with Queen Acme Set. — Connect a low resistance in series with the ammeter and run leads from it to the testing set, the positive lead to the + battery post and the negative lead to the line post C (fig. 577). Insert a standard cell TESTING AND TESTING APPARATUS 511 between the battery posts, with positive terminal to + battery post, and negative terminal to — battery post. Plug com- mutator as shown in fig. 582. Remove 10,000 ohms from rheo- stat, and 100 ohms from bridge arm B. Determine a balance in the usual way by changing the value of the resistance in the Fig. 588.— Diagram showing arrangement and connections of Leeds and Northrup fault finder It IS used to measure conductor resistance, fault resistance, to locate faults by four differ- ent tests and when used with a buzzer and telephone, to locate opens. The essential teature of the instrument is the uniform resistance AB, which lies in a circle and which has a value of about 100 ohms. By a special construction, it is so arranged that the contact can be made at any point along it, and it is therefore equivalent to a very high resistance slide wire. It has a moving contact C and a uniform scale of 1,000 divisions Xn series with this there are the two resistances E and R which may be short circuited by the switch^ U and V. E has exactly the same resistance as the wire AB. R has a resistance of 100 ohms, and is the fi.xed resistance of the bridge arrangement for resistance measurements. The resistances of 1.000 ohms and 9,000 ohms connected to the battery post are to protect the battery and the apparatus from excessive current. The 9,000 ohms may be short circuited by the switch W. u""ia 512 HAWKINS ELECTRICITY rheostat. This operation will balance the difference of pressure at the terminals of the shunt resistance against the standard cell, and its value is equal to 140 1.40 X 100 R + 100 R + 100 To determine the current flowing, divide the value of the difference of pressure thus obtained by the value of the shunt resistance. 2o VofH Swt tch-Sethnq I V. I "^ \^ \J '•~J \ ' I V ^ Res. Fig. 5S9 — Resistance measurement ■nith Leeds and Northrup fault finder. The diagram shows the proper connections and switch settings for measuring conductor resistance. As in the ordinary shde wire bridge, the resistance X between the two posts 1 and 2 is obtained from the formula X = A -i- (1.000 - A) X R. To avoid the necessity of sol\-ing in each case the fraction A -f- (1.000-.\), a table is furnished with the instrument, giving the value of this fraction for each value of A. The resistatue is accordingly delermined in each case by simply selling the contact C for a balance and reading from the table the re- sistance opposite the number corresponding to the scale reading and multiplying by loo, the ralue of R. To use an outside battery, remove the inside battery and connect the outside battery t)etween the posts Gr and Ba. The pressure of this battery should not exceed 110 volts. Jf it exceed 25 volts, open svfitch IV. EXAMPLE — With an unknown resistance connected between the posts 1 and 2. the galvano- meter showed a balance for a dial reading of 387. The number opposite 3S7 in the table is .6313; hence, X = .6313 X 100 = 63.13 ohms. TESTING AND TESTING APPARATUS 513 C "xl.S je a, 2f 3 w o g tvjO O C h „ 00*^ £t3 g o 5.5 K.WJk-'i-iUry"'«T;t/}D- flj -5^- " S " . g-c S'^^'S .«l-Ci-3„M.S^M> o o rt^ o-o 3 a>>5 [^ ■fi— 3 t;-- cp &-^ D loS, a o;s 3 I" o t; „ 0) <->J3-3 " ^^ y.S "^ " >---"-9' — ol: c 5 "-K b-^Li " a •S-o H-s "> " SOX " " ^ a- " li 4) — - 'J's So*;- >.'o o -c 2 " 3 ° •5 cits o^s > ^"- ^^-2 ^2-gt'So>P-0:Si 2JS-5.52 o-c c M ;;j3 •2S-°t5oS£^2.S°^ J3a .-"w-mOh^ cH " M "'t:'-' ™ . u a >.3. c S « fcl .3 3«^*^ E o~T3 a a; .2 c 0x1 L. c.i2 « u « o J> oote o ■o E-g ^- Ct3 05! 5? t- „ rt c lu*^ S «; 3 g.H "■S'-'tS > C Mc ° w— fc.s i2^-^& g.>,S3 "S 5 tio «J3 ai£ ■"_^ p .^X o*" _° 01 .5 5 >,J5 „-5 jj-g ~r^ " w S 4)13 fc O r3-<4;iurtr;cO~ "O o o — t3 5 o ^ c . § !5-o -o bcii 2 MtSw C E &.£ b 2'ing the resistance in arms A or B. Then by Wheatstone bridge formula: RX = AY, and L = X+Y; Y = L - X, whence X A R + A Y = R B + A ooo-^ 1000-^ ^"'V Sw/tch- Setfinq 8 [m] e-eQ [m)©— e © Fig. .596. — Murray loop method of fault location with Leeds and Xorthrup fault finder. Case I where there are two wires ha\'ing equal resistance, in one of which there is a fault. Connect and set switches as shown; join the good wire to post 1 and the faulty wire to post 2. The resistance of E is equal to that of AB. From the symmetry of the arrange- ment, it is evident that, if the fault were exactly at the junction between the good and bad wires, the contact point C would rest for a balance at 1.000 on the scale, or at 500 if the fault were half-way along the bad wire; hence, at whatever point it comes to rest, the reading divided by 1,000 and multiplied by the length of the bad wire is the distance from the instrument to the fault. EXAMPLE — In a pair of equal wires, 5.8 miles long, one is grounded. With the connections made as above, and the galvanometer balanced for the dial reading 124, the distance tc the fault is (124 X 58) ^ 1,000 = .7192 miles. Oues. How may the distance from 2 to the fault be determined in knots or miles. Ans. Divide Y by resistance per knot or mile. TESTING AND TESTING APPARATUS 517 The Varley Loop. — This is a method of locating a cross or ground in a telephone or telegraph line or other cable by using a Wheatstone bridge in a loop formed of a good wire and the faulty wire joined at their distance ends. One terminal of the battery is grounded and the other connected to a point on the bridge at the junction of the ratio arms. The rheostat arm then includes the resistance of the rheostat plus the resistance of the 9000 Swifch>-5eHir7(3. Se-e e Fig. 597. — Murray loop method of fault location with Leeds and Northrup fault finder: Case II, where the good and bad wires are unequal. The figure shows the connections. It is the ordinary Murray loop and it is evident that the resistance a, to the fault will be obtained from the formula a = (A -^ 1,000) X r, where r is the resistance of the loop, and A is the reading of the contact C on its scale. The distance d, to the fault is obtained from the formula d = Ar ^ (1,000 X M), where M is the resistance per mile of the faulty wire. EXAMPLE — A wire having a resistance M of 16.46 ohms per mUe is grounded. This wire was looped with a wire of unknown resistance and the total resistance of the loop r was measured and found to be 54.07 ohms. Connections were made as in the figure, and the reading A was found to be 332. Substituting these values in the above formula: d = (332 X 54.07) -T- (1,000 X 16.46) = 1.09 miles. fault, while the unknown arm includes the resistance of the good wire plus the resistance of the bad wire beyond the fault. When the bridge is balanced, the unknown resistances may be readily determined by a simple equation. 518 HAWKINS ELECTRICITY In making the Varley loop test, the resistance of looped cable or conductors is measured, and then connected as in fig. 598, Close the battery key and adjust R for balance. When earth current is present, the best results are obtained when the fault is cleared by the negative pole, and just before it begins to polarize. If X be the resistance from 2 to the fault, then X = X SffOUND -m r ii B*TTEi?V KV( I ^ ||.' Fig. 598. — The Varley loop test. The diagram shows the various connections. X and Y are the resistances of the cable between the fault and the points 1 and 2 respectively. L is the resistance of the good and bad cable or X + Y. also, X divided by the resistance of the cable or conductor per knot or mile gives the distance of fault in knot or miles. When the resistance of the good wire used to form a loop with the defective wire, together with that portion of the defective ^vire from the joint to the fault is less than the resistance of the defective wire from the testing station to the fault, the resistance R must be inserted be- tween point 1 and the good conductor, the defective wire being con- nected directly to point. The formula in this case is X = L + R TESTING AND TESTING APPARATUS 519 Special Loop. — This method may be used to advantage where the length of the cable or fatilty wire only is known and where there are two other wires which may be used to complete the loop. It is not necessary that the resistance of the faulty wire and the length and resistance of the other wires be known. Figs. 601 to 604 show the connections and method of testing. EXAMPLE.— All the wires in a cable 10,852 ft. long were found to be grounded so that none of them could be used as good wires. Two wires were selected out of another cable going to the same place by a different ffOOO Sv/itcf?- Setf irx] V Figs. 599 and 600. — Varley loop method of fault location with Leeds and Northrup fault finder. This method may be used as a check on the Murray methods. Connect the faulty wire to 1, and measure the resistance of the loop. Then throw switches as shown in the fig. 600. Let: a = resistance to fault, d = distance to the fault in miles, M = resistance of the faulty wire per mile, r = resistance of the loop, R = resistance of the coil R, or 100 ohms, T = A -7- (1,000— A) to be read from the table. From the Wheatstone bridge relation: a = (r-100 T) -^ (T + D. and d = (r-100 T) -J- (T+1) M. EXAMPLE — A wire having a resistance of 16.46 ohms per mile is grounded. This was looped with a wire of unknown resistance and the resistance of the loop was found to be 54.07 ohms. Connections were made as in the figure, and the reading A was found to be 234. From the table T = .3055, and substituting: d = (54.07-30.55) -r (1.3055 X 16.46) = 1.094 + miles. route and securely joined to one of the grounded wires at the distant, end. This grounded wire and one of the good ones were connected as shown in figs. 601 and 602 and the reading A was found to be 307. 520 HAWKINS ELECTRICITY Connections were then made as shown in figs. 603 and 604 and A was found to be 610. What is the value of d? A xx)rding to formula , AL 307 X 10,853 , , ., . ^ = "A = 610 = ^'^^^ ^'- The Potentiometer. — For the rapid and acctirate measure- ment of voltage, current, and resistance, the potentiometer can be recommended. Those in charge of electric light and power Svs/it'cb -Setting I y. I V-/* ^^ '^ '^ l ' I Pigs. 601 and 602. — Si)ecial loop test with Leeds and Northrup fault finder. For the first measurement connect the faulty wire to 2, either of the good wires, as Z, to 1, the post Gr to ground, and short circuit the coils R and E by closing switches U and V as in the figures . Balance in the usual way and call the dial reading A. For the second measure- ment, connect the post Gr. (disconnected from ground), to the other good wire y as shown in figs. 603 and 604. and get another balance; call this reading A'. The distance d, to the fault is determined from the simple formula d = AL -r A' where L is the length of the cable or faulty wire. companies, and also those who purchase large amounts of electrical energy are realizing, more and more, the necessity of having satisfactory' primary standards with which to check their volt-, ampere-, and watt-meters. TESTING AND TESTING APPARATUS 521 When it is realized that an error of one per cent, in a com- mercial instrument means an error of one dollar one way or the other in every one hundred dollars charged, the need of such standardization apparatus becomes at once apparent. The potentiometer, it should be noted, relies for its accuracy, only upon the constancy and accuracy of resistances and upon standard cells. WWVSV- w Svifcb-Settipg. 00^ © Figs. 603 and 604. — Special loop test as made with the Leeds and Northrup fault finder. Diagram showing connections for the second measurement. The special loop test may be used to advantage where the length of the cable or faulty wire only is known, and where there are two other wires which may be used to complete the loop. To use an outside bat- tery, connect one pole to Ba, and ground the other. The pressure of this battery must never exceed 110 volts; if it be over 25 volts, see that switch W is open. With the materials now available, and the skill which has been acquired in their manufacture, both the resistances and the standard cells are obtainable which are remarkably constant, and both can be readily checked for accuracy. 522 HAWKINS ELECTRICITY Location of Opens. — These measurements are based on the fact that the capacity of -^sires in a cable is ordinarily a measur- able quantity, which, in wire of uniform diameter, is propor- tionate to length. In making these tests, a fault finder is used together with a buzzer, dry cells to operate it, small induction coil, and telephone receiver. These instruments are to be found in any telephone exchange. It is best to locate the buzzer at some distance from the fault finder in order that it cannot be heard by the operator. V Fig. 605. — To use galvanometer of Leeds and Northrup fault finder in series with the battery: Set switches as shown, and connect between posts Gr. and 2 (see figs. 587 and 5&8). The galvanometer will have the maTi'Tniim sensibility with the pointer at 1,000 and the mini- mum at zero. Fig. 606. — To measure high resistances, such as the resistances of faults with Leeds and Xorthrup fault finder. First Method. — Arrange the switches as shown in the figure. Con- nect p)Osts Gr. and 2, turn the handle until the galvanometer needle comes to rest at an even deflection of ten divisions. Call the reading A. Connect in the unknown resistance between Gr. and 2. Now close the switch W, so that the figure 1 appears on the top of the block, and again bring the galvanometer to a deflection of ten divisions and call the reading B. Then X = (10.000 B -^ A) - 1,000. In case X be a high resistance, it will be found that the galvanometer will not deflect ten di%-isions for any position of the pointer. In such case, choose a number of di%-isions which is a factor of ten, such as 5. 2, or 1, and multiply (10,000 B -J- A) by ten di\'ided by the number chosen, as y y . y. For ex- ample, for a deflection of two di^-isions: X = ^ (10,000 B -i- A) - 1,000. The satisfactory range of the set for high resistance measurement may be increased by using an outside batterv of higher voltage. ^Vilh the contained batterj', satisfactory measurements can be made up to 1 or 2 megohms. When outside battery is used, connect one terminal to the j)Ost ba, and the other to 2 for the reading A. Connect the batter>' and imknown resistance in series between these posts for the reading B. When an outside pressure of 25 volts or over is ussd, the switch W should not be closed unless there be a resistance in series with the battery of 10,CKX) ohms or over. Second Method. — For use as a voltmeter to measure high resistances. (More convenient but not quite as accurate as first method.) Set the switches to RV. M and 10. Turn the knurled nut on the galvanometer so as to set the needle to the extreme light hand side of the scale. Connect the posts 2 and Gr. with a short piece of wire. Turn the rotating pointer on the scale until the galvanometer needle moves over about 20 scale di%'isions when the battery key is closed. Remove the connection between 2 and Gr. as the voltmeter terminals. This makes a simple way of testing for various kinds and amounts of trouble. On a wet cable a deflection of 10 to 15 di\Tsions indicates hea\->- enough trouble to locate with the fault finder. With a little care, trouble showijis only 5 or 6 divisions can be located. TESTING AND TESTING APPARATUS 523 Before attempting locations for opens it is well to make the following measurements : 1. The insulation of the broken wire and the insulation of the good wire with which it is to be compared ; This may be done as shown in fig. 606. It is best that the insulation resistance be fairly good, but experiments indicate that good results can be obtained by the methods which follow, even when the insulation is as low as 100,000 ohms, and fair results when as low as 50,000 to 100,000 ohms. 9O00 Sw'ifchSetfing /V BAD PA/R^ GOOD f^/R Pig. 607. — Diagram of connections in testing for opens with Leeds and Northrup fault finder. The apparatus required consists of fault finder, buzzer, dry cell to operate buzzer, small induction coil, and telephone receiver. Connect the battery to the primary of the induc- tion coil, one terminal of the secondary to the post Ba, and the other to the connected wires as shown. Set switches U and V so as to short circuit the two resistance coils. 2. The resistance between the two sections of the broken wire :>hould be measured. This may be done by joining the broken wire and a good wire at the distant end of the cable and measuring the resistance of the loop. To ensure close locations, this resistance should be over 100,000 ohms. Fair 524 HAWKINS ELECTRICITY locations can be made when the resistance is much lower and it is worth while to attempt it even if the resistance be as low as 10,000 ohms. The difficulty of determining the balance point increases as the resistance decreases. Oues. Describe the method of locatmg an open with a fault finder. Ans. {Case I) The broken wire \\'ill be one of a pair. Select another pair in the cable that will have the same capacity per mile and join together the mate of the broken wire and one wire Pig. 608. — Diagram of connections in testing with Leeds and Northrup fault finder for open wire in telegraph and other cables in which the wires are not grouped in pairs. Connect the broken wire to 1. Select a good wire and join to 2. Connect all o:her wires and ground them, by connecting to the cable sheath. Connect the distant end of the broken wire to the others. Ground the end of the induction coil that is not connected to the post Ba. of the other pair. Make the connections as shown in fig. 607, then depress the battery key and move the contact to the point of minimum sound in the telephone. The distance to the break is equal to LA -^ (1,000 — A), where L is the length of the good wire. TESTING AND TESTING APPARATUS 525 EXAMPLE: A cable 1.45 miles long contained a broken wire. It was found that the insulation resistance of the end of this wire was over 10 megohms, as was that of the good pair selected to test against it. The resistance between the two pieces of the good wire was also over 10 megohms. Connections were made as in fig. 607, and it was found that the balance point was 476. Accordingly from the table .9084 and 1,000 - A d = 1.45 X .9084 = 1.317 + miles. |«n|l|l| 1 r-WS" •Switch -5ctf I pg. BAD R^/R N GOOD PAIR Fig. 609. — Diagram of connections for reading Tn testing for opens with Leeds and Northrup fault finder, when broken wire and good wire are not in the same cable. Location of Opens. — {Case II) Open wire in telegraph or other cables in which the wires are not grouped in pairs. The connections are made as in fig. 608, and the measurement and calculation exactly as in the preceding case. The accuracy of the location of both of the above methods depends on the good and broken pair, or the good and broken wires having equai and uniform capacity per unit length. It is not always possible to select 526 HAWKINS ELECTRICITY wires that are alike in this respect. In such cases, as for instance, where there is no good wire in the cable containing the broken wire, and a good wire has to be selected from another cable, the method of C4ise III may be used. Location of Opens. — Case III, in which the broken wire and good Vi-ire are nor in the same cable. Connect the good wire and broken wire in the same way as shown in fig. 607, and set Pic. 610. — ^Leeds and Northnip potentiameter. It is direct reading from .000001 vott to 16 \'olts, asd with accessories tiie range may be extended to 1600 volts, and cmrents may be measured up to 3CNX) amperes. The instrument has fifteen cofls of 5 ohms eac^, wfaicfa are in series vith an extended trire aboat 190" long of equal leastance. Tbe dectrical circuits are shown in tbe diagram fig. 611. It is well for the user to open np the potenti> ometer and make himself familiar with its interior constructi<», in otxler to fnlly under- stand the operation of the rheostat and other pans. There are no con tart resastanoes in the potentiometer drcnit proper. The potentiameter has low internal resistance wfaidi gives it the maximum sensibility. Compared with faigfa resistance potentiameter, tins is especially advant.ageous in measuring the e!ectroax>tive force at tbennooonples. and the fall of potential acrc'ss s:.andard low resistances. As ooostmctad. tbe last ooe^tenth volt is covered by the extended wire and the handle which carries the oootact potat on tbe wire may be manipulated rapidly so that a fluctuating voltage may be aoctnatdy followed. When used with any cadmium cell, the pipteniiometer is direct reading. The accuracy of the potentiometer resistances can be voified with the facilities of the ofdinuy labamtory. the pointer for a balance. Call the reading A. Then connect the good wire and the broken wire at the distant end and set the TESTING AND TESTING APPARATUS 527 -Ba. Fig. 611. — Diagram showing connections of Leeds and Northrup potentiometer. The coils in the series AD are each o ohms, and between each two there is a brass block with a reamed hole. A pair of flexible cords with taper plug terminals to fit these holes is furnished. These coils can be measured with an ordinary Wheatstone bridge and thus compared with each other to a high degree of accuracy, even if the bridge be not accurate. For potentiometer work, the essential point is that they should be like each other, not that they should be accurately any particular value. In the same way the resistance of the extended wire can be compared with the resistance of the coils in AD. Its resistance should be 1.1 times the value of any coil between A and D. Outside connection with the extended wire may be made by using the posts marked BR and — BA. This adjustment for balancing an unknown electromotive force is accomplished by the manipulation of the two contact points M and M'. The coils AD are arranged in a circle, a revolving switch moving M. A checking device enables the operator to set this switch without taking his eye from the galvanometer. The resistance S is of such value that when it shunts the wire OB, the total resistance between O and B is Jq of the same urrshunted. When the shunt is applied, provided the total current remain the same, the drop between any two points on AB will be ^, of its previous value. The total current will remain the same provided the total resistance in the circuit remain the same. This is accomplished by making the coil K such that it exactly compensates for the reduction in resistance caused by plugging in the shunt coil S. The low scale is applied by moving the plug from the position 1 to the position .1. With this change the potentiometer reads from .16 volt down by indicated steps of .000005 volt. The reading is very simple. For instance, if M stand at 1.2 and M' at 1.35 revolutions, the reading is 1.2135 volts. The resistances of the instrument are wound upon metal spools, and are therefore able to dis.sipate a com- paratively large amount of energy. This allows the potentiometer to be used for pressure measurements up to 16 volts without the use of a, volt box. 528 HAWKINS ELECTRICITY pointer for a new balance. Call this A'. The connections for this reading are sho\^Ti in fig. 609. The distance to the break will be A A'L 1,000 (A - A' ) + A A' where L is the total length of the broken wire. ^jTb Wtffry A 5^ Pig. 612. — Diagraxa showing actual connections in the rheostat of Leeds and Xorthrup potentio- meter. The figure corresponds to R of fig. 611. The rheostat is mounted in the end of the potentiometer as shown in fig. 610. Rough adjustment of the potentiometer ctirrent is made by means of the variable resistance R. Fine adjustment is made by means of the variable resistance R'. It will be noted that the 23 ohm resistance of this latter rheostat is shunted by a resistance of 6.1 ohms, making possible a very fine regulation. Further, there is in series with the moving contact a resistance of 400 ohms, which makes the effect of variable contact resistance negligible. Only one cell of storage battery should be used. When this battery is fresh, the plug shown in the figure at 2R should be inserted at R. This gives the greatest resistance in the rheostat circuit. As the cell runs down, the plug should be changed to 2R. When both plugs are in, the rheostat slide wires are in series with the potentiometer circuit. EXAMPLE : A pair of wires containing one broken wire was connected with a good pair in a different cable as shown in fig. 607. The reading A was found to be 180. The good and bad wires were then joined at the distant end as in fig. 609, and the reading A was found to be 88. The total length of the bad wire MX was 1.44 miles. Required, the distance to the break. Substituting the values in the formula: d = 180 X 88 X 1.44 1,000 (180 - SS> + ISO X 88 = .211 + mile. TESTING AND TESTING APPARATUS 529 To Pick Out Faulty Wires in a Cable. — Short circuit the coils E and R with switches U and V, Set the pointer at 1,000. Connect the post Gr. to ground or the cable sheath and apply the wires one after another to the binding post 2. The gal- vanometer will deflect for a faulty wire. Fig. 613. — Diagram of the Crompton potentiometer. In this instrument the resistance con- sists of fourteen coils, each of 10 ohms, in series with a straight wire, also 10 ohms resistance, thus forming a system of fifteen equal steps. Across the whole a pressure of 1.5 volt is applied from a secondary cell, thus providing .10 volt per step. Any fraction is then tapped off by means of a radial switch on the resistance coils and a sliding contact on the wire. The standardization is performed by adjusting a resistance in series with the whole until the standard cell employed indicates, by means of the galvanometer G, a balance at the point which represents its electromotive force on the basis given above. Oues. What is a potentiometer? Ans. An arrangement of carefully standardized resistances for measuring voltages in comparison with a standard cell. It is used for accurate measurement of voltages, currents, and resistances. In place of a series of standardized resistances, a slide wire may be used as in fig. 614. Ones. Describe one form of potentiometer. Ans. As shown in fig. 614, it consists of a fine German silver wire about 3 feet long stretched between the binding posts A, B, 530 HA WKIXS ELECTRIC I T Y which are attached tx) a wooden base carrying a scale diNnded into 1,000 equal parts. There are three circuits, the terminal A being included in each, one including the batter}% and the other two the galvanometer. A three point switch connects the galvanometer in series with the standard cell SC, or the cell to be tested C, the circuits being completed by leads terminating in the sliding contacts M and S. ^^SmiCH OJUSTABLE ESISTi^NCE Fig. 614. — Digram .— C i) ZX —7" o - = =3 c ' - = o ='-S'=3 '^ 5 5^" =« ♦^ act ^='J'Sg-£2'SE Soc £■5^-5; i ki d u ct AMMETERS. VOLTMETERS. AND WATTMETERS 541 Ones. How does the winding differ in ammeters and voltmeters ? Ans. An ammeter coil consists of a few turns of heavy wire (when designed to carry the full current), while a voltmeter coil is wound with many turns of fine wire. Thus, the ammeter is of low resistance, and the voltmeter of high resistance. Fig. 623. — New moving element of Keystone instruments, weight 1.2 grams. Fig. 624. — Mo\'ing element of Keystone instruments assembled in bearing. The moving element consists of coil, counterpoise and pointer. The mechanical connections are made by means of screws and steady pins. In order to adjust for slight set or subset of spring under long use a zero adjuster is provided by means of which this set can be connected and the pointer brought back to zero. 542 HAWKINS ELECTRICITY Ques. Why is a high resistance coil used with a volt- meter? Ans. As actuall}'^ constructed, most voltmeters are simply special forms of ammeter. From Ohm's law, the current through a given circuit equals the pressure at its terminals di\'ided by its resistance. Hence, if a high resistance be connected in series with a sensitive ammeter that will measure ven.' small currents, then the current passing through the circuit is directly proportional to the voltage at its terminals, and the instrument may be cal- ibrated to read volts. Figs. 625 and 626.- such that ail the currr- where the instrumer.: . across a shtmt.as in f.„ _ __. — ._ ^;-_--. instrament and the remainder iixrou^ ihe aauiii Ques. Into what two classes may ammeters be divided ? Ans. They are classed as series or shunt according to the way they are designed to be connected with the circmt. Ques. What determines the mode of connecting am- meters? Ans. When the wire of the ammeter coil is large enough to carry the whole current, it is coimected in the circuit in series as shown in fig. 625. If, however, the wire be small, the instru- ment is connected in parallel with a shunt of low resistance, so that it only carries a small part of the current, as in fig. 626. AMMETERS. VOLTMETERS, AND WATTMETERS 543 For circuits which carry large currents, the shunt connection is always used, because otherwise the coil of the ammeter would have to be very heavy and the instrument correspondingly bulky. Oues. How are shunt ammeters arranged to correctly measure the current? Ans. The coil is arranged so that a definite proportion of the whole current passes through it. A large conductor of low re- sistance is connected directly between the two terminals or bind- ing posts of the instrument; the coil is connected as a shunt around a definite part of this main conductor; then, since the two axe connected in parallel and each branch has a definite &■' IJl[.l ^ 1,000 AMPhKh; Type B Shunt 400 Ampere Type D Shunt Figs. 627 and 628. — Westinghouse ammeter shunts. These shunts are used where heavy currents are to be measured. The shunt is connected in series with the bus bar or circuit to be measured, and its terminals are connected by means of small leads to the ammeter or other instrument. These shunts are designed to have approximately 50 millivolts drop at full rated current. They are intended primarily for Westinghouse meters, but can be used satisfactorily with any meter requiring 50 millivolts for full scale deflection. resistance, the current divides between the two branches directly in proportion to their relative conductivities, or inversely ac- cording to their resistances. The coil, therefore, takes a defi- nite part of the whole current, and the force moving it and its pointer away from the zero position is directly proportional to the whole current. Hence, by providing a proper scale, the value of the entire current will be indicated. Oues. How is a voltmeter connected ? Ans. A voltmeter is always connected to the two points, whose difference of potential is to be measured. 544 HAWKINS ELECTRICITY For instance, to measure the voltage between the two sides A and B of the circuit shown in fig. 629, one terminal of the voltmeter is connected to wire A, and the other to wire B. If the " drop " or difference in voltage through a certain length of wire L, of a circuit, as from A t-o B in fig. 630 is to be determined, one terminal of the voltmeter is connected to A and the other to B. In a similar manner is found the drop through a lamp. Ques. What is the difference between a voltmeter and an ammeter? Ans. A voltmeter measures pressure, while an animeter measures current. As actually constructed, most voltmeters are simply special forms of ammeter. a B Fig. 629. — ^Voltmeter connection for measuring the pressure in an electric circuit. The volt- meter is connected in parallel in the circuit at the point where the voltage is to be meas- ured. Fig. 630. — Voltmeter connection for measuring the "drop" or fall in voltage in a certain length of wire, as for instance, the length between the points A and B. The voltmeter is shuttled between the two points whose pressure difference is to be measured. If a high resistance be connected in series with a sensitive ammeter that will measure very small currents, then the current passing through the circuit is directly proportional to the pressure or voltage at its terminals and the instrument may be calibrated to read volts. Ques. Explain the term " calibrate." Ans. To calibrate a measuring instrument is to determine the variations in its readings by making special measiu-ements, or by comparison with a standard. Ques. Describe a solenoid or plunger ammeter. Ans. This type consists of a "plunger" or soft iron core arranged to enter a solenoid. Current being passed through the wire of the solenoid causes the core to be more or less attracted AMMETERS. VOLTMETERS, AND WATTMETERS 545 Fr.. TO LOWER SPRING -^ nO UPPER 5PRm6 631. — Weston ammeter; view showing shunt enclosed within the instrument. West n instruments are direct reading and dead beat. Although the scales have practically uniform divisions, it is not assumed in the calibration that they are uniform, and the scales are not printed or engraved. The method of calibration consists in laying out each large division of the scale bycomparing the instrument with a standard, and then inking in the division lines so found. The smaller divisions between the large ones are then equally spaced and marked by a mechanical method. KiG. 632.— Weston Portable voltmeter, inspector's style. This instrument is provided with a reversmg Icey. Instead of the regular binding posts, pins are used with which connections are .made by means of contact cups attached to flexible cords. These contact cups are convenient m making connections, or in changing quickly from one range to the other, if ine mstrument have a double scale. Connections for the different ranges are made in pre- cisely the same way as with the regular double scale voltmeters. For the upper scale values, the contact pm to the right and the front contact pin to the left being taken, and lor the lower scale values, the left contact cup is changed to the rear contact pin. 546 HAWKINS ELECTRICITY against a restraining force of gravity or springs. A pivoted pointer attached to the core indicates the current value on a graduated dial as shown in fig. 633. Ques. What are the objections to plunger instruments? Ans. They are not reliable for small readings, and are readily affected by magnetic fields. •WEIGtrr PLUNGER SOLENOID Fig. 633. — Plunger type instrument. The current to be measured passes through the solenoid, producing a magnetic effect on the soft iron plunger which tends to draw it into the coil, and thus cause the pointer to move over the graduated scale. The distance the rod moves depends on the value of the restraining force (which may be springs or gravity), the coil winding, and strength of current. The winding consists of a few turns of heavy wire for an ammeter, and a large number of turns of fine wire when constructed as a volt- meter. Since the iron has a certain amount of residual magnetism, the deflection with smaller following large currents is more than would be produced by the same current following a smaller one. The instmment therefore is less reliable than the usual types. Ques. Describe a magnetic vane instrument. Ans. It consists of a small piece of soft iron or vane mounted on a shaft that is pivoted a little off the center of a coil as shown in fig. 634. The principle upon which the instrument works is AMMETERS. VOLTMETERS, AND WATTMETERS 547 that a piece of soft iron placed in a magnetic field and free to move will move into such position as to conduct the maximum number of lines of force. The current to be measured is passed around the coil producing a magnetic field through the center of the coil. The magnetic field inside the coil is strongest near the inner edge, hence, the vane will move against the restraining force of a spring so that the distance between it and the inner edge of the coil will be as small as possible. A pointer, attached to the vane shaft moves over a graduated dial. •SCALE •POINTER SPRING ' VANE OIL Fig. 634. — Magnetic vane instrument. A soft iron vane, eccentrically pivoted within a coil carrying the current to be measured, is attracted toward the position where it will con- duct the greatest number of magnetic lines of force against the restraining force of a spring or equivalent. Ques. Describe an inclined coil instrument. Ans. As shown in fig. 635, a coil carrying the current, is mounted at an angle to a shaft to which is attached a pointer. A bundle of iron strips is mounted on the shaft. A spring re- strains the shaft and holds the pointer at the zero position when no cturent is flowing. When a cvirrent is passed through the coil, 548 HAWKINS ELECTRICITY the iron tends to take up a position with its longest sides parallel to the lines of force, which results in the shaft being rotated and the pointer moved on the dial, the amount of movement de- pending upon the strength of the current in the coil. The coils for large sizes are generally wound with a few turns of flat insulated copper ribbon. The instruments are adapted to either direct or alternating currents but are recommended for alternating currents. Oues. W hat is the principle of the hot wire instrument? Ans. Its action depends upon the heating of a conductor by the current flowing through it, causing it to expand and move an POIMTER^ SPRING iSCAl-E. MAGNETIC VANE CO\L Fig. 635. — Thompson inclined coil ammeter. It is constructed on the magnetic vane principle in which an iron vane is attracted by the magnetic field due to the coil, so as to turn itself parallel with the axis of the coil, the latter being inclined with respect to the axis of the vane. The voltmeter of this type has a similarly placed stationary coil, but in place of the iron vane, is pro%-ided with a mo%-ing coil in series with the other coil. The restraining force in each case being that due to springs. Figs. 636 and 637 show the actual construction of inclined coil instruments. index needle or pointer, the movements of which, by calibration, are made to correspond to the pressure differences producing the actuating ctirrents. Oues. ments? What are the characteristics of hot wire instru- Ans. Voltmeters of this type are not affected by magnetic fields, and as their self-induction is small, they can be used on AMMETERS, VOLTMETERS, AND WATTMETERS 549 either direct or alternating currents; but they possess certain serious defects : they consume more current than the other types ; cannot be constructed for small readings; are liable to burn out on accidental overloads ; and are somewhat vague in the readings near the zero point and are sometimes inaccurate in the upper part of the scale. Figs. 636 and 637. — Thompson inclined coil portable indicating instruments. Pig. 636, am- n^eter type P interior; fig. 637, wattmeter, type P, interior. These instruments, though primarily designed for use on alternating current circuits, may also be used on direct current circuits, by making reversed readings and taking the mean as the true indication. The voltmeters and wattmeters are constructed on the dynamometer principle and the ammeters, on the magnetic vane principle. The voltmeters and wattmeters are provided with a contact key which may be locked in position, enabling the instruments to be left conscantly in circuit. The movements of the pointer are damped by means of an air vane; the' 3 is also a friction damping device operated by a small button to check excessive osc llations of the pointer. The inclined coil instrument? are so designed that the torque is ufficiently high to insure the pointer assuming a definite position with each change in cuixent value. Oues. Describe the construction and operation of the Whitney hot wire instruments. Ans. As shown in fig. 638, a wire AX, of non-oxidizable metal, of high resistance and low temperature coefficient, passes over a pulley B mounted on the shaft C. The ends of the wire are attached to the plate E at its ends F and G, the wire being in- sulated from the plate at G. A spring H holds the wire in tension and takes up the slack due to the expansion caused by the heat- ing of the wire when a current passes through it. The current 550 //.4irA'/.V5 ELECTRICITY flows only in the portion of the wire marked A, between the plate E and the pulley B up to the point K where the connection is shown. When a cturent flows through the wire A, the spring takes up the slack, pulls A around B, and causes B to rotate upon its shaft C. It is clear, that a pointer attached to C, would indi- cate on a scale the movement of B and C, but as this movement is very slight, a magnifpng de\-ice will be required. This de\'ice Fig. 638. — piagram showing principle and construction of the Whitney hot wire instruments. The action of instruments of this type depends on the heating of a wire by the passage ot a current causing the wire to lengthen. This elongation is magnified by suitable mechan- ism and transmitted to the pointer of the instrument. consists of a forked rod L, rigidly attached to the shaft C, and carrying at its lower end a silk fibre fastened to the fork and passing around a pulley M, to which a pointer N is attached. For direct current measurements only an electromagnetic system is used. Oues. What is the principle of electrostatic instru- ments? Ans. The action of these instruments depends upon the fact that two conductors attract one another when any difference of AMMETERS, VOLTMETERS, AND WATTMETERS 551 electric pressure exists between them. If one be delicately sus- pended so as to be free to move, it will approach the other. Oues. Describe the Kelvin electrostatic voltmeter. Ans. A simple form consists, as shown in fig. 639, of a metal case containing a pair of highly insulated plates, between which a deHcately mounted paddle shaped needle is free to move. When the needle is connected to one side of a circuit and the stationary Fic. 639. — Kelvin electrostatic voltmeter; a form of instrument designed for measuring high pressures up to 200,000 volts. The instrument, as illustrated, consists of fixed and movable vanes with terminals connecting with each. These vanes which act as condensers take charges proportional to the potential difference between them, resulting in a certain attrac- tion which tends to rotate the movable disc against the restraining force of gravity. In the figure aa and b are two fixed vanes and c a movable vane, carrymg a pointer and hav- ing a proper weight at its lower end. plates to the other side, the needle is attracted and moves be- tween them as indicated by the pointer. Adjusting screws at the lower end of the needle allow it to be balanced so that its center of gravity is somewhat below the center of suspension. Gravity then is the restraining force. 552 HAWKINS ELECTRICITY The range of the instrument may be changed by hanging different weights upon the needle. B3' increasing the number of blades the in- strument can be made to measure as low as 30 volts. The form having two stationary blades and one movable blade is suitable for measuring from 200 to 20,000 volts. The quadrant electrometer or laboratory form will measure a fraction of a volt. Fig. &40. — Thompson astatic instniment without cover. When cairrent passes through the coils of the moving element, the lines of force parallel to the shaft produce a torque which tends to turn the shaft and caxise the needle to travel across the scale. This action is. of course, opposed by the magnetic field at right angles to the shaft acting on the two pieces o! magnetic metal. These astatic instruments have no controlling springs. The two small silver spirals which conduct the current to and from the armature are made of untempered silver and exert no force as springs. The actuating and restraining forces are dependent upon the same electromagnets. The dampring effect in these instruments is produced by an aluminum disc mo\'ing in a magnetic field, and is proportional to the square of the loag- net strength. Ques. Explain the construction and principle of the Thompson astatic instruments. Ans. The fields of these instruments are electromagnets wotmd for any specified voltage and provided with binding posts separate from the current posts of the instrument. The moving coils are mounted upon an aluminum disc and are located in a magnetic field which is parallel to the shaft and astatically arranged. Two small pieces of magnetic metal are rigidly AMMETERS, VOLTMETERS. AND WATTMETERS 553 mounted on the shaft and the astatic components of the magnetic field, which are perpendicular to the shaft, tend to keep the pieces of magnetic metal in their initial positions. When current passes through the coils of the moving element, the lines of force parallel to the shaft produce a torque which tends to turn the 'ihaft and cause the needle to travel across the scale. This action Fig. 041 to 642.— Multipliers for Weston standard portable voltmeters. Multipliers are re- sistance bo.xes, the coils in which are highly insulated, and are adjusted so that the readings of the instrument may be multiplied by any desired constant. Multipliers are usually constructed so that the indications of the pointer, multiplied by 2, .5, 10, 20 or .50, will give the voltage of the circuit. By the use of multipliers the range of voltmeters may be increased to any practical limit. is, of course, opposed by the magnetic field at right angles to the shaft acting on the two pieces of magnetic metal. There are thus no restraining springs, current being conveyed to the moving coil by tortionless spirals of silver wire. Thompson astatic instruments can be provided with polarity indicators, a red disc showing on the scale card where the ooles are reversed. 554 HAWKINS ELECTRICITY The effect of external fields is eliminated by the astatic arrangement of the fields and the moving parts. A field which tends to increase the torque on one side of the armature dimin- ishes it to a corresponding degree on the other side. The damping effect in these instruments is produced by an aluminum disc moving in a magnetic field. Fig. 643. — Portable multiplier for portable voltmeter. A multiplier is used for increasing the readings of voltmeters, and consists of resis.ance coils placed in a portable case. A mul- tiplier is connected in series vrith the voltmeter and must be adjusted for the instrument with which it is to be used, because the resistance coil must be a multiple of the voltmeter resistance. For instance, a multiplier wich a value of 10, used with a 6 volt voltmeter or 521 ohms would measure about 5.215 ohms; one with a value of 40. would equal about 20,860 ohms. The multiplier 10 would give a total scale value of 60. and the multiplier 40, a total scale value of 240 volts to the 6 volt instrument. A multiplier is of consider- able value in that it does away with the necessity of having a number of voltmeters of different ranges. The instrument here illustrated has a range of 150 volts, Ques. What are multipliers? Ans. These are extra resistance coils which are connected in series wnth a voltmeter for increasing its capacity or readings. They are put up in portable boxes, and must be adjusted for each particular voltmeter as the resistance of a multiplier coil must be a multiple of the resistance of the voltmeter itself. AMMETERS. VOLTMETERS, AND WATTMETERS 555 Ques. What is an electro-dynamometer? Ans. An instrument for measuring amperes, volts, or watts by the reaction between two coils when the current to be measured is passed through them. One of the coils is fixed and the other movable. Ques. Descrfbe the Siemens' electro-dynamometer. Ans. The essential parts are shown in fig. 646. The fixed coil A, composed of a number of turns of wire is fastened to a v^ertical support, and surrounded by the movable coil B of a few Figs. 644 to 645. — Western standard portable shunts. The milli-voltmeters used in connection with these shunts read directly in amperes. Shunts of different capacities can be adjusted to the same instrument, and it can, therefore, be used to measure a current of 2,000 amperes with the same degree of accuracy as a current of 1 ampere. In selecting shunts of different capacities for use in connection with one instrument it should be considered that the higher ranges must be even multiples of the lower one in order to suit the same scale on the instrument. turns, or often of only one turn. The movable coil is suspended by a thread and a spiral spring C, below the dials which are fastened at one end to the movable coil and at the other end to a milled headed screw D, which can be turned so as to place the planes of the coil at right angles to each other, and to apply torsion to the spring to oppose the deflection of the movable coil for this position when a current is passed through the coils. The 556 HAWKINS ELECTRICITY ends of the movable coil dip into two cups of mercury E, E', located one above the other and along the axis of the coils so as to bring the two in series when connected to an external circuit. The arrows show the direction of current through the two coils. An index pointer F is attached to the movable coil. The upper end of this pointer is bent at a right angle, so that it Fig. C40. — Diagram of Siemens' electro-dynamometer. It consists of two coils on a common axis, but set in planes at right angles to each other in such a way that a torque is produced between the two coils which measures the product of their currents. This torque is balanced by twisting a spiral spring through a measured angle of such degree that the coils shall resume their original relative positions. If the instrument be used for measuring current, the coils are connected in series, and the reading is then proportional to the square of the current. If used as a wattmeter, one coil carries the main current and the other a small current, which is proportional to the pressure. The reading is then proportional to the power in the circuit. Fig. 647. — Diagram showing connections of Siemens' electro-dynamometer as arranged to read watts. swings over the dial between two stop pins G, G', and rests directly over the zero line when the planes of the coils are at right angles to each other. A pointer H is attached to the torsion screw D, and sweeps over the scale of the dial. The spring is the controlling factor in making the measurement AMMETERS, VOLTMETERS, AND WATTMETERS 557 Figs. 648 to 650. — Wright demand Indicator. This is a device for registering the maximum ampere demand of appreciable duration in any electrical circuit. It may be used on either direct or alternating current circuits. The essential features and principle are as follows: A liouid is hermetically sealed in a glass vessel consisting of two bulbs connected by a "U"' tube, and a central tube called the "index" tube, connected to the upper end of the right hand side of the "U." Around the left hand or heating bulb, is placed a band of resistance metal, through which the current to be measured is passed, or a definite shunted portion of it. The heating effect of the current increases the temoerature of the left hand bulb, causing the air to expand which forces the liquid up the right hand side of the " U " tube and into the index tube, where it remains until the indicator is reset. The height of the liquid in the index tube as shownby the scale, indicates the maximum current which has passed through the indicator. It is the difference in temperature of the air in the two bulbs which causes the flow ol the liquid. Any change in e.xternal temperature causes equal effect in both bulbs and therefore does not affect the reading. 558 HAWKINS ELECTRICITY Ques. Explain the operation of the Siemen's electro- dynamometer. Ans. In fig. 646, when a current is passed through both coils, the movable coil is deflected against a stop pin, then the screw D is t-urned in a direction to oppose the action of the current until Figs. 651 and ti.')2. — Weston illuminated dial station voltmeter and an-.r:e:er. The voltmeter has two indices, a pointed index for close readings and an index called the normdl indfx . which enables a slight deviation from the normal voltage to be seen from a long distanci The "normal index " is inside the case and terminates in a circular disc of blackened alumi- num. The disc is adjusted from the outside of the case by hand, by means of the knurled knob seen on the front of the case, so that it is directly below the point of normal voltage. When the indicating index reaches the point of normal voltage, the disc of the normal index appears in the center of the circular opcming of the indicating index, a narrow ring of white being visible, encircling the disc of the normal index. The ammeter depends for its operation upon the fall of potential between two points of the circuit carr>-ing the main current, and requires a difference of only about .05 volt to give a full scale deflection. When a maximum deflection is secured, the current passing through the instrument is never more than .07 ampere irrespective of the total capacity of the instrument. A separate shunt if used which is placed at the back of the switchboard. In many cases, a special shunt can be dispensed with and a short section of the mains on the switchboard, or the mains lead- ing from the djTiamo, can be used instead. On the basis of one square inch cross section per 1.000 amperes, a length of about 5 feet of copper conductor would be required as a shunt, in which case however, this section of the conductor must be adjusted with precision. the deflection has been overcome and the coil brought back to its original position. The angle through which the pointer of the torsion screw was turned is directly proportional to the square root of the angle of torsion. To determine the current strength AMMETERS, VOLTMETERS, AND WATTMETERS 559 in amperes, the square root of the angle of torsion is multipHed by a calculated constant furnished by the makers of this in- strument. Oues. How is the electrodynamometer adapted to measure volts or watts? Ans. When constructed as a voltmeter, both coils are wound with a large number of turns of fine wire making the instrument Fig. 653. — Thompson watt hour meter (type C-6). This form is furnished with side con- nections, the line wires entering at the left and the load wires at the right. Both sides of the system are carried through the meter in all sizes up to and including the 50 ampere size. In meters of larger ampere capacities, a voltage tap is used. sensitive to small currents. Then by connecting a high resistance in series with the instrument it can be connected across the ter- minals of a circuit whose voltage is to be measured. When con- structed as a wattmeter, one coil is wound so as to carry the main 500 HAWKIXS ELECTRICITY current, and the other made with many ttims of fine vrire of high resistance sviitable for connecting across the ciraiit. With this arrangement, the force between the two coils will be pro- portional to the product of amperes by volts, hence, the in- strument will meastu-e watts. Pig. 6S4. — Interior view of Thompson watt hoar meter (type C-6). Capacity: 5 to 600 amperes, two wire, and 5 to 300 amperes, three wire; 100 to 250 vtdts. The meter is supported by three logs, the tipper one of which is keyboled. and the lower right hand one slotted. This permits rapid and aocnrate levdling as the top screw can be inserted and the meter hung thereon approximatdy lereL The right hand screw may then be i^aced in positiba and the meter adjusted as may be required before forcing the screw hosne. Oues. Describe briefly the construction of the Thompson recording wattmeter. Ar.s. It consists of four elcnienis: 1, a motor causing rota- tion; 2, a d\Tiamo pro\-iding the necessarj'^ load or drag; 3, a registering de\-ice, the function of which is to integrate the instantaneous values of the electrical energ>' to be measured. and 4, means of regulation for light and full load- AMMETERS, VOLTMETERS AND WATTMETERS ^^[\\ iiiniiiiiiiirCT g;go„,j::o"^oi2oES«J ^Sti^-I-'^.S «io E fe S S « a, ''^ fc^ C^ g,b S « rt «S 5" Si^^^S^ «'-= c^ ajS ;S o M (u "* .|l^-2§|ES§|i-l ^-^g^^u5^3->-E-S43 S gj=i!-? 2 3 "5 Mj; C 3.C2 3 « nj (u M i^--^j^^ «-C^ oj u rt f^ 562 HAWKINS ELECTRICITY Oues. What is the action of the motor in the Thomp- son watt hour meter? Ans. It rotates at very slow speed, and since there is no iron in its fields and armature, it has very little reverse voltage. Its armature current, therefore, is independent of the speed of rotation, and is constant for any definite voltage applied at its termi- '- Fig. 656. — Interior of Thompson watt hojir meter (type C-6) showing armature, small commu- tator and gravity brushes. A spherical armature moving within circular field coils is the construction adopted in this meter. The armature is wound on a very thin paper shell, stiff enough to withstand the strain due to winding and subsequent handling. The wire composing the armature is of the smallest gauge consistent with mechanical strength. The field coils, as before stated, are circular, and are placed as near each other as possible, one on either side of the armature, with the internal diameter just sufficient to give the necessar>' clearance for the rotating element. This construction prevents magnetic leak- age. Ribbon wire is employed for the field coUs, thus economizing space and further carrying out the idea of concentration. The torque of this motor being proportioned to the product of its armature and field currents, must vary directly as the energy passing through its coils. In order then, that the motor shall record correctly it is necessary only to provide some means for making the speed propor- tional to the torque. This is accomplished by applying a load or drag, the strength of which varies directly as the speed- AMMETERS, VOLTMETERS AND WATTMETERS 563 Oues. Explain the operation of the Thompson record- ing wattmeter. Ans. There being no iron in either field or armature of the motor element, no considerations of saturation are involved. The torque or pull of the armature is dependent upon the product of the field and armature strength. The strength of the field — there being no iron — varies directly with the current in the field. Thus the strength of the field with 10 amperes flowing to the load is exactly twice the strength of the field with 5 amperes flowing to the load. The strength of the armature is dependent on the voltage of the system to which it is connected, the armature eleinent of the meter being practically a voltmeter. There is, . therefore, a torque br pull varying directly with the strength of the armature multiplied by the strength of the field, or, in other words, varying directly with the watt load, and except in so far as influenced by friction, the speed of rotation varies directly with the torque or pull. The currents generated in the disc armature consist of eddy currents, which circulate within the mass of the disc. Installation of Wattmeters. — The various types of watt- meter differ so widely either in mechanical details, or operating principles, that it is customary for manufacturers to furnish detailed instructions for the installation of their meters. Such instructions should be carefully followed in all cases, but the following will be found generally applicable to all types of motor meter : 1. After unpacking the meter, and before opening the case or cover, clean the latter carefully to remove all adhering particles of dust and excelsior. 2. The proper location for the meter should be one where there is no vibration. When this location has been selected, nail or screw upon the walls, a board somewhat larger than the dimensions of the back of the meter, and upon this board hang the meter by the top hanger. 564 HAWKINS ELECTRICITY 3. After hanging the meter, open or remove the case or cover, and if necessar}', put the mechanism in order according to instructions furnished by the manufacturer. 4. In order to operate satisfactorily, the meter should hang plumb, so that the spindle of the revolving element will be vertical, and the horizontal planes through the armature and retarding disc will be level. Many complaints relative to meters being slow on light loads, are invariably due to the fact that the meters have been installed out of plumb.* Fig. 657. — Interior view of Thompson watt hour meter (t>T)e CQ). The capacities of this type raijge from 60 to 400 amperes inclusive, two wire, and 50 to 200 ami>eres inclusive, three wire, and for voltages of from 1 W) to 600 volts. These meters are miade with either front or back connections. In front connected meters the positions of the leading-in wires and cables are the same as in the type C-6, fig. 654, so that either type of meter may be installed in the same location. •XOTE. — The most practical and accurate method of plumbing a meter is to level it by meansof a small brass weight placed upon the retarding disc. Place the weight upon the front or back upper surface of the disc, close to the edge. If the disc and weight rotate toward the right, move the bottom of the meter in the same direction so as to raise the disc on the right. When the disc is level, the weight and disc will remain stationary when the weight is placed on either the front or the back of the disc. Xe.^t . place the weight on the disc close to the edge on either side. If the disc rotate towards the front, swing the bottom of the meter away from the wall or board until the disc remains stationary' when the weight is placed upon it on either side. If the disc rotate toward the back, raise it up on that side by bringing the top of the meter away from the wall or board. It is p>ossible that the second levelling operation will alter the position of the disc obtained by the first operation, therefore, the first should be repeated, and after that the second also, until the disc remains stationary when the weight is placed at any point upon its surface. This method of levelling is more reliable than any method in which a spirit level is employed. AMMETERS, VOLTMETERS, AND WATTMETERS 565 taxi r^^ c_ a"<-'"-= >>H^ M c o °£^-^ _-d cnB bo=« "S'^lc'^ji rt "1 4>^ >^ O o - .3 •doi.o'i3ou"rt Q.T3 •;: O S^.-S o-C 5 > «> _o T3 fto) c 0! -^ H ""-^ 2J o S aEi^ c^^icc'S fi 2^ o >- »« c-Sp 5 fe.g -^ 5 tD^ o-° >a a 5 u 1, rt o^ o rt oj:: c "■ 2 2^ 4, o u j; ^-d^ — §• o njr-'_iJ ^ m .*" 5j 0.2 .. O ■>-> o. „a..<2^oE^g c-r o C-G o " ,n m ^-ti g o£j=J5x 5 c p S H 56(i HAWKINS ELECTRICITY UPPER BEARING, SLEEVE \ n DIAL- LIGHT LOAD ADJUSTING COIL DAMPING DISK FRONT RETARDING MAGNETS SERIES ^.-^COILS TERMINALS CLAMPING SCREW B TERMINAL REAR RETARDING MAGNETS JEWEL SCREW Fig. f)59. — Westinghouse type CW-6 watt hour meter with cover off. This meter is of the commutator type without iron in the magnetic circuit. The spherical armature is closely surrounded by circular field coils which provide the shortest magnetic path and smallest magnetic leakage, thus securing high torque with small consumption of energ>'. The armature ivinding is wound on a hollow sphere of prepared paper which is moulded in corrugated form to secure strength. Uniform brush tension is m.aintained by gravity. Each brush consists of two small round wires placed side by side and held against the com- mutator by a small counterweight whose distance from the fulcrum is adjustable. The current winding consists of two flat coils of strap copper, one clamped rigidly on either side of the central mounting frame which supports the armature bearings. These coils are con- nected either in series or parallel, depending on the capacity. In three wire meters one of the coils is connected in series with each side of the line. The retarding element consists of a light aluminum disc rotating between two pairs of permanent magnets. The magnets are prepared by a special aging process to insure permanence. Full load adjustment is made by shifting the position of the permanent magnets. Ample light load adjustment or friction compensation is provided by means of the movable coil, which can be shifted hori- zontally or radially on loosening one screw. The meter registers directly iq kUgwatt houi* AMMETERS, VOLTMETERS, AND WATTMETERS 567 5. In making the circuit connections, be very careful that the positive lead or wire is placed in the positive binding post of the meter. This precaution is essential for insuring an accurate and sensitive measure- ment on small loads. 6 When a meter of the commutated motor type sparks at the brushes at starting, it is an indication that the commutator is dusty. Clean it with a piece of closely woven cotton tape ^-inch in width. Pig. 660. — Interior of Thompson pre- payment watt hour meter. The actu- ating force is a large flat coil spring enclosed in a barrel or drum to which its outside end is attached. The operating knob winds this main spring by turning the drum. The spring has many turns and as the operation of the device never equals one whole turn, the spring always exerts a practically constant force. The rate device consists of a small train of gears secured to the front of the frame directly back of the register. Each device is marked with the price per kw-hr. for which it should be used. The switch is of the double pole double break type with leaf contacts. The coin receptacles are placed at the back of the meter. To make an advance payment, the wind- ing knob is turned so that the arrow points upward. A quarter dollar is then inserted in the slot and the knob turned to the right, the coin serving as a key which operates the mechan- ism within the device, turning the registering wheel and placing the coin to the credit of the customer. If the circuit be open when the coin is deposited the same motion of the knob which moves the registering mechanism closes the circuit switch contained within the case. The dial contains a scale marked in plain figures over which a pointer passes indicating the number of coins remaining to the credit of the depositor. When the first coin is deposited and the knob turned closing the main switch, the pointer rests opposite the first division on the scale. If a second coin be de- posited before the current purchased with the first coin has been consumed, a second motion of the knob will bring the pointer opposite the second division on the scale. Twelve coins can thus be deposited consecutively, after which the slot is automatically closed and further prepayment cannot be made until the value of one or more coins has been consumed. Whenever energy to the value of one coin has been delivered through the meter, the escapement is mechanically released turning the pointer back one division. This process continues until all the energy has been delivered for which payment has been made. Thus the depositor can ascertain at any time how much energy can be obtained without further payment. When all energy has been delivered, the circuit switch is opened so that no more current can be obtained until one or more coins have been deposited. The indicating mechanism shows only the number of coins which stand to the credit of the customer, but, by consulting the meter dial, one can determine what fractional part of the prepayment next to be cancelled remains to the credit of the customei. A coin or washer larger than the coin for which the device is designed cannot be introduced into the receiving slot and a smaller one will not operate the device. 568 HAWKINS ELECTRICITY Meters should never be allowed to remain with their covers off, in the testing room, station, or any other place. In order to get the best service, and to give them long life they must be kept clean. How to test a meter. — A simple test for ascertaining whether a customer's meter is fast or slow, may be applied as follows: 1. Turn off the lamps and other power consuming devices in the house and then note the reading of the meter dial and the exact time of day; 2. Turn on as quickly as possible about one-tenth of all the lamps in the house and allow them to burn for about two hours; 3. At the end of two hours, turn oflf the lamps as quickly as possible and note the reading of the meter dial. The difference between the first and second readings of the dial will be the indicated consumption of two hours, and if this be greater than the amount of power that ought to be consumed by the number of lamps turned on, the meter is fast, but if it be less, the meter is slow. The best results obtained by this method are only approximations, however, on account of the variations in the watts consumed by the different makes of lamp, the uncertainty as to the actual voltage on the line at the time of the test, and the lack of knowledge as to the age of the lamps. Therefore, if the meter test within five per cent., or do not record more nor less than one-twentieth of the assumed lamp con- sumption it is safe to assume that the meter is correct as the result of the test is not likely to be any closer to the truth. ; ^ Oues. How should a roughened commutator be cleaned and smoothed ? Ans. By means of tape. NOTE. — A meter operates under more varied and exactinR conditions than almost any other piece of apparatus. It is frequently subjected to ^^bration. moisture and extremes of temperature; it must register accurately on varying voltages and various wave forms; it must operate for many months without any supervision or attention whatever; and, in spite of all these conditions, it is expected to register with accuracy from a few per cent, of its rated capacity to a 50 per cent, overload. As a meter is a type of machine, its natural tendency is to run slow; but occasionally, through accident, a meter may run fast. When a meter runs fast the con- sumer is paying a higher rate per kilowatt hcur than his contract calls for. He is being dis- criminated against. The periodic testing of meters is therefore a necessity and is an indication of the honesty of intention of the manager toward the customers of his company. Meters controlling a very large amount of revenue may be tested as often as once a month, while the ordinary run of meters should be tested at least once a year, once in eighteen months, or once in two yeais. the period varying with different companies, different types and different civic requirements. Commutator t>-pe meters. ha\'ing comparatively hea\-y moN-ing elements with consequent rapid increase in friction due to wear on the jewel and bearings, and a commutator also increasing in friction \*ith age, must have frequent and expert attention to insure their accuracy under all conditions. AMMETERS, VOLTMETERS, AND WATTMETERS 569 . 661. — Internal connections of Sangamo watt hour meter (type D). A, copper disc armature, submerged in mercury; B, bridge wire between binding posts, for main load current, when both sides of the line are carried through the meter; CT, compounding series turns around pressure circuit magnet, building up field as load increases, to compen- sate for falling off in speed otherwise found; D, aluminum damping, or brake disc, con- trolling speed of meter; E, copper contact ears, imbedded in insulating wall of mercury chamber, leading current into and out from armature; F, hardwood float on armature proportioned to give slight "lift" to entire moving system, when armature and float are immersed in mercury; H, soft steel disc above permanent magnets, riveted to fine pitch screw working in bracket above, so that adjustment of the disc up or down gives variation in damping effect of permanent magnets, and therefore of main speed. K, clamp slider with thumb screw, for obtaining light load adjustment by mo\'ing K to right or left, as may be necessary. K spans and connects parallel wires of light load adjustment, BR and RR'. MM, powerful permanent magnets, acting on disc D, giving main speed con- trol for meter. N, high resistance heav^ wire, forming part of series adjustment between armature and any shunt with which meter may be used, to set drop through meter correct for drop of the shunt. P, spirally laminated soft steel ring, moulded in mercury chamber above the armature space, to act as a return for magnetic lines of force from and to ener- gizing magnet below. R, resistance card unit, in series with pressure circuit coils; in 110 volt meters one card is used, in 220 volt meters two cards, or one card and a thermo- couple. BR, small brass wire, connected to ingoing end of pressure circuit coils and form- ing RR'and the slides K the light load adjustment. RR', high resistance wire having opposite ends connected to ears EE by low resistance wires. Current energizing the pressure circuit coils SC passes from RR' through K to BR and thence to the coils, and if K be near the end of RR' and BR, the least compensation is obtained; if near right end, maximum light load compensation is obtained. S, shaft or -spindle. In actual meter S is divided, the lower shaft carrying armature A, and the upper shaft damping disc D. SA, series resistance adjustment, for setting meter to correct drop for shunt. SC and SC, pressure coils connected in serie«. TT, binding posts at bottom of meter. Y, lami- nated soft steel yoke, carrying coils SC and SC, and giving a powerful and concentrated magnetic field on the armature. W, worm, driving recording dial train. WW, worm wheel. 570 HAWKINS ELECTRICITY Waste of Electricity in Lighting. — In large residences where a good many servants are employed or in any plaee where the power consumed is not directly under the supervision of the per- son who must pay the bills, a great deal of waste usually occurs. CYLINDRICM (BRUSHES. COMMUTATOR DIAMETER ONE-TENTH, INGHx ARMATURE COSISTS OF THREE INTER LOCKED COILS. DlREa READING ASSOCIATION :;„ TRAIN aV DRIVEN MICROMETER SCREW ADJUSTER "■" REGULATING BRUSH TENSION. BRUSH CLAMP ENABLING SWINGING BRUSHES CLEAR OF COMMUTATOR AND REENGAGING WITH- OUT ALTERING TENSION. RIGID ONE PIECE CAST ^ ALUMINUM BASE WITH i STIFFENER RIBS- SHORTSTIFF SHAFT WITH PIVOTS FRiaiON CLAMP ATTACHED LIGHT LOAD ADJUST- MENT OF WIDE RANGE. DAMPER MAGNETS -IMMOVABLY CLAMPED IN PLACE. MICROMETER MAGNETIC SHUNT ADJUSTMENT FOR CONTR0LUN6 MAIN SPEED ALUMINUM JVNEL BEARING. Pig. 662. — Interior view of Columbia watt hour meter (type T)), showing construction and principal parts and connections. The armature winding consists of three coils appro.^i- mately circular in shape. The coils are for.m wound, interlocked with one another and with the light impregnated fibre disc which serves as a spacer for them. The aluminum damper disc has the conventional anti-creep provision in the shape of the three small soft iron plugs, mounted close to the central staff, which the illustration shows. These in their revolution come successively within the influence of an adjustable iron screw which is magnetized by an extension from one of the damper magnets. The angular relationship of the armature windings and of the three iron plutrs is such that at the time that the armature is exerting a maximum torque the magnetized screw is exerting the maximum pull to hold back a given plug and conversely when the armature pull is a minimum the magnetic screw is attracting a plug with the maximum effort to cause ahead rotation. The irregularities of torque are in this way smoothed out. The commutator has three segments and is made of chemically pure silver. Each brush is formed of a length of phosphor bronze wire bent like a hair pin and secured at its "U" end to a brass sleeve, which in turn is secured to an insulated stud by a set screw. An extension on the sleeve carries a micrometer screw brush adjustment. The main speed adjustment is secured by pro\Hding a soft iron bridge plate, bridging over the extremities of each magnet end and adjustable, by means of a set screw and lock nut. to any desired distance therefrom. This gives a regular micrometer means of varying the effective magnet strength. Interposed between the series coil and the permanent magnets is a heavy soft iron shield to guard the magnets against disturbance by short circuiting. Light load adjustment is obtained by pro\nding in the coil circuit a series of small resistance spools, equipped with pin terminals, to which connection can be selectively made by means of a split bushing terminal on a flgadble cord. This series of spools is stniog on a metal artx>r located withiQ uie case. AMMETERS, VOLTMETERS, AND WATTMETERS 571 TTRKUNALS TERMINALS C0MPEW5MIN& COIL SWITCH Fig. 663. — Diagram showing internal connections of the Duncan watt hour meter. Its opera- tion depends upon the principle of the well known electro-dynamometer, in which the electromagnetic action between the currents in the field coils and an armature pro- duces motion in the latter. It also embodies the other two necessary watt hour meter elements required for the speed control and registration of the revolutions of the armature, these being embodied in the drag magnet and disc, and the meter register respectively, The motion of the armature is converted into continuous rotation by the aid of a com» mutator and brushes, the commutator being connected to the armature coils and carried on the same spindle therewith. 572 HAWKINS ELECTRICITY If the meter be read before retiring, the reading in the morning will show how much energy was consumed during the night, which will show in turn how many lamps were burning all night. A great deal of light can be saved by placing the lamps so that they will throw the light where it is needed and by placing small lamps such as 8 candle power and 4 candle power in places where not much light is needed, such as bathrooms, halls, cellars, etc. When the lamps get old and dim they should be replaced with new ones, as it costs about the same to bum an old lamp as a new one. An old 16 candle power lamp which is very dim will give only about 8 candle power and use about as much current as is required for a new 16 candle power. If the dim light be light enough, it should be replaced by an 8 candle power lamp, which will not consume as much power as the old 16 candle power. OPERA TION OF D YNA MOS 573 CHAPTER XXIX OPERATION OF DYNAMOS Before Starting a Dynamo or Motor. — When the machine has been securely fixed, it should be carefully examined to see that all parts are in good order. The examination should be made as follows : 1. The field magnet circuit should first be inspected to see that none of the wires or connections have broken or have become loose, and that the coils are correctly connected; 2. The caps of the bearings should be taken off, and these and the journals carefully cleaned of all grit and dirt. They should then be oiled, and the caps replaced and screwed up by hand only; 3. The gaps between the outer surface of the armature and the polar faces should be examined in order to ascertain whether any foreign body, such as a small screw or nail has lodged therein. If such be the case, it should be carefully removed with a bit of wire; 4. The guard plates protecting the armature windings should be removed, and the windings carefully inspected by slowly rotating the armature, to see that they are not damaged, and that the insulation is perfect. The armature should then be finally rotated by hand to see that it revolves freely, and that the bearings are securely fixed; 574 HAWKINS ELECTRICITY 5. The commutator should be examined to see that it is not damaged in any way through one or more of the segments being knocked in, or the lugs being forced into contact with one another ; 6. The brush holders and brushes should be inspected to see that the former work freely on the spindle, and that the hold off catches work properly, are clean and make good contact with the brush holders or flexible leads ; 7. Ha^•ing ascertained that the machine is not injxired in any way, and that the armattu-e revolves freely, the brushes should be adjusted. In the subsequent working of the d>Tiamo it •will of course be un- necessan,- to follow the whole of these proceedings ever\' time the machine is started, as it is extremely unlikeh' that the machine will be damaged from external causes while working without the attendant being aware of the fact. Adjusting the Brushes. — The adjustment of the brushes upon the commutator requires careful attention if sparking is to be avoided. There are two adjustments to be made: 1. For pressure; The brushes naust bear against the commutator segments with sufficient pressure for proper contact. 2. For lead. The brushes must have the proper angular advance (positive or n^ative, according as the machine is a d\-namo or motor) to prevent sparking. Oues. At what point on the commutator should the brushes bear? Ans. The points upon the commutator at which the tips of the brushes (carried by opposite arms of the rocker) bear, should OPERATION OF DYNAMOS 575 be, in bipolar dynamos, at opposite extremities of a diameter. In multipolar dynamos the positions vary with the number of poles and the nature of the armature winding. Ones. What provision is made to facilitate the correct setting of the brushes? Ans. Setting marks are usually cut in the collar of the commutator next to the bearing. Figs. 664 and 665. — Diagrams illustrating how to set brushes. Some brush holders require brushes set with the direction of rotation of the commutator, and others, set against the direction of rotation. In fig. 0G4 is shown a brush holder of the first class, which must always be set as indicated by the arrow. If set in the opposite direction, trouble will ensue, as an inspection of the figure will show, because the surface of the commutator and the brush would form a toggle joint, and the brush would tend to dig into the commutator and either break itself or bend the brush rigging. In fig. 6G5 is shown a brush holder of the second type. This brush is set against the direction of rotation, but an inspection of the cut will show that there is, in this case, no tendency for the brush to dig into the com- mutator surface. Each type of brush holder, of which there are several, should be adjusted as recommended by the manufacturer to secure proper working. Ques. How are the brushes set by these marks? Ans. The tips of all the brushes carried by one arm of the rocker are set in correct line with the commutator segments marked out by one setting mark, and the tips of the brushes carried by the other arm or arms are set in correct line with the segments marked out by the other mark or marks. If one or more of the brushes in a set be out of line with their setting mark, it will be necessary to adjust the brushes up to this mark by push- ing them out or drawing them back, as may be required, afterwards 67G II A WKINS ELECTRICITY clamping them in position. When adjusting the brushes, the armature should always be rotated, so that the setting marks are horizontal. The rocker can then be rotated into position, and the tips of both sets of brushes conveniently adjusted to their marks. In those brush holders provided with an index or pointer for adjusting the brushes, the setting marks upon the commutator are absent, length of the pointer being so proportioned that when the tips of the brushes are in line with the ex- treme tips of the pointers, the brushes bear upon the correct positions on the commutator. Fig. 666. — Method of soldering cable to carbon brush. Drill a hole in the end, also in the side of the brush, as shown in the sketch, and after thoroughly tinning the "pig-tail," place it in the end hole and fill the holes up with solder through the side hole. Another method is to drill a hole through the carbon so that the cable will just slip through, countersink the edge of the hole a little, clean the cable thoroughly and pass it through the hole. Then with any good flu.x and solder, fill the countersunk part on both sides. Oues. What should be done after adjusting the brushes to their correct positions upon the commutator ? Ans. Their tips or rubbing ends should be examined while in position to see that they bed accurately on the surface of the commutator. In many instances it will be found that this is not the case, the brushes sometimes bearing upon the i)oint or toe, and sometimes upon the heel, so that they do not make contact with the commutator through- out their entire thickness and width. The angle of the rubbing ends will therefore need to be altered by filing to make them lie flat. OPERA TION OF D YNA MOS bll Ques. How is the proper brush contact secured ? Ans. When the brushes do not bed properly they should be refitted to secure proper contact. Ques. How is the pressure adjustment made? Ans. This is effected by regulating the tension of the springs provided for the purpose upon the brush holders. Ques. With what pressure should the brushes bear against the commutator? Ans. The tension of the springs should be just suiKcient to cause the brushes to make a light yet reliable contact with the commutator. The contact must not be too light, otherwise the brushes will vibrate, and thus cause sparking; nor must it be too heavy, or they will press too hard upon the commutator, grinding, scoring and wearing away the latter and themselves to an undesirable extent, and moreover, giving rise to heating and sparking. The correct pressure is attained when the brushes collect the full current without sparking, v/hile their pressure upon the commutator is just sufficient to overcome ordinary vibration due to the rotation of the commutator. Direction of Rotation. — This is sometimes a matter of doubt and often results in considerable trouble. As a general rule, a dynamo is intended to run in a certain direction; either right handed or left handed according to whether the armature, when looked at from the pulley end, revolves with or against the direction of the hands of a clock. Dynamos are usually designed to run right handed, but the manufacturers will make them left handed if so desired. It may be necessary to reverse the direction of rotation of a ■ dynamo, if the driving pulley to which it has to be connected happen to revolve left handed, or if it be necessary to bring the loose side of the belt on top of the pulley, or to place the machine >78 HAWKIXS ELECTRICITY Figs. 667 to 669. — Method of winding cables with marlin. When connecting the feeders and djTiamo and se^^^ce leads to a switchboard, the wires are often served with marlin. By serving is meant to tightly wrap the wires of each set together with marlin. A tool for ser\-ing may be made as in fig. 667, using a piece of oak 2 ins. wide, J g in. thick and 14 ins. long, ha%-ing four holes drilled through it, as shown. The marlin is passed through the holes, commencing at the hole nearest the handle, the object being to cause a strain on the marlin at the point where it passes around the wire, so that the marUn may be wrapped tightly. It is necessary to serve the first four or five inches by hand, pushing the winding into the conduit as far as possible. This acts as an additional protection to the wres where they leave the conduit. The serving is contir.ued, as in fig. 668, to within four or five inches of the first lug by means of the serv-ir.g tool, passing the ball of marlin around the wires with the serving tool. The wires are then bent in shape, as in fig. 66;t. To serve the wires properly it is necessar>' to tie the ends of the wires taut. The wires should be straightened and run together so as to be parallel, being bound with tape at different points to keep them so. When the serving is complete the marlin should be thoroughly painted with a moisture resisting compound. The marlin ser\-ing will stiffen the wires and they can be bent very neatly to avoid touching the bus bars of the board. When painted the marlin hardens so that it is difficult to bend the wires after the paint has dried. It then requires a strong pressure to bend them. The marlin acts as an additional insula- tion and mechanical protection to the wires, and while no harm would result from the wires coming in contact with the bars while thus protected, it looks better to bend them so as to avoid touching the bars. />Hpf»^T"y ^ l.li.COllEGfl^^»^V OPERA TION OF D YNA MOS 579 in a certain position on account of limited space. The direction of rotation of ordinary series, shunt, or compound bipolar dynamos may be reversed by simply reversing the brushes without changing any of the connections, then changing the point of contact of the brush tips 180°. In multipolar dynamos, a similar change, amounting to 90° for a four pole machine, and 45° for an eight pole machine, will reverse their direction of rotation. It will be understood that under these conditions, the original direction of the current and the polarity of the field magnets will remain unchanged. Fig. 670. — Method of assembling core discs. For this operation two wooden "horses " 'should be provided to support the core at a convenient height, as shown in the illustration. This rule does not apply to arc dynamos and other machines, which have to be run in a certain direction only, in order to suit their regulating devices. If the direction of current generated by a dynamo be opposite to that desired, the two leads should be reversed in the ter- minals, or the residual magnetism should be reversed by a current from an outside sovu-ce^ JNn^^y^^y. «^ *'^if Cr^fr CAT »,«•*.«.. 580 HAWKINS ELECTRICITY Starting a Dynamo. — Ha\-ing followed the foregoing in- structions, all keys, spanners, bolts, etc., should be removed from the immediate neighborhood of the machine, and the d\Tiamo started. Figs. 671 and 672. — Tinning block for electric soldering tooL It is made with two soft bricks. One brick is used to support the soldering tool, and the other to contain the tinning material and to furnish a material which will keep the copper bit bright enough to receive its coating of "tin." Fig. 671 represents a section of the tinning brick, which is scooped out on top as shown by the lower line. Into one end of the hollow in the brick, some sal-ammoniac is placed to help tin the copper bit. Sal-ammoniac is a natural flux for copper and aids greatly in keeping the tool well tinned. Next, some melted solder is run into the hollow of the brick, and lastly enough resin to fill the ca\-ity nearly to the top. When the tool is not in use, the electricity is switched off and the tool permitted to lie in the resin. If it be desired to repair the tin coating a little when the tool is in use, the latter is rubbed on the brick below the layer of solder, and the layer of resin. If the tool be in ver>' bad condition, it may be pudied into the sal-ammoniac once or twice and then rubbed in the solder again. It requires but little heat to keep the brick and its contents ready for use. In fact, the brick is a fair non-conductor of heat and prevents the escap>e of heat from one side of the tool. When momentarily not in use, the tool remains in the solder which becomes melted underneath the layer of resin. When the copper bit becomes too hot, it will begin to volatilize the resin, thus f-alling attention to this fact, ■mbereapoa, the electhdty should be tumed oS from the tooL OPERA TION OF D YNA MOS 581 Oues. How should a dynamo be started? Ans. A dynamo is usually brought up to speed either by starting the driving engine, or by connecting the dynamo to a source of power already in motion. In the first case, it should be done by a competent engineer, and in the second case by a person experienced in putting on friction clutches to revolving shafts, or in slipping on belting to moving pulleys. l-'iG. 673. — Connections for two shunt wound dynamos to run in parallel. The positive lead of each machine is connected to the same bus bar. In starting, if but one machine is to be used, the dynamo is first brought up to speed and the voltage regulated by means of the rheostat R and the voltmeter V. The main switch is then thrown in. The connections for the field are taken off the dynamo leads so that the opening of the main switch will not open the field circuit and for this reason the field will begjin to build up as soon as the machine is started. When but one of the machines is running, the idle machine is brought up to speed with the main switch open, and the voltage regulated by means of the rheostat and voltmeter until the voltages of the machines are the same. Then the main switch is thrown in and the load on the machines (which is ascertained by the ammeters) is equalized by means of the rheostats. Should there be any great difference in voltages, the higher one will run the other as a motor without changing the direction of rotation. The field current will remain unchanged, and the armature current of the low dynamo will be reversed, which will cause it to run as a motor in the same direction as it ran as a dynamo. When dynamos feeding current to motors are to be shut down, the switches on the motors should first be opened. Otherwise some of the motor fuses will blow. As the voltage goes down the motors will draw more current to do the work. If a plant be shut down with the motor switches "in" it will generally be found impossible to start a shunt dynamo, the low resistance in the mains not allowing enough current to flow around the shunt fields to energize them. 582 HAWKINS ELECTRICITY Ques. Should the brushes be raised out of contact in starting? Ans. The brushes should not be in contact in starting if there be any danger of reverse rotation, as might happen when the dynamo is driven by a gas engine. Aside from this, it is desirable that the brushes be in contact, because they are more NEGATIVE BUS SMUMT DYNAMO SHUMT DVKAMO Fig. 674. — Connections for two shunt dynamos to run on the three «^Tre sj-stem. The two machines are connected in series, three wires being carried from them, one from the outside pole of each machine and one from the .iunciion of the two machines. The voltage between the outside wires is equal to the combined voltage of the two machines and the voltage between the outside and the central or neutral wire is equal to the voltage of the corresponding machine. If the load on each side of the system be equal, there will be no current in the neutral wire, while if the loads be unequal, the neutral wire will have to carrv' the difference in current between the two outside wires. easily and better adjusted, and the voltage will come up slowly, so that any fault or difficulty will develop gradually and can be corrected, or the machine stopped before any injury is done. Ques. How should a series machine be started ? Ans. The external circuit should be closed, otherwise a OPERA TION OF D YNA MOS 583 closed circuit will not be formed through the field magnet wind- ing and the machine will not build up. Oues. What is understood by the term " build up" ? Ans. In starting, the gradual voltage increase to maximum. A ) ( V ) (T Fig. 675. — Connections for two compound wound dynamos to run in parallel. The series fields of the machines are connected together in parallel by means of wire leads or bus bars, which connect together the brushes from which the series fields are taken. This is known as the equalizer and is shown by the line running to the middle pole of the dynamo switch. By tracing out the series circuits it will be seen that current from the upper brush of either dynamo has two connections to its bus bar. One of these leads through its own field, and the other, by means of the equalizer bar, through the fields of the other dynamo. As long as both machines are generating equally there is no difference of pressure between the brushes of either, but should the voltage of one be lowered, current from the other would flow through its fields and thereby raise the voltage, and at the same time reduce its own until both are equal. The equalizer may then be called upon to carry much current, but to have the machines regulate closely it should be of low resistance. It may also be run as shown by the dotted lines, but this will leave all the machines alive when any one is generating. The ammeters should be connected as shown. If they were on the other side they would come under the influence of the equalizing current and would indicate wrong, either too high or too low. The equalizer switch should be closed a little before the main switches are closed. Oues. How should a shunt or compound machine be started ? Ans. All switches controlling the external circuits should be opened, as the machine excites best when this is the case. If 584 HAWKINS ELECTRICITY the machine be pro\-ided with a rheostat or hand regulator and resistance coils, these latter shotdd all be cut out of circuit, or short circuited, until the machine excites, when they can be gradually cut in as the voltage rises. When the machine is giving the correct voltage, as indicated by the voltmeter or pilot lamp, the machine may be switched into connection with the external or working circuits. Oues. In starting a shunt dynamo, should the main line switch be closed before the machine is up to voltage or after? Ans. If the machine be working on the same circuit with other machines, or with a storage batter\', it is, or course, neces- sary to make the voltage of the machine equal to that on the line before connecting it in the circuit. If the machine work alone, the switch may be closed either before or after the voltage comes up. The load ^\*ill be thrown on suddenly if the s\^'itch be closed after the machine has built up its voltage, thus causing a strain on the belt, and possibly dra-^-ing water over the engine cylinder. On the other hand, if the switch be closed before the voltage of the machine has come up, the load is picked up grad- ually, but the machine may be slow or may even refuse to pick up at all. Oues. Why does a shunt machine pick up more slowly if the main switch be closed first? Ans. Because the resistance of the main line is so much less than that of the field that the small initial voltage due to the residual magnetism causes a much larger current in the armature than in the shunt field. If this be too large, the cross and back magnetizing force of the armature weakens the field more than the initial field cturent strengthens it, and so the machine cannot bxiild up. OPERATION OF DYNAMOS 585 Ones. If a shunt dynamo will not pick up, what is likely to be the trouble? Ans. The speed may be too slow; the resistance of the external circuit may be too small; the brushes may not be in proper position ; some of the electrical connections in the dynamo may be loose, broken or improperly made; the field may have lost its residual magnetism. © FTaIIx n Pigs. fi76 and 677. — Diagrams of ground detectors. Fig. 676, a ground detector switch suitable for mounting on a switch board. The two arms pivoted at their upper ends are connected with an insulating bar A and make contact at their lower ends with two brass strips and a contact button, which are connected to the bus bars and ground, respectively. When the arms are moved to the left, the positive bus bar is connected to the ground through the voltmeter V. In fig. 677 is another form of ground detector. This is known as a lamp ground detector. On a 110 volt system two ordmary lamps are connected in series, while the line connecting the lamps is connected to the ground through a snap switch vS. When current is on, the two lamps will bum with equal brilliancy, but at a lower candle power. When the switch S is closed, if the two lines be clear, the brilliancy of the lamps will not be affected, but if there be a ground on the positive side, one lamp will bum brighter, the brightness depending on the resistance of the ground. If there be a dead ground, the lamp will bum to its full candle power. Oues. What is the indication that the connections between the field coils and armature are reversed ? Ans. If the machine build up when brought to full speed, the connections are correct, but if it fail to build up, the field coils may be improperly connected. 586 HAWKINS ELECTRICITY POSITIVE BU5-BA.R -♦- FIELD RHEOSTM ARMATURE Pig. 678. — Method of correcting reversed polarity in large shunt d>Tiaiiio by transposing the shunt field leads, and then starting up the machine. As soon as the voltmeter registers any voltage, the dynamo may be stopped and the field leads restored to their original position, when it will be found that the residual magnetism in the pole pieces will usually bring the dynamo up to its jwlarity and proper voltage. This method has the disadvan- tages, of the uncertainty as to the machine building up, and that a temporar>' wire must probably be run from the switchboard to one terminal of the field circuit, which is usually connected to a terminal back of the dynamo frame, so that the flow of current through the field coils may be reversed. With d>-namos ha\-ing laminated field magnet cores of comparatively low residual magnetism, this method may suflfice, but in the case of solid field magnetic cores it is not practical. A better method is to disconnect the shunt field leads and temjjorarily extend them to some other source of direct current. If the current be of higher voltage than the coils are designed for, as for instance 1 10 volt d>T!amo and available current 500 volt, caution must be exercised and a suitable resist- ance be pro\-ided to protect the coils. A 500 volt coil, however, may be supplied from 110 volt circuit, pro%"iding the field winding to be energized is equipp>ed with a cut off switch ha\-ing a discharge resistance, so that it may be used to close and break the circuit when the temporary leads have been connected. If the field windings be not so pro\'ided, a bank of lamps or some other non-inductive resistance must be connected across the leads between the field magnet coils and the point at which the circuit is to be op>ened and closed. This is to pro%-ide a path for the discharge of the induced electromotive force. The circiiit should not remain closed more than a few seconds if the full voltage can be applied. It is well, however, to leave the current on long enough to run the machine up to about half speed and make sure, by means of a voltmeter, that the i)olarity has been corrected. ^^Tien this has been ascertained the dynamo should be stopped and the field winding leads returned to their propver terminals. Then the voltage will be brought up in the right direction, provided the work has been done correctly. OPERA TION OF D YNA MOS 587 This can be tested by connecting a voltmeter across the terminals of the armature, or by means of a magnetic needle placed at a short dis- tance from one of the pole pieces in such a position that it does not point to the north pole. If the field coils be improperly connected, the current due to the initial voltage will weaken the field magnetism and thus pre- vent the machine building up, and when the field circuit is closed the voltmeter reading will be reduced, or the magnetic needle will be less strongly attracted. Oues. What will be the result if the connections of some of the field coils of a dynamo be reversed ? Ans. If one-half the number of coils oppose the other half, the field magnetism will be neutralized and the machine will not build np at all; but if one of the coils be opposed to the others, the machine might build up, but the generated voltage will be low, and there will be considerable sparking at some of the brushes. Oues. How may it be ascertained which coil is reversed ? Ans. In all dynamos there should be an equal number of positive and negative poles, and in almost all of them the poles should be alternately positive and negative. Therefore, if a pocket compass be brought near the pole pieces, and it show that there are more poles of one kind than the other, the indication is that one or more of the coils are reversed, and the improper sequence of alternation will determine which one is wrongly connected. Oues. When a dynamo loses its residual magnetism, how can it be made to build up? Ans. By temporarily magnetizing the field. To do this a current is passed through it from another dynamo, or from the cells of a small primary battery. Usually, this will set up sufficient initial magnetism to allow the machine to build up. The battery circuit should be broken before the machine has built up to full voltage. 588 HA WKIXS ELECTRICITY Oues. What should be done if a dynamo become reversed by a reversal of its field magnetism due to light- ning, short circuit, or othen^ise? Ans. The residual magnetism should be reversed by a current from another dynamo, or from a batter}-; but if this be not convenient, the connections between the machine and the line should be crossed so that the original positive terminal of the d\Tiamo \^-ill be connected to the negative terminal of the line, and \4ce versa. POSmvE BUS-BAR KE6KnVE BUS-BVP Pig. 679. — Method of ccrrec*.:ng reversed pclarity in campoond woaad d>Tiamo. Tbe polarity may be reversed vriihcui disconnecting or rlianging the wire. The figure shows two compound d>'namos, and essentia] connections. Tbe current from any machine cofmected to the equalizer bar by its equalizer switch will di^•ide. a portioo i.Tiamo will then be ready to cnt in with the other machines as soon as the voltage has been taxwgfat up to the pcDper point, or it may be shut down until required. OPERA TION OF D YNA MOS 589 Ques. Can a dynamo be reversed by reversing the connections between the field coils and the armature? Ans. No, for if these connections be reversed, the machine will not build up. Ques. Will a dynamo build up if it become reversed? Ans. Yes. Ques. Then what is the objection to a reversed dynamo? Ans. Since the direction of current of a reversed dynamo is also reversed, serious trouble may occur if it be attempted to connect it in parallel, with other machines not reversed. Attention while Running. — When a dynamo is started and at work, it will need a certain amount of attention to keep it running in a satisfactory and efficient manner. The first point to be considered is the adjustment of the brushes. If this be neglected, the machine will probably spark badly, and the commutator and brushes will frequently require refitting to secure good contact. Ques. What may be said with respect to the lead of the brushes? Ans. The lead in all good dynamos is very small, and varies with the load and class of machine. The best lead to give to the brushes can in all cases be found by rotating the rocker and brushes in either direction to the right or left of the neutral plane until sparking commences, increasing with the movement. The position midway between these two points is the correct position for the brushes, for at this position the least sparking occurs, and it is at this position that the brushes should be fixed by clamping the rocker. 590 HAWKINS ELECTRICITY Ques. How does the lead vary in the different types of dynamo? Ans. In series dynamos giving a constant current, the brushes require practically no lead. In shunt and compound dy- namos the lead varies with the load, and therefore the brushes must be rotated in the direction of rotation of the armature with an increase of load, and in the opposite direction with a decrease of load. Fig. 6S0. — Method of taking temperature. In taking the temperature of a hot part, it is con- venient to use a thermometer in which the scale of degrees has been etched on the stem. Bind this to the heated part, ha^•ing first taken the precaution to cover the bulb with waste to prevent the radiation of heat and take the reading when the column of mercury has ceased to rise. The question which most often presents itself to the attendant is hoiv hot can the various parts of a dynamo or motor become and yet be within the safe limit. The degree of heat can be determined by applWng the hand to the various parts. If tha heat be bearable it is entirely harmless, but if the heat become unbearable to the hand for more than a few seconds, the safety limit has been reached and the machine should be stopped and the fault located. Of course when the solder begins to melt at the commutator connections and shellac begins to "fr\- out " of the armature and an odor of burnt cotton begins to pervade the air. the safe limit has been far e.xceeded. and in most cases, as a matter of fact, serious damage is the result. To be more definite, no part of the dynamo or motor should be aUo^ved to rise in temperature more than SO degrees F. above the temperature of the surrounding air. excepting in the case of commutators where no solder has been used to connect the leads. These can be allowed to rise to a still higher temperature. In cases where the dynamos are subjected to a rapidly var^'ing or fluctuating load, it is of course not possible to constantly shift the brushes as the load varies, therefore the brushes should be fixed in the positions where the least sparking occurs at the moment of adjustment. OPERA TION OF D YNA MOS 591 If at any time violent sparking occur, which cannot be reduced or suppressed by varying the position of the brushes by rotating the rocker, the machine should be shut down at once, otherwise the com- mutator and brushes are liable to be destroyed, or the armature burnt up. This especially refers to high tension machines. Oues. What should be done if the brushes begin to spark excessively ? Ans. First, look at the ammeter to see if an excessive amount of current is being delivered; second, see if the brushes make good contact with the commutator, and if the latter have a bar too high, or too low, and an open circuit. Figs. 681 and 682. — Remedies for leakage of oil from self-oiling bearings. If there be sufficient space, a metal ring may be attached to the shaft as in fig. 681. With this arrangement the high speed of the shaft will carry the oil outside of the ring and throw it off in the oil reservoir. Another way is to insert a tin apron, as shown in fig. 682 at T, which will serve to drain the oil which may creep along the shaft, and also cut off the draft from the pulley which may suck the oil out of the bearing. Sometimes a tin fan is attached to the pulley, which tends to drive the oil back into the bearing, and which also assists in keeping the bo.\ cool. Oues. What should be done if the current be excessive ? Ans. If the current exceed the rated capacity by more than 50 per cent., and continue for more than a few minutes, the main switch should be opened, otherwise the machine may be seriously injured. 592 HAWKINS ELECTRICITY Oues. How does an excessive current injure a dynamo? Ans. By causing it to overheat which destroys the insulation of the armature, commutator, etc. Lubrication. — The shaft bearings of dynamos may be lubricated by sight feed oilers or oil rings. The latter method is almost universally used. An oil well is provided in the hollow casting of the pedestals as shown in fig. 728. Oil rings revolve Fig. 683. — Imaginative view of a shaft showing its rough granular structure. In operation these minute irregularities interlock and act as a retarding force, or frictional resistance. Hence, the necessity for lubrication — a lubricant presents a thin intervening film against which the surfaces rub. with the shaft and feed the latter with oil, which is contin- uously brought up from the reservoir below. The dirt settles to the bottom and the upper portion of the oil remains clear for a long period, after which it is drawn off through the spigot and a fresh supply poured in through openings provided in the top. The latter are often located directly over the slots in which the rings are placed, so that the bearings can be lubricated directly by means of an oil cup, if the rings fail to act or the reservoir become exhausted. OPERATION OF DYNAMOS 593 Oues. What kind of oil can should be used in filling the reservoir, or oil cups? Ans. One made of some non-magnetic material such as copper, brass, or zinc. If iron cans be used, they are liable to be attracted by the field magnets, and thus possibly catch in the armature. Oues. What is the indication of insufficient lubri- cation? Ans. The bearings become unduly heated. Oues. What precaution should be taken with new dynamos? Ans. They are liable to heat abnormally and for the first few days they should be carefully watched and liberally sup- plied with oil. After a dynamo has been running for a short time under full load, its armature imparts a certain amount of heat to the bearings, a little more also to the bearing on the commutator end of shaft; beyond this there is no excuse for excessive heating. The latter may result from various causes, some of which are given with their remedies, as follows: 1. A poor quality of oil, dirty or gritty matter in the oil; 2. Journal boxes too tight; 3. Rough journals, badly scraped boxes; 4. Belt too tight; _ 5. Bearings out of line; 6. Overloaded dynamo; 7. Bent armature shaft. Oues. What is the allowable degree of heating? Ans. It may be taken as a safe rule that no part of a working dynamo should have a temperature of more than 80° Fahr. above that of the surrounding air. Accordingly, if the temperatur& of the engine room be noted before applying the thermometer to the machine, it can at once be seen if the latter be working at a safe temperature. In taking the temperature. 594 HAWKINS ELECTRICITY the bulb of the thermometer should be wrapped in a woolen rag. The screws and nuts securing the different connections and cables should be examined occasionally, as they frequently work loose through vibration. Instructions for Stopping Dynamos. — When shutting down a machine, the load should first be gradually reduced, if possible, by easing down the engine; then when the machine is supplying little or no current, the main switch should be opened. This reduces the sparking at the switch contacts, and prevents the engine racing. Fig. 684. — Diagram illustrating forces acting on a dynamo armature. In the figure the normal field magneto-motive force is in the direction of the line 1 . 2, produced by the field circuit G,if there were no current in the armature. But as soon as the armature current flows, it produces the opposing force 3, 4, which must be combined with 1, 2 to give the resulting force to produce magnetism and hence voltage. The resultant 1, .^, if 3, 4 be large enough, does not differ much from the original force 1. 2. Or, e.xpressed in a more physical way, the brushes E, F, rest on the commutator and all the turns embraced by twice the angle 6. 3, F, oppose the flow of flux through the armature core as well as all the turns embraced by twice the angle, 7, 3, E. The remaining turns distort the flux, making the pole comers at A and B denser, and at C and D rarer. So that all the effect is to kill an increase of flux, or voltage. This cross magnetism tends also to decrease the flow of flux, for the extra ampere turns required to force the flux through the dense pole tips are greater than the decreased ampere turns relieved by the reduction of flux at the other pole tips; this follows, since iron as it increases in magnetic density requires ampere turns greater in proportion than the increase of flux. When the voltmeter almost indicates zero, the brushes should be raised from contact with the commutator. This prevents the brushes being damaged in the event of the engine making a backward motion, which it often does, particularly in the case of OPERA TION O F D YNA MOS 595 a gas engine. On no account, however, should the brushes be raised from the commutator while the machine is generating any considerable voltage; for not only is the insulation of the ma- chine liable to be damaged, but in the case of large shunt dynamos, the person lifting the brushes is liable to receive a violent shock. Oues. What attention should the machine receive after it has been shut down? Ans. It should be thoroughly cleaned. Any adhering copper dust, dirt, etc., should be removed from the armature by dusting with a stiff brush, and the other portions of the machine should be thoroughly cleaned with linen rags. Waste should not be used, as it is liable to leave threads or fluff on the projecting parts of the machine, and on the windings of the armature, which is difficult to remove. Oues. What attention should be given to the brushes and brush gear? Ans. They should be examined and thoroughly cleaned. If necessary the brushes should be refitted and readjusted. All terminals, screws, bolts, etc., should be carefully cleaned and screwed up ready for the next run. The brush holders should receive special attention, as when dirty, they are liable to stick and cause sparking. All dirt and oil should be removed from the springs, contacts, pivots, and other working parts. It is advisable at stated intervals to entirely remove the brush holders from the rocker arms, and give them a thorough cleaning by taking them to pieces, and cleaning each part separately with emery cloth and benzoline or soda solution. Another point to which particular attention should be given is the cleaning of the brush rocker. This being composed wholly of metal, and the two sets of positive and negative brushes being only separated from it by a few thin insulating washers, it follows that if any copper dust given off by the brushes be deposited in the neighborhood of these washers, there is considerable liability for a short circuit of the machine to occur by the dust bridging across the insulation. 596 HAWKINS ELECTRICITY Oues. What further attention should be given? Ans. It is a good plan, when the machine has been thor- oughly cleaned and all connections made secure, to occasionally- test the insulation of the different parts. If a record be kept of these tests, any deterioration of the insulation can at once be detected, localized and remedied before it has become suffi- ciently bad to cause a breakdown. As a means of protecting the machine from any moisture, dirt, etc., while standing idle, it is advisable to cover it with a suitable waterproof cover. COUPLING OF DYNAMOS 597 CHAPTER XXX COUPLING OF DYNAMOS Series and Parallel Connections. — When it is necessary to generate a large and variable amount of electrical energy, as must be done in central generating stations, apart from the question of liability to breakdown, it is neither economical nor desirable that the whole of the energy should be furnished from a single dynamo. Since the efficiency of a dynamo is dependent upon its output at any moment, or the load at which it is worked (the efficiency varying from about 95 per cent, at full load to 80 per cent, at half load), it is advisable m order to secure the greatest economy in working, to operate any dynamo as near full load as possible. Under the above circumstances, when the whole of the output is generated by a single dynamo this can evidently not be eflfected, for the load will naturally fluctuate up and down during the working hours, as the lamps, motors, etc., are switched into and out of circuit; hence, although the dynamo may be working at full load during a certain portion of the day, at other times it may probably be working below half load, and therefore the efficiency and economy in working in such an arrangement is very low. Ques. How is maximum efficiency secured with vari- able load ? Ans. It is usual to divide up the generating plant into a number of units, varying in size, so that as the load increases, it can either be shifted to machines of larger size, or when it exceeds the capacity of the largest dynamo, the output of one 598 HAWKINS ELECTRICITY can be added to that of another, and thus the dynamos actually at work at any moment can be operated as nearly as possible at full load. Oues. What should be noted with respect to connecting one dynamo to another? Ans. It is necessary to take certain precautions (as later explained) in order that the other dynamos may not be affected by the change, and that they may work satisfactorily together. Oues. What are the two methods of coupling dynamos? Ans. They are connected in series, or in parallel. In coupling dynamos in series, the current capacity of the plant is kept at a constant value, while the output is increased in proportion to the pressures of the machines in circuit. When connected in parallel, the pressujes of all the machines are kept at a constant value, while the output of the plant is increased in pro- portion to the current capacities of the machines in circuit. Coupling Series Dynamos in Series. — Series wound dyna- mos will run satisfactorily together without special precautions when coupled in series, if the connections be arranged as in fig. 685. The positive terminal of one dynamo is connected to the negative terminal of the other, and the two outer terminals are connected directly to the two main conductors or bus bars through the ammeter A, fuse F, and switch S. If it be desired to regulate the ]»ressure and output of the machines, variable resistances, or hand regulators R, R*, may be ar- ranged as shunts to the series coils as shown, so as to divert a por- tion or the whole of the current therefrom. Series Dynamos in Parallel. — Simple series wound dynamos not being well adapted for the purpose of maintaining a constant pressure, are in practice seldom coupled in parallel; the condi- tions or working, however, derive importance from the fact that COUPLING OF DYNAMOS 599 compound dynamos, being provided with series coils, arc subject to similar conditions when working in parallel, which is frequently the case. Oues. What may be said with respect to coupling two or more plain series dynamos in parallel? Ans. The same procedure cannot be followed as in the case of plain shunt dynamos, for the reason that if the voltage of the Fig. 685. — Diagram showing method of coupling series dynamos in series. R and R' are two hand regulators which are placed in shunt across the coil terminals to regulate the pressure and output of the machine. dynamo to be coupled be exactly equal to that of the bus bars when connected in parallel, the combination will be unstable. Oues. Why is this? Ans. If, from any cause, the pressure at the terminals of one of the dynamos fall below that of the others, it immediately takes a smaller proportion of the load; as a consequence, the 600 HAWKINS ELECTRICITY ctirrent in its field coils is reduced, and a fvirther fall of pressure immediately takes place. This again causes the dynamo to re- linquish a portion of its load, and again occurs a further fall of pressure. Thus the process goes on, until finally the dynamo ceases to supply current, and the current from the other dyna- mos flowing in its field coils in the reverse direction reverses its magnetism, and causes it to run as a motor against the dri\'ing power in the opposite direction to that in which it previously ran as a dynamo. --f Fig. 686. — Diagram showing method of coupling series dynamos £n i)aralleL In the diagram A, A', are ammeters ; F, F', fuses ; S, S', switches. Under such circumstances the armature is liable to be destroyed if the fuse be not immediately blown, and in any case it is subjected to a very detrimental shock. This tendency to reverse in series dynamos can be effectually prevented by connecting the field coils of all the dynamos in parallel. Oues. How are the field coils of all the dynamos con- ^ nected in parallel? Ans. This is effected in practice by connecting the ends of all the series coils where they join on to the armattire circuit by a CO UPLING OF D YNA MOS 601 third connection, called the " equalizing connection," or " equal- izer," as shown in fig. 686. Ques. What is the effect of the equalizer? Ans. The immediate effect is to cause the whole of the current generated by the plant to be divided among the series coils of the several dynamos in the inverse ratio of their resistance, without any regard as to whether this current comes from one armature, or is divided among the whole. The fields of the several dynamos being thus maintained constant, or at any rate being caused to vary equally, the tendency for the pressure of one dynamo to fall below that of the others is diminished. Shunt Dynamos in Series. — The simplest operation in connection with the coupling of dynamos, and the one used prob- ably more frequently in practice than any other, is the coupling of two or more shunt dynamos to run either in series or in parallel. When connected in series, the positive terminal of one machine is joined to the negative of the other, and the two outer terminals are connected through the ammeter A, fuses F, F', and switch S, to the two main conductors or omnibus bars as represented in fig. 687. The machine will operate when the connections are arranged in this manner, if the ends of the shunt coils be con- nected to the terminals of their respective machines. Shunt Dynamos in Parallel. — The coupling of two or more shunt dynamos to run in parallel is effected without any difficulty. This method of coupling dynamos is one that is very frequently used. Fig. 688 illustrates diagrammatically the method of arranging the connections. The positive and negative terminals of each machine are connected respectiv^ely to two massive insulated copper bars, shown at the top of the diagram, called omnibus bars, through the double pole switches, S, S', and 602 HAWKINS ELECTRICITY the double pole fuses F, F'. Ammeters, A, A' are inserted in the main circuit of each machine, and sen^e to indicate the amoimt of current generated by each. An automatic sv^-itch or cutout, Ac, Ac', is also shown as being included in the main circuit of each of the machines, although this appliance is some- times dispensed ^\'ith. The pressure of each of the machines is regulated independently by means of the hand regulators R, R', inserted in series vrith the shunt circuit. BUS BARS ^vwww — vwww^ IG. 687. — Diagram showing method of coupling shunt djTiamos in series. The ends of the shunt coils may be connected to the terminals of their respective machine, or they may bo connected in series as shown. The shunt circuits are represented as being connected to the positi%'e and negative terminals of the respective machines, but in many cases where the load is subjected to sudden variations, and when a large number of machines is connected to the bus bars, the shunt coils are frequently connected direct to these. In such circumstances this method is preferable, as by means of it the fields of the idle dynamos can be excited almost at once direct from the bus bars by the current from the working dynamos; hence, if a heavy ]oad come on suddenly, no time need be lost in building up a new machine previous to switching it into parallel. The pressure of the lamp circuit is given by a voltmeter. COUPLING OF DYNAMOS 6U3 whose terminals are placed across the bus bars; and the pressure at the terminals of each of the machines is indicated by separate voltmeters or pilot lamps, the terminals of which are connected to those of the respective machines. Oues. Describe a better method of parallel connection. Ans. Better results are obtained by connecting both the shunt coils in series with one another, so that they form one long .siiunt between the two main conductors, the same as in fig. 687. BUS B^RS Fig. 688. — Diagram showing method of coupling shunt dynamos in parallel. When arranged in this way, the regulation of both machines may be effected simultaneously by inserting a hand regulator (R) in series with the shunt circuit as represented. Switching Dynamo Into and Out of Parallel. — In order to put an additional dynamo in parallel with those already work- ing, it is necessary to run the new dynamo up to full speed, and, when it excites, regulate the pressure by means of a hand regu- lator until the voltmeter connected to the terminals of the machines registers one or two volts more than the voltme<" 604 HAWKINS ELECTRICITY connected to the lamp circmt, and then close the switch. The load upon the machine can then be adjusted to correspond w-ith that upon the other machines by means of the hand regulator. Oues. In connecting a shunt dynamo to the bus bars, must the voltage be carefully adjusted? Ans. There is little danger in overloading the armature in making the connection hence the pressure need not be accurately adjusted. It is, in fact, common practice in central stations to judge the voltage of the new dynamo merely by the appearance of its pilot lamp. Oues. How is a machine cut out of the circuit? Ans. WTien shutting dowTi a machine, the load or cturent must first be reduced, by gradually closing the stop valve of the engine, or inserting resistance into the shunt circuit by means of the hand regulator; then when the ammeter indicates nine or ten amperes, the main switch is opened, and the engine stopped. By following this plan, the heavy sparking at the switch contacts is avoided, and the tendency for the engine to race, reduced. Oues. What precaution must be taken in reducing the current? Ans. Care must be taken not to reduce the current too much. Oues. Why is this necessary? Ans. There is danger that the machine may receive a reverse current from the other dynamos, resulting in hea\^'' sparking at the commutator, and in the machine being driven as a motor. Oues. What provision is made to obviate this danger? Ans. Dynamos that are to be run in parallel are frequently provided with automatic cutouts, set so as to automatically COUPLING OF DYNAMOS 605 switch out the machine when the current falls below a certain minimum value. Dividing the Load. — If a plant, composed of shunt dynamos running in parallel, be subjected to variations of load, gradual or instantaneous, the dynamos will, if they all have similar characteristics, each take up an equal share of the load. If, however, as is sometimes the case, the characteristics of the dynamos be dissimilar, the load will not be shared equally, the dynamos with the most drooping characteristics taking less than their share with an increase of load, and more than their share with a decrease of load. If the difference be slight, it may be readily compensated by means of the hand regulator increasing or decreasing the pressures of the machines, as the load varies. If, however, the difference be considerable, and the fluctuations of load rapid, it becomes practically impossible to evenly divide the load by this means. Under such circumstances, the pressure at the bus bars is liable to great variations, and there is also liability of blowing the fuses of the overloaded dynamos, thus precipitating a general breakdown. To cause an equal division of the load among all the dynamos, under such circumstances, it is needful to insert a small resistance in the armature circuits of such dynamos as possess the straightest characteristics, or of such dynamos as take more than their share of an increase of load. By suitably adjusting or proportioning the resistances, the pressures at the terminals of all the machines may be made to vary equally under all varia- tions of load, and each of the machines will then take up its proper share of the load. Coupling Compound Dynamos in Series. — Since com- pound dynamos may be regarded as a combination of the shunt and series wound machines, and as no special difficulties are \ GOG HAWKINS ELECTRICITY encoiintered in ninning these latter in series, analogy at once leads to the conclusion that compound dynamos under similar circumstances may be coupled together with equal facihty. Ques. How are compound dynamos connected to operate in series? Ans. The series coils of each are connected as in fig. 685, and the shunt coils are connected as a single shunt as in fig. 687, which — I 3v'S BARS RHEOSTAT ■f , . SHUNT - Fig. 6S9. — Coupling compound djiiamos in series; short shunt connection, indicate the changes that would be made for long shunt connection. The dotted lines may either extend simply across the outer brushes of the machines, so as to form a double short shtmt, or may be a shunt to the bus bars of external circuit, so as to form a double long shimt. Compound Dynamos in Parallel. — Machines of this type will not run satisfactorily together in parallel unless all the series coils are connected together by an equaUzing connection, as in series dynamos. The method of arranging the connections CO UP LING OF D YNA MOS 607 as adopted in practice, being illustrated in fig. 690. By means of it idle machines are completely disconnected from those at work. Oues. How is the equalizer connected? Ans. The equalizer is connected direct to the positive brushes of all the dynamos, a three pole switch being fitted for discon- necting it from the circuit when the machine to which it is con- ZER NO.l DYNAMO SHUNT COILS SHUNT COILS '-p" SHUNT COILS n[ Fig. 690. — Diagram showing method of coupling compound dynamos in parallel. nected is not working. The two contacts of the switch are re- spectively connected to the positive and negative conductors, while the central contact is connected to the equalizer. Switching a Compound Dynamo Into and Out of Parallel. — If the characteristics of all the dynamos be similar, and the coimections arranged as in figs. 690, or 691, the only precaution 608 HAWKINS ELECTRICITY to be observed in switching a new machine into parallel is to have its voltage equal, or nearly equal to that of the bus bars previous to closing the switch. If this be the case, the new machine will take up its due share of the load without any shock. Ques. How is a compound dynamo, running in parallel, cut out of circuit? Ans. The load is first reduced to a few amperes, as in the case of shunt dynamos, either by easing down the engine, or by SHUNT COILS r^ SHUNT COILS M Fig. 091. — Diagram showing another and better method of coupling compound dynamos in parallel. With this arrangement the idle machines are completely disconnected from those at work. The same reference letters are common in both diagrams. S, S' are switches; F, F' fuses; A, A' ammeters, which indicate the total amount of current generated by each of the machines; AC, AC, automatic switches, arranged for automatically switching out a machine in the event of the pressure at its terminals being reduced through any cause; R. R' are hand regulators, inserted in the shunt circuits of each of the machines, by means of which the pressures of the individual machines may be varied and the load upon each adjusted. The pressure at the bus bars is given by the voltmeter V, one terminal of which is connected to each of the bars; a second voltmeter may be used, to give the pressure of any individual machine, by connecting "voltmeter keys" to the terminals of each of the machines, or a separate voltmeter may be used for each individual machine. The only essential difference between figs. (190 and 691 is. that in fig. 690 the equalizer is connected direct to the positive brushes of all the dynamos, while in fig. C91 the equalizer is brought up to the switchboard and arranged between the two bus bars, a switch being fitted for disconnecting it from the circuit when the machine to which it is connected is not working. CO UPLING OF D YNA MOS 609 cutting resistance into the shunt circuit by means of the hand regulator, and then opening the switch. Previous to this, however, it is advisable to increase the voltage at the bus bars to a slight extent, as while slowing down the engine the load upon the outgoing dynamo is transferred to the other dynamo arma- tures, and the current in their series coils not being increased in proportion, the voltage at the bus bars is consequently reduced somewhat. Equalizing the Load. — When a number of compound dyna- mos of various output, size, or make, are running together in parallel, it frequently happens that all their characteristics are not exactly, similar, and therefore the load is unequally dis- tributed, some being overloaded, while others do not take up their proper share of the work. NOTE. — The action of an equalizing bar in equalizing the load on compound dynamos run in parallel may be explained as follows: 'I'he compound winding of a dynamo raises the pressure in proportion to the current flowing through it, and if , in a system of parallel operated compound dynamos without the equaUzing connection, the current given by one machine were slightly greater than the currents from the others, the pressure of that machine would increase. With this increase in pressure above the other machines, a still greater current would flow, and so raise the pressure further. The effect is therefore cumulative, and in time the one dynamo would be carrying too great a proportion of the whole current of the system. With the equal- izing connection, whatever the current flowing from each machine, the currents in the various compound windings are all equal, and so the added pressure due to the compound winding is practically the same in each machine. Any inequality in output from the machines is readily eliminated by adjusting the shunt currents by means of the shunt rheostats. When compound wound dynamos are operated in parallel, the equalizer bar insures uniform distribution among the series coils of the machines. NOTE. — To secure the best results in parallel operation, dynamos should be of the same design and construction and should possess as nearly as possible the same characteristics; that is, each should respond with the same readiness, and to the same extent, to any change in its field excitation. Any number of such machines may be operated in parallel. 1 he usual practice is to connect the equalizer and the series field to the positive terminal, though if desired, they may be connected to the negative terminal ; both however, must be connected to the same terminal. The resistance of the equaUzer should be as low as possible, and it must never be greater than the resistance of any of the leads from the dynamos to the bus bar. Sometimes a third wire is run to the switchboard from each dynamo and there connected to an eqalizer bar, but the usual practice is to run the equalizer directly between the dynamos and to place the equalizer switches on pedestals near the machines. This shortens the connections and leads to better regulation. The positive and equalizer switches of each machine differ in pressure only by the slight drop in the series coil, and in some large stations these two switches are placed side by side on a pedestal near the machine. In such cases, the equalizer and positive bus bars tire often placed under the floor near the machines, so that all leads may be as short as possible. If all the dynamos be of equal capacity, all the leads to bus bars should be of the same length, and it is sometimes necessary to loop some of them. CIO HAWKINS ELECTRICITY If the difference be small, it may be compensated by means of the hand regulator; if large, however, other means must be taken to cause the machines to take up their due proportion of the load. If the series coils of the several dynamos be pro\aded with small adjustable resistances, in the form of German silver or copper ribbon inserted in series with the coils, the distribution of the current in the latter m.ay be altered by varying the resistance attached to the individual coils. The effect of the series coils upon the individual armatures in raising the pressure may be adjusted, and the load thus evenly divided among the machines. Shunt and Compound Dynamos in Parallel. — It is not practicable to run a compovmd dynamo and a shunt dynamo in parallel, for, unless the field rheostat of the shunt machine be adjusted continually, the compound dynamo will take more than its share of the load. DYNAMO FAILS TO EXCITE 611 CHAPTER XXXI DYNAMO FAILS TO EXCITE This trouble is of frequent occurrence in both old and new machines. If a dynamo fail to excite, the operator should first see that the brushes are in the proper position and making good contact, and that the external circuit is open if the machine be shunt wound, and closed if series wound. In starting a dynamo it should be remembered that shunt and compound machines require an appreciable time to build up, hence, it is best not to be too hasty in hunting for faults. The principal causes which prevent a dynamo building up are : 1. Brushes not properly adjusted; 2. Defective contacts; 3. Incorrect adjustment of regulators; 4. Speed too low; 5. Insufficient residual magnetism; 6. Open circuits; 7. Short circuits; a. In external circuits; b. In dynamo. 8. Wrong connections; 9. Reversed field magnetism. Brushes not Properly Adjusted.— If the brushes be not in or near their correct positions, the whole of the voltage of the armature will not be utilized, and will probably be insufficient to 612 HAWKINS ELECTRICITY excite the machine. If in doubt as to the correct positions, the brushes should be rotated by means of the rocker into various points on the commutator, sufficient time being given the ma- chine to excite before mo^•ing them into a new position. Defective Contacts. — If the different points of contact of the connections of the inachine be not kept thoroughly clean and free from oil, etc., it is probable that enough resistance will be interposed in the path of the exciting current to prevent the machine building or exciting. Each of the contacts should therefore, be examined, cleaned, and screwed up tight. Oues. Which of the contacts should receive special attention? Ans. The contact faces of the brushes and surface of the commutator. These are very frequently covered with a slimy coating of oil and dirt, which is quite sufficient to prevent the machine exciting. Incorrect Adjustment of Regulators. — When shimt and compound machines are provided %\-ith field regtUators, it is possible that the resistance in circuit may be too great to permit the necessary strength of exciting current passing through the field Vi4ndings. Accordingly, the fault is corrected by cutting out more or less of the resistance. The field coils of series ma- chines are sometimes pro\-ided v\-ith short circuiting switches or resistances arranged to shunt the current across the field coils. If too much of the current be shunted across, the switch should be opened, or if there be a regulator, it shovild be so adjusted that it Vi-ill pass enough current through the field windings to excite the machine. DYNAMO FAILS TO EXCITE 613 Speed too Low. — In shunt and compound dynamos there is a certain critical speed below which they will not excite. If the nonnal speed of the machine be known, it can be seen whether the failure to excite arises from this cause, by measuring the speed of the armature with a speed indicator. In all cases it is advisable, if the machine do not excite in the course of a few minutes, to slightly increase the speed. As soon as the voltage rises, the speed may be reduced to its regular rate. 4>0 Fig. 692. — Method of testing for break by short circuiting the terminals of the machine. If the external circuit test out apparently all right, and there be no defective contacts in any part of the machine, and all short circuiting switches, etc., be cut out of circuit, the machine still refusing to excite, short circuiting the terminals of the machine fhould be tried. This should be done very cautiously, especially in case of a high tension machine. It is advisable to have, if possible, only a portion of the load in circuit, and the short circuit should be effected as shown in the figure. The short circuit may be made by momentarily bridging across the two terminals of the machine with a single piece of wire. As this, how- ever, is liable to burn the terminals, a better plan is to fix a short piece of scrap wire in one terminal, and then with another piece of insulated wire make momentary contacts with the other terminal and the short piece of wire. If the machine excite, it will be at once evident by the arc which occurs between the two pieces of wire. As the voltage of a series machine when induced to build in this manner generally rises very rapidly, great care should be taken that the contact is at first only momentary, merely a rubbing or scraping touch of the wires. The contact may be prolonged if the machine do not excite at the first contact. Compound wound machines can often be made to excite quickly by short cir- cuiting their terminals in this manner. 614 HAWKINS ELECTRICITY Insufficient Residual Magnetism. — This fault is not of frequent occurrence; it takes place chiefly when the dynamo is new, and may be remedied by passing the ciirrent from a few storage cells, or from another dynamo, for some time in the proper direction through the field coils. If a heavy current, such as is obtainable from a storage battery, be not available, and the machine be shunt or compound wound, a few primary cells arranged as in fig. 693 will generally sufl5ce. Pig. 693. — Method of overcoming insufBcient residual magnetism. The flexible "lead" L cf the dynamo p is disconnected from the positive terminal of the machine, and is connected to the negative or zinc pole of the battery B. the other or positive carbon pole being con. nected to the terminal, from which the lead was removed, and shunt circuit S. As thus; arranged, it will be seen that the battery B is in series with the armature and shunt circuit, and therefore its voltage will be added to any small voltage generated in the armature. When the machine is started, the combined voltages will probably be able to send sufficient current through the shunt to excite the machine. As the voltage rises and the strength of the current in the shunt windings increases, the flexible lead may be again inserted into the terminal from which it was removed. The battery will thus be short circuited, and may be cut out of circuit without any danger of breaking the shunt circuit, and thus caus- ing the machine to demagnetize. Open Circuits. — Dynamos are affected by open circuits in different ways, depending upon the type. Series machines re- quire closed circuit to build up, while an open circuit is necessary DYNAMO FAILS TO EXCITE G15 with the shunt machine. An open circuit may be due to: 1, broken wire or faulty connection in the machine; 2, brushes not in contact with commutator; 3, safety fuse blown or re- moved; 4, circuit breaker open; 5, switch open; 6, external circuit open. If the trouble be due merely to the switch or external circuit being open, the magnetism of a shunt machine may be at full strength, and the machine itself may be working perfectly, but if the trouble be in the machine, the field mag- netism will probably be very weak. Open circuits are most likely to occur in: 1. The armature circuit; 2. The field circuit; 3. The external circuit. When the open circuit is due to the brushes not making good contact, simple examination generally reveals the fact. Oues. What causes breaks in the field circuit? Ans. Bad contacts at the terminals, broken connections, or fracture of the coil windings. Oues. How is the field circuit tested for breaks? Ans. The flexible leads attached to the brushes are removed from their connections with the field circuit, and the latter is then tested for conductivity with a galvanometer. Oues. Where is a break likely to occur in a shunt machine? Ans. In the hand regulator through a broken resistance coil or bad contact. Very frequently the fault occurs in the connecting wires leading from the machine to the hand regulator fixed upon the switchboard, or in the short wires connecting the field coils to the terminals or brushes. 616 HA WKIXS ELECTRICITY The insulation of a broken wire will sometimes hold the two ends together 30 as to defy any but the most careful inspection or exam- ination; therefore, in order to avoid loss of time, it is ad\Tsable to dis- connect the wires if possible, and test each separately for conducri\-ity ■wnth a batten,' and galvanometer connected, as in fig. 694. If the fault be not located in the various connections, the magnet coils should be tested with the batten.- and galvanometer coupled up as in fig. 706, care being first taken to disconnect the ends of each of the coils. A faiilt)- coil will not show any deflection of the gal\'anometer. Fig. 694. — Method of testing dj-namo for short circuits. la the figure, ooe pole of the battel^ B is placed in contact with the frame of the xn.aciui:e at a point wliicb Ins ptevioady been well scraped and cleaned; the other pole is connected to one of the galvanometer temnnals as shown. The other terminal of the gal%-anomeier is connected to eadi of the dynamo terminals T T under test in turn. If a deflection of the needle be prodnoed when the gal- vanometer terminal is in contact with either, the terminals are in omtact with the frame. and they should then be removed, and the fault repaired by additional insalataoo or by rednsulating. Oues. At what point of a shunt coil does a break usually occur? Ans. At the point where the wire passes through the flanges of the spool or bobbin. DYNAMO FAILS TO EXCITE 617 Oues. How should the coil be repaired ? Ans. In most cases a little of the wood or metal of which the flanj^je is made can be gouged or chipped out, and a new connect- ing wire soldered on to the broken end of the coil without much difficulty. If it be necessary to take the magnets apart at any time, care should be taken in putting them together again to wipe all faces perfectly clean, and screw up firmly into contact, and to see that the connections of the coils are made as they were before being taken apart. Fig. 695. — Watson armature discs. Each laminaiion is made from low carbon electrical steel of high magnetic permeability. Each disc is annealed and afterwards varnished. If the faulty coil cannot be repaired quickly, and the machine is urgently required, the coil may be cut out of circuit entirely, or short circuited, and the remaining coils coupled up so as to produce the correct polarity in the pole pieces. Oues. What trouble is liable to be encountered in operating after cutting out a coil? Ans. The remaining coils are liable to heat up to a greater extent than formerly, owing to the increased current, hence it is advisable to proceed cautiously in starting the dynamo, since the temperature may exceed a safe limit. If this occur, a resist- ance may be put in circuit with the field coils, or the speed of the dynamo reduced. 618 HAWKINS ELECTRICITY Fig. 696.— Fort Wayne pedestal typ)e commu- tator truing device. When this device is used, the armature is revolved in its own bearings by means of a handle clamped to the pulley. The tool has a horizontal travel of 21 ins., (being 3 ins, wide inside the fasten- ing bolt in the base), and a vertical adjust- ment of 12 ins., adapt- ing it to machines with commutators up to 36 ins. in diameter. Fig. 697.— Fort Wayns yoke type commutator truing device for ma- chines ha\-ing brush mechanism mounted on a yoke carried by the field frame. It consists of a carriage for the tool holder ha\'ing a screw feed and a bracket for at- taching to the brush yoke. The bracket re- places two brush holder brackets on the brush yoke , and is made to fit the yoke of the par- ticular machine on which it is to be used. DYNAMO FAILS TO EXCITE 619 Oues. What kind of dynamo is affected by breaks in the external circuit? Ans. A scries dynamo. Oues. Name the kind of break that is difficult to locate. Ans. A partial break. Short Circuits. — In a series or compound dynamo a short circuit or heavy load will overload the machine and cause the fuses to blow. A shunt machine will not excite under these circumstances, for the reason that practically the whole of the current generated in the armature passes direct to the external circuit, and the difference of potential between the shunt ter- minals is practically nil. Oues. What should be done if it be suspected that the failure to excite arises from this cause? Ans. The main leads should be taken out of the dynamo terminals, then, if due to this cause, the machine will excite. Oues. What parts of a dynamo are specially liable to be short circuited ? Ans. The terminals, brush holders, commutator, armature coils and field coils. Oues. How are the terminals liable to be short cir- cuited ? Ans. The terminals of the various circuits of the machine are liable to be short circuited, either through metallic dust bridging across the insulation, or through the terminals making direct contact with the frame of the machine. The various terminals should be examined, and if the fault cannot be located by inspection, they should each be disconnected from their cir- cuits and tested with a battery and galvanometer arranged as in fig. 694, (^20 HAWKIXS ELECTRICITY Oues. What precaution should be taken with the brush holders .- Ans. Since, they are liable to be short circuited through the rocker by metallic dust lodging in the insulating washers, they should be kept clean. Oues. How are the brush holders tested ? Ans. A galvanometer and battery are connected in series with one terminal of the galvanometer connected to one set of SUPP& MAGNTIZING COIL COIL UNDER TEST LAMINATED IRON GORE Pig. 69&. — Field coil testing with telephone rectiver. In the method here shown, a telephone receiver is connected in series with two s>iimietrically placed coils A and B. Ver>- little sound will be heard when the flux through the two coils AB is the same; but if a short- circmted coil is being tested, the fluxes through the coils A, B will not t>e equal and a noise can be heard in the leceivier. brushes; the imconnected terminal of the battery is then con- nected with the other set of brushes. A deflection of the needle will indicate a short circviit. Oues. What is the effect of a short circuit in the field coils or field circuit? Ans. The machine generally refuses to e.xcite. Oues. How are the field coils tested for short circuit ? Ans. By measuring the resistance of each coil ■with an ohmmeter or Wheatstone bridge. The faulty coils will show a much less DYNAMO FAILS TO EXCITE 621 resistance than the perfect coils. The fault may also be dis- covered and located by passing a strong current from a battery or another dynamo through each of the coils in turn, and observing the relative magnetic effects produced by each upon a bar of iron held in their vicinity. The short circuit may be in the terminals or connections, and these should first be examined and tested. Some series dynamos are provided with a resistance, arranged in parallel or shunt with the field coils, to divert a portion of the current therefrom, and thus regulate the output. Pig. 699. — Watson armature complete. The armature coils are form wound, heavily insulated and so mounted on the core as to insure rapid dissipation of heat by ventilation. Each coil is protected by an insulating sheath and tape covering before mounting. The arma- ture is baked after the coils are mounted to drive out all moisture, then, while hot, is treated with insulating compound and again baked twelve hours. When making a series dynamo excite, all resistances and controlling devices should be temporarily cut out of circuit by opening the shunt circuit. Series machines have frequently a switch which short circuits the field coils. Care should be taken that this is open, or otherwise the machine will not excite. Wrong Connections. — When a machine is first erected, the failure to build up may be due to incorrect connections. The whole of these latter should therefore be traced or followed 622 HAWKINS ELECTRICITY out, and compared with the diagrams of dynamo connections given in figs. 190 to 198. Sometimes errors are made in connecting the field coils, causing them to act in opposition. This may occur when the dynamo is a new one or the coils have been removed for repairs. It may be caused either through the coils having been put on the field cores the wrong way, or through incorrect coupling up. Under these circumstances, the dynamo, if bi- polar, will fail to excite; and if multipolar, poles will be produced in the yokes, etc. It may be remedied by removing one of the coils from the core and putting it on the reverse way, or by reversing its connections. The correctness of connections of all the coils should be verifisd. In compound dynamos it sometimes happens that the machine will excite properly, but that the series coils tend to reverse the polarity of the dynamo, thus reducing the voltage as the load upon the machine increases. This may be detected when the machine is loaded by short circuiting the series coils, not the terminals. If the voltage rise in doing this, the series coils are acting in opposition to the shunt coils, and the connections of the series coils must be reversed. Reversed Field Magnetism. — This is sometimes caused by the nearness of other dynamos, but is generally due to re- versed connections of the field coils. Under such conditions the field coils tend to produce a polarity opposed to the mag- netization to which they owe their current, and therefore the machine will refuse to excite until the field connections arc reversed, or a current is sent from another dynamo or a battery through the field coils in a direction to produce the correct polarity in the pole pieces. ARMATURE TROUBLES 623 CHAPTER XXXII ARMATURE TROUBLES A large proportion of the mishaps and breakdowns which occur with dynamos and motors arise from causes more strictly wdthin the province of the man in charge than in that of the designer. The armature, being a complex and delicately built structure, is subject in operation to various detrimental in- fluences gi\ing rise to faults. IVIany of the faults which occur are avoided by operators better informed as to the electric and magnetic conditions which obtain in the running of the machine, especially the mechanical stresses on the copper inductors due to the magnetic field and the necessity of preserving proper insulation. The chief mishaps to which armatures are subject are as follows : 1. Short circuits; a. In individual coils; b. Between adjacent coils; c. Through frame or core; d. Between sections of armature; e. Partial short circuits. 2. Grounds; 3. Breaks in armature circuit. G24 HAWKIXS ELECTRICITY Short Circuit in Individual Coils. — This is a common fault, which makes its presence known by a \-iolent heating of the armature, flashing at the commutator, flickering of the light on hghting circuits, and by a smell of burning varnish or over- heated insulation. When these indications are present, the machine should be stopped at once, otherwise the armature is Uable to be burnt out. The fault is due either to metallic dust lodging in the insulation between adjacent bars of the com- FiG. 700. — Method of lofaringsh<»tcircaitedannatnrecwL Disooooect the external and field circnits from the annatore.and pass a laige cnneat — say bom 20 to 100 amperes — from a battery (B> or srnnttwr- dynamo throng the wh<^ aimatme by means of the bntdies. Then, having prevkmsty well cleaned the oommntatw. measure the difierenoe of potential between ad j acent segments all round the commutator (C), by means oS a voltmeter or galvanometer (O;. the terminals at which are connected to adjacent segments, as shown. The short arcnited coQ or coils will be located by the difference of potential between the oorre^mnding segments being little or nothing. It may be remarked, however, that this is not always a decisive test. In some cases the short circait may be intermittent, or may disappear as soon as the armatnre ceases to rotate. In sach cases, the short arcait is caused by die wire coming into contact throo^ the action of the centrifugal forces developed by the rotation of the armature. mutator, or to one or more convolutions of the coils coming into contact with each other, either through a metaUic filing be- coming embedded in the insulation or damage to the insulation. Oues. How is the fault}' coil located? Ans. When the machine is stopped, the faulty coil, if not burnt out, can generally be located by the baked appearance of the varnish or insulation, and by its excessive temperature over ARMATURE TROUBLES 625 the rest of the coils, being detected also by the baked appear- ance of the "^^arnish or insulation. Oues. What should be done if the machine do not build, and it be suspected that the fault is due to short circuited armature coils? Ans. The field magnets should be excited by the current from a storage battery or another dynamo, and, having raised the Fig. 701. — Test for break in armature lead. Clean the brushes and commutator, and apply current from a few cells of battery having a telephone receiver in circuit as shown in the figure. If the machine have more than two brushes, connect the leads to two adjoining brushes and raise the others. IS'ow rotate the armature slowly by hand and there will be a distinct click in the receiver as each segment passes under the brushes until one brush bears on the segment at fault, when the clicking will cease. In making this test, the brushes must not cover more than a single segment. brushes from contact with the commutator, the armature should be run for a short time. In stopping, the faulty coil or coils may be located by the heat generated by the short circuit. When the dynamo is started for the purpose of localizing a short circuit, precautions should be taken, and the machine only run for a few minutes at a time until the faulty coil is detected. When the faulty coil has been located, the insulation between the segments of the commutator to which its ends are connected should be carefully examined for anything that may bridge across from segment to segment, and scraped clean. If the commutator be apparently all 626 HAWKINS ELECTRICITY right, the fault probably lies in the winding. The insulation of the winding should be carefully examined, and any metallic filings or other particles discovered therein carefully removed, and a little shellac var- nish applied to the faulty part. Oues. If the insulation on adjacent conductors has been abraded, how should it be repaired? Ans. A small boxwood or other hardwood wedge, coated with shellac varnish should be driven in tightly between the wire, this will generally be sufficient. Fig. 702. — ^Bar tofaartest foropen drcoitin omlor short cucmtin tne ocal or between segments. If. in testing as in fig. 701 .d in fig. 704. This applies to drum wound armatures. Faialts of this description can frequently be discovered by a carefiJ inspection of the windings of the armature without recourse to testing. When located, the faialt can usually be re- paired with a hardwood wedge, as explained above, or a piece of mica or vulcanized fibre cemented in place with shellac \'amish. ARMATURE TROUBLES 629 Short Circuits between Sections through Frame or Core of Armature.^Detection of this fault can be effected by the methods described above, and by disconnecting the whole of the armature coils from the commutator and from each other, and testing each separately with a battery and galvanometer coupled up as in fig. 705, one mre being connected to the shaft and the other to the end of the coil under test. As a rule, there is no way of remed^Hing this fault other than unwinding the defective coils, reinsulating the core, and rewinding new coils. Fig. 705. — Method of locating short circuits between coils through armature core. The galvanometer, battery and coil to be tested are connected in series as shown, and then the unconnected terminal of the galvanometer is brought into contact with the shaft. If then some portion of the insulation of the wire has been abraded or destroyed, thus bringing the bare wire into contact with the metal core, as at A in the figure, the needle of the gal- vanometer will be deflected since a closed circuit is formed through the core and wire. If the insulation be perfect, the needle will not be deflected. It will thus be seen that in the conductivity test (fig. 700) it is necessary that the needle should be deflected, or turned, to prove that all is right, while in the insulation test the converse holds good; if the needle be deflected, it proves that the insulation is broken down. Short Circuits between Sections through Binding Wires. — This fault is the result of a loose winding, and is caused by the insulation upon which the binding wires are wound giving way, thus bringing coils at different pressures together. As a conse- quence of the hea\'y current which flows, tne binding wires are as a rule imsoldered or burned. The location of the fault can 630 HAWKINS ELECTRICITY therefore be effected b}' simple inspection. To remedy, it will be necessary to unwind and rewind on new binding wires, on bands of mica or vulcanized fibre, soldering at intervals to obviate fl^-ing asunder. Partial Short Circuits in Armatures. — This is usually due to the presence of moisture in the windings. To remedy the fault, the armature should be taken out and exposed to a moderate heat, or subjected to a current equal to that ordina- FiG. 706. — Method of testing for breaks. The instruments are connected as shown. B is the battery, G the galvanometer, ard S the coil of wire being tested. One terminal of the battery is connected to a terminal of the galvanometer, and the other to one of the ends of the coil under test. The other terminal of the galvanometer is connected to the other end of the coil, If the connecting wires be making good electrical contact with the respective terminals, and the wire of coil being tested be unbroken, the needle of the galvanometer will be deflected as soon as a closed circuit is made by the end of the coil coming into con- tact with the galvanometer terminal. If the wire of the coil be broken in some part or the ends of the connecting wires do not make good electrical contact with the tenninals. the needle will not be deflected. In order to prevent mistakes, it is ad\'isable to test the battery and galvanometer corihections and contacts by short circuiting or bringing the ends of the wire connecting the terminal of the galvanometer and negative pole or the battery together before starting to test the circuit or coil. If the needle be deflected, the connections are all right; if not deflected, there is a bad contact somewhere, which must be made good before the test can proceed. rily given by the dynamo. Under the action of heat or of this current the moisture will be gradually dispersed. When thoroughly dry, and while still warm, a coat of shellac should be applied to the whole of the windings. ARMATURE TROUBLES 631 Burning of Armature Coils. — The reason for the burning of an armature coil may be explained as follows: The coil, segments, and the short circuit between the segments form a closed circuit of low resistance so that it is only necessary to have a low pressure set up in the active portion of the coil to force a very large current through the coil and the short circuited com- mutator bars. The heating effect of this current is sufficient to burn otit the coil. "IG. 707. — Watson field coils. Automatic machinery is employed to wind these coils; after winding, they are bound with tape, then baked to expel all moisture, and while hot, are saturated with an insulating compound and again baked for twelve hours to make them practically oil and water proof. Heavy flexible: eads are brought out to avoid danger of breaking or other damage. Cutting Out Damaged Armature Coils. — To cut out a damaged coil from an armature, first, disconnect the coil from the commutator, and after cutting off the leads, insulate the exposed parts with tape. Then connect the commutator bars (which were connected with the leads) with a wire of the same size as the wire winding. To remove the coil entirely, cut the band wire or remove the wedges, and lift up a sufficient number of leads and coils to pewiait of the removal of the damaged coil. 632 HAWKINS ELECTRICITY Grounds in Armatures. — These faults occur when the armature coils become connected to the frame or core of the armature. When this grounding is confined to a single coil, it is not in itself liable to do damage. A simple method of locating a grounded coil is illustrated in fig. 708. Fig. 708. — Method of locating grounded armature coil. B is a batterj' or djTiamo circuit giving a current of a few amperes through the armature by its ovm brushes (1 and 2). At G is placed a roughly made galvanometer, to carry some 25 amperes or so, one terminal being in connection with the shaft of the armature, and the other attached to a movable brush 3. Since the function of the particular galvanometer is simply to show a deflection when a current is passing, and to mark zero when there is none, a coil of thick wire with a pocket compass in the center wrill do all that is required, but care must be taken to remove it sufficiently far away from the disturbing effects of the armature magnetism. The manner of testing is as follows: Assume a steady current to be flowing from battery- B through the armature; touch the commutator with brush 3. and a current will flow through G. Slowly rotate the armature or the brush 3 until the galvanometer G shows no deflection. The coil in contact with 3 will be found to be grounded. A hand regulator or rheostat R may be inserted in series with the battery or dynamo circuit to regulate the strength of the current passing. Oues. What is the advantage of this test? Ans. The damaged coil can be located \\'ithout unsoldering the coils from the commutator, which is sometimes a difficult operation -without proper tools; further, the fault can frequently be repaired VN-ithout disconnecting any of the wires if its exact position be determined. ARMATURE TROUBLES 633 Magneto Test for Grounded Armatures. — A magneto test for grounded armatures is not to be recommended, as armatures often possess sufficient static capacity to cause a magneto to ring even though there be no leak. This is due to the alternating current given by the magneto for when the circuit has capacity it acts as a condenser and at each revolution of the armature of the magneto a rush of current goes out and returns, charging the surfaces of the conductor alternately in opposite directions, and ringing the bell during the process. Fig. 709. — Method of binding armature winding. Complete appliances for handling annatures in making repairs are usually not available with most street railway companies, since they are so seldom required. When needed, therefore, some temporary contrivance must be resorted to for help in the dilemma. Should an armature burn out, some local concern that makes coils and rewinds armatures may be available to do the work; again, it will be necessary to send to the manufacturers for a man, as soon as coils can be made ready for the work. In no case should any but an experienced man be given charge of this work. But if there be any doubt as to whether the armature is really burnt out, let a competent man be the judge. When a large armature needs repairing, a pair of chain tongs can be used on some part of the shaft when putting in the coils, and a block and tackle, as shown, can be used, when putting on the band wires. Do not finish one band and then cut off the wire, but run it over for the next, etc. Then solder and trim off the wires. Breaks in Armature Circuit. — A partial or complete break in the armature circuit is always accompanied by heavy sparking at the commutator, but not, as a rule, by an excessive heating of the armature or slipping of the belt, and this enables the fault to be distinguished from a short circuit. The faulty part can always be readily located by the " flat " which it produces 634 HAWKINS ELECTRICITY upon the surface of the commutator. The armature circuit being open at the faulty part, heavy sparking results at every half revolution as the brushes pass over it, and as a consequence the corresponding segments become " pitted " or " flattened " with respect to the others; they may easily be discovered on examination. Breaks in the armature circuit may occur in either the commutator or in the coils of the armature. To ascertain whether it be in the latter, carefully examine the winding of the faulty coil. The defect may be sought for more particularly at the commutator end of the armature, as breaks in the wire are most frequent where the connections are made with the commutator segments. If no break can be discovered, try passing a heax'j' current through the faulty coil by means of the brushes. If a partial break exist with sufficient contact to pass a current, the coil will be heated at that point and may be discovered by running the fingers over the coil. When located, the fault may be repaired by rewinding the coil, or carefully cleaning the broken ends and jointing. The fault may also be temporarily repaired by soldering the adjacent commutator segments together without disconnecting the coil. CARE OF THE COMMUTATOR AND BRUSHES G35 CHAPTER XXXIII CARE OF THE COMMUTATOR AND BRUSHES For satisfactory operation, the brushes and commutator must be kept in good condition. To this end the main thing to l)e guarded against is the production of sparks at the brushes. If care be taken in the first instance to adjust the brushes to their setting marks, and to regulate their pressure upon the commu- tator, and afterwards to attend to the lead as the load varies, so that little or no sparking occurs, and also to keep the brushes and commutator free from dirt, grit, excessive oil, etc., the sur- face of the coinmutator will assume a dark burnished appearance and wear will practically cease. Under these circumstances the commutator will run cool, and will give very little trouble. In order to maintain these conditions it will only be neces- sary to see that the brushes are kept in proper condition and fed forward to their setting marks, as they wear away, and that the commutator is occasionally polished. If the pressure of the brushes upon the commutator be too great, or their adjustment faulty, or the commutator be allowed to get into a dirty condition, sparking will result, and, if not at once attended to and remedied, the brushes will quickly wear away, and the surface of the commutator will be destroyed. As this action takes place, in the earlier stages, the surface of the commutator will become roughened or scored, resulting in 636 HAWKINS ELECTRICITY jumping of the brushes, and increased sparking; in the later stages, the commutator w-ill become untrue and worn into ruts, moreover, owning to the violent sparking which takes place through this circumstance, the machine ■u'ill quickly be rendered useless. Oues. How is the commutator easily tested as to the condition of its surface? Ans. It is readily tested by resting the back of the finger nail upon it while in motion; the nail being very sensitive to any irregularities, indicates at once any defect. Oues. What causes grooves or ridges to be cut in the commutator? Ans. They result from using brushes with hard burnt ends which are not pliable ; also by too great a pressure of the brush upon the commutator surface. Sparking at the brushes is expensive and detrimental, chiefly because it results in burning the brushes and also the commutator, necessitating their frequent renewal. Ever>' spark consumes a particle of copper, torn from the commutator or brush. The longer the sparking continues, the greater the evil becomes, and the remedy must be applied without delay. Oues. What kind of oil should be used on the com- mutator? Ans. Mineral oil. Oues. What attention should be given to the brushes? Ans. At certain intervals, according to the care taken to reduce sparking and the length of time the machine runs, the brushes will fray out or wear unevenly, and will therefore need trimming. They should then be removed from the brush holders and their contact ends or faces examined. If not truly square. CARE OF THE COMMUTATOR AND BRUSHES 637 they should be filed or clipped with a pair of shears, the course of treatment differing with the type of brush. If the machine be fitted with metal strip brushes, frayed ends should be clipped square with a pair of shears, the ends thoroughly cleaned from any dirt or carbonized oil, and replaced in their holders. Gauze and wire brushes require a little more attention. When their position on the commutator has been well adjusted and looked after, so that little or no sparking has taken place, it is generally only necessary to wipe them, clean the brushes and clip off the fringed edges and corners with the shears, or a pair of strong scissors. If, however, the machine has been sparking, the faces will be worn or burnt away, and probably fused. If such be the case, they will need to be put in the filing clamp, and filed true. Fig. 710. — Bistell brush gear. The brushes are held in the brushholders radially and work equally well with armature running in either direction. Brushes can be renewed and adjustment made while machine is in operation. A convenient method of trimming carbon brushes, or of bedding a complete new set of metal brushes, is to bind a piece of sandpaper, face outwards, around the commutator after the current has been shut off, and then mount the carbon or metal brushes in the holders, adjusting the tension of the springs so that the brushes bear with a moderately strong pressure upon the sandpaper. Then let the machine run slowlj'' until the ends of the brushes are ground to the proper form. Care should be taken, however, that the metal dust given off does not get into the commutator connections or armature windings, or short circuiting will result. If the contact faces of the brushes are very dirty and covered with a coating of carbonized oil, etc., it will be necessary to clean them with benzoline or soda solution before replacing. 638 HAWKINS ELECTRICITY Oues. Describe a filing clamp. Ans. As usually constructed, it consists of two pieces of metal, both shaped at one end to the correct angle, to which the brushes must be filed. One of the pieces of metal (the back part) has a groove sufficiently large to accommodate the brush, which is clamped in position by the other piece of metal and a pinching screw. If the clamp be not supplied with the machine a convenient substitute can be made out of two pieces of wood about the same width as the brush. One end of each piece is sawn to the correct angle, and the brush placed between the two. Fig. 711. — Jig for filing brushes to the correct bevel; used with copper brushes to fit them to the commutator. In filing, the brush is fixed in the clamp, with the toe or tip projecting slightly over the edge of the clamp, and the latter being fixed in a vise, the brush is filed by single strokes of a smooth file made outwards, the file being raised from contact with the brush when making the back stroke. Sparking. — In all well designed machines there are certain positions upon the commutator for the bnishes at which there will be no sparking so long as the commutator is kept clean and in good condition. In other dynamos, badly designed or con- structed, sparking occurs at all positions, no matter where the CARE OF THE COMMUTATOR AND BRUSHES 639 brushes are placed, and in such dynamos it is therefore impossible to prevent this no matter how well they arc adjusted. Oues. What two kinds of sparking may be generally distinguished? Ans. One kind of sparking is that due to bad adjustment of the brushes, and a second kind, that due to bad condition of the cominutator. Fig. 712. — Commutator clamp; a useful device for holding the segments firmly in position in taking out the end rings of the commutator to repair for internal grounds. It is made of 2 X J^ inch sheet steel, with a. H inch screw. The illustration clearly shows the adjustable fastening. The notches fit around rivets on one side of each fastening, which can be moved by removing the two cotterg. The clamp is made loose or taut by screwing the bolt in the nut. Sparks due to bad adjustment of the brushes are generally of a bluish color, small when near the neutral plane, and increasing in violence and brilliancy as the brushes recede from the correct positions upon the commutator. When sparks are produced by dirty or neglected state of the com- mutator, they are distinguished by a reddish color and a spluttering or hissing. When due to this last mentioned cause, it is impossible to sup- press the sparking until the commutator and brushes have -been cleaned. In the former case, the sparks will disappear as soon as the brushes have been rotated into the neutral points. Another class of sparks appear when there is some more or less developed fault, such as a short circuit, or break in the armature or commutator. 640 HAWKINS ELECTRICITY These are similar in character to those produced by bad adjustment of the brushes, but are distinguished from the lattter by their not decreasing in violence when the brushes are rotated towards the neutral plane. Ha\-ing distinguished the classes of sparks which appear at the commutator of a dynamo, it remains to enumerate the causes which produce them. These are : 1. Bad adjustment of brushes; 2. Bad condition of brushes; 3. Bad condition of commutator; 4. Overload of dynamo; 5. Loose connections, terminals, etc.; 6. Breaks in armature circuit; 7. Short circuits in armature circuit; 8. Short circuits or breaks in field magnet circuit. Bad Adjustment of Brushes. — When sparking is produced by bad adjustment of the brushes, it may be detected by rotating or shifting the rocker, by the indication that the sparking vnll vary with each movement. To obtain good adjustment of the brushes, it will be necessary to rock them gently backwards and forwards, until a position is found in which the sparking disappears. Oues. If, in rocking the brushes, a position cannot be found at which the sparking disappears, what is the prob- able cause of the trouble? Ans. The brushes may not be set with the proper pitch, that is they may not be separated a correct distance, or the neutral plane may not be situated in the true theoretical position upon the commutator through some defect in the winding, etc. In this last named case, the brushes may be strictly adjusted to their theoretically correct positions before starting the machine ; then, wheo C4RE OF THE COMMUTATOR AND BRUSHES 641 the machine is started and the load put on, violent sparking occurs, which cannot be suppressed by shifting the rocker. If, however, one set of brushes only be observed, it will generally be found that, at a certain position, the sparking at the set of brushes under observation ceases or is greatly reduced, while sparking still occurs at the other set. When this position is found, the rocker should be fixed by the clamping screw, and the brushes of the other set at which sparking is still occurring adjusted by drawing them back or pushing them forward in their holders until a position is found at which the sparking ceases. Correct position of the brushes and the suppression of sparking is a matter of importance, and any time spent in carefully adjusting the brushes will be amply repaid by the decreased attention and wear of the brushes and com- mutator. Figs. 713 to 715. — Brushes making bad contact. A brush making a bad contact, as only at the shaded portion of figs. 713 and 714, will not allow the short circuited coil enough time to reverse, causing sparking and heating. The latter will also result from bad contact on account of the surface being too small for the current to be carried off. This form of bad contact is worse than that shown in fig. 715, where the area of contact surface only is lessened. If the brushes do not make good contact, they should be ground down. Bad Condition of Brushes.- — If the contact faces of the brushes be fused or covered with carbonized oil, dirt, etc., there will be bad contact which is accompanied by heating and spark- ing. Simple examination will generally reveal whether this be the case. The remedy is to remove the brushes, one at a time if the machine be running, clean, file if necessary, trim, and readjust. If the brushes be exceedingly dirty, or saturated with oil, it will be necessary to clean them with turpentine, benzoline, or soda solution, before replacing. Bad Condition of Commutator. — If the surface of the commutator be rough, worn into grooves, or eccentric, or if there be one or more segments loose or set irregularly, the 642 HAWKINS ELECTRICITY brushes "vsill be throvi-n into ^-ib^ation, and sparking vriR result. A simple examination of the commutator will readily detect these defects. A rough and uneven commutator is due to bad adjustment of brushes, bad construction of commutator, and to neglect generalh', If allowed to continue, it results in hea\'>' sparking at the brushes, and the eventful destruction of the commutator. The fault may be remedied by filing or re-ttuTiing the commutator. Pig. 716. — Rou^ and grooved oommutator due to impnaper brush cidjustment and failure to keep brashes in pitiper ooDditaoa. Ques. How is an untrue commutator detected? Ans. If the commutator be untrue, the fact will be indicated when the machine is slowed down by a \-isible eccentricity, or b}' holding the hand, or a stick in the case of a high tension machine, against the surface while revohdng, when any irregu- larity or eccentricity will be apparent by the \-ibration or movement of the stick. The only remedy for an imtrue com- mutator is to re-tum it in the lathe. CARE OF THE COMMUTATOR AND BRUSHES 643 Oues. What should be done in case of high segments? Ans. They should be gently tapped down with a mallet, and if possible the clamping cones at the commutator end should be tightened. If it be impossible to hammer the segments down, they should be filed down to the same diameter as the rest of the commutator, or the com- mutator re-turned. For low segments, the only remedy is to pull out the segments, or turn commutator down to their level. Oues. Explain the term " flats on the commutator." Ans. This is the name given to a peculiar fault which develops on one or more segments of the commutator. It is not confined to dynamos of bad design or construction, but frequently appears on those of the highest class, and may be recognized as a " pit- ting " or " flattening " of one or more segments. Oues. What is the effect of flats on the commutator? Ans. Sparking at the brushes. Oues. What are the causes which produce flats? Ans. Periodical jumping of the brushes due to a bad state of the commutator, bad joint in the driving belt, a flaw, or a difference in the composition of the metal of the particular bar upon which it appears. But more frequently flats may be traced to a more or less developed fault, such as a break, either partial or complete, in the armature coil. The break may occur either in the coil itself, or at the point where its ends make connection with the lug of the commutator, or at the point where the lug is soldered to the segment. Oues. What should be done in case of flats? Ans. The brushes should be examined to see if any periodical vibration take place. If such be the case, the cause should be removed, the flat carefully filed or turned out, and the brushes readjusted. 644 HAWKINS ELECTRICITY If it be due to a difference in the composition of the metal of which the segment is made, the fiat will exist as long as the particular segment is in use, and will need periodic attention. With hard drawn copper or phosphor bronze segments, this fault is rarely due to this last mentioned cause. It is more frequently due to bad soldering, of the conductors to the lugs, or of the lugs to the segments. In all cases of flats, if the disconnection in the armature circuit be not complete, and cannot be readily located, the effect of re-soldering or sweating the ends of the coils into the lugs should be tried. Flats may also frequently be cured by drilling and tapping a small hole in the junction between the lug and the segment, and inserting a small screw, or bit of screwed copper or brass wire, afterwards filing down level with the surface of the commutator. I n J f,n ^ Figs. 717 and 71S. — Method of repairing broken joint between commutator segment and lug. To repair such a break push asbestos in between adjacent bars, so that heat from the torch will not affect them. Asbestos should also be worked in at the back if possible, for the purpose of keeping solder from places where it would cause trouble. Then unsolder the armature leads from the lug and remove the latter. Next, with specially made cape chisels, cut in a slot in the commutator bar for a new lug. Care and skill are required not to de- stroy the mica insulation between the segments. The slot should be cut one-quarter to three-eighths inch deep. The connector is then soldered in place. With care a satisfac- tory-connection can be made in this way, which will last well. If it do not last, the trouble in almost every case is due to poor soldering. Short circuits sometimes occur after this operation, because of solder falling in at the back and lodging on lower connections. In large machines, the excessive current flowing is quite likeljtto melt this solder, and the machine may buck, throwing out the melted solder, after which it may be all right again. While the bar connector is out, however, asbestos should be packed in back of it to pre- vent this occurrence, which may be a serious affair. All surplus solder and the asbestos packing should be removed after the connection is finished, and the connections cleaned with compressed air. The armature should be turned over slowly, air being applied all the while. Segments Loose or Knocked In. — When the segments are loose, it is an indication that the clamping ring or cone has worked loose. This should therefore be tightened up, and the commutator r©-tumed if necessary. CARE OF THE COMMUTATOR AND BRUSHES 645 Ques. How should low commutator segments be treated ? Ans. The commutator surface may be turned down to the level of the low segment, or the latter may be pulled out again to its former level, this latter being the preferable method, if it can possibly be effected. Oues. How is a commutator segment pulled out to its correct position? Ans. A hand vise is firmly clamped to the lug, or a loop of copper wire is passed round the conductor where it joins the commutator. A bar of iron, to act as a lever, is supported on a fulcrum over the commutator, and one end of the bar is passed through the loop or vise. Pressure is applied to the other end which will generally bury the segment up to its proper position. How to Re-turn a Commutator. — In re-turning the com- mutator, the armature should first be carefully taken out of the armature chamber, avoiding knocks or blows of any kind. The whole of the winding should then be wrapped in calico or canvas before the armature is put into the lathe, to prevent any particles of metal becoming attached to the surface of the armature at the time the commutator is being turned. The armature should on no account be rolled upon the floor, or subjected to blows or knocks while being put into the lathe. In re-turning the commutator, a sharp pointed tool should be used with a very fine feed. A broad nosed tool ought not to be used, as it is liable to burr over the segments. After turning, the commutator should be lightly filed with a dead smooth file, and finally polished with coarse and fine sandpaper. After the commutator has been turned and polished, the insulation between the segments should be lightly scraped with the tang of a small file to remove any particles of metal or burrs which might short circuit the commutator. 646 HAWKINS ELECTRICITY The points where the armature ■n"ires are soldered to the lugs should also be carefully cleaned with a brush, and should then receive a coat or two of shellac varnish. While the commutator is being turned, care should be taken that the setting marks for the adjustment of the brushes are not turned out if these be present. The same care should be used in putting the armature back into the armature chamber as was used in taking it out, otherwise the insulation may be damaged. Figs. 719 and 720. — Bissell commutators. The segments are of hard drawn copper and are insulated from each other and from the shell by mica. Ques. Should the commutator be run without any lubricant? Ans. In most cases it will be found that a little lubricant is needed in order to prevent cutting the brushes, cutting the commutator ; this is especially the case when hard strip brushes are used. The quantity of oil applied should be ver}^ small; a few drops smeared upon a piece of clean rag, and applied to the commutator while running, being quite sufficient. Ques. What kind of oil should be used on the com- mutator? Ans. Mineral oil, such as vaseline, or any other hydro- carbon. Animal or vegetable oils should be avoided, as they CARE OF THE COMMUTATOR AND BRUSHES 047 have a tcndenc}'' to carbonize, and thus cause short circuiting of the commutator, with attendant sparking. Overload of Dynamo. — It may happen, through some cause or other that a greater output is taken from the machine than it can safely carry. When this is the case, the fact is indicated by excessive sparking at the brushes, great heating of the armature Figs. 721 to 72.3. — Method of repairing a large hole burned in two adjacent bars of a com- mutator. Fig. 721 shows the hole. The first operation is to clean carefully and tin the sur- face of the hole. Thetwobarsare then wedged apart and mica strips, A B, fig 722, of the re- quisite size and thickness forced in. The commutator must now be warmed up as much as possible by means of soldering irons, and strips of mica, C D, E F, fig. 723, placed at the front and back of the hole, being kept in position by pieces of wood W, solder is poured into the hole from a ladle, using a rough mica funnel to guide it. and other parts of the dynamo, and possibly by the slipping of the belt (if it be a belt driven machine), resulting in a noise. The causes most likely to produce overload are: NOTE. — In operating dynamos having metal brushes, it is of importance to keep the commutator smooth and glossy. To accomplish this, it is necessary to keep the commutator and brushes clean and free from grit, and to occasionally lubricate the commutator with some light oil, such as ordinary machine oil. 1 his should be done da'ly if the machine be in constant use. Keep the brushes resting upon the commutator with just enough pressure to insure a good firm contact. This will be found to be much less than the springs are capable of exerting. A good method to follow i n cleaning the machine is as follows: Loosen the brush holder thumb screws and tilt the brushes off the commutator (or, if box brush holders be used, take them out of their holders). Then run the machine and hold a clean cloth against the commutator. After the commutator is clean, hold against it a cloth or piece of waste moistened with machine oil and reset the brushes. If for any reason the brushes begin to cut or score the commutator, it may be readily detected by holding the finger against the commutator; the ridge may be easily felt by the finger. This should be attended to at once in the following manner: Tilt back the brushes (or if box brushes are used take them out of their holders), and hold lightly against the commutator a piece of No. 00 sandpaper well moistened with oil, passing it back and forth until the surface is perfectly smooth. Then wipe off the commutator with a clean piece of cloth or waste and lubricate with another clean piece moistened with oil and reset the brushes. 648 HAWKINS ELECTRICITY 1 . Excessive voltage ; 2. Excessive current ; 3. Reversal of polarity of dynamo; 4. Short circtiits or groiinds in dynamo, or external circuits. Oues. What is the indication of excessive voltage? Ans. It is indicated by the voltmeter, or by the brilliancy of the pilot lamp. Fig. 724. — Method of smoothing commutator with a stone. The proper stone to use is made out of white sandstone similar to that used for grindstones, but a trifle softer. It is dove- tailed into a holder, as shown in the illustration, and h«ld in place by a set screw. 'When being used, one knob is grasped in one hand and the other knob in the other hand, the stone being moved back and forth along the length of the commutator. As the stone will become coated with copper at first, it must be cleaned freque.jtly ty means of coarse sandpaper. The fine dust from the stone will get under the brushes and wear them to a very close fit. After using the stone, finish with fine sandpaper. Oues. What are the causes of excessive voltage? Ans. Over excitation of the field magnet or too high speed. In the former case, resistance should be introduced into the field cir- cuit to diminish the current flowing therein if a shunt machine: or if a series machine, a portion of the current should be shunted across the field coils by means of a resistance arranged in parallel with the series CARE OF THE COMMUTATOR AND BRUSHES 649 coils; or the same effect may be produced in both cases by reducing the speed of the armature if this be possible. If due to excessive speed, which will be indicated by a speed indicator, the natural remedy is to reduce the speed of the engine driving the dynamo, or, if this be not easily done, insert resistance into the dynamo circuit, as described above. Ques. What are the causes of excessive current? Ans. If the dynamo be supplying arc lamps, the excessive current may possibly be caused by the bad feeding of the lamps. If this be the case, the fact will be indicated by the oscillations of the ammeter needle, and the unsteadiness of the light. If incandescent lamps be in the circuit, the fault may be caused by there being more lamps in circuit than the dynamo is designed to carry. Under such circumstances, another djmamo should be switched into circuit in parallel, or, if this be not possible, lamps should be switched off until the defect is remedied. When motors are in the circuit, sparking frequently results at the dynamo commutator, owing to the fluctuating load. In such cases the brushes should be adjusted to a position at which the least sparking occurs with the average load. Ques. What may be said with respect to reversal of polarity of dynamos? Ans. When compound or scries wound dynamos are running in parallel, their polarity is occasionally reversed while stopping by the current from the machines at work. Loose Connections, Terminals, etc. — When any of the connecting cables, terminal screws, eic., securing the different circuits are loose, sparking at the brushes, as a rule, results, for the reason that the vibration of the machine tends to continually alter the resistance of the various circuits to which they are connected. When the connections are excessively loose, sparking also results at their points of contact, and by this indication the 650 HAWKINS ELECTRICITY faulty connections may be readily detected. When this sparking at the contacts is absent, the whole of the connections should be carefully examined and tested. Breaks in .\rinature Circuit. — If there be a broken circuit in the armature, as sometimes happens through a fracture of the armature connections, etc., there will be serious flashing or sparking at the brushes, which cannot be suppressed by adjust- ing the rocker. As a rule it results in the production of " flats " upon one or more bars of the commutator. Fig. 725. — Sandpaper holder for commutator. The sandpaper is made fast on top by a clamp and screw. The two face blocks are pivoted and adjust themselves to the commutator, and will fit any size of commutator. If it have four brushes, the lower block vnU go in be- tween the brush-holders. Oues. How may such sparking be reduced without stopping the machine? Ans. By placing one of the brushes of each set a little in advance of the others, so as to bridge the gap. Short Circuits in Armature Circuit. — This fault is indi- cated by sparking at the commutator, and in bad cases by an excessive heating of the armature, dimming of the light and CARE OF THE COMMUTATOR AND BRUSHES 651 slipping of the belt, and in the case of a drum armature, by a sudden cessation of the current. Short Circuits or Breaks in Field Magnet Circuit. — Either of these faults is liable to give rise to sparking at the commutator. If one of the coils be short circuited, the fact will be indicated by the faulty coil remaining cool while the perfect coil is overheated. The fault may arise through some of the Fig. 726. — Saxidpaper block. It is made to fit the surface of the commutator. At S is a saw cut into which the ends are pushed after being wrapped around the block. The latter should be cut down on the dotted lines to form a handle. connections to the coils making contact with the frame of the machine or with each other. To ascertain this, examine all the connections, and test with a battery and galvanometer. A total break in one or more of the field coils may readily be detected by means of the battery and galvanometer. A partial break is not, however, so readily discovered, for the reason that the coil wires may be in sufficiently close contact to give a deflection of the galvanometer needle. The only methods of detecting this fault is by measuring the resistance of the coils with an ohmmeter or 652 HAWKINS ELECTRICITY "VMieatstone bridge, or by placing an ammeter in circuit with each coil in turn, and comparing the amount of current flowing in each. If the partial break be not accessible, the only way to remedy the fault is to rewind the coil, and the same applies to a break in the interior of the coil. Short Circuits in Cominutator. — These are of frequent occurrence, and result in heating the armature and sparking at the brushes. They are caused either by metallic dust ot particles lodging in the insulation between the segments, or b> the deterioration of the commutator insulation. To remedy, the insulation between the segments should be carefully examined, and any metallic dust, filings, or burrs cleaned or scraped out. When the commutator is insulated with asbestos or pasteboard (as is oftentimes the case in dj'namos of European make), short circuits ver}- frequently occur through the insulation absorbing moisture or oil, which is subsequently carbonized by the sparking at the brushes. In faults of this description the only remedy is to expel all moisture from the com- mutator insulation by means of heat, and scrape out all metallic dust which may be embedded in the surface of the insulation. If this do not effect a cure, it will be necessary to dig out the insulation, as far as pos- sible, with a sharp tool, and drive in new insulation. Oil should not be used on commutators insulated ^\ith these materials, but only asbestos dust or French chalk. HEA TING 653 CHAPTER XXXIV HEATING The excessive heating of the parts of dynamos and motors is probably the most frequent and annoying fault which arises in operation. When the machine heats, it is a common mistake to suppose that any part fotmd to be hot is the seat of the trouble. Hot bearings may cause the armature or commutator to heat, or vice versa. All parts of the machine should be tested to ascertain which is the hottest, since heat generated in one part is rapidly diffused. This is best done by starting with the machine cold; any serious trouble from heating is usually perceptible after a run of a few minutes at full speed with the field magnets excited. Heating may be due to various electrical or mechanical causes, and it may occur in the different parts of the machine, as in: 1. The connections; 2. The brushes and commutator; 3. The armature; 4. The field magnet; 5. The bearing. Oues. How is heating detected ? Ans. By api)lying the hand to the different parts of the machine if low tension, or a thermometer if high tension, and also by a smell of overheated insulation, paint, or varnish. 654 HAWKINS ELECTRICITY Oues. What should be done if the odor of overheated insulation, paint or varnish be noticeable? Ans. It is advisable to stop the machine at once, otherwise the insulation is liable to be destroyed. Oues. What is the allowable rise of temperature in a well designed machine? Ans. It should not exceed 80° Fahr., above the surrounding air, and in the case of the bearings, this temperature ought not to be reached under normal conditions of working. If this limit be exceeded after a run of six hours or less, it indicates a machine either badly designed and probably with the material cut down to the lowest possible limit with a view to cheapness, or some fault or other which should be searched for and remedied as early as possible, otherwise the machine will probably be destroyed. Oues. How should the rise of temperature be measured ? Ans. It is not sufficient to feel the machine with the hand, but special thermometers must be placed on the armature wind- ing, immediately on stopping the machine, covering them with cotton or wool to prevent cooling. Readings must be taken at short intervals, and continued till no further rise of temperature is indicated. Heating of Connections. — A rise of temperature of the connections may be due to either excessive current, or bad con- tacts, or both. The terminals and connections will be excessively heated if a larger current pass through them than they are de- signed to carry. This nearly always proceeds from an overload of the dynamo, and if this be rectified, the heating will disappear. If the contacts of the diflfcrcnt connections of the dynamo be not kept thoroughly clean and free from all grit, oil, etc., and the connections themselves be not tightly screwed up, heating will result, and the con- nections may even become unsoldered. HEATING 656 Heating of Brushes, Commutator and Armature. — When heating occurs in these parts, it may be due to any of the following causes: 1, excessive current; 2, hot bearings; 3, short circuits in armature or commutator; 4, moisture in armature coils; 5, breaks in armature coils; 6, eddy currents in armature core or conductor. Fig. 727. — Ventilated commutator; sectional view showing air ducts. Air is frequently circu- lated through a commutator in order to maintain it at a sufficiently low temperature, suitable openings being provided for this purpose as shown. Oues. What may be said with respect to excessive current? Ans. "When a dynamo is overloaded, the temperature of the armature will rise to a dangerous extent, depending upon the degree to which the safe capacity of the machine is exceeded, and heavy sparking of the brushes will also result. If the overload be not removed, the insulation of the armature may be destroyed. 656 HAWKINS ELECTRICITY Ques. State some causes of hot bearings. Ans. Lack of oil ; presence of grit or other foreign matter in the bearings; belt too tight; armature not centred with respect to pole pieces; bearings too tight; bearings not in line; shaft rough or cut. Fig. 72S. — Self-oiling and self-aligning bearing. The self-oiling feature consists of rings which revolve with the shaft, and feed the latter with oil continually, which they bring up from the reservoir below. The dirt settles to the bottom, and the upper portion of the oil remains sufficiently clean for a long time, after which it is drawn off, and a fresh supply poured in through holes pro^^ded in the top. These latter are often located directly over the slots in which the rings are placed, so that the bearings can be lubricated imme- diately by means of an oil cup if the rings fail to act or the reser\-oir become exhausted. The bearing is made self-aligning by pro\-iding the bearing proper with an enlarged cen- tral portion of spherical shape, held in a spherical seat formed in the pedestal by turning, milling, or by casting Babbitt or other fusible metal around it, thus allowing the bearing to adjust itself to the exact direction of the shaft. The upper half of the box can be taken off to facilitate renewal, etc., and to permit the armature to be removed. Oues. What is the eifect of hot bearings? Ans. Besides giving trouble themselves, the heat may be conducted along the armature shaft and core, thus gi\'ing rise to excessive heating of the armature. HEATING 657 POINTS RELATING TO HOT BEARINGS 1 . Use good oil ; 2. See that oil cups or reservoirs are full and all oil passages clear; 3. In self-oiling and splash systems where the oil is used over again, it should be kept in clean condition by frequent straining; 4. Keep bearings clean and properly adjusted; 5. Maintain bearings in good alignment; 6. Avoid tight belts; 7. Examine the air gap or clearance between armature and pole faces and see that they are uniform. Oues. What troubles are encountered with short circuits in the armature or commutator? Ans. This results in sparking at the brushes, and in the heating of one or more of the armature coils, and even in the burning up of the latter if a bad case. "When the armature is overheated, and the defect does not proceed from an overload or the causes mentioned below, the dynamo should be immediately stopped and tested for this fault. Oues. What will happen with an overheated com- mutator? Ans. It will decompose carbon brushes and cover the com- mutator with a black film, which offers resistance and increases the heat. Oues. What should be done if carbon brushes become hotter than the other parts? Ans. Use higher conductivity carbon. Reduce length of brush by adjusting holder to grip brush nearer the commutator. Reinforce brushes with copper gauze, sheet copper or wires, or use some form of combined metal and carbon brush. Increase size or number of brush if necessary, so the current does not exceed 30 amperes per square inch of contact. 658 HAWKINS ELECTRICITY Brushes heat sometimes due to too much friction. They should not press against the commutator more than is necessan.' for good contact. Ques. Give some causes for heating of armature. Ans. Eddy currents; moisture; short circuits; unequal strength of magnetic poles ; operation above rated voltage, and below normal speed. Ques. What trouble is encountered with eddy currents ? Ans. Considerable heating of the whole of the armature results, which may even extend to the bearings. Fig. 729. — Eck Manchester type motor. It is a ver\' small size unit and is designed for special purposes where very little room is available. The motor occupies a space of 2 J^" X 4 J^" between bearings and develops A horse power at 2.000 R. P. M. The frame of this motor is miade of high permeability steel so as to reduce the weight to a minimum. The armature is of the hand wound bipolar t\-pe built up of thin punchings. The armature, after being wound, is baked at high temperature for a prolonged period and then dipped while hot in insulating varnish. Pully is one inch in diameter and takes a }-4 inch round belt. Weight of motor 5 J-2 iMunds. Ques. How can this be overcome? Ans. There is no remedy for eddy currents other than the piirchase of a new armature, or reconstruction. The fault may be detected bj^ exciting the field magnets and running the machine on open circuit, with the brushes raised off the commutator for some time, when the armature will be fovmd to be excessively heated. HEATING 659 Ques. How does moisture in the armature coils affect the armature? Ans. The effect of this fault being to practically short circuit the armature, a heating of the latter results. In bad cases, steam or vapor is given off. Ques. What is the effect of short circuits in the arma- ture? Ans. It produces overheating. Oues. What trouble is likely to occur when the arma- ture is not centered in the armature chamber? Ans. A heating of the bearings is liable to be occasioned through the attractive forces developed by the center of the armature core not being parallel with the centre of the armature chamber or bore, or through the core being nearer one pole piece than the other. This may result from unequal wearing of the bearings, and therefore the bearings should either be relined or the bolt holes of the bearings readjusted, or the bearings packed up until the armature is correctly centered. Oues. What happens in case of breaks in the armature coils? Ans. This fault results in local heating of the armature, for the reason that resistance is interposed in the path of the current at the fracture. It always results in sparking at the brushes, and the heating being confined to the neighborhood of the break. Oues. What are the effects of operation above the rated voltage and below normal speed? Ans. Voltage above normal is a possible cause of heating, and operation below normal speed calls for an increase of field strength and reduces the effective ventilation, thus tending to cause heating. Gc.n HAWKINS ELECTRICITY Fig. 730. — Forced system of lubrication as applied to enxine of the generatiiig set showc in fig. 443. In engines empirying the forced system of lubncation the crank pit. which is formed by the columns, is accessible through doors in the front and back of the engine. The base of the engine forms an oil tank to which is attached a sniall plunger pump dn\ en by an eccentric on the shaft. The lubricant is carried under pressure to the various parts of the engine by the mechanism shown in the accom.panying diagram. The oil is forced by a pump to a grcjve in the main bearing, and a drilk-d hole in the shaft connects this groove with the crank pin. Ftcrn the crank i)in box the oil is further fcTced to the wrist pin through the pipe Fanning along the side of the conncciing red. The passage in the crosshead allows the oil to be foiced from the wrist pin to the guides. As the oil is forced from one bearing to another, it is quite important that the bearing caps be set light, otherwise the oil will escape before reaching the last bearing After passing through the bearings, the oil is collected in the base, strained and used again. The oil she uld be free from foreign substances, and to guar*' against the introduction of any foreign matter, a strainer, which may be taken uat for examination or cleaning, is attached to the suction valve of the pump. An oil pressure of from 10 to 20 lbs. should be maintained. and may be regulated by adjusting the set screw on the relief valve of the oiling syslem. The pressure gauge need not remain in the circuit continuously. Only mineral oils should be used for lubrication. A heavy oil gives better results and prevents knocking more effectively than thin oil. An nil n-hich has been found to give good resjlts, consists of two-thirds red engine oil and one-third heavy cylinder oil. .As the oil pas!«s through the bearings repeatedly, it pradually loses its lubricating properties, becoming thick and gritty, and should be occasi irially run through a filter and mixed with new oil. The frequency of this change dep>c-nds on the oil, as well as the number of hours the engine is in operatioti, and can easily be determine by observation. The oil in the resci-voit should stand about 2 inch?s over the suction and disch.irge valves, ai,d no water should be allowed to mix with it. Should any "vater accumulate in the base-, it should be drawn off by the cock provided for the purpose before starting the engine. HEATING 661 Ques. How may the field magnets become heated? Ans. By excessive field current; eddy current in pole pieces; moisture; short circuits. Ques. What may be said with respect to excessive field current? Ans. When heating results from this cause, all the exciting coils will be heated equally. It may be due to excessive voltage, in the case of shunt dynam.os; or to an overload in the case of compound and series dynamos. In either case it may be remedied by reducing the voltage or overload. If due to the coils being incorrectly coupled up, that is, coupled up in parallel instead of in series, it will be necessary to rectify the connections or insert a resistance in series. Ques. State the causes of eddy currents in the pole pieces? Ans. This fault may be due to defective design or con- struction of the armature. Slotted armatures are particularly liable to cause this fault, if the teeth and air gap be not properly ])roportioned. The defect may also be occasioned by variation in the strength of the exciting current. If due to this latter cause, it will he aecompauied by sparkinj; at the l)rushes. If a shunt dynamo, insert an ammeter into the shunt eircuit, and note if the deflection be steady. If this he not the case, the variation in the current most probably proceeds from imperfect contacts thrown into vibration. Ques. How is the insulation affected by moisture? Ans. Moisture tends to decrease the insulation resistance, thus in effect producing a short circuit with its attendant heating. Ques. How is moisture in the field coils detected? An?: It \^ easily detected by applying the hand to the coils, 662 HAWKINS ELECTRICITY when they will be found to be damp, and in addition steam or vapor will be given off where the machine is working. The fault may be remedied by drying and varnishing the coils. Ques. What is the indication of short circuits in the field coils? Ans. This fault is characterized by an unequal heating of the field coils. If the coils be connected in series, the faulty coil will be heated to a less extent than the perfect coils ; if connected in parallel, the faulty coil will be heated to a greater extent than the perfect coils. ' The former can thus be easily located. OPERATION OF MOTORS 663 CHAPTER XXXV . OPERATION OF MOTORS In operating motors of any considerable size, whether con- nected to the public supply mains of a central generating station for combined lighting and power service, or to power service mains only, there are certain precautions to be observed in starting, stopping, and regulating the motor, in order that the efficiency of the supply, and indirectly the working of other motors and lamps connected to the mains in the immediate neighborhood, may not be affected by abnormal variations of prespiire. These precautions should be observed also to prevent any danger of the motor itself being subjected to detrimental mechanical shocks and excessive temperatures in the working parts. Before Starting a Motor: — The general instructions relat- ing to inspection and adjustment, lubrication, etc., which have already been given, should be carefully followed preparatory to starting. Starting a Motor. — In starting a motor, resistance must be put in scries with the armature because, since there is no reverse electromotive force to counteract the applied voltage when the motor is at rest, the switching of the latter direct to the motor would result in an abnormal rush of current. This, in addition to being uneconomical and productive of a drop of voltage in 664 HAWKINS ELECTRICITY the mains, would injure all except the smallest motors. Hence motors above two horse power usually require a rheostat. Ques. Describe a rheostat or " starting box." Ans. It consists essentially of a suitable resistance to be inserted at starting to reduce the initial rush of current, and which can be cut out in sections by successive movements of a lever as the speed increases. Jpffla X ,, MiMi ^» I iifli-F- COMMUTATOR 3 ''■N ^ ARMATURE (- z ^ 1 .'/ Ji|il.l.WiH 61EEVE FOR PRESSING ON PIATE Bjfl Fig. 731. — Press for forcing on and removing a commutator. Small commutators are pressed on to the shaft by a hand press. All of the larger commutators are pressed on by means of a power press. In the above figure is shown a hand press. The plate B is used in removing old commutators. It is placed back of the commutator as at a; y with the slot C over the shaft. Bolts a b are passed through the holes in the plate and secured by nuts. I'he commutator can then be forced off the shaft. In pressing on a commutator, a sleeve is placed over the shaft at O, and against the commutator. The rear end of the shaft is secured so it will withstand the pressure, and the commutator is forced on. The power presses are built on the principle of a hydraulic press. In pressing on a commuta- tor a piece of babbitt metal or soft brass should be used against the end of the shaft. The shaft should be painted with white lead before having the commutator pressed on, in order to lubricate the shaft so that the commutator will press on easily. The wiper rings are pressed on after the commutator and then the armature is ready to be connected. Ques. Describe what occurs in starting a motor. Ans. When the lever of the starting box is moved to the first contact some of the resistance is cut out of the circuit and current flows through the motor. This produces a torque and starts the armature rotating. The movement of the armature induces a reverse voltage, which, as the speed increases, gradually reduces OPERATION OF MOTORS 665 the applied ctirrent. With this reduction of current, the torque is reduced and the speed not accelerated as quickly as at first. "When the applied current has been reduced to a certain value by the increasing reverse current, the handle of the starting box is moved to the next contact, and so on till all the resistance in the starting box has been cut out, the motor then attaining its normal speed. i|'- Figs. 732 to 735. — Various starting resistances. The type of resistance used in motor starting rheostats of small size consists usually of tinned iron wire wound on asbestos tubes, as shown in f\^. 732. the tubes being firmly supported by porcelain nipples, the ends of which fit into holes in the top and bottom of the enclosing case. In starters of larger size, cast metal grids, as shown in fig. 733, are used. In addition to these types of resistance, some forms of starter are equipped with what is known as "unit" type resistance. In this form, the resistance is built up of a number of separate sections, or units, which are con- nected to form the complete starting or regulating resistance as the case may be. A single unit consists of a moulded core of \ntreous material upon which is wound the resistance wire, as shown in fig. 734. The surface of the unit is then coated with a special cement and baked. By this method the resistance material is protected from mechanical injury and is also made proof against moisture and other conditions which sometimes affect the ordinary type of resistance. In_addition to units coated with cement only, there are still other types of units, as in fig. 735. which are pro\'ided with a sheet metal covering around the cement, as a further precaution against injury. Each of the various types of resist- ance described possesses certain characteristics not shared by the others, the use of any particular type being largely governed by conditions of service. Oues. What is the difference between a starting box and a speed regulator? Ans. Motor starting rheostats or " starting boxes," are 666 HAWKINS ELECTRICITY designed to start a motor and bring it gradtially from rest to full speed. They are vot intended to regulate the speed and must nc be used for such purpose. FaiJure to obsenr this caution will result in burning out thf resistance which, in a motor starter, is sufficient to carry the current for a limited time only, whereas in the case of speed r^ulators sufficient resistance is pro\-ided to carrj- the full load current continuoush*. Pig. 736. — View of Cutler-Hammer starter with slat€ front ^elno^•ed showing open wire coil resistance. The type of resistance here used consists of tinned iron wire wound on ad>estos tubes. The bottom of the casing is perforated to secure \'entilation. Oues. used? For what kinds of service are speed regulators Ans. In cases when the speed must be varied, as in traction motors, organ blowers, machine tool drive, etc. Oues. How long does it take to start a motor? Ans. UsuaDy from five to ten seconds. OPERATION OF MOTORS 667 Ques. How is the starting lever operated ? Ans. It is moved progressively from contact to contact, pausing long enough on each contact for the motor to accelerate its speed before passing to the next. Ques. What are the conditions at starting in a series motor ? Ans. There is a rush of current, the magnitude of which depends on the amount of resistance cut out at each movement of the starting lever. BUTTON CONTACTS RENEWABLE CONTACTS Figs. 737 and 738. — Sliding contact starters. Fig. 737, starter with button contacts; fig. 738, starter with renewable contacts. Motor starters in which the successive steps of resistance are cut out by a pivoted lever carrying a contact shoe which slides over button contacts or over contact segments, are known as sliding contact starters. Button contacts are usually furnished with motor starting rheostats of small size while contact segments are used on those of greater capacity. The contact segment being held in position by two screws, is readily renewable when worn by long service or damaged by arcing. The fixed button contact is not so easily renewed but being used only on small size starters is never likely to be subjected to severe service. Some starters, however, have renewable button contacts. Ques. How are small series motors started on battery circuits? Ans. By simply closing a switch to complete the circuit, the resistance of the battery being sufficient to prevent a great rush of current while starting. 668 HAWKIXS ELECTRICITY Oues. How is a shunt motor started? Ans. In starting a shtint motor, no trouble is likely to occur in connecting the field coils to the circuit. Since the resistance of the armature is very low, it is necessary on constant voltage circuits to use a starting rheostat in series with the armatiu-e. The necessary connections are shown in fig. 756. The switch is first closed thus sending current through the field coils, before any passes through the armature. The rheostat lever P is then moved to the first Figs. 739 and 740. — Multiple switch starters. Fig. 739, starter with no voltage release; fig. 740, starter with no voltage release and circuit breaker. The multiple switch type of starter is designed to overcome the arcing on sliding contacts which, in the case of large motors would be very severe. The cutting out of each step of resistance is accomplished in the multiple switch starter by a separate carbon contact switch which breaks the circuit with a quick snappy action. contact to allow a moderate amount of current to pass through the armature. The resistance of the rheostat is gradually cut out by further movement of the lever P, thus bringing the motor up to speed. Oues. How does the reverse voltage affect the starting of a motor? Ans. When a motor is standing still, there is no reverse voltage, and the ctirrent taken at first is governed principally by the resistance of the circmt. If the motor be series wotmd, there is a momentary reverse voltage, due to self-induction while OPERATION OF MOTORS 669 the field is building up. If the motor be shunt wound, self- induction delays the current through the field coils, but that through the armature is not impeded by such cause. When the armature begins to revolve, reverse voltage is developed which increases with the speed. The resistance of the starting box may be gradually cut out as the armature comes to speed. Thus the reverse voltage gradually replaces ohmic drop in limiting the current as the motor comes to speed. Fig. 741.— Starting rheostat with no voltage and overload release. The no voltage release permits the starting lever to fly to the "off position" should the voltage fail momentarily, thus protecting the motor against damage should the voltage suddenly return to the line. The movement of the lever is due to a spring. The overload device causes the lever to back to the off position should the current exceed a predetermined ma.ximum for wliich the release is adjusted. Fig. 742. — Compound starter. Rheostats designed for the double duty of starting a motor and regulating its speed are commonly known as compound starters, the resistance pro- \'ided being a combination of armature resistance for starting duty and shunt field resistance for speed regulation. Failure to Start. — This fault, which is liable to occur in a motor of any description, is similar to failure to excite in a dy- namo, and is liable to be produced by any of the causes mentioned in connection with the latter fault, excluding insufficient speed. and insufficient residual magnetism. When a motor fails to start, it should first be ascer- tained if a supply of electrical energy be available in the 670 HAWKINS ELECTRICITV mains. This may readily be discovered by means of a voltmeter, or if low tension service, by means of the fingers bridging across the main terminals. If the supply of energy be present, the contact arm of the starter should be moved into such position that all resistance is inserted into circuit with the motor. This is important, as the motor may start suddenly while trying to ascertain the cause of the stoppage. Fig. 743. — Starting panel. In installing any kind of motor starting rheostat, it is necessary to proN-ide main line knife switch and luses in addition to the starting box. The appearance of the installation can be much improved by mounting all of these upon one panel. Having closed the switch, if the motor fail to start, it will be advisable to remove the load if possible, as the failure may arise from an overload of the machine. This being effected and the motor not starting, the terminals of the latter should be tested by the means already described for voltage. If no voltage be OPERATION OF MOTORS 671 generated, a broken circuit or a defective contact may be looked for in the main fuse, switch, or starting box. The resistance coils of the latter, through the heat developed, frequently break in positions out of sight. If a defective contact of this nature cannot readily be seen, the contact arm should be moved slowly over the contacts, as it is possible the broken coil may be cut out of circuit by this means. If a difference of pressure exist between the motor terminals, the field magnets will, if shunt or compound wound and in good order, be excited, which may be ascertained by means of a bar of iron. If no magnetism Figs. 744 to 746. — Cutler-Hammer motor starting rheostats with no voUage and overload release. Fig. 744, starter with fixed button contact, fig. 74.T, with renewable button contact, and fig. 74G, with contact segments. In construction the resistance is enclosed in a pressed steel box on which is mounted a marbleized slate panel carrying the starting lever, contacts and protective devices. By means of a calibrated scale, the overload release (shown in the lower left hand comer, figs. 744 and 74.5, and in the lower right hand comer fig. 746) can be set to break the circuit on any overload not exceeding .50 per cent, of the rated capacity of the motor. This calibrated scale can also be used for determining, with a fair degree of accuracy, the amount of current being consumed by the motor. be present, it will of course, indicate a broken or bad connection, either between the terminals of the field coils, or one or more of the coils them- selves. If the bar pull strongly, the position of the brushes upon the commutator in regard to the neutral points should be ascertained, and the rocker adjusted, if necessary, to bring them into their correct po.si- tions. If this fail to start the motor, the connecting leads from the motor terminals to the brushes and the brushes themselves should be carefully examined for broken or bad connections, and defective contact of tbi 672 HAWKINS ELECTRICITY brushes wilh the commutator. In the latter case, it may arise from a dirty state of the commutator, or from the brushes not being fed properly. If due to these causes, pressing the brushes down upon the commutator with the fingers will probably start the motor. If the failure to start arise from none of these causes, it is probably due to the field coils acting in opposition, or to a short circuited armature. This latter remark applies more especially to motors provided with drum armatures. Fig. 747. — Allen-Bradley compression type resistance units. The contact resistance between the discs composing the resistance column is suliject to variations of pressure, thereby producing proportionate resistance changes in the column as a whole. In the complete resistance unit, the resistances column is encased in a drawn steel tube,which is lined with a highly refractory cement, for purpose of insulation, fl/TorJjjz!,' ^/;« column both mcchanicdl and elec- trical protection and excluding the air which effectually prevents any combustion should the column l:>ecome red hot due to overload. The ends of the tube are closed by means of caps through which pass electrodes for making connections between the discs and exterior conductors. The steel tube, when necessary, is provided with ribs or fins for the dissipation of acquired heat. Precautions with Shunt Motors. — With motors of this type, because of the large amount of self-induction in the shunt wind- ings, it is important to note: 1, that in switching on the field magnet, the current may take an appreciable time to grow to its OPERATION OF MOTORS 673 PiG. 748. — Allen-Bradley type Z automatic motor starter. The operation of this machine is as follows: When the main switch is closed, the motor circuit is made through the fuses, resistance unit, current relay, and the motor armature. At the same time, the solenoid circuit is closed (this is connected directly across the line, and takes a current which is a small fraction of an ampere), and the plunger of the solenoid is drawn up, which produces a pressure on the resistance unit, and increases the current in the motor circuit to the predetermined value at which the current relay is set. When this value is reached the current relay operates and opens the solenoid circuit, which reduces the magnetic pull and allows the solenoid plunger to drop back slightly. This action increases the resistance in the motor circuit, which decreases the current sufficiently to allow the relay to close again. Similar cycles of operation are repeated as the motor accelerates, and each time the plunger is drawn a little farther into the solenoid, until the short circuiting contacts on the top are pushed together, which short circuits the current relay and resistance unit, making them inoperative, and completing the operation of starting the motor. It will be noted that in starting a motor with this device the current is always held down to a certain predetermined value, and it is impossible to overload the motor by too rapid starting. The current relay is calibrated in amperes, and may be set to suit existing conditions. The action of the starter being controlled by a current relay and not by an oil or air dash pot. the motor will start rapidly when under a light load, and slowly when more heavily loaded. The fuses or circuit breakers may be set at a value that will furnish protection to the motor upder running conditions. 674 HAWKINS ELECTRICITY normal value, and 2, that in s\\ntching off, especially wnth quick break swntches, high voltages are induced in the VN-indings, which may break down the insulation. Fig. 749. — Monitor starter giving automatic start with knife switch control; designed for use with printing presses. It consists of a set of solenoids connected in series and so inter- operating as to cut reastance out of the armature circuit of the motor as the apparatus it is dri\-in(j comes up to speed. This tj-pe is for small motors or where no need arises for speed regulation; there is. therefore, no adjustment of speed possible aside from an actual chajige in motor conditions. At full speed the motor is directly across the main supply line. Fig. 750. — Monitor automatic starter, equipped with relay for push button control. Oues. What provision is made so that the magnetizing current will have time to reach its normal value? Ans. The field connections are generally separated from the actual starter, and taken to the main switch, so that wherever the main switch is closed, the current flows through the field coils, before the starting lever is moved. OPERATION OF MOTORS 675 Ques. How are the connections arranged to avoid excessive voltage in the windings due to self-induction? Ans. Generally the armature and field magnet circuits are placed in a closed circuit that is never opened. In other cases, in order that the rise of voltage may not injure the insulation when the shunt is opened, a special form of main switch is sometimes used which, before breaking from the suppl}', puts a non- inductive resistance across the shunt of the motor. This is known as a flashing resistance. Pigs. 751 to 753. — Monitor control switches. Fig. 751, push button "start" and "stop" switch; fig. 752. safety lever control switch with "slow" and "fast" buttons for rotary- printing presses. This device will upon pressure of the "start" button, set the machine in motion and bring it up to the predetermined speed, either as previously set by the starter limits or by the setting of the rheostat arm. The stop button projects some dis- tance beyond any other portion of the device, in order that in case of emergency the operator may stop the machine merely by hitting the face of the switch with his open hand. The lever control switch, fig. 753, is similar in its action to the push button switch but combines two other features: locking point, and visual indication of the station from whence the press has been stopped. With the lever at the downward position, the press is locked and cannot be started from any other station. In order to make the press ready to start the lever must be raised to the central position. Thus a rnan may safely enter the press without delay by setting his station to the locked position, knowing that it cannot be started except by some one coming to that station and realizing that the press has been purposely locked. Also, by looking along the press, it is possible to tell from which station it has been locked, and proper action can be immediately taken. The safety control station is usually combined for use on large rotary presses with the "slow" and "fast " push buttons as shown in fig. 752. A pressure upon the fast or slow buttons will cause the press speed to be correspondingly accelerated or retarded, and this action will continue so long as the button is pressed. The press continues to run at the speed attained at the instant of releasing the button. Any speed may, therefore, be selected or changed to suit momentary requiremente- This gives complete control excepting reversal which is not required of such a pras& 676 HAWKINS ELECTRICITY Fig. 754. — Wiring diagram of the standard fonn of Monitor controller. A set of solenoids are connected in series and so interoperating as to cut resistance out of the armature circuit of the motor as the apparatus it is dri\-ing comes up to speed. The controller is designed to be used on all classes of work. In its simplest form, a single copper and graphite contact, is controlled by two magnets, so proportioned as to cut out the entire starting resistance when the armature current falls to a predetermined value. In the larger sizes, the number of steps controlling the resistance is increased and arranged to produce the correct acceler- ation. In every case the regulation of the starting resistance is effected entirely by the current passing to the motor without the use of dash pot, mechanical governor or delicat« cut outs. The time element is thus directly proportioned to the load and the motor brought up to speed in the shortest time consistent with the load,. but always with safe limitation of the armature current. The distinction between the current controlled starter and the starter with dash pot governor should be noted. The starter here shown limits the start- ing current to a fixed value throughout the Starting operatioa. which is an ideal oandition and prevents blowing ftises in starting. OPERATION OF MOTORS 677 Oues. How can shunt motors be controlled from a distant point? Ans. The starter and switch are placed at the desired point and the two main wires and the field wires run from that point to the motor. This requires additional wire which increases the cost and Hne loss. Regulation of Motor Speed. — Motors are generally run on constant voltage circuits. Under these conditions, the speed of series motors varies with the load and at light loads becomes excessive. Shunt motors run at nearly constant speeds. For many purposes, particularly for traction, and for driving tools, it is desirable to have speed regulation, so that motors running on constant voltage circuits may be made to run at different speeds. The following two methods are generally used for regulating the speed of motors operated on constant voltage circuits: 1. By inserting resistance in the armature circuit of a shunt wound motor; 2. By varying the field strength of series motors by switch- ing sections of the field coils in or out of circuito Oues. Describe the first method. Ans. This method is illustrated in fig. 756. When the main switch is closed, the field becomes excited, then by moving the lever P of the starting rheostat the various contacts (1, 2, 3, 4, 5), more or less of the rheostat resistance is cut out of the armature circuit, thus varying the speed correspondingly. This is the same as the method of starting a motor, that is, by variation of resistance in armature circuit, but it should be noted that when this method is used for speed regulation, a speed regulating rheostat should be used instead of the ordinary starting box, because the latter, not being designed for the purpose, will overheat and probably burn out. 67 ^ UA WKINS ELECTRICITY Fig. 755. — Monitor printing press controller. It provides var: features required in the operation of large rotarj,- presses, such ; newspapers. From any one of various stations' amilar to the c located at all desirable places about the press, the latter mj;.- 'b^e ated, slowed down or locked. It differs from other i>-pei that the solenoid has an overall marimum pull of less i'hazi the main line current directly but through pilot circuits, ■ switches; there are no sliding contacts. At the control sta::-?, guish the accelerating button from the retarding button by the ously he can in the same manner ascertain the f^: ?■•• ;n f the ';■ ■"ever whether at start, stop or safety, can be r lever of either control station is placed at stop. - and a powerful d>Tiamic brake brings the pre> or harmful strain. The start will always be • circuit, and with fuU field, and should the cu— and open the circuit to the motor. This contr: of normal speed by annatare resistance and. by ztli co:::r;:, speed of the motor. OPERATION OF MOTORS 679 Ques. Describe the second method. Ans. This niclliod of regulating the speed of a series motor is shown in fig. 757. The current through the armature will flow through all the field windings when the position of the switch lever S, is on contact 4, and the strength of the field will be the maximum. By moving the arm to contact 3, 2 etc., sections of the field winding are cut out, thus reducing the strength of field and varying the speed. OM^ Fig. 756. — Spead regulation of shunt motor by variable resistance in the armature circuit. Ques. How does the speed vary with respect to variation of field strength ? Ans. Decreasing the field strength of a motor increases its speed, while increasing the field strength decreases the speed. Under the conditions of maximum field strength, as with switch S on point 1, the torque will be greatest for any given current strength and the reverse voltage also greatest at any given speed. The current through the armature of the motor, to perform any given work, will thus be a C80 HA WKIXS ELECTRICITY Pig. 757. — Speed regolatiaa of series motor by cntthig oat sections of the field winding, bi this method the fidd winding is tai>ped at several points, divi^ng the oofl into sections and the leads from these points are connected a nnilti-point switdi of the type that woold be nsed on a rheostat. By moving the lever S, to the I^ or right, the corrent will flow throng one or more sections of the field winding, thus decreasing or increasing the ampere turns and therdiy providincr means of regulation. NOTE. — .\ compound motor may be made to run at constant speed, if the current in the series winding of the field be arranged to act in opposition to that of the shunt winding. In such case, an increase of Ic^id will weaken the fields and allow more current to flow through the armature without decreasicg the speed of the armature, as would be necessary in a shunt motor. Such motors, however, are not very c^en used, since an overload would weaken the fields too much and cause trouble. If the current in the series field act in the same direction as that in the shunt fields, the motor will slow up some when a heavy load comes on, but will take care of the load without much trouble. NOTE. — Motors have much the same faults as dynamos, but they make themselves manifest in a different way. An open field circuit will prevent the motor starting, and will cause the melting of fuses or burning out of the armature. A short circuit in the fields, if it cut out only a part of the winding, will cause the motor to run faster and very likely spark badlv. If the brumes be not set exactly opposite each other, there will also be bad sparking. If they be not at the neutral point, the motor will spark badly. Brushes should always be set at the point of least sparldng. If it become necessary to open the field circuit, it should be done slowly, letting the arc gradually die out. A quick break of a circuit in connection -xith any dynamo, or motor is not advisable, as it is very likely to break down the insulation of the machine. The ordinary starting box for motors is wound with comparatively fine wire and will get very hot if left in circuit long. The movement of the arm from the first to the last point should not occupy more than tUrty seconds and if the armature do not begin to move at the first point, the ann should be thrown back and the trooble located. OPERATION OF MOTORS 681 minimum, as well as the speed at which the motor has to run, in order to develop sufficient reverse voltage to permit this current to flow. Regu- lation of speed by varying the field strength is limited in range of action, since the field saturation point is soon reached, moreover, with too low a field strength, armature reaction produces excessive field distortion, sparking, etc. Tj Earth /iG. 758. — Speed regulation of a series motor by the method of short circuiting sections of the field winding. It will be seen that there are seven different positions for the contact springs on the barrel contacts. A. represents the armature and brushes, little A, B, and C. the divided field magnet coils, L the line connection, and G the earth connection. The diagram shows the connections for trolley car operation. Oues. How is the speed of shunt and compound motors varied with respect to the normal speed in the two methods? Ans. The first method (variable resistance in armature circuit) reduces the speed below the normal or rated speed of the machine, while the second method increases the speed above the normal. In the first method the amount of speed reduction depends partly upon the amount of resistance introduced into the armature circuit, and partly upon the load. In the second method the amount of speed increase depends entirely upon the amount of resistance placed in the shunt winding circuit. 682 HAWKINS ELECTRICITY Eighty-five per cent, is about the maximum speed reduction obtainable by armature resistance but so great a reduction is seldom satisfactory since comparatively slight increases in the load will cause the motor to stall. Shunt field regulation may be obtained up to any point for which the motor is suited, the only limitation in this case being the maximum speed at which the motor may be safely operated. It should be remembered, however, that speed increase by shunt field weakening increases the current in proportion to the increase in speed, and care should be taken not to overload the armature. Fig. 759. — Cutler-Hammer multiple switch starter with no voltage release; for use with large motors, or with motors of medium size where the starting conditions are severe or when more than fifteen seconds are required to accelerate the motor. In operation, the cutting out of each step of resistance is accomplished by a separate lever and the levers them- selves are so interlocked as to prevent closing switches except in proper order, beginning with the lever on the left. The last switch (the one on right hand side) is held by an electro- magnet w^hen closed, each of the other switches being held in the closed position by a latching device on the switch next to it. In front of each switch is placed a metal stop, so arranged as to prevent any switch being operated tmtil the one next to it on the left has been closed. These metal stops constitute the interlocking mechanism and prevent the starting of the motor in any way except by closing the switches in regular rotation, thus insuring proper resistance in the circuit and protecting the motor from excessive starting currents. When the current is interrupted, the electro-magnet releases the last switcli, which, on opening, releases the latch on the switch next to it, allowing that switch to open, and this in turn releases the next latch and so on, the switches opening automatically one after another. In starting the motor, each switch should be closed quickly and firmly, pausing a second or two before closing the next switch to give the motor time to accelerate. N'OTE. — In starting a motor, first see that the bearings contain sufficient oil and that the brushes bear evenly on the commutator. If a circuit breaker be used, close it; then close the main switch. Rotate slowly the handle of the starting rheostat as far as it will go. Care should be taken, in starting the motor, that the handle of the rheostat be not rotated too fast. To stop a motor, open the circuit breaker or switch, which wall cut in the resistance of the starting box. Never attempt to stop a motor by forcibly pulling open the starting box. Disregard of these instructions may cause burning out of the field coils. OPERATION OF MOTORS 683 Ques. How is a wide range of speed regulation secured? Ans. By a combination of the two methods. Regulation by Armature Resistance. — Speed regulators for this method of regulation, are designed to carry the normal current on any contact without overheating and when all the resistance is in the circuit, they will reduce the speed of the motor Fig. 700. — Cutler-Hammer speed regulator with no voltage release, regulation by armature resistance only, reducing speed of motor below normal. No resistance in the armature cir- cuit. No provision is made in regulators of this type for increasing the speed of the motor. The maximum speed obtamable when these regulators are used is, therefore, the normal speed at which the motor is designed to operate with no resistance in circuit. With all resistance in circuit and the motor taking normal current these regulators will reduce the speed of the motor .50 per cent. If the motor be taking less than normal current the per- centage of speed reduction obtainable will be correspondingly less. The notched fan tail extension on the lower end of the lever engages with a magnetically operated pawl to hold the lever squarely on any contact so long as the no voltage release magnet is energized. about 50 per cent, provided the motor be taking the normal current. When operating without resistance in the armature circuit, shunt wound and compound wound motors will regulate to approximately constant speed regardless of load. This 684 HAWKINS ELECTRICITY characr eristic of inherent regtilation is lost, however, when armature resistance is employed to reduce the speed of the motor, fluctuations in load resulting in fluctuations in speed, which become more noticeable as the amount of resistance inserted in the armature circuit is increased. Accordingly, it becomes necessary to move the lever of the speed regulator forward or backward to again obtain the speed at which the machine was operating before the load changed. Pig. 761. — Cntler-Haimner compound starter with no voltage and overload release. This is a staiting riieostat and field regulator combined. In operation, two levers are employed, both being mounted on the same hub i>ost and one lying directly under the other. The upper lever only is pro\"ided with a handle, but when moving from the off position to the starting position (that is to say, from left to right) the lower, or starting, lever is carried along by the upper, or speed regiilating, lever until it comes in contact with the no voltage release magnet where it is held fast by the attraction of the magnet, lea\'ing the upper lever free to be njoved backward over the field contacts, thus weakening the shunt field of the motor little by little until the desired speed is attadned. During the operation of start- ing the motor, the f.ejd resistance is short circuited by an auxiliary contact (the slotted metal strip shown near center of rheostat) but as soon as the starting lever touches the no voltage release magnet or. In other words, when the motor has been accelerated to normal speed, this short circuit is removed, and the field resistance becomes effective for speed regulation. The motor is accelerated from rest to normal speed by moving both levers from left to right, while increases in speed above normal are obtained by moving the upper lever from right to left. Only the lower, or starting lever comes into contact with the no voltage release magnet. This lever ispro\-ided with a strong spiral spring which tends always to throw the lever back to the off-position. Hence should the voltage fail, the no voltage release magnet releases the starting lever and this, in flying back to the off position, opens the armature circuit of the motor and carries the speed regulating lever with it to the off position. The upper, or speed regulating le\'er, not being influenced by the spring, though mounted on the same hub post as the starting lever, may be mov^ back and for^ »t will, or left vndefiajt«ly ia the position which gives the spe«4 desired. " ' ' OPERATION OF MOTORS 685 When the speed of a motor driving a constant torque machine is reduced by inserting resistance in the armature circuit there is no corresponding reduction in current consumed. The motor runs more sloA^ly simply because a part of the energy impelling it is shunted into the resistance and there dissipated in the form of heat. Hence, whether the motor be operatin*^ at fall SDced or half speed, the amount of current consumed is the same; the only difference being that in the one case all the energy taken from the line is expended in driving the motor while in the other case only one half is utilized for power, the other half being dissipated in the resistance. Speed regulation by armature resistance only is therefore open to two objections: 1, the difficulty of maintaining constant speed under varying load conditions, and 2, the necessity of wasting energy to secure speed reduction. These objections are, in part, Fig. 762. — Cutler-Hammer compound speed regulator with no voltage and overload release; regulation by combined armature and shunt field resistance, designed to both decrease and increase the speed of a motor. Speed reduction is accomplished by inserting resistance in the armature circuit, the maximum amount of speed reduction obtainable with these con- trollers being .50 per cent, below normal. Speed increase is obtained by inserting resistance in the shunt field circuit, the maximum amount of speed increase obtainable with these controllers being 25 per cent, above normal. offset by the fact that speed reduction by armature resistance may be applied to any motor of standard design and requires nothing more than the simplest and least expensive speed regulating rheostat. In cases where the motor will be operated nearly always at full speed, the difference in first cost of the installation may justify the use of the armature resistance method of control. As a rule, speed regulation by shunt field resistance is preferable. Regulation by Shunt Field Resistance. — Since regulation by his method is for speeds above normal, a starter must be used to bring the motor up to its rated speed. Usually the starter is 686 HAWKINS ELECTRICITY v c iJ o o - a =< - ° w 3 S ^ C O iiX o c Hi! ° .^■E -- = 3 = -7;^-5 3 O " -DO 3 -■£ " 5 " t^ 2 = p 3 — j: •' o c " i^ '^ 3 3 > V- ; -^ -J > >, o -i2 r'^ 5J:n -^ u S<--5 3 30- V^S o ;i 5 . - o -^ c~oS=t P^ £v •- ^ ^5^-5 >.t: w o £ c 0.5 2 > t a o^ F u -^ ^ cj-'r '^S t:ii 3 i 03 * £.2 „ o "10 , .2 ..J O S ^itj. 3 5 c o a-g 9 ^ ir M « 5 o "« O .J ■" J2-= Hu.-^eo UrSaoS- E = « = -^ z-e a: o ■3 fc 3 2 — t) a 1^ - J? o -.t: .2 m2 ■?^ 2° O V O u zi o > ^ £ ^ 3 a - zj r- ■" X.Z < -J f" a V5 30 C C OPERATION OF MOTORS 687 I'-'Star™ nl„fe.. -.0) ^¥ H "I •O rt .' •fi-fl j2:=-a E2 o^ I II 3-. g| g pS.^ :-£ 1 = 1 I .a-£c>"0 g "o Sl^fi ^1" -^ ""-"Erg Mic ao--^ % §:- §! I i-i III p s ^.^1 I .^2 §|i| « i^l -g^ % g« °.S° ^-^ -Sir ^ ^K^si d- 8|st3 •g^ c-2-..^ .0 ^^ 1^ ^g ^ «^ 2|a^.= "r"^ fe.5 J? '"' — O'C ° S J8£§- i!"o S&c-^ O'^ ^§>> .Sfe'«:S°« °S ES->2,_- J-^-li IM-il It'll § fe-l-T^-^"1- -I fe^«« iiii^^lljtlrJii! fe f ^'-lli iJ. Ill :^ t«o«;iiS p-Pf-H '-;.^i«°'M2 p-" I 3"!.cort^ Oi?S Og>.«3^ jA '^'t' oj C ■*-' . aj tt> 4* fl> )=('*-< "^ 688 //.4]rA'/.V5 ELECTRICITY OPERATION OF MOTORS 689 combined with the regulator, as shown in fig. 761, the device being called a compound starter. The weakening of the shunt field of a motor by the insertion of resistance in the shunt field circuit causes the armature to revolve more rapidly. One advantage of this method of control Fig. 707. — Cutler-Hammer reversible starter with no voltage release, adapted to start and operate motor at full speed in either direction, such for instance as motors driving auxiliary motions on lathes, planers and other machine tools which may rotate in either direction but always at constant speed. They are not designed to reduce the speed of the motor, but merely to start it and bring it smoothly up to full speed in either direction. Two no voltage release latching devices are provided so that the lever will be held in the full speed position in either direction so long as the voltage of the line remains constant. On failure of voltage a strong centering spring attached to the hub-post of the lever throws the latter to the central, or off position. The shunt field circuit is not opened by starters of this type. is that the motor will inherently regulate to approximately con- stant speed under widely varying load conditions. Another advantage is found in the fact that all of the current taken from the line is utihzcd for power, the changes in speed being obtained, not by dissipating a portion of the effective energy in the re- sistance (as in the case of the armature resistance method of 690 HAWKINS ELECTRICITY control) but by weakening the reverse voltage by inserting resistance in the shunt field circuit. Speed increase by shunt field weakening is limited, however, to about 10 to 15 per cent, above the normal speed in motors of standard construction. Greater ranges of speed can be obtained from motors especially designed for shunt field control but should not be attempted with motors of standard design \\'ithout first ascertaining from the manu- facturer the maximum safe speed. Combined Armature and Shunt Field Control. — Regu- lation by combined armature and shunt field resistance is by far the easiest way of obtaining a wnde range of speeds. Rheostats embodying these methods are known as compound speed regu- lators, one form being shown in fig. 762. Standard regulators can be obtained giving a wide range of speed variation, and special regulators may be constructed giN'ing practically any desired range. Selection of Starters and Regulators. — Unsatisfactory operation of these devices is, in nearl}- all cases, due to lack of precaution in selecting the proper piece of apparatus for the work to be done. One of the commonest errors is to select a rheostat of insufficient capacity. If the current required to operate the motor at full speed with no resistance in circuit be greater than the rated capacity of the rheostat, overheating of the resistance will result. An increase in temperature even to a point where the hand cannot be held on the enclosing case need cause no apprehension, but should the resistance become red hot it indicates that the apparatus is being worked far beyond its capacity, and the load on the motor should be reduced or a regulator of greater capacity substituted. If the current required to operate the motor at full speed with no resistance in circuit be less than the rated capacity of the OPERATION OF MOTORS 091 rheostat no overheating will occur, but it will not be possible to secure the full 50 per cent, speed reduction the rheostat is de- signed to give with all resistance in circuit. In ordering a starter or regulator the manufacturer should be furnished with the following information : 1. Horse power of motor with which speed regulator will be used; 2. Voltage of motor; ""IG. 768. — Various sizes of Watson commutator. The segments are punched from hard drawn copper strip and are insulated from each other and the core by amber mica, of hardness corresponding to that of the copper in order that the wear of mica and copper may be imiform. The segments are assembled in a ring under great pressure and are repeatedly heated and tightened, being finally secured and rigidly locked together. 3. Winding of motor, whether series, shunt, or compound wound; 4. Nature of the machine which motor is to operate; 5. Normal rated speed of motor to be used; 6. Maximum speed at which it is desired to operate the motor ; 7. Minimum speed at which it is desired to operate the motor ; 8. Whether controller will ever be required to reverse direction of motor or to operate it in one direction only ; t)92 HAWKINS ELECTRICITY 9. If reversible controller be desired, whether or not full range of speed control is required in both directions ; 10. Whether the regulator shall be equipped with any of the following devices: no voltage release, overload release, knife switch, fuses; 11. Whether button contacts or renewable contact segments are preferred; A Fig. 709. — Organ blower speed regulator; diagram showing operation and method of installing. A cord running from the top of the organ bellows passes over two pulleys and is then made fast to the chain furnished with the regulator. This chain passes around a sheave which turns on a post projecting from the center of the slate panel. Attached to the lower end of the chain is a weight, also furnished with the regulator. As the air is exhausted from the bellows the latter slowly collapses, drawing the rope down with it, and in so doing turns the sheave from left to right, thus cutting resistance out of circuit and increasing the speed of the motor which pumps air into the bellows. Responding to the inrush of air, the bellows expands, rela.xing the tension on the rope which is now pulled in the oppo- site direction by the weight, thus turning the sheave froin right to left, cutting resistance into circuit once more and slowing down the motor. The speed of the motor is thus automatically regulated by the bellows, with the result that a practically uniform pressure is maintained at all times. In connection with an organ blower regulator it is necessary to install a separate startifig rheostat. This is required for the reason that all organ bellows leak. During the intermissions in the musical part of the service, or at other times when the blower is not operating, the air gradually escapes and the bellows settles down, moving the rheostat arm to the right and cutting out resistance. With the motor at rest and the bellows empty all the blower regulator resistance would be short circuited and it is therefore necessary to avoid throwing the motor directly across the line when starting again. A starting rheostat with no voltage release is suitable for this purpose, and should be installed within easy reach of the organist, so that a moment or two before beginning to play he can move the lever of the starting box and get the motor into operation. Where remote control is desirable a self starter can be substituted for the manually operated starting box, in which case the entire installation can be controlled by a oush button, or single throw knife switch. OPERATION OF MOTORS C03 12. Gi\dng, also, if possible, the resistance of the shunt field cold, and the shunt field cvurent at the maximum speed required. If this cannot be ascertained, give horse power, voltage, normal speed, maximum speed required, serial number of motor and name of manufacturer. Fir. 770. — General Electric type K7 controller with cover open showing construction. The niechanism consists of a long spindle, carrying a number of heavy brass or gun metal segments, making contact for a longer or shorter time with a corresponding number of spring contacts. The spindle is provided at its upper end with a handle, and the various contacts are made by turning it through an arc of about 150°. For this method a moderate amount of resistance is employed. The first contact joins both naotors and the full amount of resistance in series across the line, and as the motors are standing still, maximum current flows so that they exert their full torque. The moment they start to revolve, the current tends to fall, due to the generation of a reverse voltage; to prevent tnis and maintain a heavy current for some time, thus obtaining rapid acceleration, the resistance is arranged so that it can be gradually reduced, until at about the fourth notch the two motors are in series without resistance across the line. To increase still further the speed in the above type of controller, the series fields may be shunted, and then the ne.xt steps place the motors in parallel with the resistance. Speed Regulation of Traction Motors. — The speed regu- lator for motors of this class is called a controller, and being located in an exposed place is enclosed in a metal casing. Con- trollers are designed to be used for starting, stopping, reversing, and regulating the speed of motors where one or more of these operations have to be frequently repeated. 694 HAWKINS ELECTRICITY The controller used with a single motor equipment is practically the same as any other single motor starting box, excepting that the resistance has sufficient carrying capacity to be left in the circuit some time. When the motor is to operate at full speed all the resistance is cut out. To reverse, a reversing notch is i)laccd in the armature or field circuit, but not in both. Oues. What provision is made to overcome the arc when the circuit is opened? Ans. A magnetic field is used with such polarity that it blows out the arc. Fig. 771. — Controller of the Rauch and Lang electric vehicles. It is of the flat radial type. Two movable copper leaf contacts of ample size make all commutations necessary to obtain the various speeds. Five speeds forward and reverse are provided. Magnetic blow out coils are used on all controllers designed for 500 volt circuits, and on types designed for lower voltages requiring more than 60 amperes normal capacity. The coils are wound with either copper wire or fiat strips of sufficient capacity to carry full load current continuously without undue heating, and after being wound they are treated with an insulating compound making them moisture proof. OPERATION OF MOTORS 695 Oues. What provision is made to prevent reversal before bringing the controller lever to the " off " position? Ans. Controllers having separate reversing cylinders arc fitted -with mechanical interlocks making it necessary to place lever in off position before reversing. POINT RESISTANCE MOTOR 1 MOTORS *R»<»TUSE FIEU) »R«*TuaE ntLO 3 -w^vr-----^>O^00^3~>CV^ 4 -V^vvvwj-^v.^-aXQ,^'(XA) ^"O*^ 5 6 -jAwvw^^v^^vp^O^OOO ^O^TXX) i-Y^-^^s^^--- \p'