The cover image was created by the transcriber and is placed in the public domain.

THE THOUGHT IS IN THE QUESTION THE INFORMATION IS IN THE ANSWER

HAWKINS
ELECTRICAL GUIDE
NUMBER
FOUR
QUESTIONS
ANSWERS
&
ILLUSTRATIONS
A PROGRESSIVE COURSE OF STUDY
FOR ENGINEERS, ELECTRICIANS, STUDENTS
AND THOSE DESIRING TO ACQUIRE A
WORKING KNOWLEDGE OF
ELECTRICITY AND ITS APPLICATIONS
A PRACTICAL TREATISE
by
HAWKINS AND STAFF

THEO. AUDEL & CO. 72 FIFTH AVE. NEW YORK

COPYRIGHTED, 1914,
BY
THEO. AUDEL & CO.,
New York.

Printed in the United States.

TABLE OF CONTENTS GUIDE NO. 4

Classification—series system—danger in series arc light system—constant current system—parallel system—arrangement of feeder and mains in parallel system—series-parallel system—center of distribution—Edison three wire system—evolution of the three wire system—balanced three wire system—balancing of three wire system—copper economy in three wire systems—Dobrowolsky three wire system—modifications of three wire system—three wire storage battery system—three wire double dynamo system—three wire bridge system—three wire three brush dynamo system—Dobrowolsky three wire system—three wire auxiliary dynamo system—three wire compensator system—extension of the three wire principle—five wire system—dynamotor—connections of balancing set—balancing coils—distribution by dynamo motor sets—boosters and their uses—auxiliary bus bar.

Preliminary considerations—various wires—copper wire—iron wire—German silver wire—standard of copper wire resistance—relative conductivity of different metals and alloys—conductors—cable for elevator annunciators—covered conductors—rubber covered conductors—rubber insulated telephone and telegraph wires—weather proof conductors—twisted weather proof wires—precautions in using weather proof conductors—slow burning wire; where used—slow burning weather proof wire; where used; how installed—miscellaneous insulated conductors—safe carrying capacity of wire—pothead wires—circular mils—square mils—mil foot—lamp foot—ampere foot—center of distribution—wire gauges—B. & S. standard wire gauge—micrometer screw gauge—calculating gauge—table of various wire gauges—table of lamp feet for rubber covered wires—lamp table for weather proof wires—symmetrical and unsymmetrical distribution—wiring table for light and power circuits—table of wire equivalents; how to use—table of cable capacities—incandescent lamps on 660 watt circuits—"tree" and "modified tree" system of wiring—distribution with sub-feeders—wrong and right methods of loop wiring—table, of amperes per motor; of amperes per dynamo—calculations for three wire circuit—three wire circuit panel board—size of the neutral wire.

The term "wiring"—open or exposed wiring—selection of wires—installation—disadvantages of open wiring—splicing—pitch of wires—crossing of wires—wiring across pipes—practical points relating to exposed wiring—methods of carrying wires, through floors; on walls—protecting exposed wiring on low ceilings—various porcelain knobs and cleats—wires used in mouldings—standard wooden moulding—kick box—usual character of moulding work—practical points relating to wiring in mouldings—tapping outlets—arc light wiring—arc lamps on low pressure service—circular fixture block—concealed knob and tube wiring; objections; method of installation—arrangement of switch and receptacle outlet in knob and tube wiring—switch boxes—rigid conduit wiring; advantage—types of rigid conduit; requirements—-conduit box—disadvantages of unlined conduit—flexible conduit wiring—Greenfield flexible steel conduit—"fishing"—insulating point—canopy insulator—fish plug and method of insertion—method of installing conduits in fireproof buildings—service entrance to rubber conduit system—condulet outlet to arc lamp—hickey—methods of bending large conduits—armoured cable wiring; features; installation.

Materials for outside conductors—tensile strength of copper wire—pole lines—pole constructing tools—wooden poles—preservation of wooden poles—preservation processes—methods of setting wooden poles in unsuitable soil—reinforced concrete poles—cross arms—lineman's portable platform—poles for light and power wires—spacing of poles—erecting the poles—guy anchors—method of raising a pole—method of pulling anchor into place—guys for poles—head and foot guying—guying corner poles—guy stubs and anchor logs—climbers—wiring the line—pay out reels—method of stringing wires—"come alongs"—tension on wires—sag table—lineman's block and fall with "come alongs"—attaching wire to insulator—splicing; American wire joint; McIntire sleeve and sleeve joint—transpositions—insulators—overhead cable construction—petticoat insulator—Clark's "antihum"—service connections and loops—method of making series, and parallel service connections—joint pole crossing—service wires.

City conditions—underground systems—various conduits—vitrified clay pipe conduit—vitrified clay or earthenware trough conduit—joints in multiple-duct vitrified clay conduit—concrete duct conduits—methods of laying conduit—method of laying single duct vitrified clay conduit—method of laying multiple duct clay conduit—wooden duct conduits—objection to use of wood; remedy; adaptation wooden built-in conduits; method of installation—wrought iron or steel pipe conduits; method of installation—porcelain bridgework or carriers—cast iron pipe and trough conduit; advantages—fibre conduits; joints—Edison tube system—underground cables—metal sheaths on underground cables—pot heads—General Electric manhole junction box—pot head connections.

Preliminary considerations—electrician's instructions—location of receptacles—ceiling buttons—hallway wiring—selection of wiring system—three wire convertible system—method of wiring—location of panel boards—current required on each floor of building—arrangement of feeders—installation of motors—largest size of feeder permissible—method of cutting pocket in floor—outlet baseboard—completed pocket—how to examine partition interiors—house plan of conduit wiring—attachment of mains to knobs—precautions in making joints—wiring for heating appliances—wiring with combination of moulding, flexible tubing or conduit in non-fire proof building—feeder system for large hotel.

Classification—Carbon flashers—wiring diagram for Dull's carbon flashers—brush flashers—knife flashers—flasher transmission gearing—simple on and off flashers—flash system of gas lighting—high speed flashers—lighting flashers—wiring diagram for flags—diagram showing method of wiring for high speed effects on single lines—method of wiring for a torch—wiring diagram for high speeds—Dull's lightning—script breakers—chaser flashers—thermo flashers—carriage calls—monogram for carriage calls—wiring diagrams for sign illumination—-National carriage call operating keyboard—clock monogram—Bett's clock mechanism for operating electric monogram time flasher—talking signs—two way thermal flasher.

Lightning rods; why sharp points are used; erection—diagram showing principle of air gap arrestor—variable gap arc breaker—location of lightning arrestor—carbon arresters with fuses for telephone lines—ground connection.

Early experiments—theory—description of storage cell—electrolyte—effect of current passing through electrolyte—types of storage batteries—Plant cells—Willard plates; wood separator—parts of "Autex" automobile cells—Faure or pasted type—comparison of Plant and Faure plates—the electrolyte; kind generally used; preparation; test; mixing acid and water; kind of vessel used;—specific gravity table—effect of deep vessel—density of electrolyte—hydrometer syringe—impurities in electrolyte—tests for impurities; chlorine, nitrates, acetic acid, iron, copper, mercury, platinum—old electrolyte—voltage of a secondary cell—charging—connections for charging—charging; first time; period; regulation of voltage—Edison cell data—frequency of charging—cadmium test—emergency connections for weak ignition battery—portable testing instruments—charge indications—two methods of charging from a direct current lighting system—colors of the plates—how best results are obtained in charging—charge voltage—two ways of charging—diagram of charging connections—how to keep charging current constant—tests while charging; after charging—charge indications—behavior of electrolyte during discharge—lead burning outfit—"boiling"; causes—hydrogen gas generator for lead burning—quick charging—charging through the night—charging period for new battery—Willard underhung battery box for automobiles—high charging rates—"National" instructions for taking voltage readings—mercury arc rectifier—capacity—table of capacity variation for different discharge rates—how to increase the capacity—discharging; too rapidly rating; maximum rate—Edison alternating current rectifier—attention after discharging—the battery room—battery attendants—points on care and management—placement of cells—how to avoid leakage—precautions when unpacking cells—assembling cells—connections—precaution in joining terminals—battery troubles—short circuiting; indication; location—overdischarge; buckling—sulphation of plates—data on National cells; American cells; Autex cells—action in idle cell—lack of capacity—how to prevent lead poisoning—low specific gravity without short circuits; causes—treatment of weak cells—pole testing paper—disconnecting cells—sulphuric acid specific gravity table—how to take a battery out of commission—Witham charging board—putting batteries into commission—cleaning jars—table of voltage charge as affected by discharge rate—condensed rules for the proper care of batteries.

Uses of storage batteries; their importance in power plants—load curve showing use of storage battery as aid to the generating machinery—parallel connection of battery and dynamo—"floating the battery on the line"—diagram showing effect of battery in regulating dynamo load—connections and circuit control apparatus—diagram showing action of battery as a reservoir of reserve power—three wire system with battery and dynamo—methods of control for storage batteries—diagram of connections for ignition outfit—variable resistance—end cell switches—diagram of connection of battery equipment for residential lighting plant—end cell switch diagram—features of end cell switch construction—end cell switch control—circuit diagram for charging battery in two parallel groups and discharging in series—reverse pressure cells; regulation—Holzer-Cabot dynamotor—boosters—application of series booster system—Bijur's battery system—load diagram—characteristics of series booster—shunt boosters; with battery—Entz' carbon pile booster system—application of shunt booster—circuit diagram for non-reversible shunt booster and battery system—compound boosters; their connections—method of charging battery at one voltage and supplying lights at a different voltage—connections of one form of differential booster—differential boosters; with compensating coil; adaptation.

CHAPTER XXXVI
DISTRIBUTION SYSTEMS

The selection of the system of transmission and distribution of electric energy from the generating plant to lamps, motors, and other devices, is governed mainly by the cost of the metallic conductors, which in many electrical installations, is a larger item than the cost of the generating plant itself. This is especially true in case of long distance transmission, while in those of the lighting plants, the cost of wiring is usually more expensive than that of the boilers, engines, and generators combined.

The principal distribution systems, are classed as:

1. Series;
2. Parallel;
3. Series-parallel;
4. Parallel-series.

Ques. What is the characteristic feature of each class?

Ans. In the series systems the current is constant, but the voltage varies. In the parallel systems, the voltage is constant, but the current varies.

Series System of Distribution.—A series system affords the simplest arrangement of lamps, motors, or other devices supplied with electric energy. The connections of such a system are shown in [fig. 783]. The current from the terminal of the dynamo passes through the lamps, L, L, L, L, one after the other and finally returns to the terminal. The current remains practically constant, but the voltage falls throughout the circuit in direct proportion to the resistance, and the difference in pressure between any two points in the circuit is equal to the current in amperes multiplied by the resistance in ohms included between them.

For example. Each open arc lamp requires about 50 volts. In the system shown in [fig. 783], the pressure measured across the brushes of the dynamo is assumed to be 1,000 volts. As this current flows through the circuit 45 volts will be actually lost in each lamp, and as the drop on the line wire is usually about 10 per cent. of the total voltage, there will be a drop of 5 volts on the conductor between any two lamps. In the circuit shown, there are twenty lamps, therefore, the difference in pressure between either terminal of the dynamo and middle point A of the circuit will be 10 lamps × 50 volts = 500 volts. Likewise, the difference in pressure between any two points on the circuit will be equal to 50 volts multiplied by the number of lamps included between them.

Fig. 783.—Series system of distribution. This is a constant current system, so called because the current remains practically constant. It is used chiefly for arc lighting.

Ques. Describe the danger in a series arc light system?

Ans. Since the total voltage of the system is equal to the sum of the volts consumed in all of the lamps, it is high enough to be dangerous to personal safety.

This is illustrated in [fig. 783]. If the line be grounded at B owing to defective insulation, the pressure of the circuit at that point will be zero, and therefore, a man standing on the ground could touch that point without receiving a shock, but if he should touch the line at the point C, he will receive a slight shock of 150 volts, as there are three lamps between the point C, and the ground connection B. Therefore, the danger of touching the circuit increases directly with the resistance between the point touched and the ground connection, so that if a man touch the circuit at the point D, he will receive a dangerous shock of 16 × 50 = 800 volts. In practice, sixty lamps are often placed on a single arc lighting circuit, so that its total pressure is about 3,000 volts, thus greatly increasing the danger of the system.

Ques. What is a constant current system?

Ans. The series system is a constant current system, and is so called because the current remains practically constant, while the voltage falls throughout the circuit in direct proportion to the resistance.

Ques. What are the principal applications of the series system?

Ans. For arc lighting, and telegraphic circuits.

Ques. What are the advantages of the series system?

Ans. In the case of telegraphic circuits only one wire is required, and for lighting and power transmission and distribution, only two wires; therefore, it is simpler and cheaper than any other system.

Ques. What is the disadvantage of the series system?

Ans. The danger due to the high voltage in installations such as arc lighting circuits.

Parallel System.—Parallel or multiple systems are usually more complicated than series systems, but since the voltage can be maintained nearly constant by various methods, practically all incandescent lamps, electric motors, and a large proportion of arc lamps are supplied by parallel systems.

The general principle of the parallel system is shown in [fig. 784]. With six lamps on the circuit, each requiring one-half ampere of current, at 110 volts, the dynamo will have to supply a current of 3 amperes at a pressure of 112 volts, and this current will flow through the circuit and distribute itself as shown on account of the lesser resistance of the wire relatively to that of the lamps. At the first lamp, the 3 amperes will divide, ½ ampere flowing through the lamp and the remaining 2½ amperes passing on to the next lamp and so on through the entire circuit. The reduction of pressure from 112 volts across the brushes to 110 volts at the last lamp is due to the resistance of the conducting wires.

Ques. What three effects are due to this drop in pressure?

Ans. 1, All the lamps or motors in the circuit receive a lower voltage than that at the dynamo, 2, some lamps or motors may receive a lower voltage than the others, and 3, the voltage at some lamps or motors may vary when the others are turned on or off.

Fig. 784.—Parallel system of distribution. This is a constant voltage system and is used principally for incandescent lighting and electric motor circuits.

The first is the least harmful and may be counteracted by running the dynamo at a little higher voltage; but the second and third are very objectionable and difficult to overcome. They are counteracted successfully in practice, however, by various methods of regulation, the use of boosters, and the operation of dynamos in parallel.

Ques. What are the principal applications of parallel or constant pressure systems?

Ans. They are used on practically all incandescent lamp and electric motor circuits, and on some arc lamp circuits.

Ques. Why is it specially applicable to incandescent lamp circuits?

Ans. Incandescent lamps cannot be made to stand a pressure much over 220 volts, and therefore have to be operated on low voltage systems.

Ques. What is the principal disadvantage of a parallel system as compared with a series system?

Ans. The greater cost of the copper conductors.

Fig. 785.—Arrangement of feeder and mains in parallel system. By locating the feeder at the electrical center, less copper is required for the mains. The cut does not show the fuses which in practice are placed at the junction of feeder and main.

Ques. What is the usual arrangement of parallel systems?

Ans. Conductors known as a feeder run out from the station, and connected to these are other conductors known as a main to which in turn the lamps or other devices are connected as shown in [fig. 785].

Ques. In what two ways may feeders be connected?

Ans. They may be connected at the same end of the mains, known as parallel feeding, or they may be connected at the opposite end of the main, called anti-parallel feeding.

The main may be of uniform cross section throughout, or it may change in size so as to keep the current density approximately constant. The above condition gives rise to four possible combinations:

1. Cylindrical conductors parallel feeding, [fig. 786];
2. Tapering conductors, parallel feeding, [fig. 787];
3. Cylindrical conductors, anti-parallel feeding, [fig. 788];
4. Tapering conductors, anti-parallel feeding, [fig. 789].

Figs. 786 to 789.—Various parallel systems. Fig. 786, cylindrical conductors parallel feeding; [fig. 787], tapering conductors parallel feeding; [fig. 788], cylindrical conductors anti-parallel feeding; [fig. 789], tapering conductors anti-parallel feeding. The term "tapering" is here used to denote a conductor made up of lengths of wire, each length smaller than the preceding length, the object of such arrangement being to avoid a waste of copper by progressively diminishing the size of wire so that the relation between circular rails and amperes is kept approximately constant. In an anti-parallel system, the current is fed to the lamp from opposite ends of the system.

Series-Parallel System.—This is a combination of the series and parallel systems, and is arranged as indicated in [fig. 790]. Several lamps are arranged in parallel to form a group, and a number of such sets are connected in series, as shown. It is not necessary for the groups to be identical, provided they are all adapted to take the same current in amperes, which should be kept constant, and provided the lamps of each set agree in voltage. For example, on the ordinary 10-ampere arc circuit, one group might consist of 5 lamps, each requiring 2 amperes at 50 volts; the next might be composed of 10 lamps, each taking 1 ampere at 100 volts, and so on.

Fig. 790.—Series-parallel system of distribution. It consists of groups of parallel connected receptive devices, the groups being arranged in the circuit in series.

Parallel-Series System.—In this method of connection, one or more groups of lamp are connected in series and the groups in parallel as shown in [fig. 791].

Fig. 791.—Parallel-series system of distribution. It consists of groups of series connected receptive devices, the groups being arranged in the circuit in parallel.

Ques. When is a parallel-series system used?

Ans. When it is desired to operate a number of lamps or motors on a line where voltage is several times that required to operate a single lamp or motor.

The parallel-series system is employed chiefly in the lighting circuit on electric traction lines; here, usually five 110 volt lamps are connected to the source of supply which has a pressure of 550 volts.

Center of Distribution.—It is important to determine the point at which the feeders should be attached to the mains in order to minimize the amount of copper required. The method employed is similar to that used in determining the best location of a power plant as regards amount of copper required. The center of distribution may be called the electrical center of gravity of the system, and is found by separately obtaining the center of gravity of straight sections and then determining the total resultant and point of application of this resultant of the straight sections.

Feeders (feeding cables or conductors) are run from the source of supply to the distributing centers, and, as these feeders are in many cases of considerable length, a substantial loss of pressure generally occurs in them. The pressure at the source of supply, however, is so regulated as to compensate for the drop in the feeders, and the pressure at the distributing centers is thus kept constant; or the same result is obtained by the use of regulating devices in the feeders. The essential condition in most systems is that the pressure at the distributing centers shall be kept practically constant, irrespective of the load.

Edison Three Wire System.—In electric lighting systems used up to about 1897, it was not considered practicable to use incandescent lamps requiring a pressure exceeding 120 volts. This limited the operating voltage of parallel systems, and necessitated the use of conductors of large size and weight, especially where the current had to be transmitted a considerable distance.

The effect of this limiting voltage is more apparent when it is clearly understood that the size of wire required to carry a current depends upon the amperes and not upon the volts.

A wire capable of carrying a current of 10 amperes at 20 volts, can carry 10 amperes at 20,000 volts or any other voltage. Therefore, since the amount of electric energy or power transmitted through a conductor is equal to the amperes multiplied by the volts, it is clear that by increasing the voltage, the power transmitting capacity of a current can be almost indefinitely increased without increasing the size of the conducting wire. This is the reason why considerations of economy dictate the use of the highest voltages possible in long distance transmissions. The voltage of the current is determined, however, by the requirements of the apparatus to be operated.

Incandescent lamps usually require a pressure of 110 volts, and the current required by a 16 candle power lamp at that voltage is about ½ ampere. Therefore if the lamp be designed for a pressure of 220 volts, the current will be reduced to ¼ ampere, and the same size of wire could be used to feed twice as many lamps.

Figs. 792 and 793.—Evolution of the three wire system. Fig. 792 shows two dynamos supplying two independent circuits. These may be connected in series as in [fig. 793], thus operating the two circuits of [fig. 792] with two wires instead of four. To balance the system in case of unequal loading, a third or neutral wire is used as shown in [fig. 794].

The saving of copper is the sole merit of the three wire system, and the object which led to its invention was to effect this economy with the use of 110 volt lamps.

Principle of the Three Wire System.—In [fig. 792], two dynamos A and B are shown supplying two independent incandescent lighting circuits, each circuit receiving 3 amperes of current at a pressure of 110 volts. It is evident that the dynamos could be connected with each other in series, and the lamps connected in series with two each, as shown in [fig. 793], thus making the two wires K and L of the two independent circuits unnecessary, as the pressure will be increased to 220 volts while the current will remain at 3 amperes, and each lamp will require ¼ ampere.

Fig. 794.—Balanced three wire system. The middle conductor, known as the neutral wire, keeps the system balanced in case of unequal loading, that is, a current will flow through it, to or from the dynamos, according to the preponderance of lamps on the one side or the other. These current conditions are shown in [fig. 797].

The amount of copper saved will be 100 per cent., but this arrangement is open to the objection, that when one of the lamps is turned off, or burned out, its companion will also go out. This difficulty is avoided in the three wire system by running a third wire N, from the junction O, between the two dynamos, as shown in [fig. 794], thus providing a supply or return conductor to any one of the lamps, and permitting any number of lamps to be disconnected without affecting those which remain. If the system be exactly balanced, no current will flow through the wire N, because the pressure toward the - terminal of the dynamo A, will be equal to the pressure from the + terminal of dynamo B, thus neutralizing the pressure in the wire. For this reason the middle wire of a three wire system is called the neutral wire, and is usually indicated by the symbol O or ± the latter meaning that it is positive to the first wire and negative to the second. If the system be unbalanced, a current will flow through the neutral wire, to or from the dynamos, according to the preponderance of lamps in the upper or lower sets. When the number in the lower set is the greater, the current in the neutral wire will flow from the dynamos as shown in [fig. 797], and toward the dynamos under the reverse condition.

In the case represented in [fig. 797], there are five lamps in circuit, requiring 2½ amperes of current at a pressure of 110 volts. The two lamps in the upper set will require 1 ampere, and the three lamps in the lower set, 1½ amperes. Since a pressure of 110 volts can force only a current of one ampere through resistance of the two lamps in the upper set, it is evident, that the additional ½ ampere required by the three lamps in the lower set will have to be supplied through the neutral wire, as shown.

Balancing of Three Wire System.—In practice it is impossible to obtain an exactly balanced system, as the turning on and off of lamps as required results in a preponderance of lamps in the upper or lower sets, and furthermore, even when the number of lamps in the two sets are equal, they may be located irregularly, thereby causing the currents to flow for short distances in the neutral line. Therefore, the larger the number of lamps in the circuit, the easier it will be to keep the system in a balanced condition.

Copper Economy in Three Wire Systems.—Theoretically, the size of the neutral wire has to be only sufficient to carry the largest current that will pass through it. A large margin of safety, however, is allowed in practice so that its cross section ranges from about one-third that of the outside line, in large central station systems, to the same as that of each outside line in small isolated systems.

If the neutral wire be made one-half the size of the outside conductor, as is usually the case in feeders, the amount of copper required is ⁵/₁₆ of that necessary for the two wire system. For mains it is customary to make all three conductors the same size increasing the amount of copper to ⅜ of that required for the two wire system.

Fig. 795.—Dobrowolsky three wire system with self-induction coil. It consists of an ordinary direct current dynamo, the armature A and pole pieces N and S of which are shown. A self-induction coil D, is connected to two diametrically opposite points of the winding of the armature A. The coil D may be carried by and revolve with the armature; but in the construction represented, it is stationary, being connected to the armature winding through the brushes CC, rings and wires JJ. The middle point of the self-induction coil D, is connected to the neutral conductor O of the three wire system, the outside conductors + and - being supplied from the brushes BB in the usual manner. The pressure at the terminals of the coil D is alternating; hence the latter, on account of its self-induction, does not act as a short circuit to the armature. Furthermore, the inductances of the two halves of the coil D being equal, the pressure of the neutral wire O is kept midway between the pressures of the outside wires + and -. When the two sides of the system are unbalanced in load, the difference in current carried in one direction or the other by the neutral wire passes freely through the coil D, since the current is steady, or varies slowly, and is therefore unimpeded by the self-induction. It is evident that the ohmic resistance of D should be as low and its self-induction as high as possible, in order that the loss of energy and the difference in voltage on the two sides of the system shall be as small as possible under all conditions.

Modifications of the Three Wire System.—By the employment of suitable arrangements, it is possible to operate a three wire system with only one dynamo. Some of the various arrangements which have been used or proposed in this connection may be briefly mentioned as follows:

Three Wire Storage Battery System, in which a storage battery is connected between the two outside wires, and the pressure of the neutral wire varied to balance the system by shifting the point at which it is connected to the battery.

Three Wire Double Dynamo System, in which a double dynamo having two armature windings upon the same core, connected to two separate commutators, is used in the same manner as two separate dynamos connected in series.

Three Wire Bridge System, in which a resistance is connected across the two outside wires, and the neutral wire is brought to a point on the resistance through a movable switch. The pressures on the two sides of the circuit are equalized by adjusting the arm of the switch for any change of load.

---

Fig. 796.—Three wire compensator system. A and B are the compensators or equalizers. They consist of auxiliary dynamos coupled together and connected to the system as shown. D is the main dynamo, and E, a booster.

Three Wire Three Brush Dynamo System, in which the neutral wire is connected to a third brush on the dynamo.

Dobrowolsky Three Wire System, in which a self-induction coil is connected to two diametrically opposite points of the armature of an ordinary direct current dynamo. The principle of this system is illustrated in [fig. 795].

Three Wire Auxiliary Dynamo System, in which the neutral wire is connected to an auxiliary dynamo which supplies a pressure one-half as great as that of the main dynamo. The auxiliary dynamo is usually belt driven by the main dynamo, and acts as a dynamo when the load is greater on the negative side of the circuit, and as a motor when the excess of load is on the positive side.

Three Wire Compensator System, in which two auxiliary dynamos A and B called compensators or equalizers, are coupled together and connected to the system as shown in [fig. 796]. Each compensator generates one-half as much pressure as the main dynamo D, and serves to equalize the pressure and the load, the compensator on the lightly loaded side operating as a motor and driving the other as a dynamo. When the system is exactly balanced, both compensators run as motors under no load, therefore, consume very little energy. In this arrangement only one booster E, is required for both sides of the system, as the compensators are connected to the outside wires at a point beyond the boosters, and therefore, sub-divide the increased difference of pressure equally between the two sides of the system.

Fig. 797.—Three wire double dynamo system having two separate windings on the same core and separate commutators A and B as shown.

Extension of the Three Wire Principle.—In order to attain still greater economy in copper, the principles of the three wire system may be extended to include four, five, six, and seven wire systems. The comparative weights of copper required by such systems are as follows:

The four wire system requires about two-ninths as much copper, and the seven wire system about one-tenth as much copper, as an equivalent two wire system; but neither is desirable, as their operation involves too much inconvenience, too many unavoidable complications, and create a possibility of accident, which more than offsets the saving in copper.

Fig. 798.—Diagram showing dynamotor connections when used as an equalizer in the three wire system. DM, dynamotor; G, generator side; M, motor side.

The Five Wire System.—This system is employed advantageously in many places in England and Europe, but has not as yet been introduced to any extent in America. It is very probable that in the future the three wire 440 volt system will be selected in preference to the five wire system.

Dynamotor.—This is a combination of dynamo and motor on the same shaft, one receiving current and the other delivering current, usually of different voltage, the motor being employed to drive the dynamo with a pressure either higher or lower than that received at the motor terminals.

The dynamotor in the direct current circuit corresponds to the transformer in the alternating current circuit.

Fig. 799.—Diagram showing connections of balancing set in three wire one dynamo system. The set consists of a motor and dynamo connected, and its operation is practically the same as a dynamotor.

Ques. How is the dynamotor used as an equalizer in the three wire system?

Ans. When thus used, the machine is connected as in [fig. 798]. When both sides of the system are balanced, there will be no current in the neutral lead N, and a small current will pass through the two armature windings of the dynamotor in series, both armatures acting as motors. If the load on one side of the system become larger than the load on the other side, there will be a greater drop in the leads connected to the overloaded side and consequently a lower voltage will exist over the larger load than exists over the smaller load. The armature winding of the dynamotor connected to the higher voltage will act as a dynamo, whose pressure will tend to raise the voltage of the more heavily loaded side.

The direction of the currents in an unbalanced three wire system that is being supplied with energy from a main dynamo is shown in the figure. The commutator at G is connected to the dynamo winding of the dynamotor and is supplying current to the upper or larger load, and the lower commutator is connected to the motor winding of the dynamotor and is taking current from the lightly loaded side.

Motor-Dynamo or Balancing Set.—A balancing set or balancer consists of a motor mechanically connected to a dynamo used to balance a three wire system. The operation of such a combination is practically the same as the dynamotor just described. The balancer is connected as shown in [fig. 799].

Fig. 800.—Holzer-Cabot type M motor-dynamo set. This combination is known as a booster, and is used to raise or lower the voltage on feeders. The motor is series wound and connected in series with one leg of the feeder. Thus, the voltage which the booster will add to the line will be directly in proportion to the current flowing in the feeder. The regulation is therefore automatic.

When an unbalanced load comes on, the voltage on the lightly loaded side rises and on the heavily loaded side drops. The machine on the light side then takes power from the line and runs as a motor driving the machine on the heavy side as a dynamo, supplying the extra current for that side. This action tends to bring the voltage back to normal and gives good regulation.

In some cases the field of each machine is connected to the opposite side of the system which gives a quicker action. This regulation is automatic and the set takes care of unbalanced loads in either direction without adjustment.

Balancing Coils.—Another method of balancing a three wire system which does away with any additional rotating machines makes use of balance coils.

Ques. Describe the type of dynamo used with balancing coils?

Fig. 801.—Diagram showing connections of balancing coil system. The dynamo used in this system is provided with both commutator and collector rings.

Ans. The regular two wire dynamo is used supplying power to the outside wires, but there are collector rings connected to the armature. These rings are much lighter than they would be for a converter as they carry only about ⅛ of the dynamo load. These rings being light are usually placed at the end of the commutator and are connected directly to the commutator bars.

Ques. How are the balancing coils constructed?

Ans. They are built of standard transformer parts, and are placed in cases similar to those of ordinary small transformers.

The coil has a straight continuous winding, both ends and a connection from the middle point of the winding being brought out of the case.

Figs 802 and 803.—Distribution by dynamo-motor sets. Fig. 802, sets in parallel; [fig. 803], sets in series. In [fig. 802], current produced by the main dynamo G, is carried to the machines by the conductors A and B to which the motor portions M are connected in parallel. These motors are provided with shunt wound field coils which may be connected to the primary or to the secondary circuit, consequently the machines run at a practically constant speed. The dynamo portions D of the transformers are connected to the secondary circuits which supply the lamps, etc., L, as indicated. The field magnets of these dynamos may also be fed by the main circuit AB, or they may be self-excited by shunt or compound winding. In [fig. 803], the motors M are all connected in series with the main dynamo G, and the dynamo elements D of the transformers, connected to the lamps, etc., L. If the current be kept constant (the dynamo G having a regulator like a series arc dynamo), and the motors M are simple series wound machines, they will exert a certain torque, or turning effort, which will be constant. It follows, therefore, that if the dynamos D be also series wound, each will generate a certain current which will be constant. If lamps or other devices, designed for that particular current, be connected in series on the secondary circuits, the dynamos D will always maintain that current, no matter how many lamps there may be. When lamps are added, the resistance of the local circuit is raised, and the current in it decreases, so that the dynamo increases its speed until it generates sufficient pressure to produce practically the same current as before. Hence this constitutes a system which is self-regulating, when lamps, etc., are cut in or out of the secondary circuits. No harm results even if the secondary be short circuited, since only the normal current can be generated. But if the secondary circuit be opened, then the machine will race, and probably injure itself by centrifugal force, because the torque of the motor M has its full value, and there is no load upon the dynamos D. To guard against this danger, some automatic device should be provided to short circuit the field or armature of the motor when its speed or reverse voltage rises above a certain point.

Ques. How are the coils connected to the dynamo?

Ans. Two coils are used and are connected to the collector rings as shown in [fig. 801], one coil across each phase. The connections from the middle points of the coils are connected together and to the neutral wire of the system.

Fig. 804.—Diagram to show correctness of balancing coil connection. In the figure, AE, BF, CG, and DH represent the balance coil and its connection for different positions of the armature of a bipolar machine.

Ques. What is the action of the coils in equalizing the load?

Ans. On balanced load, the coils take a small alternating exciting current from the collector rings as any transformer does when connected to an alternating current line with its secondary open. When an unbalanced load comes on, the current in the neutral divides, half going to each coil. This enters the coil at the middle point and half flows each way through the coil and the slip rings into the armature winding. The unbalanced current is thus fed back directly into the dynamo armature continuously.

The coils are small and can be placed back of the switchboard or below the floor, as they require no attention. The current flowing to each slip ring is 25% of the direct current in the neutral wire with the small exciting current taken by the coil added.

The coils are usually built to take care of current in the neutral equal to 25% of full load current of the dynamo with a voltage regulation not to exceed 2 per cent.

Ques. Upon what does the operation of the balancing coil system depend?

Ans. It depends on the following points: First, the impedance[1] of the coils keeps the exciting current which they take from the collector rings down to a small value as it is alternating current. At the same time the current from the neutral wire flows through the four half coils in parallel, and being direct current is impeded only by the ohmic resistance of the coils, which is low, giving only a slight loss in the coils. The common point to which the neutral wire is connected must at all times be neutral to the - and + direct current brushes.

That this common point is at all times neutral is readily shown. Referring to [fig. 804], let AE, BF, CG and DH represent the balance coil and its connection for different positions of the armature of a bipolar machine. Let O be the tap to the middle point of the winding.

Take the instant when the balance coil taps are directly under the direct current brushes as shown at position AE. It is evident that since the point O is the middle point of the coil, it is neutral between A and E. When the armature turns so that the balance coils take the position BF, the voltage drop between A and E may be divided into 4 parts, AB, BO, OF and FE. As in the first instance, O is neutral between the ends of the coil, and the voltage drop over OF equals that over OB.

Since the space AB includes the same number of armature coils as space FE and they are in fields of equal strength, the voltages across the two spaces will be equal, and the voltage over AB equals that over FE. Then adding equals: AB + BO = FE + FO and O is neutral between A and E as in the first case.

In the same way it can be shown that O is neutral between the direct current brushes for any position of the balance coil taps. One coil will operate the system, but two coils, giving four points spaced 90 electrical degrees apart, give better distribution of the current to the armature winding and better regulation of the voltage.

Boosters.—A booster may be defined as, a dynamo inserted in a circuit at a point when it is necessary to change the voltage. A booster is generally driven by a motor, the two armatures being directly coupled, although boosters are sometimes driven from the engine or line shaft.

Fig. 805.—Crocker Wheeler motor-dynamo set. There are numerous cases where such a combination is useful for furnishing a circuit with a voltage different from that of the main plant or with a voltage that can be varied independently. For storage battery charging and electrolytic work, where constant current is desirable, it forms a simple means of voltage regulation. Where a circuit of special voltage is required, the set not only supplies current at the desired pressure, but insulates the special circuit, which may be subject to more severe requirements than the main system. The advantage of the three wire distribution can be obtained from any two wire dynamo by means of a small rotary balancer or balancing transformer, which consists of two direct current machines of the same voltage, mechanically connected together with their armatures in series. Multiple voltage systems for speed regulation can also be obtained by a similar arrangement.

Ques. Explain the use of a booster?

Ans. When a number of feeders run out from a station, the longest and those carrying the heaviest loads will have so much drop on the line that the pressure at distant points is too low. It is therefore necessary to raise the pressure to compensate for the drop and this is done by inserting a booster in the circuit.

It would not be economical to raise the voltage on all the lines by supplying current from the main dynamo at higher pressure, hence the voltage is raised only on the lines which need it by means of the booster working in series with the main dynamo.

Fig. 806.—Diagram showing use of auxiliary bus bar. In order to avoid the necessity for boosters, some stations have an extra bus bar, which is kept at a higher pressure than the main bus, and to this are connected the feeders that have an extra large drop.

Ques. For what other service are boosters employed?

Ans. They are used in connection with storage battery plants for the purpose of raising the voltage of the bus bars to the pressure necessary for charging storage batteries.

Ques. What is an auxiliary bus bar?

Ans. An extra bus bar which is kept at higher pressure than the main bar.

Ques. What is the object of an auxiliary bus bar?

Ans. It is used in place of a booster as shown in [fig. 806]. One or more dynamos maintain the pressure between the auxiliary bar and the common negative bar. The feeders which need boosting are connected to the common negative bar and the auxiliary bar as shown.

CHAPTER XXXVII
WIRES AND WIRE CALCULATIONS

The wireman who is called upon to plan and install a system of wiring will find it necessary first to have a knowledge of the various kinds of wire so as to select the one best suited for the work, and to be able to make simple calculations in order to determine the proper sizes of wire for the various circuits.

Wires are generally made of circular cross section. The process of manufacture consists in drawing the material through steel dies, when its properties permit this treatment. In the case of some substances, as for instance, tin and lead, difficulties arise in the drawing process, and these are therefore "squirted."

The metals most extensively used for wires are copper and iron; German silver, tin and lead are also employed, but only at points where it is desirable to have a comparatively high resistance in the circuit.

Copper Wire.—Copper is used in nearly all cases of wiring because it combines high electrical conductivity with good mechanical qualities and reasonable price. In conductivity it is only surpassed by silver, but the cost of the latter of course prohibits its use for wiring purposes.

Copper wire is used for electric light and power lines, for most telephone and some telegraph lines, and for all cases where low resistance is required at moderate cost.

Hard drawn copper wire is ductile, and has a high tensile strength; these properties allow it to be bent around corners and drawn through tubes without injury.

Pure annealed copper has a specific gravity of 8.89 at 60° Fahr. One cubic inch weighs .32 pound; its melting point is about 2,100° Fahr.

Good hard drawn copper has a tensile strength of about three times its own weight per mile length. Thus, a number 10 B. & S. gauge copper wire, weighing 166 lbs. per mile, will have a breaking strength equal to approximately 3 × 166 = 498 lbs.

Iron Wire.—This kind of wire is largely used for telegraph and telephone lines, although it is rapidly being replaced by copper in long lines.

There are three grades of iron wire:

1. Extra best best (E. B. B.) which has the highest conductivity and is the nearest to being uniform, in quality, being both tough and pliable;

2. Best best (B. B.), which varies more in quality, is not so tough, and is lower in conductivity. It is frequently sold as E. B. B.;

3. Best (B.), which is the poorest grade made, being more brittle, and lowest in conductivity. Iron wire should be well galvanized.

German Silver Wire.—German silver is an alloy consisting of 18 to 30% nickel, and the balance about four parts copper to one part zinc. It is very largely used as a resistance material in making resistance coils, and is sold in the form of wire, and strip. The resistance of this wire varies with its composition.

The resistance of the 18% alloy at 25° C. is 18 times that of copper, and of the 30% alloy about 28 times that of copper.

The safe carrying capacity of the wire in spirals in open air for continuous duty is such that the circular mils per ampere varies from about 1,500 in No. 10 wire to about 475 in No. 30. For intermittent duty the capacity is twice as great.

Standard of Copper Wire Resistance.—Matthiessen's standard for resistance of copper wire is as follows: A hard drawn copper wire one meter long, weighing one gramme, has a resistance of .1469 B. A. unit at 32° Fahr. Relative conducting power: silver, 100; hard or un-annealed copper, 99.95; soft or annealed copper, 102.21.

A committee of the Am. Inst. Electrical Engineers recommends the following form of Matthiessen's standard, taking 8.89 as the specific gravity of pure copper: A soft copper wire one meter long and one millimeter in diameter has an electrical resistance of .02057 B. A. unit at 0°C.[2] From this the resistance of a soft copper wire one foot long and .001 in. in diameter (mil-foot) is 9.72 B. A. units at 0°C.

For every degree Fahr., the resistance of copper wire increases .2222%. Thus a piece of copper wire having a resistance of 10 ohms at 32°, would have a resistance of 11.11 ohms at 82°.

Relative Conductivity of Different Metals and Alloys.
(According to Lazare Weiler.)

Conductors.—Copper is used more than any other metal for transmitting electrical energy, and for interior wiring it is used exclusively. Copper conductors should be of the highest commercial conductivity, not less than 97%.

For conductors up to sizes as large as No. 8 B. & S. gauge, single conductors may be used, but for larger sizes the necessary conductivity should be obtained by conductors made up of strands of smaller wires. The size of these strands depend upon the size of the conductors and the conditions under which they are to be used.

Where conductors are very large (as for instance dynamo leads), and where it is essential that they should be as flexible as possible, strands as small as No. 20 or 22 B. & S. gauge may be used.

Conductors for flexible cords, pendants, fixtures, etc., should also consist of very fine strands, so that they may be perfectly pliable and flexible.

The individual strands for instance, for a No. 16 B. & S. gauge flexible cord should be as fine as No. 30.

Fig. 807.—Elevator cable for annunciators. This type of cable is designed for connecting the movable elevator car with the signal buttons upon the different floors, and is constructed so as to secure strength and flexibility.

Covered Conductors.—For most conditions of service, wires are protected with an insulating covering. Wires used in interior circuits should have a covering which shall act both as an electrical insulator and as a mechanical protection. In some instances, however, the insulating qualities are of secondary importance.

The various forms of covering now in use commercially for wires are:

1. Rubber;
2. Weather proof;
3. Slow burning;
4. Slow burning weather proof;
5. Armoured.

Rubber Covered Conductors.—This class of conductor consists of a tinned copper wire with a rubber covering, protected by an outside braiding of cotton saturated with a preservative compound.

Ques. What are the advantages of rubber insulation for conductors?

Ans. It is waterproof, flexible, fairly strong, and has high insulating qualities.

Fig. 808.—Rubber insulated telephone and telegraph wires. The inner coat of rubber should be free from sulphur or other substances liable to corrode the copper.

Ques. What are the disadvantages of rubber insulation?

Ans. It deteriorates more or less rapidly and is quickly injured by temperatures above 140° Fahr.

Ques. For what service are rubber covered conductors adapted?

Ans. For interior wiring.

Ques. Is pure rubber used?

Ans. No. The covering should be made from a compound containing from 20 to 35 per cent. of pure rubber.

It would be difficult to place pure rubber on a wire, and moreover a covering made of pure rubber would not be durable and would deteriorate very rapidly, particularly at temperatures above 120° Fahr. Accordingly, it is mixed with other materials, such as French chalk, silicate of magnesia, sulphur, red lead, etc.

Weather Proof Conductors.—In this class of conductor, the wire is protected from the weather by a waterproof covering, consisting usually of braided cotton of two or three thicknesses saturated with a moisture resisting insulating compound.

Ques. Where are weather proof conductors used?

Ans. In places subject to dampness, such as cellars, tunnels, open sheds, breweries, etc.

Fig. 809.—Twisted weather proof wires. The insulation consists of two or three thicknesses of braided cotton saturated with a moisture resisting insulating compound.

Ques. What are the advantages of weather proof conductors?

Ans. The insulation is cheap, very durable, and does not deteriorate unless exposed to high temperatures such as will melt the compound.

Ques. State the disadvantages.

Ans. The covering is more or less inflammable and is not very efficient as an insulator.

Ques. What precaution should be taken in using weather proof conductors?

Ans. On account of the inflammable character of the covering, care should be taken in wiring at points where any considerable number of conductors are brought together, or where there is much woodwork or other combustible material.

Ques. For what use are weather proof conductors especially adapted?

Ans. For outside wiring where moisture is certain and where fireproof quality is not necessary.

Obviously conductors of this class should not be used in conduits, nor in fact, in any way except exposed on glass or porcelain insulators.

Slow Burning Wire.—This class of conductor is defined as: one that will not carry fire. The covering consists of layers of cotton or other thread, all the interstices of which are filled with the fireproofing compound, or of material having equivalent fire resisting and insulating properties. The outer layer is braided and specially designed to withstand abrasion. The thickness of insulation must not be less than that required for slow burning weather proof wire and the outer surface must be finished smooth and hard.—Underwriters' requirements.

Fig. 810.—Slow burning wire, formerly known as Underwriter's Wire. The insulation is triple braided, saturated with a white fireproof compound. Solid conductor.

Ques. Where should slow burning wires be used?

Ans. In hot dry places, where ordinary insulations would be injured, and where wires are bunched, as on the back of a large switchboard or in a wire tower.

A slow burning covering is considered good enough when the wires are entirely on insulating supports. Its main object is to prevent the copper conductors coming into contact with each other or anything else.

Ques. What must be done before using weather proof wire?

Ans. Permission to use the wire must first be obtained from the local Inspection Department.

Slow Burning Weather Proof Wire.—The covering of this type wire is a combination of the underwriters and weather proof insulations. The fireproof coating comprises a little more than half of the total covering. When the fireproof coating is placed on the outside, the wire is called "slow burning weather proof."

Fig. 811.—Slow burning weather proof wire. The insulation is composed of two braids thoroughly saturated with a fire proof composition, over which is a highly polished weather proof third braid. This wire was formerly known as "fire and weather proof" wire.

Ques. How does slow burning weather proof wire compare with weather proof wire?

Ans. It is less inflammable and less subject to softening under heat.

Ques. Where should slow burning weather proof wire be used?

Ans. In places where the wires are to be run exposed and where moisture resisting quality is desired, also where at the same time it is desirable to avoid an excess of inflammable covering.

Ques. How should it be installed?

Ans. It should be set on glass or porcelain insulators.

Miscellaneous Insulated Conductors

Fig. 812.—Armoured submarine cable. This type of cable is insulated with a rubber compound containing not less than 30% of pure Para rubber. The following specifications have been adapted by various telegraph companies and the United States Government for general use.

NOTE.—The above specifications refer only to river and harbor cables. Ocean cables are of an entirely different character, and consist of "shore end," "intermediate" and "deep sea" types.

Fig. 813.—Gas engine ignition cable. This is a special cable made to stand the hard service necessary on automobiles. The conductor is composed of 36 strands of No. 27 tinned copper wire, equal to No. 14 in capacity, which gives it necessary flexibility. About this conductor are woven two layers of cotton thread. Over this are woven, in opposite directions, several layers of specially prepared tape which has been given two coatings of fine insulating varnish. Two strong braids of cotton form the outside covering, and each of these different braids is passed through a bath of insulating liquid and baked in a steam heated oven. With three layers of tape the cable will stand a test of 18,000 to 20,000 volts, and with five layers, 30,000 volts.

Fig. 814.—Paper insulated lead encased telephone cable.

Ques. For what service is slow burning weather proof wire not suited?

Ans. It is not adapted to outside work.

Safe Carrying Capacity of Wire.—All wires will heat when a current of electricity passes through them. The greater the current or the smaller the wire, the greater will be the heating effect. Large wires are heated comparatively more than small wires because the latter have a relatively greater radiating surface.

Fig. 815.—Pothead wires. The standard wire for pothead work is either No. 19, 20 or 22 B. & S. gauge, either single conductor or twisted pair, insulated to a diameter of ³/₃₂ inch over rubber, without any outer braid or protection. In the case of twisted pairs one conductor is usually made of a differently colored rubber than the other, so as to distinguish between them.

The temperature of a wire increases approximately as the square of the current, and inversely as the cube of the diameter of the wire.

The elevation in temperature of a wire carrying a current represents so much lost energy.

From these considerations it must be clear that it is important not to overload conductors in order to secure efficient working, and to avoid risk of fire on inside installations.

The Board of Underwriters specifies that the carrying capacity of a conductor is safe when the wire will conduct a certain current without becoming painfully hot.

In the following table of carrying capacity, prepared by the underwriters, a wire is assumed to have a safe carrying capacity when its temperature is not increased by the given current over 30° Fahr. above that of the surrounding air.

SAFE CARRYING CAPACITIES OF WIRES
(Maximum amperes allowed by the Underwriters.)

The lower limit is specified for rubber covered wires to prevent gradual deterioration of the high insulations by the heat of the wires, but not from fear of igniting the insulation. The question of drop is not taken into consideration in the table on [page 731].

The carrying capacity of Nos. 16 and 18 B. & S. gauge wire is given, but no smaller than No. 14 is to be used, except as allowed under rules for fixture wiring.—Underwriters' Rules.

Circular Mils.—The unit of measurement in measuring the cross sectional area of wires is the circular mil; it is the area of a circle one mil (.001 in.) in diameter.

The area of a wire in circular mils is equal to the square of the diameter in mils.

Fig. 816.—Diagram illustrating circular mils. The circular mil is used as a unit of cross sectional area in measuring wires. It is equal to the area of a circle .001 in. diameter; its value is .0000007854 square inch. In the figure the sum of the areas of the nine small circles equals the area of the large circle. This is evident from the following: Take the diameter of the small circles as unity, then the diameter of the large circle is three. Hence, the sum of the area of the small circles × (¼ π × 1²) × 9 = .7854 × 9 = 7.0686; area of the large circle = ¼ π × 3² = .7854 × 9 = 7.0686. Therefore since the area of the large circle equals the sum of the areas of the small circles, the area of a wire in circular mils is equal to the square of its diameter expressed in mils.

Thus a wire 2 mils in diameter (.002 in.) has a cross sectional area of 2 × 2 = circular mils. Accordingly to obtain the area of a wire in circular mils, measure its diameter with a micrometer which reads directly in mils or thousandths of an inch, and square the reading.

The circular mil (abbreviated C.M.) applies to all round conductors, and has a value of .7854 times that of the square mil, that is, 1 circular mil = .7854 square mil. If the diameter be expressed as a fraction of an inch, as for instance ¹/₃ in., the circular mil area may be found as follows: Reduce the fraction ¹/₃ to the decimal of an inch, multiply the result by 1,000 to express the diameter in mils, and square the diameter so expressed, thus: ¹/₃ = 1,000 ÷ 3 = .333. .333 × 1,000 = 333 mils; 333 × 333 = 110,889 circular mils.

The diameter of any wire may be found when its circular mil area is known by extracting the square root of the circular mil area.

Square Mils.—For measuring conductors of square or rectangular cross section, such as bus bars, copper ribbon, etc., the square mil is used. A square mil is the area of a square whose sides are one mil (.001 in. long) and is equal to .001 × .001 = .000001 square inch.

Fig. 817.—Diagram illustrating square mils. A square mil is a unit of area employed in measuring the areas of cross sections of square or rectangular conductors. It is equal to .000001 square inch. One square mil equals 1.2732 circular mils. The figure shows an area of nine square mils; this is equal to 9 × 1.2732 = 11.4588 circular mils.

EXAMPLE.—A copper ribbon for a field coil measures ⅝ inch by ⅛ inch. What is its area in square mils? What is its area in circular mils?

⅝ = .625 in., or 625 mils; ⅛ = .125 in., or 125 mils.

Area in square mils = 625 × 125 = 78,125.

Area in circular mils={78,125 ÷ .7854 }{or 78.125 × 1.2732} = 99,469.

Mil Foot.—This unit is used as a basis for computing the resistance of any given wire. A mil foot means a volume one mil in diameter and one foot long.

The resistance of a wire of commercially pure copper one mil in diameter and one foot long is taken as a standard in calculating the resistance of wires, and has been found to be equal to 10.79 ohms at 75° Fahr.

The calculation is made according to the following rule:

The resistance of a copper wire is equal to its length in feet, multiplied by the resistance of one mil foot (10.79 ohms) and divided by the number of circular mils, or the square of its diameter.

Expressed as a formula:

resistance in ohms = length of wire in ft. × 10.79 circular mils . . . . (1)

EXAMPLE. What is the resistance of a copper wire 1,500 feet long and having a transverse area of 10,381 circular mils?

Substituting these values in formula (1)

resistance= 1,500 × 10.79 10,381 =1.559 ohms.

The transverse area of a copper wire is found by multiplying the resistance of a mil foot (10.79) by its length in feet and dividing the result by its resistance in ohms.

This is obtained directly from the formula (1) by solving the equation for circular mils, thus:

circular mils = length of wire in ft. × 10.79 resistance in ohms . . . . (2)

EXAMPLE. What is the circular mil area of a wire 1,500 feet long and having a resistance of 1.559 ohms?

Substituting the values in equation (2)

circular mils = 1,500 × 10.79 1.559 = 10,381

Figs. 818 and 819.—Diagrams illustrating the meaning of the term lamp foot, and how to apply it in calculating a circuit. As defined, one 16 candle power lamp at a distance of one foot from the fuse block or point of supply is called a lamp foot; this is equivalent to one 8 candle power lamp at a distance of 2 feet, or one 32 candle power lamp one-half foot from the fuse block. In [fig. 819], there are four 8 candle power lamps, and the distance to center of distribution is 10 feet. The circuit then contains 4 ÷ 2 × 10 = 20 lamp feet.

Lamp Foot.—This unit facilitates laying out wiring and calculating the drop. A lamp foot is defined as one 16 candle power lamp at a distance of one foot from the point of supply. Accordingly the number of lamp feet in any circuit is equal to the number of 16 candle power lamps (or equivalent in other sizes) in the circuit multiplied by the distance in feet from the fuse block to the center of distribution.

When no point is specified, the feet are always measured from the supply point to the center of distribution. When other than 16 c.p. lamps are in the circuit they must be reduced to 16 c.p. lamps. Thus two 8 c.p. lamps would be counted one 16 c.p. lamp, one 32 c.p. lamp would be counted two 16 c.p. lamps, etc.

Ampere Foot.—From the foregoing explanation of lamp foot, the significance of ampere foot is easily understood—the two terms are in fact self-defining.

An ampere foot may be defined as the product of one ampere multiplied by one foot.

The unit ampere foot is used in figuring motor circuits or currents designed to carry a mixed load.

Fig. 820.—The center of distribution of a circuit coincides with the geometrical center of the group of lamps when the lamps are of uniform size and spaced equal distances apart. The center of distribution is here indicated by the dotted line A B.

The ampere feet of a main are found by multiplying the maximum load in amperes by the distance from the fuse block to the electrical center of the load.

Thus if the center of distribution be 50 feet from the fuse block and the maximum load is 9 amperes, the number of ampere feet is equal to 9 × 50 = 450.

Electrical Center of Distribution.—The electrical center of a circuit depends upon the distances between the lamps and the fuse block; also the relative sizes of the lamps.

It may be defined as the sum of the lamp feet for each section divided by the number of 16 candle power lamps in the circuit.

If the lamps be of uniform capacity, and placed at equal distances apart, the center of distribution will coincide with the geometrical center of the group of lamps. However, if the lamps vary in size, and be irregularly spaced, the electrical center will not coincide with the geometrical center unless the lamps be symmetrically arranged so as to compensate for the difference in sizes and spacing.

Fig. 821.—Diagram of an irregular circuit illustrating method of finding the center of distribution. Rule: Divide the sum of the lamp feet for each section by the number of 16 candle power lamps or equivalent in the circuit; the quotient gives the distance in feet from the fuse block to the center of distribution.

In such cases, as shown in [fig. 821], the electrical center can be determined by adding together the lamp feet of the several sections A, B, C, etc., of the main and dividing the result by the 16 c.p. units. Thus the lamp feet of

Section	A = 10	lamps	× 10	feet	= 100
"	B = 9	"	× 5	"	= 45
"	C = 7	"	× 6	"	= 42
"	D = 6	"	× 4	"	= 24
"	E = 5	"	× 5	"	= 25
"	F = 4	"	× 10	"	= 40
"	G = 2	"	× 5	"	= 10
which added together gives a total of					286	lamp feet.

This when divided by the ten 16 c.p. units comprising four 16 c.p. lamps and three 32 c.p. lamps, gives a little over 28½ feet as the distance from the fuse block to the center of distribution, the position of which is shown by the line M N in [fig. 821], while that of the geometrical center is shown by the line K L.

When the center of distribution is at a considerable distance from the supply circuit, and it becomes advisable to divide the wiring into two distinct elements—a feeder and one or more mains, the junction of the feeder and the mains should be located at the electrical center of the mains whenever possible. When this is done, it is obvious that the wire size of only one half the main needs to be calculated, as both halves of the main are identical.

Fig. 822.—Brown and Sharpe (B. & S.), or American Standard wire gauge. This gauge was adopted by the brass manufacturers Jan., 1858. The cut is full size, and therefore, shows the actual sizes corresponding to the gauge numbers.

Wire Gauges.—For the purpose of facilitating the measurement of wire, a number of gauges have been designed by various wire manufacturing concerns. The principal gauges used in the United States are the American or Brown & Sharp's gauge; the English standard or Birmingham gauge; Washburn & Moen's standard gauge; Imperial wire gauge; Stubs' steel wire gauge, and the U. S. Standard wire gauge.

The several gauges are here given with explanation of their use.

The American Standard or Brown and Sharp's Gauge.—This gauge is commonly designated as A. W. G. or B. & S., and has been adopted by brass manufacturers and is used mostly in measuring brass, copper, silver, German silver, and gold in both wire and plate.

Birmingham or Stub's Wire Gauge (B. W. G.).—Old English Standard and Iron Wire Gauge. Birmingham or Stubs' Iron Wire Gauge is not the same as Stubs' Steel Wire Gauge. A table of Stub's Steel Wire Gauge is given on [page 741].

Fig. 823.—Micrometer screw gauge. It consists essentially of a screw whose thread is accurately turned to a pitch of some convenient fraction of an inch or centimetre. When the screw is screwed home, the surfaces of A and B meet, and the instrument should then read zero on both the straight and the circular scale. If this be not so, there is a zero error which must be either allowed for, or corrected by means of the screw provided for that purpose. If the former course be adopted, the reading of the instrument is taken when the faces A and B are in contact, and this number added to or subtracted from the final reading according to whether the error makes the wire apparently smaller or greater than its real size. The surfaces A and B are now screwed apart and then, after the wire to be measured (which should be clean and straight) has been introduced between them, they are screwed together to lightly grip the wire. If the gauge be screwed up too tightly the value of the measurement is destroyed, since a copper wire can easily be crushed, and in addition the accurate screw may be permanently damaged. To avoid the possibility of this happening, screw gauges are provided with a ratchet which prevents an excessive force being applied to the screw. If the pitch of the screw in the gauge be ¹/₂₀th of an inch, and the circular scale consist of 50 divisions, then for each revolution of the screw, the surface B will travel a distance equal to the pitch, that is ¹/₂₀th of an inch. The graduations on an instrument of this kind are generally ¹/₁₀th of an inch on the straight scale, with shorter lines to mark the half divisions. The thickness of a wire on the straight scale can therefore be read to the nearest ¹/₂₀th inch. Each division of the circular scale represents ¹/₅₀th of a revolution of the screw, which corresponds to a change in distance between A and B, of ¹/₅₀ of ¹/₂₀ = ¹/_1,000 in. If then the reading on the straight scale be 1 and on the circular scale 35, the distance between A and B is .1 + .035 = .135 inch.

Washburn and Moen's Standard Wire Gauge.—Commonly designated as W. & M. G. has been adopted by the U. S. Steel Corporation in making their wire.

New British Standard (N. B. S.).—British Imperial English Legal Standard and Standard Wire Gauge, and is variously abbreviated by S. W. G. and I. W. G.

Roebling Gauge.—Washburn Moen, American Steel & Wire Co.'s Iron Wire Gauge.

Figs. 824 and 825.—U. S. wireman's calculating gauge; views showing both sides. On the side shown in [fig. 824], set the required number of feet on the small circle opposite the required number of amperes on the large circle, then set the small pointer at the required voltage and loss. Then on the other side ([fig. 825]) the large pointer will indicate the required size of wire in B. & S. gauge, and will also indicate the safe carrying capacity, while the wire may be gauged by slot A ([fig. 824]).

U. S. Standard Wire Gauge.—This gauge is used for measuring sheet and plate iron, and steel, by the U. S. Government in assessing duties, and in making requisitions for supplies.

Old English Standard Wire Gauge.—The old English gauge is the same as the Birmingham or Stubs' standard gauge, commonly designated as B. W. G. It is used chiefly for measuring sheet iron and steel, also soft steel and iron wire.

London Gauge.—Old English (not Old English Standard).

From the foregoing it is seen that great confusion exists with such a multiplicity of gauges and emphasizes the importance of specifying the gauge and of knowing what gauge to use.

In using the gauges known as Stubs' Gauges, there should be constantly borne in mind the difference between the Stubs' Iron Wire Gauge and the Stubs' Steel Wire Gauge. The Stubs' Iron Wire Gauge is the one commonly known as the English Standard Wire, or Birmingham Gauge and designates the Stubs' soft wire sizes. The Stubs' Steel Wire Gauge is the one that is used in measuring drawn steel wire or drill rods of Stubs' make and is also used by many makers of American drill rods.

The following table gives the diameters, in decimal parts of an inch, of the various sizes of wire corresponding to the number of gauge numbers of the different standard wire gauges used in the United States.

NOTE.—The sizes of wire are ordinarily expressed by an arbitrary series of numbers. Unfortunately there are several independent numbering methods, so that it is always necessary to specify the method or wire gauge used. The above table gives the numbers and diameters in decimal parts of an inch for the various wire gauges in general use.

Wiring Terms.—The various members of a complex wiring installation are designated feeders, sub-feeders, mains, branches, and taps.

A feeder is a stretch of wiring to which no connection is made except at its two ends.

A sub-feeder is of the same class as a feeder, but is distinguished either by being one of two or more connecting links between the end of a single feeder and several distributing mains, or by constituting an extension of a feeder.

Fig. 826. Circuit diagram illustrating names of the various parts. A circuit may consist of the following parts as defined in the accompanying text: 1, feeder, 2, sub-feeders, 3, mains, 4, branches, 5, taps. It is well to clearly distinguish between these divisions because the terms are constantly used in wiring.

A main is a stretch of wiring supplied from one or more feeders or sub-feeders and distributing current to a number of taps, or else to a number of branches.

A branch distributes current among a number of lamps, etc.

A tap almost invariably delivers current to a single lamp or other device.

Reference to [fig. 826] will make these definitions clearer. This diagram is intended merely to illustrate the above definitions and does not represent any special plan of wiring.

Figs. 827 and 828. Simplest forms of circuit, consisting of a main with one or more lamps at the end. The smallest size wire allowed (No. 14 B.&S. gauge) will generally be found amply large for such circuits. Note carefully the difference between a main and a branch by comparison with [fig. 826]. A main begins from a fuse block, while a branch is an offset from a main without any fuse block.

The simplest possible wiring installation is one in which a single lamp or compact cluster of lamps is located at the end of a main, as shown in figs. 827 and 828. In such cases calculations are almost always unnecessary, for the reason that No. 14 wire, the smallest size allowed by the underwriters, will supply several lamps at a long distance (as interior wiring goes) with a very moderate drop. For example, if the three lamps shown at the end of the main in [fig. 828], be of 16 candle power each, and the voltage of the supply circuit be 110 volts, a main of No. 14 wire would supply the lamps at a distance of 135 feet from the fuse block with a drop of only 1 per cent.

When the lamps are strung along the main, however, as in [fig. 826], it is sometimes necessary to choose the size of wire with regard to the drop, and in order to do this the main must be measured for either "ampere feet" or "lamp feet."

Wire Calculations.—The problem of calculating the size of wire will be presented here in as simple a form as possible, with explanation of the various steps so that any one can understand how the formula is derived.

In determining the size of wire, there are four known factors which enter into the calculation, viz.:

1. Length of circuit in feet;
2. Maximum current in amperes;
3. Drop or volts lost in the circuit, in % of the impressed voltage;
4. Heating effect of the current.

The calculation is based on the mil foot, which as previously explained, is a foot of copper wire one mil in diameter and whose resistance is equal to 10.79 ohms at 75° Fahr.

Fig. 829.—Wiring for lights requiring unusually long feeders.

The first step is to find an expression for the resistance of the wire which may be later substituted in Ohm's law formula. Accordingly, the resistance of any conductor is equal to its length in feet multiplied by its resistance per mil foot and the product divided by its area in circular mils, thus:

resistance in ohms = length in feet × resistance per mil foot circular mils

or

ohms = feet × 10.8 circular mils . . . . (1)

(calling the resistance per mil foot 10.8 instead of 10.79 to facilitate calculation).

LAMP TABLE FOR RUBBER COVERED WIRES

Showing the maximum number of 16 candle power 110 to 240 volt lamps in parallel that may be carried by the various sizes of wire without violating the underwriters' rules.

Now, according to Ohm's law,

volts = amperes × ohms . . . . (2)

hence, substituting in (2) the value for the resistance in ohms, as obtained in (1):

volts = amperes × feet × 10.8 circular mils

or using the usual symbols

E = I × feet × 10.8 circular mils . . . . (3)

or expressed in words, formula (3) means that the volts lost or drop between the beginning and end of a circuit is equal to the current flowing through the circuit multiplied by the product of the conductors' length in feet multiplied by the resistance of one mil foot of wire, divided by the area of the conductor in circular mils.

LAMP TABLE FOR WEATHER PROOF WIRES

Showing the maximum number of 16 candle power 120 to 240 volt lamps in parallel that may be carried by various sizes of weather proof wire without violating the underwriters' rules.

Now, since the length of the circuit is given as the "run" or distance one way, that is, one half the total length of wire in the circuit, formula (3) must be multiplied by 2 to get the total drop, that is:

E = I × feet × 10.8 X 2 circular mills = I × feet × 21.6 circular mills . . . . (4)

Solving the last equation for the unknown quantity, the following equation is obtained for size of wire:

Figs. 830 and 831.—Symmetrical and unsymmetrical distribution. When a main is supplied by a feeder, the junction of the two, if practicable, is located at the electrical center of the main, as indicated in [fig. 830], so that the distribution is symmetrical, that is, the ampere feet each way from the junction are the same. This is nearly always practicable in surface wiring, and when it is practiced it is only necessary to calculate the wire size for one-half of the main, as the other half is identical. In [fig. 830] there are four lamps on each side of the junction, J; the center of each group is at a distance, M, so that the lamp feet in each half of the main are 5 × M. The lamp feet of the feeder would be 10 × N, N being the distance from the feeder fuse block to the junction, J. In concealed work, however, it does not always happen that a feeder can be made to join a main at its electrical center; when this is not practicable, each end of the main should be figured separately. In [fig. 831], for instance, the main has five lamps on one side and two on the other, and the distances from the junction to the centers of the two groups are at unequal distances S and S'. If the distance S be 14 feet, and the lamps, 16 c. p., the lamp feet in the left hand main equals 5 × 14 = 70, while in the main to the right, taking S' at 10 feet, there are only 2 × 10 = 20 lamp feet. Hence what appears to be one continuous main in this case would have to be treated as two mains, and each part figured separately.

circular mils = I × feet × 21.6 E = amperes × feet × 21.6 "drop" . . . . (5)

The following practical example is given to illustrate the application of the formula just obtained:

EXAMPLE.—What size wire should be used on a 250 volt circuit to transmit a current of 200 amperes a distance of 350 feet to a center of distribution with a loss of three per cent. under full load?

The volts lost or drop is equal to 250 × .03 = 7.5 volts.

Substituting the given value in formula (5)

circular mils = 350 × 200 × 21.6 7.5 = 201,600.

Diameter = 2 √201,600 = 449 circular mils or .449 in.

From the table (on [page 731] or on [page 742]) the nearest (larger) size of wire is 0000 B. & S. gauge.[4]

RULE.—Multiply current in amperes by single distance and refer to the nearest corresponding number under column of actual volts lost, to find size of wire. It should also be noted that the underwriters prohibit the use of wire smaller than No. 14 B. & S. gauge, except as allowed for fixture work and pendant cord.

Ques. If the calculated size of wire be larger than any in the table how is the required area obtained?

Ans. By using two or more smaller wires in parallel, whose combined area is equal to the required area.

To facilitate finding the equivalent sizes the above table of wire equivalents has been prepared.

Ques. How is the table of wire equivalents used?

Ans. To use the table, find in the vertical column at left the size of conductor desired; then follow along horizontally until the size of wire that is desired to use for the strands, and the corresponding number at top of column will give the number of strands of that size wire required.

Fig. 832.—Break down switch for use on three wire circuit, enabling it to be operated break down fashion with the two outers connected together and the neutral wire serving as one side of the resultant two wire circuit. Such circuits must be figured as two wire installations of one half the three wire voltage. The size of the neutral wire of a three wire circuit depends on conditions of operation. Three wire circuits for occasional two wire working, must have a neutral wire whose cross section is equal to the combined cross sections of the two outer wires. This plan is useful for buildings supplied from a central station, as it will be satisfactory for two wire operation in emergencies, and for three wire, two phase or three phase distribution should the central station ultimately be changed over to either of those alternating current systems. The expense for the extra copper in the beginning will not be nearly so great as that entailed by a change in the wiring later on should developments require it. It is permissible, however, to make the cross section of the neutral wire smaller than that of each outer wire, if one be reasonably sure that there will never be any changes such as those just mentioned, and if the drop in the two outer wires do not exceed 1½ per cent. Under such conditions, it will be found a very good rule to calculate the neutral wire of a principal feeder for a maximum unbalancing of 25 per cent, that is, a condition under which the current in one outer wire will be 75 per cent of the current in the other one, the current in the neutral being 25 per cent of that in the heavier loaded outer wire.

Ques. What is the significance of the zig-zag line?

Ans. The figures below this line give the gauge numbers of two wires which will have the same conductivity as the corresponding conductor in left hand column.

TABLE OF CABLE CAPACITIES

Fig. 833.—Diagram showing capacities of cables for both open and concealed work as allowed by the underwriters.

Incandescent Lamps on 660 Watt Circuits.—The standard incandescent lamp is rated as equivalent to the light given by 16 candles, and may consume, according to type and make, from 50 to 56 watts, or from 3.1 to 3.5 watts per candle power. Therefore, a 660 watt circuit will carry thirteen 16 candle power 49.6 watt lamps, or eleven 56 watt lamps.

Fig. 834.—Diagram showing symmetrical and unsymmetrical distribution. The two 5 lamp centers are located at equal distances from the distributing pocket or cabinet, P, so that the sub-feeders, A and A', have equal values of lamp feet. The sub-feeders, B, B', have equal lengths, but as one supplies 10 lamps, and the other 16, the lamp feet are different, and each sub-feeder must therefore be figured separately. The main, G, should be considered as a part of the sub-feeder, B, in order to avoid the necessity for a fuse at the junction of the two. As it is symmetrically divided, only one-half of it would be considered. Thus, if the sub-feeder, B, were 50 feet long, and the main, G, 30 feet long, B would have 16 × 50 = 800 lamp feet and one-half of the main would have 8 × 15 = 120 lamp feet (assuming all 16 c. p. lamps). Hence 800 + 120 = 920 lamp feet should be taken as the load length and the proper size wire used for that figure, making the sub-feeder, B, of the same size as the main, G. The same procedure applies to the sub-feeder B' and main G'; also to the sub-feeder, E, and the main, F. The proper sizes of wire for the different circuits is easily found from the lamp feet table, after having calculated the lamp feet assigned the drop.

Figs. 835 and 836.—The "tree" and "modified tree" systems of wiring. The tree system consists of a feeder reducing in size and supplying mains for each floor, the general arrangement resembling the trunk and branches of a tree. Since fuses must be inserted on each floor where the size of the feeder is reduced, the system requires a large number of joints, and in the event of a fuse blowing it could not be quickly located. The tree system is not to be recommended, as it results in considerable drop, and at full load the lamps nearest the point of supply will either burn too brightly or those more remote will not give the rated candle power. In the modified tree system, [fig. 836], the size of the feeder is not reduced. With this arrangement the losses are considerably reduced owing to the much smaller losses on the feeder between those centers farthest away from the point of supply.

The proper size of wire for a 660 watt circuit will depend upon the voltage for which the lamps are made. For example: a 16 candle power lamp which consumes 56 watts on 110 volt circuit will take, 56 ÷ 110 = .5 or ½ ampere of current, while the same lamp, if made for 220 volts, will take only 55 ÷ 220 = .25 or ¼ ampere. Therefore, eleven 16 candle power 56 watt lamps will require a current of 5½ amperes at 110 volts, or 2¾ amperes at 220 volts.

According to the laws of resistance, the resistance of a round wire is inversely proportional to the square of the diameter, and if the circuit be taken at 100 feet, and the allowable percentage of drop at 1 volt, then according to formula, (5) on [page 748], the wire required for a circuit carrying eleven 16 candle power 56 watt 110 volt lamps, will have a cross sectional area of,

5.5 × 100 × 21.6 1 = 11,880 circular mils.

while the same number of lamps on a 220 volt circuit will require wire having a cross sectional area of,

2.75 × 100 × 21.6 1 = 5,940 circular mils.

In order to conform to the underwriters' requirements, No. 8. B. & S. gauge, wire must be used for the circuit carrying the 110 volt lamps, while No. 12, B. & S. wire, would be sufficient for the 220 volt circuit.

In the case shown in [fig. 829], the branch circuits A and B are identical, each supplying four 16 candle power lamps requiring 3.5 watts per candle power at 110 volts or carrying a load of 4 × 16 × 3.5 = 224 watts, = 224 ÷ 110 = 2 amperes.

Fig. 837.—Distribution with sub-feeders (multi-center distribution). The feeder connects at a central point, A, with several sub-feeders which run to distributing centers, as at B, C, D, and E. With this arrangement, compound wound dynamos may be so designed that the pressure at A will remain nearly uniform for all loads. If, for instance, the wiring be proportioned for 2% drop, the dynamos may be over compounded to that extent, and the even illumination will compensate for the extra cost in the installation.

The distance from the feeder junction or cut out to the electrical center of each branch circuit is 12.5 feet. The compact area of distribution permits the reduction of the loss of volts to 1 per cent, or 110 × .01 = 1.1 volts "drop." Then substituting in formula (5) on page 748 the values for amperes, feet and drop as obtained above

2 × 25 × 21.6 1.1 = 981 circular mils,

or a value far below that of even No. 18 wire, B. & S. gauge (see table on [page 731]), but the smallest wire allowed by the underwriters for the mains A and B is No. 14, B. & S. gauge.

In calculating the size of wire for the feeders the total load must be considered. This is equal to eight 16 candle power lamps, requiring 3.5 watts per candle power at 110 volts = 8×16×3.5 = 448 watts = 4 amperes.

The distance from the entrance cut out to the feeder cut out is 200 feet. The drop should not be greater than 1.5 per cent. or 110×1.5 = 1.6 volts. Then,

4×200×21.6 1.6 = 10,800 circular mils

a value which indicates that No. 8 wire, B. & S. gauge, must be used for the feeders in order to keep the drop within the limit of predetermined value.

NOTE.—In using this table, it is only necessary to calculate the lamp feet of the tap and take the size of wire corresponding to the nearest greater number of lamp feet in the table. The lamp feet specified by this table should not be exceeded by more than 10 per cent. Thus, if a tap measure 108 lamp feet, in 110 volt lamps, No. 12 wire would be used. But if it measure 115 lamp feet, it would be advisable to use No, 10 wire.

Constant Voltage Arc Lamp Circuits.—The branch conductor should have a carrying capacity about fifty per cent. greater than the normal current required by the lamp, so as to provide for the heavy current required when the lamp is started. The underwriters prohibit the use of any size wire under No. 12 for parallel connected arc light circuits.

Constant Current Series Arc Lamp Circuits.—The wiring for series connected arc lamps should never be concealed nor encased unless requested by the electrical inspector.

For all interior wiring of this class, approved rubber covered wire should be used, and the wire should always be rigidly supported on porcelain or glass insulators which will hold the wires at a distance of at least one inch from the surface wired over. The wires on all circuits up to 750 volts, should be kept at least 4 inches from each other, and 8 inches apart on circuits of over 750 volts. No wires carrying a current having a pressure exceeding 3,500 volts should be carried into or over any building except central stations and sub-stations.

Fig. 838.—Diagram showing "bridge wiring." This method is used in the case of two parallel mains where one feeder is ample for both. The feeder is run to a central point as shown and connected to the two mains by a so called "bridge." The arrangement clearly gives good distribution and effects a saving in copper and labor, for if the bridge were omitted, two feeders would be necessary.

Wire Calculations for Motors.—The proper size of wire for a motor may be readily determined by means of the following formula:

circular mils = H.P. × 746 × D × 21.6 E × L × K . . . . (6)

in which

H.P. = horse power of motor;
746 = watts per H.P.;
D = length of motor circuit from fuse block to motor;

21.6 = ohms per foot run in circuit where wires are one mil in diameter;
E = voltage at the motor;
L = drop in percentage of the voltage at the motor;
K = efficiency of the motor expressed as a decimal.

The average values for K are about as follows: 1 H.P., .75; 3 H.P., .80; 5 H.P., .80; 10 H.P. and over, 90 per cent.

Figs. 839 and 840.—Wrong and right methods of loop wiring. In general, when a large percentage of loss is allowed with lamps at short distances, the size of wire, calculated simply in accordance with the resistance rules, will be found too small to carry the current safely. This fact is often overlooked, and even though wires may have been correctly calculated for a uniform percentage of loss, they will become painfully hot simply because the table of carrying capacity was not consulted. The cross connection of mains wherever possible, for the purpose of equalizing the pressure, will also often reduce the heating effects of the current. An example of this is shown in the above figures. A circle of lights was wired as in [fig. 839], and after the current had been turned on, the wires of the circle became hot, and there was quite a perceptible difference of candle power between the lights near A and those near B. Investigation disclosed the fact that the loop, contrary to instructions, had been left open. A few inches of wire as shown in dotted lines remedied the fault. A better arrangement, however, is shown in [fig. 840].

EXAMPLE.—What is the proper size of wire for a 10 H.P. motor, run at 220 volts, allowable drop 2 per cent. on 200 foot circuit.

Substituting the given values in the formula on [page 758]:

Circular mils = 10 × 746 × 200 × 21.6 220 × 4.4 × .9 = 36,991.

The nearest larger value to this result, in the table of carrying capacities of copper wire ([page 731]), is 41,740, corresponding to No. 4 wire, B. & S. gauge.

In all cases the size of the wire thus formed should be compared with that allowed by the underwriters for full load current of motor, plus 25 per cent. of that current, and if the size calculated happen to be smaller than the allowable size, it should be increased to the latter, otherwise it will not pass inspection.

To determine the current required for a motor, as for instance, the 10 H. P. motor under consideration, multiply the horsepower by 746, and divide the product by the voltage of the motor multiplied by its efficiency as follows: (10 × 746) ÷ (220 × .90) = 37.6 amperes.

This value increased by 25 per cent. of itself (37.6 × .25 = 9.4 amp.) is equal to 37.6 plus 9.4 = 47 amperes. In the table of carrying capacities of copper wire ([page 731]), 46 amperes is given as the allowable carrying capacity of No. 6, B. & S. gauge, rubber covered wire; therefore No. 5 wire must be used.

Calculations for Three Wire Circuit.—In all cases of interior conduit work, and in most cases of inside open work, the main feeders from a three wire source of supply are installed on the three wire plan, and the sub-feeders and distributing mains, on the two wire plan, except where the application of the method necessitates the use of unwieldy sizes.

In laying out sub-feeders and mains, the total load, under normal operating conditions, should be divided as nearly as possible into two equal parts, and one part connected on each side of the neutral part of the entrance cut out, or the neutral bus bar of the switch board or panel board in an isolated plant, thus making the load on each side of the neutral wire of the feeder as near equal as possible.

Fig. 841 shows a three wire panel board with connection for 12 mains; those shown in solid lines as A, B, C, etc., being connected between the neutral wire and the negative wire of the feeder, and those shown by dotted lines as G, H, I, etc., being connected between the neutral wire and the positive wire of the feeder. The total load consists of ninety-one 16 candle power lamps, which are so distributed that the positive wire of the feeder carries the current for 46 lamps, and the negative wire, 45 lamps, the neutral wire carrying the difference or current for 1 lamp.

The proper size of wire for the mains may be calculated as already explained, but in calculating the outer wires of the three wire feeder, the neutral wire should be disregarded and the outer wires connected as a two wire circuit carrying the total load of 91 lamps at the over all pressure of 220 volts.

EXAMPLE.—Ninety-one 16 candle power lamps consuming 3.1 watts per candle power at a pressure of 110 volts, will require a current of

16 × 3.1 × 91 110 = 41 amperes.

The distance from the entrance cut out to the main or feeder switch is 200 feet, then for a 2 per cent. drop, or a loss of 110×.02=2.2 volts, the cross sectional area of the wire will be,

41 amperes × 200 feet × 21.6 2.2 volts = 80,509 circular mils.

Fig. 841.—Three wire circuit panel board with connections for 12 mains. The wires shown in solid lines as A, B, C, etc., are connected between the neutral wire and the negative wire of the feeder, and those shown by dotted lines, as G, H, I, etc., are connected between the neutral wire and the positive wire of the feeder.

The joint resistance of the lamps on a three wire system, however, would be four times greater than on a two wire system; consequently the resistance of the outer wires of the feeder in this case will be four times greater for the same percentage of loss, and the cross sectional area of each of the outer wires will be, 80,509÷4=20,127 circular mils. According to the underwriters' rules, this value compels the use of No. 6 B. & S. gauge wire.

If the lamp voltage, 110 volts, be used, the two wire formula (5) given on [page 748] must be modified to,

circular mils = amperes × feet × 21.6 drop × 4

but if an over all voltage, (that is, the voltage between the outer wires), of 220 volts be used, the two wire formula does not require any modification.

The proper size of wire may also be calculated by means of the formula

drop 2 × distance × amperes = resistance per foot . . . . (1)

Example.—If in calculating a three wire feeder, the over all voltage be 220 volts, the drop = 4.4 volts, twice the distance = 400 feet, and the current = 20.5 amperes, then,

4.4 volts 400 feet × 20.5 amperes = .0005365 ohms per foot.

In the table of the properties of copper wire which gives the resistance of various sizes of wire, it will be noted that at all of the given temperatures No. 8 wire has a resistance greater than the value just calculated, therefore, No. 6 B. & S. gauge wire should be used for the outer wires of the feeder. In the table the resistance is given per 1,000 feet, hence it is only necessary to move the decimal point to obtain the resistance per foot.

If the calculation be based on the lamp voltage, 110 volts, the formula (1) must be modified to

drop × 4 2 × distance × amperes = resistance . . . . (2)

In this case, drop = 2.2 volts, 2 × distance = 400 feet, and current = 41 amperes, then,

2.2 volts × 4 400 feet × 41 amp. = 8.8 16,400 = .005364 ohms.

Size of the Neutral Wire.—In three wire circuits, the size of the neutral wire will depend to a great extent upon operating conditions. In the case of installations which occasionally have to be worked as two wire systems, the cross section of the neutral wire should be equal to the combined cross section of the two outer wires.

For interior wiring which must pass inspection, the neutral wire must always be twice the size of one of the outside wires. However, for general distribution, if it be reasonably sure that the system will always be worked three wire and that the drop in the two outer wires does not exceed 1½ per cent., the cross section of the neutral wire may be made smaller than that of one of the outer wires. In such a case the size of the neutral wire may be calculated for a maximum unbalancing of 25 per cent., when the current in one of the outer wires is 75 per cent. of the current in the other outer wire.

For instance, suppose that in a balanced system, the total load on each of the outer wires of a feeder be 211 amperes, and that on account of certain operating conditions, this load has to be divided unequally so as to put 242 amperes on one of the outer wires, and 181 amperes on the other outer wire. In this case the neutral wire will carry 60 amperes, or 25 per cent. of the current carried by the heavier outer wire.

If the drop in the outer wires exceed 1½ per cent., the cross section of the neutral wire will have to be equal to or larger than that of each of the outer wires, otherwise the drop in the neutral wire will exceed ½ volt with an unbalancing of 25 per cent.

CHAPTER XXXVIII
INSIDE WIRING

The term wiring is commonly understood to mean the methods employed in laying the conductors used for the transmission and distribution of electrical energy for lighting, power, and other purposes. Interior wiring, comprises the various methods of installing the conductors from the entrance devices in the walls or other parts of the buildings to the lamps, motors, and other electrical apparatus within the buildings.

The different methods of interior wiring may be conveniently grouped into the following general classes:

1. Open or exposed wiring;
2. Wires run in mouldings;
3. Concealed knob and tube wiring;
4. Rigid conduit wiring;
5. Flexible conduit wiring;
6. Armoured cable wiring.

Open or Exposed Wiring.—This method of wiring possesses the advantages of being cheap, durable and accessible. It is used a great deal in factories, mills and buildings where the unsightly appearance of the wires exposed on the walls or ceilings is of no consequence.

Ques. What kinds of wires are suitable for this method of wiring?

Ans. Either rubber covered or slow burning weather proof wire.

Rubber insulation should always be used where the wire is in a damp place, such as a cellar, and either weather proof or rubber insulation may be used to protect it against corrosive vapors.

Figs. 842 to 844.—Open or exposed wiring. Fig. 842, wires passing through beams. The holes are bored at an angle and wire run through in zig-zag course. Porcelain tubes are used where the wire passes through beams; [fig. 843], cleat work across beams, the cleats are carried by boards attached to the beams; [fig. 844], method of carrying wires on cleats around beams.

Ques. How are the wires installed?

Ans. They are laid on some cornice, wainscoting, beam, or other architectural feature suitable for the purpose, by means of porcelain knobs and cleats, as shown in figs. 842 to 844.

Porcelain knobs should preferably be of the two piece type ([fig. 863]) in which the wire is carried between the upper and lower portions rather than being tied to a one piece knob with a tie wire as in [fig. 860]. Various porcelain knobs and cleats are shown in figs. 860 to 866.

Ques. What are the disadvantages of open wiring?

Ans. The wiring is not sufficiently protected from moisture, and the effects of fire which will destroy the insulation of the wires; it is also liable to mechanical injury.

Figs. 845 to 847.—Splicing. Figs. 845 and 846, making a wire splice, and the twist completed; [fig. 847], a wrapped joint on large wire. Splicing of wires or joining a branch to a main wire should always be made by twisting the wires together or twisting one wire around the other, so that the joint will be mechanically strong enough to carry all strain which may come upon it, without any soldering. The joint should then be carefully tinned and soldered in order to give good electrical contact and to avoid corrosion along the contact surface. Where wires are too large to be twisted together, the ends are given a short bend and the two wires wrapped firmly together with a smaller bare copper wire, after which the joint is thoroughly tinned and soldered, preferably by pouring hot solder over the joint. The joint is then insulated by wrapping it with two layers of pure rubber, and three layers of tape, sufficient to make the insulation thickness equal to that of the wire, after which the whole joint should be painted with water proof paint.

Ques. How far apart should the wires be placed?

Ans. When installed in dry places and for pressures below 300 volts, the insulators should separate the wires 2½ inches from each other and ½ inch from the surface along which they pass. For voltages from 300 to 500, the wires should be separated four inches from each other and one inch from the surface along which they pass.

If the wiring be in a damp place, the wires should be at least one inch from the surface.

Figs. 848 to 850.—Crossing of wires. Where wires cross each other, tubes should be used except in case of large stiff wires as in [fig. 848]; here one wire may be bent down and carried under the other; [fig. 849], short bushing strung on the wire—this method is usually unsatisfactory, especially where a large number of wires cross each other; [fig. 850], wires crossing each other through tubes. Flexible tubing, such as circular loom may be used in crossing wires in dry locations. Insulators should always be provided where wires cross to support the wires, thus preventing the upper wires sagging and touching those below.

Figs. 851 to 853.—Methods of wiring across pipes. The wires should preferably run over rather than under the pipes. Fig. 851 shows crossing with circular loom, and [fig. 852], one in which a tube is used. Both of these methods are satisfactory in the case of gas pipes, but for steam pipes or water pipes which are liable to leak or sweat and drip moisture, the crossing should be above as shown in [fig. 853]. On side walls where vertical wires run across horizontal water pipes, the latter should be enclosed and the moisture deflected to one side.

Ques. How should wires be protected when run vertically on walls?

Ans. They should be boxed in or run in a pipe as shown in [fig. 854], the covering extending 6 feet above the floor.

When placed inside a box there should be a clearance of at least one inch around the wires; the box should be closed at the end as shown, and the wires protected where they enter the top with bushings. When the wires are placed in a pipe they should be first encased in a piece of flexible tubing that will extend from the insulator below the end of the pipe to the first one above it.

Fig. 854.—Methods employed in open wiring when run vertically on walls. Either a box casing or iron pipe should be used to protect the wires. The covering need only extend six feet above the floor.

Ques. What kind of incandescent lamp receptacle or wall socket is best adapted to exposed wiring?

Ans. One which does not have exposed contact ears, an approved form being shown in [fig. 859].

Practical Points Relating to Exposed Wiring.—Some of the principal points which should be remembered in this connection, together with the methods which may be applied to special cases, may be briefly stated as follows:

1. In interior wiring no wires smaller than No. 14 B. & S. gauge should be used, except as allowed by the underwriters, and no more than 660 watts should be allowed to a circuit.

2. Tie wires should have an insulation equal to that of the conductors which they secure.

3. In all cases, whether the wires be run on knobs, split insulators, or cleats, the wires should be supported at intervals of at least 4½ feet, and if exposed to mechanical injury, the supporters should be placed at closer intervals.

4. Wires run on bare ceilings of low basements, especially where they are liable to injury, should be protected by two wooden guard strips as shown in [fig. 858]. The protective strips should be at least ⅞ inch in thickness and slightly higher than the knobs, insulators, or cleats. The two circuit wires should not be run closer than 6 inches apart, and wires run near water tanks must be rubber covered so as to render them moisture proof.

5. Cleats should be used for the wiring of stores, offices, or buildings having flat ceilings, provided the wiring is installed in dry locations.

6. When the installation is exposed to dampness or acid fumes such as those developed in stables, bakeries, etc., the wires should run on knobs or split insulators, and should be rubber covered.

Figs. 855 to 857.—Methods of carrying wires through floors. In passing through floors (or walls) the wires often come in contact with concealed pipes or other grounded material, hence the only way they can be properly protected is by making the bushing continuous. This may consist of continuous porcelain tubes as shown in [fig. 855], or short bushings may be arranged in iron pipes as in [fig. 856]. The method followed in case of an offset in the wall is shown in [fig. 857]. Sometimes the floor can be taken up and an iron conduit, properly bent, put in place, the wires being reinforced with flexible tubing. Another method is to attach the wires to insulators; in this case the floor must not be put down until the wiring has been examined by the inspector.

7. When wires are run at right angles to beams which are more than 4½ feet apart, a running board should be used and the wires cleated to it as shown in [fig. 843]. It is desirable, however, to avoid the use of running boards, whenever possible by running the wires parallel with the beams, thus reducing the cost of insulation.

8. In factories or other buildings of open mill construction, mains of No. 8 B. & S. gauge or larger wire, where they are not exposed to injury, may be placed about 6 inches apart and run from timber to timber, not breaking around, and may be supported at each timber only.

Fig. 858.—Method of protecting exposed wiring on low ceilings by two guard strips.

Fig. 859.—Receptacle suitable for use with open wiring, the requirement being that the contact ears should not be exposed.

9. The best location for feeders is on the walls. In dry buildings the fire and weather proof wire can be used with safety; but covered wire must be used on buildings subject to any form of dampness. In all cases where feeders are run on the walls, they should be protected from mechanical injury by boxings at least 6 feet high on each floor. If floor switches be used, they may be mounted on the front of the boxing. In such cases, the holes in the boxing through which the wires pass to the switches should be provided with porcelain bushings.

10. The rosettes, receptacles, sockets, snap switches, etc., used in connection with exposed wiring should conform in all respects to the standards specified by the underwriters.

Figs. 860 to 866.—Various porcelain knobs and cleats. In open work various forms of these devices are used.

Fig. 867.—Porcelain tube for entrance of wire into a building. There must be a drip loop outside to drain off water, and the hole through which the conductor passes must be bushed with a non-combustible, non-absorptive insulating tube slanting downward toward the outside. The object of the inclination is to allow any water that might enter the tube to gravitate to the drip loop.

Fig. 868.—Interior bushing. Wires must be separated from contact with walls, floors, timbers or partitions through which they may pass by non-combustible, non-absorptive, insulating tubes, such as glass or porcelain, except at outlets where approved flexible tubing is required. Bushings must be long enough to bush the entire length of the hole in one continuous piece, or else the hole must first be bushed by a continuous water proof tube. This tube may be a conductor, such as iron pipe, but in that case an insulating bushing must be pushed into each end of it, extending far enough to keep the wire out of contact with the pipe.

Wires Run in Mouldings.—Wooden mouldings are extensively used in connection with the wiring of stores, factories and buildings. The advantages of this type of construction are: simplicity, cheapness, and accessibility, and when the moulding is run straight and accurately mitred it makes a neat job. Any class of wooden moulding wiring, however, is not sufficiently impervious to moisture to render it suitable for use in damp places, and it is liable to be crushed or punctured. Furthermore, it is naturally very combustible. These difficulties are overcome to a certain extent by impregnating the moulding with some kind of moisture repellant, or by coating it both inside and out with water proof paint. Hardwood moulding should be used wherever possible, but soft wood moulding usually conforms much better to the wall line.

Fig. 869.—Standard wooden moulding for encasing wires. Wooden moulding must not be used in concealed or damp places, nor be placed directly against a brick wall where sweating may introduce moisture that may ultimately cause a short circuit. Wooden moulding for concealing electrical conductors is prohibited by ordinances in some cities.

Ques. For what conditions is wiring in mouldings suitable?

Ans. It is adapted to installations in which the wires have to be laid after the completion of the buildings.

Ques. Describe the moulding usually employed.

Ans. It is made of hardwood in two pieces, a backing and cap, so constructed as to thoroughly encase the wire.

It should provide a one-half inch tongue between the conductors and a solid backing which should not be less than three-eighths of an inch in thickness under the grooves; it should be able to give suitable protection from abrasion.

The inside of the moulding and the cap must have at least two coats of waterproof material, or else the whole moulding must be impregnated with moisture repellant.

Only one conductor is placed in a groove.

The backing is secured to the walls or ceilings by means of wire nails. The wires are then laid in the grooves and the capping put in place and fastened by small brads.

The wires should be continuous, and only rubber covered wire should be employed.

Wooden moulding is made in a great variety of size and design. The general appearance of this type of moulding being shown in [fig. 869].

Ques. What other kind of moulding is used?

Figs. 870 to 872.—Metal moulding. An approved form consists, as shown, of two pieces: base ([fig. 870]), and cap ([fig. 871]), so formed as to snap together, the cap snapping over the base as in [fig. 872]. The entire moulding should be galvanized or coated with a rust preventive. When the base is held in place by screws or bolts from the inside surface, depressions must be provided so that the heads of the screws will be flush with the surface of the moulding.

Ans. Metal moulding, as shown in figs. 870 to 872.

Metal moulding is permitted on circuits requiring not more than 660 watts and where the pressure is not over 300 volts. Special fittings must be used with this class of moulding so that it is continuous both mechanically and electrically. The moulding should be grounded. The installation rules are practically the same as those governing conduit work.

Ques. What is a kick box?

Fig. 873.—"Kick box;" a device used to protect wires encased in porcelain tubes where they pass through floors.

Ans. A fitting, as shown in [fig. 873], for protecting wires at the points where they enter or emerge from the floor.

Ques. How is moulding work installed on brick or plaster walls which are liable to dampness?

Ans. A backing board must be put on before the moulding is used.

Ques. How should moulding be placed on a ceiling with respect to appearance?

Ans. The appearance is improved if the moulding be carried through to the side of the room, even if part of it be not used. This will give a neat and finished appearance to the ceiling as shown in [fig. 874].

Moulding should always be run in as inconspicuous position as possible, and if it be necessary to run it on the open ceiling, it should be arranged so as to form regular panels. Often it can be run so as to take the place of a picture moulding or as a part of the baseboard so that it becomes merely a part of the wooden trim of the building; and in certain cases it should be made of material to match the rest of the trim.

Fig. 874.—Treatment of moulding work on ceilings. All installations should be planned out so as to conform to symmetrical designs, as far as practicable with the proper distribution of the lights, etc., and all runs finished off, whenever necessary, by "dead" mouldings continued to the walls to improve the appearance. In the figure the sectioned portions show the location of the dead moulding. Sometimes, especially in the wiring of private houses the use of special moulding is necessary. In such cases the shape and kind of wood should match that of the finish or trim of the room, and the receptacles should be stained to match the moulding. When the moulding is run along the walls, the capping may be made to match the trim or the picture moulding already in place, thus giving an apparently concealed job. In this kind of work the feeders can be run through the spaces between the walls, and if flexible tubes such as circular loom or flexiduct be used, no splice box will be necessary where the system of wiring changes and single braided rubber wire may be used throughout.

Ques. What is the usual character of moulding work?

Ans. Usually, a certain part of the work will be run as concealed, that is, inside the partitions, the wires being "fished" from the moulding to the outlet.

Practical Points Relating to Wiring in Mouldings.—The following practical points will be found useful in the satisfactory execution of any class of wiring with wooden moulding:

1. Wooden moulding should never be concealed, and should not be used in damp places or in buildings subject to acid fumes, such as ice houses, breweries, or stables, etc.

2. Wooden moulding selected for use should be formed of good straight stock and free from knots, knot holes and other imperfections. The saving effected in the lower cost of second hand moulding does not compensate for the additional cost increase in its working.

3. When wooden moulding is used in connection with solid pipe or flexible tube conduit, an iron box or conduitlet must be installed where the system of wiring changes, as shown in [fig. 875]. The pipe conduit is secured to the box by means of lock nuts, with porcelain bushings or flexible tubes protecting the wires. In all cases the loom should run up to the moulding.

Fig. 875.—Method of tapping outlets for feeder circuits when wiring with wooden moulding.

Arc Light Wiring.—All wiring for high voltage arc lighting circuits should be done with rubber covered wire. The wires should be arranged to enter and leave the building through an approved double contact service switch which should close the main circuit and disconnect the wires in the building when turned "off" and be so constructed that they will be automatic in their action, not stopping between points when started and to prevent arcing between points under any circumstances, and should indicate plainly whether the current is "on" or "off." Never use snap switches for arc lighting circuits. All arc light wiring of this class should be in plain sight and never enclosed, except when required, and should be supported on porcelain or glass insulators which separate the wires at least one inch from the surface wired over. The wires should be kept rigidly at least eight inches apart, except, of course, within the lamp, hanger board or cut out box or switch. On side walls, the wiring should be protected from mechanical injury by a substantial boxing, retaining an air space of one inch around the conductors, closed at the top (the wires passing through bushed holes) and extending not less than seven feet from the floor. When crossing floor timbers in cellars or in rooms, where they are liable to be injured, wires should be attached by their insulating supports to the under side of a wooden strip not less than one-half an inch in thickness.

Arc Lamps on Low Pressure Service.—For this service there should be a cut out for each lamp or series of lamps. The branch conductors for such lamps should have a carrying capacity about 50 per cent. in excess of the normal current required by the lamp or lamps to provide for the extra current required when the lamps are started or should a carbon become stuck without over fusing the wires. If any resistance coils be necessary for adjustment or regulation, they should be enclosed in non-combustible material and be treated as sources of heat; it is preferable that such resistance coils be placed within the metal framework of the lamp itself. Incandescent lamps should never be used for resistance devices. These lamps should be provided with globes and spark arresters, as in the case of arc lamps on high voltage series circuits, except when the closed arc lamps are used.

4. Wooden moulding should never be run in elevator shafts, or shafts of any kind, and should never be run on the inner side of the outside walls of the buildings, as these locations are usually subject to dampness.

5. In laying out feeders it is usually cheaper to use iron conduit in a shaft, than to run moulding through the floor timbers.

6. When tapping outlets for feeder circuits, an iron outlet box with cover should be used, as shown in [fig. 886]. The one splice box is held up to the outlet box already installed by means of two long screws, and the loom is run right up to the moulding so as to leave no exposed wire.

Fig. 876.—Circular fixture block for outlet from moulding work on ceiling.

7. Wherever fixture outlets are installed, a circular fixture block as shown in [fig. 876] should be used, to give a good support for the fixture and to make a neat backing for the fixture canopy. The wires should be brought through the fixture without cutting and disfiguring the canopy.

Concealed Knob and Tube Wiring.—This method of wiring should be discouraged as far as possible, as it is subject to mechanical injury, is liable to interference from rats, mice, etc. As the wires run according to this method are liable to sag against beams, laths, etc., or are likely to be covered by shavings or other inflammable building material, a fire could easily result if the wires become overheated or short circuited.

Fig. 877.—Concealed knob and tube wiring. The wires are carried on porcelain knobs attached to the beams. If run perpendicular to the beams, holes are bored in the latter and porcelain tubes with a shoulder at one end, inserted in the holes through which the wires pass. The knobs should support the wires at least one inch from the surface over which they run, and should not be spaced further than 4½ feet apart. The wires should be attached with tie wires having an insulation equal to that of the conductor which it secures to the knob. The use of split knobs does away with the necessity of using tie wires. The conductors must be at least 5 inches apart and it is better to support them on separate beams when possible. Each wire must be encased in a piece of flexible tube at all switches, outlets, etc., and this piece of tubing should be sufficiently long to extend from the last insulator and project at least one inch beyond the outlet.

Concealed knob and tube wiring is still allowed by the Underwriters, although many vigorous attempts have been made to have it abolished. Each of these attempts has met with strong opposition from electric light companies and contractors, especially in small villages and towns the argument being that it is the cheapest method of wiring, and if forbidden, many places which are wired according to this method would not be wired at all, and the use of electricity would therefore be much restricted, if not entirely dispensed with in such communities. This argument, however, is only a temporary makeshift obstruction against progress, and in the near future, no doubt, concealed knob and tube wiring will be forbidden by the underwriters.

Figs. 878 to 880.—Methods of making fixture outlets in concealed knob and tube wiring. A cleat consisting of a piece of board at least ⅞ in. thick, should be nailed between the joists or studs into which the wood screws supporting the electrolier can be secured. Holes are then bored through the cleat, through which the flexible tubing can pass. With a combination gas and electric fixture as shown in [fig. 879], no cleat is necessary, because the gas pipe supports the fixture. The flexible tubing should be wired to the gas pipe, to prevent displacement by artisans who have occasion to work around the outlet.

Ques. Describe the method of concealed knob and tube wiring.

Ans. It consists in running the wires concealed between the floor beams and studs of a building, knobs being used to support the wires when run parallel to the beams or studs, and porcelain tubes, when run at right angles through the beams, or studs as shown in [fig. 877].

In this method of wiring, usually nothing need be disturbed on the first floor as the various outlets can be reached from the basement and from the second floor.

Fig. 881 and 882.—Arrangement of switch and receptacle outlets in knob and tube wiring. In wiring for switches, flexible tubing must be used on the conductor ends from the last porcelain support, as shown, the same as on conductor ends for other outlets. A pressed steel switch box should be used to encase each flush switch mechanism, even though it already be encased in porcelain. A ⅞ in. wood cleat or cleats are arranged to support the switch box. These wooden cleats should not be set out flush with the outer edges of the studs, but should be set about ⅜ in. back, as illustrated, to allow a space in which the plaster can take a "grip."

For instance, if it be necessary to make an outlet for the center fixture in the parlor, a strip of flooring can be removed from the floor above so as to expose the beams. Then the wireman can bore two holes through each of the beams, insert porcelain tubes therein, slip the wires through the outlet and replace the strip of flooring.

Various simple methods may be employed for carrying the wires to the outlets on the side walls. For example: a small hole can be made in the wall, and the wire may be dropped through the spaces between the walls, or they may be pulled up from the basement by means of a cord lowered with a weight attached to its end. Outlets for switches and base receptacles may be provided for, in a similar manner.

Figs. 883 and 884.—Elevation and sectional view showing arrangement of switch outlet in concealed knob and tube wiring.

Fig. 885.—Arrangement of surface switch in concealed knob and tube wiring. For a surface snap switch outlet, an iron box is not necessary, but a ⅞ in. cleat must be installed to hold the tubing in place and to provide a proper support for the screws that hold the switch. In wiring old buildings where supporting cleats were not provided back of the plaster, a ¾ in. wooden block or plate should be installed on the surface, to which the switch can be attached.

Ques. What are the advantages of concealed knob and tube wiring?

Ans. Its cheapness, especially in wiring completed buildings, and the absence of any wires or casings on the walls or ceilings.

Ques. What kind of wire must be used?

Ans. Wire having an approved rubber insulating covering.

Figs. 886 to 888.—Switch boxes for concealed knob and tube wiring. These are for flush switches and are formed from sheet steel. A single switch box can be expanded for any number of switches, by using the proper number of spacers. Single and double switch boxes can be supplied already assembled and are used where feasible, because it is cheaper to buy them this way than to assemble them. Holes partially punched, which can be knocked out with a hammer blow, are provided in the sides and back through which the flexible conduit wire protection can be extended.

Rigid Conduit Wiring.—The installation of wires in conduits not only affords protection from mechanical injury, but also reduces the liability of a short circuit or ground on the wires producing an arc which would set fire to the surrounding material; the conduit being of sufficient thickness to blow a fuse before the arc can burn through the conduit.

Ques. Describe the unlined type of conduit.

Ans. It consists of an iron or steel pipe, similar in size, thickness, and in every other way to gas pipe, except that special precautions are taken to free it inside from scale or any irregularities; it is then coated inside with enamel, outside it is sometimes enameled and sometimes galvanized.

Ques. Describe the lined type of conduit.

Ans. It usually consists of a plain iron pipe lined with a tube of paper which has been treated with an asphaltic or similar compound; this paper tube is cemented or fastened to the inside of the iron pipe so that it forms practically an integral part of the same.

Ques. What are the advantages of unlined conduit?

Ans. It is cheaper, because having no lining a smaller size of conduit can be used for any given size of conductor; it is also cheaper to install, as it can be bent, threaded, and cut more readily than the lined conduit. Wires may be more easily inserted and withdrawn as the inside is smoother than that of the lined conduit.

NOTE.—Conduits for inside wiring which are subject to inspection, must have an inside diameter of not less than ⅝ inch. They must be continuous from outlet to outlet or to junction bores, and must properly enter and be secured to all fittings, and the entire system be mechanically secured in position. In case of service connections and main wires, this involves running each conduit continuously into a main cut out cabinet or gutter surrounding the panel board as the case may be. Conduits must first be installed without the conductors, and be equipped at every outlet with an approved outlet box or plate. Outlet plates must not be used where it is practicable to install outlet bores. The outlet box or plate must be so installed that it will be flush with the finished surface, and if this surface be broken, it shall be repaired so that it will not show any gaps or open spaces around the edge of the outlet box or plate. In buildings already constructed where the conditions are such that neither outlet box nor plate can be installed, these appliances may be omitted by special permission, providing the conduit ends are bushed and secured. It is suggested that outlet boxes and fittings having conductive coatings be used in order to secure better electrical contact at all points throughout the conduit system. Metal conduits where they enter junction boxes, and at all other outlets, etc., must be provided with approved bushings or fastening plates, fitted so as to protect wire from abrasion, except when such protection is obtained by the use of approved nipples, properly fitted in boxes or devices. Conduits must have the metal of the conduit permanently and effectually grounded. Conduits and gas pipes must be securely fastened in metal outlet boxes so as to secure good electrical connections. If conduit, couplings, outlet boxes or fittings having protective coating of insulating material, such as enamel, be used, such coating must be thoroughly removed from threads of both couplings and conduit and from surfaces of boxes and fittings where the conduit is secured in order to obtain requisite good connection. Where boxes used for centers of distribution do not afford good electrical connection, the conduits must be joined around them by suitable bond wires. Where sections of metal conduit are installed without being fastened to the metal structure of buildings or grounded metal piping, they must be bonded together and joined to a permanent and efficient ground connection. Junction boxes must always be installed in such a manner as to be accessible. All elbows or bends must be so made that the conduit or lining of same will not be injured. The radius of the curve of the inner edge of any elbow must not be less than three and one-half inches. Must have not more than the equivalent of four quarter bends from outlet to outlet, the bends at the outlets not being counted.

Fig. 889.—Conduit box showing arrangement for combination side outlet with open cover. Outlet or junction boxes are of two general types: 1, those which are made for a particular position and have a given number of outlets, and 2, those which have a variable number of outlets which are plugged with metal discs in such a manner that the latter can be knocked out by a slight blow of a hammer. The illustration shows a universal plugged steel conduit box, which can be used as a straight electric, or combination gas and electric, ceiling or side wall outlet, or for flush rotary or push button switches, or for flush receptacles. When rigid conduits are used, they are screwed to the outlets by means of lock nuts and washers. In the case of flexible conduits, the entering ends of the conduits are provided with clamp bushings which are secured to the outlet by means of lock nuts. All outlet boxes are fitted with covers of various designs, which permit their use for various types of construction such as ceiling and wall work in lath or plaster, fireproofing ceiling work, etc., while many designs of outlet plates and receptacle plates may be obtained from the supply houses to satisfy the requirement of any special case.

Ques. What are the disadvantages of the unlined conduit?

Ans. The Underwriters require the use of double braided conductors instead of single braided which are allowed for lined conduits.

Ques. Where may unlined conduits be used?

Ans. In buildings where the conduit is not liable to corrosive action.

Flexible Conduit Wiring.—Flexible conduits are used to advantage in many cases where rigid conduits would not be desirable. It is especially adapted to completed buildings where it is desired to install the wiring by "fishing" without greatly disturbing the walls, floors, or ceilings.

Figs. 890 and 891.—Greenfield flexible steel conduit; [fig. 890] single strip type; [fig. 891] double strip type. The former ([fig. 890]) is formed with a single strip of galvanized steel, interlocked and gasketed in such a manner as to be suitable for concrete construction. The double strip type ([fig. 891]) is constructed of a concave and convex steel strip, spirally wound upon each other in such a manner as to interlock their concave surfaces. Thus the convex surfaces of the two strips form respectively the outer and inner surfaces of the conduit. This construction insures a smooth interior surface, thus reducing the possibility of friction in the drawing in of conductors. A gasket is provided between the inner and outer strips rendering the conduit moisture proof. This form of flexible conduit is especially adapted to use where the wiring is installed after completion of building, because it is very flexible.

Ques. How is a flexible conduit installed by "fishing"?

Ans. It is "fished" under floors, in partitions between the floor and ceiling, by making pockets in the floors, walls or ceilings, say every 15 or 20 feet, and fishing through first a stiff metal wire called a "snake," and then attaching the conduit to same and pulling the conduit in place from pocket to pocket.

Fig. 892.—Insulating joint. This fitting is used in fixture work. The part A screws on to the gas pipe and B to the fixture. The parts are separated by insulating material E, and the outside of the joint is covered with moulded insulation D. In connecting fixtures to the wiring, all wires should be kept away from the gas pipe above the joint, but they may be bunched in below the insulating joint after the wires have been spliced, soldered, and taped. It is important to protect the gas pipe at this point. Insulating joints should be tested before being used.

Fig. 893.—Canopy insulator. This fitting should be installed wherever there are metal ceilings against which the canopies of fixtures might come. The canopy is the brass cup shaped piece used at the top of fixtures to cover the joint, and is simply an insulating ring placed between the canopy and the ceiling. It is in contact with the fixture; hence, it is important that it be insulated from metal ceilings, or else all the benefits derived from an insulating joint will be lost.

Ques. How is the conduit fished on vertical runs?

Ans. A chain or weighted string is used which is dropped from the outlet to the floor and its lower end located by sound of the chain end or weight striking the floor.

Fig. 894.—Section of flooring illustrating use of fishing hook. In fishing wires, punch a hole through the plastering at the required position, being careful that there is no studding at that place. Use a brad awl and cut the hole large enough to permit running of the wires. With a short length of small brass spring wire, push through the opening a few inches of number 19 double jack chain such as is used for general fishing purposes, first having connected the end of the chain with a piece of heavy linen thread. Run out the thread between the laths and the outside wall until the chain touches the floor beneath; move the thread and locate the chain by the sound; bore a hole through the baseboard or floor, as the case may be, toward the chain. Use a two or three foot German twist gimlet. With a small brass spring wire bent at the end in the shape of a hook, fish for the chain and draw it out. At the other end of the thread attach the wire and draw it through with the thread. Passing under the floor bore a second hole through the floor as near the other as possible. Run into this a piece of snake or fishing wire with a hook at the end, until it comes to an obstruction. Locate the obstruction by sound. In running wires under the flooring first carefully examine all parts and find the direction in which the beams and timbers run, and run the wires parallel with these. After locating the end of the fishing wire see if the obstruction be a timber; if so, find the center and bore from the middle diagonally through it in the direction of the fishing wire. Drop the jack chain and thread through the hole; fish for it and draw it through hole number 2; attach the insulated wire and draw it back. Starting hole number 3, bore hole number 4 diagonally through the timber in the direction in which the wire is to be run, making holes 3 and 4 form an inverted "V" through the timber. Run the fishing wire through hole number 4 until it meets an obstruction. If at the end of the room, bore through the floor, drop the chain, fish it out, attach wire and draw it home. Putty up holes after having finished the work, in case of hard finish, plug them up with wood. In lightly built houses it is often found easier to take off the moulding above the baseboard and run the wire under it. In such cases care should be taken to break off the old nails, as any attempt to drive them out would cause a bad break. In closets and around chimneys it is usually found easy to work. A "mouse" or lead weight attached to a string may often be dropped from the attic to the cellar ceiling through the space outside the chimney.

Ques. What is the difference between flexible conduit and flexible tubing?

Ans. Flexible conduits are made of metal while flexible tubing is non-metallic.

Ques. Describe a flexible conduit.

Ans. It is a continuous flexible steel tube composed of convex and concave metal strips, wound spirally upon each other in such a way as to interlock their concave surfaces.

Ques. What are the advantages of this form of flexible conduit?

Ans. It possesses considerable strength and can be obtained in long lengths (50 to 200 feet); elbow fittings are not required as the conduit may be bent to almost any radius. The fissures of the conduit provide some ventilation; this is an advantage in some places and a disadvantage in others.

Figs. 895 to 897.—Greenfield flexible steel conduit and fish plug, showing method of insertion. Fish plugs are made for ⅜ inch, ½ inch, and ¾ inch conduit and are useful in drawing in the conduit in finished buildings where it is desired to fish it under doors or in partitions. After the conduit has been cut off square in the special vise, the fish plug may be screwed into the tube and the fish wire or drawing-in line should then be attached to the eyelet on the end of the plug.

Ques. In what places are flexible conduits not desirable?

Ans. In damp places.

Ques. Why?

Ans. Because of the fissures.

Practical Points Relating to Inside Conduit Wiring.—The following instructions apply to the installation of wiring in both rigid and flexible conduit:

1. All conduits should be made continuous from one junction or outlet box to another, or to the various fixtures. A conduit installation is made a complete system by the use of outlets, outlet boxes, switch or junction boxes, and panel boxes with doors and locks, which serve to thoroughly protect the circuit at all points.

Figs. 898 to 901.—Pull boxes and their use in conduit work. A pull box is a convenient device used for the purpose of avoiding the disadvantages of having too many bends in one continuous line of conduit; too many bends will give trouble when the conductors are drawn in. Pull boxes are also useful in places where the arrangement of the conduit is such that trouble would be experienced in bending it to a fit, and also in the case of conduits which are first run on a side wall and then have to be carried across the ceiling at right angles to the wall. Fig. 898 shows an example of objectionable bends, and [fig. 899], the method of overcoming the difficulty by the use of a pull box. It is evident that it would be impossible to make some of these bends so as to permit the drawing in of the conductors. This difficulty is overcome, as shown, by placing a pull box on the wall, with its top close to the ceiling. A board B, having the proper size holes for the conduits is fastened to the front of the box and close to the ceiling. After the conductors have been drawn into the conduits along the wall as far as the pull box, they can be readily pulled away from the box through the holes in the board into the corresponding conduit on the ceiling. Fig. 901, shows the use of a pull box in a case where it is necessary to run conduit through partitions at right angles to each other. Pull boxes can be designed to suit any condition liable to occur in practice, and when properly used will always save much time and labor. Locknuts should be placed on the ends of all conduits, both inside and outside the pull box in order to prevent their being displaced when drawing in the conductors. After all the conductors have been drawn into the conduit, all the outlets should be plugged up with wood or fibre plugs made in parts to fit around the wires and cables, and the outlets given a coating of some compound which will render the whole system air tight and moisture proof. A final test should then be made to ascertain that there are no grounds on the different parts of the wiring, and that the insulation comes up to the requirements of the underwriters. The metal of all conduits, and the sheathing of steel armoured cables should be effectually and permanently grounded.

2. In the installation of interior conduit wiring, the tubes are usually put in place as soon as the partitions of the buildings have been constructed. In non-fireproof buildings, the tubes are usually supported from the underside of the floor beams, but in fireproof buildings they are placed on top of the floor beams and under the floor as in [fig. 902].

3. When conduit is used in damp places, lead encased wires should be used, and the wires drawn in very carefully so as to prevent any injury to the casings.

4. For wiring installations in buildings constructed entirely of reinforced concrete, the preliminary work should be laid out during the progress of the building operations so as to avoid, as much as possible, the necessity of drilling holes in the finished concrete work.

Fig. 902.—Method of installing conduits in fire proof buildings. The installation of the conduit includes the placing of all outlet boxes, and when this has been completed, the lathing or plastering work is executed, and after that is finished, the wire is pulled into the tubes, and the receptacles, switches, etc., put in position. The work of pulling in the wires may be greatly facilitated by the use of pull boxes as shown in [figs. 899] and [901].

5. For concealed wiring, the location of all the outlets should be marked by sheet iron tubes large enough to hold the conduits. These tubes should be properly plugged, and set in the false work before the concrete is poured in. In a similar manner, threaded pieces of conduit of the proper size, should be placed in the false work for risers.

6. For exposed wiring on concrete walls and ceilings, suitable cast iron supports should be set in the moulds at regular intervals. When liberally used, these supports will also serve as good supports for other pipes.

7. Where a conduit line terminates on the outside of a building some suitable fitting such as a pipe cap should be used, as shown in [fig. 903], to prevent the entrance of moisture into the conduit system. A variety of devices suitable for this purpose are available at supply houses; but those having porcelain covers which spread the wires the proper distance apart are the most satisfactory.

8. Where it is desirable or necessary to continue open wiring from conduits, or where the character of the wiring makes it necessary to bring the wires over from the conduit, as in an arc lamp, neat and safe work can be done by use of a suitable form of condulet as shown in [fig. 904].

Fig. 903.—Service entrance to interior conduit system; showing method of preventing moisture reaching the interior of the conduit system.

Fig. 904.—Outlet to arc lamp from conduit by use of condulet. The wires are brought out from the conduit system at a distance of 2½ inches apart. Conduits are made in a great variety of design with interchangeable porcelain covers which render them adaptable to almost all cases requiring the installation of outlet boxes.

9. Where a conduit line terminates in a switch or panel box, the lining or casing of the panels should be of iron, and the conduit firmly secured to it so as to make good electrical contact. Vertical lines of conduit should be fastened to the wall or other supports in such a manner as to prevent the weight of the conduit coming on the panel box, and each length of conduit installed should be fastened so as to bear only its own weight. The best method of fastening conduit to brick walls is by the use of expansion bolts and screws. In the case of fire brick ceilings or other plastered walls, toggle bolts should be used. When conduits are run on wooden or iron beams, various kinds of pipe hanger may be employed.

10. There are numerous devices on the market for bending conduit for the making of elbows, offsets, etc., but the majority possess the disadvantage that the conduit must be taken to them to be bent. In the case of the smaller sizes, this difficulty is avoided by the use of some form of conduit bender such as shown in figs. 910 and 911.

Figs. 905 to 909.—Sprague multilet covers. Fig. 905, six wire porcelain cover; 906, P & S. rec. cover; 907, cover for five ampere snap switch; 908, G. E. and P. & S. rec. cover; 909, cover for ten ampere snap switch.

11. In all cases, the interior diameter of the conduit installed should be amply sufficient to permit of the wires being drawn in easily, thus providing a substantial raceway for the conductors. The practice of pulling wires through conduit by means of a block and tackle is very objectionable. It is evident that if the wires be pulled in by the application of much force the insulation is very liable to become damaged; furthermore, much difficulty will be experienced in pulling them out again, especially in warm places where the heat tends to soften the lining of the conduit, and also the rubber covering of the wire. Powdered soapstone put in the pipe while the wires are being drawn in will lessen the friction and permit the wire to go in more readily.

Fig. 910.—Ordinary form of hickey or conduit bender. It consists of a piece of one inch steam pipe about three feet long with a one-inch cast iron tee screwed onto one end of the pipe. This device is used as follows: the conduit to be bent is placed on the floor and the tee slipped over it. The workman then places one foot on the conduit close to the tee, and pulls the handle of the bender towards him. As the bending progresses, the workman should take care to continually move the bender away from himself, to prevent the buckling of the conduit.

Fig. 911.—Commercial form of hickey or conduit bender.

Figs. 912 and 913.—Methods of bending large conduits. A substantial support is necessary which may consist, as in [fig. 912], of two pieces of 2 × 4 studding A and B securely fastened to an upright. The conduit is placed under the block A and over the block B, and then bent by a downward pressure exerted at C, the conduit in the meantime being gradually advanced in the direction D to give a curve of the required radius. The method shown in [fig. 913], may be used wherever a ring A can be attached to a beam or girder by means of clamps or otherwise to serve as a support. In this case the conduit is slipped through the ring and placed on the top of blocking B. The bending is accomplished by means of a block and tackle rigged to an overhead beam as shown. Where ring supports cannot be arranged, the application of frame bending methods give the most satisfactory results.

Armoured Cable Wiring.—Where a conduit system cannot be conveniently installed, armoured cable is used. Armoured cable is made by winding steel strips over the insulated conductors, the latter being permanently retained inside the steel casing. Armoured cable is manufactured in long lengths, the actual lengths being determined by convenience in handling.

Figs. 914 and 915.—Greenfield flexible steel armoured conductors. The armour is composed of convex and concave galvanized metal strips, wound spirally upon each other and over the insulated conductors. A gasket is placed between the inner and outer metal strips, thus further rendering the conductor moisture proof.

Fig. 916.—Greenfield flexible steel armoured lead covered conductors for use in wet places, such as breweries, packing houses, cold storage buildings, coal breakers and the like, and for underground construction, in which classes of work these materials are being extensively and satisfactorily used.

Ques. What are the features of armoured cable?

Ans. It is flexible and the conductors are well protected from mechanical injury. While this form of wiring has not the advantage of the conduit system—namely, that the wires can be withdrawn and new wires inserted without disturbing the building in any way whatever—yet it has many of the advantages of the flexible steel conduit, and it has some additional advantages of its own. For example, in a building already erected, this cable can be fished between the floors and in the partition walls, where it would be impossible to install either rigid conduit or flexible steel conduit without disturbing the floors or walls to an extent that would be objectionable.

Figs. 917 to 920.—Greenfield flexible conduit tools. Special tools are necessary for installing this type of conduit. Fig. 917, universal reamer; [fig. 918], bushing tool; [fig. 919], cable armour cutter; [fig. 920], vice for holding conduit. To remove cable armours, clamp the conductor firmly in the armour cutter and with a pair of cutting pliers back the armour off, one strip at a time, to the point of contact with the cutting edge of the tool. The vise for holding conduit takes all sizes. The conduit can be cut with an ordinary hacksaw. To protect the insulation against any possible injury while the wire is being drawn in, a soft metal bushing should be inserted in the end of the tube and secured permanently thereto by means of the bushing tool. The bushing provided for this purpose has an outside thread, which permits its being screwed into the end of the tube and then expanded by the use of the tool. The tool should always be used after the bushing has been screwed into the pipe, then the bushing tool should be inserted.

Ques. How should armoured cable be installed?

Ans. It should be continuous from outlet to outlet, without being spliced and installed on the loop system. Outlet boxes should be installed at all outlets, although, where this is impossible, outlet plates may be used under certain conditions. Clamps should be provided at all outlets, switch boxes, junction boxes, etc., to hold the cable in place, and also to serve as a means of grounding the steel sheathing.

Ques. Is armoured cable wiring expensive?

Ans. It is less expensive than the rigid conduit or the flexible steel conduit, but more expensive than cleat wiring or knob and tube wiring, and is strongly recommended in preference to the latter.

CHAPTER XXXIX
OUTSIDE WIRING

In the equipment of lighting and power plants, the cost of the outside wiring represents a considerable proportion of the total investment, sometimes costing more than the engines, boilers and dynamos.

A thorough knowledge of outside wiring is therefore necessary to properly proportion and install the wires so that the system will prove economical and safe.

Materials for Outside Conductors.—Copper wire is now considered to be the most suitable material not only for the transmission of current for electric light and power purposes, but also for telegraph and telephone lines, in place of the iron wire formerly employed.

Hard drawn copper wire is used in outside construction, because its tensile strength ranges from 60,000 to 70,000 pounds or about twice that of soft copper. This is desirable to withstand the stresses to which the wire is subjected which, in the case of long spans, are considerable.

The table on the next page gives the tensile strength, in pounds per square inch of cross section, hard drawn copper wire of various sizes B. & S. gauge.

The metal aluminum possesses certain advantages as a material for overhead wires. Its conductivity is about .6 that of copper. The specific gravity of aluminum is about 2.7, while that of copper is 8.89, so that a given volume of copper will weigh 3.3 times more than an equal volume of aluminum, and copper wire of given length and resistance would be about twice as heavy as an aluminum wire of equal length and resistance.

There are several disadvantages, such as, low tensile strength, high electro-positive quality of the metal, higher electrostatic capacity, etc.

Pole Lines.—In the majority of cases overhead conductors are supported by wooden poles. In tropical countries, however, such as India, Central America, etc., where wood is rapidly destroyed by the ravages of white ants and other insects, iron poles are almost exclusively used for telegraph, telephone, and other electric transmission lines. The form of iron pole generally adopted consists of tapering shells of sheet iron of convenient length, riveted together at their ends and set into cast iron base plates which are buried in the ground.

Figs. 921 to 929.—Pole construction tools. Fig. 921, long handled digging shovel; [fig. 922], digging bar; [fig. 923], crow and digging bar; [fig. 924], tamping and digging bar; [fig. 925], wood handle tamping bar; [fig. 926], slick digging tool; [fig. 927], post hole augur; [fig. 928], carrying hook; [fig. 929], tamping pick.

Wooden Poles.—On account of their size and straightness, various species of northern pine, cedar and cypress are especially suitable for large poles. Chestnut, which can be readily sawed and hewed is a very good material for smaller poles. Sawed redwood is extensively used in California.

Preservation of Wooden Poles.—The preservation of wooden poles employed in line work is a matter of importance. Decay of the pole at or near the soil line is caused primarily by various forms of bacteria or fungi, and in some cases by insects. Bacteria and fungi attack either dead or living timber. In the case of dead timber, such as that of poles, they attack the walls of the cells and cause the familiar rot or decay which eventually destroys the usefulness of the pole.

It is well known that the rapid multiplication or action of the bacteria and the growth of the fungi are induced by a certain per cent. of moisture and the heat of the sun, that is, the portion of the pole at or near the soil line is alternately moistened and dried. Therefore, in order to protect it against this action, it is necessary to sterilize the pole by the application of an antiseptic which will penetrate the pores of the wood.

Preservation Processes.—There are several processes which may be successfully employed for the preservation of poles or other exposed timber. The best known of these are the creosoting, burnettizing, kyanizing, carbolizing, and vulcanizing processes.

In England, creosoted poles showed no sign of decay at the end of 35 years of service. In the United States they have an average life of 22 years. In Europe impregnation with copper sulphate has been extensively used, but this impregnation must take place within a few days after cutting down the tree.

Uniformly good results have been obtained by impregnation with corrosive sublimate, involving simply immersion in the liquid from ten to fourteen days. German authorities state that the average life of such poles is about 17 years, compared with 14 years for natural or untreated poles.

The application of pitch and tar oftentimes results in more harm than good. It is authoritatively stated, however, that in Europe wooden poles are effectively protected by painting them with tar up to about 2 feet above, and down to about 1½ feet below the soil line. The painted parts of the pole are then covered with a cloth which after being nailed to the pole, is also painted with tar. Finally a zinc plated sheet of iron painted on both sides with tar, is placed around the cloth and tightened to the pole.

The saving due to the use of sterilized poles is 40 per cent. of the cost of unsterilized poles. The comparison is made on the following basis: Cost of pole, $5 each; sterilizing, $1.25; renewal of sterilized pole in 24 years, unsterilized pole, in 12 years.

Figs. 930 to 932.—Pole line construction tools. Fig. 930, split wooden handle post hole auger; [fig. 931], cant hook; [fig. 932]. socket peavey.

Methods of Setting Wooden Poles in Unsuitable Soil.—In places where salt is plentiful and cheap, such as the Great Salt Lake region in Utah, it has been found that the liberal use of salt mixed with the dirt filling tamped in around the foot of the pole is very effectual in preventing decay below the soil line.

Where poles have to be planted in low, swampy ground, or where the climatic conditions are such that timber decays rapidly, it has been found advantageous to place the poles in concrete settings. This method is extensively employed in various parts of the Southern States, square poles being placed in settings about 7 feet deep and 3½ feet square. In very soft ground the employment of a concrete setting is sometimes impracticable. In such cases piles are driven deep into the soil, and the pole bolted to the part of the pile extending above the ground.

Reinforced Concrete Poles.—The strongest point in favor of concrete poles is their durability. Untreated wooden telephone and telegraph poles have to be replaced by new poles about every six or seven years, depending on the percentage of moisture in the soil, the drier the soil, the longer being the life of the pole. Concrete poles are not affected by soil conditions, and if properly made will last indefinitely.

Figs. 933 to 935.—Glass insulator and insulator pin and bracket. The insulator here shown is of the pony double petticoat type. Insulator pins are used with cross arms, brackets are attached direct to the pole.

One form of reinforced concrete pole consists of a skeleton frame work of four corrugated iron rods covered with ordinary concrete. The pole is octagonal in shape, 30 feet long, and provided with mortises for cross arms, the latter being fastened in place by means of iron bolts. It is stated that they are less expensive than pine poles, and that each pole can be manufactured at the point on the line at which it is to be installed or planted.

In Canada, reinforced concrete poles are made square on account of the ease of making, and also on account of the steel economy permitted thereby. All poles are made at the point of erection. They are moulded in wooden forms, in a horizontal position, the top side being left open and finished with a trowel. The concrete is composed of one part of Portland cement, two parts clean sharp sand, and four parts broken stone. A 35 foot pole for ordinary line work weighs about 2½ tons and a 50 foot pole about 5 tons.

Cross Arms.—The familiar cross arms for stringing wires are usually attached to the poles before they are erected. They are commonly made from yellow pine wood, generally 3¼ x 4¼ inches, and are freely coated with good mineral paint as a preservative. Attachment is made to the pole by cutting a gain one inch deep and of sufficient breadth to allow the longest side of the cross arm to fit accurately. It is then secured in place by a lag screw, with a square head, so that it may be driven into place with a wrench.

Fig. 936.—Cross arm which carries the insulator pins. The standard cross arm is 3¼ x 4¼ inches, double painted, and bored for 1½ inch pins and two ½ inch bolt holes. Telephone arms are 2¾ x 3¾ inch, bored for 1¼ inch pins and two ½ inch bolts.

The cross arm is further secured to the pole with braces. These are flat strips of wrought iron or low carbon steel, 30 inches long, ¼ inch thick and 1¼ inches wide, according to standard specifications. Holes are bored at points one inch from either end, one for attaching to the pole, the other for attaching to the cross arm; two braces forming a triangle with the cross arm for the base and with the apex at the point of connection to the pole. Like all other iron work used on pole lines, the braces are carefully "galvanized," so as to stand three immersions of one minute each in a saturated solution of copper sulphate without showing copper deposits, the color being black at the completion of the test.

Before the cross arm is set in place the gain is carefully painted with white lead. As it is important that cross arms on a line of poles, particularly when there are several on each pole, should be at equal distances from the ground as well as being uniformly spaced, it is necessary that some measuring instrument should be used to accomplish this. Such an instrument is the ordinary template, which is a length of board carrying a pointed block at one end, to correspond exactly with the top of the pole, and also cross cleats nailed at precisely the same intervals below it as it is proposed attaching the cross arms. The template, laid upon a pole, shows where to cut the gains.

In planting the poles it is customary to so arrange them that the cross arms on alternate poles shall face in opposite directions, for the purpose of equalizing the strain on the line. On curves, however, all cross arms are placed on the side of the pole facing the middle of the curve.

Ques. What provision is made for attachment of the wires?

Ans. The cross arms are bored with holes for the insertion of the insulator pins, which are made of locust wood and threaded at the upper end to receive the glass insulator.

The cross arm is made of such a length as to accommodate the number of pins to be inserted. An arm for two pins is made three feet long, according to the standard usually followed, with holes for the pins at center points three inches from either end and a space of 28 inches between them in the center.

Ques. How must electric light and power wires be placed when wired on telephone or telegraph poles?

Ans. They must not be put on the same cross arm with the telegraph, telephone, or similar wires, and when placed on the same pole with such wires the distance between the two inside pins of each cross arm must not be less than twenty-six inches.

Fig. 937.—Portable platform with rigging as used by linemen in wiring and making repairs.

Poles for Light and Power Wires.—In selecting the style of pole necessary for a certain class of work, the conditions and circumstances should be considered. Poles may be divided into three classes, the size of wire to be carried being one of the important considerations.

First Class.—Main line of poles should range in length of from 30 to 35 feet with 6 inch tops. The height of trees, of course will have to be considered in many cases.

Second Class.—Town lighting by arc lights. All poles should have at least 6 inch tops. The corner poles should have 6½ inch tops, and wherever the cross arms are placed on a pole at different angles, the pole should have at least a 6½ inch top. A 30 foot pole is sufficiently long for the main line, but it would be advisable to place 35 foot poles on corners.

Third Class.—Where heavy wire, such as No. 00, is used for feeder wire, the poles should have at least 7 inch tops. Where mains are run on the same pole line the strain is somewhat lessened, and poles of smaller size will answer.

Cull Poles.—All poles that are smaller at the top than the sizes agreed upon, are troubled with dry rot, large knots and bumps, have more than one bend, or have a sweep of over twelve inches, should certainly be classed as cull poles. Specifications for electric light and power work should be, and in many cases are, much more severe than those required by telegraph lines. A cull pole, one of good material, is the best thing for a guy stub, and is frequently used for this purpose. A cedar pole is always preferable to any other, owing to the fact that it is very light in comparison to other timber, and is strong, durable, and very long lived.

Pole Setting.—In erecting poles, it seems to be the universal opinion of the best posted construction men that a pole should be set at least five feet in the ground, and six inches additional for every five feet additional length above thirty-five feet; also additional depths on corners. Wherever there is much moisture in the ground, it is of much value to paint or smear the butt ends of the pole with pitch or tar, allowing this to extend about two feet above the level of the ground. This protects the pole from rot at the base. The weakest part of the pole is just where it enters the ground. Never set poles further than 125 feet apart; 110 feet is good practice.

Painting.—When poles are to be painted, a dark olive green color should be chosen, in order that they may be as inconspicuous as possible. One coat of paint should be applied before pole is set, and one after pole is set. Tops should be pointed to shed water.

Spacing the Poles.—In general, the spacing of poles, like their dimensions, is regulated by the weight of the lines they are designed to carry—the heavier the lines the greater the number of poles. The spacing of poles also depends on their liability to injury from storms and wind in any given locality, and the nature of the service. Poles for a telephone line may be spaced twenty to fifty to the mile—that is, from about 260 to 100 feet apart.

Figs. 938 to 941.—Pole line construction tools. Fig. 938, pike pole; [fig. 939], raising fork; [fig. 940], mule pole support; [fig. 941], jenny pole support.

Erecting the Poles.—Since each pole on a properly constructed line is sawed to the right length and carefully shaped before it is finally inserted in the ground, it is necessary that the holes be dug to as nearly the required depth as possible. Holes for poles are dug very little wider than their diameter at the butt, and the depth is usually computed according to the nature of the soil and the weight of the proposed line. Excavation, while sometimes accomplished with patent post hole augers, or even dynamite, is usually done with a long handled digging shovel, and the earth removed with a spoon shovel, such as is shown in [fig. 921].

Fig. 942.—Guy anchor log in position.

Fig. 943.—Stombaugh guy anchor. It is made of cast iron and can be screwed into the ground like an auger.

Wherever required by the nature of the soil, a "grouting" or foundation of loose stones is formed in the bottom of the hole, and, in marshy or springy ground, a base of concrete and cement is laid, with filling of the same material around the pole, when raised.

Ques. How are the poles transported to the holes?

Ans. They are rolled or carried on hooks similar to those used for carrying blocks of ice, except for a long handle for lifting the load at either side.

Fig. 944.—Method of raising a pole. When the pole has been properly placed, it is seized by several linemen. As soon as the top of the pole is raised high enough to permit the pikes to be thrust into the pole, it is then raised to a vertical position. At about 50° the butt end slides into the hole. The earth is then filled in around the pole and firmly tamped down. Eight or ten poles are about as many as can be set by the average gang in a day.

Ques. How are the poles raised and placed in the holes?

Ans. A piece of timber is inserted in the hole as a slide to prevent crumbling of the earth as the pole is slid into place. The end is raised by hand sufficiently to allow the "dead man," or pole hoist, to be placed beneath, and this is moved along regularly as the pole is lifted with pike poles, until it slides into place through the force of gravity.

Fig. 945.—Method of pulling an anchor into place before the guy wire is fastened to the top of the pole, thus obviating the liability of pulling the pole out of plumb.

This accomplished, the pole is held in a perpendicular position by pikes in the hands of assistants, or planted in the ground around it, while the earth is carefully shoveled into the hole and thoroughly packed down with a tamper.

Guys for Poles.—Where poles are subject to severe strains which might throw them down and break the wires, guy cables are largely employed, these being attached near the top and secured either to the base of the next pole, to a suitable guy stub or post, or to a guy anchor, which is buried about eight feet in the earth and held down by stones and concrete.

Ques. Under what conditions is it necessary to guy poles?

Ans. They are guyed at corners in order to thoroughly secure the poles so that no strain may come on the cornerwise span. It is also necessary to guy a line where it is to be deflected from a straight path, as when rounding a hill, water course or railway curve, in order to neutralize the pull of the wires, tending to incline the poles toward the center on which the arc is described; also when descending a hill.

Figs. 946 to 948.—Methods of guying corner poles. The proper guying of corner and terminal poles is especially important; on corners and curves, the guys should be stronger and more frequent and should be placed on the outer side as shown in the diagrams.

Fig. 949.—Head and foot guying of a pole line in descending a hill.

Guy Stubs and Anchor Logs.—In guying a line under such conditions, each pole is connected by a suitable cable to a guy post or "stub," or to an anchor log. Standard rules specify stubs between 18 and 25 feet, with exact limits as to circumference measures at the top and at a point 6 feet from the butt, according to the kind of wood used.

Figs. 950 to 952.—Lineman's tools. Figs. 950 and 951, Eastern pole climbers, with and without strap for attaching to legs; [fig. 952], portable vise with strap for pulling up the slack in splicing.

Thus, guy stubs of cedar or juniper, either 18 or 25 feet in length, must have a circumference of 22 inches at the top and of 32 inches 6 feet from the butt; stubs of chestnut must measure 24 inches in the first, and 34 in the second, while those of cypress require 28 in the first, and in the second, 39 inches for an 18 foot length, and at least 41 for a 25 foot length. In planting guy stubs the same rules are followed as hold for poles, every means being adopted to promote security of construction except that the stub is raked or tilted against the strain on the guy cable.

Wiring the Line.—The erection and guying of the poles of a line as well as the attachment of the cross arms and the screwing on of the insulators are completed before the stringing of the line is begun. It is particularly essential that the pull on poles of a given line be accurately calculated, and that each one be guyed accordingly before the line is strung, in order to avoid the danger of an undue strain upon the wires in attempting to rectify the condition afterward. It is a good working rule that the wires should be subjected to no stress other than the weights of their own spans after they have been attached to the poles.

Figs. 953 and 954.—Pay out reels. Fig. 953, type used for telephone or telegraph work; [fig. 954], type used for electric light work.

Ques. Describe how the wires are strung.

Ans. In stringing the lines, either one or the full number of wires may be put up at the same time. When one line only is to be strung, the operation consists simply in reeling the wire and running it off from a hand reel, such as is shown in [fig. 953] or [954]. At each pole the wire is drawn up to its place, pulled out to the desired tension, and attached to the insulator.

In the operation of stringing a number of lines at once, the method is different. The reels are placed at the beginning of a section, each wire being inserted and secured through a separate hole in a board, which is perforated to correspond with the spacing of the insulators on the cross arms. A rope is then attached to this running board, which is drawn by a team of horses through the stretch to be wired, being lifted over each pole top in turn. When a certain length has thus been drawn out the wires are drawn to the required tension between each pair of poles and secured to the insulators.

Fig. 955.—One form of "come along." The wire is inserted between jaws and is held fast when tension is applied to the ring.

Fig. 956.—An improved form of "come along" or wire stretcher. The jaws which grip the wire are smooth and remain parallel in closing, thus the wire is not scratched or indented, as with circular jaws having teeth.

Ques. How much tension must be put upon the wires?

Ans. In applying tension to the wires as they are strung on the poles, it is the rule to allow some sag. The amount of sag to be allowed varies with different line hangers.

A typical case quoted by one or two authorities gives a sag of four inches at the center of each 130 foot span for a given size of wire, at a given temperature. A more general rule is to make the tension on a wire as it is drawn up between each pair of poles equal to one-third of its breaking weight. Thus No. 10 B.& S. gauge, would be drawn to about 163 pounds, and No. 12 to about 102 pounds. The temperature at the time of stringing and the distance between the poles are, however, important considerations in applying tension and allowing for sag. Thus, one construction company specifies a dip of 10 inches in summer and 8 inches in winter for spans of 130 feet, or 40 poles to the mile. Several authorities specify figures about as given in the above table for No. 14 iron or copper wire.

Fig. 957.—Wireman's "come along" with hook and tackle.

Ques. How is the wire drawn out?

Ans. In drawing out the wire, it is customary to use a wire clamp, or "come along." This tool is attached to a block and tackle, or drawn in by hand, and, as soon as the proper force has been applied, the wire is held, while the lineman secures it to the insulator.

Fig. 958.—Lineman's block and fall with "come alongs" for stretching wire and holding same when making splices.

Figs. 959 and 960.—Approved method of attaching wire to an insulator; elevation and plan of insulator and tie. The line wire is first laid in the groove of the insulator, after which a short piece of the same size of wire is passed entirely around to hold it in place, then it is twisted to the line at either side with pliers.

Another contrivance for this purpose is the pole ratchet, by which the wire is drawn tight and held until attached to the pole.

Ques. How are the wires attached to the insulators?

Ans. An approved method is shown in figs. 959 and 960. Standard rules specify that all wires shall be tied to the side of the insulators toward the pole, except on the insulators next to the pole, where they are to be attached on the opposite side. On curves, however, it is required that all wires shall be arranged so that the strain shall be against the insulator and not on the wire.

Fig. 961.—American wire joint. This is a simple method of connecting the ends of the sections of wire by tightly twisting the ends around each other for a few turns; it is the standard Western Union wire joint.

Figs. 962 and 963.—McIntire sleeve and sleeve joint. An approved method of making the joints of telephone lines is by the use of some form of sleeve, such as is shown in [fig. 962]. This consists of two copper tubes of the required length, and of sufficient inside diameter, to admit the ends of the wires to be joined, fitting tightly. The tubes are then gripped with a tool, shown in [fig. 964], and twisted around one another, so that the wires are securely joined and locked, as shown in [fig. 963].

Ques. How are the wires spliced?

Ans. There are several methods of splicing wires. Fig. 961 shows the American wire joint, and [fig. 963] the McIntire sleeve joint. In making a joint, the two ends are gripped by come alongs and drawn up to the proper tension with tackle as shown in [fig. 958]. The joint is then made as shown in the illustrations.

Transpositions.—In some classes of circuit, as for instance telephone lines, the current is often seriously affected by electrostatic induction from other lines, and also from power circuits, owing to the fact that the surfaces of the wires form, as it were, so many charging plates of an electrical condenser, with the intervening air as the insulating layer or dielectric.

Fig. 964.—McIntyre's twisting clamp for wires 00 to 16 B. & S. gauge.

Fig. 965—Method of making a "transposition." This is usually done by means of transposition insulators, which are either double insulators, one being screwed to the pin above the other, or else such caps as are shown in [fig. 967]. Such insulators are intended to act as circuit breakers, the particular wire to be transposed being cut and "dead ended," or tied around, on both the upper and lower grooves of the cap. The free end of each length is then passed back and around the insulator and twisted, or sleeve jointed to the other limb of its own circuit.

The telephonic current changes the pressure of its own charging surface as frequently as it alternates, and this fact in itself is amply sufficient to account for a vast weakening of the current before it reaches its destination. The only practicable method of overcoming this annoyance in pole lines is by the arrangement known as "transposition," which is, briefly, the practice of regularly shifting the relative position of the two limbs of each circuit as regards other wires in the same pole system, as shown in [fig. 965].

For short lines and pole systems with only a few wires it is not necessary to transpose very frequently. On longer lines it has been found amply sufficient to transpose once every quarter mile; that is to say to change the relative position of the wires of the different circuits at posts situated about that distance apart. This does not mean, however, that each pair of wires is transposed so often, but that on ordinary sized systems, the transposition of some one circuit is amply sufficient to secure balanced relations and effectually counteract the effects of cross induction. It is a matter which must be carefully calculated and planned in each particular instance in order to secure the best advantages.

Fig. 966.—Telegraph and telephone line glass insulator.

Fig. 967.—Type of insulator used in making a transposition.

Insulators.—Glass and porcelain are employed almost universally for supporting overhead wires. Insulators made of these materials are superior to those made of other material such as hard rubber, or various compounds of vegetable or mineral matter, with the exception perhaps of mica insulators used on the feeders of electric railway lines.

Fig. 968.—Tree insulator. This type of insulator is especially useful in connection with temporary or repair work, or where the wires pass through trees having numerous branches. The illustration shows a Cutler tree insulator lashed to the trunk of a tree. It is made of a single piece of glass, and is provided with a slot which the wire cannot leave accidentally. The back of the device is concave and provided with ribs which prevent sliding. It can be readily slipped over wires already in place, is available for electric light circuit, and will take wires up to ½ inch, in diameter.

Figs. 969 and 970.—Overhead cable construction. In some cases, particularly on short lines exposed to inductive disturbances from power and other electrical circuits, it is usual to string the cables on poles such as usually carry the bare conducting wires. It is not necessary, however, to insulate the cable in any way; consequently it is merely hung to a supporting wire rope or cable, called the "messenger wire," being attached either with some form of hanger, such as is shown in figs. 969 and 970, or by loops of tarred marline. The marline is sometimes wound over the cable and messenger wire from a bobbin, but frequently it is merely wound on by hand. Cables used in such overhead construction consist of bundles of wires, the pairs twisted together. The size most often used is No. 19, B. & S., which is about .03589 inch in diameter, weighs 20.7 pounds, and has a specific resistance of about 8 ohms to the mile.

Glass insulators are generally used on low tension lines, and porcelain insulators on high tension lines, the latter type being usually stronger and less brittle. Porcelain is more expensive than glass, and its opacity prevents the detection of internal defects which would be readily observed through glass.

Fig. 971.—Clark's "antihum;" a device designed to prevent the humming of telegraph wires.

Ques. What is a petticoat insulator?

Ans. An insulator which has one, two or three deep flanges or "petticoats" around the base for the purpose of increasing the leakage path from the line to the pin.

Both glass and porcelain insulators may be the double or triple petticoat type which may be cast or moulded solid, or made in two or more parts which are subsequently cemented together.

Service Connections and Loops.—Whenever it is necessary to tap an overhead conductor for service connection, the method of connection will depend upon the character of the circuit. In the case of a parallel circuit, an extra insulator must be placed on the cross arm so as to prevent the service main putting a side strain on the main line conductor. In the case of a series circuit the main line conductor is usually dead ended at the nearest pole and a loop taken to the point of service, as shown in [fig. 972].

Fig. 972.—Method of making a series "loop" service connection.

Fig. 973.—Parallel service connection. Service wires tapped to the main wires, are run to insulators on an auxiliary cross arm, thence to insulators on the side of the building, and through the drain tube to the service switch.

Fig. 974.—Joint pole crossing, showing wires of two lines crossing each other. Four guard wires (shown heavier than the others) extend for one span either side of the joint pole parallel to the wires of the lower circuits and protect them from contact in case of a break in the wires of the upper circuits. These guard wires are insulated. The minimum distance between high and low tension wires should be three feet. Five is better. The end guards, which prevent wires slipping off ends of cross arms and dropping on the lower wires, should extend about six inches above the level of transmission line.

Ques. What are service wires?

Ans. Wires which enter a building.

CHAPTER XL
UNDERGROUND WIRING

In large cities, the best method of running wires for all varieties of electrical power transmission is to place them underground. Many city authorities have made this method of wiring compulsory by law, because of the difficulty in approaching a burning building, the danger from crossed and falling wires, and the disfigurement of the streets where there is a network of overhead wires.

The expense of installing an underground system is very great in comparison with that of overhead construction, but the cost of maintenance is much less and the liability of interruption of service greatly reduced.

Underground Systems.—An underground system of electrical conductors is composed of three essential elements:

1. The conductor itself, which is almost invariably of copper;

2. The insulation, which is either in the form of a complete covering of insulating material, or simply insulated supporting points;

3. The tube or conduit, which constitutes the mechanical protection against the effects of the severe shocks, weather conditions, etc., to which the system is naturally exposed.

The various underground systems may be divided into three classes:

1. Lead encased cables laid directly in the ground;
2. Solid or built in systems;
3. Drawing in systems.

Ques. What may be said of the first mentioned construction?

Ans. Where cables are laid directly in the ground, the metallic covering, consisting usually of a lead tube, which is placed over the insulation is depended upon for mechanical protection. Such cables are largely used for short private lines and the first cost is less than that of the others, but in case of repairs it has to be dug up.

Ques. Describe the drawing in system.

Ans. In this construction the cables are drawn in after the conduits are built. The conduit of the drawing in system may consist of various forms of pipe or troughs of iron, earthenware, concrete, wood or fibre, while those of the solid or built in systems are composed of either iron tubes or concrete trenches.

Conduits.—The principal qualifications of a good conduit are freedom from disintegration by the action of fire, water, acids, alkalies, or electrolysis; second, a smooth interior surface so as to permit of the easy drawing in of the cables; and third, a design which will permit of its economical installation in crowded streets. There are numerous kinds of conduit of which may be mentioned:

1. Vitrified clay pipe conduits;
2. Vitrified clay or earthenware trough conduits;
3. Concrete duct conduits;
4. Wooden duct conduits;
5. Wooden built in conduits;
6. Wrought iron or steel pipe conduits;
7. Cast iron pipe and trough conduits;
8. Fibre conduits.

Fig. 975.—A few forms of vitrified clay pipe conduits; view showing single and multiplex types. The dimensions of each duct are about 3½ × 3½. The lengths vary from two to three feet.

Vitrified Clay Pipe Conduit.—Various forms of vitrified clay conduit appear to possess the qualifications, desirable in underground construction, to a higher degree than any other type. They are made in both single and multiple duct, as shown in [fig. 975], the single type being about 3½ inches in diameter, or 3½ inches square, and 18 inches long. Multiple conduit is made in two, three, four, six and more sections, ranging from 2 to 3 feet in length.

Ques. For what conditions is the single conduit especially adapted?

Ans. It is most suitable for use where the sub-surface conditions are characterized by a great crowding of gas, water, and other pipes, as the conduits can be divided into several layers so as to cross over or under such pipes, and many other sub-surface obstructions which are present in the streets of large cities and towns.

Ques. What are the features of the multiple duct conduit?

Ans. It can be laid somewhat cheaper than the single duct type, especially in lines of about two to four ducts; it is, therefore, most suitable for use in outlying communities where the streets are comparatively free from many sub-surface obstructions.

Ques. How is the conduit laid?

Ans. In laying conduit, a trench is dug, usually sufficiently wide to allow the placing of three inches of concrete on each side of the ducts, and sufficiently deep to hold at least thirty inches of concrete on top of the upper layer of concrete forming the conduit, and to allow for three inches of concrete in the bottom. The trench is graded from some point near the middle of the block to the manhole at each intersection, or from one manhole to the next manhole, at a gradient not less than 2 inches to 100 feet.

Ques. How are single duct conduits laid?

Ans. The tiles of the several ducts are placed close together, and the joints plastered and filled with cement mortar consisting of one part of Portland cement to one part of sand. When the conduit is being laid, a wooden mandrel about four or five feet long, three inches in diameter, and carrying a leather or rubber washer from three to eight inches larger at one end is drawn through each duct so as to draw out any particles of foreign matter or cement which may have become lodged in the joints, and also to insure good alignment of the tiles, as shown in [fig. 977].

Single duct conduits are usually laid by bricklayers. This fact accounts for the somewhat greater cost of the single over the multiple conduit which is usually laid by ordinary laborers. One good brick-layer and helper, however, will lay from 200 to 300 feet of single duct conduit per hour.

Practically the same standard of construction is maintained on all conduit lines from two ducts up to twenty-five ducts, as many of the smaller lines may extend for miles into the outlying districts, and contain transmission lines of the maximum working voltage.

Fig. 976.—Vitrified clay or earthenware trough conduit; this type of conduit consists of troughs either simple or with partitions, the latter type being shown in the figure.

Vitrified Clay or Earthenware Trough Conduit.—It consists of troughs either simple or with partitions as shown in [fig. 976]. They are usually made in tiles 3 or 4 inches square for each compartment, with wall about one inch thick. The length of the tiles ranges from two to four feet. Each of the two foot form duct troughs weighs about 85 pounds. When laid complete, the top trough is covered with a sheet of mild steel, about No. 22 gauge, made to fit over the sides so as to hold it in position, and then covered over with concrete.

Joints in Multiple-duct Vitrified Clay Conduit.—In laying multiple duct earthenware conduit, the ducts or sections are centered by means of dowel pins inserted in the holes at each joint, which is then wrapped with a six inch strip of asphalted burlap, or damp cheese cloth, and coated with cement mortar as shown in [fig. 978]. Economy of space and labor constitutes the principal advantages derived from the use of multiple duct conduit.

Fig. 977.—Method of laying single duct vitrified clay conduit. The tiles of the several ducts are placed close together as shown in the figure, and the joints plastered and filled with cement mortar consisting of one part Portland cement and one part sand.

Concrete Duct Conduits.—These are usually constructed by placing collapsible mandrels of wood or metal in a trench where the ducts are desired and then filling the trench with concrete. After the concrete has solidified, the mandrels are taken out in pieces, leaving continuous longitudinal holes which serve as ducts. Some builders produce a similar result by placing tubes of sheet iron or zinc in the concrete as it is being filled into the trench. These tubes have just enough strength to withstand the pressure to which they are subjected, and are, therefore, very thin and liable to be quickly destroyed by corrosion, but the ducts formed by them will always remain unimpaired in the hardened mass of concrete.

Wooden Duct Conduits.—In this type of conduit, the ducts are formed of wooden pipe, troughing, or boxes, and constitute the simplest and cheapest form of conduit. A pipe conduit consists of pieces of wood about 4½ inches square, and three to six feet long, with a round hole about three inches in diameter bored through them longitudinally. As shown by [fig. 979] a cylindrical projection is turned on one end of each section, which, when the conduit is laid fits into a corresponding recess in one end of the next section. The sections are usually laid in tiers, those of one tier breaking joint with those of the tiers above or below.

Fig. 978.—Method of laying multiple duct vitrified clay conduit. The sections are centered by the dowel pins shown in the cut.

The trough conduit consists of ducts about 3 inches square made of horizontal boards and vertical partitions, usually of yellow pine about one inch in thickness. This form of conduit can be laid in convenient lengths of 10 or 12 feet, or it can be built along continuously.

Ques. What is the objection to the use of wood for conduits?

Ans. The decay of the wood tends to form acid which corrodes the lead sheath of the cable.

Fig. 979.—Wooden pipe type of conduit. It consists of pieces of wood about 4½ inches square, and three to six feet long, with a wide hole about three inches in diameter, bored through them longitudinally.

Ques. How can this be prevented?

Ans. The decay of the wood can be prevented to a certain extent by the application of sterilizing processes, thereby preserving it in fairly good condition for about ten to fifteen years.

Ques. For what service is wooden conduit best adapted?

Ans. For temporary installations which will be discontinued before the wood decays.

Wooden Built-in Conduits.—Within recent years several forms of wooden built-in conduit have been designed and successfully used for permanent work. They possess several advantages over any of the duct systems, the chief of which are high insulating quality, the capability of using bare wire and rods for underground conductors, and reduced cost. An approved form of wooden built-in conduit is shown in [fig. 980].

Fig. 980.—Perspective view of wooden built-in conduit. It consists of an outer rectangular casing of wood which is lined inside with impregnated felt.

Ques. How are wooden built-in conduits installed?

Ans. A wooden trough is laid in a trench about 18 inches deep. Porcelain carriers as shown in figs. 981 and 982 are placed in the trough at intervals of 4 to 5 feet, to act as bridgework for supporting the conductors. This bridgework is placed on and is surrounded by impregnated felt or similar material, and the spaces between the carriers, after the conductors have been placed in position on them, is filled with voltax, which hardens rapidly and forms a solid insulating material throughout the conduit.

Wrought Iron or Steel Pipe Conduits.—These are formed of pipes similar to gas or steam pipes, with screw or other connections. They are laid either simply in the earth, or in hydraulic cement, and are the strongest and one of the most satisfactory forms of underground conduit. An appropriate standard of this kind of work is shown in [fig. 983].

Figs. 981 and 982.—Porcelain bridgework or carriers for supporting underground conductors.

Ques. What is the ordinary method of construction?

Ans. A trench, the width of which will depend upon the number of pipes to be laid, is first dug in the ground, and after its bottom has been carefully leveled, is braced with side planking and filled to the depth of two to four inches with a layer of good concrete, consisting of two parts of Rosendale cement, three parts of sand, and five parts of broken stone capable of passing through a one and one-half inch mesh. This concrete is well secured in place and forms the bed for the lowermost layer or tier of pipes. Ordinary wrought iron pipe is employed, in 20 foot lengths about three to four inches in diameter, depending upon the size and number of cables they are intended to carry. After the last tier of pipes have been put in place, and a layer of concrete from two to four inches placed over it, a layer of two inch yellow pine planking is laid over the whole.

Fig. 983.—Cross section of wrought iron pipe conduit laid in hydraulic cement.

The pipe connections consist of a taper screw thread coupling which can be easily made up as the pipes are laid, and which forms a tight joint.

The pipes in each tier are usually laid from ½ to ¾ of their diameter apart, and when the first tier is in place, the spaces between and around the pipes are filled in with concrete which is carried up over the pipes to a depth of about one-half a diameter to form the bed for another tier of pipes.

Ques. What is the principal object of the top covering of planks.

Ans. To protect the conduit against the tools of workmen making later excavations.

Practical experience shows that workmen will dig through concrete without stopping to investigate as to the character of the obstruction, but under similar circumstances, will invariably turn away from wood.

Ques. How are the pipes treated before being laid?

Ans. They are dipped in tar to protect the outside surface from rust.

Ques. What is the most satisfactory form of lined iron pipe?

Ans. Pipe lined with cement. The internal surfaces of these pipes are usually covered with a lining of pure Rosendale cement about ⅝ inch thick and containing no sand. The internal surface of the cement lining does not offer much friction to the introduction or withdrawal of the conductors.

These pipes are laid in cement or concrete in the same manner as plain iron pipe, and are given a coating of tar on the outside to prevent rusting.

Cast Iron Pipe and Trough Conduit.—Cast iron pipe for underground conduits is similar to ordinary wrought iron pipe, except that it is thicker. The additional thickness is necessary to make the strength equal to that of wrought iron; it is therefore heavier to handle and more expensive.

Ques. Describe a cast iron trough conduit.

Ans. It consists of shallow troughs of cast iron in six foot lengths, laid directly in the earth so as to form a system of continuous troughing in which the conductors are placed and then covered over by cast iron covers which are bolted to the trough.

Ques. What advantages does this form of conduit possess over the duct type?

Ans. First, the cables can be laid directly in place, thus eliminating any chance of injury during the process of drawing in, and second, the cables are easily accessible at any point by simply removing one or two of the sectional cast iron covers, thus permitting of their being readily inspected and repaired.

Fig. 984.—Fibre conduit. It consists of pipes made of wood pulp, having about the same thickness as cast iron pipe. Slip joint conduit for electrical subways is three inches inside diameter. The socket joints keep the lengths centered and make it easier to lay than a mere butt joint. It is laid in cement like iron pipe.

Branch connections can be made with greater facility than in the case of any duct system, so that it is especially suitable for distribution systems were it not for the fact that it is so expensive as to be practically prohibitive.

Fibre Conduits.—This form of conduit consists of pipes made of wood pulp impregnated with a bituminous preservative and insulating compound. These pipes are laid in concrete in a manner similar to iron pipe. Fibre conduits are made in sizes ranging from 1 inch to 4 inches in diameter and from 2½ to 5 feet in length, with walls ranging from ¼ to ½ inch in thickness.

Ques. Name the three types of fibre conduit.

Ans. The socket joint type, as shown in [fig. 984], the sleeve type, [fig. 985], and the screw joint type, [fig. 986].

Ques. What is the usual method of laying the socket joint type of fibre conduit?

Ans. After the trench has been dug to the required width and depth, depending upon the number or pipes to be placed in a tier and the number of tiers, a bed of concrete about 3 inches deep is placed on the bottom and a line drawn on one side for the alignment of the first line of pipes. The other lines of pipe or ducts are laid parallel to the first line, and are separated from it and from each other by means of ¼ inch or ½ inch wooden or iron pegs. The pipes are well grouted and covered with a layer of concrete to the depth of ¼ or ½ inch, and the next tier laid in place in the same manner. When the final tier of pipes has been installed, it is covered with a layer of concrete about 2 to 3 inches deep.

Fig. 985.—Sleeve joint type of fibre. Both the socket type ([fig. 984]), and the sleeve type here shown are easily aligned without the use of a mandrel.

Ques. What is done when it is necessary to cut a length of pipe to break joints, or to enter a manhole?

Ans. The remaining part of the length may be utilized by using a fibre conduit sleeve having an inside diameter ½ inch greater than the pipe being used on the system.

These sleeves are furnished by the manufacturers at a nominal charge per foot. They are about four inches in length and fit over the ends of the abutting pipes, so that they make tight joints and give perfect alignment.

Although its employment is not permitted where fireproof regulations are in force, fibre conduit is now being extensively used in other places, and is giving satisfactory service. It is not affected by moist earth and is impervious to the action of acids, alkalies, and gases. As it is not subject to expansion and contraction, leakage is practically eliminated, and since it is a very good insulator, troubles due to stray currents are reduced to a minimum. It is extremely light, comparatively non-breakable, and can be accurately laid at the rate of 12,000 duct feet per day by a gang of common laborers, consisting of two layers and three helpers.

Fig. 986.—Screw joint type of fibre conduit. This method of connection will form a tight line and is suitable for running under the lawns of private houses and parks, under the streets of towns and villages, and in other places where the cost of building electric subways is prohibitive.

Edison Tube System.—Of the various built in or solid underground conduit systems other than those already described under wooden conduit systems, the most satisfactory are the Edison tube system, the Crompton naked conductor system, the Kennedy system, which is a modification of the Crompton and the Callender systems.

Ques. Describe the Edison tube system.

Ans. It consists of a series of iron tubes or pipes containing one or more copper conductors which are placed therein before each complete section or pipe leaves the factory, so that they only need to be joined together to form a continuous line of underground conduit with conductors in place. The arrangement of wires and the details of the Edison tube system are shown in figs. 987 to 989.

Fig. 987.—Cross section of Edison "feeder" tube. This runs from the power station to the centers of distribution, and contains two principal conductors and a smaller conductor to serve as a neutral wire, and also three insulated cables of seven strands of No. 19 B. W. G. wire each. These cables form independent circuits and enable the voltages at the distant end of the feeder to be read at the central station. For this reason they are commonly called pressure wires.