CHAPTER LXVII

MANAGEMENT

The term "management," broadly speaking, includes not only the actual skilled attention necessary for the proper operation of the machines, after the plant is built, but also other duties which must be performed from its inception to completion, and which may be classified as

1. Selection; 2. Location; 3. Erection; 4. Testing; 5. Running; 6. Care; 7. Repair.

That is to say, someone must select the machinery, determine where each machine is to be located, install them, and then attend to the running of the machines and make any necessary repairs due to the ordinary mishaps likely to occur in operation.

These various duties are usually entrusted to more than one individual; thus, the selection and location of the machinery is done by the designer of the plant, and requires for its proper execution the services of an electrical engineer, or one possessing more than simply a practical knowledge of power plants.

The erection of the machines is best accomplished by those making a specialty of this line of work, who by the nature of the undertaking acquire proficiency in methods of precision and an appreciation of the value of accuracy which is so essential in the work of aligning the machines, and which if poorly done will prove a constant source of annoyance afterward.

The attention required for the operation of the machines, embracing the running care and repair, is left to the "man in charge," who in most cases of small and medium size plants is the chief steam engineer. He must therefore, not only understand the steam apparatus, but possess sufficient knowledge of electrical machinery to operate and maintain it in proper working order.

The present chapter deals chiefly with alternating current machinery, the management of direct current machines having been fully explained in Guide No. 3, however, some of the matter here presented is common to both classes of apparatus.

* * * * *

=Selection.=--In order to intelligently select a machine so that it will properly harmonize with the conditions under which it is to operate, there are several things to be considered.

1. Type; 2. Capacity; 3. Efficiency; 4. Construction.

The general type of machine to be used is, of course, dependent on the system employed, that is, whether it be direct or alternating, single or polyphase.

Thus, the voltage in most cases is fixed except on transformer systems where a choice of voltage may be had by selecting a transformer to suit.

In alternating current constant pressure transmission circuits, an average voltage of 2,200 volts with step down transformer ratios of 1/10 and 1/20 is in general use, and is recommended.

For long distance, the following average voltages are recommended 6,000; 11,000; 22,000; 33,000; 44,000; 66,000; 88,000; and higher, depending on the length of the line and degree of economy desired.

In alternating circuits the standard frequencies are 25, and 60 cycles. These frequencies are already in extensive use and it is recommended to adhere to them as closely as possible.

In fixing the capacity of a machine, _careful consideration should be given to the conditions of operation both_ =present= _and_ =future= in order that the resultant efficiency may be maximum.

Most machines show the best efficiency at or near full load. If the load be always constant, as for instance, a pump forcing water to a given head, it would be a simple matter to specify the proper size of machine, but in nearly all cases, and especially in electrical plants, the load varies widely, not only the daily and hourly fluctuations, but the varying demands depending on the season of the year and growth of the plant's business. All of these conditions tend to complicate the matter, so that intelligent selection of capacity of a machine requires not only calculation but mature judgment, which is only obtained by long experience.

In selecting a machine, or in fact any item connected with the plant _its construction should be carefully considered_.

Standard construction should be insisted upon so that in the event of damage a new part can be obtained with the least possible delay.

The parts of most machines are _interchangeable_, that is to say, with the refined methods of machinery a duplicate part (usually carried in stock) may be obtained at once to replace a defective or broken part, and made with such precision that little or no fitting will be required.

The importance of standard construction cannot be better illustrated than in the matter of steam piping, that is, the kind of fittings selected for a given installation.

With the exception of the exhaust line from engine to condenser, where other than standard construction may sometimes be used to reduce the frictional resistance to the steam, the author would adhere to standard construction except in very exceptional cases. Those who have had practical experience in pipe fitting will appreciate the wisdom of this.

For installations in places remote from large supply houses, the more usual forms of standard fittings should be employed, such as ordinary T's, 45° and 90° elbows, etc.

In such locations, where designers specify the less usual forms of standard fittings such as union fittings, offset reducers, etc., or special fittings made to sketch, it simply means, in the first instance that they usually cannot be obtained of the local dealer, making it necessary to order from some large supply house and resulting in vexatious delays.

As a rule, those who specify special fittings have found that their making requires an unreasonable length of time, and the cost to be several times that of the equivalent in standard fittings.

An examination of a few installations will usually show numerous special and odd shape fittings, which are entirely unnecessary.

Moreover, a standard design, in general, is better than a special design, because the former has been tried out, and any imperfection or weakness remedied, and where thousands of castings of a kind are turned out, a better article is usually the result as compared with a special casting.

In the matter of construction, in addition to the items just mentioned, it should be considered with respect to

1. Quality; 2. Range; 3. Accessibility; 4. Proportion; 5. Lubrication; 6. Adjustment.

It is poor policy, excepting in very rare instances, to buy a "cheap" article, as, especially in these days of commercial greed, the best is none too good.

Perhaps next in importance to quality, at least in most cases, is _range_. This may be defined as _scope of operation_, _effectiveness_, or _adaptability_. The importance of range is perhaps most pronounced in the selection of tools, especially for plants remote from repair shops.

For instance, in selecting a pipe cutter, there are two general classes: wheel cutters, and roller cutters. A wheel cutter has three wheels and a roller cutter one wheel and two rollers, the object of the rollers being to keep the wheel perpendicular to the pipe in starting the cut and to reduce burning. It must be evident that in operation, a roller cutter requires sufficient room around the pipe to permit making a complete revolution of the cutter, whereas, with a wheel cutter, the work may be done by moving the cutter back and forth through a small arc, as illustrated in figs. 2,786 and 2,787. Thus a wheel cutter has a _greater range_ than a roll cutter.

Range relates not only to ability to operate in inaccessible places but to the various operations that may be performed by one tool.

PROPERTIES OF STANDARD WROUGHT IRON PIPE

--------------------------+------+-----------------+ | | | | | | | | | | | | Diameter |Thick-| Circumference. | | ness.| | --------+--------+--------+ +--------+--------+ Nominal| Actual | Actual | |External|Internal| internal|external|internal| | | | --------+--------+--------+------+--------+--------+ Inches | Inches | Inches |Inches| Inches | Inches | --------+--------+--------+------+--------+--------+ ⅛ | .405 | .27 | .068 | 1.272 | .848 | ¼ | .54 | .364 | .088 | 1.696 | 1.144 | ⅜ | .675 | .494 | .91 | 2.121 | 1.552 | ½ | .84 | .623 | .109 | 2.639 | 1.957 | ¾ | 1.05 | .824 | .113 | 3.299 | 2.589 | 1 | 1.315 | 1.048 | .134 | 4.131 | 3.292 | 1¼ | 1.66 | 1.38 | .14 | 5.215 | 4.335 | 1½ | 1.9 | 1.611 | .145 | 5.969 | 5.061 | 2 | 2.375 | 2.067 | .154 | 7.461 | 6.494 | 2½ | 2.875 | 2.468 | .204 | 9.032 | 7.753 | 3 | 3.5 | 3.067 | .217 | 10.996 | 9.636 | 3½ | 4. | 3.548 | .226 | 12.566 | 11.146 | 4 | 4.5 | 4.026 | .237 | 14.137 | 12.648 | 4½ | 5. | 4.508 | .246 | 15.708 | 14.162 | 5 | 5.563 | 5.045 | .259 | 17.477 | 15.849 | 6 | 6.625 | 6.065 | .28 | 20.813 | 19.054 | 7 | 7.625 | 7.023 | .301 | 23.955 | 22.063 | 8 | 8.625 | 7.982 | .322 | 27.096 | 25.076 | 9 | 9.625 | 8.937 | .344 | 30.238 | 28.076 | 10 | 10.75 | 10.019 | .366 | 33.772 | 31.477 | 11 | 12. | 11.25 | .375 | 37.699 | 35.343 | 12 | 12.75 | 12. | .375 | 40.055 | 37.7 | --------+--------+--------+------+--------+--------+ ------+-------------------------+---------------+--------+-------+------ | | | | | | | | | | | | | Length | | | |Length of pipe |of pipe |Nominal| Diam.| Transverse areas. | per square |contain-|weight |Number | | foot of |ing one | per | of ------+--------+--------+-------+-------+-------+ cubic | foot. |thread Nom. |External|Internal| Metal |Ext'nal|Int'nal| foot. | | per intern| | | |surface|surface| | | inch ------+--------+--------+-------+-------+-------+--------+-------+ of Inches|Sq. ins.|Sq. ins.|Sq.ins.| Feet | Feet | Feet |Pounds |screw ------+--------+--------+-------+-------+-------+--------+-------+----- ⅛ | .129 | .0573| .0717| 9.44 |14.15 |2513. | .241 | 27 ¼ | .229 | .1041| .1249| 7.075 |10.49 |1383.3 | .42 | 18 ⅜ | .358 | .1917| .1663| 5.657 | 7.73 | 751.2 | .559 | 18 ½ | .554 | .3048| .2492| 4.547 | 6.13 | 472.4 | .837 | 14 ¾ | .866 | .5333| .3327| 3.637 | 4.635 | 270. | 1.115 | 14 1 | 1.358 | .8626| .4954| 2.904 | 3.645 | 166.9 | 1.668 | 11½ 1¼ | 2.164 | 1.496 | .668 | 2.301 | 2.768 | 96.25 | 2.244 | 11½ 1½ | 2.835 | 2.038 | .797 | 2.01 | 2.371 | 70.66 | 2.678 | 11½ 2 | 4.43 | 3.356 | 1.074 | 1.608 | 1.848 | 42.91 | 3.609 | 11½ 2½ | 6.492 | 4.784 | 1.708 | 1.328 | 1.547 | 30.1 | 5.739 | 8 3 | 9.621 | 7.388 | 2.243 | 1.091 | 1.245 | 19.5 | 7.536 | 8 3½ | 12.566 | 9.887 | 2.679 | .955 | 1.077 | 14.57 | 9.001 | 8 4 | 15.904 | 12.73 | 3.174 | .849 | .949 | 11.31 |10.665 | 8 4½ | 19.635 | 15.961 | 3.674 | .764 | .848 | 9.02 |12.34 | 8 5 | 24.306 | 19.99 | 4.316 | .687 | .757 | 7.2 |14.502 | 8 6 | 34.472 | 28.888 | 5.584 | .577 | .63 | 4.98 |18.762 | 8 7 | 45.664 | 38.738 | 6.926 | .501 | .544 | 3.72 |23.271 | 8 8 | 58.426 | 50.04 | 8.386 | .443 | .478 | 2.88 |28.177 | 8 9 | 72.76 | 62.73 |10.03 | .397 | .427 | 2.29 |33.701 | 8 10 | 90.763 | 78.839 |11.924 | .355 | .382 | 1.82 |40.065 | 8 11 |113.098 | 99.402 |13.696 | .318 | .339 | 1.450|45.95 | 8 12 |127.677 |113.098 |14.579 | .299 | .319 | 1.27 |48.985 | 8 ------+--------+--------+-------+-------+-------+--------+-------+----

Open construction should be employed, wherever possible, so that all parts of a machine that require attention, or that may become deranged in operation, may be accessible for adjustment or repair.

The design should be such that there is ample strength, and the bearings for moving parts should be of liberal proportions to avoid heating with minimum attention.

A comparison of the proportions used by different manufacturers for a machine of given size might profitably be made before a selection is made.

The matter of lubrication is important.

Fast running machines, such as generators and motors, should be provided with ring oilers and oil reservoirs of ample capacity, as shown in figs. 2,788 to 2,794.

All bearings subject to appreciable wear should be made adjustable so that lost motion may be taken up from time to time and thus keep the vibration and noise of operation within proper limits.

=Selection of Generators.=--This is governed by the class of work to be done and by certain local conditions which are liable to vary considerably for different stations.

These variable factors determine whether the generators must be of the direct or alternating current type, whether they must be wound to develop a high or a low voltage, and whether their outputs in amperes must be large or small. Sufficient information has already been given to cover these various cases; there are, however, certain general rules that may advantageously be observed in the selection of generators designed to fill any of the aforementioned conditions, and it is well to possess certain facts regarding their construction.

=Ques. Name an important point to be considered in selecting a generator.=

Ans. Its efficiency.

=Ques. What are the important points with respect to efficiency?=

Ans. A generator possessing a high efficiency at the average load is more desirable than a generator showing a high efficiency at full load.

=Ques. Why?=

Ans. The reason is that in station practice the full load limit is seldom reached, the usual load carried by a generator ordinarily lying between the one-half and three-quarter load points.

=Ques. How do the efficiencies of large and small generators compare?=

Ans. There is little difference.

=Ques. How are the sizes and number of generator determined?=

Ans. The sizes and number of generator to be installed should be such as to permit the engines operating them being worked at nearly full load, because the efficiencies of the latter machines decrease rapidly when carrying less than this amount.

=Ques. What is understood by regulation?=

Ans. The accuracy and reliability with which the pressure or current developed in a machine may be controlled.

It is generally possible if purchasing of a reputable concern, to obtain access to record sheets on which may be found results of tests conducted on the generator in question, and as these are really the only means of ascertaining the values of efficiency and regulation, the purchaser has a right to inspect them. If, for some reason or other, he has not been afforded this privilege, he should order the machine installed in the station on approval, and test its efficiency and regulation before making the purchase.

=Installation.=--The installation of machines and apparatus in an electrical station is a task which increases in difficulty with the size of the plant. When the parts are small and comparatively light they may readily be placed in position, either by hand, by erecting temporary supports which may be moved from place to place as desired, or by rolling the parts along on the floor upon pieces of iron pipe. If, however, the parts be large and heavy, a traveling crane such as shown in fig. 2,797, becomes necessary.

=Ques. What precaution should be taken in moving the parts of machines?=

Ans. Care should be taken not to injure the bearings and shafts, the joints in magnetic circuits such as those between frame and pole pieces, and the windings on the field and armature.

The insulations of the windings are perhaps the most vital parts of a generator, and the most readily injured. The prick of a pin or tack, a bruise, or a bending of the wires by resting their weight upon them or by their coming in contact with some hard substance, will often render a field coil or an armature useless.

Owing to its costly construction, it is advisable when transporting armatures by means of cranes to use a wooden spreader, as shown in fig. 2,798 to prevent the supporting rope bruising the winding.

=Ques. If an armature cannot be placed at once in its final position what should be done?=

Ans. It may be laid temporarily upon the floor, if a sheet of cardboard or cloth be placed underneath the armature as a protection for the windings; in case the armature is not to be used for some time, it is better practice to place it in a horizontal position on two wooden supports near the shaft ends.

=Ques. What kind of base should be used with a belt driven generator or motor?=

Ans. The base should be provided with V ways and adjusting screws for moving the machine horizontally to take up slack in the belt, as shown in fig. 2,799.

Owing to the normal tension on the belt, there is a moment exerted equal in amount to the distance from the center of gravity of the machine to the center of the belt, multiplied by the effective pull on the belt. This force tends to turn the machine about its center of gravity. By placing the screws as shown, any turning moment, as just mentioned, is prevented.

=Ques. How should a machine be assembled?=

Ans. The assembling should progress by the aid of a blue print, or by the information obtained from a photograph of the complete machine as it appears when ready for service. Each part should be perfectly clean when placed in position, especially those parts between which there is friction when the machine is in operation, or across which pass lines of magnetic force; in both cases the surfaces in contact must be true and slightly oiled before placing in position.

Contact surfaces forming part of electrical circuits must also be clean and tightly screwed together. An important point to bear in mind when assembling a machine is, to so place the parts that it will not be necessary to remove any one of them in order to get some other part in its proper position. By remembering this simple rule much time will be saved, and in the majority of instances the parts will finally be better fitted together than if the task has to be repeated a number of times.

When there are two or more parts of the machine similarly shaped, it is often difficult to properly locate them, but in such cases notice should be taken of the factory marks usually stamped upon such pieces and their proper places determined from the instructions sent with the machine.

=Ques. What should be noted with respect to speed of generator?=

Ans. Each generator is designed to be run at a certain speed in order to develop the voltage at which the machine is rated. The speed, in revolutions per minute, the pressure in volts, and the capacity or output in watts (volts × amperes) or in kilowatts (thousands of watts) are generally stamped on a nameplate screwed to the machine.

This requirement frequently requires calculations to be made by the erectors to determine the proper size pulleys to employ to obtain the desired speed.

=Example.=--What diameter of engine pulley is required to run a dynamo at a speed of 1,450 revolutions per minute the dynamo pulley being 10 inches in diameter and the speed of engine, 275 revolutions per minute?

The diameter of pulley required on engine is 10 × (1,450 ÷ 275) = 53 inches, nearly.

=Rule.=--To find the diameter of the driving pulley, _multiply the speed of the driven pulley by its diameter, divide the product by the speed of the driver and the answer will be the size of the driver required_.

_Example._--If the speed of an engine be 325 revolutions per minute, diameter of engine pulley 42 inches, and the speed of the dynamo 1,400 revolutions per minute, how large a pulley is required on dynamo?

The size of the dynamo pulley is

42 × (325 ÷ 1,400) = 9¾ inches.

=Rule.=--To find the size of dynamo pulley, _multiply the speed of engine by the diameter of engine wheel and divide the product by the speed of the dynamo_.

_Example._--If a steam engine, running 300 revolutions per minute, have a belt wheel 48 inches in diameter, and be belted to a dynamo having a pulley 12 inches in diameter, how many revolutions per minute will the dynamo make?

The speed of dynamo will be 300 × (48 ÷ 12) = 1,200 rev. per min.

=Rule.=--When the speed of the driving pulley and its diameter are known, and the diameter of the driven pulley is known, the speed of the driven pulley is found by _multiplying the speed of the driver by its diameter in inches and dividing the product by the diameter of the driven pulley_.

=Example.=--What will be the required speed of an engine having a belt wheel 46 inches in diameter to run a dynamo 1,500 revolutions per minute, the dynamo pulley being 11 inches in diameter?

The speed of the engine is 1,500 × (11 ÷ 46) = 359 rev. per min. nearly.

=Rule.=--To find the speed of engine when diameter of both pulleys, and speed of dynamo are given, _multiply the dynamo speed by the diameter of its pulley and divide by the diameter of engine pulley_.

=Ques. How are the diameters and speeds of gear wheels figured?=

Ans. The same as belted wheels, using either the pitch circle diameters or number of teeth in each gear wheel.

=Ques. What should be noted with respect to generator pulleys?=

Ans. A pulley of certain size is usually supplied with each generator by its manufacturer, and it is not generally advisable to depart much from the dimensions of this pulley. Accordingly, the solution of the pulley problem usually consists in finding the necessary diameter of the driving pulley relative to that of the pulley on the generator in order to furnish the required speed.

=Ques. What is the chief objection to belt drive?=

Ans. The large amount of floor space required.

=Ques. How may the amount of space that would ordinarily be required for belt drive, be reduced?=

Ans. By driving machines in tandem as in fig. 2,810, or by the double pulley drive as in fig. 2,811.

=Ques. What is the objection to the tandem method?=

Ans. The most economical distance between centers cannot be employed for all machines.

=Ques. What is the objectionable tendency in resorting to floor economy methods with belt transmission?=

Ans. The tendency to place the machines too closely together. This is poor economy as it makes the cleaning of the machines a difficult and dangerous task; it is therefore advisable to allow sufficient room for this purpose regardless of the method of belting employed.

=Ques. What is the approved location for an alternator exciter?=

Ans. To economize floor space the exciter may be placed between the alternator and engine at S in fig. 2,811.

=Belts.=--In the selection of a belt, the quality of the leather should be first under consideration. The leather must be firm, yet pliable, free from wrinkles on the grain or hair side, and of an even thickness throughout.

If the belt be well selected and properly handled, it should do service for twenty years, and even then if the worn part be cut off, the remaining portion may be remade and used again as a narrower and shorter belt.

Besides leather belts, there are those made of rubber which withstand moisture much better than leather belts, and which also possess an excellent grip on the pulley; they are, however, more costly and much less durable under normal conditions.

In addition to leather and rubber belts, there are belts composed of cotton, of a combination of cotton and leather, and of rope. The leather belt, however, is the standard and is to be recommended.

Equally important with the quality of a belt is its size in order to transmit the necessary power.

The average strain under which leather will break has been found by many experiments to be 3,200 pounds per square inch of cross section. A good quality of leather will sustain a somewhat greater strain. In use on the pulleys, belts should not be subjected to a greater strain than one eleventh their tensile strength, or about 290 pounds to the square inch or cross section. This will be about 55 pounds average strain for every inch in width of single belt three-sixteenths inch thick. The strain allowed for all widths of belting--single, light double, and heavy double--is in direct proportion to the thickness of the belt.

=Ques. How much horse power will a belt transmit?=

Ans. The capacity of a belt depends on, its width, speed, and thickness. _A single belt one inch wide and travelling 1,000 feet per minute will transmit one horse power; a double belt under the same conditions, will transmit two horse power._

This corresponds to a working pull of 33 and 66 lbs. per inch of width respectively.

=Example.=--What width double belt will be required to transmit 50 horse power travelling at a speed of 3,000 feet per minute?

The horse power transmitted by each inch width of double belt travelling at the stated speed is

3,000 1 × ----- × 2 = 6, 1,000

hence the width of belt required to transmit 50 horse power is

50 ÷ 6 = 8.33, say 8 inches.

=Ques. At what velocity should a belt be run?=

Ans. At from 3,000 to 5,000 feet per minute.

=Ques. How may the greatest amount of power transmitting capacity be obtained from belts?=

Ans. By covering the pulleys with leather.

=Ques. How should belts be run?=

Ans. With the tight side underneath as in fig. 2,814.

=Ques. What is a good indication of the capacity of a belt in operation?=

Ans. Its appearance after a few days' run.

If the side of the belt coming in contact with the pulley assume a mottled appearance, it is an indication that the capacity of the belt is considerably in excess of the power which it is transmitting, inasmuch as the spotted portions of the belt do not touch the pulley; and in consequence of this there is liable to be more or less slipping.

Small quantities of a mixture of tallow and fish oil which have previously been melted together in the proportion of two of the former to one of the latter, will, if applied to the belt at frequent intervals, do much toward softening it, and thus by permitting its entire surface to come in contact with the pulley, prevent any tendency toward slipping. The best results are obtained when the smooth side of the belt is used next to the pulley, since tests conducted in the past prove that more power is thus transmitted, and that the belt lasts longer when used in this way.

=Ques. What is the comparison between the so called endless belts and laced belts?=

Ans. With an endless belt there is no uneven or noisy action as with laced belts, when the laced joint passes over the pulleys, and the former is free from the liability of breakage at the joint.

=Ques. How should a belt be placed on the pulleys?=

Ans. The belt should first be placed on the pulley at rest, and then run on the other pulley while the latter is in motion.

The best results are obtained, and the strain on the belt is less, when the speed at which the moving pulley revolves is comparatively low. With heavy belts, particular care should be taken to prevent any portion of the clothing being caught either by the moving belt or pulleys, as many serious accidents have resulted in the past from carelessness in regard to this important detail. The person handling the belt should, therefore, be sure of a firm footing, and when it is impossible to secure this, it is advisable to stop the engine and fit the belt around the engine pulley as well as possible by the aid of a rope looped around the belt.

=Ques. Under what conditions does a belt drive give the best results?=

Ans. When the two pulleys are at the same level.

If the belt must occupy an inclined position it should not form a greater angle than 45 degrees with the horizontal.

=Ques. What is a characteristic feature in the operation of belts, and why?=

Ans. Belts in motion will always run to the highest side of a pulley; this is due partially to the greater speed in feet per minute developed at that point owing to the greater circumference of the pulley, and also to the effects of centrifugal force.

If, therefore, the highest sides of both pulleys be in line with each other, and the shafts of the respective pulleys be parallel to each other, there will be no tendency for the belt to leave the pulleys when once in its proper position. In order that these conditions be maintained, the belt should be no more than tight enough to prevent slipping, and the distance between the centers of the pulleys should be approximately 3.5 times the diameter of the larger one.

=Ques. What minor appurtenances should be provided in a station?=

Ans. Apparatus should be installed as a prevention against accidents, such as fire, and protection of attendants from danger.

In every electrical station there should be a pump, pipes and hose; the pump may be either directly connected to a small electric motor or belted to a countershaft, while the pipes and hose should be so placed that no water can accidentally reach the generators and electrical circuits. A number of fire bucket filled with water should be placed on brackets around the station, and with these there should be an equal number of bucket containing dry sand, the water being used for extinguishing fire occurring at a distance from the machines and conductors, and the sand for extinguishing fire in current carrying circuits where water would cause more harm than benefit. To prevent the sand being blown about the station, each sand bucket, when not in use, should be provided with a cover.

Neat cans and boxes should be mounted in convenient places for greasy rags, waste, nuts, screws, etc., which are used continually and which therefore cannot be kept in the storeroom.

While it is important to guard against fire in the station, it is equally necessary to provide for personal safety. All passages and dark pits should therefore be thoroughly lighted both day and night, and obstacles of any nature that are not absolutely necessary in the operation of the station, should be removed. Moving belts, and especially those passing through the floor, should be enclosed in iron railings. If high voltages be generated, it is well to place a railing about the switchboard to prevent accidental contact with current carrying circuits, and in such cases it is also advisable to construct an insulated platform on the floor in front of the switchboard.

=Switchboards.=--The plan of switchboard wiring for alternating current work depends upon the system in use and this latter may be either of the single phase, two phase, three phase, or monocyclic types. The general principles in all these cases, however, are practically identical.

Fig. 2,820 shows the switchboard wiring for a single phase alternator. As an aid in reading the diagram, the conductors carrying alternating current are represented by solid lines, and those carrying direct current, by dotted lines.

The exciter shown at the right is a shunt wound machine. By means of the exciter rheostat, the voltage for exciting the field winding of the alternator is varied; this, in turn, varies the voltage developed in the alternator since the main leads of the exciter are connected through a double pole switch G to the field winding of the alternator.

A rheostat is also introduced in the alternator field winding circuit to adjust the alternator pressure. It may seem unnecessary to employ a rheostat in each of two separate field circuits to regulate the voltage of the alternator, but these rheostats are not both used to produce the same result. When a considerable variation of pressure is required, the exciter rheostat is manipulated, whereas for a fine adjustment of voltage the alternator rheostat is preferably employed.

Sometimes a direct current ammeter is introduced in the alternator's field circuit to aid in the adjustment.

The main circuit of alternator after being protected on both sides by fuses, runs to the double pole switch K. These fuses serve as a protection to the alternator in case of a short circuit at the main switch. It will be noticed the fuses are of the single pole type and are mounted a considerable distance apart; this is to prevent any liability of a short circuit between them in case of action. Enclosed fuses are now used entirely for such work, since in these there is no danger of heated metal being thrown about and causing damage when the fuse wire is melted. Enclosed fuses are also more readily and quickly replaced than open fuses, the containing tube of each being easy to adjust in circuit, and when the fuse wire within is once melted the tube is discarded for a new one.

The main circuit after passing through the main switch is further protected on both sides by circuit breakers. Leaving these protective devices, the left hand side of the circuit includes the alternating current ammeter, and then connects with one of the bus bars. The right hand side of the circuit runs from the circuit breaker to the other bus bar. As many feeder circuits may be connected to the bus bars and supplied with current by the alternator as the capacity of this machine will permit. If, however, there be more than one feeder circuit, each must be wired through a double pole switch.

In alternating current work the pressures dealt with are much greater than those in direct current installations, so that proportionate care must be taken in the wiring to remove all possibility of grounds.

To locate such troubles, however, should they occur, a ground detector is provided. For this class of work the ground detector must be an instrument especially designed for high pressure circuits. Two of its terminals should be connected to the line wires and the third, to ground; in case of a leak on the line, a current will then flow through the detector and by the position of the pointer the location and seriousness of the leak may be judged.

A step down transformer is also rendered necessary for the voltmeter and the pilot lamps, owing to the high voltage in use. The primary winding of the transformer is connected across the main circuit of the alternator. This connection should never be made so that it will be cut out of circuit when the main switch is open, for it is always advisable to consult the voltmeter before throwing on the load by closing this switch.

=Ques. How does the switchboard wiring for a two phase system differ from the single phase arrangement shown in fig. 2,820?=

Ans. It is practically the same, except for the introduction of an extra ammeter and a compensator in each of the outside wires, and in the use of a four pole switch in place of the two pole main switch.

The ammeters, of course, are for measuring the alternating currents in each of the two phases or legs of the system, and the compensators are two transformers with their primary coils in series with the outside wires and their secondary coils in series with each other across the outside wires. The transformers thus connected are known as compensators or pressure regulators, and as such compensate for the drop in pressure on either side of the system.

=Ques. How is the four pole main switch wired?=

Ans. Its two central terminals which connect directly with the line wires, are joined together by a conductor, and from this point one wire is led off. This wire, together with the two outside wires, form the feeders of the system.

=Ques. How many voltmeters are required for the two phase system?=

Ans. One voltmeter is sufficient on the board if a proper switching device be employed to shift its connections across either of the two circuits; otherwise, two voltmeters will be necessary, one bridged across each of these respective circuits.

The same reasoning holds true in regard to ground detectors, so that one or two of these will be required, depending upon the aforementioned conditions.

=Ques. What are the essential points of difference between the single phase switchboard wiring as shown in fig. 2,820, and that required for a three wire three phase system?=

Ans. The three phase system requires the use of a three pole switch in place of the two pole switch; the insertion of an ammeter, a circuit breaker, and a compensator in each of the three wires of the system; the presence of two ground detectors instead of one, and the addition of a voltmeter switch if but one voltmeter be provided, or else the installation of two voltmeters, connected the one between the middle wire and outer right hand wire, and the other between the middle wire and outer left hand wire.

=Ques. Mention a few points relating to lightning arresters.=

Ans. In most cases where direct current is used they are mounted on the walls of the station near the place at which the line wires enter. If they be mounted outside the station at this point, special precautions should be taken to keep them free from moisture by enclosing them in iron cases, but no matter where they are located it is necessary that they be dry in order to work properly.

If possible, one place should be set aside for them and a marble or slate panel provided on which they may be mounted.

Wooden supports are undesirable for lightning arresters on account of the fire risk incurred; this, however, may be reduced to a minimum by employing skeleton boards and using sheets of asbestos between the arresters and the wood.

In parts of the country where lightning is of common occurrence and where overhead circuits are installed which carry high pressures, heavy currents, and extend over considerable territory, it is advisable to have the station well equipped with lightning arresters of the most improved types.

In each side of the main circuit, between the lightning arrester connections and the switchboard apparatus there should be connected a choke coil or else each of the main conductors at this point should be tightly coiled up part of its length to answer the same purpose.

A quick and effective way of coiling up a wire consists in wrapping around a cylindrical piece of iron or wood that part of the conductor in which it is desired to have the coils, the desired number of times, and then withdrawing the cylindrical piece. The coils, each of which may contain 50 or 200 turns, thus inserted in the main circuit introduce a high resistance or reluctance to a lightning current, and thus prevent it passing to the generator; there will, however, be an easy path to earth afforded it through the lightning arrester, and so no damage will be done. Coils of the nature just mentioned may advantageously be introduced between the generator and switchboard to take up the reactive current developed upon the opening of the circuit, and in the case of suspended conductors, the coils may be used to take up the slack by the spring-like effect produced by them.

The safety of the operator should be especially considered in the design of high pressure alternating current switchboards.

Such protection may be secured by screening all the exposed terminals, or preferably by mounting all the switch mechanism on the back of the board with simply the switch handle projecting through to the front; by pushing or pulling the switch handle, the connections can thus be shifted either to one side of the system or to the other.

=Ques. Upon what does the work of assembling a switchboard depend?=

Ans. It depends almost entirely upon the size of the plant, varying from the simple task of mounting a single panel in the case of an isolated plant, to the more difficult problem of supporting a large number of panels in a central station.

=Ques. When the material chosen for a switchboard must be shipped a considerable distance, what form of board should be used?=

Ans. The board units or "slabs" should be of small dimensions, to avoid the liability of breakage and expense of renewal when a unit becomes cracked or machine injured.

Ordinarily, switchboards vary from five to eight feet in height and the widths of the panels vary from five to six feet. In some boards the seams between the slabs run vertically, and in others horizontally. In order to render the assembling of the switchboard as simple as possible, and its appearance when finished the most artistic, these seams should run horizontally rather than vertically. The edges of each of the slabs should also be chamfered so that there will be less danger of their breaking out when being mounted on the framework.

=Ques. In assembling a switchboard, how should the lower slabs be placed, and why?=

Ans. They should be suspended a little distance from the floor to prevent contact with any oil, dirt, water or rubbish that might be on the floor.

=Ques. How are the slabs or panels supported?=

Ans. They are carried on an iron or wooden framework with braces to give stability.

The braces should be securely fastened at one end to the wall of the station, and at the other end to the framework of the board, as shown in fig. 2,836.

To fasten the switchboard end of the brace directly to the slate, marble or other material composing the board is poor practice and should never be attempted.

If the station be constructed of iron, these switchboard braces must be such that they will thoroughly insulate the board and its contents from the adjoining wall.

=Ques. What is the usual equipment of a switchboard?=

Ans. It comprises switching devices, current or pressure limiting devices, indicating devices, and fuses for protecting the apparatus and circuits.

On some switchboards are also mounted small transformers for raising or lowering the voltages, and lightning arresters as a protection from lightning. In addition to the apparatus previously mentioned nearly all switchboards carry at or near their top two or more incandescent lamps provided with shades or reflectors, for lighting the board.

=Ques. What should be done before wiring a switchboard?=

Ans. The electrical connections between the various apparatus mounted on the face or front of the board, are made on the back of the board. It is necessary that these connections be properly made else considerable electrical power will be wasted at this point. The wiring on the back of the board should therefore be planned out on paper before commencing the work.

In laying out the plan of wiring care must be taken to allow sufficient contact surface at each connection; there should be not less than one square inch of contact surface allowed for each 160 amperes of current transmitted.

For the bus bars, which, by the way are always of copper, one square inch per 1,000 amperes is the usual allowance; this is equal to 1,000 circular mils of cross sectional area per ampere.

Every effort should be made to give the bus bars the greatest amount of radiation consistent with other conditions, in order that their resistances may not become excessive owing to the heat developed by the large currents they are forced to carry. Suppose, for instance, the number of amperes to be generated is such as to require bus bars having each a cross sectional area of one square inch. If the end dimensions of these bars were each 1 inch by 1 inch, there would be less radiating surface than if their dimensions were each 2 inches by ½ inch.

=Operation of Alternators.=--The operation of an alternator when run singly differs but little from that for a dynamo.

As to the preliminaries, the exciter must first be started. This is done in the same way as for any shunt dynamo. At first only a small current should be sent through the field winding of the alternator; then, if the exciter operates satisfactorily and the field magnetism of the operator show up well, the load may gradually be thrown on until the normal current is carried, the same method of procedure being followed as in the similar case of a dynamo.

On loading an alternator, a noticeable drop in voltage occurs across its terminals. This drop in voltage is caused in part by the demagnetization of the field magnets due to the armature current, and so depends in a measure upon the position and form of the pole pieces as well as upon those of the teeth in the armature core. The resistance of the armature winding also causes a drop in voltage under an increase of load.

Another cause which may be mentioned is the inductance of the armature winding, which is in turn due to the positions of the armature coils with respect to each other and also with respect to the field magnets.

=Alternators in Parallel.=--When the load on a station increases beyond that which can conveniently be carried by one alternator, it becomes necessary to connect other alternators in parallel with it. To properly switch in a new machine in parallel with one already in operation and carrying load, requires a complete knowledge of the situation on the part of the attendant, and also some experience.

The connections for operating alternators in parallel are shown in fig. 2,843. In the illustration the alternator A is in operation and is supplying current to the bus bars. The alternator B is at rest. The main pole switch B' by means of which this machine can be connected into circuit is therefore open.

Now, if the load increase to such extent as to require the service of the second alternator B, it must be switched in parallel with A. In order that both machines may operate properly in parallel, three conditions must be satisfied before they are connected together, or else the one alternator will be short circuited through the other, and serious results will undoubtedly follow.

Accordingly before closing main switch B, it is necessary that

1. The frequencies of both machines be the same; 2. The machines must be in synchronism; 3. The voltages must be the same.

=Ques. How are the frequencies made the same?=

Ans. By speeding up the alternator to be cut in, or change the speed of both until frequency of both machines is the same.

=Ques. How are the alternators synchronized or brought in phase?=

Ans. The synchronism of the alternators is determined by employing some form of synchronizer, as by the single lamp method of fig. 2,843, or the two lamp method of fig. 2,845.

=Ques. In synchronizing by the one lamp method, when should the incoming machine be thrown in?=

Ans. It is advisable to close the switch when the machines are approaching synchronism rather than when they are receding from it, that is to say, the instant the lamp becomes dark.

=Ques. What are the objections to the one lamp method?=

Ans. The filament of the lamp may break, and cause darkness, or the lamp may be dark with considerable voltage as it takes over 20 volts to cause a 100 volt lamp to glow.

=Ques. What capacity of single lamp must be used?=

Ans. It must be good for twice the voltage of either machine.

=Ques. What modification of the synchronizing methods shown in the accompanying illustrations is necessary when high pressure alternators are used?=

Ans. Step down transformers must be used between the alternators and the lamps to obtain the proper working voltages for the lamps.

=Ques. How is the voltage of an incoming machine adjusted so that it will be the same as the one already in operation?=

Ans. By varying the field excitation with a rheostat in the alternator field circuit.

=Ques. How may two or more alternators be started simultaneously?=

Ans. After bringing each of them up to its proper speed so as to obtain equal frequencies, the main switches may be closed, thereby joining their armature circuits in parallel. As yet, however, their respective field windings have not been supplied with current, so that no harm can result in doing this. The exciters of these machines after being joined in parallel, should then be made to send direct current simultaneously through the field windings of the alternators, and from this stage on the directions previously given may be followed in detail.

=Ques. What are the conditions when two or more alternators are directly connected together?=

Ans. If rigidly connected together, or directly connected to the same engine, they must necessarily run in the same manner at all times.

When machines connected in this way are once properly adjusted so that they are in phase with each other, their operation in parallel is even a simpler task than when they are all started together but are not directly connected.

=Ques. When an alternator is driven by a gas engine, what provision is sometimes made to insure successful operation in parallel?=

Ans. An amortisseur winding is provided to counteract the tendency to "hunting."

=Ques. What is the action of the amortisseur winding?=

Ans. Any sudden change in the speed of the field, generates a current in the amortisseur winding which resists the change of velocity that caused the current.

The appearance of an amortisseur winding is shown in the cut below (fig. 2,850) illustrating the field of a synchronous condenser equipped with amortisseur winding.

=Ques. How are three phase alternators synchronized?=

Ans. In a manner similar to the single phase method.

Thus the synchronizing lamps may be arranged as in fig. 2,581, which is simply an extension of the single phase method.

=Ques. Are three lamps necessary?=

Ans. Only to insure that the connections are properly made, after which one lamp is all that is required.

=Ques. How is it known that the connections of fig. 2,851 are correct?=

Ans. If, in operation, the three lamps become bright or dark _simultaneously_, the connections are correct; if this action takes place _successively_, the connections are wrong.

If wrong, transpose the leads of one machine until simultaneous action of the lamps is secured.

=Ques. What is the disadvantage of the lamp method of synchronizing?=

Ans. Lack of sensitiveness.

=Ques. Which is the accepted lamp method, dark or brilliant?=

Ans. In the United States it is usual to make the connections for a dark lamp at synchronism, while in England the opposite practice obtains.

With the dark lamp method, the breaking of a filament might cause the machines to be connected with a great phase difference, whereas, with the brilliant lamp it is difficult to determine the point of maximum brilliancy. This latter method, therefore may be called the safer.

=Ques. What may be used in place of lamps for synchronizing?=

Ans. Some form of synchroscopes, or synchronizers.

=Ques. How does the Lincoln synchronizer work?=

Ans. The construction is such that a hand moves around a dial so that the angle between the hand and the vertical is always the phase angle between the two sources of electric pressure to which the synchronizer is connected.

If the incoming alternator be running too slow, the hand deflects in one direction, if too fast, in the other direction. When the hand shows no deflection, that is, when it stands vertical, the machines are in phase. A complete revolution of the hand indicates a gain or loss of one cycle in the frequency of the incoming machine, as referred to the bus bars.

=Cutting Out Alternator.=--When it is desired to cut out of circuit an alternator running in parallel with others, the method of procedure is as follows:

1. Reduce driving power until the load has been transferred to the other alternators, adjusting field rheostat to obtain minimum current; 2. Open main switch; 3. Open field switch.

=Ques. What precaution should be taken?=

Ans. _Never_ open field switch before main switch.

=Ques. What is the ordinary method of cutting out an alternator?=

Ans. The main switch is usually opened without any preliminaries.

=Ques. What is the objection to this procedure?=

Ans. It suddenly throws all the load on the other alternators, and causes "hunting."

=Ques. What forms of drive are especially desirable for running alternators in parallel, and why?=

Ans. Water turbine or steam turbine because of the uniform torque, thus giving uniform motion of rotation.

With reciprocating engines, the crank effect is very variable during the revolution, resulting in pulsations driving the alternator too fast or too slow, and causing cross current between the alternators.

=Ques. Is a sluggish, or a too sensitive governor preferable on an engine driving alternators in parallel?=

Ans. A sluggish governor.

=Alternators in Series.=--Alternators are seldom if ever connected in series, for the reason that the synchronizing tendency peculiar to these machines causes them to oppose each other and fall out of phase when they are joined together in this way. If, however, they be directly connected to each other, or to an engine, so that they necessarily keep in phase at all times, and thus add their respective voltages instead of counteracting them, series operation is possible.

NOTE.--According to the practice of the General Electric Co., 2½ degrees of phase difference from a mean is the limit allowable in ordinary cases. It will, in certain cases, be possible to operate satisfactorily in parallel, or to run synchronous apparatus from machines whose angular variation exceeds this amount, and in other cases it will be easy and desirable to obtain a better speed control. The 2½ degree limit is intended to imply that the maximum departure from the mean position during any revolution shall not exceed 2½ ÷ 360 of an angle corresponding to two poles of a machine. The angle of circumference which corresponds to the 2½ degree of phase variation can be ascertained by dividing 2½ by ½ the number of pole; thus, in a 20 pole machine, the allowable angular variation from the mean would be 2½ ÷ 10 = ¼ of one degree.

=Transformers.=--These, as a whole, are simple in construction, high in efficiency, and comparatively inexpensive. Their principles of operation are also readily understood.

The efficiency of a transformer, that is, the ratio between full load primary and full load secondary is greatest when the load on it is such that the sum of the constant losses equals the sum of the variable losses.

In general, transformers designed for high frequencies and large capacities are more efficient than those designed for low frequencies and small capacities. As a whole, however, a transformer leaves but little to be desired as regards efficiency, a modern 60 cycle transformer of 50 kilowatts capacity or more possesses an efficiency of approximately 98 per cent. at full load and an efficiency of about 97 per cent. at half load.

=Ques. How should a transformer be selected, with respect to efficiency?=

Ans. One should be chosen, whose parts are so proportioned that the point of maximum efficiency occurs at that load which the transformer usually carries in service.

In many alternating current installations, comparatively light loads are carried the greater part of the time, the rated full load or an overload being occurrences of short durations. For such purposes special attention should be given to the designing or selecting of transformers having low core losses rather than low resistance losses, because the latter are then of relatively small importance.

=Ques. What kind of efficiency is the station manager interested in?=

Ans. The "all day efficiency."

This expression, as commonly met with in practice, denotes _the percentage that the amount of energy actually used by the consumer is of the total energy supplied to his transformer during 24 hours_. The formula for calculating the all day efficiency of a transformer is based upon the supposition that the amount of energy used by the consumer during 24 hours is equivalent to full load on his transformer during five hours and is as follows:

5w E = ------------- 24c + 5r + 5w where E = the all day efficiency of the transformer, w = the full load in watts on the primary, c = the core loss in watts, r = the resistance loss in watts.

=Ques. What are the usual all day efficiencies?=

Ans. The average is about 85 per cent. for those of 1 kilowatt capacity, 92 per cent. for those of 5 kilowatts capacity, 94 per cent. for those of 10 kilowatts capacity, and about 94.5 per cent. for those of 15 kilowatts capacity.

=Ques. What becomes of the energy lost by a transformer?=

Ans. It reappears as heat in the windings and core.

This heat not only increases the resistances of the windings and core, producing thereby a further increase of their respective losses, but in addition causes in time a peculiar effect on the iron core which is intensified by the reversals of magnetism constantly going on within it.

After about two years' service, the iron apparently becomes fatigued or tired, and this phenomenon is called aging of the iron. Since the life of the transformer depends to a great extent upon this factor, the conditions responsible for its existence should as far as possible be removed. Means must therefore be provided in the construction to radiate the heat as quickly as it is generated.

=Ques. What kind of oil is used in oil cooled transformers?=

Ans. Mineral oil.

=Ques. How is it obtained?=

Ans. By fractional distillations of petroleum unmixed with any other substances and without subsequent chemical treatment.

=Ques. What is the important requirement for transformer oil?=

Ans. It should be free from moisture, acid, alkali or sulphur compounds.

=Ques. How may the presence of moisture be determined?=

Ans. By thrusting a red hot iron rod in the oil; if it "crackle," moisture is present.

=Ques. Describe the Westinghouse method of drying oil.=

Ans. It is circulated through a tank containing lime, and afterwards, through a dry sand filter.

=Ques. What is the objection to heating the oil (raising its temperature slightly above boiling point of water) to remove the moisture?=

Ans. The time consumed (several days) is excessive.

=Ques. What effect has moisture?=

Ans. It reduces the insulation value of the oil. .06 per cent. of moisture has been found to reduce the dielectric strength of oil about 50 per cent. "dry" oil will withstand a pressure of 25,000 volts between two 9½ inch knobs separated .15 inch.

=Ques. What is understood by transformer regulation?=

Ans. It is the difference between the secondary voltage at no load and at full load, and is generally expressed as a percentage of the secondary voltage at no load.

=Ques. What governs its value?=

Ans. The resistance and reactance of the windings.

=Ques. How may the regulation be improved?=

Ans. By decreasing the resistances of the windings by employing conductors of greater cross section, or decreasing their reactance by dividing the coils into sections and closely interspersing those of the primary between those of the secondary.

NOTE.--_The term_ ="regulation"= as here used is synonymous with "drop." The _voltage drop_ in a transformer denotes the drop of voltage occurring across the secondary terminals of a transformer with load. This drop is due to two causes: 1, the resistance of the windings; and 2, the reactance or magnetic leakage of the windings. On non-inductive load, the reactive drop, being in quadrature, produces but a slight effect, but on inductive loads it causes the voltage to drop, and on _leading current loads_ it causes the voltage to rise. As the voltage drop of a good transformer is very small even on inductive load, direct accurate measurement is difficult. It is best to measure the copper loss with short circuited secondary by means of a wattmeter, and at the same time the voltage required to drive full load current through. From the watts, the resistance drop can be found, and from this and the impedance voltage, the reactive drop may be calculated. From these data a simple vector diagram will give, near enough for all practical purposes, the drop for any power factor, or the following formula may be used which has been deduced from the vector diagram. _________________________ D = √(W + X)^{2} + (R + P)^{2} - 100 where R = % resistance drop; X = % reactive drop; P = % power factor of load; W = % wattless factor of ________ load (√1 - P^{2}); D = % resultant secondary drop. For non-inductive loads where P = 100 and W _____________________ = 0, D = √X^{2} + (100 + R)^{2} - 100. In the case of leading currents it should be considered negative.

In transformers where there is a great difference in voltage between the primary and secondary windings, however, this remedy has its limitations on account of the great amount of insulation which must necessarily be used between the windings, and which therefore causes the distances between them to become such as to cause considerable leakage of the lines of force.

=Ques. How does the regulation vary for different transformers, and what should be the limit?=

Ans. Those of large capacity usually have a better regulation than those of small capacity, but in no case should its value exceed 2 per cent.

=Ques. What advantages have shell type transformers over those of the core type?=

Ans. They have a larger proportion of core surface exposed for radiation of heat, and a shorter magnetic circuit which reduces the tendency for a leakage of the lines of force into the air.

Both types have advantages and disadvantages as compared with the other. In the shell type, there is less magnetic leakage, but also less surface exposed for radiation, and greater difficulty in providing efficient insulation between the two circuits; in the core type there is more surface exposed for radiation and less difficulty in insulating the windings, but there is also a great leakage of the lines of magnetic force into the outer air.

=Ques. How are the windings usually arranged?=

Ans. As a rule, there is only one primary winding but the secondary winding is generally divided into two equal sections, the four terminals of which are permanently wired to four connection blocks which may be connected so as to throw the secondary sections either in parallel or in series with each other at will.

=Ques. What is necessary for satisfactory operation of transformers in parallel?=

Ans. They must be designed for the same pressures and capacities, their percentages of regulation should be the same and they must have the same polarity at a given instant.

One may satisfy himself as to the first of these conditions by examining the name plates fastened to the transformers, whereon are stamped the values of the respective pressures and capacities of each.

Although equal values of regulation is given as one of the conditions to be satisfied, transformers may be operated in parallel when their percentages of regulation are not the same. Ideal operation, however, can be attained only under the former state of affairs. Suppose, for instance, a transformer having a regulation of two per cent. be operated in parallel with another of similar size and design but having a regulation of one per cent. The secondary pressures of these transformers at no load will of course be the same, but at full load if the secondary pressure of the one be 98 volts, that of the other will be 99 volts. There will, therefore, be a difference of pressure of one volt between them which will tend to force a current backward through the secondary winding of the transformer delivering 98 volts. This reversed current, although comparatively small in value, lowers the efficiency of the installation by causing a displacement of phase and a decrease in the combined power factor of the transformers.

=Ques. Describe the polarity test.=

Ans. The test for polarity consists in joining together by means of a fuse wire, a terminal of the secondary winding of each transformer, and then with the primary windings supplied with normal voltage, connecting temporarily the remaining terminals of the secondary windings. The melting of the fuse wire thus connected indicates that the secondary terminals joined together are of opposite polarities, and that the connections must therefore be reversed, whereas if the fuse wire do not melt, it shows that the proper terminals have been joined and that the connections may be made permanent.

The object of this test is, obviously, not to determine the exact polarity of each secondary terminal, but merely to indicate which of them are of the same polarity.

=Motor Generators.=--In motor generator sets, either the shunt or series wound type of motor may be employed at the power producing end of the set, but the field of the generator is either shunt or compound wound, depending upon whether or not it is desired to maintain or to raise the secondary voltage near full load. In either case a rheostat introduced in the shunt field winding of the generator will be found very essential. Both generator and motor are so mounted on the base that their respective commutators are at the outer ends of the set. By this means ample space surrounds all of the working parts, and repairs can readily be made.

Motor generators are frequently used as boosters to raise or boost the voltage near the extremities of long distance, direct current transmission lines. Of these, electric railway systems in which it is desired to extend certain of the longer lines, form a typical example.

Owing to the great cost of changing such a system over to one employing alternating current, or storage batteries, or of constructing an additional power station, these solutions of the problem are usually at variance with good judgment and the amount of money at hand. The choice then remains between the purchase of additional wire for feeders, the connection of a booster in the old feeders, or the installation of both larger feeders and a booster. Of these, it is generally found that either the second or the third mentioned alternative meets the conditions most satisfactorily.

A booster installed in a railway system for the purpose just mentioned, would have a series wound motor, and the conditions to which it must conform would be as follows: The motor having a series winding must provide for the full feeder current passing through both armature and field windings.

Owing to the varying loads on a railway system, due to the frequent starting and stopping of cars, the feeder current varies between zero and some such value as 150 amperes. This fluctuation of current through the field winding will, in ordinary cases, vary the magnetization of the pole pieces from zero almost to the point of saturation; that is, the maximum feeder current will so nearly fill the magnet cores with lines of force that it would be quite difficult to cause more lines of magnetic force to pass through them.

So long as the point of saturation is not reached, however, the proportion of current to field strength remains constant, and therefore the ratio of amperes to volts will not vary.

The severe fluctuations of the feeder current would, if the motor were shunt or compound wound, cause most serious sparking and various other troubles, but in a series motor where the back ampere turns on the armature that react on the field vary in precisely the same proportion as the ampere turns in the field, there exists at all times a tendency to balance the active forces and produce satisfactory operation. If, however, the field magnet cores be very large, they cannot so quickly respond, magnetically, to changes in the strength of the current, and there is then greater liability of the armature reaction momentarily weakening the field and thereby producing temporary sparking.

=Ques. Are motor generators always composed of direct current sets?=

Ans. No.

=Ques. Describe conditions requiring a different combination.=

Ans. For purposes where for instance direct currents of widely different voltages are to be obtained from an alternating current circuit, and it is desired to install but one set, a motor generator consisting of an alternating current motor such as an induction motor, and a dynamo must necessarily be employed.

In such sets, it is common to find both motor and dynamo armatures mounted on a common shaft, and the respective field frames resting on a single base, although for connection on a very high pressure alternating current circuit, separate armature shafts insulated from each other but directly connected together, and separate bases resting on a single foundation, are usually employed to afford the highest degree of insulation between the respective circuits of the two machines.

=Ques. What is the objection to a set composed of alternating current motor and alternator?=

Ans. The commercial field that would be naturally covered by such a set is better supplied by a transformer.

=Ques. Why?=

Ans. Because a transformer contains no moving parts, and is therefore simpler in construction, cheaper in price, and less liable to get out of order.

=Dynamotors.=--A dynamotor differs from a motor generator in that the motor armature and the generator armature are combined into one, thereby requiring but one field frame. Since the motor and generator armature windings are mounted on a single core, the armature reaction due to the one winding is neutralized by the reaction caused by the other winding. There is, consequently, little or no tendency for sparking to occur at the brushes, and they therefore need not be shifted on this account for different loads.

=Ques. How is a dynamotor usually constructed?=

Ans. It is usually built with two pole pieces which are shunt wound.

=Ques. Why does the voltage developed fall off slightly under an increase of load?=

Ans. Because a compound winding cannot be provided.

=Ques. Describe the armature construction and operation.=

Ans. It consists of two separate windings; one of which is joined to a commutator mounted on one side of the armature for motor purposes, and the other to the commutator on the other side of the armature for generator purposes.

By means of two studs of brushes pressing on the motor commutator, current from the service wires is fed into the winding connected to this commutator, and since the shunt field winding is also excited by the current from the service wires, there is developed in the generator winding on the rotating armature a direct voltage which is proportional to the speed of rotation of the armature in revolutions per second, the number of conductors in series which constitute the generator winding, and the total strength of the field in which the armature revolves. This pressure causes current to pass through the generator winding and the distributing circuit when the distributing circuit to which this winding is connected by means of its respective commutator, brushes, etc., is closed.

=Ques. How is a dynamotor started?=

Ans. It is connected at its motor end and started in the same manner as any shunt wound motor on a constant pressure circuit.

=Ques. What precautions should be taken in starting a dynamotor?=

Ans. The necessary precautions are, to have the poles strongly magnetized before passing current through the motor winding on the armature; to increase gradually the current through this winding, and not to close the generating circuit until normal conditions regarding speed, etc., are established in the motor circuit.

=Ques. How is the current developed in the machine regulated?=

Ans. It can be regulated by the introduction of resistance in one or the other of the armature circuits, or by a shifting of the brushes around the commutator.

=Ques. Are dynamotors less efficient than motor generators of a similar type?=

Ans. No, they are more efficient.

=Ques. Why?=

Ans. Because they have only one field circuit and at least one bearing less than a motor generator.

A motor generator has at least three bearings, and occasionally, four, where the set consists of two independent machines directly connected together.

=Rotary Converters.=--An important modification of the dynamotor is the rotary converter. This machine forms, as it were, a link between alternating and direct current systems, being in general a combination of an alternating current motor and a dynamo.

It has practically become a fixture in all large electric railway systems and in other installations where heavy direct currents of constant pressure are required at a considerable distance from the generating plant. In such cases a rotary converter is installed in the sub-station, and being simpler in construction, higher in efficiency, more economical of floor space, and lower in price than a motor generator set consisting of an alternating current motor and a dynamo which might be used in its place, it has almost entirely superseded the latter machine for the class of work mentioned.

=Ques. What is the objection to the single phase rotary converter?=

Ans. It is not self-starting.

=Ques. What feature of operation is inherent in a rotary converter?=

Ans. A rotary converter is a "reversible machine."

That is to say, if it be supplied with direct current of the proper voltage at its commutator end, it will run as a direct current motor and deliver alternating current to the collector rings. While this feature is sometimes taken advantage of in starting the converter from rest, the machine is not often used permanently in this way, its commercial application being usually the conversion of alternating currents into direct currents.

=Ques. How does a rotary converter operate when driven by direct current?=

Ans. The same as a direct current motor, its speed of rotation depending upon the relation existing between the strength of the field and the direct current voltage applied.

If the field be weak with respect to the armature magnetism resulting from the applied voltage, the armature will rotate at a high speed, increasing until the conductors on the armature cut the lines of force in the field so as to develop a voltage which will be equal to that applied.

Again, if the field be strong with respect to the armature magnetism, resulting from the applied voltage, the armature will rotate at a low speed. If, therefore, it be desired to operate the converter in this manner and maintain an alternating current of constant frequency, the speed of rotation must be kept constant by supplying a constant voltage not only to the brushes pressing on the commutator, but also to the terminals of the field winding.

=Ques. How does it operate with alternating current drive?=

Ans. The same as a synchronous motor.

=Ques. What is the most troublesome part and why?=

Ans. The commutator, because of the many pieces of which it is composed and the necessary lines along which it is constructed, its peripheral speed must be kept within reasonable limits.

=Ques. What should be the limit of the commutator speed?=

Ans. The commutator speed, or tangential speed at the brushes should not exceed 3,000 feet per minute.

=Ques. Name another limitation necessary for satisfactory operation.=

Ans. The pressure between adjacent commutator bars should not exceed eight or ten volts.

If the commutator bars be made narrow in order to obtain the necessary number for the desired voltage with the minimum circumference and therefore low commutator speed, the brushes employed to collect the current are liable to require excessive width in order to provide the proper cross section and yet not cover more than two bars at once.

=Ques. How can the commutator speed be kept within reasonable limits, other than by reducing the width of the commutator bars?=

Ans. By using alternating current of comparatively low frequency.

For a rotary converter delivering 500 volt direct current, the proper frequency for the alternating current circuit has been found to be 25 cycles per second.

=Ques. When a rotary converter is operated in this usual manner on an alternating current circuit, how can the direct current be varied?=

Ans. It may be varied (from zero to a maximum) by changing the value of the alternating pressure supplied to the machine, or it may be altered within a limited range by moving the brushes around the commutator, or in a compound wound converter by changing the amount of compounding.

Under ordinary conditions, varying the voltage developed by changing the voltage at the motor end is not practical, hence the voltage developed can be varied only over a limited range. In addition to this, the voltage developed at the direct current end bears always a certain constant proportion to the alternating current voltage applied at the motor end; this is due to the same winding being used both for motor and generator purposes. In all cases the proportion is such that the alternating current voltage is the lower, being in the single phase and in the two phase converters about .707 of the direct current voltage, and in the three phase converter about .612 of the direct current voltage. It is thus seen that whatever value of direct current voltage be desired, the value of the applied alternating current voltage must be lower, requiring in consequence the installation of step down transformers at the sub-station for reducing the line wire voltage to conform to the direct current pressure required.

=Ques. What is the efficiency of a rotary converter?=

Ans. It may be said to have approximately the same efficiency as that in the average of the same output, although in reality the converter is a trifle more efficient on account of affording a somewhat shorter average path for the current in the armature, reducing in consequence the resistance loss and the armature reaction.

=Ques. May a converter be overloaded more than a dynamo of the same output, and why?=

Ans. Yes, because there is usually less resistance loss in the armature of the converter than in the armature of the dynamo.

Thus, a two phase converter may be overloaded approximately 60 per cent., and a three phase converter may be overloaded about 30 per cent. above their respective outputs if operated as dynamos.

=Ques. Describe how a converter is started.=

Ans. There are several methods any one of which may be employed, the choice in any given case depending upon which of them may best be followed under the existing conditions.

If it be found advisable to start the converter with direct current, the same connections would be made between the source of the direct current and the armature terminals on the commutator side of the converter as would be the case were a direct current shunt motor of considerable size to be started; this naturally means that a starting rheostat and a circuit breaker will be introduced in the armature circuit.

The shunt field winding alone is used, and this part of the wiring may be made permanent if, as is usually the case, the same source of direct current is used normally for separate field excitation.

The direct current may be derived from a storage battery, from a separate converter, or from a motor generator set installed in the sub-station for the purpose.

An adjustable rheostat will, of course, be connected in the field circuit for regulation. Before starting the converter, however, it is necessary to do certain wiring between the terminals on the collector side of the machine and the alternating current supply wires, in order that the change over from direct current motive power to alternating current motive power may be made when the proper phase relations are established between the alternating current in the supply wires and the alternating current in the armature winding of the converter.

In order that proper phase relations exist, the armature of the converter must rotate at such a speed that each coil thereon passes its proper reversal point at the same time as the alternating current reverses in the supply wires. This speed may be calculated by doubling the frequency of the supply current and then dividing by the number of pole pieces on the converter, but a far more accurate method of judging when the converter is in step or in synchronism with the supply current consists in employing incandescent lamps as shown in fig. 2,872.

=Ques. How is a polyphase converter started with alternating current?=

Ans. This may be done by applying the alternating pressure directly to the collector rings while the armature is at rest. There need be no field excitation; in fact the field windings on the separate pole pieces should be disconnected from each other before the alternating voltage is applied to the armature, else a high voltage will be induced in the field windings which may prove injurious to their insulation. The passage of the alternating current through the armature winding produces a magnetic field that rotates about the armature core, and induces in the pole pieces eddy currents, which, reacting on the armature, exert a sufficient torque to start the converter from rest and cause it to speed up to synchronism.

=Ques. How much alternating current is required to start a polyphase converter?=

Ans. About 100 per cent. more than that required for full load.

=Ques. How may this starting current be reduced?=

Ans. Transformers may be switched into circuit temporarily to reduce the line wire voltage until the speed become normal.

In conjunction with this method, the method of synchronizing shown in fig. 2,872 may be used, thus, in starting, there is an alternating current between the brushes which pulsates very rapidly, but when synchronism is approached, the pulsations become less rapid until finally with the converter in step with the alternator the pulsations entirely disappear.

The light given by the lamps thus connected indicates accurately the condition of affairs at any one time, varying from a rapidly fluctuating light at the beginning to one of constant brilliancy at synchronism.

=Ques. If the armature of the starting motor have a starting resistance, how must this be connected?=

Ans. It should be connected in series with the armature inductors before the alternating voltage is applied.

As the motor increases in speed, the starting resistance is gradually short circuited until it is entirely cut out of circuit.

NOTE.--Some converters are provided with a small induction motor for starting mounted on an iron bracket cast in the converter frame, and whose shaft is keyed to that of the converter. Allowing for a certain amount of slip in the induction motor, the field of this machine must possess a less number of magnet poles than the converter in order to enable the latter machine to be brought to full synchronism. To start the induction motor, it is simply necessary to apply to its field terminals the proper alternating voltage. The bracket, and therefore the motor, is usually mounted outside the armature bearing on the collector side of the converter.

=Ques. Describe the usual wiring for the installation of a rotary converter in a sub-station.=

Ans. Commencing at the entrance of the high pressure cables, first there is the wiring for the lightning arresters, then for the connection in circuit of the high tension switching devices, from which the conductors are led to bus bars, and thence to the step down transformers.

On a three phase system the transformers should be joined in delta connection, as a considerable advantage is thereby gained over the star connection, in that should one of the transformers become defective, the remaining two will carry the load without change except more or less additional heating. Between the transformers and rotary converter the circuits should be as short and simple as possible, switches, fuses, and other instruments being entirely excluded. The direct current from the converter is led to the direct current switchboard, and from there distributed to the feeder circuits.

=WATTMETER ERROR FOR A LOAD OF 1,000 VOLT-AMPERES= (For a lag of 1 degree in the pressure coil)

+------------+----------+-------+-------------------+ | | | |Error of indication| |Power factor|True watts| Error | in per cent | | | | | of true value | +------------+----------+-------+-------------------+ | 1. | 1,000 | .3 | 0.03 | | .9 | 900 | 7.6 | 0.85 | | .8 | 800 | 10.5 | 1.31 | | .7 | 700 | 12.5 | 1.78 | | .6 | 600 | 13.9 | 2.32 | | .5 | 500 | 15.1 | 3.02 | | .4 | 400 | 15.9 | 3.98 | | .3 | 300 | 16.6 | 5.54 | | .2 | 200 | 17.1 | 8.55 | | .1 | 100 | 17.3 | 17.30 | +------------+----------+-------+-------------------+

NOTE.--In the iron vane type instrument when used as a wattmeter, the current of the series coil always remains in perfect phase with the current of the circuit, provided series transformers are not introduced. The error, then, is entirely due to the lag of the current in the pressure coil, and this error in high power factor is exceedingly small, increasing as the power factor decreases. In the above table it should be noted that the value of the error as distinguished from the per cent. of error, instead of indefinitely increasing as the power factor diminishes, rapidly attains a maximum value which is less than 2 per cent. of the power delivered under the same current and without inductance. It should also be noted that the above tabulation is on the assumption of a lag of 1 degree in the pressure coil. The actual lag in Wagner instruments for instance, is approximately .085 of a degree, and the error due to the lag of the pressure coil in Wagner instruments is, therefore, proportionally reduced from the figures shown in the above tabulation.

=Ques. In large sub-stations containing several rotary converters how are they operated?=

Ans. Frequently they are installed to receive their respective currents from the same set of bus bars; that is, they may be operated as alternating current motors in parallel. They are also frequently operated independently from single bus bars, but very seldom in series with each other.

=Ques. How may the direct current circuit be connected?=

Ans. In parallel.

NOTE.--In motor testing, by the methods illustrated in the accompanying cuts, it is assumed that the motor is loaded in the ordinary way by belting or direct connecting the motor to some form of load, and that the object is to determine whether the motor is over or under loaded, and approximately what per cent. of full load it is carrying. All commercial motors have name plates, giving the rating of the motor and the full load current in amperes. Hence the per cent. of load carried can be determined approximately by measuring the current input and the voltage. If an efficiency test of the apparatus be required, it becomes necessary to use some form of absorption by dynamometer, such as a Prony or other form of brake. The output of the motor can then be determined from the brake readings. The scope of the present treatment is, however, too limited to go into the subject of different methods of measuring the output of the apparatus, and is confined rather to methods of measuring current input, voltage, and watts. The accuracy of all tests is obviously dependent upon the accuracy of the instruments employed. Before accepting the result obtained by any test, especially under light or no load, correction should be made for wattmeter error. See table of wattmeter error on page 2,075.

=Ques. What provision should be made against interruption of service in sub-stations?=

Ans. There should be one reserve rotary converter to every three or four converters actually required.

=Ques. Why does a rotary converter operate with greater efficiency, and require less attention than does a dynamo of the same output?=

Ans. There is less friction, and less armature resistance, the latter because the alternating current at certain portions of each revolution passes directly to the commutator bars without traversing the entire armature winding as it does in a dynamo; there is no distortion of the field and consequently no sparking, or shifting of the brushes, since the armature reaction resulting from the current fed into the machine and that due to the current generated in the armature completely neutralizes each other.

=What electrical difficulty is experienced with a rotary converter?=

Ans. Regulation of the direct current voltage.

=Ques. How is this done?=

Ans. It can be maintained constant only by preserving uniform conditions of inductance in the alternating current circuit, and uniform conditions in the alternator.

While changes in either of these may be compensated to a certain extent by adjustment of the field strength of the converter, they cannot be entirely neutralized in this manner; it is therefore necessary that both the line circuit and the alternator be given attention if the best results are to be obtained from the converter.

=Ques. What mechanical difficulty is experienced with rotary converters?=

Ans. Hunting.

=Ques. What is the cause of this?=

Ans. It is due to a variation in frequency.

The inertia of the converter armature tends to maintain a constant speed; variations in the frequency of the supply circuit will cause a displacement of phase between the current in the armature and that in the line wires, which displacement, however, the synchronizing current strives to decrease. The synchronizing current, although beneficial in remedying the trouble after it occurs, exerts but little effort in preventing it, and many attempts have been made to devise a plan to eliminate this trouble.

NOTE.--Three phase motor test; polyphase wattmeter method. This is identical with the test of fig. 2,882, except that the wattmeter itself combines the movement of the two wattmeters. Otherwise the method of making the measurements is identical. If the power factor be known to be less than 50 per cent., connect one movement so as to give a positive deflection; then disconnect movement one and connect movement two so as to give a positive deflection. Then reverse either the pressure or current leads of the movement, giving the smaller deflection, leaving the remaining movement with the original connections. The readings now obtained will be the correct total watts delivered to the motor. If the power factor be known to be over 50 per cent., the same methods should be employed, except that both movements should be independently connected to give positive readings. An unloaded induction motor has a power factor of less than 50 per cent., and may, therefore, be used as above for determining the correct connections. For a better understanding of the reasons for the above method of procedure, the explanation of the two wattmeter method, fig. 2,882, should be read. The power factor may be calculated as explained under fig. 2,882. Connect as shown in fig 2,882. The following check on connection may be made. Let the polyphase induction motor run idle, that is, with no load. The motor will then operate with a power factor less than 50 per cent. The polyphase meter should give a positive indication, but if each movement be tried separately one will be found to give a negative reading, the other movement will give a positive reading. This can be done by disconnecting one of the pressure leads from the binding post of one movement. When the power factor is above 50 per cent. then both movements will give positive deflection.

=Ques. What are the methods employed to prevent hunting?=

Ans. 1, the employment of a strongly magnetized field relative to that developed by the armature; 2, a heavy flywheel effect in the converter; 3, the increasing of the inductance of the armature by sinking the windings thereon in deep slots in the core, the slots being provided with extended heads; and 4, the employment of damping devices or amortisseur winding on the pole pieces of the converter.

=Ques. What method is the best?=

Ans. The damping method.

The devices employed for the purpose are usually copper shields placed between or around the pole pieces, although in some converters the copper is embedded in the poles, and in others it is made simply to surround a portion of the pole tips.

In any case its action is as follows: The armature rotating at a variable speed has a field developed therein which is assumed to be also rotating at a variable speed; the magnetism of this rotary field induces currents in the copper which, however, react on the armature and oppose any tendency toward a further shifting of the magnetism in the armature and therefore prevent the development of additional currents in the copper. Since copper is of low resistance, the induced currents are sufficient in strength to thus dampen any tendency toward phase displacement, and so exert a steadying influence upon the installation as a whole.

=Electrical Measuring Instruments.=--In the manufacture of most measuring instruments, the graduations of the scale are made at the factory, by comparing the deflections of the pointer with voltages as measured on standard apparatus. The voltmeters in most common use have capacities of 5, 15, 75, 150, 300, 500 and 750 volts each, although in the measurement of very low resistances such as those of armatures, heavy cables, or bus bars, voltmeters having capacities as low as .02 volt are employed.

The difference between the design of direct current voltmeters of different capacities lies simply in the high resistance joined in series with the fine wire coil. This resistance is usually about 100 ohms per volt capacity of the meter, and is composed of fine silk covered copper wire wound non-inductively on a wooden spool.

In the operation of an instrument, if the pointer when deflected do not readily come to a position of rest owing to friction in the moving parts, it may be aided in this respect by gently tapping the case of the instrument with the hand; this will often enable the obstruction, if not of a serious nature, to be overcome and an accurate reading to be obtained.

=Ques. Describe a two scale voltmeter.=

Ans. In this type of instrument, one scale is for low voltage readings and the other for high voltage readings; on these scales the values of the graduations for low voltages are usually marked with red figures, while those for high voltages are marked with black figures. A voltmeter carrying two scales must also contain two resistances in place of one; a terminal from each of these coils must be connected with a separate binding post, but the remaining terminal of each resistance is joined to a wire which connects through the fine wire coil with the third binding post of the meter. The two first mentioned binding posts are usually mounted at the left hand side of the meter and the last mentioned binding post and key at the right hand side.

The resistance corresponding to the high reading scale is composed of copper wire having the same diameter as that constituting the resistance for the low reading scale, but as the capacity of the former scale is generally a whole number of times greater than that of the latter scale, the resistances for the two must bear the same proportion.

=Ques. How is a two scale voltmeter connected?=

Ans. In the connection of a two scale voltmeter in circuit, the single binding post is always employed regardless of which scale is desired. If, then, the voltage be such that it may be measured on the low reading scale, the other binding post employed is that connected to the lower of the two resistances contained within; if, however, the pressure be higher than those recorded on the low reading scale, the binding post connected to the higher of the two resistances contained within is used.

NOTE.--Three phase alternator load test. By means of the connection shown in fig. 2,888, readings of armature current and field amperes can be obtained with any desired load. The field current can be varied also so as to maintain constant armature voltage irrespective of load; or the field current may be kept constant and the armature voltage allowed to vary as the load increases. The connections may also be used to make a temperature test on the alternator by loading it with an artificial load. In some cases after the alternator is installed the connection may be used to make a temperature test, using the actual commercial load the alternator is furnishing.

Inasmuch as the capacities of the scales are usually marked on or near the corresponding binding posts, there will generally be no difficulty in selecting the proper one of the two left hand binding posts.

=Ques. How is a two scale voltmeter connected when the binding posts are not marked?=

Ans. If only an approximate idea is possessed of the voltage to be measured, it is always advisable to connect to the binding post corresponding to the high reading scale of the meter in order to determine if the measurement may not be made safely and more accurately on the low reading scale. In any case, some knowledge must be had of the voltage at hand, else the high reading portion of the instrument may be endangered.

_Too much care cannot be taken to observe these precautions_ whenever the voltmeter is used, for the burning out or charring of the insulation either in the fine wire coil or in the high resistance of the meter by an excessive current, is one of the most serious accidents that can befall the instrument.

If a voltmeter has been subjected to a voltage higher than that for which it was designed, yet not sufficiently high to injure the insulation, but high enough to cause the pointer to pass rapidly over the entire scale, damage has been done in another way. The pointer being forced against the side of the case in this manner, bends it more or less and so introduces an error in the readings that are afterward taken.

The same damage will be done if the meter be connected in circuit so the current does not pass through it in the proper direction, although in this case the pointer is not liable to be bent so much as when it is forced to the opposite side of the meter by an abnormal current, since then it has gained considerable momentum which causes a severer impact. The extent of the damage may be ascertained by noting how far away from the zero mark the pointer lies when no current is passing through the instrument. If this distance be more than two-tenths of a division, the metal case enclosing the working part should be removed and the pointer straightened by the careful use of a pair of pinchers.

=Ques. What should be noted with respect to location of instruments?=

Ans. If they be placed near conductors carrying large currents, the magnetic field developed thereby will produce a change in the magnetism of the instruments and so introduce an error in the readings.

=Ques. How should portable instruments be wired?=

Ans. The wires must be firmly secured to the supports on which they rest, so as to reduce the possibility of their being pulled by accident, and so causing the instruments to fall.

A fall or a rough handling of the meter at once shows its effect on the readings, for as much harm is done as would result from a similar treatment of a watch.

The hardened steel pivots used in all high grade voltmeters are ground and polished with extreme care so as to secure and maintain a high degree of sensitiveness. The jewels on which the moving parts revolve are of sapphire, and they too must necessarily be made with skill and carefulness; if, therefore, the jewels become cracked and the pivots dulled by careless handling, the meter at once becomes useless as a measuring instrument.

=Ques. How should readings be taken?=

Ans. The deflection of the pointer should be read to tenths of a division; this can be done with considerable accuracy, especially after a little practice.

For very accurate results, a temperature correction should be applied to compensate the effect which the temperature of the atmosphere has upon the resistance of the meter when measurements are being taken. In ordinary station practice the temperature correction is negligible, being for resistance corresponding to the high scale in first class meters, less than one-quarter of 1 per cent. for a range of 35 degrees above or 35 degrees below 70 degrees Fahrenheit.

=Ques. What attachment is sometimes provided on station voltmeters used for constant pressure service?=

Ans. A normal index.

=Ques. What precaution must be taken in connecting station voltmeters?=

Ans. Care must be taken to guard against any short circuiting of the voltmeter, which, would mean a short circuiting of the generator, and as a result the probable burning out of its armature.

The high resistance of the voltmeter prevents any such occurrence when it is connected in the proper way, but should one side of the circuit be grounded to the metal case or frame of the meter, a careless handling of the lead connected with the other side of the circuit would produce the result just mentioned.

=Ques. Why do station voltmeters indicate a voltage slightly lower than actually exists across the leads?=

Ans. Since they are usually connected permanently in circuit; a certain amount of heat is developed in the wiring of the instrument.

The effect of this heat increases the voltmeter resistance and consequently reduces the current below that which otherwise would pass through the meter; since the deflections of the pointer are governed by the strength of the current, station voltmeters invariably indicate a voltage slightly lower than that which actually exists across their leads.

NOTE.--=Checking up of a recording wattmeter.= This may conveniently be done by noting the deflections at short intervals on an ammeter connected in circuit, and also the readings on the dial of the recording wattmeter during this period. If this test be continued for an appreciable time, the product of the pressure in volts, the current in amperes, and the time in hours, should equal the number of watthours recorded on the counters of the dial.

NOTE.--=Transformer testing.= In the early days of transformer building, before the commercial wattmeter had been perfected, leakage or exciting current was the criterion of good design. After the introduction of the wattmeter, core loss became the all important factor, and for a long time the question of leakage current was lost sight of. With the introduction of silicon steel, leakage or exciting current again assumed prominence. Keeping in mind the fact that all characteristics of a transformer are of more or less importance, it is essential that the user of such apparatus have at hand the necessary facilities for making tests of all such variable quantities. The tests which all users of transformers should make, are given in this chapter.

=Ques. Can direct current be measured by an alternating current voltmeter?=

Ans. Yes.

NOTE.--=Transformer copper loss test.= The usual and best method of obtaining copper losses is to separately measure the primary and secondary resistance and calculate from these the primary and secondary copper losses. For general diagram of connections and discussion of the drop method, see fig. 2,875. The current should be kept well within the load current of the transformer to avoid temperature rise during the test. In other words, the resistance of the coil is the voltage across its terminals divided by the current. The resistance of the primary coil can be measured similarly. The copper loss in watts in each coil will then be the product of the resistance and the square of the rated current for that coil. The total copper loss will be the sum.

=Ques. What would be the effect of placing a direct voltmeter across an alternating current circuit, and why?=

Ans. There would be no deflection of the pointer owing to the rapid reversals of the alternating current.

=Ques. What are the usual capacities of alternating current voltmeters?=

Ans. They are 3, 7.5, 10, 12, 15, 20, 60, 75, 120, 150, 300 and 600 volts, but these capacities may each be increased by the use of a multiplier.

=Ques. How are station voltmeters usually attached to the switchboard?=

Ans. They are usually bolted to the switchboard by means of four iron supports mounted on the back of the instrument; two of these are fastened near each side of the case.

Under certain conditions, however, as in paralleling of alternators, it is convenient to have the alternating current voltmeter mounted on a swinging bracket at the side of the switchboard. The voltmeter may then be swung around in any desired direction so as to enable the attendant to keep informed of the voltage while switching in each additional alternator.

=Ques. How should an ammeter be operated to get accurate readings, and why?=

Ans. It should be cut out of circuit except while taking a reading, because of the error introduced by the heating effect of the current.

In an ammeter having a capacity of 50 amperes, the error thus introduced will be less than 1 per cent. if connected continuously in circuit with a current not exceeding three-quarters this capacity.

An ammeter of 100 amperes capacity may be used indefinitely in circuit with less than 1 per cent. error up to one-half its capacity, and for five minutes at three-quarters capacity without exceeding the 1 per cent. limit.

The 150 scale ammeter may be left in circuit for an indefinite length of time at one-third its full capacity, and for three minutes at one-half its full capacity, with a negligible error.

Ammeters of 200 and of 300 ampere capacities must not continuously carry more than one-quarter of these loads respectively if the readings are to have an accuracy within 1 per cent. nor more than one-half these respective number of amperes for three minutes if the same degree of accuracy be desired.

In order to cut or shunt the ammeter out of circuit when not in use, it is customary when wiring the instrument in place, to introduce a switch as a shunt across it; this switch is kept closed except when a measurement is being taken.

When currents larger than 300 amperes have to be measured, ammeter shunts are generally employed, although ammeters up to 500 amperes capacity are manufactured.

=Ques. What is used in place of instrument shunts for high pressure alternating current measurements?=

Ans. Instrument transformers.

=Ques. What important attention should be periodically given to measuring instruments?=

Ans. They should be frequently tested by comparison with standards that are known to be correct.

Electrical measuring instruments, owing to the nature of their construction and the conditions under which they must necessarily be used, are subject to variations in accuracy. This feature is an annoying one on account of the difficulty of detecting it; a meter may, as far as appearances go, be in excellent working order and yet give readings which are not to be relied upon.

Ridiculous as it may appear, the average station attendant may frequently be seen straining his eyes to read to tenths of a division on the scale of a meter which, if subjected to test, would show an inaccuracy of over 2 per cent.

In testing a meter, by comparing it with a standard, in order to obtain the best results there should be one man at each meter so that simultaneous readings may be taken on both instruments, and the man at the standard meter should maintain the voltage constant while a reading is being taken, by means of a rheostat in the field circuit of the generator supplying the current.

Each meter should be checked or calibrated at five or six approximately equidistant points over its scale; the adjustable resistance being varied each time to give a deflection on the standard meter of an even number of divisions and the deflection on the other meter recorded at whatever it may be. Having obtained the necessary readings, the calculation of the constant or multiplying factor of the meter undergoing test is next in order.

This may best be shown by taking an actual case in which a 150 scale voltmeter is being tested to determine its accuracy. The data and calculations are as follows:

Readings on Readings on Constant standard meter meter tested 150 149.2 150 ÷ 149.2 = 1.005 125 125.0 125 ÷ 125.0 = 1.000 100 98.9 100 ÷ 98.9 = 1.011 75 73.6 75 ÷ 73.6 = 1.019 50 50.0 50 ÷ 50.0 = 1.000 25 24.8 25 ÷ 24.8 = 1.008 ------ 6.043

Average constant for six readings, 6.043 ÷ 6 = 1.007.

It may be stated in general that before taking the readings for this test, the zero position of the pointer on the meter tested should be noted, and if it be more than two-tenths of a division off the zero mark, the case of the meter should be removed and the pointer straightened.

Furthermore, it will be noticed from the readings here recorded that the test is started at the high reading end of the scale; this is done in order that the pointer may gradually be brought up to this spot, by slowly cutting out of circuit the adjustable resistance, and thus show whether or not the pointer has a tendency to stick at any part of the scale. If the meter seem to be defective in this respect, it should be remedied either by bending the pointer or scale, or by renewing one or both of the jewels, before the comparison with the standard is commenced.

It is obvious from the readings recorded for the 150 scale voltmeter, that as compared with the corresponding deflections of the standard, the former are a trifle low.

In order to determine for each observation how much too low they are, it is necessary to divide each reading on the standard by the corresponding reading on the meter tested. The result is the amount by which a deflection of this size on the meter tested must be multiplied in order to obtain the exact reading. This multiplier is called a constant, and as shown, a constant is determined for each of the six observations.

The average constant for the six readings is then found, and this is taken as the constant for the meter as a whole; that is, whenever this 150-scale voltmeter is used, each reading taken thereon must be multiplied by 1.007 in order to correct for its inaccuracy.

The most convenient and systematic way of registering the constant of a meter is to write it, together with the number of the meter and the date of its calibration, in ink on a cardboard tag and loop the same by means of a string to the handle or some other convenient part of the meter.

NOTE.--=Transformer polarity test.= A test of importance in the manufacture of transformers, and sometimes necessary for the user, is the so called _banking_ or _polarity_ test. The transformers from any particular manufacturer have the leads brought out in such a manner that a transformer of any size can be connected to primary and secondary lines in a given order without danger of blowing the fuses due to incorrect connections. All manufacturers of transformers, however, do not bank transformers in the same way, so that it is necessary in placing transformers of different makes to test for polarity. This is done as shown in fig. 2,906. One transformer is selected as a standard and the leads of the second transformer connected as indicated in the diagram. If the transformers be 1,100-2,200 volts to 110-220, two 110 volt lamps are connected in the secondaries of the transformers as indicated, while the primary of the transformer is connected across the line. In transformers built for two primary and two secondary voltages, it is necessary to test each primary and each secondary. The diagram shows the method of connecting one 2,200 volt coil and one 110 volt coil to the transformer to be tested. When the primary circuit of the transformer under test is closed, and if the secondary leads of the 110 volt coil under test be brought out of the case properly, the two 110 volt lamps should be brightly illuminated. If, on the other hand, the two 110 volt terminals have been reversed, no current will flow through the lamps. If these two terminals be found to be brought out correctly, transfer the secondary leads of the transformer under test to the second 110 volt coil. Upon closing the primary circuit, the lamp should again be brightly illuminated. Repeat this process with each of the secondary coils and the other primary coil, and if the lamps show up brightly in every case on closing the primary circuit, all leads have been properly brought out. If on any tests the lamps do not light up brightly, the leads on the transformer must be so changed as to produce the proper banking.

=Ques. What are the usual remedies applied to a voltmeter to correct a 3 or 4 per cent. error?=

Ans. They consist of straightening the pointer, varying the tension of the spiral springs, renewing the jewels in the bearings, altering the value of the high resistance, and, in the case of a direct current instrument, strengthening the permanent magnet.

=Ques. How is the permanent magnet strengthened?=

Ans. After detaching it from the instrument, wrap around several turns of insulated wire, and pass through this wire for a short time 3 or 4 amperes of direct current in such a direction as to reinforce the magnet magnetism.

=Ques. How may the value of the high resistance of a voltmeter be altered?=

Ans. Determine the resistance of the voltmeter and add or subtract, according as the reading is high or low, a certain length of wire whose resistance is in per cent. of the voltmeter resistance the same as the per cent. of error.

NOTE.--The complete calibration of a two scale voltmeter does not, as might be supposed, necessitate that the readings on both scales be checked with standards, for since the resistance corresponding to the one scale is always some multiple of the resistance of the other, the constants of the two scales are proportional. For instance, if S = the reading at the end of the high scale of the voltmeter; S^{1} = the reading at the end of the low scale of the voltmeter; R = the resistance in the meter corresponding to the high scale; R^{1} = the resistance in the meter corresponding to the low scale; K = the constant for the high scale, and K^{1} = the constant for the low scale. Then

SK ÷ R = S^{1}K^{1} ÷ R^{1}

from which

K^{1} = SKR ÷ S^{1}R

That is to say, if the respective resistances corresponding to the two scales be known, and the constant of the high scale be determined by comparison with a standard, then by aid of these known values and the maximum readings on the two scales, the constant of the low scale may be calculated. It is also possible to calculate the constant of the high scale if the constant of the low scale be known, together with the values of the resistances corresponding to the two scales; for from the equation previously given.

K = RS^{1}K^{1} ÷ R^{1}S

=Ques. What is a frequent cause of error in an alternating current meter, and why?=

Ans. The deterioration of its insulation, which permits the working parts of the instrument coming in contact with the surrounding metal case.

A convenient method of testing for deterioration of insulation is shown in fig. 2,905.

=How to Test Generators.=--In the operation of electrical stations, many problems dealing with the generators installed therein can be readily solved by the aid of characteristic curves, which bear a relation to the generators similarly as do indicator diagrams to steam engines.

In steam engineering, a man who did not fully understand the method of taking an indicator diagram would be considered not in touch with his profession, and in electrical engineering the same would be true of one ignorant of the method of obtaining characteristic curves.

The necessary arrangement or connection of the generator from which it is desired to obtain a characteristic curve, consists in providing a constant motive power so that the machine may be run at a uniform speed, and when the field magnets of the generator are separately excited the field current from the outside source must also be maintained constant, preferably by a rheostat connected in the field of the auxiliary exciting machine. It is also necessary in every case that means be provided for varying the main current of the generator step by step from zero to maximum. This may best be done by employing a water rheostat, as shown in fig. 2,909.

=Ques. What instruments are needed in making a test of dynamo characteristics?=

Ans. A voltmeter, ammeter, speed indicator, the usual switches and rheostats.

=Ques. How is the apparatus connected?=

Ans. It is connected as shown in fig. 2,910.

=Ques. Describe the test.=

Ans. Having completed the preliminaries as in fig. 2,910, the test should be started with the main circuit of the generator open. Then, in the case of the shunt machine, the speed should be made normal and the field rheostat adjusted until the voltmeter reading indicates the rated voltage of the machine at no load and readings taken. The electrodes of the water rheostat should be adjusted for maximum resistance and main circuit closed, and a second set of readings taken. Several sets of readings are taken, with successive reductions of water rheostat resistance. The results are then plotted on coordinate paper giving the characteristic curve shown in fig. 2,908.

=Ques. What does the characteristic curve (fig. 2,911) show?=

Ans. An examination of the curve shows that the highest point of the curve occurs at no load or 0 amperes; that as the current is increased, the voltage drops, first slightly to the point B and then rapidly until the point E is reached, when any further lowering of resistance in the main circuit to increase the current causes not only a rapid decline in the voltage but also of the current until both voltage and current become approximately zero.

In some generators, a very slight current results even when the terminals of the machine are actually short circuited; that is, due to residual magnetism in the pole pieces, the lower portion of the curve often terminates, not exactly at zero, but at a point some distance along the current line.

The working portion of the curve is from A to C, at which time the machine is supplying a fairly constant voltage. From C to E shows a critical condition of affairs, while the straight portion D O represents the unstable part of the curve caused by the field current being below its proper value.

The position of the point C determines the maximum power the machine is capable of developing, being in this case (47.5 × 25) ÷ 746 = 1.59 horse power.

=Ques. How may the commercial efficiency of a generator be determined?=

Ans. To obtain the commercial efficiency, the _input_ and _output_ must be found for different loads.

The input may be found by running the generator as a motor at its rated speed, loading it by means of a Prony brake. The generator must be stripped of all belting or other mechanical connections, supplied with its normal voltage and full load current, and the pressure of the Prony brake upon its armature shaft or pulley adjusted until the rated speed of the armature is obtained. The data thus obtained is substituted in the formula.

2π L W R input in brake horse power = ---------- (1) 33,000

in which

L = length of Prony brake lever; W = pounds pull at end of lever; R = revolutions per minute.

The output or electrical horse power for the same load is easily calculated from the formula

amperes × volts output in electrical horse power = --------------- (2) 746

After obtaining value for (1) and (2) the commercial efficiency for the load taken is obtained from the formula

output commercial efficiency = ------ (3) input

Having obtained the commercial efficiency, the difference between the ideal 100 per cent. and the efficiency found will be due to certain losses in the generator. These losses may be classified as

1. Mechanical. 2. Electrical.

The mechanical losses are the friction of the bearings and brushes, and air friction. The electrical losses consist of the eddy current loss, hysteresis loss, armature resistance loss, and field resistance loss.

In testing for these losses, the generator to be tested should be belted to a calibrated motor which latter machine should preferably be of the constant pressure, shunt wound type.

The friction of the bearings and belt of the generator are determined together by raising the brushes off its commutator and running it at the rated speed by means of the calibrated motor.

The amount of power as ascertained from the calibration curve of the motor for the voltage and current used therein when driving the generator as just explained, is a measure of these two losses. The power thus used is practically constant at all loads and is about 2 per cent. of that necessary to drive the generator at full load.

The friction of the brushes can very conveniently be determined next by lowering them on the commutator and giving them the proper tension.

The increase in power resulting from the greater current that will now be taken by the motor to run the dynamo at its rated speed, will be a measure of this loss. In general, its value will be about .5 per cent. of the total power required to drive the dynamo at full load, and this also will remain constant at all loads.

The friction of the air upon the moving armature of the dynamo cannot be determined experimentally, but theoretically this loss is small and may be estimated as .5 per cent.; it is also constant at all loads.

The core loss may be determined experimentally by exciting the field magnets of the dynamo with the normal full load field current through the magnet coils, and noting the increase of power required by the motor to maintain the rated speed of the dynamo thus excited under no load, over that necessary under the same conditions with no field excitation. This increase of power will be the value of the core loss. The core loss is approximately 3 per cent. of the power required to operate the dynamo at full load, and it is constant at varying loads. If it be desired to divide the core loss into its component parts, it is necessary also to run the dynamo under the same conditions as before with field excitation but at half its rated speed. If, then,

H = the power lost in hysteresis at rated speed, E = the power lost in eddy currents at rated speed, T = the power lost in hysteresis and eddy currents at rated speed, S = the power lost in hysteresis and eddy currents at half speed.

there may be formed the two following equations:

H E T = H + E, and S = --- + ---, 2 2

from which the elimination of H will give E = 2T - 4S.

The value of the eddy current loss thus found will be about 1½ per cent., and constant at all loads.

Having previously ascertained the power lost in both eddy currents and hysteresis, and knowing now the power lost in eddy currents alone, it is easy to find that lost in hysteresis by simply subtracting the latter known value from the former. The value of the hysteresis loss is therefore approximately 1½ per cent., and it is constant at different loads.

There yet remains to be determined the armature resistance loss and the field resistance loss. As for the calibrated motor, this may be disconnected from the dynamo, as it need not be used further in the test.

The armature resistance is the resistance of the armature winding of the dynamo, between the commutator bars upon which press the positive and negative brushes. Assume that the value of the armature resistance be known, call this value R ohms, together with that of the full load armature current, which is also known and which call I amperes, this is sufficient data for calculating the armature resistance loss at full load. It is evident that to force the full load current I through the armature resistance R will require a pressure of R volts, and that the watts lost in doing so will be the voltage multiplied by the current. The armature resistance is consequently

IR × I = I^{2}R watts

or, expressed in horse power it is

I^{2}R ÷ 746

At full load it is usually about 2 per cent. of the total power required to drive the generator fully loaded. The armature resistance loss varies in proportion to the load, in fact, as the last expression shows, it increases as the square of the armature current.

The field resistance loss is calculated in the same manner as just explained for the armature resistance loss, it being equal in horse power to the square of the full load field current multiplied by the resistance of the field winding and divided by 746. In a shunt dynamo it is practically constant at 2 per cent. of the total power at full load, but in a series or in a compound generator it will vary in proportion to the load.

HAWKINS PRACTICAL LIBRARY OF ELECTRICITY

IN HANDY POCKET FORM PRICE $1 EACH

_They are not only the best, but the cheapest work published on Electricity. Each number being complete in itself. Separate numbers sent postpaid to any address on receipt of price. They are guaranteed in every way or your money will be returned. Complete catalog of series will be mailed free on request._

=ELECTRICAL GUIDE, NO. 1= Containing the principles of Elementary Electricity, Magnetism, Induction, Experiments, Dynamos, Electric Machinery.

=ELECTRICAL GUIDE, NO. 2= The construction of Dynamos, Motors, Armatures, Armature Windings, Installing of Dynamos.

=ELECTRICAL GUIDE, NO. 3= Electrical Instruments, Testing, Practical Management of Dynamos and Motors.

=ELECTRICAL GUIDE, NO. 4= Distribution Systems, Wiring, Wiring Diagrams, Sign Flashers, Storage Batteries.

=ELECTRICAL GUIDE, NO. 5= Principles of Alternating Currents and Alternators.

=ELECTRICAL GUIDE, NO. 6= Alternating Current Motors, Transformers, Converters, Rectifiers.

=ELECTRICAL GUIDE, NO. 7= Alternating Current Systems, Circuit Breakers, Measuring Instruments.

=ELECTRICAL GUIDE, NO. 8= Alternating Current Switch Boards, Wiring, Power Stations, Installation and Operation.

=ELECTRICAL GUIDE, NO. 9= Telephone, Telegraph, Wireless, Bells, Lighting, Railways.

=ELECTRICAL GUIDE, NO. 10= Modern Practical Applications of Electricity and Ready Reference Index of the 10 Numbers.

=Theo. Audel & Co., Publishers. 72 FIFTH AVENUE=, =NEW YORK.=