LIBRARY

OF THE

UNIVERSITY OF CALIFORNIA.

GIFT OK

Q& Wtrcis.

Received ~fctc<J l8 9

Accession No. 7 / 7 6~ Class No.

MODERN
SWITCHBOARDS

AND THE APPLIANCES USED THEREON; TOGETHER
WITH AN HISTORICAL RESUME' OF EARLY PRAC-
TICES AND EXPEDIENTS, INDICATING THE ADVANCE
RECENTLY MADE IN THIS CLASS OF ELECTRICAL
APPARATUS; AND DATA ON APPROVED METHODS
OF CONSTRUCTION

ALBERT B. HERRICK

ILLUSTRATED BY NUMEROUS CUTS, DRAWINGS AND
DESIGNS, ESPECIALLY PREPARED FOR THIS PUBLICATION

THE CUTTER ELECTRICAL AND MANUFACTURING
COMPANY, PHILADELPHIA, U.S.A., MDCCCXCVIII.

t^ :

THE CUTTER ELECTRICAL AND
MANUFACTURING COMPANY

PRICE, THREE DOLLARS

PRESS OF

EDWARD STERN & CO., INC.

PHILADELPHIA.

CONTENTS

INTRODUCTORY.

The advance of the art of switchboard construction. Old methods of operating. New

methods, 7

CHAPTER I.

CIRCUIT BREAKING DEVICES. The early methods of severing the circuit. Various
switch forms. Slow break snap switches and switches with auxiliary contacts. Physical
properties of contacts. Types of switch contacts ; American, English and German
types. Mechanical and electrical properties of contact surfaces, 12

CHAPTER II.

SWITCHBOARD CONSTRUCTION. Switchboard attendance. Gallery construction and
detail. Relation of switchboards to distributing systems. Fireproof construction.
Exposure. Framing and insulation of switchboards. Method of connecting conductors
and bus bars. Conductor losses. Material for electrical engineering. Impurities of cop-
per. Dimensions of bus bars. Weights and current capacities. Alloys and their con-
ductivity. Brass and other copper alloys. Switchboard material and their necessary
properties. Insulators wood, slate, marble and onyx. Method of drilling. Special
method of switchboard construction. Insulating in high tension switchboards, .... 21

CHAPTER III.

SWITCHBOARD APPLIANCES. Potential measurements. Requirements of switchboard
instruments. Method of testing for errors. Checking methods. Instrument movements.
Solenoid. Permanent magnet type. Instruments actuated by rise in temperature.
Astatic voltmeters. Recording instruments. Voltmeters. Comparative pressure
indicators. Indicating wattmeters. Integrating wattmeters. Dynamo galvanometers.
Dynamo regulators. Automatic regulators, 41

CHAPTER IV.

PROTECTIVE DEVICES. Fuses. Physical conditions under which fuses operate. Con-
dition where fuses are useful. Cutter circuit breakers operation and reliability. Types
of I-T-E circuit breakers. Lightning arresters. Characteristics of lightning discharges.
Proper method of lightning protection. Magnetic type of lightning arrester. Types of
lightning arresters. Non-arcing lightning arresters 60

INTRODUCTORY

^ j= - ji| HE evolution of the switchboard has necessarily fol-
^" lowed the progress of the various systems of distribu-

Jtion, as a necessary adjunct for the controlling and
connecting of the different circuits, and it is to-day
necessary for the collection, distribution and control of
output in any electric light, power or railway system.

In order to fully comprehend the present switchboard practice, an
historical resume may be necessary. The earliest attempt at a
collection of apparatus for the purpose of controlling a distributing
system was at Menlo Park, in 1879, when Edison made the first
commercial application of low-tension currents to a multiple arc dis-
tributing system. The effect of assembly here was especially the
following the diagram of connections with bare copper wires con-
necting the different plug devices.

The points of serviceability and utility came later on in the art,
and the necessity of measuring devices for the current flow and
potential, as well as the protecting devices against abnormal flow
of current, soon asserted itself.

The first central station erected had the apparatus for circuit
and dynamos strewn around the four walls of the station, regard-
less of utility.

In 1883, Mr. Luther Stierenger, at the Louisville Exposition,
designed the distribution circuits so that they and their dependent
apparatus were concentrated at one point. All switches, fuses, and
bus bars were brought together, and the points of utility and con-

INTRODUCTORY

venience were realized ; this afterward developed into switchboard
construction.

The placement of the switchboard in the older practice was left
as one of the last points to be considered in laying out a station,
but now the proper position has been found to have such direct
bearing on the facility of handling the apparatus which it controls
that the proper placement has become a primary matter, and the
engineer has need of all his ability and judgment in considering
the conditions which determine the best possible position. Origi-
nally wires secured to the wainscoting and plugs and crude instru-
ments were connected in by splices. A fire hazard developed from
this combination; the switchboard was then placed at a distance
from the wall, and on a skeleton frame, in the construction of which
as little wood as possible was used. This sufficiently reduced the
hazard until fire-proof methods of construction were developed.
At first the bus bars and their connecting cables were kept on the
front of the board until back connections were necessitated on the
score of safety, space and appearance.

Wood was universally used for supporting instrument cases,
regulators and equalizers, as well as for insulation, up to 1889,
when, on account of the high fire hazard placed on such construc-
tion, slate and porcelain were substituted for wood, as the insulating
and supporting structure.

The operation of the early switchboards was comparatively
simple, and the attendance and space necessary were not considered
items to be taken into account ; but on enlarging the field of current
distribution systems, careful designing had to be done, in order to
make the switchboard as compact, simple and easily operated as
possible ; also, as the current required by the expanding system

INTRODUCTORY

increased, new methods had to be introduced in handling the dyna-
mos and feeders of increased capacity.

Fig. B shows the dynamo board of the old Adams Street Sta-
tion, Chicago, 111., which was an advanced type at the time of its
construction, with exposed bus bars and all apparatus assembled on
a wooden facing, using wood for insulation.

Fig. C shows the feeder board for the same station, including
wooden frame equalizers, which are now relegated to the scrap
heap in nearly every station, as external systems of distribution
have become more complex and require economical methods to
obtain uniform potential on mains, to meet the varying demands on
the system.

THE FIRST CIRCUIT BREAKER

MADE BY
THE CUTTER COMPANY

o
o

d
a

< E

O u

Q H
W <

W H

6 04

*%
CO

CIRCUIT BREAKING DEVICES

N the evolution of circuit breaking devices, it can
be clearly shown that it is by the slow process of
the survival of the best forms, that the present
stage of the art has been reached.

The initial letter illustrates the primitive form of circuit rup-
turing device, when the conductor itself is severed.

The switching devices that were adopted in the early days were
mere enlargements of those used in the feeble carrying current art.

The well-known plug came directly from the telegraph plug,
and the first switching devices were derived from the telegraph
switch, only enlarged to accommodate the greater volume of cur-
rent flow.

The necessity for a mechanical device, which could open the
circuit without requiring the conductors to be
moved, was met by having two circuit termi-
nals end in two insulated plates, the adjacent
ends of which were connected together by
the insertion of a plug, as shown in Fig. i.
The adoption of this form of device to the
heavier current art required larger terminals and more area of
plug connection.

Fig. 2 shows the form commonly used. The functional weak-
ness of this connection was that the adjacent breaking surfaces
were too near each other, so that on withdrawing the plug the
arc followed, and could be maintained between the terminals.

CIRCUIT BREAKING DEVICES

FIG. 2

The arc was extinguished originally by being blown out ; but as the
current density ran up, other means were
used, and it was a familiar sight to see a sand-
box handy, so that if the arc was too fierce to
be blown out, it could be extinguished by
throwing a handful of sand on it. Two plugs
were also connected in series, one to break the
main circuit, and the other to extinguish the

resultant arc.

\J The first invention in switching mechanisms

was the introduction of a plug and flash plate

t\ which were normally depressed below the surface

of the plug switch base; but on with-

I I A, drawing the plug this false insulation

plug and flash plate severed the arc by
springing between the terminals. Fig.
3 shows the next step where, by separa-
ting the circuit terminals, the arc cannot
hold; in this switch is also shown the ele-
ments of the knife switch, which soon fol-
lowed.

The circuit breaking devices were, during
the same period, following along other lines FlG 4

which were in their inception a modification
of Fig. 4, the strap key, and Fig. 5, the
telegraph key. Fig. 6 shows the earliest
form adopted from this origin, for heavy
current-carrying devices, and was first made
in 1879, for headboard switches for the Edison "Z" dynamos.

FIG.

FIG. 5

CIRCUIT BREAKING DEVICES

FlG -

In this switch the contact faces consist of two abutting surfaces,

one being rigid on the switch base and the other
attached to a lever which forms one terminal of
the circuit, being held positively in open or shut
position by means of a spring compressing a
stop in a two-way jockey plate, as shown.

Fig. 7 shows another basic form, from which
was probably developed
FlG - 6 the design of the switch

shown in Fig. 8, but this development took
place at a later period.

From the switch shown in
Fig. 3 the current-carrying
contact became a flexible
plate, bearing against a lever
which was withdrawn, and
here the advance was made
of having a wiping contact,
and the arc was not drawn on the current-carrying surfaces. Fig.
8 shows the first mechanical
form of this switch, and
Fig. 9 another form where
the pivot forms one ter-
minal. These switches
were next designed in the
vertical plane, in order to reduce the room
occupied, and took the forms shown in
Figs. 10 and n. At this time the high
potential large current art required another function in switching

FIG. 8

FIG. 9

FIG. 10

FIG. it

CIRCUIT BREAKING DEVICES

devices, which was the severing of the current with great rapidity,
in order that the arc should
not follow after the severing
member of the switch.

Fig. 12 shows the first
attempt to attain this object.
The device consists of two
flexible contacts which are connected by a conducting blade, the

lower edge of which has an insu-
lated blade which, on the withdrawal
of the conducting blade, introduces
an insulating medium between the
terminals of the switch. Also for
rapid removal, on withdrawing the blade, a
spring is put under compression, which snaps
the connecting blade away from the circuit ter-
minals.

The next form of switch, Fig. 13, is an adap-
tation made from Fig. 8, with the
addition of a mechanical device
to throw the switch quickly by means of compress-
ing a spring when the handle is thrown over,
which will withdraw the blade from its contacts
and snap it in open position.

After these, numerous devices were adopted,
having for their function the putting under tension
the blade as it was withdrawn from the terminals
or clips. Fig. 14 shows the first form, where the
blade is first under tension from a spring on the handle, which

FIG. 12

FIG 13

OF THK

UNIVERSITY

CIRCUIT BREAKING DEVICES

FIG 14

FIG. 15

then engages with the blade, withdraws and snaps on being
relieved from the contact friction. Fig. 15 is
another form where the blade is withdrawn under
tension by means of a spiral
spring.

Fig. 1 6 is
still another
form where

the blade is snapped from contact by a
spring under compression. Fig. 17 has
an auxiliary snapping tongue, which
finally severs the circuit.

Fig. 1 8 shows a snap switch where the
blade is divided into two parts entering
one clip, the
upper half
leaving the
contacts and
placing the

lower half under tension, which is
withdrawn with a snap.

Fig. 19 is a variation from Fig.
1 8, in having the snapping and mov- FIG 17

ing blades enter two different
clips, while a tension is placed
between them on withdrawing the
blade fixed to the handle.

Fig. 20 shows auxiliary con-
tacts composed of carbons. The purpose of these is to effect the

FIG. 16

FIG. 1 8

CIRCUIT BREAKING DEVICES

same result as that practically obtained by quickly snapping the
blade from its contact. This arrangement intro-
duces a high resistance in the circuit before sever-
ing it, and the points of arcing at breaking the
circuit are between the carbons. The result
is an arc of high resistance, which is easily
extinguished. A fuse has been
bridged by a switch blade
which will blow after opening FlG - I9

the switch, and in this way protect the switch termi-
nals. On high inductive circuits with alternating
currents, reactive devices have been placed between
the terminals of a switch which opens the main cir-
cuit, and then the reactive device is cut out. All
FIG. 20 these arrangements have for their object the reduction
in volume of the main current before finally opening the circuit.

SWITCH TERMINALS

The positions of the parts and members of the switch are
largely a matter of convenience, but the method of designing in
order to decrease the drop through the switch
with the least possible amount of material is one
of contacts and their physical properties. This
matter has been given a great deal of thought,
and it can be said that there is yet no standard
form of contact.

The Germans, English and Americans have

each evolved their own type, each with its distinctive merits.
Taking up the original type of contact, which consists of a rigid

SWITCH TERMINALS

FIG. 22

FIG. 23

member sliding between two surfaces, the first form brought out was
a blade fitted between two rigid walls of metal, as
shown in Fig. 21. This was used in electro-
plating practices, where the switch was not dete-
riorated by arcing effects. The next step, Fig. 22,
was to have one side of the terminal, into which

the blade entered, flexible, and
holding the blade against a rigid
support in order to allow a slight movement of
the contact, due to inequalities of the contact sur-
faces. Fig. 23 shows where both of these clip
surfaces have been made flexible
in order to further increase the contact surfaces
between the terminal and its blade.

Fig. 24 shows another method of
securing the flexible contacts to the
terminal plate. Fig. 25 shows the
same general form of construction,
but more flexible in its bearing on the contact sur-
faces, and the flexible contacts soldered into grooves
cut in the terminal plate. Fig. 26 is a form used in
small-capacity switches, where the flexibility of contact
y: is obtained by a continuous metal plate

fife^. bent to form both contact clips. Fig.

ft 27 shows the same result obtained by

k ^^^L reversing the conditions found in Fig.
^^IBl 25, the moving arm, in this case, being

Fic - 27 the flexible contact member and the terminal being

the rigid contact member. The object of this arrangement was to

FIG. 24

FIG. 26

SWITCH TERMINALS

FIG. 28

FIG. 29

reduce the drop on the switch, due to the discontinuous connection
between the flexible contact member and the ter-
minal. Fig. 28 shows a multiplication of the same
general construction as Fig. 25, in order to increase
the area of the contact without
introducing the inflexibility of
one large moving switch blade.
Fig. 29 is the element from which the Ger-
man and English types of switch emanated.
They consist of a flexible contact surface, bearing on a rigid ter-
minal surface ; in this case the contact is a flexible
brush, wiping over the terminal.
This has been reduced in heavier
sizes of switches to a number of
flexible fingers, bearing against the

contact terminals, as shown in Fig.

FIG - 3 r- .LI i

30. .big. 31 shows the same ele-

ment of construction, but where the blade is intro-
duced between the rigid terminals. Fig. 32 shows
the construction which effects the same flexibility between the dif-

ferent fingers of two contact surfaces, the two sur-

faces being in this case the sepa-

rate terminals of the circuit.
Fig- 33 shows the recent

American improvement in

contact surfaces where the

movable contact member con-

sists of a laminated surface of
a great number of individual spring contacts, which, on their intro-

FIG. 31

SWITCH TERMINALS

duction between contact surfaces, offer individual flexible contact
points for carrying the current from the movable member to the termi-
nals, and in this way presenting a large surface under compression.

CONTACT SURFACES

The materials which form the contacts should possess the quality
of not mutually abrading each other when rubbing together. Two
similar metals are liable to bite and tear the contact surfaces, and
they should be selected so as to have different physical character-
istics, in order that they will wear well together.

The conductivity of a contact surface is dependent upon three
values : the pressure which they bear on each other, the character
of surface exposed to the conductivity of current, and the character
of metals forming the contact areas.

As a conductor, a contact surface behaves as if it were com-
posed of a multitude of points on a flexible warped surface, and as
the pressure is increased, more of the points come in contact, and
the pressure plays a more important part in the conductivity of a
contact than the area. With a rise in temperature of the termi-
nals, the actual drop between the contact will fall, if there is no
local potential due to the metals in the contact setting up a local
thermo-electro-motive force, or Peltier effects.

In the case of copper and zinc, and their alloys, there is no
appreciable thermo effect, but for lead-aluminium, lead-iron, tin-
aluminium, tin-iron, bismuth-iron, bismuth-aluminium, in combi-
nations, there is a very appreciable resistance over normal, due to
local counter electro-motive forces ; the resistance of contact rises
rapidly with time, and all local currents tend to depreciate the sur-
faces of contact through which they act.

SWITCHBOARD CONSTRUCTION.

OCATING the switchboard is determined by the relative
positions of the operating machinery in the plant. It is
desirable that all points of control, namely, the throttle
of the engine, the switchboard and the dynamo
commutators, be near together, and so arranged
that they can be easily reached by an attendant
without squeezing past fly-wheels or belts.

In short, do not arrange your apparatus so as

to put the attendant under any physical hazard.

When trouble arises, it is necessary to take care of several

things at once, and their close proximity to each other renders

the attendant more eflicient. This is

especially true in isolated plants. In

moderate-size stations the functions

of each attendant are less involved,

and the switchboard and genera-
tors are generally under one man's

care. In the larger stations the

general practice is to have one attend-
ant whose only duty is to take care

of the switchboard. The modern

tendency, both abroad and at home,

is to place the switchboard on a

gallery or in an elevated position,

where the attendant has 'a full view of the operation of the

FIG. 35

GALLERY DESIGN AND LOCATION

PLATE -3*.

generators under his charge, and can act promptly in a case of
emergency.

The following shows a number of designs of gallery construc-
tions ; also stairways leading to gallery ; both spiral and straight.
Designs of railings are shown in Plate 34; Fig. 35 shows a design of
a gallery of cantilever construction; Fig. 36 shows a double-decked
switchboard, reached by a spiral stairway, and Fig. 37 shows a

plain .railing for
a slightly ele-
vated gallery.
Each case of gal-
lery construction
is entirely de-
pendent upon
the architectural
arrangement of
the station, and
these illustra-
tions are given
only to indicate
certain prac-
tices. In this
location of the
switchboard, if
the dynamo
leads are run
above the dyna-
mo to an under-
ground system,

FROM DESIGNS BY HERR1CK & BURKE.

SWITCHBOARD LOCATION

or underneath the dynamo to an overhead system of distribution,
extra copper in these two cases is not necessary to conduct the
current to the gallery. Where the current has to be carried up
to the gallery and back again, there is considerable length of con-
ductor used, only on account of the gallery location.

Where the handling of 100,000 amperes is concerned, as in
some of the larger three-wire systems, where distribution is
underground, the indications are that future practice will be to
locate the bus bar in the same plane as the distribution and supply
systems, and to operate the switching mechanism from the gallery
by mechanical or transmission methods, thereby saving a large
expense in conductors and waste of energy in transmission, there
being located on the gallery only the measuring instruments and
regulating devices, and the levers to operate the circuit-changing
switches. Compressed air is used at present for mechanically open-
ing circuits at high speeds at a distance from the operator.

In some cases, the external distributing system will be the
determining factor for the location of the switchboard, in order

to have short internal conductors and
reduce internal losses; where the cur-
rent flow is large, the matter of conduc-
tor lengths becomes a very important
factor.

There is another marked general ten-
dency in switchboard construction, that
is, to make the whole structure absolutely
FlG - 3 fire-proof; there is no reason why a

central station should contain any combustible material other than
the necessary waste and oil.

CONSTRUCTION DETAILS

There is at the command of the electrical engineer to-day fire-
proof material for every form of switchboard and insulation, neces-
sary in central station construction. Asbestos aifords a very good

fire-proof covering for the conductors, and pre-
vents them from conducting fire to different
parts of the building; low-tension conductors
may be bare, and supported by porcelain insu-
lators (see Fig. 38) or bus bars supported by
marble, as shown in Figs. 39 and 40.
FlG 39 Every advance advocated by insurance

inspectors in this direction is mutually important for the central
station manager to preserve this class of property from destruction
by fire, besides being an economical investment in the way of
obtaining lower insurance rates on this class of risks.

Do not place the switchboard under steam pipes, or have an
exposed window open on the back of the board. Do not allow
automatic sprinklers to be placed over the board, for if they should
act, the current leakage through the wet
surfaces would cause a far worse hazard
than could normally exist with a properly
constructed switchboard.

The switchboard should be supported
away from the wall, in order to have the
back connections accessible ; the underwri-
ters' rules require such placement in several districts. The dis-
tance between the wall and the back of the board has not been suf-
ficient in the larger boards, as there should be three and one-half
feet in the clear in order that the attendant can properly work
behind the board, and not make false connections with his tools.

FRAMING

Wooden frames, known as the skeleton form of construction,
have been used for the supporting of switch-
board instruments. This step was taken to
reduce the amount of combustible material used,
and to separate the switchboard from the panel-
ing or woodwork of the station. Fig. 41 shows
method of joining and making skeleton switch-
boards.

FlG - 4I Oak or ash is the material gener-

ally used. The horizontal slats are placed at such
distances apart that the apparatus can be readily
secured to them. This form of construction was very
much in vogue some years ago, but, on account of the
fire hazard, it was abandoned where switchboard con-
struction was seriously considered.
The next step was to substitute slate for the slats,
and still use the wooden vertical supports, with
the instruments and devices mounted on the slate.
Later, I-beams, channel bars or "L" iron were
substituted for the wood to support the marble or
slate ; in this way the space occupied by the supports
for the board and bus bars was much
less, and the clearances behind the board
were more favorable to make good connections.

Figs. 42 and 43 show methods of securing mar-
ble or slate to the verticals.

Standard steel sections being used, the switch-
board can be readily erected; but the precaution to
have them insulated from the building structure should always be

FIG. 42

FIG. 43

FIG. 44

FRAMING AND CONNECTIONS

borne in mind where iron framework is used. This can be readily
done by supplying foot-plates of marble, as shown in Fig. 44, and
having the guys or expansion bolts, which stay it from the wall,

insulated by a coupling, as shown in

Fig- 45-

In railway, high potential, and three-
wire systems with grounded neutral,
it is very essential to have the iron
framework carefully insulated from the building structure, in
order to bring up the ground resistance, as well as to prevent any
jumping of current or running discharges behind the board to
ground ; also the hazard of injury to attendants working behind a
switchboard which is insulated from the ground is greatly reduced.

CONNECTIONS

Methods of making connections between conductors behind a
switchboard are of various kinds, depending on the form and purpose
of the conductor. These various methods are connected together
and shown in Plate 46; some of them are standard, and some
have inherent weaknesses which have led to their abandonment.

The first shown is the familiar wrapped splice, which is used
with bare conductors, where they are both served with copper wire
and soldered together.

The second is the sleeve, where a thin brass tube is slipped over
both ends to be connected and soldered.

The third form is a clamp connection, where both conductors
are parallel and clamped together in one connector.

The fourth connection is where the wire is turned under the
head of the bolt and screwed down.

PLATE 46

CONNECTIONS

The clamp connection shown in the fifth is the "V type,
where two pieces with "V recesses clamp the wire.

The sixth is an obsolete connection, where a sleeve is split and
provided with a taper thread on the outside, over which screws a
taper nut, and compresses the sleeve over the copper rod. The
weakness in this connection consists of the fact that when the con-
ductor heats, it expands and stretches the clamping nut so as to
loosen the connection. All connections which have in their incep-
tion the surrounding of a solid conductor with a sleeve which is
not strong enough to resist stretching under expansion, will event-
ually give trouble by heating.

The seventh shows the regular lug connection, and the eighth a
connection often provided for in the terminals of back-connected
switches; if the stud screws into the terminal and the nut locks
these together, the arrangement is satisfactory; but if the stud
passes through the switch with a nut only on top, and the bearing
on the bottom a small shoulder, this method of connecting will
eventually give trouble.

The ninth form shows where a threaded stud is secured to a bus
bar by means of two nuts, which form a good connection. Angles
are usually formed in bus bars by means of bolts; steel or iron bolts
should always be used for this purpose.

The tenth form is the method used when a bus bar connects to
a bus rod.

The eleventh connection is when the end of this rod is cut to a
taper ol twenty degrees, and a female taper lug is bolted down on
it. Where these surfaces are ground together, it makes a very
good form.

The twelfth shows the German method of securing a cable to a

CONNECTION EFFICIENCIES

lug, by forcing taper screws into the stranding, and in this way
expanding the cable and securing contact.

There are a great number of other methods used, of connections
for special purposes, but not of general application to switchboard
connections.

We have in the case of a contact an effective negative coeffi-
cient for temperature ; we may explain this in this way :

As the contacts expand, they tend to present more surface of
contact, and are under contact at a higher pressure than when at
normal temperatures. The effect of the increased efficiency of a
joint at elevated temperatures is very clearly shown when the parts
in contact are held by a steel bolt. Either brass or copper expands
faster than the iron bolt, and under these conditions you can enor-
mously increase the pressure on the contact and decrease the losses
at this joint. The conductivity of the iron bolt is not of as much
importance as this increased pressure by unequal expansion of the
different parts of the conductor.

To take the volts drop on a connection, and multiply it by the
amperes passing, will give the watts at that particular current den-
sity ; but where it is a matter of contact surfaces, the drop will not
follow as quickly as a current rises. To assume that this is pro-
portional leads us into very grave errors.

In specifying any system of conductors for switchboard and
operating devices, the losses should be expressed at full load, in
volts drop.

BUS CONDUCTORS

There has been an erroneous idea that by laminating and allow-
ing large radiating surfaces, conductors can in this way be kept

BUS CONDUCTORS

cool, and consequently the losses reduced ; some have advocated
forcing the density up to as high as three thousand amperes per
square inch for copper. The supposed gain is keeping the temper-
ature down so that the resistance will not increase, due to the tem-
perature coefficient for that conductor; but there are constant losses
in energy, which, if saved by using better proportioned conductors,
would pay a handsome interest on the investment for the additional
copper. An example will illustrate this fallacy more forcibly :

Suppose we had six thousand amperes to carry one thousand
hours in one year 40 feet. With bus bars at a current density of
eight hundred and fifty square mils per ampere, and using five
bus bars in multiple, 2 x % inch x 40 feet. R=. 0000668. The
watts lost will be 2,404 per hour, or 2,404 kilowatt hours per year,
which if produced at a cost of one cent per kilowatt hour, the loss
will cost $24.04 per year. This bus bar weighs 771 pounds, and
will cost, erected, approximately $308.00.

Take the same case as above, but using a density of three thou-
sand amperes per square inch, or three hundred and thirty square
mils per ampere, and we will increase the radiating surface by
using a bus bar 2 x V 8 inch x 40 feet, which will have a resistance
=.0001688. The watts lost per hour will be six thousand, and the
kilowatt hours per year, six thousand. The cost of production is
one cent per kilowatt hour ; this will make the lost cost $60.00 per
year. The weight of the bus bar is 155 pounds, and the. cost to
erect $70.00; the difference in the losses is $35.96, and the differ-
ence between the two investments is about $238.00, or for the
additional expenditure of $238.00, which would be necessary in
order to have the current density 850 square mils per ampere, this
additional investment will earn fifteen per cent, by the economy in

BUS CONDUCTORS

waste energy, effected by this additional expenditure in copper.
Again, in the case cited, the laminated copper bus, one-eighth of an
inch thick, has to dissipate .33 of a watt per square inch of surface,
whereas the larger bus has to take care of only .2 of a watt per
square inch of surface exposed. In this case the larger bus is
working more advantageously regarding ultimate temperature
obtained, and will increase resistance less, due to this rise in tem-
perature, which will again be in favor of the larger bus bar.

This example is worked out for the reason that a great deal of
engineering is done on what is known as the least first cost basis,
regardless of what this extravagant economy costs in operation.
Such cases are more clearly demonstrated by practical examples
than by generalities.

The materials of electrical engineering, especially those used for
conductors, are more often put in by faith than by test. The
station manager who will have his boiler-plate tested, which will
not represent more than one-fortieth of the capital invested in the
plant, will neglect to have his copper tested, which will represent
anywhere from thirty to sixty per cent, of the capital invested in the
plant ; yet poor conductivity in the copper distributing system may
seriously affect the dividend which should accrue to the installation,
and the current uselessly frittered away in heating the conductors.

It will be a wise plan, and should always be required, where
there is any considerable investment of copper to be made, to have
submitted by the manufacturer a sample piece of fixed dimensions,
delivered and tested for its conductivity, and if it is to be used for
overhead work, it should be also tested for tensile strength.

The conductivity of copper is seriously affected by the presence
of other metals, even in very small quantities, especially arsenic

BUS CONDUCTORS

and tin. The most important impurity which will appear if the
copper is not properly handled in smelting is the sub-oxide of cop-
per. The brittleness of electrolytic copper is generally due to the
presence of copper hydride formed during deposition. The
low conductivity of over-refined fused copper is due to the pres-
ence of carbide of copper, which is formed in the presence of carbon
as soon as the sub-oxide disappears. Copper impurities can only
be detected first by the physical properties of the copper, and
second by a chemical test.

The steel and iron manufacturers fill specifications requiring
fixed physical properties in their products so should the copper
producer be required to fill both electrical and mechanical condi-
tions, which are so important to the successful operation of a plant
from a commercial standpoint.

Other conductors, such as iron, aluminium, etc., have been pro-
posed, but in all cases the conductivity has been so low that the
mass to carry any given current is from seven to thirteen times that
for an equivalent copper bus bar, and this larger conductor requires
much more space than can be afforded behind the switchboard ; the
insulating expense and the cost per unit of current carried is
greater than with copper at the present market prices.

The dimensions of bus bars are generally selected by the cur-
rent which they have to carry, and the connections which have to
be made to them ; two copper busses bolted together w r ill carry
about one hundred and eighty amperes per square inch of contact
section, and the cross-section carries approximately twelve hundred
amperes per square inch. Consequently the dimensions of the bus
selected should be such that it will present proper area of contact
for connections, without making them excessively long.

BUS CONDUCTORS

The current which is found in average practice, which can be
carried economically by copper bus bars, is given in the table
below; these current densities are covered by ordinary central
station practices where the load factor is not greater than fifty per
cent. The resistance per foot, the area in circular mils and square
mils, and the weight per foot are given.

COPPER BAR DATA

SIZE

AMPERES

CIRCULAR
MILS

SQUARE
MILS

OHMS
PER FOOT

WEIGHT

PER
FOOT

i x ^ in.

433

318310

25OOOO

.0000336

i^x>4

530

397290

3 i 2000

.0000269

1. 21

*y* x y\ "

626

477465

375000

.OOOO223

i-45

13^ x ft "

725

556400

437000

.OOOOI92

1.70

i^x y a

676

596830

468750

.0000179

1.82

i^ x # "

798

7l62OO

562500

.OOOOI49

2.18

13^ x % "

916

835600

656250

.0000128

2-54

2 x % "

1035

954930

750000

.OOOOII2

2.92

2%xf/

H54

1074300

843750

.OOOOO995

3-27

2^ X X "

1500

I59I550

1250000

.OOOOO672

4.86

2^ X % " 1715

1989440

1562500

.00000537

6.07

2 X^ "

No. oooo B. & S.
y z in. Round

K "

K " "

j (i u

1222

257

305
426

560

861

1273240
2II6OO
25OOOO
390625
562500
IOOOOOO

IOOOOOO

.00000840
.0000505
.OOOO428
.OOOO273
.OOOOI9O
.OOOOIO7

3.89
.64
.76
1.18
1.71
305

Bus bars are formed by having the copper billet drawn through
graduated dies. If these dies are not properly graduated, or are
dull, the surface of the copper becomes torn and reduces the use-

BUS CONDUCTORS

fulness for bus bar work, as a good connection cannot be made on
an abraded surface. If the surface of the bus bar is warped it
should be straightened, in order to make a good contact. Specifi-
cations for bus bars should be drawn so that they can be rejected
for any of the above mechanical imperfections. Bus bars are
drawn soft, medium or hard, as ordered. Medium is the bus bar
usually ordered. Ordinary tools and ordinary tool speeds will
work copper satisfactorily. If the tools tear instead of cutting the
metal, use milk as a lubricant, and the cutting will be perfectly
smooth.

COMPOSITIONS

In any mixture of other metals with copper, the resultant alloy
will be reduced in conductivity below the mean of the different
metals forming the mixture. There has been no law yet discovered
between the conductivity of the different metals and the resultant
alloy, from which the conductivity of the alloy can be predeter-
mined. It will be seen in the table on following page that the con-
ductivity of these alloys is more rapidly reduced when any quan-
tities of tin are added to copper, than with an equal portion of zinc,
except in the instance when tin is present in very small quantities,
just sufficient to make the metal flow into a mould ; this will give
the highest conductivity of metal that casts readily, except in the
case of one-half per cent, of silver and ninety-nine and one-half
per cent, of copper, which gives a conductivity of eighty-nine
per cent.

No old metals should be remelted where conductivity is required,
as they have very low conductivities, and metal otherwise good
may be burned in the pot and seriously affect the resultant com-

COMPOSITIONS

position. Brass is a mixture of two parts copper and one part zinc,
and from the peculiar properties of this mixture, it is probable that
there is a chemical combination at this ratio, and not a purely
mechanical mixture, as with most alloys. So-called cast copper

COPPER
PER CENT.

ZINC
PER CENT.

TIN

PER CENT.

CONDUCTIVITY COM-
PARED WITH COP-
PER AS 100 PER CENT.
CONDUCTIVITY

VALUE AS A CON-
DUCTOR COMPARED
WITH PURE COPPER
AT 20 CTS. PER LB.

98.44

94.49
88.89
86.67
82.54
75.00
73-30
67.74

oo.oo

98-59

93-98
90.30

89.70
88.39
87.65
85.09
16.40

I. 5 6

5-51
II. II

r 3-33

17-5

25.00

36.70
32.26

100.00

46.88
33-32
25.50
30.90
29.20
22 08
22.27
25.40

27-39
62.46
19.68
12.19
IO.2I
12.10
10.15
8.82
12.76
11-45

42.22
59-20
78.40
64.60
68.40
81.40
89.80
78.60

73-oo
32.00
101.60
164.00
195.80
165.20
197.00
204.00
159.80
1 74.60

cents

i i

i i
i t
.1
it

u
Ii

1 1
II
II
U

, I
II

1 I

1.41
6.02

9.70
10.30

ii. 61

12.35
14.91
83.60

IOO.OO

varies all the way from eighteen per cent, to eighty-nine per cent,
conductivity of pure copper, and a number of methods are used, as
well as mixtures, which will make the copper flow into a mould
and conform to the pattern.

COMPOSITIONS

In switchboard construction the current-carrying appliances
should be so designed that they can be worked directly out of pure
copper stock ; do not use cast metal where the current densities are
high, and where the economy of material and labor is to be con-
sidered.

The bronzes are employed in special cases for their mechanical
properties rather than their electrical properties, as their conduc-
tivities are low.

SWITCHBOARD MATERIAL

The actual surface on which the switchboard appliances are
assembled should have the physical properties of possessing
mechanical strength, insulating, and be fire-proof; besides, it is
necessary that it be drilled readily and have a good surface on
which to mount the diiferent appliances, which are combined
together to form the switchboard.

Wood possesses the mechanical strength and insulation when
dry, but it is not fire-proof. There have beeri a number of attempts
made to impregnate the wood with silicate of soda, or other fire-
proof compounds, or even to paint it with fire-proof paint, in order
to make it incombustible. Wood has also been covered with
asbestos paper, in order to serve the same purpose, but none
of these methods have proved satisfactory.

The natural stones offer the best material for switchboard sur-
faces, and, above all, slate is the most generally used. The insula-
ting qualities of slate vary with diiferent mines, and also with dif-
ferent parts of the same mine ; but generally speaking, a uniform
color, gray or brown slate, without any marked seams or veins, will
be found to be a fair insulator. Slates which fracture readily

SWITCHBOARD MATERIAL

along the veins, and in which the fractured surface shows a semi-
metallic color, are always treacherous to use. Slate should have a
bright, laminated fracture along the plane of its natural cleavage,
and if it has a dull fracture, it is very liable to be too porous, and
will absorb moisture, which will reduce its insulating value. As a
rule, soapstone is too soft and fragile to be used for switchboard
purposes, and does not hold enameling well. Slate surfaces are
treated with enamel paint, and different imitations of marble and
wood are in this way made, the enamel being burnt in and hardened
and polished to a surface.

Among the marbles every variety can be found in regard to
color and finish, but not many of them possess high insulating
qualities. White Vermont is soft and fragile, as well as being too
porous to exclude moisture, and a drop of oil will spoil a slab ; for
these reasons, it is very rarely used for switchboard purposes.

The granites are too hard to drill and holes will drift when hard
spots are reached.

There are a number of Tennessee marbles, such as the gray and
the champion pink, both of which have, as a rule, high insulating
qualities and are easily worked ; they also make very pretty com-
binations with dull coppered finished appliances.

Of the foreign stones, Italian marble and Mexican onyx make
very good surfaces, and as a rule are good insulators. The former
is probably more extensively used for switchboard construction
than any other marble in America.

To drill marbles,, use the slow-speeded twist drill ; there will be
no trouble in drilling marbles in this way, if the drill is not allowed
to choke with the marble dust, because when this happens the drill
gets hot and draws the temper. Where large holes are to be cut,

SWITCH BOARD MATERIAL

the quickest way is to use an iron pipe, the right size outside, with
teeth cut on the lower edge ; rotate this in the drill chuck and feed
with emery ; in this way the hole can be ground through very
quickly. When there is trouble with the holes breaking out in the
back, with large size drills, first drill an eighth of an inch centre
hole and then drill with a large drill from both faces.

TILE CONSTRUCTION

Glazed tiles have been built up as a wall, forming a surface for
the switchboard, the tiles being laid in cement, and the holes
drilled through with hand drills, where required for securing appa-
ratus; also cast-iron frames, in which are set porcelain blocks
moulded with holes, so that apparatus can be thus secured and
insulated.

SECURING AND MOUNTING SWITCHBOARDS

Natural stones, when secured to wooden structures, are very
liable after a time to crack, by the warpage of their supports. In
securing marble to iron structures, the only precaution necessary is
to support the marble slab under the securing bolt, so that the mar-
ble is not sprung to conform to the surface of the iron work to
which it is secured. Asbestos washers between the marble and
iron, through which the securing bolts pass, form a very good bed-
ding for marble slabs, and will support them permanently without
danger of fracture. Natural stones can be ground and polished to
within one thirty-second of an inch, and when so ordered, all slabs
for one board can be cut from one block, so that all veining can be
matched up, and the color will be uniform. It is necessary, in

SECURING AND MOUNTING SWITCHBOARDS

natural stones, to submit them to a test; the test usually made,
where only low potentials are used, is by means of a magneto.
There are two flat surfaces furnished with the leads to the magneto,
and these are pressed on adjacent surfaces, or opposite surfaces of
the marble or stone to be tested ; if the magneto does not ring up
with this connection, it is passed. All that this test indicates is
that there is not a distinct metallic vein on the surface of the slate.
It is the veins passing through the slate which form the conductors
that render the slate useless as an insulator. The best way to test
these veins is to use a sharp-pointed terminal, and stab the vein at
two points close together; if, under this condition, you cannot ring
the magneto, the veins are probably non-metallic. Where slate or
any veined material is to be used for high potential, it should be
submitted to the actual potential stress, and should require over
fifty volts per mil potential before it breaks down ; no marble
should be exposed to potentials over six hundred volts, without the
live terminals being separated from the marble itself, by micanite
or other equivalent insulator, as marbles and slates leak excessively
in moist weather, when they are acted on directly by high poten-
tials. This leaking heats the marble, disintegrating it, and in some
cases explodes it.

SWITCHBOARD APPLIANCES

OLTAGE measurements should be made with
greater precision than any other electrical
measurements in central station work, espe-
cially in low potential multiple arc distribu-
ting systems ; a constant potential is necessary
on the consumption circuits, in order to main-
tain an effective service and increase the life of incandescent lamps.
The conditions which surround measuring instruments in
switchboard work must be taken into consideration in their selec-
tion. Those voltmeters having magnetic torque for their indica-
tion must be shielded from external magnetic influences when
near iron structures or field magnets. This is generally taken care
of in their design, either by shielding or working under such
intense magnetic conditions that stray fields will not be of suffi-
cient magnitude to introduce a commercial error.

With a varying voltage, it is essential that the instrument be
dead beat in its action, and quickly respond to any impressed elec-
tro-motive force at its terminals.

The current flow through a voltmeter should be as small as
possible, and the temperature coefficient of the resistance, in series
with the moving system, should be low, and should give a negligi-
ble temperature error for the instrument to be of any real value.

An indicating instrument must be theoretically correct in prin-
ciple, accurate throughout the whole range, simple in construc-
tion, and so proportioned to the work which it has to do that it

SWITCHBOARD APPLIANCES

will not be eternally getting out of order, and will stay in correct
calibration.

To check up the accuracy of voltmeters, the temperature error
is found by checking two voltmeters against each other when both
of them are cold. Then place the voltmeter across the potential
at which it operates for a number of hours, until its temperature
has attained a maximum. Again check these two voltmeters, the
cold against the hot voltmeter, and if the readings now differ from
each other more than one per cent., the voltmeter should be rejected.

Another source of error in voltmeters is due to pivot friction,
which is caused by the pivot wearing against the jewel bearing and
abrading the surface, and increasing the friction between these two
surfaces ; this error should not be found in new instruments, as it
is one which increases with the ageing of the instrument. In
order to keep this friction as low as possible, the moving system
should be light, and in order to test friction errors, the instrument
should be moved quickly, in order to oscillate the index hand
note whether the hand always comes back and registers with the
zero mark current should be applied to the instrument, and dif-
ferent parts of the scale tested in the same way. The important
parts of the scale where friction is found largest are near the standard
readings, where it is continually used. The voltmeters should
always be tested for the friction error in the same position in which
they are used.

Means are generally provided on the switchboards for checking
up the different voltmeters against each other, by means of plugs
and a pressure switch. This should be done at stated intervals, in
order to maintain the accuracy of these instruments, which are so
important to the proper operation of the system.

SWITCHBOARD APPLIANCES

For alternating currents, the voltmeter should be calibrated at
the same frequency as that on which it is used. In some types of
voltmeters a change of frequency introduces a considerable error in
alternating current instruments ; especially where they are applied
to circuits having a large power factor, they should be carefully
checked for errors.

INSTRUMENT MOVEMENTS

The different principles employed in the movements of electrical
instruments can, as a rule, be used for both volt and current meas-
urements. The magnetizing force, in the case of the voltmeter,
is supplied with a great number of turns of fine wire, and, in the
case of the ammeter, is supplied with a few turns of coarse wire.

The oldest type of a measuring instrument was a permanent
magnet pivoted over a conductor, which magnet was deflected by
currents passing through the conductor, and these deflections were
calibrated.

The movements of commercial instruments, as now used for
switchboard work and commercial measurements, are only those
which will be hereafter described .

The solenoid type, in which an iron core is
sucked into convolutions, through which the cur-
rent circulates, has a pointer attached to the core,
and these deflections are calibrated. Fig. 47
shows an instrument employing the solenoid
principle; in this case, the core is supported on
a knife edge, counterbalanced by the pointer FlG - 47

and an adjustable weight, and can be calibrated to give a
fairly uniform scale for equal increments of current. This

FIG. 48

INSTRUMENT MOVEMENTS

design is one which is used by the Brush Company on their arc
circuits.

Fig. 48 shows another form of solenoid type; in this case the

solenoid is circular and concentric to
the curve core which swings around
the axis of support, in response to cur-
rent changes.

By properly balancing this system
by weights, a fairly uniform scale can
be secured, or, for special purposes, the
scale can be expanded at any point
desired. This movement was used by
the Edison Company.

These two movements are specially
useful for current indications; they have an inherent error in
measuring direct current, due to the iron core not responding
readily to slight changes of current, which makes an increased
reading lower than it should be and a decreased reading higher
than it should be. This magnetic lag is called "hysteresis," and
where there is any mass of iron employed in the moving system
the accuracy is not sufficient for volt indications, except where
alternating currents are measured, and hysteresis is, under this
condition, not an appreciable error, but in this case, the precaution
must be taken to laminate the core, so that local currents will not
be induced in it.

Both of the systems above described are naturally heavy in
their construction, so that the coefficient of friction of the instru-
ment is large, and gives an error which is practically identical
with hysteresis.

INSTRUMENT MOVEMENTS

FIG. 49

Fig. 49 shows still another instrument of the solenoid type,
which is used specially for alternating current work, being provided
with a laminated core. This movement is
used by the Westinghouse Company.

The perfecting of devices for the accu-
rate commercial measurements of electrical
currents and potentials is due to the untir-
ing efforts of Mr. Edward Weston, who
undertook, in 1880, to seriously develop for
the electrical industries instruments by
which the electrical quantities involved in
electrical engineering could be accurately
determined. After having experimented
over the whole field of the possible forms of
instruments, he determined on the type of
instrument shown in Fig. 50, believing it to embody the best possible
principles for the measuring systems of commercial instruments.
This type is applicable to the measurement of direct currents; the

measuring coil consists of a number of
turns of fine wire, wound on a rectangular
form, and this coil rotates in a concentric
annular space, through which passes a
permanent magnetic field. When current
enters this coil, it tends to deflect it from
a fixed position, which is reacted by dif-
ferential springs; in this way deflections
can be obtained, which are proportional to the current flowing.

"A" represents the internal iron core; "B" the moving coil;
U C' the springs, and "D" the pointer.

FIG. 50

INSTRUMENT MOVEMENTS

Where this instrument is used as a voltmeter, it has resistance
in series with the moving coil. When used as an ammeter, it
measures the difference of potential across a shunt, which differ-
ence varies with the current flowing. These instruments are made
in a number of forms for switchboard work the round dial type
for isolated plants, illuminated dial type for central station work,
and the edgewise type for heavy current
work. All these are shielded by being
closed in iron boxes, which form the case.
Fig. 51 shows another instrument hav-
ing the elements of a moving system rota-
ting in a permanent magnetic field, this
movement being restrained by springs

which also carry the current into the moving systems. The instru-
ment shown in this figure is known as the Kennelly Ammeter,
manufactured by the Edison Manufacturing Company. This instru-
ment consists of a flat permanent horseshoe magnet of semi-circu-
lar shape, with its poles brought out into semi-circles, and separated

about one-sixteenth of an inch. In this
narrow air-gap, and working in a strong
magnetic field, is the disc armature "A."
The windings are laid radially and symmet-
rically over the upper surface of the disc,
and when the current passes through these
radial lines, a magnetic pull is set up in the

plane of the disc, which causes the armature to turn ; the deflections
are proportional to the current flowing through the system.

Fig. 52 shows another method by which currents are measured
by indirect means, due to the expansion of the conductor on being

INSTRUMENT MOVEMENTS

heated by the current passing through it and the expansion and
contraction of this conductor being multiplied and actuating a
pointer, which indications are calibrated. This figure shows several
strands of wire, through which the current to be measured passes,
one strand of which actuates the multiplying device, to which is
attached the pointer. A disc is attached to the moving system,
which enters a semi-circular air-tight box, and acts as an air dash-
pot, which reduces the oscillation of the meter and makes it
approximately dead beat. This instrument is known as the Hoyt
hot wire type.

The Cardew type of instrument has the conductor wound
around the spindle of the indicating hand, and rotates this around
as it expands or contracts; this movement is again calibrated.
These instruments have special uses for shipboard work, as they
are not affected in their indications by any external movement;
twisted strips of different metals and spiral springs used torsion-
ally have been used for the moving devices of
measuring instruments.

Fig. 53 shows what is known as the Wirt
type, and consists of a magnetic field which
is unsymmetrical to the moving iron system.
The current is so disposed around the moving
system that very little iron is used, and the

r 1 r i Fir " 53

hysteretic error, by careful treatment of the iron,
is reduced to a negligible value. As the current flows through the
loop surrounding the system, the iron tends to include a greater
number of lines of force, and this moving effort is counteracted by
weights or springs, so that the deflections are proportional to the
current flowing. Where this system is used for a voltmeter, a

INSTRUMENT MOVEMENTS

number of turns wound on an elliptical spool, and the moving sys-
tem placed concentrically to one of the curved ends, makes this
form adaptable to volt indications. Hoffmann & Braum, Thomson -
Houston and others use variations of the same principle for
indicating movements.

ASTATIC VOLTMETERS

With high potentials there exists sufficient attraction between
surfaces of dissimilar potentials so that a moving system can be

operated by these surface charges.

Fig. 54 shows a type, originally designed by
Sir William Thomson, in which a moving vane
enters between two charged surfaces; this sys-
tem is pivoted and provided with adjustable
weights, so that voltages can be read up to any
potential which is only limited by the striking
distance between the charged plates. This in-
strument requires no flow of current for the
indications, and the scale is not proportional.
Other arrangements of the same principle have been made, one
by Mr. Kennelly, where the system is suspended and consists of a
horizontal vane, which carries vertical circular sec-
tors of aluminum, which rotate in grooves formed
by curved brass plates. Current is carried into a
moving vane by a bi-filar suspension. When these
adjacent surfaces are charged, they tend to rotate
the vane proportional to the charging potentials. FIG. 55

This last arrangement, on account of the fibre suspension, is hardly
adaptable to switchboard construction. See Fig. 55 for details.

FIG. 54

4 8

ASTATIC VOLTMETERS

FIG. 56

The permanent magnet type of instruments is not suitable for
the measurement of alternating currents. Mr. Weston devised the
instrument shown in Fig. 56, which has, in
common with his permanent magnet type of
instrument, a rotating coil on which is
wound the convolutions carrying the cur-
rent to be measured, and the current is
carried into the moving system by spiral
springs. The magnetic field in which this
system rotates is also formed by the current
to be measured, and consists of two hollow helixes which enclose
the moving system. There is no iron in this instrument, and it is
not affected in its readings by changes in frequency. The oscilla-
tions of the meter can be damped by a brake which is depressed on
pushing down the contact key. This instrument does not give a
proportional scale, but one that is expanded in the middle.

RECORDING VOLTMETERS

In order that a permanent record may be
kept of the variation of voltage on any system
of potential, it is important that a visual con-
tinuous record be kept of the potential, and
there are several instruments on the market
for this purpose.

The one shown in Fig. 57 consists of a
voltmeter movement of the solenoid type, in
which the plunger is supported on springs,
and has an index extension, on the end of which is a pen, by
means of which a line of aniline ink can be traced on a dial.

FlC. 57

RECORDING VOLTMETERS

This dial is moved around at a uniform rate by clockwork, and the
variations of voltage are in this way registered permanently.

The Richard Brothers' indicator records in the same way as the
above-described Bristol indicator, but the actuating mechanism con-
sists of an iron vane attracted to the poles of an electro-magnet, and
the pointer carrying the aniline ink is attached to this system.

VOLTMETER RELAYS

In order to control voltages where variable speeds are delivered
to the dynamos, such as water-power or gas engines, it is necessary
that an automatic method be used to regulate the generator thus
driven, as compounding will not affect regulation for variable
speeds.

A voltmeter relay is used for the purpose of actuating the rheostat
which controls the field circuits. There are a number of methods
by which this relay is operated, but generally by the solenoid and
plunger principle, the solenoid being wound with fine wire, which
is placed across the potential to be controlled. The plunger is
restrained by a calibrated spring, and makes connection with one
contact when the voltage is high, and with another contact when
the voltage is low. A relay can be constructed so as to regulate
within one-half of a volt.

The circuit thus controlled by the plunger may operate through
the means of a motor or electro-magnet; the arm of the rheostat,
cutting in or out resistance in the field circuit, maintains, in this
way, a constant potential at the brushes of the generator. Such
relays are also used to light signal lamps on the switchboard, to call
the attention of the attendant to the fluctuation of voltage from the
fixed standard.

VOLTMETER RELAYS

The central station in Paris, France, has its potential output
almost entirely operated by automatic devices, which are controlled
by automatic relays.

COMPARATIVE PRESSURE INDICATORS

In feeder distribution where the network of mains are supplied
by feeders whose terminal pressure should be the same throughout
the system, the pressure wires are brought back from the feeder
ends and can be directly placed across the voltmeter and the
potential measured, or the difference of potential may be compared
against the selected standard feeder, by means of a comparative
pressure indicator. The construction of this indicator is one having
a permanent magnet needle pivoted, and is actuated by one solenoid
across the standard feeder and one solenoid across the feeder pres-
sure, to which it is connected ; the effect on the magnetic system is
differential, and the potential read on the instrument is the differ-
ence of potential that exists at the terminal of the measured feeder
from that of the potential at the terminal of the standard feeder.

INDICATING WATTMETERS

In order to determine the power on any electrical circuit, it is
necessary to multiply the current flowing by the
volts pressure. An instrument which records this
product is shown in Fig. 58. The main current
flows around the moving system, and the potential
on this circuit flows through the moving system, and
the inter-reaction of these two forces is calibrated,
according to their mutual forces, directly into a scale FlG -

of watts which is the power element of the circuit being measured.

INDICATING WATTMETERS

This instrument has special uses in the low potential systems,
where the load curve in amperes does not indicate the power out-
put of the station; for as the current consumption rises, the external
distribution losses increase, which is compensated for by raising
the bus pressure, so that an indicating wattmeter is the only indi-
cation of output which, when plotted on a curve, is comparable to
the station's economical performance.

INTEGRATING WATTMETERS

In any variable load delivery, to multiply the current by the
volts for a limited number of observations during the day, in order
to determine the watt output, never leads to reliable results being
obtained, unless, of course, the plant happens to be operating under
constant load. For these measurements instruments of the inte-
grating type have been devised, which operate by the combined
effect of the current flow and potential difference ; their rotations
vary with the product of these two forces.

Their construction is, in all cases, quite similar to an ordinary
shunt motor, the conditions of field and armature being reversed.
The field is in series with the load; the armature, which is of high
resistance and generally of Siemens' construction, with a commutator
of a few segments, is placed "in shunt across the line," the shunt
being taken off beyond the field coils. An outside resistance is placed
in this shunt circuit, to reduce the current in the armature to the
necessary small quantity, and to prevent any appreciable waste
across the line. Also, a few shunt turns in some types act accumu-
latr .ly with the series turns, to compensate for friction of armature
rotation.

INTEGRATING WATTMETERS

Motor meters, as a class, rotate far more rapidly than is
allowable in practice. It has been necessary, therefore, to
introduce a "drag" or resistance to rotation, to slow the meter
to a reasonable speed. This has been done in several ways,
and perhaps the most common method is to attach a number of
air fans to the shaft. These are quite largely used, and with fair
success.

The resistance of an air fan to rotation is approximately pro-
portional to the square of the speed ; therefore, this device is only
fitted for combination with such meters as have a torque increasing
with the square of the current. However, since the torque of such
meters does not quite reach the square, the retarding effect
increases rather too rapidly, and has a tendency, although not
always pronounced, to cause the speed of the meter to fall off pro-
portionately on high loads.

Another method of "drag," which has been used with some
success, is the rotation of a small fan in a liquid a method
perhaps rather better than the previous one, since resistance to
rotation falls below the square of the speed when the liquid
itself begins to rotate. Much depends in this case upon the

shape of the receptacle containing the
fluid.

A third method consists in rotating a
small inefficient dynamo, generally a mere
disc turning between permanent or electro-
magnets. This resistance is of course
directly proportional to the speed, and there-
fore to the torque. Applying this "drag"
friction solves the principal difficulty to contend with.

INTEGRATING WATTMETERS

Fig. 59 shows the type of Thomson integrating wattmeter, which
is adapted for both alternating and direct watt indications, and
made especially for switchboard work. The main current enters
the meter through the "U' turn of copper, which produces the
field in which the armature rotates. The potential across the
armature varies with the voltage, and the damping effect is pro-
duced by a copper disc rotating in a permanent magnetic field.
The summation of the revolutions are recorded on a dial.

Fig. 60 shows an integrating meter which is specially used to
measure alternating current. This meter is essentially an induc-
tion meter. The entire current to be measured passes through the
primary, and an alternating field of force is developed
in the direction of the axis of that coil. At the same
time an alternating current is induced into the second-
ary, and this induced current develops another field of
force in the direction of the axis of the second coil, and
therefore at an angle to the primary, which produces a FlG 6o
resultant field, which is constantly shifting; this field acts on a
disc, and the rotations are calibrated to correspond with the cur-
rents flowing, and the indications on the dial will be practically
ampere hours. The retarding devices are air vanes.

DYNAMO GALVANOMETERS

In the control of dynamos, so that they may be thrown into
service together, a means must be provided so that it is known
when the potential of the generator is the same as that of the sys-
tem on which it is to be thrown.

In small switchboards, a voltmeter switch is provided so that it
can be connected across the terminals of the dynamo, and its

DYNAMO GALVANOMETERS

potential adjusted until it is the same as the system to which it is
to be connected. For this purpose, in the older central station
practice, dynamo galvanometers were used which were connected
across one pole of the dynamo switch, the other pole being closed ;
when the dynamo is at the same potential the galvanometer reads
zero and the machine can be thrown in on the circuit.

Another zero method which is used for this purpose consists of
a differential galvanometer, one winding of which is placed across
the bus potential and the other across the open dynamo switch.
When the potential on both windings is equal, the hand stands at
zero, and the potential of the machine is right for throwing in.

The incipient objection to a zero instrument for this purpose is
that it gives the same indications for no current as it does for the
proper adjustment of the generator; this has led to mistakes being
made by the switchboard attendants, which has resulted in these
methods being abandoned.

Now, each generator is provided with a small direct reading
voltmeter, which is calibrated in volts and is provided with a small
pilot switch, where there are several potentials used, so that the
voltmeter switch will indicate the proper positions in which to
throw the main dynamo switches when the dynamo potential is
equal to the bus potential.

DYNAMO REGULATORS

The regulation of dynamos requires a variation of the current
strength flowing through the field circuits, in order that they may
be thrown into service, and also that the loads may be properly
divided between the generator when working in multiple arc, and
the proper potential maintained on the consumption circuits. In

DYNAMO REGULATORS

shunt dynamos their output is varied by a resistance in series with
the shunt field, which resistance can be varied to increase or
decrease the current flowing through the field circuit. The
mechanical device used for varying this field circuit primarily con-
sists of a continuous resistance, from which taps are taken at
various points and brought to contact blocks, over which the con-
tact arm traverses, thus cutting out or in resistance in this circuit.

The ordinary form is shown in Fig. 61, which consists of a cir-
cle of segments ; the current is introduced through the pivot of the
lever, and thence to contact buttons, and through
the resistance to the other terminal of the regula-
tor. The fault with this device, as usually connec-
ted, is that if the lever contact leaves the contact
buttons the field circuit is open; if this happens
when the machine which is controlled is working
in multiple with others, serious damage may
result. Again, in large generators many thousand volts may be
induced in the field coils on the rupture of the field circuit, tending
to break down the field insulation. For this reason the permanent
connection should be made in all such regulators from the point
"A" to "B," so that the field circuit is always continuously
through the regulator, even if the lever is detached.

In any case where a field circuit of a generator over 50 K. W.
is broken, this should always be done slowly, and the arc drawn in
order that a partial circuit is maintained through the arc, so as to
prevent the discharge of the field from rising to abnormal potential.
In general, the field switches are provided with field circuit con-
tacts, and on withdrawing the switch it enters another set of con-
tacts, which throws a non-inductive resistance across the field

FIG. 61

DYNAMO REGULATORS

terminals, so that when the field circuit is broken the discharged
potential is suppressed.

The Siemens-Halske Company use a carbon auxiliary contact
device, which allows the field circuit to be gradually broken
between carbon points automatically, and suppresses in this way a
sudden surging of dangerous pressure from the field discharges.

There has been a tendency for small units especially, to make
the dynamo regulator compact. This, in itself, is a very good
feature; but with the same watts lost as the size decreases, the tem-
perature which the regulator attains, increases and uses the wire
nearer its fusing temperature, practically results in too many break-
downs.

A larger factor of safety should be made in dynamo regulators,
and much more than has been allowed in the past. A limiting
temperature of ninety degrees Centigrade for enamel or grid con-
struction, and fifty degrees Centigrade for open spiral coil construc-
tion, should not be exceeded. This latter form of construction has
been used largely in central station work, simply because the factor
of safety has not been sufficient in other types of regulators.

The cost of a regulator does not exceed three per cent, of the
cost of the generator it controls in units, of the size of generator
used in central station work. To increase the factor of safety of
the regulator strengthens the weakest part of the regulating system
and greatly insures reliability of service.

In regard to the total resistance of a regulator, this should be
such that the dynamos' potential can be reduced below the normal
at no load, when separately excited at normal potential. This is
an important point, for a number of manufacturers obtain their
rheostat data from tests on generators which admit of their being

DYNAMO REGULATORS

self-excited, and with this connection it obviously takes much less
resistance to control the generator than when separately excited.
In units from one hundred kilowatts up, the field currents become
of considerable magnitude, and the regulators to control them have

to be of special construction in order to properly
carry and control these currents. The contacts
and the connections are larger, and generally
switchboards employing units of this size neces-
sitate compactness; consequently the dial form
of regulator is not suitable.

Fig. 62 shows a regulator in which the con-
tact clips are arranged in two parallel staggered
steps with a sliding contact bridging them,
which contact is moved by a pinion in the
contact head meshing into a rack on the side
of the regulator.
There is also another form, in which the steps are arranged
around in two parallel quadrants, and so staggered that eighty
changes of resistance can be effected by moving the
regulator arm through forty degrees. Both of the
above regulators are connected to their resistance in
what is known as a loop connection, which, in effect,
is the same as drawing a contact along two legs of a
U-shaped resistance, and in both regulators the
moving of the contact upward increases the potential
or load on the machine.

In the quadrant type of regulator, it is also neces-
sary to open the field circuit, and to prevent this being opened
accidentally when the machine is in operation, an interlocking

FIG. 62

FIG. 63

DYNAMO REGULATORS

device is used, which prevents the field switch being withdrawn or
thrown in unless the regulating lever is at the lowest point (see
Fig. 63).

AUTOMATIC FIELD REGULATORS

Where dynamos are driven at a variable speed, the shunt field
has to be brought up for a drop in speed, and down for a rise
in speed.

There are several automatic regulators which can assist regula-
tion considerably, if the variations are not too sudden. They
employ an automatic relay which is operated by a given change in
voltage, which, in turn, closes a circuit through a solenoid or motor,
which actuates the contact lever to throw in or take out resistance ;
in this way compensation is made for speed variations. The con-
tact lever is usually provided with a dash-pot or other equivalent
damping device to prevent hunting of the regulator and surging
in the electro-motive force of the generator controlled. Compound-
ing will not take care of speed variations in a dynamo.

PROTECTIVE DEVICES

ROTECTION from abnormal flows of cur-
rent, in a circuit which has a limited current-
carrying capacity, is necessary for the safety
of the system and the structure which con-
tains it.

In any electrical distribution, there is a
fixed consumption circuit, and the conductors
are proportioned to the current supplied, and

will take care of all normal demands with a predetermined loss.
However, as the conductor system depends upon insulation to keep
the current in the conductors themselves, and to limit the current
in quantity by the consumption devices in that circuit, if the insu-
lation fails or the current demand becomes excessive, this circuit
then should be automatically severed.

The proper method to be employed for the protection of such
circuits depends on the potential of the circuit, the current supply
methods, and the character of apparatus to be protected, and the
different devices which are used for the purpose of protecting cir-
cuits each have their own special sphere of usefulness.

FUSES

Much has been written on this matter, and there is hardly
another "example in electrical engineering where so much has been
done in the laboratory without comprehending the true practical
function of the device and its inherent limitations.

FUSES

The rating of the fuse is the determination of six variables :
its specific resistance, its temperature coefficient, the cooling effect
of the terminals, the conditions for dissipating heat by convection
and radiation, the specific heat of the metal, and the latent heat
required by the fuse metal for its volatilization; the two last
have more direct bearing on the time element of the fuse than on
its rating. In use, the aging effects on the fuse are found to be the
oxidization due to the elevated temperature at which it is used,
and also its molecular stability changing under the variations of
temperature.

These variables have all to be considered in the rating of a fuse ;
and to make a fuse for a specific number of amperes, in order that
it can be connected with any kind of terminals, be blown in any
position and at any external temperature or condition of moisture,
has obviously led to the present doubt as to the accuracy of the
fuse. In fact, any fuse can be so arranged that by conditions
external to the fuse itself, it can be made to carry continuously a
current flow one hundred per cent, above its rating.

The usefulness of the fuse as an automatic device is only real-
ized when adapted to that duty for which it was originally intended,
namely, to protect the conductor system from injury. The fuse
being made the weakest point in the circuit, it should have such
reliability as to not allow such an excessive current to flow through
the conductor, so as not to create a fire hazard or injure the insula-
tion of the conductor. It would be much more practical to num-
ber fuses to correspond with the size of wire which they are
designed to protect, and be of such rating that they will blow
between those currents above which the wire is allowed normally
to carry, and below that current which will injure the conductor or

FUSES

its insulation, and there is plenty of margin between these two
current capacities to allow for all the variables which will alter the
current-carrying capacity of the fuse.

To send out a fuse marked to blow at a given number of
amperes, or even fuse wire sent on spools, rated in amperes has led
to the belief that they could be accurately calibrated ; but without
fixing the conditions of the length and terminals and holder, they
in practice do not give reliable service. It is a physical impossi-
bility to be sure that results can be reached in practice between
one hundred per cent, of each other; yet to protect the conductors
from excessive heating, due to abnormal current flow, is within
their legitimate use ; but to suppose that by overloading a fuse ten
or twenty per cent, that it will blow, is not within the original
intention or the present possibilities of this device.

Determining the sphere of usefulness of the fuse is a simple
question of cause and effect. On circuits of high or low potential,
not carrying currents over thirty amperes in most instances, an
effective service can be rendered by the fuse, and it will operate
quickly enough to fulfil the ordinary demand of a circuit- rupturing
device. At these capacities the resistance of the fuse can be con-
siderable without undue loss, as the heating of the fuse increases as
the square of the current flowing multiplied by the resistance of
the fuse ; also the thermal conduction of the heat and the areas of
radiation are low, and the mass of metal to be volatilized small ; so
fuses up to these capacities can be given a fairly reliable rating, and
are in some measure independent of the variables, which assume
large proportions in fuses at higher current-carrying capacities.

On low potential systems fuses are used extensively for the
reason that they can be of such capacity that they will not operate

FUSES

unless there is an extraordinary demand on the system, for the
current in these systems rises to an abnormal amount when the
conductors are either grounded or crossed ; but their use is being
abandoned in the large network of underground mains from cen-
tral lighting stations and copper strips substituted. It is consid-
ered better practice to-day to maintain current when a short-circuit
occurs on these mains until it is burned out, rather than hazard the
continuity of the service due to fuses blowing.

Fuses should not be considered as a reliable protective device,
as it is a very important fact that, when the time arises for their
action, it should be prompt; for it is evident that, by continuing
the condition of short-circuit, excessive strains may be brought on
the current-generating machinery and engine, as this strain has to
be maintained long enough to raise the temperature of the fuse to
the melting point and then supply sufficient energy to volatilize
the fuse before the circuit is ruptured. Modern practice indicates
that more than the external circuit itself has to be considered, as
the reaction on the generating system, which is supplying the ser-
vice, becomes a very important factor. In compound generators,
where the effect of sudden rise in current demand on the gene-
rators is accumulative, it makes it very necessary that the cur-
rent be severed quickly, as soon as it has reached a fixed value.

In modern central station practice the promptness of action
of an automatic circuit -rupturing device becomes its most
important function, where protection is to be afforded to the
generating apparatus. The distributing system, however, can
be subjected to a much greater variation in current flow, and
stand the strain for a longer period; so that in this class
of protection the time element of circuit-breaking is not so

FUSES

important. The fuse generally consists of a metal having a low
fusing point, the combination of tin, lead and bismuth being those
usually employed in the alloy. These are soldered to hard end
terminals, usually of copper, the terminals having ears provided so
that they may be clamped to the contact blocks. Fuses are also
made for high potential service of a number of strands of fuse wire,
each enclosed with a rubber tube. Also copper is used in the form
of wire, but it has a very large time constant, and has to be
maintained at considerable temperature if operated near its rated
capacity. Copper fuses have also been used, wound in the form of
a spiral, so that in rupturing the circuit the arc is drawn in a mag-
netic field, due to the convolutions of the fuse, and in this way
extinguished. In order to reduce the radiating effect, fuses are
enclosed in glass tubes, and in some cases are surrounded by a
chemical which combines with the fuse metal itself on reaching a
fixed temperature, and in this way reducing the resultant arc on
the breaking of the circuit. In this same class are also fulminate
fuses, which have, in close proximity to the fuse, a fulminate which
explodes when the fuse reaches a certain temperature, and in this
way ruptures the circuit.

AUTOMATIC CIRCUIT BREAKERS

A great many methods and devices have been employed for the
purpose of mechanically rupturing a circuit by its own effect, after
it has reached a predetermined point. The magnetic effect of the
current itself is generally used to actuate an armature, which in
turn releases the switching device. Expansion of conductors has
also been used to disconnect the circuit, but the time constant of

AUTOMATIC CIRCUIT BREAKERS

any heating effect, caused by the current itself, is far too slow to
be effective, as is shown in the case of fuses.

In order to accomplish the result desired in a circuit breaker,
the time between which the abnormal current rises in the cir-
cuit to be controlled and the actual opening of the circuit
should be as short as possible; in other words, the time of open-
ing should become less and less in proportion as the strength of
flow increases, for the reason that the circuit which it controls
may be short-circuited, and this current can quickly assume
large proportions. It is important that any circuit-rupturing
device should act as quickly as possible after the current has
commenced to rise above the set value, so that the rupturing
device can break the circuit before the current flow has attained
such volume as would be destructive to any rupturing contact;
and the quicker the circuit breaker acts, the greater the
advantage to both the generating apparatus and the distribut-
ing system.

The actual breaking of the circuit should be performed posi-
tively and with certainty ; yet to sever a high potential would
necessitate drawing an arc which would seriously interfere with
the proper action of the contact surfaces. Provision must be made
so that this arc will occur where it cannot damage the current-
carrying parts of the breaking mechanism.

THE CUTTER CIRCUIT BREAKER

This circuit breaker is the successful issue of many years'
extensive research into the problem of automatically severing of

THE CUTTER CIRCUIT BREAKER

electric circuits, and the perfected device as we have it to-day is
the outcome of tests made under practical conditions, rather than
the product of laboratory experiment.

Plate "C' shows the construction of this circuit breaker in
side view and in part section. The main current circulates around
the solonoidal coil "B" and tends to draw into the solenoid the
movable plunger "C." The initial position of this plunger in the
solonoid is determined by the adjusting screw "M." When the
current is sufficient to overcome the weight of the plunger it is
drawn into the coil with constantly increasing velocity, due to
intensified magnetic action, as the polar distance or air space is
decreased. When nearing the upward limit of its travel, having
acquired a high momentum, it impinges upon the trigger U N'
through the medium of the push pin " E." The immediate result
of this is the release of the switch arm by the displacement of the
retaining catch " F." The upper projection "H" of the trigger
u N ' is thrust against the striker plate " K," thereby utilizing
the energy of the current to start the movement of the switch arm.
This movement is intensified and sustained beyond the point of
final rupture between the switch contacts by the thrust of the
spring "O," which is released from compression by the initial
action of the trigger. Thus the contact arm is thrown away from
the contact terminal, and the circuit is opened.

Plate " D " shows the disposition of the main contacts and of
the auxiliary carbon contacts through which the current flows after
the copper switch plates have been severed. Upon these carbon
surfaces the current is finally ruptured. By this arrangement the
metallic contacts are preserved from the deleterious effects of an
arc, the circuit being finally ruptured between the auxiliary carbon

-U I-T-E

Circuit Breakers

CUTTER ELECTRICAL
AND MFG.

PLATE C

HJ2 SANSOM ST.
PHILADELPHIA, U.S.A.

THE CUTTER CIRCUIT BREAKER

contacts. The efficiency of carbon for a final break is due to the
fact that the vapor resultant upon the formation of an arc has a
high resistance, and, owing to the refractory nature of this sub-
stance, but a relatively small volume is volatilized by the action of
the arc, which, in addition, introduces into the partially severed cir-
cuit a counter electro-
O) motive force, approxi-

mating forty volts.

Among the impor-
tant results attained
by the development
of the Cutter circuit
breaker are : First,
the great reduction in
the time elapsing
between the occur-
rence of the excessive
current and its final
interruption. In the
event of a fault or
derangement in the distributing system the current tends to rise
with extreme rapidity; the prompt interruption of the circuit,
breaking the current before it has attained its full abnormal value,
thus becomes a matter of the greatest importance. Some recent
tests of these instruments show that the circuit breaker responded
wathin five one-hundredths of a second after the occurrence of a
short-circuit, and the current which would normally reach two
hundred amperes was severed before eighty amperes flowed through
the circuit breaker, showing that with normal inductance of gener-

PLATE D

PLATE E
I-T-E CIRCUIT BREAKER

STANDARD SWITCHBOARD TVPE. SINGLE POLE. 5 TO 200 AMPERES

THE CUTTER CIRCUIT BREAKER

ating apparatus the circuit breaker acted more quickly than the
generators could respond to the demand ; consequently, abundant
protection was afforded.

The responsiveness of the device upon the occurrence of a
gradual overload is no less marked. Friction and the conforma-

tion of the magnetic circuit, if not properly taken care of in the
design of a circuit breaker, will allow the current to creep
beyond the point at which the device is adjusted to operate before
the force of the solenoid is sufficient to actuate the plunger. In
this instrument the plunger is free to move without appreciable
friction, and, being worked far below the point of its magnetic
saturation, it is extremely sensitive to current changes. This,
in combination with the structural features alluded to, insures
a positiveness of action which altogether precludes the possi-
bility of "floating," as it has been termed. It may be hardly
necessary to add that the circuit breaker is a very efficient pro-
tection against damage by lightning.

Plate "E" shows the standard switchboard type of the Cutter
circuit breaker, of from five to two hundred amperes capacity, in
an open position; Plate U F," a larger instrument of the same type,
of from two hundred to twelve hundred and fifty amperes capacity,
closed, while Plate U G V shows a still larger circuit breaker of
standard switchboard type, single pole, of from fifteen hundred
to three thousand amperes. All of these are intended for direct
current only and are single pole, for use on circuits of six hundred
volts or less, the long, clear break making them peculiarly adapted
to the severe conditions incident upon street railway work. They
are equally suited for lighting and power circuits, up to and
including six hundred volts.

PLATE F

I-T-E CIRCUIT BREAKER
STANDARD SWITCHBOARD TYPE. SINGLE POLE. 200 TO 1,250 AMPERES

PLATE G

I-T-E CIRCUIT BREAKER

STANDARD SWITCHBOARD TYPE. SINGLE POLE. 1,500 TO 3,000 AMPERES

THE CUTTER CIRCUIT BREAKER

Plate "H v is a double pole circuit breaker, of a capacity of
from thirty to two hundred amperes, for voltages of two hundred
and fifty or less. As illustrated, it is intended for front connec-
tions, making it suitable for panel board work. For use on
switchboards it would be used with back connections.

Plate "!'' is a larger instrument of the same type, having a
capacity of from two hundred to fifteen hundred amperes, double
pole, for use upon circuits having a voltage of two hundred and
fifty or less, equally adapted for lighting or power.

Plate "J v represents a circuit breaker of from two hundred to
six hundred amperes capacity, double pole, double coil. Not only
is this instrument designed to open both sides of the circuit, but,
having two coils, it will be operated upon the occurrence of an
overload upon either side as well as by an overload affecting both
sides of the line, thereby insuring absolute protection of the circuit
under any and all conditions.

Plate U K" represents a type of instrument especially designed
to meet the severe condition of opening an alternating current
circuit of two thousand volts or less. It is made in single pole
only, and has a clean, wide double break of ten inches. In this
instrument the coils of the smaller sizes are wound with covered
magnet wire, and in the larger sizes the coils are made of open,
bare rectangular copper. This type of instrument is made up to
a capacity of two hundred amperes.

We have in the foregoing treated of the Cutter circuit breaker
of an overload type only. With but a small increase in the size,
and without in any way affecting the simplicity of the overload
instrument, an underload function may be added. A principle
analogous to that which insures certainty of action in the overload

PLATE H

I-T-E CIRCUIT BREAKER
MIDGET SR. TYPE. DOUBLE POLE. 30 TO 200 AMPERES

PLATK I

I-T-E CIRCUIT BREAKER
STANDARD SWITCHBOARD TVPK. DOUBLE POLE. 250 TO 1,500 AMPERES

PLATE J

I-T-E CIRCUIT BREAKER
500 AMPERE. DOUBLE POLE. DOUBLE COIL

PLATE K

I-T-E CIRCUIT BREAKER

ALTERNATING CURRENT TYPE FOR HIGH VOLTAGE. SINGLE POLE ONLY

5 TO 200 AMPERES

PLATE L
I-T-E CIRCUIT BREAKER

DIRECT CURRENT, no, 220 AND 500 VOLTS. 5 TO 25 AMPERES

PLATE M
I-T-E CIRCUIT BREAKER

DIRECT CURRENT, EITHER SINGLE OR DOUBLE POLE. 4,000 TO 8,000 AMPERES

THE CUTTER CIRCUIT BREAKER

is made use of in the underload, the actuation of the trigger upon
the occurrence of the underload being effected by the blow of an
armature moving under spring pressure. It will be seen that
whether the cause of operation be an underload, a "sneak current,"
or a heavy short-circuit, the trigger will be acted upon with a ham-
mer blow, and never, in any case, without a free preliminary
movement of the actuating body.

The underload device is made in two forms, one of which,
operating only upon the interruption of the current supply, is
especially suited for motor protection, while the second form,
which operates upon the occurrence of a predetermined minimum
flow, is peculiarly adapted for use in connection with storage
batteries.

Plate "L" shows an overload circuit breaker having the under-
load function added. The type shown is intended for direct cur-
rents of from five to twenty-five amperes, single pole. This form
is regarded as the best for storage battery protection.

Plate U M." This type of instrument is specially designed for
circuits of very large capacity, either single or double pole. The
construction is such that the current is carried through laminated
contacts in series with the actuating coil of the instrument. The
main break is between the laminated contacts, and is followed almost
simultaneously by the auxiliary break, which is in shunt with the
main contacts, and is protected by a final carbon break of ample
capacity. All the features which have made the smaller types so
succesvsful are preserved in this device, while the arrangement of
the laminated contacts and the general design are so accurately
proportioned that the circuit breaker can be set with no greater
effort than is required in closing a circuit breaker of fifty amperes.

PLATE N

I-T-E CIRCUIT BREAKER
OVERLOAD AND "No VOLTAGE." 5 TO 25 AMPERES

THE CUTTER CIRCUIT BREAKER

Plate "N' shows an overload circuit breaker having a "no
voltage v function added. This type is intended for direct current
of from five to twenty-five amperes, and is regarded as best for
motor protection. They are made for one hundred and ten, two
hundred and twenty, and also five hundred volts.

Plate "O." This instrument has been specially designed for
use in connection with Edison three-wire currents, also for three-
phase alternating current power circuits. This type of circuit
breaker is made of a capacity from five to two hundred amperes,
for use on circuits of two hundred and fifty volts or under.

I T-H CIRCUIT HKEAKKR
FOR USE ON CARS

PLATE O

I-T-E CIRCUIT BREAKER
STANDARD SWITCHBOARD TYPE. TRIPLE POLE.

LIGHTNING ARRESTERS

IGHTNING arresters, in their function, bear
the same relation to the potential stress of the
station as the circuit breaker bears to the cur-
rent demand, and their purpose is to offer to a
high potential discharge a path to ground where
it can pass off, rather than enter the station or
the protected consumption devices, and break
down their insulation in the effort of this dis-
charge reaching the earth.

Lightning discharges have peculiar characteristics which make
it comparatively easy to divert them. A discharge of lightning is
supposed to be a rush of high potential which possesses an enor-
mous frequency, and consequently inductive circuits arrest its flow;
in this way discharges can be damped back from entering the
station, and there will be an accumulation of potential adjacent to
the inductive portion of this circuit. If an air gap be connected
here, and one side connected to the charged line, and the other side
of the gap connected to the ground, the discharge will jump this
gap and pass to ground and be equalized, rather than force its way
through the throttling inductive circuit. In order to accomplish
this result, the conductor is wound into a number of convolutions,
not over twenty, on a four-inch diameter mandrel, and these turns
are interposed between the lightning arrester and the apparatus to
be protected, preferably very near the lightning arrester itself.
This protects the station from the discharge ; but when the dis-

LIGHTNING ARRESTERS

charge passes the air gap, it so reduces the resistance of this gap,
that when an active circuit is thus protected, the main current
follows to ground and maintains an arc over the gap. For the pur-
pose of again rupturing this circuit, several devices are used.

THE THOMSON-HOUSTON LIGHTNING ARRESTER

Fig. 78 shows type of magnetic lightning arrester. Under nor-
mal conditions the current from the generator passes through the
electro-magnetic windings to the right hand wing of the arrester,
and then to line, the left hand wing being connected to earth. The
result of the current flowing through this electro-magnet produces

a strong magnetic effect at the ends of the mag-
net, which is projected across the air gap ; this
forces the arc away along the curving wing
edges, until it becomes too attenuated to be
maintained by the machine potential. Practi-
cally this occurs instantly, and the arrester is
then ready for another lightning stroke. Each
arrester protects its own side of the line, and
FlG 78 therefore two are required for each circuit. The

illustration shows the form of arrester usually used on switchboards
for central station protection.

WESTINGHOUSE TANK ARRESTER

The action of this arrester is to maintain an artificial ground of
fairly high resistance on the lines to be protected, only when they
are hazarded, and this type of leak arrester has been developed in
order that there be no actual severance of the ground connection
from the circuit to be protected ; for in an arrester possessing an air

WESTINGHOUSK TANK ARRESTER

+ BU53

gap, an abnormal potential must exist before the device operates.
Here the systems to be protected are connected to a leak to ground
during times of danger, and has electrically the effect of bringing
the line to the level of the earth ; for through this leak both line

and earth are main-
tained at the same
potential, and dis-
charges pass off to
ground through this
leak. An inductive
circuit is usually
interposed between
the apparatus to be
protected and the
external line and
leak, in order to
force this discharge
FIG. 79 to ground.

A leak arrester

allows the surging of induced potential in the line protected, due
to the inductive effects of charged clouds over the line to follow
the same potential values as the earth over which they are strung.
The type of leak arrester is only suitable to protect circuits where
one side of the circuit only is to be protected. Fig. 79 shows this
type of arrester, and the tank leak is plugged when the system is
threatened. There are generally three tanks employed, having
between them inductive circuits, "A," " B," and " C," so that
if the discharge passes one, it is restrained still further by the
other two.

WESTINGHOUSE TANK ARRESTER

There are a great number of types of arresters using fuses with
air gaps, the fuse being severed by the current following the
discharge ; but as discharges follow each other rapidly, continuous
protection is desirable.

There are also condenser arresters in which the lightning dis-
charge passes from one plate to the other ; but these plates being
in series, the initial potential of the circuit cannot maintain a dis-
continuous arc.

THE NON-ARCING LIGHTNING ARRESTER

It has been discovered that with certain metals, when they form
the surfaces for a spark gap, an alternating current will not main-
tain an arc between these surfaces, due to the
cooling effect of the terminals and the character
of the vapor of the metal in
the arc. Fig. 80 shows the
type of the Westinghouse
Non-Arcing Arrester, which
is constructed embodying
this principle, and which

FIG. 80

consists of a number of cylinders of non-arc-
ing metal, forming several gaps between them,
which are jumped by the high potential dis-
charge ; the alternating current does not follow
and maintain an arc across these spaces or
gaps, and this form of device is largely used
in alternating current switchboards, where it
is considered policy to place the arrester on the switchboard
itself.

FIG. Si

THE NON-ARCING LIGHTNING ARRESTER

Yet another form of arrester is shown. In Fig. 81, in this case,
the current, in its passage to the ground after following the dis-
charge through the gaps, actuates an electro-magnet which pulls
an arm, and lengthens the air gap until the arc is extinguished.

I860

THE LOW TENSION SWITCHBOARD

F the elements of design, common to both the
isolated plant and the large low tension cen-
tral station switchboard, both can be treated
together. Taking the simpler forms first, and
afterwards considering the special conditions
necessary to be met by the switchboard, when
used to control the more intricate methods of
central station operation and distribution-
methods recently introduced to give the proper
potential supply over the expanding areas of the external network
of conductors, and also for working of the storage battery in
connection with the generators of the station.

The isolated plant usually consists of several generators and a
number of feeders to centres of distribution. The actual assembly
of the apparatus on the switchboard, and its relative position, is so
largely determined by the local conditions to be met, and the con-
venience in handling, that to illustrate the different methods of
assembly would not be of much value, but the method of connect-
ing the different appliances is common to all the systems.

Diagram i gives the connections where a shunt dynamo is oper-
ated on a simple two-wire system. In this case the two terminals
of the dynamo are brought to the switchboard through the dynamo
leads to a double-pole switch.

The connections are usually made to the switch, so that the
dynamo feeds through the fuses and then through the switch

THE LOW TENSION SWITCHBOARD

mechanism to the bus bar, so that if the switch is pulled, it can be
fused when there is no current on the switch. Circuit breakers
are usually inserted in the dynamo lead before it is tapped on the
dynamo switch ; the main current is here taken, one leg to the bus

bar, and the other leg to the
shunt dynamo amperemeter
"A," and then to the other
bus bar. The dynamo regula-
tor "R" is connected in series
with the shunt field, the field
wire being brought from the
dynamo for this purpose; one
end of the field being connected
to one lead at the dynamo, and
the other end connected to the
other dynamo lead at the switch-
board. The voltmeter "V" is
connected across the bus bars,
and the ground detector
"GTL," in one-hundred-and-
F d S H ten-volt system, consists of two
lamps in series across the bus
Diagram No. 1. bar > a middle connection being

taken between the two lamps to
ground. Generally the ground
connection is provided with a plug " P," so that the system is only
grounded when the test is being made.

In connecting two or more dynamos that are to be worked in
multiple, provision has to be made so that these dynamos can be

THE LOW TENSION SWITCHBOARD

thrown together without any fluctuation to the potential of the
system.

Diagram 2, Fig. i, shows a method where a dynamo galvano-
meter is used for this purpose, and is so connected that the dynamo
feeds into the bus bar one side direct, and the other side through
the dynamo galvanometer, which usually bridges the open dynamo
switch, two single pole switches being used when this method is
employed for throwing in dynamos.

It is evident that when no current flows, the potential on the
dynamo and the bus bar must be the same, when the dynamo can
be thrown on the system. Fig. 2 shows another zero method
where the potential for the generator and the potential on the
buses both act on the same magnetic system differentially; when
the currents through them both are equal, the needle of the differ-
ential galvanometer " D G ' will point to zero, and the dynamo is
ready to be thrown in.

Both of these zero methods possess the inherent objection that
they give the same indication when there is no current flowing
through the instrument as when they are ready to be thrown in,
and mistakes have arisen from this cause, and have led to volt-
meters being used for this purpose. Each generator is supplied
with a pair of contact buttons of a voltmeter switch "V S,"
which are connected direct to the two leads of the dynamo, and
the voltmeter can be connected directly to any machine and its
potential raised until it is the same as the bus, when it can be
thrown in.

In compound generators, where more latitude can be allowed in
the potentials of the generators being thrown in, pilot lamps "PL"
may be sufficient to give the proper indication when they are con-

No.

sl s.

G L

o o o
I

F.o3.

/ww

DG.

?v.s.

THE LOW TENSION SWITCHBOARD

nected across the dynamo leads. These two last connections are
shown in Diagram 2, Fig. 3.

The above connections show those used for shunt generators,
where the rheostat is the only method of regulation ; but isolated
plants are usually provided with compound wound machines, and
provision has to be made on the switchboard for equalizing connec-
tions, as it is the usual practice in isolated plants to have the equal-
izing bus on the switchboard.

The proper connection of the equalizer plays an important part
in compound dynamo regulation, and the resistance of the leads
from the dynamos to the switchboard, both equalizing and series
leads, should be calculated as follows :

In all generators of the same size and having the same com-
pounding characteristics, the resistance from the equalizing bus to
the bus connected to the series side would be equal for all genera-
tors. Where these generators are different sizes, the drop of
potential on the equalizer and series leads should be the same for
all compound generators, when these generators are working
together and carrying their full load. Where the dynamos are not
of the same type nor the same compounding characteristics, the
resistance of the equalizing circuit will have to be such that the
drop from all generators will be the same when they take their
maximum load together; but it is hardly possible to compound
dynamos of very different characteristics together, so that they will
pick up their proportional load equally among themselves, and
only the best approximate condition can be obtained, when they
will take their full load together.

If it is necessary to operate compound generators on two or
more potentials, each potential must have its independent equalizing

THE LOW TENSION SWITCHBOARD

bus, so that the equalizing leads can be shifted to the equalizing bus
to correspond with the potential bus, on which that dynamo is to be
operated, in order that the proper regulation can be effected by the
series field. Diagram 3 gives the connections for two compound
generators connec-

L <flGTL(

ted in multiple.
This diagram shows
a three-pole switch
used for this pur-
pose. The third or
middle clip should
be made longer, so
that the equalizing
connection can be
established before
the generator is
thrown in on the
system. An equal-
izing bus is provided
in this diagram,
which is the usual
practice, where
there are a number
of machines work-
ing together.

Another method, which is sometimes used, is to connect the
equalizer and also the dynamo lead from the equalizer side of the
generator to a double pole switch. This can be thrown in and the
dynamo brought up to potential, and a single pole switch can be

TTeecb?v 5wi+cl\e3.

No3.

THE LOW TENSION SWITCHBOARD

closed on the open side of the generator to throw the machine in
on the system. This method of connection has its advantage : if
there is much distribution drop, and if the generators are over-
compounded to take care of it, and a number of machines work
together in multiple to supply the full load when this maximum
drop occurs, it is evident that when one machine is operated to
supply the minimum load, the generator will normally give too
high a potential, and hand regulation will be necessary in order to
obtain the proper potential ; whereas, if the series and compound
connection were left in circuit of the idle machines, they would
reduce the normal over-compounding of the active machine as the
load on the plant decreased. In this way better regulation can be
produced and the over-compounding of the machine is somewhat
controlled. This also avoids ever putting in a compound machine
in circuit without first equalizing.

Another method is to have the voltmeter by which the genera-
tors are brought up to potential, not connected across the machine
until the equalizer switch is thrown, and in this way avoid putting
in the generator without first equalizing.

The circuit breaker should always be connected in the lead on
the opposite side of the dynamo from the series winding, for it
would require a double pole circuit breaker to open the circuit on
the compounding side of the machine. Diagram 4 shows the con-
nection for a three-wire two-dynamo system, which, in effect, is two
dynamos in series, the positive of one brush being connected to the
negative of the next, which forms the neutral connection. Besides
the apparatus required on the two-wire switchboard, the three-wire
system requires two voltmeters, one on each side of the system, and
the three-wire ground detector should have two lamps in series

THE LOW TENSION SWITCHBOARD

between the positive and negative bus to ground, and one lamp
between neutral bus to ground; the ground detector should be
provided with a double throw switch, so that the positive or nega-

GTL

TOT

tive side of the system can be tested with the same detector, three-
pole switches being used for all feeders.

Sometimes the neutral connections are made between the
dynamos themselves, and the neutral bus is only used behind
the feeder board and connected to the common dynamo neutral in

THE LOW TENSION SWITCHBOARD

the dynamo room. This saves nearly one-third of the copper in
leads over that required to bring every terminal of the dynamo to
the switchboard.

Where dynamos are to be used on either side of the system, they
have to be provided with a double throw switch, so that the)'
can be connected on either side, and the connections must be
made so that the polarities will be right for either position of the
switch.

Diagram 5 shows the connection for two sets of compound
dynamos connected to a three-wire system. Here the only depart-
ure from Diagram 4 is that an equalizer has been added, one for
each side of the system, and of course if these machines are to be
used on either side of the system, a double throw three-pole switch
would be required, so that the dynamos on one side of the system
always equalize on the same bus.

There are a number of methods used where, from a single
dynamo of two hundred and twenty volts, a three-wire system is
operated. These systems require an auxiliary device which would
equalize the loads between the two sides of the system. It is evi-
dent that if two dynamos were connected together in series across a
22o-volt two-wire system, and the neutral taken from the common
connection of these two dynamos, and if an unbalanced load were
operated, the dynamo on the highly loaded side of the system would
tend to operate as a dynamo and pump current into that side of the
system, and balance the system external to the generator itself.
The switchboard for this system, as far as the generators go, is sim-
ply a two-wire switchboard, but the equalizer or compensator is con-
nected across from the positive to the negative, and the neutral taken
from the common junction of these two to the feeder switches only.

-HIU

-HU

THE LOW TENSION SWITCHBOARD

Diagram No B A .

A storage battery has also been proposed for affecting an equal-
ization of the load between the two sides of the three-wire system,
but the connections in this
case are the same as an
equalizer. A five-wire sys-
tem is usually connected as a
three-wire system, except in
having four dynamos in series with connection between each gen-
erator, and the dynamos connected between the different buses at
the back of the board. The connections for the above methods are

shown in Diagram 6A.

It is often important,
in isolated plant work,
that the motors, espe-
cially if elevator motors be
used during light loads,
be operated on indepen-
dent generators; these
loads fluctuate violently
and the generators can-
not regulate quickly
enough in order that the
constant potential be kept
on the lighting system.

The switchboard is
separated up into two bus
systems lighting and
power and each genera-
te 82 tor is provided with a

100

THE LOW TENSION SWITCHBOARD

double throw switch, operating on power in one position and light-
ing in another.

In large apartment houses, public buildings and asylums, it is
also advisable to separate the public lighting from the private light-
ing, and supply it by the same or different generators. This is
taken care of by providing the generators with double throw
switches, and also supplying the feeders with double throw switches,
if they are to be fed from different sources of supply.

Some methods of making back connections for isolated plant
switchboards are shown in Fig. 82.

101

LOW TENSION CENTRAL STATION SWITCHBOARDS

EFINITE control of potential over a large net-
work of low tension conductors has required
a variety of methods in the handling of these
feeders at the switchboard, in order to pro-
duce this variation of potential economically.
The first means resorted to in order to
create this difference of potential on the
feeder terminals was to insert an adjustable
resistance to compensate for the unequal losses occurring in the
feeders, so that the current would be delivered to the mains at a
uniform potential throughout the system. This method, however,
is uneconomical, both as regard the amount of energy consumed
for the purpose of regulation and the valuable space occupied by
this method.

Fig. B, of the introductory, shows this method with equalizers,
as installed in the old Adams Street Station of the Chicago Edison
Company. This method has now become practically obsolete ; but
as the low potential systems have been expanding over large areas
to include a greater number of customers, it has again become
necessary to deliver current at the station ends of these feeders at
different potentials, to compensate for the unequal loading of the
system. This is affected under some conditions by running the
dynamos at different potentials, which supply independent busses,
the feeders being so arranged that they can be thrown on any of
these busses and supplied with current at the proper potential to

102

LOW TENSION CENTRAL STATION SWITCHBOARDS

make good the feeder losses and deliver their current to the mains
at a uniform potential.

When the units in the station are small, they can be divided up
on the different busses with comparatively high economy ; but as
the central station business has increased, and the plants have
adopted large units, both on the score of economy for the first cost
and operation and to increase the kilowatt output to the square
foot of floor space, these practical conditions have greatly altered
modern central station switchboard design. The condition of oper-
ating the units at an economical load must be maintained, while
the feeder service requires several potentials to be delivered to it.

The booster system has been devised in order to change the
potential of the current delivered by the generators U D D" in
Diagram 6. In order to clearly understand the connections for the
system, the following explanation is made regarding its action in
the case of the three-potential system:

In this case, the medium potential bus is fed directly by the
units operating the station, and the current is delivered from the
medium bus to the high potential bus, through the armature of a
dynamo whose field is separately excited, the current capacity of
the armature being sufficient to carry the loads, at which it would
be uneconomical for one of the units to operate. The current, in
passing through this armature, has its initial potential raised, and
the amount of this increased potential is controlled by a field regu-
lator "R," in series with the dynamos' separately excited field; the
currents which supply the bus of lower potential than the medium
bus also flow through an armature, but in this case the initial
potential of the current is reduced before it is delivered to the
low potential bus. This dynamo is operating as a motor, and

10;

LOW TENSION CENTRAL STATION SWITCHBOARDS

the potential of the current passing through it can be regulated
by means of a field rheostat. Shunt wound boosters are usually
used in low potential work which is controlled by the shunt
field only.

In the three-wire system two boosters are required for each
potential, and it is the usual practice to couple these boosters
together, forming one continuous line of shafting, to which is also
coupled a motor for the purpose of keeping the boosters up to full
speed, and making good the losses due to transformation and the
unequal loading of the high and low busses.

Diagram 6 shows the method employed for connecting boosters
to a three-wire system.

It is evident that if one dynamo could be arranged to give
several potentials, the efficiency of output would be raised and
the losses inherent in the operation of the boosters would be saved.
There have been several methods proposed for this purpose, but
when a series wound armature rotates in a multipolar field, and a
load falls on one circuit supplied by a section of this armature, a
redistribution of magnetism will occur, caused by the unequal load-
ing of this symmetrical armature ; this will give an unequal poten-
tial delivery to the circuits from this armature, and a variation of
potential beyond the control of the regulator. If this condition is
avoided in the design of the machine, it becomes both expensive in
first cost and uneconomical in operation ; but with a symmetrical
external distribution system, with regard to the station and feeders
which are tapped radially into a concentric main, and if a load falls
on one part of this main, and that section of the multipolar gen-
erator has its field increased, the distribution of magnetism in the
concentric field will be identical with that of the distribution of

LOW TENSION CENTRAL STATION SWITCHBOARDS

current in the concentric main ; in this way a compounding effect
can be obtained external to the dynamo itself.

In certain cases, economy may be shown by inserting regulating
resistances between the high bus, and busses of lower potentials may
this way be obtained where the current demands on the busses of
lower potential are small, and the losses in these resistances are
less than that caused by a partly loaded engine and generator
working, on this reduced potential. This method also allows keep-
ing the generator efficiency high by working all the units under
the maximum possible loads for the different daily variable outputs
of the station; this result can be obtained by the proper switch-
board design, which will also effect a large operating economy.

In other methods of producing the several potentials required
for close regulation on the external distribution system, the storage
battery is one which is coming into use, and here the different
busses obtain their potentials from different cells of the same series
of storage batteries, and the potentials on these busses can be regu-
lated by the battery-regulating switches. This gives the only
method by which the regulation potential on the feeders can be
affected without changing the loading on the generator.

As the feeder has to be supplied with several potentials, switch-
ing arrangements have to be provided so that these feeders can be
connected at will to the various potential busses supplied by the
generating apparatus. Where only two potentials are required, a
.double throw switch is used and the feeder is brought to the centre
clip, when connection is made to either potential.

Where three potentials are used, double pole switches can also
be used by grouping the nearby or low resistance feeders, so that
they can be connected to common bus and low bus, and the outline

1 06

LOW TENSION CENTRAL STATION SWITCHBOARDS

or high resistance feeders, by means of a double throw switch, can
be connected to common and high bus; the dynamos would, of
course, have to have three-way switches in this case, so that they
could be operated on any bus, as required.

Where the distribution system requires considerable variation in
potential, a double throw switch could be used for a considerable
number of potentials for the different feeders, if they are properly
grouped, and if the drop is so slight at light loads that all feeders
can be supplied from the common bus. Under these conditions all
the busses can be tied together by tie switches, or all feeders can
be thrown on one bus. W T here the feeders
themselves have to be supplied with more
than two potentials during the load fluctua-
tions on the station, several forms of switches
have been devised for this purpose. The
simplest form is a double pole double throw
switch, having each blade on an indepen-
dent handle ; in this way a feeder can be supplied from any of four
potentials, and also two busses can be connected together by this
switch at low loads. See Fig. 83. The radial form of switch for

the same purpose has different bus bars con-
nected to clips and disposed concentrically
around the centre ; the switch plate is pivoted
so that it can be thrown in on any of these
busses when swung around in position. For
this construction, see Fig. 84.

Another form has been developed for this
purpose, where the switch blade is detachable, and can be engaged
with several terminals which align with the different bus terminals

107

LOW TENSION CENTRAL STATION SWITCHBOARDS

and the movable blade inserted ; in this way the feeder can be con-
nected to any bus. See Fig. 85.

In central station switchboards, there are several general
methods of distributing the controlling and regulating of apparatus.
One way is to concentrate all the regulators and field switches at
one point, and the dynamo switches, am-
meters and galvanometers on the dynamo
switchboards.

The most prevailing method is to
assemble the regulator, field switch,
dynamo galvanometer, dynamo switch
and amperemeter on the dynamo board,

all mounted on the same panel. The voltmeters are placed in
some conspicuous position with the pressure switches; by using
edgewise instruments, a density of four hundred kilowatts per run-
ning foot was obtained in the switchboard designed for The Chicago
Edison Company by the author. The character of bus bar connec-
tions required for the carrying of one hundred and eighty thousand
amperes from the lower dynamo board to the upper three-potential
feeder board, which was located above the dynamo board in the
Chicago Edison Company's station, is shown in Fig. 87.

The matter of field connections has to be carefully considered,
for with large low potential units, a high induced potential is
created when the field circuit is broken ; this will tend to break
down the insulation, and is the only way in which a hazardous
potential can be produced in a low potential system.

The method of connecting the field in any generator will
depend upon the system which is supplied by that unit. The self-
exciting method is when the field is connected directly across the

1 08

D
O

Q
W

O
O
<
U

V
K

as
W
K

ai
U
ca

LOW TENSION CENTRAL STATION SWITCHBOARDS'

terminals of the generator ; the rheostat is included in this circuit,
and when the armature is rotated and creates a potential, due to the
residual magnetism in the field, it sends a current circulating
through the field windings, and in this way builds up the genera-
tor's potential. The bus exciting method is where the field is con-
nected to the bus bars and excited by a potential external to its
own. This method has the advantage of bringing the machine
quickly up to the proper potential, and always of the right polarity;
but when the field is to be withdrawn from the bus bar, after the
machine is shut down, the field must first be short-circuited through
a non-inductive resistance, such as a lamp bank, before it is broken,
in order to reduce the potential of discharge.

The advantage of the self-exciting method is in the feature that,
when the dynamo is shut down, the field dies away with the fall of
potential on the generator ; but in large multipolar units, the poten-
tial may rise very slowly, and for this reason bus exciting is
largely used.

Separately exciting methods have no difference in their connec-
tion from the bus exciting, except a separate generator is used for
field exciting. A condition arises where it is very advisable to
have the field exciting of the generators so arranged that any of
the generators of the station can be used temporarily as a sepa-
rate exciter, when there occurs a very severe load or a short-circuit
on the external distribution system. It is necessary to provide this
method in order to hold up the potential of the generators working
on a short-circuit, as the reaction of the armature under these con-
ditions is so great that it kills the field, and the generator loses its
potential when the field is excited, and the distribution is supplied
from the same bus bar, and the station falls flat. Diagram 7 shows

LOW TENSION CENTRAL STATION SWITCHBOARDS

how by using a special field switch, all the fields can be excited
from any generator. This method is only advised to be used in the
case of emergency or short-circuit, and the proper connections can

be quickly made before the
potentials have fallen, by in-
serting a dynamo on the field
bus and withdrawing the
switch that connects the field
and main busses together.

Another method, known
as the Donshea method,
combines the good features
of both the bus exciting and
separately exciting methods. For connections see Diagram 8.
The field switch "A" interlocks with one leg to the dynamo
switch; to the other leg the field is permanently connected. If
dynamo switch "B" and field switch "A" are
thrown, the field of dynamo "B" is excited by the
pressure on the busses. The dynamo is thrown on
the system by closing dynamo switch " C." This
switch is so arranged that when the main blade is
withdrawn, it carries with it field switch " A,"
which is also electrically connected to it. When
these interlocking switches are withdrawn, the field
circuit is from the terminal of the dynamo, through
field resistance box, field switch and dynamo switch
blade, then back to the other brush of the dynamo. In this posi-
tion of switches the dynamo is self-excited and the field will die
away with the voltage on the armature.

III

LOW TENSION CENTRAL STATION SWITCHBOARDS

The Potter method, Diagram 9, is applicable to compound
generators, and by reason of the field being also excited by the
series winding when thrown across the equalizer and positive side
of the system, it will be in multiple with the series fields of the
other dynamos in operation, and the cur-
rent will be diverted to it, exciting the
fields. By the aid of this initial excitation,
the dynamo can be brought up readily to
the proper potential by adjusting the
shunt field, and when its proper potential
is reached, it can be thrown in multiple
with other generators.

By practising this method of exciting dynamos, compound
machines cannot be thrown in multiple without first connecting the
series field, if the positive pole and equalizer be connected each
to one leg, by double pole switches. For these connections, see
Diagram 9.

112

RAILWAY SWITCHBOARDS

NOWING the extreme conditions
that arise in railway practice, it
is required of the electrical en-
gineer, in designing the switch-
board for this service, to fully
protect the generating apparatus
from the shocks due to sudden
overloads, and an automatic
circuit breaker is in this case
a necessity. The current
passes through this circuit breaker, then through the ammeter
shunt or the ammeter system itself to the negative bus bar.

In railway practice, the positive side of the machine, where the
trolley is positive, is connected through the series winding. The
equalizing connection is taken to the middle point of the switch to
the equalizing bus, but the present practice in power stations is to
equalize at the dynamo, and the equalizing switch either mounted
on the frame of the dynamo itself or on a pedestal by the side of
the dynamo. In other cases again, the equalizer is tied together
permanently between all the dynamos. The disadvantage of hav-
ing the equalizer opened is that there is a danger of the machine
being thrown in circuit before it is equalized. In order to provide
against this accident, several suggestions have been made; one is
to make the switch at the dynamo double pole, through which are
carried both the equalizer and positive connections; by means of

RAILWAY SWITCHBOARDS

this connection, the generator cannot be thrown in from the gallery

or switchboard without having the equalizer thrown in first.

Another method has been proposed where the throttle of the

engine is connected to the equalizer switch, so that when it is open

it closes the equalizer
switch ; in this way the
generator cannot be
thrown in before it is
equalized.

The field of the
railway generator may
be connected up in two
ways; the one shown
in full lines is the bus-
excited method, and
the one shown in dot-
ted lines is the self-
excited method. A
dynamo galvanometer
or voltmeter is ar-
ranged across the dy-
namo terminals of the
dynamo switch, in
order to show when the
generator is of the

right potential to be connected in on the system. Diagram 10

shows these connections.

It is also usual to allow for a panel between the dynamo and

feeder panels, on which to mount the main ammeter, integrating

114

R A I L\V A Y S\YI TC H BO AR DS

wattmeter, voltmeters and pressure switches. The positive is only
taken to the feeder board, and the feeders are provided with a single
pole switch, ammeter, circuit breaker and reactance coil to choke
back any lightning discharges and force the arrester to operate.

The dynamo panels should be provided with a small double pole
lighting switch, where the station is lighted from the power genera-
tors, so that any generator can light the station independent of the
power bus. This lighting circuit should be looped inside of the
circuit breakers.

The present practice indicates that the best results obtained
are when the lightning arresters are located as near the point
where the feeders enter the station as possible. Behind the
switchboard is not the proper place for the lightning arresters as
a rule.

The panel form of construction is now universally adopted, the
apparatus being mounted on an upper panel, with a foot plate about
twenty inches high below it. These panels are made interchange-
able for the different units and feeders, and the extension of a
switchboard only requires that the bus bar and iron frame be
extended. This gives a very flexible method, and amply provides
for the future growth of the system.

Fig. 90 shows an assembly of a modern form of street railway
switchboard. It consists of an edgewise dynamo regulator "A,"
dynamo switches "B" the positive switch in this case being dou-
ble throw, as this board is arranged for two potentials field switch
"C," dynamo galvanometers U D" and amperemeters "E," also
circuit breakers "F." This panel was designed by the author to
take care of two two-hundred-kilovvatt railway generators; its
width is twenty-six inches, and all connections are made on the

RAILWAY SWITCHBOARDS

Q Q

/ /

^JAS. ^

' S

back of the board, as shown in the side and rear views. The
equalizing in this case is done at the generator, and the copper bus

bars are supported on
the back by cast-iron
brackets and insulated
from them by marble
blocks.

It is very useful in
some cases to be able to
separate the feeder sys-
tems, so that they can
be supplied by indepen-
dent generators, where
extra demands of traffic
require a higher poten-
tial to be obtained on the congested part of the system, in order
that the schedule may be restored. To effect this result, the dynamo
should be provided with a double throw switch, and also the
equalizing system should be double, and the equalizer switch double
throw. If the feeder switches are also provided double throw, the
feeders can be operated on independent generators when required.
It is the usual practice to tie all rail and return grounds to a
common negative bus ; but to reduce electrolysis, in some cases, the
ground returns which are tapped directly to the water or gas-pipe
systems are brought to one ground bus, and the rail or return
feeders are connected into a separate ground bus.

Generators are connected between the pipe return ground and
the positive pole of the system, and the rail return ground to the
positive pole of the system ; these generators are maintained at

116

RAILWAY SWITCHBOARDS

such potentials, that the pipe systems are always kept at a lower
negative potential than the rail system, so that all sneak currents
will flow to and through the pipe system and be taken from the
pipe system at the station where they can do no harm, for electroly-
sis takes place if they leave the pipe system and re-enter the earth
in their attempt to return to the station.

In regard to the conductors behind the board, these are supported
on porcelain insulators, or threaded through porcelain blocks as a
rule, and in this case weather-proof insulation is sufficient for the
conductor itself. All conductors should be stranded, and even the
field wires should be a stranded conductor, in order to reduce the
hazard that a solid conductor used here would increase. In some
cases lead-covered leads are used, but where bare rubber is used for
the insulation, great care should be exercised to prevent oil from
reaching these conductors.

The thickness of the marble used for the switchboard surface
should not be less than one and one-half inches where the circuit
breaker opens on the board, as this blow will crack thinner marble.
Care should also be taken not to drill too many holes in line, either
horizontally or vertically, as it may seriously weaken the marble or
slate through this line.

Exposed terminals of different potential, adjacent to one
another, should be taped and insulated, or so shielded that no
spark can jump between them, for when the circuit breaker opens
on overloads there may be quite a rise in potential on the dynamo,
which sometimes starts a flaring arc between exposed adjacent sur-
faces, which may produce damaging results.

So closely allied to the switchboard connections is the ground
connection of lightning arresters, and so many good lightning

117

RAILWAY SWITCHBOARDS

arresters have been condemned on account of their poor ground
connections, that a word here in regard to this important point will
not be amiss. Every obstruction offered to the flow of this dis-
charge by the ground wire subjects the station apparatus to an
electrostatic stress, tending to break it down at its weakest point,
and every means should be used so that the lightning discharge
can jump the spark gap of the arrester and pass to ground. With
this high frequency which a discharge possesses, it has a tendency
to travel on the surface, rather than on the interior of the wire, and
in this way choke its own passage ; this effect increases as the
diameter of the wire increases, and consequently only a small wire
is used for the ground conductor, No. 6 being the usual size. A
bend in a conductor greatly increases its self-induction, conse-
quently the wire should be as straight as possible from the point of
connection at the lightning arrester to the ground connections.
Carrying this wire parallel to or near masses of iron will also tend to
retard by self-induction the passage of this discharge to earth. To
use a water-pipe system for earth is not the best practice ; but where
it is necessary, a brass lug can be clamped to the water-pipe and
the contact surfaces amalgamated ; into this lug solder the ground
wire. After the connection is made, it should be painted over
with two coats of air-drying asphalt varnish. No ground connec-
tions that are used for any other purpose should be used for the
lightning arrester ground. No part of an iron structure or piping
through the building should be used for the purpose of this con-
ductor. The ground conductor should be connected to the water
system, as near its entrance to the earth as possible.

A ground near running water or naturally moist earth will give
the best results, but in all cases it must be below the frost-line.

118

RAILWAY SWITCHBOARDS

If these cannot be secured, a hole can be sunk in the ground until
water is reached. A copper plate two by two feet, with the con-
ductor firmly soldered to it, will, in ordinary cases, be adequate for
lightning ground. Loose waste metal does not materially increase
the actual contact area of the earth plate. If such material is used
for the earth plate, each piece should be connected with the ground
conductor itself. The best material to use to get a low resistance
ground is broken coke; this should be tamped well in the bottom
of the hole to a depth of about two inches, and then the copper
plate laid on this, and about four inches more coke laid over the
ground plate and tamped well. Dirt can then be thrown over this
and tamped lightly.

BACK VIEW OF
SPECIAL RAILWAY
SWITCHBOARD FOR
MARIETTA STREET
RAILWAY,
MARIETTA, O.
BUILT AND INSTALLED BY

WALKER Co.
CLEVELAND, O.

FRONT VIEW OP

SPECIAL RAILWAY
SWITCHBOARD FOR
MARIETTA STREET
RAILWAY,
MARIETTA, O.
BUILT AND INSTALLED BY

WALKER Co.
CLEVELAND, O.

ALTERNATING CURRENT SWITCHBOARDS

ENERAL features in the construction of alter-
nating current switchboards, wherein
they depart in their designs from those
already described, will be considered.

Increasing the potential brings in a
hazard to the attendant which must be
taken care of, also the leakage between
terminals in a board designed to control
over one thousand volts, if the marble
itself is only depended upon for insula-
tion. All terminals and screws or bolts
holding these terminals to the marble
should be insulated from the marble itself
by mica or micanite, or an equivalent in-
sulation ; fibre is not of any value here, as it is a poor insulator when
under compression. This method of insulating will greatly reduce
the surface leakage, which may become serious in damp weather.

All exposed terminals on the front of the board should be
screened from the switchboard attendant. Several fatal accidents
have occurred from the neglect of this point, as the attendant has
fallen against the exposed terminals with serious results. If these
points be borne in mind, the alternating current switchboard
becomes very simple in design.

The severing of high potential circuits, when alternating, is not
so difficult, as the current reversals do not maintain such a fierce

1 20

ALTERNATING CURRENT SWITCHBOARDS

arc between these switch terminals, as in the case of much lower
direct potentials.

In the switches used, it is usual to place over them a marble
slab, having the free switch blade in front, and with recesses in
this marble which screen the clips. Another form of construction
is to have only the switch handle on the front of the board, and all
the switch mechanism arranged behind the board, with a handle
projecting in front which can be pushed or pulled to shift the con-
nections from one side of the system to the other.

Alternating current machines, not. producing a character of cur-
rent which will magnetize their own field, require separate exciters
and a separate field system. Also as alternating current dynamos,
they are not, as a rule, run in parallel ; this again separates the
feeder system, so that each feeder can be supplied by any generator.
The character of the current also allows of regulating devices
which increase or diminish the potentials supplied that feeder, by
inserting in it transforming, regulating, or compensating de-vices.

In long-distance transmission, where high potential is necessary
in order to reduce the copper line costs, the alternators deliver the
current at normal potentials to step-up transformers, which again
increase the potential and reduce the current for a given kilowatt
output.

All these characteristics and peculiarities of alternating current
systems have to be taken care of in the switchboard design,
besides the different connections required by the two-phase, three-
phase, polyphase and monocyclic. First taking up the field con-
nections and exciting methods where there are several alternators,
it is better economy to use one exciter of sufficient capacity for all
the alternators than a separate exciter for each machine; provid-

ALTERNATING CURRENT SWITCHBOARDS

ing two exciters of sufficient capacity gives a duplicate exciting
system.

A rheostat is used in the exciter fields, and all the alternators on
this exciter can have their potential raised or lowered together.
Each alternator is again provided with a rheostat in its field circuit,
.so that each alternator can be independently regulated. Where
the alternators are not worked in multiple, a pair of horizontal bus
bars are used for each single phase machine. The feeder is pro-
vided with a double throw switch, the middle clip of which has the
feeder connected to it. The other terminals are provided with a
plug receptacle, and also the bus from the generators. In this way
the feeder can be connected to any bus, or changed to any other
dynamo by first plugging in to the proper generator and then
throwing over the switch of that feeder. Where the alternators
are worked in parallel, only a double pole switch is required for
the feeders, as alternators are added as the load of the feeders
increases.

In order to accommodate the different combination of phases,
the proper switching arrangements are shown in the diagrams for
these different systems.

In working alternators in parallel, it is necessary to have both
the potential and the period at which it occurs in step, in order that
the generators may be thrown together and work synchronously.
Two generators considerably out of step will jump together
when connected in multiple, but will throw considerable strain
on the armature and driving mechanism. It is also evident
that a potential indicator will not be useful in putting alternators
together, consequently a synchronizer is necessary. Fig. 91 shows
a form of visual synchronizer, in which there are two primaries,

122

ALTERNATING CURRENT SWITCHBOARDS

one actuated by the electro-motive force of one generator, and the
other by the electro-motive force of the other generator ; each pri-
mary induces an electro-motive force of fifty
volts. When these two primaries are acting
on the same secondary, and both currents
are in phase, the electro-motive force will be
one hundred volts, and the lamp will be
maintained at full candle-power. The con-
nections usually employed are those to
reverse one of these primaries, so that the
induced electro-motive forces oppose each
other, and the lamps are out when the

FIG. 91

generators are in phase. The objection to this connection
is, that if the lamp happens to break or the circuit be open,
the alternators will be thrown in when they are probably out
of step.

An acoustic synchronizer is also used, in which the currents
from the two machines to be thrown together act oppositely on a
diaphragm. When there is a phase difference between the two
currents, the diaphragm is in vibration, but when the phase on the
two machines is synchronous, the acoustic syn-
chronizer does not emit a sound, as the attrac-
tions are equal and opposed. Fig. 92 shows
this form of synchronizer.

The phase indicator is an instrument that
shows the angular difference which occurs
between the maximums of two varying currents.
Fig. 93 shows an instrument for this purpose, in which the current
from the two sources, when they are out of phase with each other,

FIG. 92

123

ALTERNATING CURRENT SWITCHBOARDS

tend to rotate an armature ; this effort is proportional to the phase
difference.

Voltmeters for alternating currents do not, as a rule, measure

directly the potentials of the cir-
cuits, but a transformer is used with
a known ratio of transformation,
which reduces the potential of the
circuit to fifty or one hundred volts.
A voltmeter is connected across
the secondary of the transformer,
and the machines regulated by this
voltmeter.

Alternating ground detectors
act in the same manner when one
end of the primary is connected to
the line and the other end to ground; if there is any leak on
the system, a current will flow, and for the brilliancy of the
lamp, the amount of external resistance to ground can be
judged.

In regard to conductors used for alternating currents, ones hav-
ing diameter larger than one-half inch should be stranded, as
beyond this size there is considerable more drop than that due to
the ohmic resistance of a conductor. Owing to the tendency of the
alternating currents to distribute themselves unequally across a sec-
tion of a conductor, and flowing more on the external surface rather
than the interior of the conductor, this effect decreases as the
frequency is reduced. Feeders were formerly protected by fuses,
which were enclosed in a fuse box having a semi-circular groove in
which the fuse was laid.

FIG. 93

124

ALTERNATING CURRENT SWITCHBOARDS

Circuit breakers are now being used extensively on the alter-
nators, and afford them the same protection which is so necessary
in railway practice.

The connections for the most prevalent systems used in practical

work are shown.

Diagram No. 1 1 il-
lustrates the method of
connecting a single
phase system having a
common field circuit
and independent dy-
namo circuits. Trac-
ing out the field con-
nections, the current
generated at the ex-
citer "E" passes to
the terminals of the
switches "S," "ES,"
and when this double
throw switch makes
connection with the
field busses, the cur-
rent is returned back
to the exciter through
the other pole of the

switches. In this way the fields can be operated on either exciter
and shifted from one to the other without opening the field cir-
cuits. These connections are entirely independent of the arma-
ture and its potentials, one-hundred-and-ten-volt system being

Diagram NO I I

125

ALTERNATING CURRENT SWITCHBOARDS

generally used for exciting; the construction of this part of the
alternating current switchboard can be followed out on the lines
given for low potential switchboard construction.

Regarding the generator terminals, one is connected to the cir-
cuit breaker, then to
one pole of the dou-
ble pole switch, and
carried through the
ammeter to that alter-
nator's bus; the other
terminal of the alter-
nator is also carried to
the double pole switch
to that alternator's
bus. In this system,
each machine is pro-
vided with a pair of
busses, extending pref-
erably behind the
board, so that the
feeders can be connec-
ted to these different
busses.

Where only two

alternators are used,
it is evident that double throw double pole switches will make the
proper connections ; but when more alternators are used, it is the
usual practice to provide plug terminals on the horizontal dynamo
busses for each feeder, and also on the double throw feeder switch,

126

ALTERNATING CURRENT SWITCHBOARDS

so that this feeder can be plugged into any dynamo; by having
this feeder double throw, in order to change from one dynamo to
another, the idle terminals of the switch can be plugged into the
dynamo to which this feeder is to be transferred, and in this way it

can be shifted quickly
without dipping the
voltage on the light
circuits.

Voltmeters are gen-
erally provided for
each alternator, and
an ammeter for each
feeder ; also, where
the external distribu-
tion requires, a com-
pensator is introduced
into the feeder circuit.
By varying the in-
ductance of the com-
pensator, the voltage
on the external sys-
tem can be varied ;
a ground detector
is also placed on this
board with a multi-
point switch, so that any circuit can be tested for grounds.

Diagram No. 12 shows the connections used where the alterna-
tors are run in parallel ; in this case the field connections are the
same as those shown in Diagram No. n. A double pole switch

TT IT HI

127

ALTERNATING CURRENT SWITCHBOARDS

can be used for the alternators, and all the alternators connected
to a single pair of bus bars; the feeders only require a double
pole switch, and in order to throw the alternators together a syn-
chronizer has to be added and connected between the alternators.
Lightning arresters are
not shown in these
diagrams'; as they
are usually connected
to the feeder at its
entrance to the sta-
tion.

Diagram No. 13
shows the connec-
tions required by the
two-phase three-wire
system, with the al-
ternators operating in
.parallel, and Diagram
No. 14 shows the con-
nections for a two-
phase four-wire sys-
tem. The connec-
tions for the two-
phase three-wire sys-
tem are the same as

those used for the monocyclic and three-phase systems, the
middle wire only being used for power circuits in the mono-
cyclic system, and the two outside wires for lighting sys-
tems.

128

ARC LIGHT AND SPECIAL SWITCHBOARDS

ISTORICALLY the arc light switchboard
was the first combination of apparatus for
the distribution of current foi illuminating
purposes which could be strictly called a
switchboard. The requirements to-day
have altered very little the constructional
features from those originally installed,
with the exception that the number of
lights carried by a single dynamo has
steadily risen as the art of insulation
became more perfected, and to-day a one-

hundred-and-fifty lighter is in practical operation, which means a
potential of approximately seven thousand five hundred volts.

The same care in insulating the terminals from the switchboard
itself has to be exercised, but the current quantity is low and does
not produce a very serious arc when the current is broken.

The arrangement of the switchboard must be so flexible that
any dynamo can be connected to any circuit, and any circuits can
be looped together on any dynamo. The general arrangement to
effect this is to bring all dynamo terminals to a series of plug
receptacles which align with all the lamp circuit terminals. There
are also transfer busses provided, as shown in Diagram 15. Here
No. i machine is connected to No. i circuit; No. 2 machine
operates No. 2 and No. 3 circuits in series; No. 3 machine plugs
on the transfer bus, and circuits Nos. 4, 5 and 6 are in series, and

129

ARC LIGHT AND SPECIAL SWITCHBOARDS

the end of circuit No. 6 is transferred back to machine No. 3 by
means of transfer bus. This shows the general combinations
required on a switchboard. A number of forms of plugs have
been devised for connecting these circuits. Fig. 94 is the oldest

type, which is simply
a plug with the dy-
namo or lamp cir-
cuit cable directly at-
tached.

Fig. 95 shows a
lock form, in which
the plug is inserted
between the spring
and a notched latch,
for the purpose of
connecting these cir-
cuits together.

Fig. 96 is a plug
only, without any
connecting cable to
it; the dynamo ter-
minals are located on
one face, and the
lamp terminals on
another opposite face

about three inches apart. The plug is long enough to con-
nect these two terminals, it passing through a spring bushing
on the front face and registering with a tapering recess in the
back face.

130

FIG. 94

ARC LIGHT AND SPECIAL SWITCHBOARDS

Fig. 97 shows the form of plug, which, besides having the plug
contact, also has a sleeve, which, when the plug is
withdrawn, slides over the exposed plug terminal and
screens this contact from the arcing
effect.

The lamp circuit is provided with
an ammeter, which generally has in
connection with it a polarity indicator,
so that the attendant can see whether
he has plugged in his circuits correctly.
Each leg of the circuit should be pro-
tected with lightning arresters.

The dynamo controller is gener-
ally located on a pedestal near the machine it con-
trols ; but in small plants these controllers are

also assembled on the switch-
board face with the rest of the
apparatus. The general
method of making these
switchboards to-day is to sep-
arate the two potentials en-
tirely, the positive usually
being on the upper side of the
board with the positive trans-
fer busses, and the negative
on the lower side of the board.
In mounting these terminals
FIG. 9 6 on the switchboard face, the

same precautions should be exercised as are given for high poten-

FiG. 95

ARC LIGHT AND SPECIAL SWITCHBOARDS

tial switchboards. A pressure switch is often used in connection
with the arc light circuit, in order to determine the pressure across the
terminals of the circuit, and also the number of lamps on the circuit.
By having a switch arranged so that one side of the voltmeter
can be connected to ground, the insulation to ground can be tested
while these circuits are in operation. Only the highest grade of

insulated wire should
ever be used to make
these switchboard con-
nections, and the flexible
cables in front should be
stranded of wires, not
larger than twenty -two.
There are a great
number of special
switchboards designed to
fulfil special conditions,
but they are not so
generally used as to
make their description
of value. A few cases
have been selected to
illustrate this class
where originality of design has been displayed. Fig. 98 shows the
accepted form built for the Navy, where compactness was one of
the essential features. It was also important that any circuit could
be supplied from any generator independently.

The switchboard shown illustrates the arrangement for three
generators and eight feeders ; the knife switch employed here has a

FIG. 98

132

ARC LIGHT AND SPECIAL SWITCHBOARDS

movable blade which can engage in the feeder terminal and be
thrown so that any dynamo can be connected to any feeder. In this
case the potentials are separated on two sides of the board right
and left-r-and the adjacent terminals are brought very closely
together on account of them being the same potential. The

dynamos are connected in multiple
by the switches at the bottom of the

cut, and the instruments and regu-
lators are assembled on another board.
In the production of scenic effects,
fine gradations of illumination are
required, in order that the desired
effect can be produced. This requires
inserting a variable resistance
in series with each of the lamp
circuits to be controlled, and
the regulators are interlocking,
so that any group or groups of
lamps can be varied together.
Feeder switches and pilot lamps
which are across the different

circuits controlled, are assem-
bled together for this special
FIG. 99 work.

Fig. 99 shows one of the new types of switchboards designed
for this purpose.

In alternating current work the resistance is substituted by a
choking coil, which has a variable choking effect on the lamps in
series with it.

ARC LIGHT AND SPECIAL SWITCHBOARDS

It has been the author's attempt in writing this essay on
"Modern Switchboards" to give a practitioner's review of the
art, and bring out such points in construction, appliances and
connections as are necessary for their proper construction and
design.

CIRCUIT BREAKERS
AND THEIR USE IN POWER TRANSMISSION

BY \V. H. TAPLEY
Chief Electrician U. S. Government Printing Office, Washington, D. C.

Reprinted from The Electrical Engineer by permission of Mr. Tapley and the Editors

When the application of individual electric motors to driving
machinery became firmly established in the manufacturing world,
and was conceded to be a more economical method of power trans-
mission than belting with long lines of shafting, second only to the
motor, and how properly to connect it to the machine which it was
to operate, was the subject of suitable protection both to motor and
machine.

The first thing that suggested itself, and naturally, was to pro-
tect the motor in the same way as lighting circuits, namely, to
introduce a suitable fuse. This was done, and where motors were
belted the results attending overloads were rather of an annoying
and aggravating nature than anything which could really be called
serious; yet when gearing and the direct application of armature
to the main driving shaft of a machine began to supersede the belt,
it was only a short time before the fact that a fuse was not an
adequate protection became forcibly impressed upon the advocates
of electrical power transmission.

As the art advanced, the thing to which the electrical engineer
would turn for a rational solution of the problem was the electric
current itself. How well this has been accomplished is shown by

CIRCTIT BREAKERS AND THEIR USE IX POWER TRANSMISSION

the successful introduction of the circuit breaker, now so univer-
sally used in all large power plants. That the magnetic property
of the electric current was the means best adapted for the actuation
of the protective device, and gravitation the most reliable force for
governing its operation, is seen by the great superiority of the cir-
cuit breaker which depends entirely upon these forces, over those
in which the effect of the actuating current is subject to variation
due to extraneous conditions.

PROTECTION AND WHAT IT SHOULD BE, AS APPLIED TO
A LARGE MANUFACTURING PLANT

In treating of this subject the tendency of the engineer has
been to regard it almost entirely from an electrical point of view,
incidentally, if at all, considering that which affected the real suc-
cess of the manufacturing establishment employing motors, namely,
constant service and lowest cost of production. Before entering
further into this matter, let us see what is absolutely required to
give a manufacturing establishment protection worthy of that name,
when using electrical power transmission.

First To secure the protection of electrical apparatus from
motor to generator.

Second To provide a method which will afford ample protec-
tion for the machinery to which electric motors are attached.

Third To secure a freedom from interruption of production,
and avoid the exasperating delay which is experienced in replacing
any part of the protective device after the same has been called
into service.

Fourth After protecting everything in the shape of machinery,
the safety of building and electrical apparatus and providing

136

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

against the stoppage of production, the mat-
ter of reducing to the lowest possible point
the liability of accident to the operators
required to handle either motors or machin-
ery must be considered, as indeed this is a
matter of supreme importance.

That all the above-mentioned features are
ever present, confronting the engineer, who
is responsible for the successful operation of
a manufacturing plant, is confirmed by the
large number of so-called protective devices
already offered to the public.

At present a very large part of the labor of
the electrical engineer and the manufacturers
of this kind of electrical apparatus has been
directly in one line, that of protecting the
electrical apparatus from the effects of over-
heating and the building from fires which
might occur from heavily overloaded circuits.

As this feature is commanding a large
share of attention in the electrical press and
manufacturing world, it would seem best to
devote our time in this article to the field
suggested in the last three of the foregoing
propositions, which, if satisfactorily solved,
of necessity cover all the ground now under
consideration by engineers, on the subject
of proper and positive protection to electrical
apparatus as applied to transmission of power.

'37

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

The pronounced success during the past two or three years of
the direct application of electric motors to all kinds of machinery
has put this method of power supply so far in advance of other
methods that, notwithstanding the comparatively high first cost, it
is now considered the most economical method, and should be
adopted by every large manufacturing plant where the work is, in
any sense, of an intermittent nature.

Let us now look from the electrical side to that of the manufac-
turing plant proper, and see if protection is not even more impor-
tant and imperative here, where very much larger sums of money
are invested, and until now have been wholly neglected except by
insurance from fire.

To suggest something which may form a topic for discussion, let
us take the case of a printing-press to which is directly connected
an electric motor. The cost of the printing-press is, roughly speak-
ing, $3,000.00, and that of the motor equipment $300.00. (These
are nominal figures, which vary with the class of press used and
character of work required from it; the cost of motors also varies
considerably, but these figures are conservative and fully within the
figures for which good apparatus can be purchased.) Allowing the
electrical to be one-tenth the cost of the mechanical installation,
does it not seem strange that it has always been the motor which it
has been the sole idea of the engineer to protect, notwithstanding
its cost is insignificant as compared with the value of the machine
to which it is attached? Is it any wonder that the manufacturers
of costly machinery, such as printing-presses, have looked with
doubtful eye upon the method of direct motor application, whether
it be by gearing or having the armature of the motor keyed to the
main shaft of the press? The manufacturer well knew that for a

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

short time the motor was capable of producing perhaps five times
its rated output, and realized that if this period covered only a few
seconds, there was a great probability that it would be sufficient to
ruin the press should anything occur which would tend to stop it
suddenly. He did not feel that there was even the protection
which is afforded the presses when driven by belts, for these would
slip when called upon to do much more than the normal work of
driving the press.

The writer takes the same ground as the machinery builder, and
when the representatives of companies manufacturing electrical appa-
ratus were asked about this matter, they invariably assured the pur-
chaser that a fuse inserted in the circuit supplying the motor with
current would provide against all possible trouble of this kind. It
was tried ; the fuse worked in some cases and we began to take cour-
age, thinking that perhaps we were too particular and that the fuse
afforded the required protection. It was seen, however, that the blow-
ing of the fuse might serve to protect the motor but not the press.

It is impossible to change over from one method of operating
machinery to another without meeting failures, due perhaps to
nervousness on the part of the operator when called upon to do a
thing for the first time ; and certainly it was so when motors were
first used in this manner. The sudden turning on of the control-
ler naturally blew the fuse, which too often was sufficiently
increased in size to prevent this annoyance. But at its best the
fuse served only to protect the motor, the requirements of the
motor-driven machine being altogether overlooked. To better
appreciate the shortcomings of the fuse in this respect, it is only
necessary to understand the conditions under which it operates to
open the circuit in which it may be placed.

139

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

The effective energy which must be supplied to the fuse is made
up of the following quantities : First, heat sufficient to raise the
temperature to that of the melting point of the fuse ; and second,
an additional amount of heat proportional to the mass and latent
heat of fusion of the fuse, while in addition to this, the heat, being
radiated by the fuse and its terminals, must be supplied. It will be
seen from a consideration of these facts that the fuse requiring a
relatively large excess of energy to effect its operation will permit
a proportional excess of power to be supplied to the motor. The
damage which may result from this may perhaps be more readily
seen by an example. Suppose a foreign body gets into the work-
ing parts of a machine which is directly connected to a motor.
The power supplied by the motor is now expended in the wreck-
ing of the machine, the weakest parts yielding first to the strain.
Only a very short time is necessary for the execution of great dam-
age. The possibility of damage is then limited only by the energy
required to blow the fuse.

It was only upon the advent of the modern circuit breaker that
protection worthy the name was secured for motor-driven machin-
ery. Owing to the much lessened energy required in the opera-
tion of this device, the time required, upon the occurrence of an
abnormal flow, for the opening of the circuit is minimized, while,
in addition to this, the heavier overloads are made to contribute
some of their energy to the acceleration of the circuit-opening
switch, thereby still further decreasing the time of opening.

It may thus be seen that by the use of a properly constructed
circuit breaker the excess of power which may be communicated
to the machine is vastly reduced as compared with the fuse. In
fact, the time element is so lessened that the possibility of damage

140

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

to machine as well as to motor is practically limited to that due to
their combined momentum. What this is in each individual case
makes it necessary to decide whether an auxiliary break to take
care of the same is necessary or not. This is a question for the
mechanical engineer to solve ; but whatever this may be, it does not
affect the principles set forth above, and only serves to bring out
more clearly how necessary it is to shut off the current instantly
and thus prevent the machine from acquiring any additional
momentum.

As the question of machinery protection has been given so little
consideration, it was deemed advisable to bring to the reader's
notice, in rather a minute way, all the possibilities that it may have,
in order to give to the subject the importance which we feel it
deserves. In the foregoing, a reference was made to the compara-
tive cost of a printing-press and that of the necessary electrical
equipment to drive the same; /. e., $3,000.00 for the former and
$300.00 for the latter a relative value often to one, which justifies
the statement that the protection of machinery is a much more
important consideration in electrical transmission than that of the
motor. Can the electrical engineer afford to neglect this important
feature of machinery protection and still hope that his customer
will secure satisfactory results? for, after all, it is necessary that the
new system, as a whole, shall be made more productive, and there-
by more profitable, than the old.

It may be assumed that our third proposition is intended more
especially for the manufacturer than the engineer ; but, taking the
ground that that which is of importance to the buyer concerns the
seller also, we believe it is worthy of the close attention of both.
As the writer has had an extended experience upon the application

141

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

of electricity to printing machinery, he hopes to be able to treat
this branch of a manufacturing business in a more positive manner
than he could do should he endeavor to extend its scope into a
more general or theoretical field.

Referring once again to the printing-press, with its electrical
equipment cost (which we assume to be respectively $3,000.00 and
$300.00), let us see what the production of such a press should be
when it is running for three hundred days in the year on a fairly
good class of printing, and what costly affairs stoppages of presses
are, no matter what the cause. A press should earn an average of
$10.00 a day, or $3,000.00 a year. This is not intended to express
net earnings, but simply the average gross earnings for the press on
commercial work, and we assume that it is running on such work
continuously. It will be seen that delays caused by an accident to
a press may prove much more expensive to the printer than
the actual cost of the repairs to the press itself, as a delay of a week
means a loss of $10.00 a day, or $60.00; and serious accidents
often mean a month of working days, or $260.00, aside from the
cost of repairs, which experience has shown are not to be lightly
considered.

Nor is it to be forgotten that the press to which an accident
usually occurs is generally running on a piece of work which must
be completed within a given time. This means that we must lift
the form and place it upon another press which has to be "made
ready," perhaps interfering with other work, and all this addi-
tional cost must be borne by the manufacturer without any return
for the same. Is not the manufacturer fully justified, then, in
demanding that his machinery and output be equally considered
with that of the electrical apparatus in the matter of protection?

142

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

If the protection of apparatus worth $300.00 is deemed so impor-
tant as to occupy, as it undoubtedly does, the attention of the fore-
most electrical engineers, are we not justified in taking the position
that protective devices should be so constructed as to fully protect
the manufacturer at all points, and not stop with the electrical
equipment alone?

By the use of the highest-grade circuit breaker now offered to
the public, which fulfils in a very satisfactory manner all the
requirements thus far considered, such a saving may be made, not
only to the motor, but to the machine which it drives, that the loss
occasioned by stoppages and on account of repairs will be practi-
cally eliminated, and the device will pay for itself many times in
the first year.

The delays incident to the blowing and replacement of fuses are
perhaps more annoying in newspaper and publishing offices (where
mails have to be met and where the time for the completion of a
particular piece of work is limited) than in most classes of manu-
facturing; but in any case the saving is so important as to amount
to very much more than the cost of adequate apparatus. Indeed,
it would seem to be the best paying insurance which the manufac-
turer could possibly obtain. Such delays as we have been consid-
ering are practically unknown when the fuse is replaced by a
thoroughly mechanically and electrically constructed circuit
breaker.

With the more universal adoption of the individual motor in
electrical power transmission comes also the question of protection
to operatives employed in handling the machinery. This is of so
much importance that most of the States in this country have
appointed inspectors to visit all manufacturing establishments and

143

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

see that the proper precautions are used, and every means employed
to lessen the danger to the employees, of whatever nature it may be.
With all the precaution taken to prevent accidents, it is impossible
to do away with them entirely. Strange as it may seem, the care-
lessness or foolhardiness of the employees themselves is mainly
responsible for most of the accidents which occur to-day, thanks to
the hearty co-operation of inspectors and employers in their endea-
vors to make accidents practically impossible. Yet I know of noth-
ing to prevent a man from placing his hands in close proximity to
running gears, and if he happens to have a piece of waste in his
hands and the same gets caught, the resulting damage is only limi-
ted by the quickness with which the machine can be stopped.
Several such cases have come under my personal observation, and
only the prompt opening of the circuit breaker prevented very
serious results. In one case a man's fingers were pulled into a train
of gears, he endeavoring to clean the press while in motion. His
fingers were badly jammed, but the opening of the circuit breaker
stopped the press before any serious damage was done, and thus
saved the man three fingers. Another case was that of a man
who got his arm caught between the two cylinders of a press. He
was fixing the packing on one of the cylinders and motioned to the
feeder to reverse the press. Instead, she started it ahead suddenly
with the result that his arm was drawn in between the revolving
cylinders, and again, owing to the instantaneous opening of the cir-
cuit breaker the pressman escaped with a severely bruised arm,
instead of crushed bones as we had all expected.

These serve as examples upon which to base a requirement that
protection to operatives is not a matter of minor importance, and
should not be put aside with the remark: "Let the employees keep

144

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

their hands out of the machinery we cannot protect everything
and everybody." To be sure, we have no electrical or mechanical
device which will make brains for the ignorant or prevent careless
operatives from getting hurt ; yet devices can be made, as we have
seen, which reduce the results of such carelessness to a minimum,
and their employment should become more general as the success-
ful operation of them becomes better known.

Having thus considered in detail the protection demanded by a
large manufacturing plant using individual motors as the method
of power supply, we are confronted with the proposition : can these
demands be met to the satisfaction of all concerned to the builder
of machinery, as well as of electrical apparatus, and to the manu-
facturer using them ? If such is the case, what is the device neces-
sary to fulfil the requirements, and is it commercially obtainable?

With reference to the fuse, we have only to read any of the
scores of valuable papers \vritten upon their action to fully justify
us in \vriting against it "not satisfactory' and passing on. As
the current itself may be the means of protection, as well as that of
propulsion, the method of opening the circuit electrically should be
employed. This has been successfully accomplished in the modern
circuit breakers which operate on the inverse time element rather
than the constant time limit.

The ultimate requirements of a circuit breaker are that we can
rely upon it to do all that we have shown it should do, and operate
successfully, not once, twice, or for a month, but always. When
this ceases to be the case, the magnetic circuit breaker will be
superseded by some other means of protection.

To produce such an instrument the highest skill, electrical and
mechanical, is required. Long study of the existing state of the

CIRCUIT BREAKERS AND THEIR USE IN POWER TRANSMISSION

art and the conditions under which circuit breakers operate is
necessary, and the closest attention must be given to every detail
of manufacture.

With this an accomplished fact, such results are not the only
reward, however, nor should they be. The public will cheerfully
pay, not only for the labor and material used in its production, but
also a profit sufficient to encourage the maker and enable him to
continue the work, for the perfect is never obtainable and is only
reached approximately. In electrical science, perhaps more than
in any other, we are never able to write the word " Finis."

146

THE DEVELOPMENT OF THE CIRCUIT BREAKER

T is an old saying that fire is a good servant but a poor
master a homely setting forth of the broad principle
that Nature's forces become valuable to us only as we
learn not merely how to harness them, but also how
to properly restrain them in harness. In general, the
more effective the force when under control the greater the danger
to be apprehended should it overcome restraint. In pursuance of
these principles, it is seen that the application of the physical forces
to man's needs is invariably handicapped by man's inability to keep
these forces to their proper channels.

An illustration of this is seen in the history of steam engineering.
For many years after steam was first used as a motive power, its use
was restricted to very low pressures, that it might be the more readily
confined, and it was only with the application of the safety valve
and improved methods of boiler and engine construction that the
high pressures of later years became safe and practicable.

The domain of electricity furnishes us with another example.
The one difficulty which must be overcome before the long-distance
transmission of electricity becomes a complete success, lies in the
absence of proper insulation, and it is only when this need shall be
fully met that electrical transmission over long distances will become
a prime factor in modern engineering.

These two instances are certainly ample warrant for the state-
ment that the work of the engineer consists, not only in the devis-
ing of means by which Nature's forces may be directed into the

f ( r/5 ^^

L, H <

>-i CJ

>-) <j in

U H ^~

W K

^ B

*""* Crf

THE DEVELOPMENT OF THE CIRCUIT BREAKER

service of man, but also in the adaptation of further means to the
end that these forces may be readily restrained and kept from effect-
ing damage.

As has been hinted, this fact is eminently true in the domain of
electrical engineering. It is not surprising, therefore, that as the
uses of electricity have vastly increased, and as the means for
employing it have multiplied, there has been a marked development
in the methods and devices employed for the protection of electric
circuits.

In the pioneer days of electrical engineering, the introduction of
strips of fusible metal into a circuit was supposed to afford that cir-
cuit ample protection in the event of an excessive flow of current.
This device, however, was only good enough while there was nothing
better to take its place. The objections to this method of protec-
tion are best appreciated by those most familiar with its results ; for,
aside from theoretical considerations, experience has taught that the
melting of a fuse is dependent, not simply upon the passage of a
certain volume of current, but also on the manner in which the
fuse is connected into circuit, and the relation between it and the
walls and cover of the fuse block. In very many cases the time
consumed in the melting of the fuse, after the occurrence of a dan-
gerous overload, was sufficient to allow great damage to be done in
other portions of the circuit. Further than this, the replacing of
fuses involves delays frequently inadmissible in this age of haste.
These are but the more evident reasons why the fuse has given
place upon the "Modern Switchboard'' to the circuit breaker, an
instrument which, in its approved form, involves no such uncertain-
ties of action as are inherent in the fuse. Let it not be supposed,
however, that the circuit breakers which were first introduced pos-

149

I-T-E CIRCUIT BREAKER

MIDGET JR. SINGLE POLE. 5 TO 25 AMPERES

THE DEVELOPMENT OF THE CIRCUIT BREAKER

sessed all the varied points of excellence which are realized in the
perfected device ; for while even the earliest forms offered a decided
advance over former methods of protection, they still left room for
improvement in many respects.

Those familiar with the first circuit breakers will remember that
in their design the automatic features were secured at the expense
of conductivity; in other words, a quality of primary importance
for normal operation was sacrificed in order to provide for abnormal
conditions.

Another, and no less vital weakness, evident not only in the first
circuit breakers, but also in many later forms, was due to the fact
that their operation was not entirely independent of everything
save the actuating current. This was because such variables as the
tension of springs and the friction between metallic surfaces were
allowed to enter into or to influence the adjustment of these devices,
and naturally resulted in much uncertainty in their operation.

One by one, however, these and other faults have been overcome,
and the circuit breaker, in its most advanced form, combines with a
conductivity not surpassed by that of the best switches of corres-
ponding capacity, an accuracy of operation which compares favorably
with that of the best ammeters, and a certainty of action compar-
able with that of gravitation itself.

Added to these developments there has been a vast increase in
the scope of the circuit breaker. An instance of this is seen in the
adaptation of the instrument to alternating current service. Until
recently this has been considered to be out of the question, some of
the reasons being that the earlier forms offered too great an im-
pedance to alternating currents of even ordinary frequency, or
heated unduly, owing to eddy currents or to hysteresis in the iron

THE DEVELOPMENT OF THE CIRCUIT BREAKER

portions, while some of the instruments gave rise to a disagreeable
humming noise when traversed by the alternating current.

However, after a careful study of the special conditions involved,
these difficulties were traced to their sources and their remedies
effected by adaptations of design and materials to the peculiar
demands of the alternating current, resulting in the production of
an instrument as perfect in its operation as the better known
direct current circuit breaker.

Until recently, the operation of the circuit breaker has been
limited to opening the circuit in the event of an overload ; but in
response to the demands of the ever-progressive engineer, there has
been produced an instrument which effects the same end, upon the
occurrence of a predetermined minimum flow or "underload."
From this it was but a natural step to a combination of these func-
tions in a single instrument, resulting in the production of a circuit
breaker, operating upon the occurrence of either an underload or an
overload, and this was accomplished without a sacrifice of any of
the features entering into either the efficiency of operation or
ease of manipulation, which characterize the simple overload instru-
ment.

The single pole circuit breaker was a well-established success
before a satisfactory double pole instrument was produced. The
chief obstacle lay in the difficulty of obtaining a proper insulation
between poles, without sacrificing the strength and the absolute
rigidity necessary on account of the automatic action of the instru-
ment.

Before persistent effort, however, these difficulties vanished, and
the engineer has now at his command a double pole circuit breaker
which fully meets the most severe requirements.

152

THE DEVELOPMENT OF THE CIRCUIT BREAKER

While these few cases indicate, in a measure, the lines along
which the development of the circuit breaker has extended, there
have been improvements of almost if not quite equal importance
in directions other than those which have been cited. The circuit
breaker has been adapted to service in circuits employing the
highest pressures of ordinary practice, while in point of size it has
been made to deal with the largest and smallest currents with equal
efficiency and ease. A whole power station may be "opened up"
by the operation of a single massive instrument, or the circuit of a
single incandescent lamp may be automatically broken by a tiny
device which may be covered by the hand.

To those familiar with the development of electrical apparatus,
it is hardly necessary to mention that the continued progress of the
circuit breaker has been effected only at the expense of time,
thought and money, unstintingly applied. While it is no trifle
in itself, the circuit breaker largely depends for its ultimate perfec-
tion on a multitude of details of seeming insignificance. Even such
apparently unimportant matters as the depth of a contact, the
exact temper of a switch blade, or the diameter of a bearing pin,
are carefully thought over and planned, while the quality, quantity
and disposition of the various metals entering into the electrical
and magnetic portions of the instrument are made the objects of
careful and exhaustive test. The designs of the circuit breaker are
the outcome of long experience and close research, while the
exactness of their construction and the beautiful finish, by virtue
of which they enhance the appearance of even the handsomest of
"Modern Switchboards," is the product of the most careful work-
manship, aided by the most improved machinery that skill can
devise or money secure. w. M. SCOTT, M. E.

i53